• frank
    18.1k
    Sometimes knowledge enters philosophical consideration in the form of a propositional attitude. The SEP says it well:

    Propositional attitude reporting sentences concern cognitive relations people bear to propositions. A paradigm example is the sentence ‘Jill believes that Jack broke his crown’. Arguably, ‘believes, ‘hopes’, and ‘knows’ are propositional attitude verb and, when followed by a clause that includes a full sentence expressing a proposition (a that-clause) form propositional attitude reporting sentences. Attributions of cognitive relations to propositions can also take other forms. For example, ‘Jack believes what Jill said’ and ‘Jack believes everything Jill believes’ are both propositional attitude ascriptions, even though the attitude verb is not followed by a that-clause. Some philosophers and linguists also claim that sentences like ‘Jill wanted Jack to fall’, ‘Jack and Jill are seeking water’, and ‘Jack fears Jill’, for example, are to be analyzed as propositional attitude ascribing sentences, the first saying, perhaps, something to the effect that Jill wants that Jack falls, the second that Jack and Jill strive that they find water, and the third that Jack fears that Jill will hurt him. But such analyses are controversial. (See the entry on intensional transitive verbs.)

    Having a successful theory of propositional attitude reports is important, as they serve as a converging point for a number of different fields, including philosophy of language, natural language semantics, philosophy of mind, metaphysics, and epistemology.

    In this article, we examine attempts to deal with a puzzle about propositional attitude reporting sentences that was first posed by Gottlob Frege in his 1892. Subsequent literature has been concerned with developing a semantic theory that offers an adequate treatment of this puzzle. We present the main theories and note the considerations that count in their favor and some of the problems that they face.

    1. Frege’s puzzle
    — sep article on propositional attitude reporting

    I recently did a thread in Frege's puzzle, but I'd be happy to revisit it if anyone's interested.
  • Ludwig V
    2.2k
    To better understand the ready-existent regulations by which something operates is not the same as pigeonholing everything into rules of one’s own creation.javra
    The search for definitions is often a matter of codifying what we actually do. It is a very hard thing to do perfectly, partly because the rigidity that goes along with that can end up in conflict with the more flexible and dynamic practices of actual use.

    a mystic’s understanding of reality at large cannot be shared in the complete absence of JTB knowledge regarding this understanding, via which the understanding could then be convincingly communicated to others.javra
    There are various understandings of the world and some of the things in it that can't be communicated through propositional true/false knowledge. But there are other ways of communicating - poetry, pictures, music, dance.

    More mundanely, though, most understandings among adult humans in a society are commonly held by all individuals (e.g., the understanding of which side of the road to drive on).javra
    Surely that example is easy to communicate in common-place ways. What is harder to communicate by means of articulate rules is different. Curiously, how to use words is one of them. But how to be respectful or friendly are not like that, either.

    But consider how kids learn language: they must come to their own understanding regarding what words in their proper contexts signify. One cannot impart this understanding to children directly (in contrast to how a JTB can be directly imparted among adults), but can only lead the way toward it via affirmations and negations regarding what is correct. This until the understanding clicks.javra
    Yes, but you know when the understanding clicks because you know when the child is using the words correctly.

    JTB, on the other hand, will require a) belief (that is both true and endlessly justifiable in valid manners in principle), b) some measure of understanding, and c) awareness.javra
    Doesn't this show that all three are interwoven as different aspects of knowledge?

    The five justificatory routes identified in JTB+U illustrate this point.Sam26
    I was very pleased to see you include testimony. Because it enables us to pass on what we know It is critical to our practice of knowledge. We all stand on the shoulders of others and our society would be greatly impoverished if testimony were not an effective way of communicating it. However, accommodating it in the standard JTB framework is tricky. It requires acceptance of fallible justifications.

    In this way, JTB+U brings Wittgenstein’s therapeutic insight into constructive form: it dissolves confusions about “know” by looking at use, and it offers a framework that captures the grammar of knowledge as it is lived in our forms of life.Sam26
    Yes, I agree with that.

    Sometimes knowledge enters philosophical consideration in the form of a propositional attitude.frank
    I thought that knowledge just is an attitude to a proposition. In what other form could it enter into philosophical consideration? I think it is useful to see "know" and "believe" in the context of "think", "suppose", "imagine", "deny", "assert" and Frege's puzzle is indeed a puzzle.


    Some philosophers and linguists also claim that sentences like ‘Jill wanted Jack to fall’, ‘Jack and Jill are seeking water’, and ‘Jack fears Jill’, for example, are to be analyzed as propositional attitude ascribing sentences, — sep article on propositional attitude reporting
    Well, there's some truth to that. But I think that it misses the point and over-extends a useful idea. It would be a bit misleading, wouldn't it, to parse "I wish I had a red flower for a buttonhole" as expressing a positive attitude to the proposition "I have a red flower for my buttonhole"; the object of my positive attitude is the red flower, once it appears in my buttonhole.
  • Sam26
    3k
    I was very pleased to see you include testimony. Because it enables us to pass on what we know It is critical to our practice of knowledge. We all stand on the shoulders of others and our society would be greatly impoverished if testimony were not an effective way of communicating it. However, accommodating it in the standard JTB framework is tricky. It requires acceptance of fallible justifications.Ludwig V

    Much of our knowledge comes through testimony (books, lectures, person-to-person, etc). Testimony is often undervalued and misunderstood. Every area of study, including science, must rely on testimony. What's lacking is the knowledge of how (it's a skill) to evaluate and appreciate its value.

    Almost all justification is fallible, not just testimony. Why? Because most knowledge relies on probabilistic reasoning, including science.
  • Ludwig V
    2.2k
    Much of our knowledge comes throughSam26
    Yes, that's quite right. I think, though, that philosophers have always been more interested in how new knowledge is acquired, so tend to focus on that. What they don't pay enough attention to, in my opinion, is how important the spread of knowledge is and how dependent new knowledge is on knowledged that has already been acquired.

    Almost all justification is fallible, not just testimony. Why? Because most knowledge relies on probabilistic reasoning, including science.Sam26
    That's complicated. Some probabilistic reasoning is absolutely certain. The odds of a coin toss are exactly and without doubt 1/2. Empirical probabilities less so, although in practice they seem to work quite well. I don't know how reliable Bayesian probabilities are, but, given the difficulty of verifying them (in one-off cases), I set even less store by them. But note that probabilities have no meaning unless and until there are outcomes - at which point the probability becomes 1 or 0.
    Fallible justifications are a bit different. In many cases, no proper estimate can be made because the outcomes cannot be listed and weighed. In others, the fallibility is only a possibility and disappears when the outcome is known.
    Then there are the Wittgensteinian hinge propositions and other indubitable propositions. How far they stretch I am not clear. But it does seem likely that much of our everyday knowledge is more probabilistic than we like to recognize. (People treat the time their train leaves as a certainty, but they know full well that trains are often disrupted by one thing or another.) They ignore that because they can do nothing about it and it doesn't happen too often.
  • Sam26
    3k
    Some probabilistic reasoning is absolutely certain.Ludwig V

    That claim is self-contradictory under the way I’m defining these terms. Absolute certainty means knowledge held with 100% confidence, without the possibility of error. This is the domain of sound deductive reasoning, where a valid argument with true premises guarantees its conclusion—what we call a proof in the strict deductive sense. The word absolutely here signifies “without restriction” or “without qualification.”

    By contrast, probabilistic reasoning always carries qualification: no matter how small the probability of error, there remains some chance that the conclusion is false. Thus, probabilistic conclusions can approach certainty in degree, but they can never be absolutely certain in kind.
  • DifferentiatingEgg
    753
    The deepest conceit of philosophy: that metaphysics and epistemology are about the world as it is.

    Every epistemology carries within it a metaphysics, just as every metaphysics presupposes an epistemology. Philosophy often speaks of metaphysics as if it were an eternal architecture of reality, but “disorders” such as dementia or schizophrenia reveal that both epistemology and metaphysics rest on fragile, human scaffolding. When the mind deteriorates, the capacity to know and the categories of what is known collapse together: soap and marinara no longer belong to distinct orders, voices leak from thought into the world, self and other lose their boundary. What this shows is that metaphysics is not an independent order of being, nor epistemology a neutral method of knowing, but two faces of the same fragile ordering principle — a set of boundaries the mind must uphold to make sense of experience. When those boundaries dissolve, what we call “absurd” is simply lived reality; what we call “truth” is revealed as a maintenance project of cognition itself.

    Yet epistemology and metaphysics both are really about the world as the mind can sustain. Consequently not about the world as is.
  • Sam26
    3k
    The problem with your account is that it tries to dismiss epistemology while simultaneously advancing an epistemological thesis. To say that “truth is a maintenance project of cognition” is itself a claim about the nature of truth and knowledge, exactly the kind of claim epistemology exists to evaluate. In other words, you rely on the very framework you’re trying to undermine. Pointing to mental illness as evidence doesn’t solve this contradiction either, because interpreting dementia or schizophrenia as “revealing” something about knowledge already presupposes a stable epistemic standpoint from which such judgments can be made. If all categories collapse with the mind, then the very distinction between “collapse” and “order” also collapses, leaving you with no grounds to assert your conclusion.

    What your view really demonstrates is not the futility of epistemology, but its necessity. The fact that cognition can fail does not mean that truth is nothing more than a maintenance project; it means that human beings sometimes lose their grip on truth. To confuse the breakdown of knowledge with the nature of knowledge itself is like confusing a malfunctioning compass with the nonexistence of north. Far from undermining epistemology, your example underscores why we need it: to distinguish between distorted perception and genuine understanding, between the fragile scaffolding of a disordered mind and the enduring structures of reality.
  • Ludwig V
    2.2k

    Yes, I understand that position. I won't pick it apart here.

    I wanted to point out that mathematical probability is mathematically certain. It has uncertain elements, but even there, the degree of uncertainty is certain. We can consider coin-tossing as an abstract structure and see the implications of the rules, without any appeal to empirical probabilities. So the statement "There is a 50% probability that the coin will land heads" and "The coin will land either heads or tails" are mathematically certain.

    All that changes when we consider empirical probabilities. Here, nothing is certain and the language changes to reflect that. However, there are some elements that need to be regarded as certain. In an actual game of coin-tossing, we do need to be certain of the result of each toss; if there are any uncertain outcomes (the coin lands and stays on its side, or turns into a bubble), they need to be discounted, or included as a third possible outcome. (Where we are basing our estimates of life expectancy on past data, that, too, needs to be certain.) Probabilistic reasoning needs some things to be (or be regarded as) certain, though never mathematically certain.

    Bayesian probability is different again, since it assigns a probability to a single outcome. I'm not at all clear what meaning can be attached to that. But it does seem to me that the outcome of the experiment must be (treated as) certain, or there is no result and the prediction fails.
  • DifferentiatingEgg
    753


    You’re right that to say “truth is a maintenance project of cognition” is itself an epistemic claim. But that doesn’t undermine my point... it reinforces it. The fact that I can’t step outside the framework of justification to make my claim is precisely what I mean when I call truth a “maintenance project.” To describe truth is always to participate in it, never to stand above it.

    When I point to dementia or schizophrenia, I’m not saying they reveal “the essence of knowledge” from some Archimedean standpoint. I’m saying their breakdowns highlight the contingency of the boundaries we ordinarily take for granted. You call this “confusing the breakdown of knowledge with the nature of knowledge.” I’d say: the breakdown discloses the nature. Knowledge is not a mirror of eternal structures; it’s the fragile activity of maintaining categories against the ever-present possibility of their collapse. We can never stand outside the scaffolding of the framework and measure it. We can only ever uphold it from within, patching and justifying where need be.

    In that sense, I’m not dismissing epistemology but radicalizing it: epistemology is not the neutral arbiter between “distorted” and “genuine,” but itself part of the scaffolding, an instrument of maintenance that only works for as long as the categories hold. Our claims about knowledge are bound up in the same fragile maintenance they describe.
  • Sam26
    3k
    The following is a point of clarity.

    There are four uses of certainty that we need to be aware of, and they are the following:

    1. Subjective certainty
    This is the psychological sense: conviction, the felt impossibility of doubt. I may be subjectively certain that a friend will keep a promise, or that my team will win. But conviction alone is not knowledge, because one can be subjectively certain and still wrong.

    2. Hinge certainty
    These are the background beliefs that stand fast and frame our practices of doubt and justification. “The world has existed for a long time,” “I have two hands,” “Memory is generally reliable.” They are not justified in the ordinary way, nor are they really doubted in practice. They are arational bedrock, conditions of doubt and inquiry.

    3. Epistemic certainty
    This is when a belief is not only true and believed, but also justified and undefeated within the relevant practice. It is defeater-resistant knowledge. Epistemic certainty is what JTB+U aims to secure: not infallibility, but a belief whose justificatory structure is strong enough to withstand error-signals in its domain.

    4. Absolute certainty
    Here belong the logical and moral necessities: “2+2=4,” “A thing cannot both be and not be in the same respect,” or “murder is always wrong.” These are not contingent truths but necessary ones, either analytic or rooted in moral grammar. Doubt here would not be corrigible error but conceptual breakdown.

    So:
    Subjective certainty = conviction.

    Hinge certainty = arational bedrock.

    Epistemic certainty = knowledge in the JTB+U sense.

    Absolute certainty = logical or moral necessity.
  • Ludwig V
    2.2k

    All right. That's very clear.

    By contrast, probabilistic reasoning always carries qualification: no matter how small the probability of error, there remains some chance that the conclusion is false. Thus, probabilistic conclusions can approach certainty in degree, but they can never be absolutely certain in kind.Sam26
    So where does probabilistic reasoning fit? An example of a conclusion that can approach certainty in degree, but never be absolutely certain in kind. Bear in mind, that probability is, by definition, defined by an outcome, of which the probability, by definition, is 1 or perhaps 0. (I'm not saying the outcome always has to happen, just that each probability defines an outcome.)
  • Sam26
    3k
    So where does probabilistic reasoning fit? An example of a conclusion that can approach certainty in degree, but never be absolutely certain in kind. Bear in mind, that probability is, by definition, defined by an outcome, of which the probability, by definition, is 1 or perhaps 0. (I'm not saying the outcome always has to happen, just that each probability defines an outcome.)Ludwig V

    Probabilistic reasoning does not yield absolute certainty (sense 4: logical/moral necessity). It is not the kind of certainty we find in “2+2=4” or “murder is always immoral,” where denial collapses into contradiction. Instead, probability operates in the register of epistemic certainty (sense 3: defeater-resistant knowledge in practice). A high probability, backed by justification and understanding, can approach a state where we rightly treat the belief as knowledge, even though it remains in principle corrigible.

    This distinction is crucial. To say “I am certain at the 99.9% confidence level that the medicine will work” is not to claim an absolute necessity. It is to claim a belief that has survived defeater screening, cross-checked by testimony, reasoning, and sensory evidence, and that functions with the kind of stability that makes action rational. In other words: probabilistic reasoning belongs in the realm of practical epistemic certainty, not absolute certainty.

    Your remark about outcomes is well taken: once an outcome occurs, its probability collapses to 1 (happened) or 0 (didn’t happen). But notice that before the fact, what matters for knowledge is not the metaphysical status of the outcome but the justification of belief leading up to it. The grammar of “know” ties us to what is justifiable before the outcome is fixed. This is why probabilistic reasoning fits comfortably within JTB+U: truth (the event does or does not occur), belief (the agent commits to its likelihood), justification (the probabilistic model and evidence), and understanding (grasp of how probabilities function).

    So the answer is: probabilistic reasoning cannot reach absolute certainty, but it can yield epistemic certainty robust enough for knowledge in practice. That’s why we entrust planes to aeronautical engineering or treatments to clinical trials. We are not misusing “know” when we say, “We know this drug works,” even though the claim is probabilistic. We are speaking in the epistemic register, not the absolute one.
  • Sam26
    3k
    Just to reiterate...

    Probability and the Varieties of Certainty

    One common source of confusion in epistemology is the place of probability. Some treat probabilistic reasoning as a weaker substitute for knowledge, as though it can never rise above opinion. Others mistakenly assume that probability can deliver the same kind of necessity we find in logic or mathematics. Both views rest on a failure to distinguish the varieties of certainty.

    Recall the four senses:

    Subjective certainty — conviction, the felt assurance of belief.

    Hinge certainty — the arational bedrock that makes doubt and inquiry possible.

    Epistemic certainty — defeater-resistant knowledge in practice, stable enough to guide action.

    Absolute certainty — logical or moral necessity, where denial collapses into contradiction.

    Probabilistic reasoning clearly does not yield absolute certainty. No degree of evidence can turn an inductive conclusion into a logical necessity. To say “this drug has a 95% chance of success” is not to assert that its success is absolutely necessary. The grammar of probability already places it outside the category of absolute truths like “2+2=4.”

    But this does not mean probability is epistemically weak. Probabilistic reasoning is one of the chief ways in which we arrive at epistemic certainty. When a belief is supported by convergent evidence — statistical studies, repeated trials, predictive accuracy — it can withstand defeaters and function with the kind of reliability that makes action rational. We rightly say we know airplanes will stay aloft, even though our confidence is grounded in probability, not logical necessity. The strength of the claim lies not in its absoluteness but in its resilience across error-signals and justificatory routes.

    It is also important to note the role of outcomes. Once an event occurs, its probability collapses to 1 or 0. The coin either lands heads or it does not. But knowledge-claims are not made after the fact; they are made in the stream of life, before outcomes are settled. What matters epistemologically is not the metaphysical status of the event but the justification for belief leading up to it. Probability is the language we use to track that justification, and when paired with understanding, it satisfies the conditions of JTB+U.

    In this light, probability is not a lesser form of knowing but one of the principal ways human beings secure knowledge. We live in a world where outcomes are not always transparent, but where patterns and regularities can be detected and acted upon. Probabilistic reasoning translates those patterns into degrees of epistemic certainty. It belongs not to the realm of absolute necessity but to the realm of practice — the very place where JTB+U is meant to operate.
  • Joshs
    6.4k


    Sam26

    You’re right that to say “truth is a maintenance project of cognition” is itself an epistemic claim. But that doesn’t undermine my point... it reinforces it. The fact that I can’t step outside the framework of justification to make my claim is precisely what I mean when I call truth a “maintenance project.” To describe truth is always to participate in it, never to stand above
    DifferentiatingEgg

    I’m with you all the way here. I would just add that truth is not a unitary concept but changes its sense according to the context of its use. Human beings are in the business of construing recognizable patterns in the swirl of experience and then drawing from our memory of those patterns to anticipate new events. This is a richer and more fundamental process than what is contained within the thin and derivative notions of propositional truth and justification. We don’t primarily make truth-epistemic claims, we project expectations and wait to see if the way events unfold do it in a way that is more or less inferentially consistent with our anticipations. Events will never duplicate those expectations, so even when our hunches are verified, we must adjust these patterns to accommodate the novel features of the events we recognize. This is not epistemology, it is context and situation-based sense-making.
  • Sam26
    3k
    I think we’re converging on a similar point. I would agree that “truth” does not wear a single face. Its criteria shift depending on the language-game: in a courtroom truth is tied to testimony and records, in science it is tied to predictive success but also to the testimony and documentation that communicate, test, and replicate those predictions, and in mathematics it is tied to logical necessity. To borrow Wittgenstein’s term, these are family resemblances rather than a unitary essence.

    Where I’d want to add a note of caution is that the factivity of truth still matters across those contexts. However we construe it, “p” being true always implies that things are as “p” says they are. Otherwise we lose the very grammar that distinguishes knowledge from conviction.

    On the question of pattern recognition: I see this as closely related to what Wittgenstein called hinges, or what I sometimes call hinge-beliefs. These are certainties that do not themselves stand in need of justification but make justification possible. For example, we do not reason our way to believing that objects persist when out of sight, or that other human beings have inner lives, these are taken for granted in our dealings. They are beliefs in the broader sense, which includes both propositional claims (“The earth existed before my birth”) and non-propositional, prelinguistic certainties (the infant’s trust in the caregiver, or the body’s implicit grasp that the ground will hold). Pattern recognition provides much of the raw material for these hinges: it is how we find stability in experience, and those stable expectations are what make reasoning and justification possible.

    And here is where my JTB+U framework comes in. Propositional truth on its own is “thin,” as you put it. What matters is how it is embedded in justification and uptake, how belief is connected to truth in ways that others can test, and how the agent understands what the claim involves. So I wouldn’t discard propositional truth, but enrich it. It becomes one thread in the weave of truth, belief, justification, and understanding, all grounded in the certainties that stand fast in our forms of life.
  • Joshs
    6.4k
    ↪JoshsI think we’re converging on a similar point. I would agree that “truth” does not wear a single face. Its criteria shift depending on the language-game: in a courtroom truth is tied to testimony and records, in science it is tied to predictive success but also to the testimony and documentation that communicate, test, and replicate those predictions, and in mathematics it is tied to logical necessity. To borrow Wittgenstein’s term, these are family resemblances rather than a unitary essence.

    Where I’d want to add a note of caution is that the factivity of truth still matters across those contexts. However we construe it, “p” being true always implies that things are as “p” says they are. Otherwise we lose the very grammar that distinguishes knowledge from conviction.
    Sam26

    If we’re sticking with the later Wittgenstein we want to be careful here ( otherwise feel free to ignore this :grin: ).
    The grammar of truth isn’t defined across instances connected by family resemblances. It is defined by the particular instantiations within the larger family. We dont consult an already given conceptual grammar (“p” being true always implies that things are as “p” says they are) and then apply it to a family of instances. There is always something particular we are doing with ‘“p” being true always implies that things are as “p” says they are’, and this grammar doesnt just give us unique criteria but a unique sense of meaning of the phrase ‘ “p” being true always implies that things are as “p” says they are’.

    Your wording suggests “truth” is a concept with multiple “faces” or modes of operation. For Wittgenstein, this risks hypostatizing truth as a thing. He would prefer to dissolve that urge: “truth” isn’t an object with different guises, but simply a word whose uses vary across practices. And every particular use, even within a single “practice,” can shift the sense. If meaning is use, then it’s not just different practices (“law,” “science,” “math”) that set the sense of “truth,” but also the fine-grained contexts within those practices.
  • Sam26
    3k
    I think you’re right to warn against reifying “truth” into a thing with different guises. Wittgenstein’s reminder is well taken: meaning is use, and every particular use of “truth” has to be understood within its practice and even within its immediate context. To that extent, I agree we don’t apply a pre-given conceptual grammar from outside.

    That said, I would still hold that in the epistemic use of “true,” there is a minimal grammatical constraint that runs through its particular instantiations. However much the criteria differ — testimony in law, predictive success in science, proof in mathematics — the word “true” here always functions in contrast to “false,” and that contrast implies that the proposition fits the way things stand. This is not to hypostatize truth as an object with “faces,” but to recognize the mirroring role that the grammar of epistemic truth presupposes.

    So I wouldn’t want to treat “truth” as having an essence across all uses. I agree that outside epistemology — in poetic, moral, or expressive speech — the sense of “truth” shifts without necessarily carrying factivity. But within epistemology, its factive core seems indispensable. Without it, we lose the very grammar that lets us correct, doubt, or justify claims. So my position is: meaning is use, but among the uses of “truth” there is a distinctive epistemic family where factivity is non-negotiable, even though its criteria are locally defined.
  • Sam26
    3k
    One of the recurring problems in AI research is the ambiguity of “knowledge.” Systems can be designed to store facts, to process probabilities, or to simulate reasoning, but what it means for an AI system to know something is rarely clarified. Engineers often collapse knowledge into probability scores, while philosophers insist that knowledge requires more than high confidence. This is precisely where a Wittgensteinian extension of the classical JTB model — JTB+U — could offer guidance.

    The first benefit lies in distinguishing different senses of certainty. Human beings navigate between convictional certainty (“I know she loves me”), epistemic certainty (“I know the bridge will hold”), and absolute certainty (“2+2=4”). An AI system that fails to recognize these differences risks either inflating its probabilistic outputs into necessity or downplaying genuine knowledge as “only probable.” Incorporating the grammar of certainty into AI reasoning could allow systems to represent not just confidence levels but also kinds of certainty — treating mathematical truths differently from empirical predictions, and both differently from convictional expressions.

    The second benefit is the awareness of hinges. Wittgenstein insisted that justification does not go on forever. At some point there are certainties that “stand fast”: that the external world exists, that memory is generally reliable, that other people have inner lives. These are not optional premises; they are conditions of sense. An AI designed without hinge-awareness will endlessly seek justification where none is required, or worse, collapse into regress. A hinge-sensitive AI could be programmed to recognize its own bedrock assumptions and to operate within them while remaining open to correction at higher levels. This would make its reasoning both more efficient and more human-like.

    The third benefit arises from the “+U” in JTB+U — understanding. Current AI systems can store truths, assign probabilities, and even generate justifications of a sort, but they often lack uptake. They can state that Paris is the capital of France, but they do not consistently grasp what follows from that claim, when to apply it, or how to detect when an error-signal demands revision. Embedding “understanding” as a criterion would shift AI development away from mere statistical prediction toward competence in language-games: the ability to apply concepts correctly across cases, to give and ask for reasons in context, and to adapt when defeaters arise.

    Finally, JTB+U emphasizes contextual justification. Testimony functions differently in science than in law; sensory evidence is weighed differently in everyday life than in a laboratory. An AI that can track these language-game differences will be able to tailor its justificatory standards to the domain in which it operates, rather than treating knowledge as homogeneous. This would prevent both under- and over-reach: no longer dismissing courtroom testimony as “unscientific,” while not confusing legal plausibility with empirical law.

    In short, JTB+U offers a conceptual framework that could reshape AI epistemology. It would push systems beyond probability scores toward a layered understanding of certainty; beyond blind data-processing toward hinge-awareness; beyond rote recall toward genuine uptake; beyond one-size-fits-all reasoning toward practice-specific justification. If humans are to design AI that integrates into our forms of life, the grammar of “knowledge” must be respected. Far from being a mere philosophical abstraction, epistemology in this Wittgensteinian key could guide the next generation of intelligent systems.
  • Sam26
    3k
    How Epistemology Can Help Align AI

    A Wittgensteinian JTB+U Framework for Responsible Intelligence

    Introduction: Fear Without Framework
    One of the greatest public anxieties about artificial intelligence is not simply that it makes mistakes, but that it makes them with the wrong kind of confidence. AI systems already generate falsehoods while sounding certain, apply reasoning across domains without context, and produce outputs we cannot always distinguish from knowledge. People worry that such systems may spin out of control — not only because they are powerful, but because we lack a shared standard for when their outputs count as genuine knowledge.

    What if part of the solution lies not only in engineering but in epistemology? Philosophy has wrestled for millennia with the conditions of knowledge. The classical model of Justified True Belief (JTB) remains the backbone: one knows something if it is true, believed, and justified. Yet this model alone is too thin for the complexities of modern reasoning. In my work I extend JTB with a fourth element — Understanding (U) — and frame it through Wittgenstein’s later philosophy. The result, JTB+U, offers a grammar of knowledge that could help discipline AI reasoning and reassure human users.

    The Core: JTB+U
    The JTB model works well because it matches our practices: we do not call someone knowledgeable if they voice a lucky guess, nor if they cling to a false belief. We demand truth, belief, and justification together. My addition of +U (Understanding) makes explicit what was always implicit: that knowledge also requires uptake. A system must not only produce a true justified belief but must also grasp the concept well enough to apply it competently across contexts.

    In human life, understanding is what allows us to use knowledge in reasoning, explanation, and correction. Without it, justification can be hollow or accidental. For AI, adding +U means more than citing sources or probabilities — it means showing conceptual grasp: when to apply a claim, what follows from it, and when it fails.

    Certainty: Degrees and Kinds
    AI systems often present outputs as if they were absolute, when in fact they are probabilistic. Here Wittgenstein’s analysis of certainty helps. We can distinguish at least four senses:

    Subjective certainty — conviction or felt assurance.

    Hinge certainty — arational bedrock that makes doubt possible (e.g., the world exists).

    Epistemic certainty — defeater-resistant knowledge in practice, stable enough to guide action.

    Absolute certainty — logical or moral necessity, where denial collapses into contradiction.

    Probabilistic reasoning belongs to epistemic certainty: not infallible, but robust against correction and sufficient for rational action. By contrast, mathematics and logic express absolute certainty in kind, not in degree. An AI system grounded in this grammar would know how to communicate not only confidence scores but what kind of certainty is at stake. That would go far in preventing false impressions of infallibility.

    Hinges: Bedrock for AI Reasoning
    Another danger is regress: how do we know that an AI’s reasoning is not flawed at some hidden level? Wittgenstein’s answer was that justification does not go on forever. At some point there are certainties that “stand fast”: basic beliefs not themselves justified but making justification possible.

    Humans take for granted that the world existed before our birth, that memory is generally reliable, that other people have minds. An AI system, too, must operate with hinge assumptions: that its arithmetic is consistent, that its data streams are genuine, that communication has meaning. If AI could make such hinges explicit, humans would better understand what the system presupposes, and where corrections could intervene. Transparency at the hinge level could dissolve much of the “black box” fear.

    Language-Games: Contextual Justification
    Wittgenstein also reminded us that justification is not timeless but practice-indexed. What counts as evidence in a courtroom differs from what counts in a laboratory. Human beings navigate these shifts in grammar fluidly; AI systems do not.

    Here the JTB+U framework highlights five primary routes of justification:

    1) Testimony

    2) Logic (inductive and deductive reasoning)

    3) Sensory experience

    4) Linguistic training

    5) Pure logic (boundary-setting only)

    Each route is weighted differently across practices. In law, testimony and records dominate; in science, prediction and replication; in daily life, sensory access and trust. An AI system that tracks these contextual differences could avoid misfires — no longer treating courtroom testimony as “unscientific,” nor confusing legal plausibility with empirical law. Context-sensitive justification is not optional; it is essential for responsible reasoning.

    Understanding as Uptake
    Perhaps the most radical demand of JTB+U is that knowledge requires understanding. For AI, this entails more than generating correct statements. It requires showing uptake:

    * applying concepts correctly across cases,

    * responding to defeaters,

    * grasping the inferential consequences of claims,

    * recognizing when claims no longer fit.

    Without +U, AI outputs risk remaining surface mimicry — statistical echoes of human reasoning. With +U, they could approximate conceptual competence: the ability to play the language-game of giving and asking for reasons.

    Why Epistemology Matters for AI Safety
    Technical controls are vital, but without an epistemological grammar they risk being blind. JTB+U offers a framework for:

    * Transparency — AI can show what kind of certainty it claims.

    * Trust — users can see hinge assumptions and justificatory routes.

    * Humility — AI can learn to say “I don’t know” when defeaters arise.

    * Alignment — systems become participants in human language-games, not alien oracles.

    AI will never be “safe” if we cannot tell when it knows what it says it knows. A Wittgensteinian epistemology does not solve all engineering problems, but it clarifies the standards by which outputs should be judged. If adopted, it could help AI evolve from a powerful but opaque prediction machine into a responsible reasoning partner.

    Conclusion: Epistemology in Practice
    Philosophy is often accused of irrelevance, but here it has something urgent to offer. The JTB+U framework, enriched by Wittgenstein’s grammar of hinges and language-games, provides exactly the clarity needed at the frontier of AI. It disciplines knowledge-claims, distinguishes types of certainty, contextualizes justification, and demands understanding. If we are serious about aligning AI with human values, we must first align it with the human grammar of knowledge. Epistemology, far from being a museum piece, may be one of the keys to AI’s safe future.
  • Sam26
    3k
    The Wittgensteinian Toolkit

    If metaphysics provides the horizon of intelligibility, Wittgenstein’s later philosophy supplies the instruments for navigating within it. His method is therapeutic rather than speculative: instead of offering new theories of knowledge, it clarifies how the word know functions in the practices where sense and doubt are learned. The tools are deceptively simple—language-games, grammar, forms of life, family resemblance, and hinges—but they work together, reinforced by the ideas of rule-following, public criteria, and the river-bed imagery of On Certainty. Taken as a whole, this toolkit grounds epistemology not in metaphysical essence but in lived intelligibility.

    Language-games: where “know” has meaning
    Wittgenstein begins with the recognition that meaning arises from use. Words are not labels for private entities but moves within human activities. To call something knowledge is not to identify a mysterious inner state but to perform a specific act inside a game with shared rules—asserting, challenging, correcting, teaching. Each domain has its own grammar of knowing: the scientist cites data, the witness swears an oath, the child repeats a lesson. There is no single essence of know; the unity lies in family likeness across these games. By examining language-games, we shift the epistemological question from “What is knowledge?” to “How do we use ‘know’ here?” The philosopher’s task becomes descriptive rather than legislative.

    Grammar: the logical order that frames sense and doubt
    Grammar, for Wittgenstein, is not syntax but the underlying logic that determines what counts as sense. It tells us what it means to justify, to verify, to doubt. Grammar sets the boundaries of intelligible talk: we can meaningfully say “I know this is a hand,” but not “I know I am in pain.” The latter confuses private experience with public criteria. Grammar thus replaces the metaphysician’s search for foundations with attention to rule-governed use. To grasp a concept grammatically is to know what would count as applying it correctly, what would make its application nonsensical, and how mistakes show themselves. Epistemology becomes a study of conceptual grammar—the order that lets sense and error exist at all.

    Forms of life: the public foundations of justification
    Every grammar stands on a background of practices, gestures, and agreements that are not themselves justified but simply enacted. These are our forms of life: the biological and cultural patterns within which reasoning has meaning. Justification presupposes such a background. We can question a witness, but not the practice of questioning itself; we can test a thermometer, but not the institution of measurement that makes testing possible. Forms of life are not propositions but the lived matrix of criteria. They explain how justification can be public without being infinite: our shared ways of acting already contain the standards that make giving reasons possible.

    Rule-following and public criteria
    Within a form of life, rules acquire authority through training and correction. To follow a rule is not to consult an inner diagram but to participate competently in a practice where others can see and correct our moves. This dissolves the picture of justification as an internal state of certainty. What shows that I understand a rule is not introspection but my ability to go on correctly when circumstances vary. Knowledge therefore carries a social dimension: it is demonstrated in action and confirmed by public criteria. Wittgenstein’s reminder—there is no inner pointing to meaning—cuts against the Cartesian idea that knowing is primarily a private event. The grammar of know is visible in use, not hidden in consciousness.

    Family resemblance: unity without essence
    Philosophers often seek the single property that makes all instances of knowledge what they are. Wittgenstein advises a different posture: notice the overlapping similarities among uses instead of chasing a universal definition. Just as “games” share no one essence but display a network of resemblances—competition, rules, play—so too “knowledge” unites diverse activities by kinship rather than identity. This insight loosens the grip of essentialism that has haunted epistemology since Plato. It allows JTB, later extended to JTB+U, to function as a grammatical model rather than as a metaphysical claim. Its durability lies in family likeness across contexts, not in a timeless form.

    Hinges and the river-bed: the arational background of reason
    On Certainty adds the final and most radical instrument. Every act of justification presupposes propositions that stand fast—hinges that make doubt meaningful. These are not beliefs we know but certainties we act from. They form the river-bed in which the current of reasons flows. To question them all at once would be to lose the very distinction between sense and nonsense. Examples include “There is an external world,” “Objects persist,” “Words have stable meanings,” and, in your metaphysical horizon, “Consciousness is the condition for any appearing at all.” Hinges are arational, not irrational; they lie beneath justification, providing the stage upon which justification can occur.

    This insight explains why epistemic regress ends without circularity. When reasons run out, we do not reach an arbitrary stopping point but the practical background that makes reasoning possible. To call something a hinge is to mark the transition from saying to doing, from proposition to practice. Wittgenstein’s point is grammatical: the verb to know presupposes a contrast with possible doubt; remove that contrast, and the word loses its function. Hinges, then, are the conditions of meaningful knowledge, not propositions we could ever justify or refute.

    Doubt as practice-bound
    Because hinges define the limits of intelligible doubt, skepticism must itself be seen as a language-game. Doubting that the external world exists or that one has a body is not a deeper form of inquiry but a misuse of grammar—like playing chess without a board. Doubt has sense only where the framework of certainty stands fast. This reframes epistemology: instead of searching for indubitable foundations, we describe the hinges that already stabilize our reasoning. Certainty here is not a psychological feeling but a logical role. It anchors rather than concludes.

    The toolkit as re-grounding for epistemology
    When these instruments work together, they convert epistemology from a quest for absolute justification into a grammar of responsible use. Language-games reveal diversity; grammar maps the logic of sense; forms of life show why justification is public; rule-following and criteria keep meaning from collapsing into subjectivity; family resemblance preserves unity without essence; and hinges secure the background that makes any of this possible. The result is an epistemology both humbler and stronger: humbler because it recognizes limits, stronger because it grounds knowledge in the lived regularities that precede theory.

    Wittgenstein’s method thus complements rather than replaces the classical JTB framework. JTB describes the explicit structure of epistemic accountability; the toolkit shows the soil in which that structure grows. Together they transform the question of knowledge from a search for metaphysical foundations into an examination of the practices that sustain sense. When we later add understanding as the fourth component, the +U will draw directly on these insights: understanding is nothing other than fluency in a language-game, mastery of its grammar, and attunement to its hinges.
  • Janus
    17.6k
    The "U" is a given—without understanding nothing gets off the ground in the first place, so I don't see how it adds anything when it is always already implicit in the JTB.
  • Sam26
    3k
    It’s true that understanding (U) has always been presupposed in talk about knowledge, but presupposition is not the same as articulation. Classical JTB leaves “understanding” implicit, folded into justification as though the two were indistinguishable. Yet practice shows the difference matters. One can have reasons without grasping how those reasons work, just as one can repeat a proof without understanding its logic. The “+U” is not a new ingredient added to JTB; it makes explicit a condition that was always there but often overlooked.

    In this sense, JTB+U performs a Wittgensteinian clarification: it dissolves the illusion that justification alone guarantees comprehension. “U” distinguishes genuine justification from parroting, algorithmic correctness, or social conformity. Philosophically, that difference is now urgent—especially in an age where machines can simulate justification without understanding.

    This is important, because it's easy to suppose your point is correct.
  • Sam26
    3k
    Let's add more depth: You are correct that understanding is already implicit in JTB. The classical model could never function without it, for to justify a belief is already to grasp the connection between reasons and conclusion. Yet in most philosophical treatments this grasp is left unarticulated, as though understanding were guaranteed whenever justification appears. My point in making it explicit is not to alter the structure of JTB but to bring to light what it silently depends on. Understanding is the hinge upon which the whole framework turns, the capacity to follow reasons as reasons, to see how evidence supports a claim rather than merely to repeat that it does. Without that inner relation, justification becomes mechanical and belief drifts toward imitation.

    When we examine how the word know functions in our language-games, we see that understanding is built into its grammar. To say “I know that p” normally implies that I can use p rightly in related contexts, that I see its sense and can extend it coherently. We withdraw the word know when this fluency is missing, as when someone parrots a theorem or repeats an argument they do not comprehend. Thus, understanding is not an optional embellishment but part of what gives the concept of knowledge its life. JTB has always assumed it, much as a door presupposes its hinge even when the hinge is unseen.

    Making that hinge explicit matters because it prevents epistemology from mistaking outward form for inner mastery. Classical analyses of justification tend to focus on external adequacy: the quality of reasons, the truth of premises, the logical connection among propositions. But none of these suffice if the knower lacks comprehension of how those reasons work. The +U draws attention to this internal dimension, protecting JTB from its own abstraction. It ensures that the model remains tied to lived competence rather than to formal correctness alone.

    From a Wittgensteinian perspective, this move simply applies his method to epistemology itself. By describing how know operates in our practices, we notice that understanding is already part of its use. Knowledge is shown in one’s ability to go on rightly when the explicit rule runs out, to recognize when an argument holds and when it does not. The philosopher’s task is not to add new parts to the machine but to illuminate what makes it run. In this light, JTB+U is not a new engine but a clearer description of the one we have always been using.

    Naming understanding thus gives epistemology self-awareness. It acknowledges the quiet condition that makes justification possible and holds belief in contact with truth. To call the model JTB+U is to make explicit the hinge on which it already turns. The addition changes nothing in principle, yet it changes everything in practice: it restores to the concept of knowledge the depth and accountability that modern epistemology, in its fixation on external form, too easily neglects.
  • Sam26
    3k
    I'm currently writing a paper to make more explicit my view on the epistemology of JTB+U, and the following is a portion of that paper.

    The endurance of the classical triad lies not in philosophical inertia but in functional necessity. Wherever human beings act together, we rely—often without notice—on the grammar that JTB names. Truth secures contact with reality; belief marks personal assent; justification connects the two through publicly defensible reasons. Remove any one element and the practice collapses. The model persists because its logic is lived before it is theorized. Science, law, and ordinary life all enact its pattern, however differently they speak.

    Consider science first, not as an institution but as a disciplined form of life. A physicist claims to know that a certain particle has a particular charge. What makes that claim knowledge rather than opinion is not conviction but the convergence of truth and justification. The statement corresponds to how the world behaves; the methods used to reach it—calibration, replication, peer review—provide the warrant that secures belief against mere luck. The scientist’s belief, her internal assent, is indispensable but not decisive. If later evidence overturns the finding, we say that the community was mistaken, not that belief altered truth. This readiness to correct itself shows the health of the practice: knowledge here is corrigible without being arbitrary.

    In science, JTB functions as a structure of accountability. Truth guards against fantasy; justification institutionalizes humility. A result that cannot be checked or replicated fails the justificatory test no matter how confidently it is asserted. Likewise, a correct result reached by accident—a lucky guess or an uncalibrated instrument—fails the epistemic test because justification is missing. These boundaries mirror the grammar that JTB codifies: luck without reason is not knowledge, and reason without truth is illusion. The laboratory simply dramatizes what ordinary cognition already presupposes.

    The same structure governs law, though expressed in another language-game. A jury must decide whether an accused person committed a crime. The verdict involves belief, but that belief counts as knowledge only when tethered to evidence that meets public standards—cross-examination, corroboration, admissibility. Truth is represented by the facts as they occurred; belief is the jurors’ collective assent; justification is the process that links the two—the rules of evidence and reasoning that make the verdict intelligible. The system acknowledges the fallibility of justification through appeals and retrials, mechanisms that reflect the same epistemic humility found in science. Even where human judgment replaces experiment, the grammar remains constant: knowledge requires a claim to be both true and justifiable to others.

    Legal reasoning also shows why knowledge must be public to be binding. The witness who swears to tell the truth invokes all three conditions. Truth corresponds to what occurred; belief is the sincerity of the witness; justification is the evidentiary frame that renders the testimony credible. A statement may be true yet unjustified if it cannot be corroborated, or justified yet false if the witness is mistaken. These permutations display the interlocking checks that give the concept its durability. We could abolish juries or rewrite evidentiary codes, but any workable system of accountability would still presuppose the same triad.

    Outside formal institutions, JTB continues to guide everyday reasoning. When I say that I know the bus leaves at eight fifteen, I am implicitly asserting that the schedule is accurate (truth), that I believe it (assent), and that I have grounds—experience, confirmation, or testimony—to warrant the claim. If the bus departs at eight thirty, I was mistaken, not lying. The grammar of correction—“I thought I knew”—shows that knowledge is a normative status, not a psychological state. We retract know precisely when one of its components fails.

    Such examples reveal that JTB is not a theoretical invention but a description of how epistemic life is already organized. Each time we distinguish between “He believes it” and “He knows it,” we invoke this grammar. Justification serves as the hinge between subjective conviction and objective warrant, allowing communities to coordinate understanding without collapsing into solipsism. The continuity of this pattern across domains explains its longevity. Cultures differ in what they treat as good evidence, yet all maintain a distinction between reliable and unreliable belief. That constancy points to a deep pragmatic truth: societies cannot act coherently without a shared grammar of epistemic accountability.

    The triad’s durability also shows its resilience to technological change. Algorithms, like humans, are judged by truth, belief-analogues, and justification. A predictive model that forecasts rain tomorrow is evaluated by whether it rains (truth), how confidently the system assigned probability (belief-analogue), and whether the reasoning behind the prediction—its data and parameters—holds up (justification). We demand these checks even of machines because they mark the boundary between information and knowledge. The form of life shifts, but the grammar remains.

    Durability here should not be confused with rigidity. Each component balances and corrects the others. Truth disciplines belief; belief animates truth; justification mediates between them. Remove truth and we drift into relativism; remove belief and we are left with sterile description; remove justification and we lapse into dogma. The triad functions as a self-correcting system, not a closed one. It remains stable precisely because it allows error to be recognized and remedied within practice.

    This balance can be seen in education, medicine, and ordinary trust. In education, to know a theorem is not merely to recite it but to understand its proof and application—a glimpse of the +U that will later enter our model. In medicine, to know a diagnosis is to link symptoms to mechanisms through reasoning that others can review. In friendship, to know someone’s character is to form a belief responsive to evidence and revised by experience. Across these contexts, the same triadic structure persists: truth as correspondence, belief as personal assent, justification as the publicly defensible bridge between them.

    Wittgenstein’s contribution is to make this structure visible rather than reinvent it. He shows that the power of JTB comes from its embeddedness in language-games. The standards of justification differ between a physicist and a parent, yet both belong to forms of life where reasons can be given and challenged. By examining those practices, we see that the grammar of know is stable not because it is timeless but because it is lived. When philosophers attempt to replace JTB, they usually reintroduce it under another name. Even coherence theories and reliabilism preserve its skeleton: truth as target, belief as stance, justification as route. The differences lie in emphasis, not in structure.

    The persistence of JTB across changing practices is therefore a mark of its depth. It names the minimal conditions under which knowledge remains intelligible. Even when justification fails or belief wavers, we appeal back to this framework to make sense of the failure. It functions like the grammar of a language: rarely noticed when used well, instantly felt when violated. To reject JTB entirely would be to forfeit the distinction between knowledge and luck, between reasoned warrant and accidental truth.

    For this reason the model endures. It does not demand perfection but provides a standard by which imperfection can be recognized. Its durability comes from its fit with the rhythms of human reasoning, from the fact that it mirrors the way we already keep our words in contact with the world. In this light, JTB is not a relic but a record of the conditions that make rational discourse possible. We can now examine its limits—the ambiguities of justification and the regress of reasons—and prepare for the Wittgensteinian insight that will dissolve those tensions without abandoning realism.
  • Sam26
    3k
    The following can't be overstated:

    We are entering an age in which the very conditions of knowing are being rewritten. Information now multiplies faster than human understanding can absorb it, and systems that simulate reasoning already shape what most people take to be true. The problem is no longer access to data but the loss of criteria for weighing it. Algorithms can imitate justification, narratives can mimic coherence, and conviction can be manufactured at scale. In such an environment, epistemology becomes a public necessity, not an academic luxury. To know what knowledge is, to see how truth, belief, justification, and understanding interlock, is the only safeguard against a world where persuasion replaces reason and where the grammar of “knowing” itself is quietly altered. Clarifying that grammar is the task before us.
  • Janus
    17.6k
    In this sense, JTB+U performs a Wittgensteinian clarification: it dissolves the illusion that justification alone guarantees comprehension. “U” distinguishes genuine justification from parroting, algorithmic correctness, or social conformity. Philosophically, that difference is now urgent—especially in an age where machines can simulate justification without understanding.

    This is important, because it's easy to suppose your point is correct.
    Sam26

    I've taken a while to respond, because there is a fair bit of subtlety, nuance in this question about understanding. I agree one cannot be justified in believing something without understanding how the justification works—that is, understanding how the (purported) facts that constitute the justification entail the belief.

    Someone could parrot an explanation of how a belief is justified without really 'getting' the explanation. It is very difficult, though, to say just what "getting" an explanation consists in other than the feeling or sense of getting it.

    In What Computers Can't Do and What Computers Still Can't Do Hubert Dreyfus argued that computers will never be genuinely intelligent because they cannot understand context.

    Yet the LLMs do seem to be able to do that, even though I cannot imagine how it would possible that they do that. Is it just a matter of parroting so sophisticated as to be able to fool us into thinking they do understand context?

    It begs the question as to how we grasp context, and I don't have an answer for that, but can only think that it must somehow be a matter of feeling. I can't imagine a computer having a feeling for context—but then what do I know?

    Anyway...interesting stuff!
  • Sam26
    3k
    Someone could parrot an explanation of how a belief is justified without really 'getting' the explanation. It is very difficult, though, to say just what "getting" an explanation consists in other than the feeling or sense of getting it.

    In What Computers Can't Do and What Computers Still Can't Do Hubert Dreyfus argued that computers will never be genuinely intelligent because they cannot understand context.

    Yet the LLMs do seem to be able to do that, even though I cannot imagine how it would possible that they do that. Is it just a matter of parroting so sophisticated as to be able to fool us into thinking they do understand context?

    It begs the question as to how we grasp context, and I don't have an answer for that, but can only think that it must somehow be a matter of feeling. I can't imagine a computer having a feeling for context—but then what do I know?
    Janus

    That’s a fair point, and an important point. When I say that someone can parrot an explanation without getting it, I don’t mean that “getting” is a hidden mental glow or some private feeling that accompanies comprehension. I mean that understanding shows itself in how a person can go on rightly — how they use what they’ve learned in new circumstances, recognize when it breaks down, and repair it. It’s not the presence of a sensation but the manifestation of mastery within a shared practice.

    That’s where I think the difference lies between human understanding and what large language models are doing. They can reproduce the surface grammar of context, the statistical pattern of what normally follows what, but they don’t inhabit the life-grammar that gives those patterns weight. They have no stake, no risk, no embodied continuity of experience. Our grasp of context isn’t just cognitive; it’s lived. It shows up as what I’d call a feeling for relevance: a readiness to respond that draws on our bodily attunement to the world and to one another.

    So yes, LLMs can simulate contextual understanding impressively, but simulation isn’t participation. Their outputs can mirror what understanding looks like, yet nothing in them corresponds to the hinge-layer that makes sense possible in the first place. Our “feeling for context” isn’t a mysterious extra — it’s the human way the background of meaning becomes visible in use.
  • Sam26
    3k
    I just finished my paper on epistemology which I'll post in here a section or two at a time.

    Post #1

    Justified True Belief Plus Understanding: A Wittgensteinian Extension

    Samuel L. Naccarato

    Abstract
    This paper reexamines the classical model of knowledge as Justified True Belief (JTB) and argues for its refinement through the addition of Understanding (+U). Drawing on Wittgenstein’s later philosophy—especially the concepts of language-games, grammar, and hinge propositions—it shows that justification operates within shared forms of life rather than in isolation. The JTB+U framework retains truth, belief, and justification but grounds them in public criteria and interpretive competence: knowing is not only having reasons but grasping how reasons function within practice. Three guardrails (No-False-Grounds, Practice-Safety, and Defeater Screening) and five justificatory routes (Testimony, Logic, Sensory Experience, Linguistic Training, Pure Logic) formalize this discipline of reliability. The analysis situates epistemology amid today’s information landscape, where distinguishing knowledge from persuasion has become urgent. By integrating Wittgenstein’s insights with modern concerns about data, AI, and scientific reasoning, the paper presents an epistemology that is self-correcting, communal, and humane. JTB+U preserves realism without dogmatism, acknowledging that understanding is both the hinge of justification and the safeguard of discernment in an age of unmoored information.

    I. Why Epistemology Still Matters
    It is fashionable to treat epistemology as an exhausted discipline—an inheritance of an earlier age, once useful for justifying science or theology but now displaced by data and probability. Yet the problem of knowledge has not vanished; it has multiplied. In a world where opinion spreads faster than evidence and conviction can be engineered by algorithms, the need to distinguish knowing from believing has never been greater. We inhabit a culture that prizes information but neglects understanding; it equates confidence with competence. Epistemology, properly understood, is not a relic but the grammar of orientation—our means of finding footing amid proliferating claims to truth.

    We are entering an age in which the very conditions of knowing are being rewritten. Information now multiplies faster than human understanding can absorb it, and systems that simulate reasoning already shape what most people take to be true. The problem is no longer access to data but the loss of criteria for weighing it. Algorithms can imitate justification, narratives can mimic coherence, and conviction can be manufactured at scale. In such an environment, epistemology becomes a public necessity, not an academic luxury. To know what knowledge is—to see how truth, belief, justification, and understanding interlock—is the only safeguard against a world where persuasion replaces reason and where the grammar of “knowing” itself is quietly altered. Clarifying that grammar is the task before us.

    To speak of knowledge presupposes an order of assessment. We do not call every opinion knowledge, nor every true remark knowledgeable. The distinction is not moral but functional: society, law, and science depend on reliable ways of sorting appearance from reality. When a court accepts testimony, when a physician interprets a scan, when a citizen evaluates a headline, the same question arises in different dress: what warrants belief? Epistemology matters because these decisions, repeated daily, determine whether our shared practices remain rational or collapse into echo.

    The classical model—Justified True Belief—still captures the skeleton of these practices. Truth ensures contact with reality; belief marks the personal uptake of that contact; justification provides the public warrant. Each condition blocks a familiar failure: without truth, we cling to illusion; without belief, we mouth what we do not hold; without justification, we risk luck or hearsay. The triad endures not because Plato decreed it but because human life still relies on the difference between being right and merely happening to be right. JTB describes, in grammatical form, the criteria by which we hold one another epistemically accountable.

    The word know itself does not wear a single face. We use it to express both epistemic and non-epistemic relations: I know the capital of France, I know how you feel, You should know better than that. Some of these are claims to justification; others are gestures of familiarity, empathy, or expectation. Wittgenstein reminds us that meaning follows use: the epistemic sense of know is only one branch within a broader family of uses. Clarifying that branch—seeing how it functions within our language-games—prevents us from mistaking conviction, or mere agreement, for knowledge. JTB+U is concerned solely with the epistemic use, where truth, belief, justification, and understanding converge within a form of life.

    Yet the simplicity of JTB conceals a difficulty sharpened by modernity. One can satisfy its letter without satisfying its spirit. A belief may be true and even justified by available evidence yet still lack the kind of grasp that distinguishes genuine knowledge from the echo of authority. Students can recite correct answers; machines can compute them; neither necessarily understands what they affirm. The missing element is not another reason but an internal relation—the capacity to use, extend, and situate what one claims to know. This fourth component, understanding (U), restores life to the classical framework and reconnects epistemology with practice.

    Understanding bridges knowing that and knowing how. It is not a separate species of knowledge but a measure of conceptual uptake: one understands when one can move fluently within the logical and practical consequences of a claim. To say “I know how a lever works” or “I know what justice requires” is to claim competence beyond the recital of propositions. This element of grasp allows knowledge to resist mere imitation. Adding U to JTB does not alter the structure of knowledge; it clarifies the dimension tacitly presupposed whenever philosophers spoke of insight, judgment, or wisdom—the ability to navigate a concept’s grammar, not merely repeat it.

    The urgency of this refinement becomes clear when justification falters under the weight of information. In an age of machine learning, deepfakes, and algorithmic persuasion, justification alone cannot guarantee comprehension. We can outsource calculation but not understanding. Even a flawless predictive system, if it cannot recognize the bounds of its certainty or the hinges on which its reasoning turns, mistakes correlation for truth. JTB+U therefore serves both as an epistemic model and as a moral warning: intelligence without understanding is cleverness without orientation.

    The revival of epistemology thus requires a shift in emphasis—from metaphysical speculation to grammatical description. Wittgenstein’s later philosophy provides the tools for that shift. His method was not to invent new theories of knowledge but to examine how the word know functions within our language-games. In doing so, he showed that justification is always contextual, embedded in the forms of life that sustain meaning, and that doubt operates only against a background of what stands fast. Our standards of justification are not free-floating; they draw their authority from these inherited practices. JTB+U continues this line by formalizing what Wittgenstein left implicit: that understanding is active mastery of a grammar, not possession of a mental state. Knowledge is not a mysterious substance but a practice: moving rightly within the space of reasons.

    This also clarifies why epistemology cannot be replaced by psychology or neuroscience. No scan or algorithm can decide whether a belief is justified; that judgment belongs to the public web of reasons in which meaning is maintained. To know something is to occupy a position within that web—to give and ask for reasons intelligibly to others. Epistemology names the effort to keep that web intact, to prevent private conviction from masquerading as objective warrant. When those boundaries collapse, language loses traction and communities lose trust.

    This project is conservative in spirit but progressive in aim. It seeks not to discard the classical framework but to ground it more deeply in lived practice. Adding understanding acknowledges that knowledge is never exhausted by propositions; using Wittgenstein’s methods locates those propositions within the activities that give them sense. The result is an epistemology both stricter and more humane: strict because it demands defeater-sensitivity and conceptual clarity, humane because it recognizes that justification is always conducted within a form of life.

    Although the framework I develop here is not metaphysical in method, it is not divorced from metaphysical depth. JTB+U does not appeal to a single essence or hidden ontology to give knowledge its meaning, yet it presupposes that justification and understanding unfold within a reality intelligible to mind. Metaphysics, in this sense, forms a horizon rather than a foundation: it shapes what can be meant without dictating how meaning is given. The grammar of knowledge may be clarified without invoking ultimate reality, but such clarity need not deny that a deeper order exists—what I later call the ultimate hinge of consciousness, the background that makes any grammar of understanding possible.

    If epistemology is to remain a living discipline, it must show how understanding anchors responsible judgment across domains—from the sciences and ethics to education and emerging forms of artificial intelligence. Only by refining what it means to know can we preserve a shared orientation toward truth in an age increasingly shaped by uncertainty.

    In what follows, I review why the JTB framework remains the most durable articulation of knowledge in practice, then trace its internal tensions—especially the ambiguity of justification and the regress of reasons—and show how Wittgenstein’s notion of hinges dissolves them without skepticism. From there, I introduce the +U element, outline the guardrails that keep justification reliable, and distinguish the several uses of certainty that reveal where knowledge properly resides. The paper concludes by showing how this Wittgensteinian extension preserves knowledge as a corrigible yet indispensable concept—one capable of guiding both human and artificial intelligence toward a more responsible understanding of truth.

    II. The Wittgensteinian Toolkit—Clarifying Our Grammar of Knowledge
    If metaphysics provides the horizon of intelligibility, Wittgenstein’s later philosophy supplies the instruments for navigating within it. His method is therapeutic rather than speculative: instead of offering new theories of knowledge, it clarifies how the word know functions in the practices where sense and doubt are learned. The tools are deceptively simple—language-games, grammar, forms of life, family resemblance, and hinges—but they work together, reinforced by the ideas of rule-following, public criteria, and the river-bed imagery of On Certainty. Taken as a whole, this toolkit grounds epistemology not in metaphysical essence but in lived intelligibility.

    Language-games: where “know” has meaning
    Wittgenstein begins with the recognition that meaning arises from use. Words are not labels for private entities but moves within human activities. To call something knowledge is not to identify a mysterious inner state but to perform a specific act inside a game with shared rules—asserting, challenging, correcting, teaching. Each domain has its own grammar of knowing: the scientist cites data, the witness swears an oath, the child repeats a lesson. There is no single essence of know; the unity lies in family likeness across these games. By examining language-games, we shift the epistemological question from “What is knowledge?” to “How do we use ‘know’ here?” The philosopher’s task becomes descriptive rather than legislative.

    Grammar: the logical order that frames sense and doubt
    Grammar, for Wittgenstein, is not syntax but the underlying logic that determines what counts as sense. It tells us what it means to justify, to verify, to doubt. Grammar sets the boundaries of intelligible talk: we can meaningfully say “I know this is a hand,” but not “I know I am in pain.” The latter confuses private experience with public criteria. Grammar thus replaces the metaphysician’s search for foundations with attention to rule-governed use. To grasp a concept grammatically is to know what would count as applying it correctly, what would make its application nonsensical, and how mistakes show themselves. Epistemology becomes a study of conceptual grammar—the order that lets sense and error exist at all.

    Forms of life: the public foundations of justification
    Every grammar stands on a background of practices, gestures, and agreements that are not themselves justified but simply enacted. These are our forms of life: the biological and cultural patterns within which reasoning has meaning. Justification presupposes such a background. We can question a witness, but not the practice of questioning itself; we can test a thermometer, but not the institution of measurement that makes testing possible. Forms of life are not propositions but the lived matrix of criteria. They explain how justification can be public without being infinite: our shared ways of acting already contain the standards that make giving reasons possible. Language-games and forms of life are mutually defining: our forms of life give stability to our games, and the games, in turn, articulate the forms of life they express.

    Rule-following and public criteria
    Within a form of life, rules acquire authority through training and correction. To follow a rule is not to consult an inner diagram but to participate competently in a practice where others can see and correct our moves. This dissolves the picture of justification as an internal state of certainty. What shows that I understand a rule is not introspection but my ability to go on correctly when circumstances vary. Knowledge therefore carries a social dimension: it is demonstrated in action and confirmed by public criteria. Wittgenstein’s reminder—there is no inner pointing to meaning—cuts against the Cartesian idea that knowing is primarily a private event. The grammar of know is visible in use, not hidden in consciousness.

    Family resemblance: unity without essence
    Philosophers often seek the single property that makes all instances of knowledge what they are. Wittgenstein advises a different posture: notice the overlapping similarities among uses instead of chasing a universal definition. Just as “games” share no one essence but display a network of resemblances—competition, rules, play—so too “knowledge” unites diverse activities by kinship rather than identity. This insight loosens the grip of essentialism that has haunted epistemology since Plato. It allows JTB, later extended to JTB+U, to function as a grammatical model rather than as a metaphysical claim. Its durability lies in family likeness across contexts, not in a timeless form.

    Hinges and the river-bed: the arational background of reason
    On Certainty adds the final and most radical instrument. Every act of justification presupposes propositions that stand fast—hinges that make doubt meaningful. These are not beliefs we know but certainties we act from. They form the river-bed in which the current of reasons flows. To question them all at once would be to lose the very distinction between sense and nonsense. Examples include “There is an external world,” “Objects persist,” “Words have stable meanings,” and, in your metaphysical horizon, “Consciousness is the condition for any appearing at all.” Hinges are arational, not irrational; they lie beneath justification, providing the stage upon which justification can occur.
    This insight explains why epistemic regress ends without circularity. When reasons run out, we do not reach an arbitrary stopping point but the practical background that makes reasoning possible. To call something a hinge is to mark the transition from saying to doing, from proposition to practice. Wittgenstein’s point is grammatical: the verb to know presupposes a contrast with possible doubt; remove that contrast, and the word loses its function. Hinges, then, are the conditions of meaningful knowledge, not propositions we could ever justify or refute.

    Doubt as practice-bound
    Because hinges define the limits of intelligible doubt, skepticism must itself be seen as a language-game. Doubting that the external world exists or that one has a body is not a deeper form of inquiry but a misuse of grammar—like playing chess without a board. Doubt has sense only where the framework of certainty stands fast. This reframes epistemology: instead of searching for indubitable foundations, we describe the hinges that already stabilize our reasoning. Certainty here is not a psychological feeling but a logical role. It anchors rather than concludes.

    The toolkit as re-grounding for epistemology
    When these instruments work together, they convert epistemology from a quest for absolute justification into a grammar of responsible use. Language-games reveal diversity; grammar maps the logic of sense; forms of life show why justification is public; rule-following and criteria keep meaning from collapsing into subjectivity; family resemblance preserves unity without essence; and hinges secure the background that makes any of this possible. The result is an epistemology both humbler and stronger: humbler because it recognizes limits, stronger because it grounds knowledge in the lived regularities that precede theory.

    Wittgenstein’s method thus complements rather than replaces the classical JTB framework. JTB describes the explicit structure of epistemic accountability; the toolkit shows the soil in which that structure grows. Together they transform the question of knowledge from a search for metaphysical foundations into an examination of the practices that sustain sense. When we later add understanding as the fourth component, the +U will draw directly on these insights: understanding is nothing other than fluency in a language-game, mastery of its grammar, and attunement to its hinges.
  • Sam26
    3k
    Post #2 Continuing with paper...

    Justified True Belief Plus Understanding: A Wittgensteinian Extension

    Samuel L. Naccarato



    III. JTB in Practice—Strengths and Durability
    With this toolkit in hand, we can now return to JTB itself, no longer as an abstract formula but as a practice embedded in forms of life. The next section examines how the triad of truth, belief, and justification operates within science, law, and ordinary reasoning, showing why it has endured across centuries of intellectual change. JTB persists because our language-games still require the grammar it names.

    The endurance of the classical triad lies not in philosophical inertia but in functional necessity. Wherever human beings act together, we rely—often without notice—on the grammar that JTB names. Truth secures contact with reality; belief marks personal assent; justification connects the two through publicly defensible reasons. Remove any one element and the practice collapses. The model persists because its logic is lived before it is theorized. Science, law, and ordinary life all enact its pattern, however differently they speak.

    To dismiss testimony out of hand is to practice a selective skepticism that defeats itself. Every belief, even in science, relies on others’ reports—records, measurements, observations, and reasoning transmitted through language. The issue is not whether testimony can yield knowledge but under what conditions it does. When sources are transparent, independent, and open to challenge, testimony functions as a public route of justification, not a private appeal to authority. A framework that denied this would unravel its own basis, for even the principles of logic and experiment reach us through human transmission.

    Philosophers have long distinguished between knowledge by acquaintance—the direct familiarity one has with a person, color, or sensation—and knowledge by description, which depends on propositions or reports. The former is immediate and non-inferential; the latter mediated and linguistic. Both fit naturally within the JTB grammar: acquaintance anchors belief in perception and experience, while description anchors belief in testimony and reasoning. What Wittgenstein’s method adds is the reminder that these are not competing species but overlapping uses of know, each governed by its own criteria of justification within a form of life.

    Consider science first, not as an institution but as a disciplined form of life. A physicist claims to know that a certain particle has a particular charge. What makes that claim knowledge rather than opinion is not conviction but the convergence of truth and justification. The statement corresponds to how the world behaves; the methods used to reach it—calibration, replication, peer review—provide the warrant that secures belief against mere luck. The scientist’s belief, her internal assent, is indispensable but not decisive. If later evidence overturns the finding, we say that the community was mistaken, not that belief altered truth. This readiness to correct itself shows the health of the practice: knowledge here is corrigible without being arbitrary.

    In science, JTB functions as a structure of accountability. Truth guards against fantasy; justification institutionalizes humility. A result that cannot be checked or replicated fails the justificatory test no matter how confidently it is asserted. Likewise, a correct result reached by accident—a lucky guess or an uncalibrated instrument—fails the epistemic test because justification is missing. These boundaries mirror the grammar that JTB codifies: luck without reason is not knowledge, and reason without truth is illusion. The laboratory simply dramatizes what ordinary cognition already presupposes.

    The same structure governs law, though expressed in another language-game. A jury must decide whether an accused person committed a crime. The verdict involves belief, but that belief counts as knowledge only when tethered to evidence that meets public standards—cross-examination, corroboration, admissibility. Truth is represented by the facts as they occurred; belief is the jurors’ collective assent; justification is the process that links the two—the rules of evidence and reasoning that make the verdict intelligible. The system acknowledges the fallibility of justification through appeals and retrials, mechanisms that reflect the same epistemic humility found in science. Even where human judgment replaces experiment, the grammar remains constant: knowledge requires a claim to be both true and justifiable to others.

    Legal reasoning also shows why knowledge must be public to be binding. The witness who swears to tell the truth invokes all three conditions. Truth corresponds to what occurred; belief is the sincerity of the witness; justification is the evidentiary frame that renders the testimony credible. A statement may be true yet unjustified if it cannot be corroborated, or justified yet false if the witness is mistaken. These permutations display the interlocking checks that give the concept its durability. We could abolish juries or rewrite evidentiary codes, but any workable system of accountability would still presuppose the same triad.

    Outside formal institutions, JTB continues to guide everyday reasoning. When I say that I know the bus leaves at eight fifteen, I am implicitly asserting that the schedule is accurate (truth), that I believe it (assent), and that I have grounds—experience, confirmation, or testimony—to warrant the claim. If the bus departs at eight thirty, I was mistaken, not lying. The grammar of correction—“I thought I knew”—shows that knowledge is a normative status, not a psychological state. We retract know precisely when one of its components fails.

    Such examples reveal that JTB is not a theoretical invention but a description of how epistemic life is already organized. Each time we distinguish between “He believes it” and “He knows it,” we invoke this grammar. Justification serves as the hinge between subjective conviction and objective warrant, allowing communities to coordinate understanding without collapsing into solipsism. The continuity of this pattern across domains explains its longevity. Cultures differ in what they treat as good evidence, yet all maintain a distinction between reliable and unreliable belief. That constancy points to a deep pragmatic truth: societies cannot act coherently without a shared grammar of epistemic accountability.

    The triad’s durability also shows its resilience to technological change. Algorithms, like humans, are judged by truth, belief-analogues, and justification. A predictive model that forecasts rain tomorrow is evaluated by whether it rains (truth), how confidently the system assigned probability (belief-analogue), and whether the reasoning behind the prediction—its data and parameters—holds up (justification). We demand these checks even of machines because they mark the boundary between information and knowledge. The form of life shifts, but the grammar remains.

    Durability here should not be confused with rigidity. Each component balances and corrects the others. Truth disciplines belief; belief animates truth; justification mediates between them. Remove truth and we drift into relativism; remove belief and we are left with sterile description; remove justification and we lapse into dogma. The triad functions as a self-correcting system, not a closed one. It remains stable precisely because it allows error to be recognized and remedied within practice.

    This balance can be seen in education, medicine, and ordinary trust. In education, to know a theorem is not merely to recite it but to understand its proof and application—a glimpse of the +U that will later enter our model. In medicine, to know a diagnosis is to link symptoms to mechanisms through reasoning that others can review. In friendship, to know someone’s character is to form a belief responsive to evidence and revised by experience. Across these contexts, the same triadic structure persists: truth as correspondence, belief as personal assent, justification as the publicly defensible bridge between them.

    Wittgenstein’s contribution is to make this structure visible rather than reinvent it. He shows that the power of JTB comes from its embeddedness in language-games. The standards of justification differ between a physicist and a parent, yet both belong to forms of life where reasons can be given and challenged. By examining those practices, we see that the grammar of know is stable not because it is timeless but because it is lived. When philosophers attempt to replace JTB, they usually reintroduce it under another name. Even coherence theories and reliabilism preserve its skeleton: truth as target, belief as stance, justification as route. The differences lie in emphasis, not in structure.

    The persistence of JTB across changing practices is therefore a mark of its depth. It names the minimal conditions under which knowledge remains intelligible. Even when justification fails or belief wavers, we appeal back to this framework to make sense of the failure. It functions like the grammar of a language: rarely noticed when used well, instantly felt when violated. To reject JTB entirely would be to forfeit the distinction between knowledge and luck, between reasoned warrant and accidental truth.

    For this reason the model endures. It does not demand perfection but provides a standard by which imperfection can be recognized. Its durability comes from its fit with the rhythms of human reasoning, from the fact that it mirrors the way we already keep our words in contact with the world. In this light, JTB is not a relic but a record of the conditions that make rational discourse possible. We can now examine its limits—the ambiguities of justification and the regress of reasons—and prepare for the Wittgensteinian insight that will dissolve those tensions without abandoning realism.
bold
italic
underline
strike
code
quote
ulist
image
url
mention
reveal
youtube
tweet
Add a Comment

Welcome to The Philosophy Forum!

Get involved in philosophical discussions about knowledge, truth, language, consciousness, science, politics, religion, logic and mathematics, art, history, and lots more. No ads, no clutter, and very little agreement — just fascinating conversations.