Thanks

Tool 2 - the grammar check, and grammar here in Wittgenstein’s sense, not in the schoolbook sense. He doesn’t mean punctuation or sentence diagrams. He means what are the rules of use for an expression, what role does it play, what counts as a sensible move, and what counts as a category mistake.

You take a philosophical statement, and ask, what kind of sentence is this supposed to be? Is it reporting a fact, giving a rule, expressing a commitment, giving a standard, drawing an inference, or doing something else? Sometimes philosophical trouble comes from treating one kind of sentence as if it were another.

It’s for spotting when a statement only looks like it’s saying something, when it’s really the result of our words getting detached from the contexts or settings that give them their point. You can also think of it as a way of asking, what would count as understanding this claim? What would we do with it?

For example, suppose someone says, “I’ve got a pain in my foot.” That’s not something you normally verify by looking for evidence in the same way you’d verify “there’s a nail in my shoe.” You might ask where it hurts, or whether it’s sharp or dull, or you might offer help. Imagine someone says, “I know I’m in pain because I observed it.” Such a sentence has the wrong grammar. It treats pain like an object discovered by inner observation, and it makes the person’s relation to their pain look like the relation to something external. You can feel the temptation, but the sentence is already sliding into a picture that isn’t expressing how we actually talk and respond.

Consider a philosophical example, “I know the external world exists.” It looks like an ordinary knowledge claim, like “I know there’s a tree in the yard.” But if you run a grammar check, you’ll ask, what would count as checking it, what would count as correcting it, what would count as evidence for or against it, and what would it mean to doubt it in the ordinary ways we doubt things? This is exactly where On Certainty starts to bite. Wittgenstein’s point isn’t that the proposition is false. It’s that in many contexts it doesn’t behave like a normal empirical claim at all. It’s closer to something that stands fast in the background of inquiry, the kind of thing you don’t typically confirm because it’s part of what makes confirmation possible.

A grammar check asks whether we’re trying to do philosophy with a statement/proposition that’s outside it's normal use. It looks like a straightforward statement of fact, but it’s functioning more like a rule, or a framework commitment, or a hinge. And once you see that, much of the philosophical pressure is dissipates.

Tool 1 is the simplest and, I think, the most important: “Look and see.”

When a philosophical question starts to feel deep, Wittgenstein’s first move is often to stop, and look at how the words are actually used in ordinary situations. Instead of guessing that there must be some hidden thing the word refers to, he'll point out what we already know how to do with it.

What it’s for, it’s for breaking the spell of abstract pictures. A lot of philosophy starts when we take a word that works perfectly well in our everyday life, remove it from its normal setting, and then demand an explanation of what it really is.

Think about the word game. Most people assume that a concept must have a strict definition. So, they ask, what is the essence of a game? But if you actually look, you find board games, card games, Olympic games, children’s games, solitaire, chess, tag, etc. There isn’t one feature shared by every case, and that discovery isn’t a defect. It’s a reminder that our concepts don’t always work by strict definitions, they often work by overlapping similarities (family resemblances).

Take the philosophical example, “What is meaning?” It can sound like we’re asking for a hidden object, a mental item, or a thing attached to a word. Wittgenstein’s move is to say, don’t posit anything yet. Instead look at what we call meaning in real life. We explain a word, we correct someone’s misuse, we translate, we follow an instruction, we misunderstand and then get it right, we use a word in a new context and people either accept it or reject it. The meaning isn’t a ghostly extra. It shows itself in the role the expression plays in our shared practices (forms of life).

If you want a quick test for whether “look and see” is needed, try the following: when you ask your question, do you immediately feel pulled toward a hidden mechanism or a deep entity that must be behind the scenes? If you do, you’re probably in Wittgenstein’s territory. The next tool, the grammar check, is what he uses to say exactly where the question goes off the rails.

Thanks for all of the replies. I'm trying to think of another subject for a thread. My philosophical focus tends to be very narrow, but hopefully I'll think of something that's interesting.

I don’t buy your reading of Wittgenstein. It takes his rule following comments and turns them into a kind of norm skepticism, as if Witt were saying there are no binding standards in a practice, only “creative re establishment” in each use case. That’s not what he’s doing.

Wittgenstein’s point isn’t that practices don't have any authority to correct us. His point is that the authority doesn’t come from some interpretation behind the rule, like rails laid in advance. It comes from how we’re trained, how we correct, what counts as getting it right, and how we actually go on together. If you deny that any regularities or shared expectations can bind, you don’t get a deeper Wittgenstein, you get the complete collapse of rule following, which is precisely the kind of picture Wittgenstein is fighting against.

Also, the “language on holiday” move is being misapplied. The holiday isn’t “making a general remark about what must be in place for doubt or inquiry to make sense.” Moreover, On Certainty is full of exactly that kind of diagnosis. The holiday is when words are detached from their practical moorings and kept afloat by a philosophical picture that can’t be cashed out in the activity. So, saying “a doubt misfires if it cancels the conditions of checking and correction” isn’t grounding meaning in an abstract template, it’s describing what makes the words doubt, check, settle, and improve do any work in the first place.

Finally, “normativity is re established in each use” sounds attractive, but if you take it seriously it wipes out the very distinctions that make language games possible. A word works when it can guide what comes next and make sense of responses, challenges, and correction. That requires more than fresh enactment. It requires a stable practice for the notions of success and failure to have application.

I’m not appealing to “prior criteria” in the sense of a metaphysical essence of improve or doubt. I’m appealing to the ordinary fact that in inquiry, improvement talk is answerable to how the practice handles error, correction, and learning. If someone insists that nothing could ever count as settling anything, that isn’t a daring new use that “creatively re establishes” normativity. It’s a use that removes the success conditions of the very activity it’s pretending to describe. That’s exactly what Wittgenstein calls out, not something he licenses.

From a Wittgensteinian view, I agree with the method, viz., look at use. But “look at use” doesn’t mean every use is equally in order or valid, or that we can’t diagnose when a word has lost its grip.

In the actual language-games where we talk about improving inquiry, “improve” is tied to things like learning, avoiding mistakes, tracking error, increasing reliability, making progress, even if the metric shifts from case to case. If someone usesimprove while also insisting that nothing could ever count as settling, correcting, or learning anything, then the word is no longer doing the work it normally does. That’s exactly the kind of grammatical diagnosis Wittgenstein makes, not a stipulative definition, but an observation that the proposed use has detached from the practice that gives it sense.

My point isn’t “here’s the essence of improve.” It’s that in our epistemic practices, improve has a role, and that role presupposes some intelligible notion of getting things right versus wrong. If you cancel that, you haven’t extended the grammar, you’ve broken it.

One way to address of the “why stop?” question is to notice a structural pattern that shows up outside epistemology too. Gödel showed that in any formal system strong enough to do arithmetic, there are truths the system can’t settle using only its own internal rules. You can settle a particular undecidable claim only by stepping out of the framework's standpoint, i.e., going meta, adding axioms, thereby widening the framework. But then the same limitation shows up again at the new level. The demand for total closure keeps moving.

I think something structurally similar is happening in the hinge discussion. Ordinary inquiry works because some things stand fast: not because we proved them in the ordinary way, but because they are what make correction intelligible. When someone asks for reasons for everything at once, they’re not just asking for a better justification inside the practice, they’re shifting to a meta demand for a standpoint that can validate the whole practice without presupposing it. You can do meta clarification, and sometimes you should, but you don’t get a final, once-and-for-all foundation that stops the question forever. Like Gödel, the attempt to force total closure tends to generate an endless “one more level” move.

The point, of course, isn’t “don’t ask meta questions.” it’s distinguish between meta work that improves our ability to detect error inside epistemic inquiry, and meta demands that try to secure inquiry from outside, by a standard that can’t itself be justified without reintroducing the very background it’s trying to suspend.

what I’m doing when I make standards explicit and introduce guardrails. That’s improvement inside an epistemic practice - refining what counts as evidence, tightening defeater sensitivity, clarifying error-signals. But hinge talk is aimed at a narrower point. It’s about what must remain in place for any practice of error and correction to be intelligible. If someone proposes a redesign that still preserves the possibility of settling anything, fine. But if the proposal is effectively: “treat every check as suspect in principle, and every standard as illegitimate unless justified by a further standard,” then the redesign isn’t more reflective, it removes the very success conditions of epistemic inquiry. That’s not a prohibition. It’s a diagnosis of self-undermining.

So I’m not saying the extra-game question “could we improve this?” is meaningless. I’m saying: some hinges are revisable and are exactly where improvement debates live, while bedrock hinges are what make the debate possible in the first place
— Sam26

I agree with the overall direction of your response, but it seems to over-intellectualize in places, explaining where it only needs to describe. Rather than having to decide which questions are “allowed” or “forbidden,” to map hinges once and for all, to discard a bad analogy in favor of the right one, we need only look at how words like reason, doubt, improvement, and justification are actually used in our lives. There is no answer in advance to whether the question “Could our epistemic practices be improved?” is coherent. Sometimes it is coherent, sometimes it is idle, sometimes it is revolutionary, sometimes it is nonsense, and which it is depends entirely on the language-game being played. In actual life, rules are sometimes followed blindly, sometimes revised, sometimes ignored, sometimes negotiated. There is no sharp line between playing a game and redesigning it; there are just different activities with different criteria. — Joshs

Sure, we should look at how words like reason, doubt, improvement, and justification actually get used. But I think you’re using that point to dodge the problem.

Nobody here is trying to be the language police, deciding in advance what questions are allowed. The issue is simpler, viz., some moves stop functioning as doubt because they wipe out what would count as checking or settling anything. If you say, “Every check is suspect, every standard is illegitimate unless it’s justified by a further standard,” then you haven’t made inquiry deeper, you’ve taken away the ground where inquiry happens. At that point doubt becomes a posture, not an activity with any conditions of success.

And yes, in real life rules get followed, revised, ignored, negotiated, all of that. Fine. But revision still has to leave us with a difference between “we got it right” and “we got it wrong,” otherwise the idea of improvement doesn’t even have a target. That’s the hinge point. It’s not “mapping everything once and for all.” It’s just noticing that some philosophical questions keep the vocabulary of inquiry while canceling the thing that gives that vocabulary meaning.

I’m not saying, “it never makes sense to ask whether our practices can improve.” I’m saying: improvement talk is meaningful when it still leaves room for correction. But when the improvement proposal is really “nothing can ever settle anything,” then it’s not meaningful, it’s self-defeating.

The question “can our justificatory practices be improved?” is not only coherent, it’s exactly what I’m doing when I make standards explicit and introduce guardrails.
— Sam26

Yes.

I’m not denying meta-level understanding. I’m saying the meta-level understanding changes what kind of thing you’re doing, i.e., explaining the rules of the game, not making another move within the epistemic framework
— Sam26

In my opinion, this is where the chess analogy breaks down. Whereas in chess there is a clear separation between playing the game and explaining the rules, I don’t think this distinction holds for rational inquiry.
— Esse Quam Videri

even in literal games, rules can change, but note how they change: you can redesign chess, or create variants, but for any given game the rules stand fast while you’re playing. They aren’t propositions being assessed move by move, they are what make it possible for a move to be legal or illegal. And when someone says, “we should change the rules,” it’s often not clear what “improvement” even means without importing standards that aren’t internal to chess at all, enjoyment, fairness, aesthetic unity, and so on. That doesn’t make the question incoherent. It just shows that redesign and rule following are different activities.
— Sam26

If we group these quotes together, I think we get a good picture of the issue. It reinforces my notion that there's nothing wrong with the case you make in your paper. The question of whether a game is a good analogy or metaphor is quite separate from the question of whether you've provided a more perspicacious understanding of JTB. I believe you have.

As to games . . . If a game is something whose rules can be questioned and/or improved (from a standpoint outside the game, of course), then it is not a good analogy for a practice governed by "bedrock hinges." I think all three of us would agree with this. Improvement or inquiry outside a set of rules is presumably governed by a further set of rules; otherwise the idea of "improvement" would be hard to explain. So I think you want to avoid suggesting that our ordinary epistemic practice is like a game with rules. Up to a point -- the point of foundational hinges -- it is; we usually play within those rules. But we can readily move to a different level at which the idea of improvement can't get a grip, since we'd be asking for "reasons to improve" that put into question what it would mean to improve. We've struck a bedrock hinge. But there is no literal game like that; the analogy does break down at that level. If chess were such a game, for instance, we'd be forced to say that a suggestion to improve chess can't be made because "improvement" only has meaning within the rules of chess.

So the main point I would press you on is the final, bolded sentence in your quote above. Why is redesign not a rule-following activity? It doesn't follow (all of) the rules of the practice being redesigned, but surely there are rules nonetheless, even for using concepts like enjoyment, fairness, aesthetic unity, et al. Again, think of your own paper: In the name of a set of rules you carefully employ (and could no doubt explain if asked), you offer changes to the (subset of) rules that seemed to characterize JTB. But this "redesign" of JTB absolutely is a rule-following activity. If it weren't, we readers would get pretty impatient with you! If someone challenged you to lay out your justification for the improvements, you'd do it. You'd strongly resist the idea that such a challenge was incoherent, that it called into question the very idea of justification.

In a sentence, then: There is no game whose rules cannot be candidates for improvement; therefore rational discourse as a whole is not a game. — J

You’re mixing three different things and then acting as though the mix refutes my point. It doesn’t.

I would say Redesign is rule-following is a dodge.
Of course, redesign has norms, consistency, coherence, non-contradiction, etc. I didn't deny that. The point is whether redesign is rule-following in the same sense as the practice being redesigned. It's not. When you change chess you may use things like good game design, but you aren't making another legal chess move. You’ve shifted your activities. Saying “there are still rules” doesn’t answer the point, it changes the subject. If your objection is merely “there are norms at the meta-level,” then congratulations, everybody agrees, and nothing I said changes.

Second, you’re equivocating on “inside rational discourse.”
You keep saying, “rational inquiry includes self-reflection, so meta-level reflection is still inside inquiry.” Fine, but that’s just a slogan unless you say what makes it the same kind of inquiry. Here’s the problem as I see it. If every rule is always a candidate for improvement by demanding a further justification, you’ve just built another infinite escalation. At some point, you either stop, or you pretend you don’t stop while relying on what you refuse to acknowledge. That’s exactly what hinge talk is diagnosing. Not “don’t reflect,” but you can’t keep demanding a justification for the conditions of justification without smuggling those conditions in again.

Your JTB point is a category mistake. My paper redesigns JTB only at the level of how we handle justification in practice. That's all within the space of epistemic assessment. It isn’t an attempt to justify the possibility of justification from nowhere. So, your line “you’d strongly resist the idea that such a challenge was incoherent” misses the mark. I resist some challenges as incoherent, specifically those that cancel the very criteria by which the challenge could be evaluated. That’s not being evasive. That’s basic logic.

Now the sentence that really gives the game away is the following:

“There is no game whose rules cannot be candidates for improvement; therefore rational discourse as a whole is not a game.”

It's just assertion, and it’s wrong in an im0portant sense. The hinge point isn't “no rules can ever be discussed.” It’s that the norms that make discussion, mistake, correction, and improvement intelligible cannot all be put on trial at once without emptying those words of useful content. You can always say you’re challenging everything. But if you’re still using better, worse, reason, defeat, and correction as if they have some traction, then you’re relying on what you claim to suspend.

So no, the chess analogy isn’t claiming rational discourse is literally a game. It’s forcing a distinction you keep trying to blur, viz. that clarifying the conditions of intelligibility isn't the same thing as arguing for a claim within those conditions or parameters. You can have meta-level norms without turning bedrock conditions into ordinary premises. And pretending otherwise is exactly how the issue of global doubt and endless “improvement” talk becomes performative rather than really answerable.

es, you understand my point exactly. Sticking with actual games, there are cases where game rules have in fact been changed to improve the game, or at least change it in a way that pleases its players better. (Money in Free Parking, in "Monopoly"!). Could a game like chess -- our chosen analogy -- be improved through rule changes? I frankly have no idea, but the point is that the question isn't incoherent or meaningless. It's perfectly possible to inquire of chess players, and by extension of the game of chess itself, whether improvement is possible. And if we do that, we aren't asking whether there's a way to make the bishop move "better" along the diagonal. The criteria for "better" are outside any particular rule. We might ask, Should there be only 7 pawns? That would change the rules, not clarify them.

How might we try to answer? What would this "should" mean? This is where it gets interesting, and moves us into the whole issue of rational inquiry. There are surely aspects of entire games that can be evaluated in terms of cleverness, enjoyment, a kind of artistic unity. Where do those criteria come from? That's unclear, but we know they aren't internal to any game as such. There is no rule in chess that specifies how to increase enjoyment, or even whether enjoyment is part of the game.

So the person who claims that the chess analogy holds for empirical inquiry appears to be saying that all these extra-chess questions can't be asked. We're urged to see the empirical practice of seeking justifications as the game, or the same as rational inquiry, such that to ask for reasons why we perform the practice as we do is to "ask for reasons for being reasonable," which is incoherent.

Now I'm not saying this is wrong. Sam26 makes a strong argument for how hinges operate in our epistemic practices, and I think we all agree that justification must stop somewhere, otherwise we do fall into incoherence. But what I am saying is that I don't think the (literal) game analogy shows us the right picture of what is going on. We need a better image or explanation for the shape of epistemic practice that would make clear why it is identical with rational practice itself. A game analogy doesn't show this -- unless you really do believe that to ask "Could chess be improved?" is a meaningless question. — J

I agree that the question “Could chess be improved?” isn’t meaningless, and I’m not committed to the view that every extra-game question is nonsense (some are some aren't). The misunderstanding is where that point is misunderstood, as if hinge talk were meant to forbid reflection or redesign.

First, it helps to clarify types of hinges. Some are what I’d call foundational hinges; they can shift over time as a framework changes. Others are bedrock hinges, the sort that don't show up as a candidate for epistemic assessment at all, for example: I am an object among objects, objects persist, there is a world in which checking and correction make sense. When these change, it’s not like discovering a counterexample. It’s more like losing the stage on which counterexamples could even count as counterexamples. That difference is significant because your “could the rules be improved?” question is mainly about the first kind, the revisable, upper-level foundational hinges.

Second, even in literal games, rules can change, but note how they change: you can redesign chess, or create variants, but for any given game the rules stand fast while you’re playing. They aren’t propositions being assessed move by move, they are what make it possible for a move to be legal or illegal. And when someone says, “we should change the rules,” it’s often not clear what “improvement” even means without importing standards that aren’t internal to chess at all, enjoyment, fairness, aesthetic unity, and so on. That doesn’t make the question incoherent. It just shows that redesign and rule following are different activities.

Now bring that back to epistemic practice. The question “can our justificatory practices be improved?” is not only coherent, it’s exactly what I’m doing when I make standards explicit and introduce guardrails. That’s improvement inside an epistemic practice - refining what counts as evidence, tightening defeater sensitivity, clarifying error-signals. But hinge talk is aimed at a narrower point. It’s about what must remain in place for any practice of error and correction to be intelligible. If someone proposes a redesign that still preserves the possibility of settling anything, fine. But if the proposal is effectively: “treat every check as suspect in principle, and every standard as illegitimate unless justified by a further standard,” then the redesign isn’t more reflective, it removes the very success conditions of epistemic inquiry. That’s not a prohibition. It’s a diagnosis of self-undermining.

So I’m not saying the extra-game question “could we improve this?” is meaningless. I’m saying: some hinges are revisable and are exactly where improvement debates live, while bedrock hinges are what make the debate possible in the first place.

I'll reply soon.

I’m not denying meta-level understanding. I’m saying the meta-level understanding changes what kind of thing you’re doing, i.e., explaining the rules of the game, not making another move within the epistemic framework (here I mean within the framework I've constructed in the paper).
— Sam26

In my opinion, this is where the chess analogy breaks down. Whereas in chess there is a clear separation between playing the game and explaining the rules, I don’t think this distinction holds for rational inquiry. To explain and justify the rules of rational inquiry is not to step outside of the game, but to deepen one’s understanding of the game itself, since inquiry includes the capacity for self-reflection on its own conditions. To place meta-level reflection entirely outside of epistemic normativity is to acquiesce to conventionalism. The claim that hinges can be appropriated by reason as necessary conditions of inquiry is not a claim about how we happen to play the game, but about what must be the case for judgment, error, and correction to be possible at all, and that is much something stronger than the chess analogy suggests. — Esse Quam Videri

I don’t think the chess analogy breaks down; I think it exposes the exact pressure point, viz., what counts as staying in the same game.

Inquiry does include self-reflection, but self-reflection doesn’t automatically remain within the same normativity. In chess, you reflect while you play, and you can revise your strategy, you can even decide to adopt a different opening. None of that touches the rules. Once you start asking what has to be in place for terms like move, illegal, mistake, and correction to apply at all, you’re not improving your play within the game. You’re spelling out the background rules that make what counts as a move in the first place. That’s not a conventionalist retreat, it’s a category distinction, which are the standards that govern ordinary epistemic claims, and they're not the same as the standards that govern clarifications of the conditions of those standards.

The worry about “acquiescing to conventionalism” only has standing if “outside the game” means “arbitrary social choice.” But that’s not what I mean. A hinge can be arational in role and still non optional. The point of the hinge diagnosis is precisely that these commitments are not mere conventions we could swap out at will, they are what gives judgment, error, and correction their force. A transcendental claim like “these are necessary conditions of inquiry” may be true in a structural sense, but it still doesn’t follow that the hinge has been appropriated into the ordinary space of reasons as a claim supported by evidence, alternatives, and defeaters. It has been explained as a condition of that inquiry.

I’d put it this way: meta reflection can deepen inquiry, but it can do so in two different modes. One mode stays inside the practice and improves our assessments, better evidence, sharper defeater handling, more precise concepts. The other mode articulates the background conditions without which assessment can’t gain a foothold. That second mode is not conventionalism, it’s not “how we happen to play,” but neither is it ordinary epistemic justification. It’s an explanation of possibility conditions, not a move competing with other moves.

With regard to hinges, I take it we agree that inquiry always proceeds from what stands fast; the remaining question for me is whether what stands fast is merely an arational background, or is meta-rational in the sense that the subject can come to reflectively understand why such commitments are unavoidable given the structure of knowing. If the latter, then such background hinges can themselves be appropriated into the game of giving and asking for reasons. — Esse Quam Videri

.

Calling hinges arational doesn’t mean they’re irrational, blind, or immune to ideas. It means they don’t operate as moves in our justificatory practices. In chess, the rule bishops move diagonally isn’t something you conclude from evidence or defend against objections inside the game. It’s what makes the game playable. You can explain it, even justify why we adopt it, but none of that turns the rule into a move you play on the board.

The sense of arational I’m using, viz., is that hinges are arational because they are conditions of intelligibility for ordinary epistemic assessment, not candidates for it. You can give a perfectly rational, clarifying account about why they have to be in place, but the hinge itself isn’t “supported by reasons” in the same way that an empirical claim is, because reasons already presuppose the background that makes support, defeat, check, and correction doable.

I’m not denying meta-level understanding. I’m saying the meta-level understanding changes what kind of thing you’re doing, i.e., explaining the rules of the game, not making another move within the epistemic framework (here I mean within the framework I've constructed in the paper).

The paper needs some revisions, but I think it could be submitted to...

1) Episteme (Cambridge), which is a general epistemology journal.

2) Synthese (Springer), which is another good match.

3) Ergo (Open Access, no author fees)

Your comment about "practice" is something I've been thinking about, so it's something to consider.

I never thought Gettier had something important to say about JTB, but it took a while to figure out exactly how the problem manifested itself.

Gettier does not overturn JTB; it signals the need to make explicit features of justification that the classical formulation left implicit.
— Sam26

From what I can tell, the Ten Coins situation does not have “features” at all, to even make explicit.

It is a standing within a practice, fixed by public criteria that settle what counts as competent support in the context.
— Sam26

This case doesn’t even have any criteria, or mechanics, or judgments (what you might call “linguistic training”) for the relation between coins and jobs. It is obviously philosophy trying to shoehorn formal logic onto a situation without any viable alternative. The fact that it is an imagined world actually does not matter. Wittgenstein creates simple situations (like picking a color of flower) but it is to show the consequences of imposing forced criteria by contrasting that with what we would need of a wider context of criteria and mechanics (even imagined) for a situation. Now the criteria for justification are all well and good, but this doesn’t even get off the ground; it just seems like a lot of work to say correlation is not causation. — Antony Nickles

The Ten Coins case is thin, and that is part of the point. Gettier creates a situation where the justification is basically a detachable bit of formal support that can be preserved while the world shifts underneath it. That’s exactly why I say Gettier is trading on an impoverished picture of justification: the case is set up so that there are no real practice-level mechanics for what counts as competent justification, no standards for error and correction, and no disciplined way to track mistakes. It isn’t exposing a flaw in JTB, it’s exposing what happens when we treat justification as some free-floating relation between propositions rather than as objective justification inside the practice of epistemology.

That also answers your “correlation is not causation” point. The moral isn’t merely “don’t confuse correlation and causation.” The moral is that the classical JTB slogan can be misread as if J were satisfied by any arguable support, even if the support is structurally incapable of carrying the conclusion across relevant mistake-conditions. I claim that once you make objective justification explicit as practice-governed, with defeater sensitivity and correction built into it, the Ten Coins style justification is revealed as too thin to count as knowledge. I’m not doing extra work to rescue JTB from Gettier. I’m saying Gettier only lands if we let justification be that thin in the first place.

You rightly emphasize that inquiry presupposes what stands fast, but you tend to treat these certainties as outside epistemic assessment altogether. How, then, do you account for intellectual conversion—those moments when what once stood fast becomes questionable and inquiry reorganizes itself at a deeper level? Are your hinges provisional horizons, or final grammatical boundaries? — Esse Quam Videri

Conversion is real, and it’s actually a good test of what I believe Wittgenstein means by hinges. Some hinges that stand fast for us are local and revisable (like the rules of chess), and when they shift the whole field of inquiry gets reorganized. But not everything that stands fast is like that. There are also deeper certainties (I'm an object separate from other objects) that function as conditions of intelligibility for doubt and checking in the first place, and those don’t shift in the same way, because if they did the activity of inquiry would collapse.

My answer is: hinges aren’t all on one level (some are foundational, but others are bedrock). Some are provisional horizons within a practice, the ones that can change as inquiry advances. Others are grammatical boundaries in the strict sense, the background without which “evidence,” “error,” “correction,” and “defeater” stop having any role. Intellectual conversion is usually a reorganization among the first kind, a shift in what was taken for granted within the system.

This is why I call (and others) hinge certainty arational. It’s not that a hinge is sacred or immune by decree. It’s that hinges typically aren’t the kind of things that are decided by the ordinary routes of objective justification. When they genuinely change, it’s less like refuting a claim and more like adopting a new framework.

Thank you for these clarifying remarks. I have one additional follow-up question: in your account, is objective justification sufficient for knowledge, or is it a necessary but fallible condition whose success still depends on the independent fulfillment of conditions?

You say, quite reasonably, that epistemology cannot guarantee orientation toward reality by introspection alone, and I agree entirely. But I would be similarly reluctant to say that orientation toward reality is guaranteed by practice instead.

The residual worry here is this: practices can be corrigible, sensitive to defeat, and historically successful, and yet still fail to deliver truth in particular cases. It seems that at some point we must appeal to ‘being’ (what-is-the-case) in order to explain how a judgement can fully satisfy the norms of well-governed practice and yet still fail to be true. Practice can regulate responsibility, but success still depends on how things are. I'm not saying that you are refusing to make such an appeal, only that I didn't see it stated explicitly anywhere in your paper. — Esse Quam Videri

Objective justification is necessary for knowledge, but it’s not a guarantee, because truth is the success condition. For example, I say in the paper “Truth remains the success condition for knowledge. To say that a belief is true is to say that the world is as the proposition represents it.”

However, even a well-governed practice is subject to failure: a belief can meet the standards and still turn out false. That’s not a defect in the idea of objective justification; it’s part the fallible character of our justificatory system.

What objective justification does secure is the right to claim "I know..." the right to treat the belief as knowledge, given our best efforts. That’s why I point to the guardrails: they don’t make truth automatic; they discipline the way justification can fail. “No False Grounds excludes cases in which the support is defective… Practice Safety excludes cases in which truth is reached only by luck… Defeater Screening excludes cases in which the belief cannot retain standing under relevant challenge.”

This is also why I keep insisting that the target is epistemic certainty, not absolute certainty: “much of what we count as knowledge is not secured by absolute certainty,” and JTB+U clarifies “how fallibility and knowledge coexist.”

If you want one line: objective justification governs responsibility and standing, truth governs success, and my claim is that we can have real knowledge without infallibility because our practice of justification aims at disciplined, defeater-resistant stability, while still understanding that “how things are” can definitely surprise us.

1. On Justification vs. Judgment
You speak eloquently about justification within practice, but where, precisely, is judgment?

You describe justification as a public standing governed by grammar and practice, but you do not clearly distinguish justification from the reflective act of judgment by which the subject affirms that the conditions for truth are fulfilled. Do you intend judgment to be absorbed into justification, or is it an irreducible moment you have not yet made explicit? If it is absorbed, how do you avoid collapsing epistemic success into conformity with practice?

Main concern: Knowledge is not exhausted by correct use or standing; it culminates in an act of judgment that affirms being. I would argue that act cannot be replaced by grammar without loss. — Esse Quam Videri

I don’t want “judgment” to disappear, and I’m not trying to replace it with grammar. I’m trying to locate its role in the process.

On my view, judgment is the act of taking p to be true, the moment a person makes a commitment about belief X. Objective justification is what makes that commitment responsible, not a mere statement. So judgment and justification aren’t rivals. Judgment is the affirmation of X, justification is the warrant for that affirmation in a practice that includes standards for evidence, mistake, correction, and defeat. Without judgment, you don’t have a claim at all. Without objective justification, you have conviction, guesswork, or mere assent, even if you can produce something that looks like a justification.

That also answers the worry about collapse into conformity. My account doesn’t say “whatever a practice treats as justified is thereby knowledge.” A practice can be defective, insulated, or sloppy. That’s exactly why I make the constraints explicit, viz, the practice has to be one where error is possible, correction is intelligible, defeaters are taken seriously, and standards are answerable to failure modes. If those conditions aren’t in place, then you can have judgment and even conformity, but you don’t have objective justification in my sense.

On your last line, “an act of judgment that affirms being,” I’d put it a bit differently. Judgment can be oriented toward what is the case, and in that sense, it aims at reality (the facts), but epistemology can’t guarantee that orientation by introspection alone. That’s why I keep the paper aimed at epistemic certainty, objective justification, not at Cartesian absolute certainty. Judgment is irreducible as a human act, but the epistemic status of the judgment depends on whether it is governed by the right standards of practice rather than merely produced with confidence.

As I'm going through your paper -- which is extremely good -- I want to clarify one thing: Do you consider that traditional JTB is supposed to guarantee knowledge? Or is its goal more modest -- to provide grounds for claiming knowledge? — J

I take traditional JTB to be doing something more modest than guaranteeing knowledge in any infallible sense. I treat it as a grammar for when a claim to know is responsibly made, a true belief with objective justification in the relevant practice of what justification entails within the 5 methods I describe. That’s why I say JTB “mirrors the way we distinguish between mere belief and belief that has a secure place in our shared life,” and why it remains “a natural starting point” for thinking about knowledge.

This is also where it helps to separate JTB as a definition from JTB in practice. As a definition, it gives a clean schema. In practice, justification is not a simple box-check, it’s what your claim can actually justify inside a practice that has standards for evidence, error, correction, and defeat, and that treats some challenges as relevant and some not.

Those standards of practice (justification) aren’t private feelings or inner impressions, they’re “displayed in our shared procedures of correction and agreement, in what counts as getting it right and what counts as needing revision.”

I don't claim that JTB guarantees knowledge unless you're speaking about deductive reasoning (it's absolute in a restricted way). Most of our knowledge is inductive and so it's mostly probabilistic. It really depends on your method of justification. The method I provide can do both.

So in my framework, I talk about four senses of certainty, and these help explain the two different uses of "I know." First, there's epistemic certainty, which is really about having objective justification, something that stands up to public criteria and can’t easily be defeated. Then there's subjective certainty, which is more about personal conviction, it's the feeling of being sure about something from your own perspective. There's absolute certainty, which is tied to logical or moral necessity, i.e., things that simply cannot be otherwise. And finally, there's hinge certainty, viz., those arational bedrock commitments that make all these other kinds of certainty possible.

Now, when we say, "I know," we can be using it in that subjective sense, expressing a personal conviction, or we can be using it in the epistemic sense, pointing to something that meets those public standards of justification. My approach tries to show how these different senses of certainty all fit together. By grounding them in hinge certainty, we can see how both the subjective expression of "I know" and the epistemic use of "I know" are part of a larger, integrated picture."

Instead of framing hinges as a metaphysical claim about reality, let’s consider how they function structurally. Just as Gödel’s incompleteness theorems show that certain limits are built into formal systems, Wittgenstein’s hinges show that certain fundamental assumptions are built into our epistemic practices.

In other words, hinges aren’t there to prove anything metaphysical. They’re there to show where our practices of justification find their foundational footing. By drawing a structural parallel to Gödel, we see that these structural boundaries are not arbitrary; they’re intrinsic to how our epistemic language
games work, and hoow systems of belief work generally.

The takeaway being: when we talk about hinges, we’re pointing out that certain stopping points are part of the grammar of justification. They help us see why pushing certain doubts beyond those points stops being an epistemic move and becomes a different kind of game entirely. That’s the structural parallel to Gödel’s insight, and it’s what gives hinges their power.

I guess I can't get away from hinges. :grin:

My perhaps obsessive concern with the appearance/reality question springs from my dislike of the term "reality" in philosophy. It's not that I think there's "more to it than this," but the opposite: I think there's less to it. I don't think we should say that epistemic justification can show us what is "real" -- though see Sam26's point above, about how "truth," "reality" and "constraint" are all aiming at the same role in this discussion. Which is part of why I try to avoid "reality" as a term. I don't think it does much independent work. And at worst, it can blur the distinction between questions about being, and questions about thinking. — J

I don’t think we need to drop the contrast the term is trying to gesture at. The point isn’t “justification shows us the Real,” it’s that justification is answerable to something beyond mere endorsement or conviction. If truth does that job for you, then we can speak in those terms: objective justification is what entitles a claim to be treated as true within a practice, meaning it survives the practice’s tests, correction procedures, and defeaters. In that sense, reality isn’t doing much work. It’s just a way of reminding ourselves that error is possible, that correction is not merely a change of opinion, and that inquiry aims at what's true or justified.

I’m with you on minimizing metaphysical overtones. We can keep the functional point, answerability to constraint, without treating reality as some heavyweight philosophical concept.

If we think a belief is justified and it turns out it isn’t, that is on us for not looking closer, digging deeper (barring unforeseen issues). It might have been a snow-job, trendy, or made up of false data. We might initially think a belief is justified but the work was plagerized (literally mimicked), but that is a judgment of something other than justification (it is still justified, just not by them). It doesn’t help that a weak job of justification is nevertheless done solely, genuinely by the claimant. We only judge their understanding of justification by their demonstration of it. There is no other criteria for understanding because it is not an independent quality of a person, it is a logical distinction—thus why you can have (demonstrated) a pretty good or excellent understanding. You may have training, experience, etc., but still not understand how justification actually works, which is just another way of saying you don’t do it well. It could be a mistake, but if they actually suck at it, no one is going to say they understand it. — Antony Nickles

First, on the plagiarism case. I don’t think it’s right to say, “it’s still justified, just not by them.” If the justification depends on borrowed work they can’t actually own, then what they have is at best a true claim with borrowed support, not objective justification done by them. Objective justification isn’t just that good reasons exist somewhere in the world. It’s that the person can take responsibility for the support in the way the practice requires, including answering for sources, handling challenges, and tracking what it would mean to correct one's claim. Plagiarism is a clean case of mimicking the conclusion while lacking competent justification.

Second, on “there is no other criteria for understanding.” I agree that understanding isn’t a private inner thing. But “demonstration” needs to include more than producing an argument that looks good. In a lot of cases, the difference between doing it well and doing it badly shows up when the person is pressed on mistake conditions: what would count against this, what alternatives are there, what would defeat it, what would you revise if X were true. Someone can generate a weak or even superficially strong justification and still be unable to navigate those checks. That’s why I connect understanding to the practice’s error and correction structure, not to an extra mental property.

I’m with you, viz., understanding isn’t an independent psychological quality. But it’s also not identical to any single performance of justification. It’s competence across the relevant challenges, the ability to sustain objective justification when the practice does what practices do, probe, test, correct, and sometimes expose that what looked like support but wasn't.

Continuing with the final post of the paper.
Post #15

10. Conclusion

The classical model of knowledge as justified true belief remains a sound starting point, not because it answers every philosophical worry, but because it captures the grammar of our ordinary epistemic life. When we call something knowledge, we are not merely reporting a psychological state of conviction. We are placing a belief within a practice of justification and marking that it has succeeded in being true. This is why the model has endured. It reflects what we already do when we distinguish between mere opinion and beliefs that genuinely have standing.

The refinements developed here do not replace that model; they make explicit what the classical formulation leaves implicit. Justification does not operate in isolation. It functions within shared language-games and presupposes a background of bedrock certainties that stand fast for us. These certainties are not items of knowledge, and they are not established by further justification. They are foundational in a non-epistemic sense. They set the conditions under which doubting, checking, and justifying have their point. This is why the demand to justify everything does not express greater rigor. It reflects a misunderstanding of the role that what stands fast plays in our epistemic life.

Understanding, likewise, is not an optional addition to justification; it is internal to it. To justify a belief is to use the relevant concepts correctly within a practice, to move competently among their connections, and to recognize what counts as correction and withdrawal when the practice requires it. The beetle in the box makes the point vivid. If we treat understanding as an inward item, something to which one privately points as the basis of epistemic standing, we detach justification from the criteria that give it life. We do not strengthen knowledge by relocating its basis to the private interior. We dissolve the conditions under which the concept of knowledge functions at all.

These clarifications also reposition the role of Gettier in epistemology. Once justification is understood as a standing within an epistemic practice, disciplined by criteria of correct use and constrained by what stands fast, Gettier-style cases lose their supposed significance. They do not show that the classical model is inadequate, but that many discussions of Gettier rely on a conception of justification that fails to reflect how our epistemic practices actually operate.

To make these points concrete, I distinguished five routes through which justification typically proceeds: testimony, logical inference, sensory experience, linguistic training, and pure logic in its boundary-setting role. These routes are not ranked by epistemic importance. They reflect the order in which justificatory support most commonly appears in our language-games. Alongside these routes, I described three guardrails that express the discipline internal to justification: No False Grounds, Practice Safety, and Defeater Screening. These guardrails do not add new conditions to knowledge. They clarify what it is for support to count as justificatory within a practice and help explain why some beliefs that look well supported fail to have the standing required for knowledge.

The structural parallel with Gödel reinforces the same lesson from a different angle. Gödel’s results show that formal systems have limits that arise from their internal structure, limits that are not defects but conditions of the system’s character. Wittgenstein’s remarks on hinges show that justificatory practices have limits as well. Not everything that makes justification possible can itself be justified. The parallel is structural, not mathematical, but it is instructive. It helps us see that the presence of limits does not entail skepticism. It marks the conditions under which epistemic life can proceed.

The application to artificial intelligence illustrates why these distinctions matter now. Artificial systems can produce true statements, sometimes with impressive reliability. Yet knowledge is not merely the production of truths. It is true belief that stands within a practice of justification, governed by routes, constrained by guardrails, and framed by what stands fast. AI systems can assist human knowers and function as powerful instruments within our epistemic practices. But to treat them as knowers is to blur the grammar of knowledge at exactly the point where clarity is most needed.

What emerges, then, is an account of knowledge that is realist without dogmatism. Truth remains the success condition. Justification remains a practice governed by shared criteria. Bedrock certainties stand fast without becoming items of knowledge. Understanding is not a private achievement but a competence displayed in use. The result is not a new theory erected on top of the classical model, but a clearer view of its working parts and of the background that makes them possible. If there is a practical upshot, it is this: when the appearance of support is everywhere, the task is to recover the discipline of justification and to keep the grammar of “know” clear enough to do its work.

Continuing with paper...
Post #14

9. Objections and Stress Tests

Before closing, it is worth testing the framework against a few cases that are often used to pressure classical accounts of knowledge. The point is not to chase intuitions, but to show how the routes and guardrails clarify why some beliefs have the standing required for knowledge and why others do not, even when the surface looks similar.

Fake barn environment. Consider the case in which a person looks at what appears to be a barn, forms the belief “that is a barn,” and happens to be looking at the only real barn in an area filled with convincing façades. The belief is true, and from the person’s point of view the perceptual situation seems ordinary. Yet the belief lacks the standing of knowledge. JTB+U does not need a new condition to explain this. The belief proceeds through sensory experience, but the environment has altered the standing of that route. Practice Safety is decisive. In a setting saturated with decoys, the belief would easily have been false under nearby variations that the practice treats as relevant. The problem is not that perception stops functioning, but that the ordinary stability required for knowledge is not present. Defeater Screening also matters. The relevant defeater is built into the environment itself, namely that many barn-like objects are not barns. The point is not what the person privately considered, but what standing the belief has within a practice once that defeater is in play.

Testimony under distorted informational conditions. A second pressure point concerns testimony in an environment where repetition is treated as standing. Here the route is still testimony, but testimony has standing only within practices that supply criteria of credibility, provenance, and correction. No False Grounds blocks a common failure. Testimony can look supportive while resting on fabricated reports, altered media, or untraceable sources. A belief placed on such grounds can be compelling and socially reinforced, yet the support is defective at the point where the practice treats the defect as disqualifying. Practice Safety and Defeater Screening complete the diagnosis. A claim can be true by coincidence and widely repeated, while remaining unstable under ordinary informational variation. A claim can also remain persuasive only because relevant challenges are excluded rather than addressed. In either case, what is missing is not sincerity or intensity of conviction, but the standing a belief must have within a practice of justification to count as knowledge.

Human and AI hybrid cases. A third test concerns cases in which a person uses an artificial system as an aid. The temptation is to treat fluent output as knowledge, or to treat the user as having knowledge simply by receiving an answer. The framework clarifies the difference. A person can acquire knowledge with the help of AI only if the belief formed on the basis of the output is placed within a practice of justification that satisfies the guardrails. No False Grounds matters because an output can include invented citations or a false claim doing essential work. Practice Safety matters because slight prompt changes can produce incompatible outputs, which signals instability. Defeater Screening matters because relevant counter-considerations can be present and must be addressed within the practice if the belief is to have standing. This also shows the proper role of understanding. If the output is treated as a substitute for conceptual competence, then the belief can have the appearance of support while lacking the internal structure required for genuine justification. AI can be a powerful instrument within human epistemic life, but that does not collapse the distinction between producing true sentences and knowing.

What these tests show. These cases do not require a patch to JTB. They show that when justification is treated as a public practice with disciplined constraints, the difference between knowledge and lucky success is not mysterious. Truth remains the success condition. What varies is the standing of a belief within a practice, and that standing depends on the routes by which it is supported and on the guardrails that discipline that support. In this way, JTB+U does not replace the classical model. It clarifies what the model already presupposes when we speak carefully about what it is to know.

Continuing with paper...
Post #13

8. Broader Consequences for Epistemic Life Today

The refinement offered by JTB+U is not meant to remain at the level of conceptual reconstruction. Its point is to clarify how our epistemic practices actually work, and to make visible the distinctions that are now repeatedly blurred. In many contemporary settings, the appearance of support is treated as if it were justification, fluency is treated as if it were understanding, and conviction is treated as if it were certainty. The result is not merely disagreement. It is a weakening of the grammar of knowledge, a drift in the very criteria by which we distinguish knowing from persuading, and truth from mere plausibility.

One consequence of the framework is that it restores the place of justification as a public practice. When justification is treated as something private, or as an inner feeling of confidence, the discipline of epistemic life collapses into rhetoric. JTB+U makes explicit that justification depends on shared criteria, on the ability to locate a belief within an established route of support, and on the willingness to submit that standing to correction. This is why the language-games of knowledge depend on practices of challenge and withdrawal. A belief that cannot be corrected within a practice is not thereby strengthened. It is severed from the ordinary conditions under which justification has its point.

A second consequence is that the model clarifies the role of testimony in a world saturated with information. Testimony is not a lesser route. It is among the most common routes in ordinary life. Yet the present environment often treats testimony as interchangeable with assertion, as if the mere existence of a claim in circulation were enough to give it standing. JTB+U makes clear that testimony has standing only within the practices that grant it, and that this standing depends on criteria that are often ignored in modern informational contexts. When those criteria are weakened, testimony does not disappear. It becomes unstable, and epistemic life becomes susceptible to persuasion that imitates the surface of justification.

A third consequence concerns the probabilistic character of justification. Much of what we count as knowledge is not secured by absolute certainty. Our justificatory practices are often graded, and they frequently operate under conditions of limited information. This is not an embarrassment to epistemology. It is part of the grammar of our epistemic life. The mistake is to treat this fallibility as if it implied that knowledge is impossible, or that the classical model must be abandoned. JTB+U instead clarifies how fallibility and knowledge coexist. We can have knowledge without having what philosophers sometimes treat as conclusive proof, because the standing required for knowledge is determined within a practice, under disciplined constraints, against a stable background of bedrock certainties.

This is also where the distinction between the different senses of certainty matters. Subjective certainty is conviction. It can be intense, and it can be sincere, yet it does not settle anything about truth. Hinge certainty is bedrock, it stands fast and makes doubt possible, yet it is not knowledge. Epistemic certainty is the kind of stability a belief can have within a practice of justification, where the belief is resistant to relevant challenge and supported in the way the practice requires. Absolute certainty belongs to logic and necessity. Modern discourse often collapses these into one undifferentiated notion of certainty, and the collapse produces confusion. Conviction is treated as evidence. Bedrock is treated as dogma. Logical necessity is treated as a demand for knowledge. JTB+U separates these senses and returns each to its proper use.

A further consequence is that the framework explains why disagreement can persist even among sincere and competent thinkers. Disagreements are not always disputes over evidence. They can arise from differences in the background against which evidence is assessed, from differences in how concepts are being used, and from differences in which defeaters are treated as relevant. None of this makes truth relative. It shows that our practices of justification are complex and that the stability of knowledge depends on more than the production of arguments. When we recognize this, we are less tempted to treat disagreement as evidence of irrationality, and more able to locate what is actually at issue.

Finally, the framework provides a disciplined response to the current pressure to treat epistemic life as a contest of narratives. Persuasion is not the same as justification. A persuasive claim can be memorable, emotionally forceful, and socially reinforced, while still lacking standing within a practice of justification. JTB+U gives us a way to say this without moralizing. It identifies where the grammar breaks down. The guardrails make the point concrete. No False Grounds blocks claims whose support depends on what is not so. Practice Safety blocks claims that succeed only by coincidence or rhetorical timing. Defeater Screening blocks claims that remain compelling only because relevant challenges have been excluded or ignored.

Wittgenstein’s distinction between criteria and mere signs is useful here. Many things accompany knowledge: confidence, fluency, repetition, even social approval. Yet these are not what justify a claim. They are at best symptoms, and often only disguises. The criteria for knowledge belong to the practice: the routes by which justification is given, the guardrails that discipline it, and the ways a claim can be corrected, withdrawn, or defended when challenged. When those criteria are replaced by signs, epistemic life becomes vulnerable to persuasion that imitates the surface of justification.

In this sense, the account is realist without dogmatism. It affirms truth as the success condition. It affirms the public character of justification. It affirms the necessity of bedrock certainties without treating them as items of knowledge. It affirms fallibility without conceding skepticism. It also encourages a kind of epistemic humility that is not a retreat from truth but an acknowledgement of the limits built into our practices. The point is not that we should doubt everything. The point is that we should recognize what must stand fast for justification to function, and then take seriously the discipline by which beliefs earn their standing as knowledge within the language-games we share.

I agree with you, though many would disagree. I was curious where you would fall on the question. Sounds like we broadly agree on these issues. — Esse Quam Videri

Even if you were omniscient someone would disagree. It means nothing, don't you think?

I feel like I keep repeating myself. When I say, “the conditions that make doubt intelligible,” I mean the basic setup that makes checking and being checked possible. For any practice of inquiry to work, it has to treat some outcomes as settled and others not. Otherwise “I doubt,” “I tested,” and “I corrected myself” become empty statements.

We don’t have to start with the heavy phrase “appearance vs reality.” We can start with something leaner. For instance, settled vs unsettled, correct vs incorrect, passes the check vs fails the check. Those distinctions are already enough to rule out global doubt, because global doubt tries to remove the very idea that anything could ever count as settled.

“Truth vs opinion” can express the same structure, but only if truth means “what would be correct even if no one endorsed it.” If “truth just means “what our group happens to treat as correct,” then the difference between error and correction disappears into sociology. So, the real commitment isn’t a grand metaphysics of Reality with a capital R, it’s the thinner claim that correctness answers to something beyond mere endorsement. Call that truth, call it reality, call it constraint, it’s the same role.

I believe that’s the ontological answer here, not a theory of what exists, but the insistence that inquiry isn’t just opinion-management, it’s answerable to what settles these epistemological questions.

This is a strong reply, and I agree with much of it, but I don't think it gets to the heart of J's concern. My interpretation is that J is not questioning whether global doubt is incoherent, but is asking why grammar should be considered sufficient to settle the issue. In other words, what explains why grammar imposes the limits it does?

I would argue that Wittgenstein's anxieties over transcendental reification make it difficult for him to adequately address this question. I don’t think the deepest explanation can be grammatical. Grammar registers the limits, but it doesn’t generate them. These ultimately need to be grounded in the structure of our normative/epistemic acts themselves: to doubt, correct, or inquire is already to be oriented toward what is the case, toward conditions of fulfillment that distinguish seeming from being. An act of doubt misfires when it asks for fulfillment while cancelling the conditions of fulfillment.

So doubting is a form of judgment guided by reasons. Reasons presuppose the possibility of correcting mistakes by attending to data and testing insights. If you globally deny the existence of any constraint on the data of experience then you undercut the very idea of error, correction, learning and also doubt itself. Inquiry is intelligible only as a self-correcting process of answering questions about what is the case and is therefore rendered unintelligible under the assumption that there is nothing in principle that can settle such questions. — Esse Quam Videri

If someone hears grammar as some self-contained explanation, they might ask: why does grammar have that authority? Your answer is, it doesn’t float free. Grammar is the surface expression of deeper constraints built into what it means to doubt, inquire, and correct at all. In that sense, grammar gives us limits, but the limits are generated by the structure of normative acts, what you call conditions of fulfillment, the difference between seeming and being, and the possibility of error and correction.

This doesn’t really oppose the Witt line, it strengthens it. The hinge point isn’t language settles reality, it’s that global doubt misfires because it cancels the very conditions that make doubt an intelligible, or an answerable activity. The constraint isn’t merely linguistic, and it isn’t a transcendental reification either. It’s built into the logic of inquiry as a self-correcting practice, so if you deny in principle that anything can count as settling questions, you haven’t adopted a stricter epistemology, you’ve made the whole enterprise of error, learning, and correction unintelligible.

I’d frame it this way, grammar is sufficient to diagnose the misfire because of how it tracks the role our concepts play, but a deeper explanation is much morre practical and normative, the structure of what it is to seek fulfillment, be answerable to correction, and distinguish appearance from reality. That’s exactly why global skepticism collapses. Hopefully this answers your concern.

Continuing with paper...
Post #12

7. JTB+U and Artificial Intelligence: Why AI Does Not “Know”

The present interest in artificial intelligence has brought an old temptation back into view. We are inclined to treat fluent performance as if it were knowledge, and to treat the production of correct sounding answers as if it were understanding. This temptation is understandable. The outputs of large language models can resemble the surface of competent human speech. They can summarize, explain, and argue, and they can do so in a way that often passes casual scrutiny. Yet the resemblance is grammatical only at the level of appearance. When we look more closely, we see that the ordinary criteria for knowledge are not satisfied, not because the machine lacks a private mental state, but because it does not stand within the practices that give the concept of knowledge its use.

Truth remains the success condition for knowledge, and nothing in what follows weakens that point. An artificial system can produce a true statement, sometimes with striking reliability. But knowledge is not merely the arrival at truth. Knowledge is true belief that stands within a practice of justification, and the standing is not a decorative label. It depends on the routes by which the belief is supported, the guardrails that discipline that support, and the background of bedrock certainties that makes the whole practice possible. This is the first reason the language of knowledge becomes slippery when we apply it to machines. The system produces assertions, but it does not participate in the language-games in which assertion, challenge, withdrawal, and justification have their life.

This is also where the role of bedrock certainties becomes decisive. Human justification presupposes a background that stands fast for us. These certainties are not items we know. They are the inherited conditions under which doubting and knowing take place. They form a hierarchy in the sense that some stand deeper than others, and they are displayed in action rather than defended by argument. The point is not that a machine lacks a set of stored assumptions. It is that the machine is not trained into a form of life in which such certainties function as the background of justification. An AI system does not stand within the practices that define what counts as a mistake, what counts as correction, and what counts as the withdrawal of a claim. It can be updated, constrained, and fine-tuned, but this is not the same as occupying the human space in which bedrock certainties show themselves as what stands fast.

The five routes also clarify the difference. When a person justifies a belief through testimony, inference, sensory experience, or linguistic training, the support is situated within a practice in which the believer can be held responsible to standards. These standards are public and they include the possibility of being corrected in the relevant way. A language model can mimic the outward form of these routes. It can cite sources, draw inferences, and use perceptual language, but these are linguistic gestures, not placements within the practice itself. The model does not have testimony in the human sense, since it is not a participant in the practices that give testimony its standing. It does not infer in the human sense, since it does not operate with the conceptual competence that makes an inference a movement within a language-game rather than a pattern of token transitions. It does not perceive, and so it does not have sensory experience as a route of justification. It displays linguistic training in the limited sense that it has been trained on linguistic material, but this is not the kind of training by which a human learner comes to grasp the use of a concept within a form of life. It is closer to the acquisition of a statistical profile of usage than to the possession of a concept.

This is why the distinction between statistical competence and conceptual competence matters. A model can be statistically competent, in the sense that it produces language that fits patterns in its training data. It can do this at scale and with impressive fluency. But conceptual competence is not the possession of patterns. It is the ability to use a concept correctly within a practice, to respond to correction, to recognize when a challenge is relevant, and to withdraw a claim when the practice requires it. These are not private mental achievements. They are displayed in the way one stands within a language-game. The machine can be made to output a retraction. It can be prompted to list possible objections. Yet these are outputs, not the standing of a belief within a practice of justification.

The guardrails bring the point into sharper focus. No False Grounds matters because a model can generate support that looks acceptable and yet includes a false claim doing essential work. Practice Safety matters because a model’s correct output may be the result of a fortunate match rather than stable standing, especially in domains where the system has not been constrained by reliable sources. Defeater Screening matters because, while a model can generate lists of objections, it does not occupy the public discipline in which defeaters arise as challenges that change the standing of a belief. The model can simulate the discourse of justification, but it does not stand within a practice where its claims are owned, corrected, and withdrawn in the way that our language-games require.

None of this implies that AI is useless in epistemic life. The opposite is true. Artificial systems can be powerful instruments within human practices of justification, especially when their outputs are constrained by reliable data and when they are treated as aids rather than as knowers. They can help gather information, surface patterns, and organize arguments. But this usefulness does not collapse the grammatical distinction between producing true sentences and knowing. To treat the machine as a knower is to project the grammar of our concept of knowledge onto something that does not meet its criteria of use.

This is why JTB+U is especially valuable in the present environment. It gives us a disciplined way to distinguish persuasion from justification, fluency from conceptual competence, and the appearance of support from genuine standing within a practice. It also helps explain why the language of certainty is often misused in discussions of AI. A model can produce confident sounding claims, and this can resemble subjective certainty. But hinge certainty is not confidence, and epistemic certainty is not mere persistence under repetition. The kinds of certainty that matter to knowledge are rooted in practices and in what stands fast within them. The machine does not inhabit that structure, even when its outputs resemble the surface of a human epistemic performance.

For these reasons it is better to say that artificial systems can produce true statements, and can assist human beings in practices of justification, without saying that they know. The temptation to speak otherwise is understandable, but it blurs the grammar of knowledge at exactly the moment when we most need it to be clear.

Continuing with paper...
Post #11

6.4 The Guardrails as Clarifications of Justification

Taken together, these guardrails clarify the shape of justification as it functions. No False Grounds excludes cases in which the support is defective at a crucial point. Practice Safety excludes cases in which truth is reached only by luck. Defeater Screening excludes cases in which the belief cannot retain standing under relevant challenge. None of this adds a new condition to JTB. It makes explicit the discipline that our justificatory practices already embody.

This is also where the role of understanding, as I use the term, becomes sharper. Understanding is not a separate achievement layered on top of justification. It is the conceptual competence through which justificatory support has its proper use within a language-game. The guardrails describe the constraints that this competence must respect if a belief is to have the standing required for knowledge. When a belief violates these constraints, it may still be true, and it may still feel compelling, but it does not stand as knowledge within our practices of justification.

With these guardrails in view, the classical model is not weakened but clarified. We can see why some beliefs that look well supported nevertheless fail to constitute knowledge, without treating Gettier-style cases as decisive objections. The next step is to apply this clarified grammar to contemporary pressures, including the temptation to describe artificial systems as knowers and the need to preserve the ordinary concept of knowledge in an information environment that often rewards persuasion over justification.

Rule-following brings the point into sharper focus. To use a concept correctly is not to consult a private rule, nor to match an inner image, but to have been trained into a practice where “correct,” “mistake,” “same,” and “different” have their use. The standards that govern justification are therefore not hidden in the mind. They are displayed in our shared procedures of correction and agreement, in what counts as getting it right and what counts as needing revision. This is why justification is public in its grammar even when it concerns matters that are privately experienced. If there were no practice in which correctness could be shown, there would be no sense to the claim that a belief is justified.

Of course there is formal logic, which has it uses, but I would think more important is the internal logic of a practice, some of which are particular versions of ones you mention. As I said above, there are criteria for what we consider to be an apology, a sufficient one to categorize it as an apology. There are constraints for correctness, boundaries for appropriateness. The implications of certain acts within a practice have specific implications. All this is to say that formal logic is but one practice, no better, more important, or more necessary/powerful than others. Of course, its independence and certainty make it more desirable. — Antony Nickles

Formal logic is one practice among others, and a lot of what governs our epistemic life is the internal logic of practices, what counts as an apology, what counts as correction, what counts as evidence, and so on. That’s why I talk about objective justification as part of the five ways we justify our beliefs, rather than treating justification as a purely formal relation between propositions.

Formal logic isn’t more important in the sense that it's some master key to knowledge, but it has a particular role. It doesn't supply the whole grammar of justification on its own, because most practices involve standards of relevance, error, correction, and defeaters that aren’t reducible to formal entailment. I’m happy to say, logic is indispensable for certain jobs, but it’s not the whole story, and it shouldn’t be treated as the model for every kind of justification. Finally, I make a distinction between pure logic and the logic of deductive and inductive reasoning.

I agree with your “some are and some aren’t.” The stopping point isn’t always as clearly delineated as a finish line. Take “there is an external world.” The point isn’t that this is a proposition we happen to feel very confident about. It’s that the skeptical demand to doubt it globally can’t be carried out without using the very grammar it is trying to suspend. To be able to doubt in a way that counts as doubt, you need criteria for checking, what would count as error, what would count as correction, and what would count as a defeater. But those notions are already world involving, they presuppose stable objects, re identification across time, public conditions of correction, and a contrast between appearance and reality. The global doubt doesn’t undercut a particular belief, it undercuts the practice conditions that give doubt its meaning. That’s why it’s hinge territory.

As for the cultural question, I think cultural conditioning in a limited sense is only natural (how much depends on the context). Different cultures can weight different routes, emphasize different norms of authority, etc. But that’s not the same as saying the grammar of justification is an empirical study of how this group thinks. The hinge point is structural, any community that has a practice of giving and asking for reasons, correcting mistakes, and distinguishing seeming from being will have some things that stand fast in order for the practices to function. So yes, there can be variation in which commitments play hinge roles and how they’re expressed, but the need for a bedrock background isn’t optional, and it’s not reducible to psychology. Just as the hinge background of chess (board and pieces) aren't optional if you're playing traditional chess.

A lot of recent hinge work has been shifting away from hinges as a magic bullet against skepticism (I don't claim it as a magic bullet, but I do think it shows how global skepticism misfires) and toward practice facing issues: trust, testimony, deep disagreement, and the way background commitments shape what even counts as evidence and correction. Coliva’s work on “hinge trust,” and the newer “social hinge epistemology” literature, are good examples.

Where I separate myself. Contemporary hinge epistemology often gets stuck arguing over one global theory of what hinges are, or over whether hinge frameworks really deliver the anti-skeptical payoff people want. You can see that pressure in recent work that argues the prominent “framework” formulations don’t succeed on their own terms. My paper isn’t mainly trying to win that internal hinge debate. I use hinges to mark the structural background of epistemic assessment, and then I focus on what actually disciplines claims to propositional knowledge inside language-games and forms of life.

My distinctive move: I start with JTB as a familiar grammar for propositional knowing, then I tighten the “J” condition with explicit guardrails, and I treat “+U” as a clarification of justificatory standing, not a fourth ingredient. The separate the person who can track mistake conditions, defeaters, and correction, from the person who can only recite the right conclusion. That’s also why testimony matters so much in my account, it’s not a lesser substitute for real knowledge, it’s a primary route that has to be disciplined by the same practice governed rules.

One more divergence, and I’m happy to own it. I treat hinges as arational and foundational, and I also think they carry ontological commitments. That’s stronger than a purely “methodological” reading, but it explains why global skepticism misfires, viz., it tries to call into question the very background that gives doubt, checking, and correction their sense. (Whether you want to call that a “solution” or a “dissolution” is secondary.)

I’m not addressing knowledge as a skill here, knowing how to build a cabinet, assemble a gun, play a tune, etc. That’s important, but it’s a different topic. My focus is propositional knowledge, claims of the form “I know that p,” where questions about justification, error, correction, defeaters, and Gettier pressures actually arise.

On that score, I agree that knowing and understanding aren’t inner objects we possess. They’re statuses we attribute within practices, and practices have standards for what counts as getting it right and what counts as correction. That’s why I talk about justificatory standing rather than private confidence or conviction.

Where the “+U” bites is this: in the propositional cases I’m focused on, understanding is competence with mistake-conditions. It’s being able to say what would count against the claim, what would defeat it, and what would correct it. That’s what keeps justification from collapsing into citation, authority, or lucky success.

And on authority: yes, propositional knowledge often rests on testimony and expertise, but authority isn’t self-certifying. It’s answerable to provenance, track record, and defeaters. If someone can’t track those, then there's an obvious problem, and that is exactly what many Gettier style worries are seeming to expose.

Continuing with paper...
Post #10

6.3 Defeater Screening
The third guardrail concerns challenge. Justification is not only a matter of placing a belief within support. It is also a matter of whether that standing survives relevant disruption. Defeater Screening names the fact that, within our practices, certain challenges count as undermining, and when they are present and undefeated the belief no longer has the standing required for knowledge.

This does not mean that one must answer every skeptical possibility. Our language-games do not treat every imaginable doubt as relevant. What matters is the kind of defeater that the practice itself recognizes as bearing on the standing of the claim. Sometimes a defeater shows that the grounds were false. Sometimes it shows that the route was misapplied, that the inference did not hold, that the perceptual situation was misleading, that the testimony lacked standing, or that a concept was applied outside its proper range. In each case the issue is not private reflection but public standing. A belief is not counted as knowledge when it stands under an undefeated challenge that the practice takes to be disqualifying.

Defeater Screening therefore belongs to the grammar of justification. It is part of what it is for a belief to stand properly within a practice, rather than to be held in a way that collapses under the first relevant counter-consideration.

No, the hinges are neither habitual nor pragmatic/convenient. But . . . "By which we take ourselves to be in contact" -- that's the rub. What is the practice being described which can result in contact with reality, aka, that which ontology studies? This question isn't limited to Witt-related thought, of course, but nor do the Wittgensteinian moves render it unproblematic.

Another way of saying it: To carry ontological commitments -- which I agree that hinges do -- is not to be part of what ontology studies or describes. There remains the question of the status of our epistemic practices as they relate to what we're pleased to call "reality." That is an Ur-metaphysical question, so possibly out of bounds for the Witt line of thought? — J

I think that’s the right pressure point, and I don’t want to dodge it with Wittgenstein slogans.
When I say by which we take ourselves to be in contact with reality, I’m not trying to smuggle in a practice that gives us access to Being. I mean something more minimal, i.e., that our ordinary epistemic practices already operate with a contrast between getting it right and getting it wrong, between correction and mistake, between appearance and reality. Hinges are part of what makes that contrast intelligible. They don’t secure contact with reality, they are the background commitments that make the very idea of contact, miscontact, error, and correction usable.

So, I agree that carrying ontological commitments is not the same as doing ontology. Hinge talk isn’t an ontology and it doesn’t settle ontology. What it does is clarify the foundational/bedrock commitments under which ontological discourse, or any discourse can even get off the ground. In that sense it’s neither merely a pragmatic convenience nor a metaphysical proof. It’s a claim about the grammar (Wittgensteinian grammar of course).

On the fundamental metaphysical question about the status of our practices in relation to reality: yes, that question remains. Wittgenstein doesn’t abolish it. What his line of thought does is block a certain way of posing it, the way that tries to demand a justification for the whole framework while still using the framework’s notions of justification, evidence, and correction. If you ask the Ur-question as a metaphysical project, fine. My point is just that it’s no longer an ordinary epistemic demand, and it can’t be answered by ordinary justificatory moves, because those moves presuppose the very standing-fast background under discussion.

I’d say it this way. Hinges have ontological commitments, and in that sense they touch ontology. But they don’t deliver ontology. They set the stage on which ontological arguments can be intelligible, and they explain why some global demands for justification misfire: not because reality is off limits, but because the demand is asking for a kind of validation that cannot be supplied without circularity or infinite regress.

I don't want to make the thread all about hinges, but I also don't want to not take questions about hinges.

My own view is different, but I think that's Wittgenstein's take as I interpret it.

A rule does not interpret itself. Yet we have rules that set up novel interpretations. Following a rule can involve treating something as if it were something more. The move is essentially to build a new language game on the back of another. And something like this seems implicit in a form of life. The whole remains embedded in human activity, in a form of life. — Banno

I agree with most of that.

Wittgenstein does think his approach bears on the foundations of mathematics: of course, the temptation is to imagine that the rule, or the proof, carries its own application and its own interpretation independent of what we actually do. A rule doesn't interpret itself, it's not an aside, it's aimed at that picture.

At the same time, youa'e right that we can introduce further rules that effectively stabilize new ways of speaking. We can take an earlier practice and add a counts as norm that extends it. In this sense, following a rule can include treating a construction as if it were something more, because we have adopted criteria that make that treatment correct within the extended game.

But the Wittgensteinian idea is that this isn't a metaphysical ascent to a realm of completed entities. It's a reworking of our practice (what we do), still embedded in human activity and a form of life. The novelty comes from what we now allow as a correct move, not from discovering a new kind of object behind the calculus.