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

Metaphysics & Epistemology
195

That’s an interesting point, because it touches on a genuine interpretive question about what Witt means by “acting” at the bottom of the language game.

You’re right that Witt’s language games include actions, gestures, training, and shared practices. The builders shouting “Slab!” is a language game inseparable from the activity of handing stones. I don’t dispute that. And you’re right that grammar in Witt’s sense is not a set of explicitly formulated rules but the normative structure of our practices. I also agree with that.

But I think your reading proves too much. If “language” is expanded to include all coordinated activity, all trained responses, all patterns of going on correctly or incorrectly, then the word no longer distinguishes anything. A dog fetching a ball is participating in a pattern of coordinated activity with implicit norms. If that counts as language, then everything conscious beings do becomes language and the distinction between linguistic and non-linguistic dissolves. Witt doesn’t want that. He draws distinctions between what can be said and what shows itself, between propositions and what lies beneath them. Those distinctions lose their force if “language” swallows everything.

The child example is where I think the disagreement becomes sharpest. You say the point of OC 128 is that the child participates in a practice with implicit norms first, and only later can articulate those norms. I agree that children are embedded in social practices from birth. But certain engagements with the world aren’t things the child acquires from those practices. An infant reaches for an object it expects to persist. Nobody trained the child that objects persist. The child brings that certainty to the practice rather than acquiring it from the practice. It’s what makes the training possible in the first place. A child who didn’t already act as though objects persisted couldn’t be trained into a language game involving objects. The engagement precedes and enables the normatively structured activity, not the other way around.

The key difference is between normatively structured activity and activity that precedes normative structure. Within a language game, you can go on correctly or incorrectly. There are norms, even if they are implicit. But the certainty that objects persist, that other beings exist, that experience is continuous, these aren’t norms you can get right or wrong. There’s no “going on incorrectly” with respect to object persistence. You don’t fail at it the way you fail at a language game. There is only the catastrophic breakdown Witt calls mental disturbance, and that’s not an error within a practice. It’s the collapse of what makes practice possible. This is the distinction between mistake and mental disturbance that Witt develops in OC 71-74. A mistake can be fitted into what you know aright. But when something at this level gives way, there’s no surrounding framework within which to locate the error.

OC 204 says acting lies at the bottom of the language game. I take “at the bottom” seriously. The acting isn’t inside the language game. It’s what the language game rests on. If it were already part of the language game, it wouldn’t be at the bottom of it. It would just be more of the game. The whole force of OC 204 is that giving grounds comes to an end, and what you hit when it ends is not another proposition, not another norm, not another move in the game. It’s acting. The bottom is not more game. The bottom is something the game stands on. This is consistent with OC 253, where Witt says at the foundation of well-founded belief lies belief that is not founded. Not founded within the practice. Not founded at all. That’s the level I’m pointing to.

196

The infant reaches, tracks, grasps and follows. Caregivers respond, correct, reinforce and stabilize patterns. Over time, these interactions solidify into practices where objects are treated as persisting. At no point do we need to posit a prior belief that makes the practice possible. What makes the practice possible is a capacity for being trained into patterns of activity, not a pre-existing certainty.

I don’t deny that ‘correct and incorrect’ do not apply to forms of life. What I am calling into question is a certain essentialism you seem to be smuggling into your understanding of a form of life as a ‘bottom’ the game stands on. Can we simplybassume that a form of life which makes intelligible a series of language games is the ‘sam’ form of life as we move from game to game? Or is there a certain essentialism implied by that assumption? Isnt this where grammar comes in? How do we know what the ‘same form of life’ means?

We don’t “assume” sameness in the sense of an underlying essence; we recognize continuity through overlapping practices, and grammar is what makes that recognition possible.
Witt is not positing a stable substrate, like a human nature or invariant structure, that remains identical as we move from one game to another.

Instead, the unity of a “form of life” is what he elsewhere calls family resemblance: overlapping patterns of activity, agreement in judgments, shared reactions, ways of training and correcting. There is no single feature that all language-games share, but there are enough interconnections that we treat them as belonging together.

Grammar is what articulates the connections across practices. It tells us what counts as the same, what counts as a continuation, what counts as a shift.

For example, what counts as the same object across contexts? What counts as the same person? What counts as continuing correctly from one situation to another? These are not settled by some pre-given form of life. They are stabilized by grammatical practices—by how we actually go on using words like “same,” “object,” “person,” etc.

So grammar doesn’t rest on an already unified form of life; rather, it is part of what holds a form of life together as something we can treat as unified at all.

197

If in the beginning is the deed then an expectation is not there from the beginning. The infant has no expectation that what it reaches for will persist. It sees something and tries to grasp it. The reaching is instinctual. (OC 475)

A language-game does not “rest on” acting. Acting is part of a language-game.

PI 23: “Here the term “language-game” is meant to bring into prominence the fact that the speaking of language is part of an activity, or of a form of life.”

198

Let’s see if I can address some of your concerns.

On the infant, you say what makes the practice possible is a capacity for being trained into patterns of activity, not a pre-existing certainty. But this doesn’t hold up empirically. Infants as young as a few months show surprise when objects seem to vanish behind screens. This has been documented extensively since Piaget’s original work and refined by other researchers. Nobody trained the infant to expect the object to still be there. No caregiver reinforced that expectation. The infant brings it to the interaction. Your story about caregivers responding, correcting, and stabilizing patterns is true for many things, but object persistence isn’t one of them. And this matters because if the infant already acts as though objects persist before any training has occurred, then the certainty is a condition of the training, not a product of it. The capacity to be trained presupposes a world in which training makes sense, and that world already includes persisting objects, other beings, and continuity of experience.

On essentialism, I understand the concern but I think it’s misplaced. I’m not positing a fixed human nature or an invariant substrate beneath our practices. I’m describing conditions of engagement that show themselves in acting. Object persistence isn’t a doctrine about what humans are. It’s how every creature that engages with objects behaves. The dog, the infant, the adult all act as though objects persist, not because they share an essence but because engaging with objects just is treating them as persistent. You can’t engage with an object while doubting its persistence. The engagement and the certainty are the same thing. Calling this a condition of engagement isn’t essentialism. It’s a description of what engagement with the world looks like at its most basic level.

On grammar holding the form of life together, I think you’ve stated something true but drawn the wrong conclusion. You say grammar articulates the connections across practices, that it tells us what counts as the same, what counts as a continuation, what counts as a shift. I agree. But for grammar to do this work, sameness and continuation have to already be operative in the world the grammatical practices are about. The child reaches for the toy again. The dog returns to the bowl. These aren’t grammatical achievements. Nobody has articulated what counts as “the same toy” or “the same bowl” for the dog. The creature is already tracking persistence and sameness before grammar enters the picture. Grammar can articulate what’s already operative. It can’t create what it articulates.

You say grammar doesn’t rest on a form of life but is part of what holds a form of life together. But then what holds grammar together? You need shared reactions, agreement in judgments, common responses. Witt says this himself at PI 242. And where do those shared reactions come from? They come from beings who already engage with a shared world. The agreement in judgments that grammar presupposes isn’t itself a grammatical achievement. It’s the ground on which grammar stands. This is exactly what OC 204 is saying. Giving grounds comes to an end, and what’s at the end isn’t more grammar, more norms, more articulation. It’s acting. The acting comes first. The grammar comes after.

199

I want to step back and address something about where this thread keeps going.

I said at the beginning that I’d be working in three registers: what the text directly supports, what it implies but doesn’t fully develop, and where the distinctions lead when taken further than Witt himself goes. That still holds. The sequential reading is the backbone, and I take the exegesis seriously. But the project was never solely about what Witt meant.

OC is unfinished. Witt was working things out, testing formulations, circling back, trying again. He didn’t deliver conclusions. He left notes. Any reading of those notes involves interpretation, and honest interpreters will disagree. That’s fine. But I’ve noticed a pattern in this thread where substantive philosophical distinctions get pulled back into textual disputes that can’t be resolved. Someone makes a point about the text, I respond with my reading, they respond with theirs, and the conversation circles without advancing. Meanwhile the philosophical question, which is the point, sits untouched.

So let me be direct about what I’m doing. When I distinguish three levels of hinges, or argue that experience is more fundamental than the propositional framework it supports, or draw out the implications of OC 204 for what lies at the bottom of the language game, I’m not claiming that Witt explicitly said these things. I’m claiming that the text points in these directions, and that following those directions produces distinctions that do real philosophical work. The question I’m interested in is not whether we all agree on what Witt meant by “acting” in OC 204. The question is whether the distinction between normatively structured activity and activity that precedes normative structure illuminates something about the foundations of knowledge. I think it does, and I’ve given reasons for thinking so.

If you disagree with my reading of a particular passage, that’s fine, but to keep going back over the same disagreements is a waste of time. If the disagreement is purely textual, if it comes down to “Witt meant X not Y” with no further philosophical consequence, then I’d rather acknowledge the interpretive difference and move on to the question that matters, i.e., does the distinction work? Does it solve a problem? Does it illuminate something the text leaves in shadow? Those are the questions that can actually be settled by argument, unlike questions about what an unfinished text really meant.

The exegesis serves the philosophy. Not the other way around.

200

The solitary development of perceptual
achievements in the individual does not belong to what Wittgenstein is concerned with in PI and OC. If you want to say that such achievements make such things as forms of life and hinges possible, that is one thing. But then these can’t themselves be called hinges or forms of life. Furthermore, if we examine how the development of object permanence is understood by Piaget, we don’t find actions which simply conform to supposed facts of the physical world.

In works like The Construction of Reality in the Child, object permanence is a gradually constructed invariant that emerges through sensorimotor schemes—grasping, tracking, hiding/recovering, coordinating perception and action.
object permanence is approached asymptotically through constructive activity. Early cognition treats objects as tied to action (what is grasped, seen, interacted with “here and now”).

Progressively, the child builds stable coordination schemes that allow objects to be treated as persisting across occlusion, delay, and movement This stability becomes more robust over time, never as a single “moment of attainment,” but as a progressive equilibration. Object permanence is never just flipped on; it becomes increasingly de-centered, stable, and reversible in thought, there is no final “click” where the object suddenly becomes fully constituted forever. It is an ongoing constructive stabilization of invariance across action schemes.

It is not as if the child were approximating a representation of a mind-independent object. For Piaget, what is being constructed is not a belief about objects but a sensorimotor structure of objectivity itself. So the “asymptote” is not toward a representation matching reality, but toward a stable coordination between action and environmental invariants. Put differently, the subject progressively constructs the object in increasing adequate ways. As a result, the ‘certainty’ accompanying any particular developmental stage in the interpretive grasp of the object via sensorimotor and later cognitive coordinative activity is merely provisional.

Even though Piaget doesn’t treat object permanence as a pre-given certainty but as a progressively stabilized achievement that never becomes a simple propositional possession, he retains a quasi-Hegelian teleological presumption about how the world acts to constrain our range of actions with respect to it, such that over time our construction of the object becomes progressively more equilibrated. Consistent with his treatment of perceptual stages, Piaget would say that forms of life are schemes of thought which evolve in the direction of subsuming inclusivity.

I happen to think Wittgenstein’s account of language games is more consistent with Husserl and Merleau-Ponty’s treatment of perceptual
development than with Piaget’s teleological model, but I also think Piaget comes closer to Wittgenstein than you do with regard to the constructed basis of action in the achievement of object permanence and other ‘pre-hinge’ certainties.

201

A Note on the Framework

The word “framework” appears throughout this thread and it’s one of the most misunderstood ideas in the entire discussion. The most common misunderstanding is to think of the framework as a collection of propositions, a set of basic beliefs that form the starting point of inquiry. On this picture the framework looks like traditional foundationalism, viz., self-evident propositions at the bottom of a chain of justification, each one inside the space of reasons, supporting everything built on top of it. The basic beliefs are examined, found to be warranted, and the rest of knowledge is derived from them.

Witt rejects this picture, but what he rejects is the traditional version, not the idea that there is a foundation. There is a foundation. It’s just that the foundation is acting, not propositions, and it sits outside the space of justification rather than inside it. The ungrounded way of acting from OC 110, the prelinguistic certainties shown in bodily engagement with the world, the shared epistemic routes through which inquiry operates, these are genuinely foundational. Everything else rests on them. But they can’t be stated as a set of propositions and examined from a neutral standpoint, because they are the condition under which stating, examining, and evaluating are possible. The foundation is real. What’s wrong is the traditional picture of what a foundation looks like.

Consider what the framework actually includes.

At the deepest level are the prelinguistic certainties. The continuity of experience. The persistence of objects. The existence of other beings. These are not propositions anyone has formulated and accepted. They are beliefs shown in action, in the infant reaching for the toy, in the dog going to its bowl, in the body navigating space. They were in place before language, before propositions, before anything resembling a chain of reasoning could get started. No one decided to accept them. No one weighed evidence for them. To call them “starting propositions” is to mistake a description of the certainty, which we produce when we do philosophy, for the certainty itself, which was operative long before anyone did philosophy. These certainties are foundational in the fullest sense. They are what makes inquiry, language, and thought possible. But they are not foundational in the traditional sense, because they are not propositions inside the space of justification awaiting examination. They are the ground on which the space of justification is built.

Alongside these, in any being who has language, are the nonlinguistic certainties. The certainty that my body will respond when I stand up, that the ground will hold, that my hands are at the ends of my arms. These are the same kind of certainty as the prelinguistic ones, bodily, shown in action, not propositional. But they operate in a being who also has language. When I get up from a chair without checking my feet, that certainty is nonlinguistic even though I could put it into words if pressed. The certainty isn’t in the words. It’s in the getting up.

Then there are the linguistic hinges. The earth has existed for a long time. Physical objects persist. My name is such and such. These have propositional form and look like empirical claims, but they don’t function as empirical claims. No one investigates whether the earth has existed recently. These propositions stand fast within our language-games, not because we’ve verified them but because they’re part of the inherited background within which verification proceeds. They are held fast by what lies around them in the system, as OC 144 puts it. Their certainty is relational, drawn from their position in a network of mutual support. They are part of the foundation, but they function differently from the prelinguistic and nonlinguistic certainties. They can shift. The riverbed moves. What was once a linguistic hinge can become an ordinary empirical claim, and what was fluid can harden. The prelinguistic certainties don’t shift in this way. They are the hard rock beneath the riverbed.

And the framework includes the basic epistemic routes: logic, sensory experience, testimony, and linguistic training. These are the routes through which we engage with reality, gather evidence, evaluate claims, and conduct inquiry. They are shared across frameworks, which is why argument has traction even between people who hold different world-pictures. They belong to the foundation alongside the prelinguistic certainties, not as propositions about how inquiry should proceed but as the capacities and practices through which inquiry does proceed.

The framework, then, is not a list of these items. It’s the whole thing in operation: the acting, the certainties shown in the acting, the practices built on those certainties, and the epistemic routes through which those practices function, all working together as the foundation of inquiry. It’s a foundation you can’t write down and step back to examine, because writing it down and examining it are activities that rest on it.

The deepest source of confusion is treating the products of the framework as though they were components of the framework. The framework makes inquiry possible. Inquiry produces results, things like scientific findings, mathematical proofs, historical discoveries, everyday knowledge claims. These results live inside the space of justification. They can be checked, challenged, revised, and overturned. People look at propositions like “the earth has existed for a long time” and see something that looks like a scientific finding, and they conclude that science built the framework and could in principle revise it entirely. But science operates within the space the framework holds open. The child learning science already trusts that objects persist, that other people exist, that sensory experience connects us to the world, that testimony from teachers can be relied upon. None of that was established by science. All of it was in place before science arrived. Science can reshape linguistic hinges over time, hardening some and making others fluid, as OC 96 describes. But the prelinguistic certainties, the nonlinguistic certainties, and the shared epistemic routes were not produced by science and cannot be revised by science, because every act of scientific investigation already depends on them.

The same confusion generates the skeptic’s challenge. The skeptic asks, how do you know the framework is correct? What justifies your starting point? The question feels powerful because within the framework every claim requires justification, and the skeptic is simply extending that requirement to the framework itself. But the extension is illegitimate, and it’s illegitimate for a specific reason. Traditional foundationalism invited the skeptic’s question because it placed the foundation inside the space of justification, as a set of propositions that were supposed to be self-evidently warranted. Once the foundation is propositional, asking for its justification is grammatically coherent even if unanswerable. But Witt’s foundation is not propositional. It’s acting. And acting is not the kind of thing that can be justified or unjustified. It’s the condition under which justification operates. Asking for the justification of the acting is asking what holds the ground up. The ground is what holding things up looks like. The framework is what justification looks like. There is no outside from which to apply the demand. The framework is, as OC 105 says, the element in which arguments have their life.

This also explains why relativism doesn’t follow from Wittgenstein’s account, even though it can look as if it does. If the framework isn’t justified by evidence, it might seem as though any framework is as good as any other. But this is wrong for three reasons. First, the prelinguistic certainties are framework-invariant. The continuity of experience, the persistence of objects, the existence of other beings are not features of one framework among many. They’re the conditions of there being a framework at all. Any world-picture that violated them wouldn’t be a world-picture. Second, the shared epistemic routes give argument real traction across frameworks. Logic, sensory experience, and testimony can expose where the facts don’t align within a framework, where the internal logic strains, where the practices fail. Argument reaches its limit only when the disagreement is traced all the way down to divergent linguistic hinges that lie outside the space of justification. But that limit comes at the end of a long process of rational engagement, not at the beginning. Third, frameworks engage with reality through practice, and some engage more faithfully than others. The rational criteria Witt points to in OC 92 and OC 147, simplicity, symmetry, a picture proving itself everywhere, are not proof, but they are not arbitrary either. Over time, through sustained experience, the accumulated pressure of how well a framework works exerts real force. The world pushes back.

The framework is not a set of propositions that need justifying, and it is not the absence of a foundation. It is a foundation of a different kind from what traditional epistemology expected, not self-evident propositions inside the space of reasons, but the acting, the certainties shown in the acting, the practices and epistemic routes through which inquiry proceeds. It is genuinely foundational. Everything rests on it. But it is foundational in the way the ground is foundational, something you stand on and act from, not something you examine and certify before you’re willing to stand.

202

Even though you say the foundation is “acting,” the structure of your account still preserves a classic asymmetry; action goes out, and then something independent, “the world,” “reality,” “how things are”, pushes back and constrains it. So you get a two-level model.

On one side, embodied or pre-propositional activity; on the other, a fully formed world that has determinate facts which those activities must accommodate. The actions are then shaped, corrected, or stabilized by that independent order.

The issue for me is that this reintroduces, through the back door, that there is a fully determinate factual order which is simply “there anyway,” and which our practices must eventually answer to as something independent and fixed.

On my reading of Witt, it is not that there is first a determinate world and then action that may or may not fit it. Rather, what counts as “a determinate world with constraints” is itself partly an achievement of the very practices in question. The sense of resistance, correction, failure, and “the world saying no” is internal to a form of life, not something delivered from outside it.

Your picture is that action meets independent reality, and reality constrains action. Mine is that
coordinated activity produces a field of stability, resistance, breakdown, and correction. Within within that field, “world constraining action” is one of the internal grammars that emerges. The feeling that “the world pushes back” is not evidence of an external metaphysical layer but a structured feature of practice: certain sequences of action become self-disrupting, fail to continue, or cannot be integrated into ongoing coordination. Those failures are then interpreted as world-resistance.

Constraint is not an external veto but a local stabilization condition inside a productive process. What looks like “the world imposing structure” is often the sedimentation of repeated successful coordinations, what persists is what can be repeated, not what independently exists and then forces repetition.

So the key difference is you preserve an ontological independence of “the world as constraint-bearing facticity, whereas I treat “constraint” as an emergent relational feature of action-systems themselves.
You’re pulling Wittgenstein in a realist direction, and I’m pulling him away from that reading. I admit that there is textual evidence for either approach.

203

Josh, I think you’re conceding the substantive point while framing it as an objection.

You say if perceptual achievements like object permanence make forms of life and hinges possible, that is one thing, but then they can’t themselves be called hinges or forms of life. Fine. Call them pre-hinges, call them conditions of engagement, call them whatever you prefer. The label doesn’t matter. What matters is that something is operative before the normatively structured practices arrive, and that something is what makes the practices possible. You’ve granted that. The disagreement is now about terminology, not substance.

On Piaget, you’re right that object permanence develops gradually through sensorimotor activity. I don’t dispute the developmental story. But you draw the wrong conclusion from it. The fact that the infant’s coordination with objects develops progressively doesn’t mean persistence itself is constructed from nothing. The infant’s sensorimotor schemes are already engagements with a world in which objects persist. The grasping, tracking, and reaching presuppose a world in which there are things to grasp, track, and reach for. Piaget’s own account depends on this. You can’t have sensorimotor schemes operating on nothing. Even at the most primitive stage, the infant is engaging with a world that has structure. What develops is the infant’s coordination with that structure, not the structure itself.

You say the certainty accompanying any particular developmental stage is merely provisional. But provisional in what sense? The infant’s tracking becomes more stable and more coordinated over time. That’s development. But the infant never goes through a stage of engaging with a world where objects don’t persist. There’s no stage where persistence is absent and then gets constructed. There’s a stage where the infant’s grasp of persistence is less coordinated and a later stage where it’s more coordinated. The persistence is there throughout. What changes is the infant’s ability to track it across occlusion, delay, and movement.

You say Piaget doesn’t treat object permanence as a pre-given certainty but as a progressively stabilized achievement. But stabilization presupposes something being stabilized. If there were nothing there to stabilize, nothing the sensorimotor schemes were engaging with, there would be nothing for the progressive coordination to be coordination with. Piaget himself recognizes this when he talks about environmental invariants that constrain the range of actions. Those invariants aren’t constructed by the child. They’re what the child’s constructions are responsive to.

Your reference to Husserl and Merleau-Ponty is interesting, but I think it supports my position more than yours. Merleau-Ponty’s account of the body-subject in the Phenomenology of Perception is precisely an account of pre-reflective, pre-linguistic engagement with the world that grounds later cognitive and linguistic achievement. The body is already in the world, already oriented, already engaged, before any normatively structured practice begins. That’s what I’ve been calling non-linguistic hinges of bodily certainty. Merleau-Ponty would not say the body’s engagement with the world is a grammatical achievement. He’d say grammar comes after, as an articulation of an engagement that was already fully operative before articulation began. If you think Witt is closer to Merleau-Ponty than to Piaget on this, then you’re closer to my reading than you realize.

204

Schemes dont operate on nothing, but what they operate on cannot be separated from the contributions we make to how what appears does so for us. The world neither simply imposes itself on us in raw , uninterpreted fashion, nor is it merely conjured up in our minds. Sensorimotor structures are an alchemy produced by the meeting between subject and world.We can’t say anything about the supposed structure the world has ‘in itself’.

The structures Piaget is describing are schemes of coordination between a subject and the world. That is the only structure that matters for perceptual development. We could take a cue from Husserl here, who describes the perceptual constituting of persisting spatial objects as a kind of idealizing abstraction. In actuality there is never the perception of a self-identically persisting thing, but instead relatively consistent patterns produced by the response of an aspect of the world to our actions on it. We idealize this more or less consistent flow of experience into the abstraction we call persisting objects, construct empirical scientific mathematical models out of our abstractive activities, and then treat these derivative abstractions as if they were the grounds for our perceptions. One might consider taking to heart Merleau-Ponty’s(1968) critique of the idea of the object-in-itself.

…the identity of the thing with itself, that sort of established position of its own, of rest in itself, that plenitude and that positivity that we have recognized in it already exceed the experience, are already a second interpretation of the experience…we arrive at the thing-object, at the In Itself, at the thing identical with itself, only by imposing upon experience an abstract dilemma which experience ignores

205

The questions you’re raising about the contributions of subject and world to perceptual experience, about whether we can separate what the world gives from what we bring to it, these are genuine philosophical questions. But they aren’t questions OC is asking. As I see it Witt would say the debate about the object-in-itself is a confusion generated by a philosophical picture, and he’d want to dissolve it rather than take sides. My interest is in following where OC leads, and OC leads to the inherited background, the world-picture, the riverbed, and the relationship between what stands fast and what flows.

As I’ve said in earlier posts, many of these interpretive questions can’t be resolved. The text is unfinished, Witt was working things out rather than delivering conclusions, and honest interpreters will read it differently. That debate has no final answer, and continuing to push on it takes us in circles. What I’m trying to do, both in this thread and in my next book, is something different from what the secondary literature has done. No existing book gives a sequential remark-by-remark reading of OC. Every study I know of pulls remarks from across the text to support a thematic argument. My aim is to start at OC 1 and read forward, following the movement of Witt’s thought as it develops, letting the text speak for itself as much as possible. Where the text opens up questions it doesn’t fully answer, I’ll follow those questions where they lead and say so when I do. But the backbone is the sequential reading, and that’s where I want to keep the focus.

206

It’s true that Wittgenstein in On Certainty doesn’t explicitly stage the question in the familiar Kantian form “what is contributed by the subject vs. what comes from the object-in-itself?”, and cautions against that framing. But it doesn’t follow that the issue simply disappears. Rather, it gets displaced and reconfigured.

The very notions Wittgenstein works with, “what stands fast,” “the riverbed,” “world-picture,” “agreement in judgments,” “forms of life”, are doing work that touches exactly that boundary, even if they refuse to describe it in representational or metaphysical terms.

You can see this if you look at what “standing fast” actually does. When Wittgenstein says certain things are exempt from doubt, he is not just making a grammatical point about how we use words; he is also marking out a limit-condition for intelligibility. And the moment you start talking about limit-conditions for intelligibility, you are implicitly in the territory of: what must be in place for there to be a world for us at all? That is already very close to the terrain of “contribution,” even if the vocabulary is different.

You’re right that Wittgenstein would reject the picture of two separable inputs; raw world-data plus subjective structuring. But my question is, even if that picture is rejected, isn’t Wittgenstein still negotiating the relation that picture was trying l to capture?

I think Wittgenstein transforms the problem from ‘what part comes from the world vs what part comes from the subject’ to ‘what must be taken for granted in our practices for “world,” “object,” “fact,” and “experience” to have sense at all’.

The “world” is no longer a thing-in-itself standing over against a subject; it shows up as what resists, stabilizes, and is presupposed in practice. And the “subject” is no longer a constituting mind; it shows up as participation in shared activities, training, and responses.
So the question of contribution becomes implicit in a different form. The question becomes: what in these practices is variable (the “river”)?
What is non-negotiable (the “riverbed”)? How does the shifting relate to the fixed?

This is why my earlier concern about “productivity” versus “constraint” matters. Reading Wittgenstein the way you do, the riverbed starts to look like something given that constrains us, almost like a softened version of “the world as it is anyway.” If we
push in my direction, the riverbed is not simply given; it is what has sedimented out of practice, what has become so entrenched that it functions as necessity.

If the riverbed is treated as independent constraint, then you’re right: the subject/world contribution debate is a confusion to be dissolved. But if the riverbed itself is an achievement of practice, then the old question reappears in a transformed way: not “who contributes what,” but “how do practices generate something that appears as an independent world that constrains them?”

207

Josh, you’re raising interesting questions and I appreciate your responses. I think you’re right that Witt transforms the subject/world question rather than simply dissolving it, and I think the riverbed metaphor is one of the places where that transformation is seen in the text. Where we disagree is on whether the deepest level of the riverbed sedimented out of practice or was already in place before practice began. I’ve given my reasons for thinking it’s the latter, and you’ve given yours for thinking otherwise. I don’t think we’re going to resolve that here, and I’d rather let the disagreement stand and get back to the text. OC 95-98 is where Witt develops the riverbed metaphor directly, and I’d rather let him speak for himself than continue debating what he would have said.

I’ll be starting a new thread on this subject.

209

That is a good description of what is critical in Wittgenstein and absent in Kant. But I don’t think it is a matter of one set of problems reemerging in a new place. Kant’s question about what makes experience possible is still there. Wittgenstein separates his investigations from it. Key to the divergence is the distinction between the logical and the psychological. “Participation” is described in following way in Philosophical Investigations:

  1. “So you are saying that human agreement decides what is
    true and what is false?”—It is what human beings say that is true and false; and they agree in the language they use. That is not agreement in opinions but in form of life.

  2. If language is to be a means of communication there must
    be agreement not only in definitions but also (queer as this may sound) in judgments. This seems to abolish logic, but does not do so.— It is one thing to describe methods of measurement, and another to obtain and state results of measurement. But what we call “measuring” is partly determined by a certain constancy in results of measurement.

The sharpest departure from Kant is the role of logic. On Certainty is consistent in how logic works in language games:

  1. If one says, “Perhaps this planet does not exist, and this light phenomenon comes about some other way”, one nonetheless has // needs // an example of an object that does exist. This does not exist, as, for example, … does.
    Or should one say that sureness is just a constructed point which some things approximate more, others less? No. Bit by bit, doubt loses its sense. That’s just how this language‐game is.
    And everything that describes a language‐game belongs to logic.

I do take your point that there is:

The role of logic does not confer a recognition of that condition, at least in terms of “knowing” in On Certainty:

  1. If “I know {that this is my hand}” is construed as a grammatical proposition, then obviously the ‘I’ can’t be important. And it really means “In this case there is no doubt” or “The phrase ‘I don’t know’ makes no sense in this case.” And, of course, it follows from this that “I know” makes none either.

  2. Here “I know” is a logical insight. Only realism can’t be proved by it.

Wittgenstein acknowledges the “territory” you speak of and the limits of his method regarding it in Philosophical Investigations, Pt 2, XII:

Xll
If the formation of concepts can be explained by facts of nature, should we not be interested, not in grammar, but rather in that in nature which is the basis of grammar?—Our interest certainly includes the correspondence between concepts and very general facts of nature.
(Such facts as mostly do not strike us because of their generality.) But our interest does not fall back upon these possible causes of the formation of concepts; we are not doing natural science; nor yet natural history—since we can also invent fictitious natural history for
our purposes.

I am not saying: if such-and-such facts of nature were different people would have different concepts (in the sense of a hypothesis).
But: if anyone believes that certain concepts are absolutely the correct ones, and that having different ones would mean not realizing something that we realize—then let him imagine certain very general facts of nature to be different from what we are used to, and the formation of concepts different from the usual ones will become intelligible to him.

Compare a concept with a style of painting. For is even our style of painting arbitrary? Can we choose one at pleasure? (The Egyptian, for instance.) Is it a mere question of pleasing and ugly?

By seeking the natural ground making language possible, Sam26 is pursuing a theory of psychology.