Frameworks, Hinges, and What Lies Beyond

Metaphysics & Epistemology
1

This thread grows out of a question that kept surfacing in the On Certainty reading thread but pulling the sequential discussion off track. The question is big enough to need its own space, so I’m giving it one.

The Phenomenon

Start here. You’re sitting in a chair right now. You didn’t test it first. You didn’t check whether it would hold you, examine its joints, calculate its load tolerance, or reason from past experience to the conclusion that this particular chair on this particular occasion would bear your weight. You sat down. Before that, you walked into the room. You didn’t establish that the ground would continue to hold you with each step. You reached for your coffee. You didn’t first work out whether the cup would still be there when your hand arrived, or whether your hand would do what you intended. You started reading this post. You didn’t pause to verify that the words would continue to mean what they meant a moment ago, or that the person who wrote them exists.

None of this is carelessness. It’s what being a functioning creature looks like. We act, and our acting is saturated with certainty, but not the kind philosophy usually means by that word. This isn’t knowledge. It isn’t assumption (you didn’t adopt any of this). It isn’t epistemic certainty, the kind established by investigation and grounded in evidence. It’s the certainty that lives in the acting itself, in the sitting, the reaching, the trusting. Wittgenstein spent the last eighteen months of his life circling this phenomenon, and it turns out to be remarkably hard to talk about, not because it’s obscure but because the word “certainty” does at least four different things in this neighborhood and using it without qualification invites confusion at every turn.

The Language Problem

Here’s the first difficulty, and it’s not a failure to find the right words. It’s a substantive philosophical point.

Most of our philosophical vocabulary (“knowledge,” “assumption,” “justification,” “proposition”) was built for work inside a framework. These are terms for things we do once we’re already up and running: forming beliefs, testing claims, weighing evidence, drawing conclusions. But the certainty that shows itself in your sitting without testing isn’t inside the framework. It’s what the framework rests on. And our language doesn’t have clean terms for this, because the terms were designed for the epistemic work that happens after the certainty is already in place. We’re trying to describe, from inside, what “inside” rests on. The terminological difficulty is a symptom of the problem, not a failure to solve it.

Wittgenstein felt this himself. At OC 76 he says his aim is “to give statements one would like to make here but cannot make significantly.” Not won’t make. Cannot. The available forms of expression resist what he’s trying to say.

A second language warning. The secondary literature on OC introduced the term “hinge proposition.” Witt never used it. The image comes from OC 341–343, where he compares certain things to the hinges on which a door turns, viz., the door moves, the hinges stay put. That image is useful. But “hinge proposition” smuggles in exactly what Witt is working against. It takes the certainty your sitting-without-testing expresses, certainty shown in acting, and repackages it as a proposition, a sentence with a truth value, the kind of thing you can believe or doubt or justify. And then the whole discussion slides back into the epistemological machinery Witt was trying to get beneath.

This thread will use “hinge” on its own. No “proposition” appended. When the discussion requires a term for the formulated, linguistic versions, I’ll say so explicitly.

Three Levels

Look again at what’s going on when you sit without testing. There are layers here, and they aren’t all the same kind of thing.

At the deepest level, objects persist, other beings exist, experience continues. You don’t hold these as positions. You couldn’t function without them. A creature that genuinely doubted whether objects continued to exist when unobserved wouldn’t be a creature with an interesting philosophical view. It would be a creature that couldn’t act at all. These aren’t commitments you’ve undertaken. They’re conditions of being a creature that acts in a world.

At a second level, the ground holds, your body responds to your intentions, space and time have the structure they seem to have. These are still prelinguistic (a dog operates with all of them) but more specific than bare object-persistence. They can, in principle, fail. The ground could give way. Your body could stop responding. But they don’t function as hypotheses you’re betting on. They function as the background against which hypotheses become possible.

At a third level, formulations. “The earth has existed for a very long time.” “Physical objects continue to exist when unperceived.” “My name is Sam.” “I have never been to the moon.” These are the ones that look like propositions, and this is where the confusion starts. They can be stated, debated, doubted. But Witt argues they play a peculiar logical role (OC 136). They don’t function the way ordinary empirical claims function. They’re more like the scaffolding that holds the structure in place while work goes on inside it, and some of this scaffolding can shift over time. What counts as unshakeable in one era can become revisable in another, and vice versa (OC 96–97, the riverbed metaphor, where what was bed becomes channel and what was channel becomes bed).

Here’s the crucial point - when I describe a hinge propositionally (“objects persist,” “the ground holds”) I haven’t shown that the hinge is a proposition. The description comes after the fact. It’s a map of something that isn’t itself a map. The dog sits on the chair without testing it too, and the dog has no propositions.

All three levels are forms of hinge certainty, the certainty that stands fast, shown in what we do. But they differ in important ways. The deepest hinges are prelinguistic, operative before language enters the picture at all. The second level continues to operate nonlinguistically even in creatures who have language. The third level is linguistic, held fast by what lies around it in the system, and it can shift over time in ways the deeper levels cannot.

Moore and the King

The passage that crystallizes the framework problem is OC 92:

“However, we can ask: May someone have telling grounds for believing that the earth has only existed for a short time, say since his own birth? — Suppose he had always been told that, — would he have any good reason to doubt it? Men have believed that they could make rain; why should not a king be brought up in the belief that the world began with him? And if Moore and this king were to meet and discuss, could Moore really prove his belief to be the right one? I do not say that Moore could not convert the king to his view, but it would be a conversion of a special kind; the king would be brought to look at the world in a different way.”

What Wittgenstein is describing here is a limit. Not a limit of Moore’s cleverness, or of the evidence available to him, but a structural limit on what argument can accomplish when two people don’t share a framework. Moore can present his evidence. But his evidence counts as evidence because of standards that are themselves part of his framework. The king has different standards, not because the king is irrational, but because standards don’t float free of the framework they belong to.

Moore cannot prove his view to the king. He might convert him. But conversion is not proof. Proof operates within a shared framework. Conversion is a shift from one framework to another. And Wittgenstein is remarkably honest about what this means, i.e., at the level where frameworks differ, argument runs out.

A Convergence

Here’s what makes this more than a problem for Wittgenstein scholars.

Kuhn, working independently on the history of science, arrived at something structurally parallel. Scientific paradigms are not simply collections of theories. They include the standards by which theories are evaluated, the exemplars that define what a good explanation looks like, even the ontology that determines what exists to be explained. When paradigms conflict, the disagreement can’t be resolved by appeal to the evidence, because what counts as evidence is partly determined by the paradigm. This is the incommensurability problem. It’s OC 92 in a lab coat.

Gödel, working in formal logic, showed that any sufficiently powerful formal system contains truths it cannot prove from within. The system cannot ground itself entirely using its own resources. No amount of internal rigor closes the gap.

Three thinkers, three domains, one structural insight: no rule-governed system can fully ground itself from within. The framework’s resources are insufficient to justify the framework.

The Question

So, the framework has limits. This much seems hard to deny once you see it.

But what follows from this? That’s where the real work begins, and it’s what this thread is for.

Witt describes the structure. But he also, I think, points toward the answer, even if he doesn’t follow it all the way. This is where I’ll be developing a reading that goes beyond what Wittg straightforwardly says.

OC 204: “Giving grounds … 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.”

Acting. Not reasoning, not judging, not arriving at conclusions. Acting. The deepest ground of the framework is experiential, not propositional. The creature acts, and its acting is the certainty. The believing and the doing are one thing at this level, and both are prior to the framework that describes them.

If the deepest hinges are experiential rather than propositional, and if the framework’s limits are structural, then the question becomes: is the limit absolute? Is there nothing beyond the framework but more framework? Or can experience cross boundaries that argument cannot?

This thread will explore that question. We’ll look at different kinds of frameworks (scientific, religious, philosophical, political) and ask what holds them up, where they run into their own limits, and whether there’s anything honest to say about what lies beyond them.

The OC reading thread will continue its sequential work. Framework questions that arise there can be brought here. And the argument I’ll be developing, that experience is more fundamental than any framework and that it can, under the right conditions, extend the framework from within, will get tested in open discussion.

2

The difficulty at OC 76 is about having a false belief that is not simply a mistake. A mistake can be corrected. (72) It is a mental disturbance (71). The person who falsely believes there is a table there is quite certain that it is there.

What he says is:

… some propositions are exempt from doubt, are as it were like hinges on which those turn. (341)

No one is smuggling in the term ‘proposition’, you are intentionally leaving it out in order to make it fit your claim about prelinguistic hinges.

You fault others for continuing to return to this when in fact it is you doing this.

They do not just look like propositions, they are propositions. Ones that:

… have a peculiar logical role in the system of our empirical propositions.

Again, you attempt to leave propositions out.

The king does not believe the world began with him based on experience. Catholics do not believe in transubstantiation based on experience.

3

After the born-again Christians, we now have the born-again Certaintists. :smiley:

Your entire, exhaustingly long post can be summarized in a single word: learning .

No, the chair and the floor were never eternally certain. The very first time—when you were just a few months old—that you explored the floor with your rubbery little feet or tried to climb a chair with your rubbery body, you were filled with proprioceptive uncertainty and hesitation. It was only after a while of physical trial and error that your mother came along and said, “I am certain you can walk by yourself now.” Epistemologically speaking, there is absolutely nothing more to say about “certainty.”

  • If you genuinely want a robust, working psycho-logical model of how “certainty” is actually constructed—how internal logical structures confront external novelty through physical action (I will grant you that specific point)—simply go read Jean Piaget. And never, ever look back at these small-time thinkers (the ‘W’ guys). :smiley:

  • If you actually want a metaphysical framework that rigorously articulates and dissolves every single aporia your thread is agonizing over: look into the MCogito system. Its precise concepts—meta-causality, the differential speed of Being output, categorical codes, the Totality, and recursive embodiment—systematically answer every single one of these pseudo-mysteries.

Thread closed. (Again).

4

I’m looking forward to your exploration of the frameworks question. I agree with you that these questions are much broader than mere interpretations of Witt.

One preliminary question: When you say,

isn’t it the case that it’s precisely because we couldn’t function without them that we do hold them as positions? Granted, we rarely have to state them as positions, and even more rarely try to defend them. But if I were to be asked, Is it the case that objects persist and other beings exist?, I would answer Yes, this is my position, based on what I believe to be rational grounds. So, I’m not clear why a hinge based in experience can’t also be held as a propositional position. Is it because of the peculiar difficulty involved in trying to justify such a position? Now if the position in question concerned, say, a feature of logical argumentation, such as the LONC, I could see the problem of trying to defend it while simultaneously using it. As you say, we’d be “trying to describe, from inside, what ‘inside’ rests on.” But I don’t see that contradiction being raised in a position such as “Objects persist.”

5

Jay, I think the issue turns on the difference between expressing something as a proposition and treating it as a position within inquiry.

I agree that something like “objects persist” can be stated propositionally. I also agree that, if challenged, I may offer what look like rational grounds for it. But it doesn’t follow that the proposition functions as an ordinary position whose standing is established by evidence in the usual way. That is what I want to resist.

My view, which goes beyond strict exegesis of Witt, is that the background is prior not only to justification but prior to language, reflective thought, and articulated positions altogether. It comes first. It’s already there as the enabling condition of any saying, doubting, inferring, checking, or defending. What we later express in propositions is, in part, an articulation of something more primitive than propositions.

So, I would distinguish three levels.

First, there is the most primitive background, prior to language and prior even to explicit thought. This is not yet a proposition, not yet a judgment, not yet a position. It is the condition for there being a coherent field of experience, continuity, responsiveness, and engagement at all.

Second, there is the lived, non-linguistic level, where this background shows itself in action, expectation, orientation, trust, and unreflective involvement with the world. At this level, what stands fast isn’t something one first says and then believes. It’s something already operative in how one moves through life.

Third, there is the linguistic level, where some of what stands fast can be formulated propositionally, for example “objects persist” or “other beings exist.” At this stage we can state such things, discuss them, and even defend them. But the proposition is secondary. It doesn’t create the certainty. It gives linguistic form to something already functioning at a deeper level.

That’s why I don’t think the key question is whether a hinge can be put into propositional form. Clearly it can. The question is what role it’s playing when expressed. If “objects persist” is functioning as a hinge, then it isn’t, in the first instance, an ordinary conclusion reached from evidence. Rather, it belongs to the background that makes epistemic reasoning possible in the first place.

This also explains why I don’t think the issue is formal contradiction, as it may be with something like the law of non-contradiction. In the case of “objects persist,” the difficulty is more basic. Any observational or rational grounds I might cite already presuppose a world of enduring objects, enduring observers, enduring memory, and enduring standards of comparison. So, the problem isn’t that I utter a contradiction in defending the proposition. The problem is that I’m trying to justify, from within, something that already belongs to the framework within which justification has its sense.

So yes, I grant that a hinge can be articulated as a proposition. I grant too that one may offer supporting considerations for it. But I deny that this makes it merely an ordinary position among others. The articulated proposition is a later expression of a background that is necessarily prior to language, thought, and argument. In that sense, what stands fast is deeper than the propositions in which we sometimes try to state it.

That, at least, is the direction I want to push the discussion. It is not only an interpretation of Witt, but an attempt to develop what I take to be one of the deepest implications of his remarks.

6

Every sign requires the possibility of functioning outside its ‘original’ context (form of life). In this sense, a statement such as ‘I am’—however unquestionable it may seem, and however integral it may appear to our way of life—is always shaped by the nature of the sign. ‘I am’ can function even if I am, in fact, dead. We see this particularly clearly in posthumous autobiographical works. This is our alienation from ‘our’ language. In reality, the phrase ‘our language’ must be used with caution. In this sense, we are always, as it were, written by another, even in our most deeply rooted propositions, which we take for granted as certainties. This means that there is always a doubt—perhaps a very unconscious one—as to whether our language can function and is appropriate. Certainty arises from the repetition of how our propositions function and how well they suit our way of life. But that functioning is never guaranteed (because the nature of sign)

I suppose that is why you have turned to experience as a safe space where you can set aside that doubt. However, according to an analysis of immediate experience, it is structured by traces and signs. Memory is always at work in personal experience; the remembering, for example, that my hand is there may be a sign of something that is no longer present (the experience of the hand being seen there). There are people who confuse dreams with memories. Not to mention people who cannot feel some of their limbs and need verification to know that their limbs are there and in the right place. This tells us something about the nature of perception or experience: it is not an absolute source of certainty insofar as its context (dream or reality, perception or non-perception) may not be clear. This is characteristic of every sign (here, the memory): it is not absolutely rooted in a specific context, even if that sign is the memory itself. The ‘certainty’ that my hand is still there is never absolutely guaranteed. Only repetition and habit give us confidence and a degree of certainty.

7

Thanks, this makes it much clearer. You’re using “justification” in a big-picture sense that includes the very world in which we find ourselves as part of the grounds for justifying anything, not merely one position over another:

Whether that world does indeed presuppose enduring objects et al. is another question, but it can wait. Carry on!

8

Jay argues that because we can state something like “objects persist,” it can be held as a position supported by rational grounds. As Sam responds, feom Wittgenstein’s perspective, that conflates two different things:a grammatical proposition that can be uttered, and a claim that functions as a hypothesis within inquiry.

The problem isn’t merely that they are hard to justify. It’s that they are part of what gives “justification” its sense in the first place. They’re conditions of the game, not moves within it.

You argue for three levels

1)pre-linguistic background
2)lived non-linguistic engagement
3)linguistic articulation

I think Wittgenstein would likely reject the first level entirely. For him there is no need to posit a “pre-linguistic condition of possibility”, because the “background” is already visible in what we do, say, and count as evidence. The bedrock is not beneath language, it is in the way language is used.

What about infants and object permanence? Piaget shows that infants gradually acquire expectations like objects continue to exist when hidden, and objects move continuously through space. Wittgenstein wouldn’t deny that this is pre-lingustic, but he would also say that describing this as “object permanence” already places it within our conceptual (grammatical) framework.

In other words, the infant does not have the concept “object persists”, we interpret its behavior through our concept of an object.

So for Witt there are pre-linguistic behaviors and expectations, but calling them “object permanence” is already a grammatical interpretation

Grammar applies “all the way down” in the sense that whenever we describe, interpret, or make sense of experience we are operating within grammar.

But grammar doesnt apply all the way done in the sense that. not all behavior or experience itself is linguistic or conceptual.

9

No, not quite. Merely being able to state something doesn’t mean it can be sensibly explained or defended. There is no “because” relationship involved. I was saying something more, or more particular: that in the case of “Objects persist” I can not only state the position but offer (not necessarily compelling) reasons for why I hold it. I can also answer truthfully that it is my position, as opposed to my necessary presupposition – or that that is at least how it seems to me.

To this, I think Sam’s and your counter is that the entire concept of what it means to justify something depends on presupposing certain conditions (which is true), and that among these is the condition “Objects persist.”

I think this is extremely debatable, but again, I don’t want to sidetrack Sam’s thread.

The important distinction here is that, of course, we can say all sorts of things in English that can’t be “held as a position supported by rational grounds,” but the statement in question is in my opinion not one of them.

10

Josh, this is a good objection.

You’re making two claims. First, that whenever we describe or interpret pre-linguistic behavior, we’re operating within grammar. Calling infant behavior “object permanence” is already a grammatical interpretation, ours, not the infant’s. Second, that Witt would therefore reject the first level entirely, because the bedrock is already in the way language is used, not beneath it.

The first claim is right. The post says so explicitly. When I describe a hinge propositionally (“objects persist,” “the ground holds”) I haven’t shown that the hinge is a proposition. The description comes after the fact. It’s a map of something that isn’t itself a map. The infant doesn’t have the concept “object persists.” We interpret its behavior through our concepts. No disagreement there.

But the second claim doesn’t follow from the first. The fact that we can only describe the pre-linguistic using language doesn’t mean there’s nothing pre-linguistic to describe. And you concede this yourself at the end of your post, viz., “not all behavior or experience itself is linguistic or conceptual.” That concession is the first level. You’ve granted its existence while arguing Witt would reject it.

Look at what OC 204 actually says. Giving grounds comes to an end, and the end is not certain propositions striking us immediately as true. It is our acting, which lies at the bottom of the language-game. The language-game has a bottom, and that bottom is acting, not more language. If the bedrock were already in language use, as you suggest, Witt wouldn’t need to say that acting, not propositions, is where grounds come to an end. He’d just say the bedrock is a way of using language. He doesn’t say that. He says acting.

The dog sharpens this. The dog sits on the chair without testing it. The dog goes to its bowl expecting food. The dog has no grammar. The certainty is real and operative. You can say, correctly, that our description of the dog’s certainty is grammatical. But the dog’s certainty doesn’t wait for our description in order to operate. The dog acts with hinge certainty at the pre-linguistic level, and the acting is prior to anything we say about it.

Now, your concern might be that I’m positing something hidden beneath language, a metaphysical foundation doing secret work. But that’s not what the three levels are doing in the opening post. They describe what’s visible in action, which is exactly what you say Witt wants. The infant reaching for the hidden toy is visible. The dog going to its bowl is visible. The visibility is the point. What the three levels say is that not everything visible in action is linguistic, and that some of what’s visible in action is a condition of language getting started at all. OC 204 says this. Whether there’s more to say about what grounds these pre-linguistic certainties is a question for later in the thread, not something the opening post needs to settle.

One more thing. You say grammar applies all the way down in the sense that whenever we describe, interpret, or make sense of experience, we’re operating within grammar. Agreed. But “whenever we describe” is the key qualifier. Grammar governs description. It doesn’t govern everything that’s there to be described. The infant acts before it has grammar. The dog acts without grammar. We describe both using grammar. The three levels honor both of these facts, viz., the pre-linguistic reality of the acting and the grammatical character of any description we give of it. That’s the distinction the post is drawing when it says the terminological difficulty is a symptom of the problem, not a failure to solve it. It’s why the post leads with examples (sitting, reaching, trusting) before introducing any labels, and why it flags the language problem as a substantive philosophical point.

11

Jay, let me clarify.

First, a correction. You describe my position as saying that justification “depends on presupposing certain conditions” and that “objects persist” is among them. That’s not what I’m arguing. “Presuppose” implies something adopted in advance, a condition you take on before proceeding. But the certainty that objects persist isn’t something you or anyone presupposes. It’s not adopted at all. It’s shown in what you do. You reach for the cup. The dog goes to its bowl. The infant grasps the toy. Nobody presupposed anything. The certainty is in the acting. Calling it a presupposition pulls it back into the propositional, as though somewhere behind the scenes you’ve signed off on “objects persist” before getting on with your day. That’s exactly the picture I’m arguing against.

Now to your main point. You say you can not only state “objects persist” but offer reasons for holding it, and that it’s your position rather than your necessary presupposition. I don’t doubt any of this. You can formulate the claim, you can argue for it, and it feels like a position you hold. But notice what’s happening. When you formulate “objects persist” as a claim and offer reasons for it, you’re operating at the third level, the level of linguistic hinges. At that level, yes, it looks like a proposition and behaves like one. You can state it, defend it, treat it as something you believe on the basis of reasons.

But the certainty that objects persist doesn’t live at that level originally. It lives in your reaching (the acting apart from propositions) for your coffee without checking whether the cup is still there. It lives in the dog going to its bowl. It’s shown in acting, prior to any formulation. You didn’t arrive at “objects persist” by weighing evidence. You act with that certainty, and then, when a philosopher asks, you can formulate it as a claim and offer reasons. The reasons come after the fact. The certainty was already operative in everything you did, including in the very activity of giving reasons.

The proposition is real. The reasons you can give for it are real. But they’re descriptions of something that was already in place before any describing happened. That’s the map/territory distinction the opening post is drawing. You can make a very good map. But the territory is the acting, and the acting came first.

Your instinct that you hold this as a position is worth taking seriously, though. It tells us something about how the third level works. Linguistic hinges can feel like positions because they’re statable and defensible. But their certainty doesn’t come from the reasons you can give for them. It comes from below, from the pre-linguistic and nonlinguistic hinge certainty that was operative before you ever put it into words. OC 204 is the passage that makes this visible. Acting lies at the bottom of the language-game. Not positions. Not reasons. Acting.

12

To start, I didn’t find you post “exhaustingly long.”

Well, as I understand it, it is knowledge–it’s intuition. It’s the knowledge we get from sitting, standing still, and moving around submerged in the world for (in my case) 74 years. Most of it probably comes from direct experience without intentional learning. I learned that things fall long before I heard about gravity. Some comes from being taught and socialized by others. Some I figure out for myself by observation, rational reflection, and experimentation. And some likely is built into our human nature, e.g. Kant’s a priori knowledge of space and time.

I think this is true and I think it explains why many philosophers reject intuition as a, maybe the, fundamental way of knowing.

I agree, object permanence is not a “position” but, as I understand it, it does have to be learned, at least to an extent. Some is also innate. And babies do function without it with much help from their parents.

I wonder if these distinctions make sense, at least as fixed levels. It seems to me they are always interacting with each other. Knowledge at Levels 1 and 2 probably often started out at Level 3 until it became habitual, e.g. learning to tie your shoes or tell time. As I’ve gotten older, my Level 1 certainties about he world have been challenged by changes in my physical condition. I sometimes have vertigo which has led to some falls. I now find myself much less certain, in a non-linguistic, non-reflective way, about the solidity of the world.

13

There is a sense of “I know” that fits what you’re describing. In OC 7 Witt says his life shows that he knows or is certain that there is a chair over there, a door, and so on. He gives simple examples, telling a friend to take that chair, to shut the door. His life, what he does, shows the knowing. That’s real, and I wouldn’t deny it. But it isn’t intuition. Intuition suggests a faculty, a way of grasping truths that bypasses ordinary justification but still delivers knowledge. That keeps the certainty inside the epistemic framework, just arriving through a different door.

What Witt is pointing at is something else. The “knowing” shown in your life, in 74 years of sitting and standing and moving through the world, isn’t a special way of accessing truth. It’s conviction shown in acting. It’s the certainty of the reaching, the stepping, the sitting. And it isn’t accumulated from experience the way you’re suggesting. An infant acts with the same certainty. A dog acts with the same certainty. You weren’t sitting there at age two gathering evidence for the persistence of objects and drawing inductive conclusions. You were reaching, grasping, acting. The certainty was in the acting from the start, not arrived at through experience but operative as the condition under which experience could happen at all.

The distinction matters because calling it intuitive knowledge makes it sound like a very deep or very old epistemic achievement, something you built up over a lifetime of being submerged in the world. Some of what you describe IS like that. You learned that things fall long before you heard about gravity. That’s genuine learning from experience, and it does accumulate over a lifetime. But the certainty that the ground will hold when you step isn’t in the same category. You didn’t learn it. It was never in question. It’s the condition under which learning about falling, about gravity, about anything at all, takes place.

OC 205 makes this sharp, viz., the ground is not true, nor yet false. Knowledge is true (that’s one of its conditions). If the ground isn’t true or false, it isn’t knowledge in the epistemic sense. It’s the certainty that makes epistemic knowledge possible. You can call it “knowing” if you mean what OC 7 means, conviction shown in a life. But it isn’t intuition, and it isn’t the product of accumulated experience. It was there before the accumulating started.

You mention Kant’s a priori knowledge of space and time. That’s a useful comparison because Kant saw that something had to be in place before experience could get started. But Kant still calls it knowledge and locates it in the structure of the mind. Witt is pointing at something more radical. In OC 204 acting lies at the bottom of the language-game. Not knowing, not intuiting. Acting. The certainty is in what the creature does, and it’s prior to any framework that describes it.

14

Thanks for the thoughtful reply. I hope this doesn’t sound like a linguistic quibble but . . . Aren’t you speaking metaphorically when you talk about a “certainty” that “lives” somewhere? You claim that a dog experiences this same phenomenon. Is it possible to say more directly what this means? I’m by no means sure about any of this, but my tentative position going into this conversation would be that psychological certainty (which is my best guess at what you’re talking about) is not the kind of certainty that philosophy is interested in, or not primarily. But as I say, I’m not sure what to do with the idea of certainty living somewhere.

That said, I think you’re onto something with the three levels. I’ll stay engaged with it.

OK. I was responding, or thought I was, to this:

How should I have understood that insight?

This is also a good quote for highlighting the language question. What do you mean by something being in place, and what sort of something is it? Again, I’m guessing you’re referring to a psychological conviction, but I’m not sure. And I do sympathize with how hard it is to find language for experiences at this level.

15

Jay, good questions and a fair catch.

You found me using “presuppose” in an earlier post and you’re right to flag it. That was sloppy. What I should have said is that any observational or rational grounds I might cite are already operating within a world of enduring objects, enduring observers, enduring memory. The enduring isn’t presupposed. It’s shown in the activity of citing grounds at all. You don’t adopt “objects endure” before you start investigating. You investigate, and your investigating is already the acting of a creature for whom objects endure. The certainty is in the investigating, not behind it.

Now to your main questions. You ask whether I’m talking about psychological certainty, and the honest answer is that it depends on what we mean by psychological. The word is ambiguous in the same way “certainty” is, and the ambiguity matters.

If “psychological” means belonging to the subject, first-person, experiential, then yes, this is psychological. Witt says so in OC 174, “I act with complete certainty. But this certainty is my own.” OC 42 says conviction is a mental state (while knowledge is not). OC 86 replaces “I know” with “I am of the unshakeable conviction” and finds nothing lost. The certainty I’m describing belongs to the subject. It’s mine in the way my acting is mine.

But if “psychological” means a feeling I introspect on and report, an inner state that could accompany any content regardless of truth, then no. That’s what I’ve been calling subjective certainty in the OC thread. You feel sure. You might be wrong. The feeling floats free of the facts. Moore holds up his hands and announces “I know,” and what’s actually operative is this kind of certainty, the feeling of sureness. Witt pushes against it throughout OC.

The conviction shown in acting is different from both of these descriptions while sharing something with the first. Put your hand on the table. You reached, and the surface was there, and at no point did you entertain the possibility that it wouldn’t be. That’s conviction. It belongs to you. It’s first-person and experiential. But you weren’t feeling anything in particular about the table. You weren’t introspecting on your sureness. You just reached. The conviction is in the structure of the reaching itself, in the fact that your body went out toward a surface without hesitation. It belongs to the subject without being a feeling the subject reports on. The dog going to its bowl makes this vivid. The dog’s conviction is real, belongs to the dog, is in that sense psychological. But the dog isn’t feeling sure. The dog is going to its bowl, and the going is the conviction.

So when you guessed “psychological certainty” you were in the right neighbourhood. What I’d resist is hearing “psychological” as “just a feeling of being sure,” because that collapses the phenomenon back into the subjective certainty Witt is working against. The certainty I’m pointing at is first-person, experiential, the subject’s own. It’s conviction. But it’s conviction shown in a life, not conviction reported as a feeling.

I grant the language is imperfect. Witt’s aim is to give statements one would like to make but cannot make significantly (OC 76). The terminological difficulty is real and it’s a symptom of the problem, not a failure to solve it. “Psychological” invites one hearing. “Not psychological” invites the opposite misunderstanding. The phenomenon sits between our available categories, which is exactly why Witt spent eighteen months working on it.

And “something being in place before description” means the reaching was happening before anyone said anything about it. The infant reaches for the hidden toy. No proposition. No framework. Later we say “the infant expects the object to persist.” That description is real and useful, but it comes after the fact. The infant was acting with conviction, and the conviction didn’t wait for anyone to formulate it.

Again, part of the problem is working out the correct way to describe these ideas, which is what Witt was trying to do.

16

Excellent elaborations, thanks. Couple of responses:

Isn’t there another possibility? My investigating may be the action of a creature to whom it appears that objects endure, but this creature isn’t in fact certain. So it investigates. Or do you maintain that to be sentient at all is to be certain of such hinges? Or, another possibility: Perhaps I will only understand in retrospect, after having investigated, that my actions revealed me to be a creature for whom objects endure. It was the case, always will be, but I didn’t know it when I began.

This distinction between “psychological” as a feeling state and “psychological” as subjective yet not a report about feelings is incredibly important. I think your discussion of it is really good. All it leaves me curious about is, again, what language we should deploy here. I’m glad you (and Witt) agree that the language problem is real, and difficult. Where do we locate the force of subjective certainty without claiming it’s a feeling, one we could have more or less strongly? For now, “conviction shown in a life, not conviction reported as a feeling” will do fine for me. (I also think it is more than merely acting a certain way, but that’s another story.)

17

Forgive me for interrupting:

Conviction is teleological. It presupposes that I sense something is going to happen. I wonder how such an intuition is possible without memory coming into play? When we do something with conviction, we necessarily make assumptions about what is going to happen. In particular, there is a “mental picture” of what is going to happen; there is an expectation. How can an expectation make sense unless we recall similar situations that we project into the future? Take the first time a baby walks: it must necessarily have the intention to walk, which implies that it remembers how other humans around it walk.

But it seems to me that there is a structural doubt, which causes us distress or anxiety about things that might happen. This doubt stems from the uncertainty that memory itself generates. What I mean is, in our minds we are taking the memory into new territory, into a new context; this is what we call expectation. And that always generates doubt. The memory, or a series of them, are shown here as signs that are grafted onto a new context. That is why there is always doubt, because we experience the uprooting of the sign from its ‘original’ context.

18

While considering these various reactions, where is a connection to the actual text?

19

Why the Framework Matters: Regress, Circularity, and Gödel

Epistemology has been stuck on the same problem for a very long time. Every justification seems to need a further justification. You believe P. Why? Because of Q. Why Q? Because of R. And so on. The regress never terminates. Or it circles back: you believe P because of Q because of R because of P. Neither outcome is satisfying. The regress gives you an infinite chain with no foundation. The circle gives you a foundation that supports itself, which isn’t support at all.

The two major epistemological traditions are responses to this problem. Traditional foundationalism says the regress terminates in self-evident or self-justifying propositions, basic beliefs that don’t need further justification. Coherentism says forget the regress, what matters is that your beliefs form a coherent web, each supporting the others. Both are trying to find the bottom within the space of propositions and justification. And both fail. Foundationalism can’t explain what makes a basic belief self-justifying without either smuggling in further justification or simply declaring “this one doesn’t need it.” Coherentism can’t explain why a perfectly coherent set of beliefs has any connection to reality. A coherent fiction is still fiction.

The framework shows why both fail. They’re looking for the bottom in the wrong place. They assume the bottom must be a proposition, a belief, something that lives in the space of justification. It must be a special proposition that justifies itself (foundationalism) or a set of propositions that justify each other (coherentism). But the bottom isn’t a proposition at all.

Go back to the chair. You’re sitting in it. The certainty that it holds you isn’t a belief you’ve justified. It isn’t a self-evident proposition. It isn’t part of a coherent web of beliefs you’ve constructed. It’s shown in your sitting. It’s in the acting. OC 204: acting lies at the bottom of the language-game. OC 205: the ground is not true, nor yet false.

That last line is the key. If the ground is not true, nor yet false, then it isn’t a proposition. Propositions are true or false. The ground is neither. It’s outside the space where truth and falsity, justification and doubt, operate. And this is why the regress stops. Not because you’ve found a super-proposition that justifies itself, but because you’ve reached something that isn’t in the justification game at all. You can’t ask “but what justifies your sitting without testing the chair?” The question doesn’t apply. Justification is something that happens within the space the acting makes possible. The acting itself is the condition under which justification proceeds.

The regress, in other words, was never going to be solved from within epistemology. It was generated by the assumption that everything at the bottom must be propositional, that the foundation of knowledge must itself be knowledge. Wittgenstein dissolves it by showing that the foundation isn’t knowledge. It’s hinge certainty, shown in acting, and it’s a different kind of thing from anything the regress is chasing.

Now here’s where it gets interesting.

Gödel showed in 1931 that any formal system powerful enough to express basic arithmetic contains true statements it cannot prove from within its own resources. The system cannot establish its own consistency. This isn’t a defect of any particular system. It’s a structural feature of formal systems as such. No amount of added axioms closes the gap, because the expanded system generates its own new unprovable truths.

This has the same structure as the framework problem. A formal system is a rule-governed framework. Its axioms and inference rules are its hinges, the things that stand fast while everything else moves. And Gödel shows that these hinges cannot be justified by the system they support. The system’s resources are insufficient to ground the system. Sound familiar?

Kuhn arrives at the same place from the history of science. A scientific paradigm includes the standards by which theories are evaluated. When paradigms conflict, you can’t resolve the disagreement by appeal to evidence, because what counts as evidence is partly determined by the paradigm. The paradigm cannot justify itself using its own evaluative standards. Its resources are insufficient to ground itself.

Three thinkers, three domains, one structural insight. Witt in epistemology: the framework’s hinges cannot be justified within the framework. Gödel in formal logic: the system’s consistency cannot be proved within the system. Kuhn in the history of science: the paradigm’s standards cannot be validated by the paradigm’s own standards. Each is discovering the same limit from a different direction.

What makes this more than an interesting parallel is what it tells us about where the ground actually is. Gödel’s theorem doesn’t mean mathematics is groundless. It means the ground of a formal system isn’t itself formal. Mathematicians know their systems are consistent, but they know it from outside the system, through mathematical practice, through experience, through the fact that the system works and continues to work. The consistency is shown in the practice, not proved by the axioms.

Witt is saying the same thing about epistemic frameworks. The framework isn’t groundless. It has ground. But the ground isn’t a proposition within the framework. It’s the acting, the hinge certainty, the lived engagement with the world that makes the framework possible. The ground is shown in what we do, not stated in what we say.

This is why the infinite regress was never solvable on its own terms. The regress asks: what justifies your foundation? The question assumes the foundation is the kind of thing that could be justified, a proposition, a belief, a claim. But the foundation is acting. The question doesn’t reach it. It’s like asking what formal proof establishes the consistency of your formal system. Gödel showed the question can’t be answered from within. Wittgenstein showed the same thing about justification. And in both cases, the fact that the question can’t be answered from within doesn’t mean there’s no ground. It means the ground is of a different kind than what the system can express.

20

Jay, two good questions and an interesting aside.

First: could the creature be one for whom it merely appears that objects endure, without actually being certain? Could it investigate precisely because it isn’t certain? Think about what that creature would look like. It would hesitate before reaching. It would check whether the cup was still there before extending its hand. It would test the ground before stepping. That creature would be one for whom object persistence is a hypothesis under investigation, something it’s trying to establish. But that’s not what we see. The infant reaches without hesitation. The dog goes to its bowl without checking. You put your hand on the table without entertaining the possibility that the surface wouldn’t be there. The absence of hesitation isn’t a failure to investigate. It’s the certainty. There’s no gap between the appearing and the being certain at this level, because the certainty isn’t a conclusion the creature has drawn from how things appear. It’s the character of the acting itself.

Your second suggestion is closer, i.e., that perhaps you only understand retrospectively that your actions revealed you to be a creature for whom objects endure. There’s something right about this. The description is retrospective. You act, and then later, when a philosopher asks, you can look back and say “my reaching showed that I was certain the surface would be there.” But the certainty wasn’t waiting to be discovered. It was operative the whole time. You don’t find out retrospectively that you were certain. You describe retrospectively what was already shown in the acting. The description is after the fact. The certainty is in the fact.

Yes, to be a sentient creature acting in a world is to act with hinge certainty. Not because sentience entails a philosophical position, but because a creature that didn’t act with this certainty couldn’t act at all. It couldn’t reach, step, navigate, feed. The certainty is a condition of being a functioning creature, not a conclusion that functioning creatures happen to arrive at.

I appreciate what you say about the language. “Conviction shown in a life, not conviction reported as a feeling” does the work for now, and I agree that finding better language for this is one of the things the thread should keep working on. Witt never found a term he was satisfied with either. “Stands fast” at OC 116 might be his best attempt. It captures the firmness without implying a feeling or an epistemic achievement.

Your parenthetical interests me. “More than merely acting a certain way” points toward something I also think is there, but you’re right that it’s another story. I’d be glad to take it up when you’re ready.