I meant to highlight this earlier. It’s an important takeaway from this conversation.
I wouldn’t say anything about a ‘hinge’ unless there is a problem with the physical hinge that allows the door to move. I don’t think that complying with the request to close the door has anything to do with what Wittgenstein says about hinges or conviction. Sometimes we act with conviction other times we do not.
On the Difficulty of Precise Formulation in This Territory
A question that has come up repeatedly in this thread and the exegesis thread is whether the difficulty of articulating what Witt is pointing to reflects a problem with the thinking itself. I want to address this directly, because the answer affects how the work should be approached. The difficulty is real, but it isn’t a sign of confusion. It’s a sign that the phenomena being described haven’t been described precisely before, and the work of precisifying them is much harder than the work of refining vocabulary for territory that has already has a use.
Precise description is available in principle. Language isn’t inherently inadequate for what Witt and this thread are trying to say. The difficulty comes from the novelty of what is being described rather than from a limitation of language. The phenomena at the bedrock of hinge certainty haven’t been clearly isolated in the philosophical tradition. Witt is the first to attend to them with sustained focus, and even he was working at the limits of what he could formulate. The secondary literature has taken up pieces of what he pointed to, but no one has developed a terminology that cleanly distinguishes the levels, tracks their different modes of operation, and shows how they relate without collapsing into each other. That is a genuine gap, not a limitation of language.
This means the work has two dimensions that depend on each other. The phenomena need to be seen clearly, and vocabulary adequate to them needs to be developed. Neither can be completed independently. You cannot get the vocabulary right without seeing the phenomena clearly, and you cannot see the phenomena clearly without vocabulary that lets them come into focus. This is why formulations keep needing revision, and why pressing on terms like “conviction,” “belief,” or “absolute” often reveals distinctions that the current vocabulary hasn’t been marking. Each time a term strains under examination, two things are happening. Something about the phenomenon is becoming visible that the term was not capturing, and a better formulation is being felt for. The noticing and the formulating inform each other.
What this means in practice is that when a formulation feels inadequate, the question to ask is not whether to give up on precision. It is whether the phenomenon has been seen clearly enough yet for the right articulation to be possible. Sometimes the answer is yes, and the work is finding the words. Sometimes the answer is no, and the work is looking at the phenomenon more closely until the right articulation becomes available. Both are legitimate stages, and they alternate throughout any original philosophical project. The progress is real even when it is slow, because each increment of precision makes further phenomena accessible, which in turn demands further precision.
This is what original work in this territory looks like. Witt himself was doing it, which is why OC circles and qualifies and tests formulations that get abandoned or revised. The thread is continuing that work rather than finishing it. The taxonomy of four senses of certainty, the three levels of hinge certainty, the distinction between structural and formal absoluteness, and similar articulations are increments of precision developed by attending closely to the phenomena and feeling for vocabulary adequate to them. The difficulty is not a defect in the thinking. It is the character of the work.
Do you think that in order to understand what I’m talking about when I ask you to close the door, you need to bring to bear a framework of intelligibility that makes what I am saying sensible and familiar to you?
“A framework of intelligibility”? I must understand what a door is and what it means to close it.
From On Certainty:
- You must bear in mind that a language‐game is, so to speak, something unpredictable. I mean: it has no grounds. Isn’t reasonable (or unreasonable).
It’s just there – like our life.
- My life consists in my being content with various things.
- I should like to regard this assurance not as something akin to hastiness or superficiality, but as a form of life. (That’s very badly expressed and probably badly thought out as well.)
- Our knowledge forms a large system. And only within this system has any single item the value we give it.
From Blue Book:
Frege ridiculed the formalist conception of mathematics by saying that the formalists confused the unimportant thing, the sign, with the important, the meaning. Surely, one wishes to say, mathematics does not treat of dashes on a bit of paper. Frege’s idea could be expressed thus: the propositions of mathematics, if they were just complexes of dashes, would be dead and utterly uninteresting, whereas they obviously have a kind of life. And the same, of course, could be said of any proposition: Without a sense, or without the thought, a proposition would be an utterly dead and trivial thing. And further it seems clear that no adding of inorganic signs can make the proposition live. And the conclusion which one draws from this is that what must be added to the dead signs in order to make a live proposition is something immaterial with properties different from all mere signs.
But if we had to name anything which is the life of the sign, we should have to say that it was its use.
page 9
As a part of the system of language, one may say “the sentence has life”. But one is tempted to imagine that which gives the sentence life as something in an occult sphere, accompanying the sentence. But whatever would accompany it would for us just be another sign.
page 10
1 Like
Thank you for your generous response Sam.
So this implies that you’re not doing epistemology, but offering a transcendental account of its conditions of possibility. But then you still have the pre-linguistic ground working as some kind of explanation for epistemic practice. That’s a tension, at the very least: what is explicitly placed outside the space of reasons is still being used to explain how that space is possible.
And that means we can distinguish between two claims:
-
Animals, infant humans, and mature humans all exhibit more or less reliable, non-reflective, action-guiding dispositions.
-
Human epistemic practices are best explained by those dispositions, in the sense that they form a basis for them.
Claim (1) is uncontroversial, but claim (2) leaves out how pre-reflective habits actually explain justifications, certainties, and knowledge. And claim (2), the way you have argued it, seems to be asserting that there is some non-linguistic layer in humans that can explain epistemic practice, whereas I’d argue that once you’ve got language, there’s no pure animal layer available for epistemological explanation—the distinction between mere behaviour and justified belief is already internal to language, i.e., we can’t start with raw animal certainty and then just add justification on top.
This assumes that Wittgenstein is only rejecting propositional foundationalism, but still pointing to something beneath language games. I don’t think he is doing that. His ground is like the “bottom” he mentions elsewhere—the end of the chain of justifications, and yet internal to our languagey form of life. To “give grounds” is to justify, and Wittgenstein’s point is that ultimately, one’s justifications are not further justified. To give grounds need not be understood as finding the deeper and deeper layers beneath.
I like the way that Wittgenstein mixes up and undermines his own metaphors. This is a good one:
- I have arrived at the rock bottom of my convictions. And one might almost say that these foundation–walls are carried by the whole house.
So your base-superstructure metaphor doesn’t work for me. I think Wittgenstein has in mind something more like a system of mutually supporting practices. Justifications are on the same level as, and embedded within, our practices and their associated certainties. When justifications come to an end, they fizzle out towards the edges, where they merge into more-or-less certain practices. The benefit of this metaphor is that it doesn’t tempt one into traditional epistemology, as I think yours does: instead of animal certainty and practices holding up a stack of justifications, they hold it in place in another way (I’m avoiding introducing a single prime spatial metaphor of my own here) such that even the “top” justification is close to, and held in place with, the surrounding practices. In other words, rather than vertical and stratified, justification is distributed.
Granted, but this only shows that there is reliable pre-linguistic behaviour, not that the concept of certainty or hinge certainty apply.
But I think this only succeeds in supporting claim (1), which isn’t controversial. And I think if you’re doing (2), there does need to be some explanatory connection—but I’m not persuaded this is a coherent project, because such a connection would be tantamount to bringing the supposedly pre-linguistic into the realm of the linguistic.
I’ve bolded the crucial claim that I’m not sure about. The way Wittgenstein describes hinges in OC, they are well within linguistic practice. Of course, elsewhere he talks about animal certainty:
One might say: “‘I know’ expresses comfortable certainty, not the certainty that is still struggling.”
Now I would like to regard this certainty, not as something akin to hastiness or superficiality, but as a form of life. (That is very badly expressed and probably badly thought as well.)
But that means I want to conceive it as something that lies beyond being justified or unjustified; as it were, as something animal.
I take §359 to show (and show that he is aware) that when he tries to describe this certainty as “animal”, he is being drawn into a conception that doesn’t actually do anything to explain the role of hinge certainties.
In my view, when Wittgenstein says in PI §19, “And to imagine a language means to imagine a form of life,” I take it to mean something close to that language is our form of life, not that language entails a form of life prior to and beneath it.
When he gets to animal certainty, I think he knows it says no more than claim (1) and therefore represents a blind alley. What I think he does show successfully is that where to doubt and where not to doubt are not something given separately from language. He makes it clear that these are learned together. They’re part of the same process. It seems difficult, then, to maintain a pre-linguistic certainty as a basis.
Just as in writing we learn a particular basic form of letters and then vary it later, so we learn first the stability of things as the norm, which is then subject to alterations.
This games proves its worth. That may be the cause of its being played, but it is not the ground.
- But isn’t it experience that teaches us to judge like this, that is to say, that it is correct to judge like this? But how does experience teach us, then? We may derive it from experience, but experience does not direct us to derive anything from experience. If it is the ground for our judging like this, and not just the cause, still we do not have a ground for seeing this in turn as a ground.
Notice the sharp distinction between a cause and a ground. Crudely speaking, I think we can say that there must be animal causes, but I don’t think animal dispositions can explain knowledge, justification, and certainty.
You often slide from grounds as justifications or reasons (“giving grounds”), some of which stand fast, to the ground as what lies beneath and supports the language-game. I think Wittgenstein means the first only.
Bravo. And it takes a certain humility to do this kind of work. You have to be willing to look somewhat stupid. (Witt was wonderful at this.) You have to be able to keep pressing simple questions no matter how often you’re told that the answers are obvious. You have to remain unconverted! – until you genuinely are.
Jamal, this is the kind of response that makes the thread worth having. There are several distinct challenges, and I’ll take them in order, but I should start by clarifying the framework I’m working within, because some of what you’re reading as tension makes more sense once that framework is explained.
You read my distinction between the chronological/ontological claim and the epistemic claim as implying I’m not doing epistemology but offering a transcendental account of its conditions. I’d push back. The hinge picture isn’t a departure from epistemology. It’s part of a framework I’ve worked out elsewhere, viz., JTB+U, where U is competent application understood through Witt’s grammatical and practice based reminders. JTB+U is recognizably first order epistemology. It addresses Gettier by requiring not just true justified belief but the demonstrated competence to apply the concept across cases, withhold it from near misses, and accept correction. It addresses regress by grounding justification in hinges that aren’t candidates for justification themselves but conditions that make justification possible. The prelinguistic hinges aren’t free standing transcendental claims. They’re the foundation (not traditional foundational ideas) that lets JTB+U handle structural problems traditional epistemology hasn’t been able to solve from within.
So when you say I’m not doing epistemology, I’d say I’m doing epistemology that uses Witt and the hinge picture to address problems that have plagued JTB since Plato. Not transcendental philosophy in the Kantian sense, i.e., where conditions of possibility are derived by analyzing what must be the case for experience. Closer to descriptive work in Witt’s spirit, attending to what is the case in human and animal engagement with the world, but the descriptive work has epistemological consequences because what it describes resolves problems epistemology hasn’t been able to solve internally. The supposed tension between placing the hinge outside the space of reasons and using it to explain epistemic practice dissolves once this is in view. The hinge functions as a condition whose recognition resolves the regress, not as a premise from which epistemic practice is derived. Identifying a condition isn’t the same as offering a justification.
Now to your specific challenges.
On Claim 1 vs Claim 2. You’re right that Claim 1 is uncontroversial and Claim 2 is the substantive one. But the version of Claim 2 you’re attributing to me is stronger than what I need. I’m not claiming animal dispositions explain knowledge, justification, and certainty in the way propositional foundations were supposed to explain belief in classical foundationalism. I’m claiming something weaker and structural, viz., that human epistemic practice is the kind of thing that could only develop in creatures who already engaged with the world in certain ways before language arrived. That engagement’s a precondition of the practice, not a foundation that explains the content of the practice. The content (i.e., what counts as good reasoning, evidence, defeaters) is internal to the practice and can only be analyzed from within it. JTB+U operates at that level. The hinges sit beneath as the condition that lets the practice function.
On the metaphor. You’re right that base and superstructure is the wrong picture, and I should’ve been clearer that I’m not committed to it. The riverbed and chess images capture it better. The board, pieces, and rules aren’t proved before each move. They’re the enabling frame that makes moves intelligible as moves. They aren’t beneath the moves holding them up. They surround them. Witt’s “these foundation walls are carried by the whole house” does the same work. So the metaphor’s conceded, with the qualification that I still want to maintain that some of what’s distributed in the picture is prelinguistic engagement, not just linguistic practice.
On the dog and the infant. This is where the real disagreement sits. You grant they exhibit reliable behavior but deny that the concepts of certainty or hinge certainty apply to them, citing OC 358-359 where Witt flags his attempt to describe certainty as “animal” as “very badly expressed and probably badly thought as well.” You read that as Witt recognizing a blind alley.
I read it as Witt recognizing the difficulty of formulation, not the failure of the substantive claim. The “badly expressed” is about the language available to him, the same struggle OC 76 names when he says his aim is “to give statements one would like to make here but cannot make significantly.” The substantive insight, viz., that certainty at this level is something animal, something prior to justification, isn’t retracted. He says he wants to conceive it that way. The qualifier’s about how hard it is to say what he wants to say in the inherited vocabulary, not about whether what he wants to say is true.
The harder question’s whether “certainty” can apply where there’s no language. You’re saying once you have language, the distinction between mere behavior and justified belief is internal to language, so there’s no prelinguistic concept of certainty that could apply to the dog. But the certainty I’m attributing to the dog isn’t epistemic certainty, which I agree is internal to language and the practices of justification. It’s hinge certainty at the prelinguistic level, i.e., a different phenomenon. Calling it certainty isn’t a category mistake. It’s noticing that what’s going on in the dog’s engagement with the world has the same structural role hinge certainty has in human practice. Both are what makes engagement with the world possible at all. Both are shown in acting. Both lie outside the space of justification. The continuity of the structural role is what justifies extending the term, even though the dog has no language.
On cause and ground. You’re right that Witt draws the distinction sharply at OC 130. But I don’t think he’s saying animal dispositions can only be causes. The cause/ground distinction is about whether something gives rational support to a conclusion or merely brings it about. Witt’s point at OC 130 is precisely that the ground of our judging isn’t something that gives rational support in the ordinary sense. It’s the ungrounded way of acting from OC 110. That way of acting isn’t a justification we appeal to. It’s the condition under which appealing to justifications is possible. Calling it a ground rather than a cause doesn’t put it inside the space of reasons. It’s what OC 105 calls the element in which reasons have their life. Within JTB+U, this is what stabilizes justification. Justification proceeds along five routes (viz., testimony, logic, sensory experience, linguistic training, pure logic), each with its own criteria, each operating within practices, and the practices themselves rest on hinges that aren’t products of justification but conditions of it.
On the charge that I slide between two senses of “ground,” that’s fair, and I should be more careful. There’s the sense of ground as justification or reason within a practice, which Witt discusses throughout OC and which JTB+U analyzes through its account of justificatory routes. And there’s the sense of ground as what makes the practice possible, which the hinge chapter addresses. Witt uses “ground” for both. When he says justification comes to an end, he’s using the first sense. When he says the end is an ungrounded way of acting, he’s using something closer to the second, with the careful qualification that it’s ungrounded, not a further ground. What I’m pointing to with the prelinguistic hinges is the second sense, and you’re right that calling it a ground risks confusion. “Condition” is probably the cleaner word.
Now the deeper question. You worry that “such a connection would be tantamount to bringing the supposedly pre-linguistic into the realm of the linguistic.” I don’t think it has to be. The description of the prelinguistic in language is one thing. The prelinguistic itself is another. We can describe the dog’s engagement with the world, and the description’s linguistic, but the engagement isn’t. When I say the infant’s reaching shows hinge certainty, I’m describing in language something that operates without language. The fact that the description’s linguistic doesn’t make the described phenomenon linguistic, any more than describing a sunset in words makes the sunset linguistic. The map and the territory remain distinct.
What I’m not committed to, and where you’re reading me as more ambitious than I am, is the idea that the prelinguistic engagement explains the content of human epistemic practice in a way that lets us derive the practice from the engagement. I don’t think that’s possible, and I don’t need it. JTB+U handles the content on its own terms, by analyzing truth, belief, justification, and competent application within the practices where they have their life. What the prelinguistic hinges add is the answer to the regress problem and the explanation of why JTB+U doesn’t collapse into either skepticism or dogmatism.
So to summarize. I’m doing epistemology, viz., JTB+U as developed in chapters 6 and 7 of From Testimony to Knowledge. The prelinguistic hinges are part of that framework, doing specific work in solving the regress and stabilizing the practice based account of justification. I’m conceding the base superstructure metaphor and adopting something closer to the riverbed and chess images. I’m conceding the imprecision of “ground” and trying to use “condition” where appropriate. I’m holding fast to the claim that hinge certainty applies to prelinguistic creatures, on the grounds that the structural role’s the same and the OC 358-359 passage shows Witt struggling with formulation rather than retracting the substantive claim.
The disagreement that remains is whether “certainty” and “hinge” can do work outside language games. You think they can’t. I think they can, on the grounds that what they pick out has the same structural role at the prelinguistic level that it has at the linguistic level. That’s the real point of contention.
1 Like
Earlier, I pointed to the map vs territory binary as a key underpinning of your treatment of the relation between linguistic justification and their condition of possibility. I argued that this split between what is really there and our model of what is there invokes the risk of skepticism and an infinite regress. As long as we remain within the thinking of the map-territory binary, the only way to put a floor under the regress is by fiat: we can be certain the floor is there even if we can’t justify it propositionally.
It may be helpful at this point to express my opposition to this reading of Wittgenstein through the words of Lee Braver in his book on Wittgenstein and Heidegger, ‘Groundless Grounds’. According to Braver, our language games are arbitrary because they are not grounded in the real facts of the world. The reason for this is not because there is a split between the map and the territory, our schemes and an independently real world.
The primary way our language-games are arbitrary is that we cannot justify them in the most natural, obvious way, namely, by grounding them in reality. We want to say that, for example, we pick these colors to be primary because they really are primary; that’s the way things themselves are and our taxonomy of the world mirrors its inherent articulation, letting us “carve at the joints.” But in his later work Wittgenstein emphatically and repeatedly rejects this notion, often using the word “arbitrary” to denote
this rejection: “the rules must be laid down arbitrarily, that is, are not to be read off from reality like a description. For when I say that the rules are arbitrary, I mean that they are not determined by reality in the way the description of reality is. And that means: it is nonsense to say that they agree with reality.”The point isn’t that we can never know whether our beliefs match up with the world itself, nor that we in fact know that they do not. Rather, Wittgenstein is making the kind of move often made by great philosophers: he is rejecting the conceptual framework within which the
comparison makes sense at all.Experience radically underdetermines what we make of it. Without
meaning-objects anticipating their own proper use, the same set of facts can give rise to multiple classifications or sets of rules, even to different takes on what counts as a fact. The bizarre behavior Wittgenstein likes to imagine often shows how people could apply rules or interpret experience or continue patterns in all sorts of ways; this exposes our behavior as being one possibility among many.
Like Foucault’s invocation of Borges’ fantastic Chinese taxonomy, Wittgenstein argues that “the value of such games is that they destroy prejudices; they show that ‘it need not always be this way.’” Our classifications don’t mirror the way things are, not because they’re wrong but because there is no Way Things Are; “there is no absolute
similarity” since “everything is analogous to everything else,” nor is there absolute simplicity or exactness (PI §88). The overpowering self-evidence of our way of doing things is due to our familiarity with it, rather than its fit with the world.We cannot justify our language-games by appealing to Reality Itself because it is only through some game or other that we access reality in order to determine its significance. “I cannot use language to get outside language. . . . Grammatical conventions cannot be justified by describing what is represented. Any such description already presupposes the grammatical rules.” Without self-classifying meaning-objects, the world does not tell us how to describe it. We can of course compare various particular claims to the world, but only on the basis of some description, which cannot simply be read off of the world but requires “a great deal of stagesetting.”
Josh, Braver is answering a question I’m not asking.
Braver’s question is the following: can we justify our linguistic frameworks by comparing them with reality? His answer is no, because we only access reality through some framework or other, so the comparison has no independent standpoint. I agree with him. That’s a good answer to that question.
My question is different. I’m not asking whether our descriptions match reality. I’m asking what was there before descriptions started. Braver’s entire argument takes place within a world where language is already up and running. He’s asking about the relationship between language and what language describes. But the relationship between language and what language describes only becomes a problem after language exists. Before that, there’s no relationship to worry about, because there’s no language. There’s just a world with creatures acting in it.
Think of it this way. Braver says we can’t step outside language to compare our descriptions with reality. Fine. But nobody stepped inside language either. Language arrived. Something was already going on when it got there. Braver’s argument tells us important things about what we can and can’t do with language once we have it. It tells us nothing about what was happening before we had it.
The Witt Braver draws on is the Witt of the PI, working on rules, meaning, and the arbitrariness of our classifications. The Witt of OC goes somewhere different. OC 204 doesn’t say acting lies at the bottom of our descriptions of reality. It says acting lies at the bottom of the language-game. The game has a bottom, and the bottom isn’t more game. OC 205 says the ground is not true, nor yet false. True and false are what propositions are. The ground isn’t a proposition. It’s beneath the space where Braver’s argument operates.
Braver says there’s no “Way Things Are” because our classifications are one possibility among many. But this is a point about how language carves up what it encounters. It’s not a point about whether there’s anything to encounter. The world that creatures engaged before language existed wasn’t a “Way Things Are” in Braver’s sense, because it wasn’t a classification. It was just the world. Uncarved. Undescribed. Already there. Already being acted in with conviction.
So I’d put the question back to you. Do you think something was going on before language? If yes, then we agree on the ground, and the argument is about how to characterise it. If no, then I’d want to know what you think language arrived into.
He is not simply arguing that the reason we can’t justify linguistic frameworks by comparing them with reality is because our access to it is limited by our frameworks. That’s a familiar Kantian-style move: reality remains in place as something like a fixed background, while our epistemic situation prevents direct comparison. But Braver is doing something more radical.He dissolves the dilemna by questioning whether the very idea of “reality as something that could serve as a standard for comparison” is coherent in the way you seem to assume.
For you, “reality” is still a kind of independent, pre-articulated something, what you call the “uncarved world.” Even if we can’t step outside language to check our descriptions against it, it’s still there as a determinate baseline. As you say, “I’m not asking whether our descriptions match reality. I’m asking what was there before descriptions started.” But Braver is arguing that once you describe that “something” as uncarved, undescribed, already there, you’re still characterizing it, and doing so from within a conceptual scheme. Braver’s point would be that this move doesn’t escape the problem; it just redescribes it in a more minimal vocabulary.
I think what comes before the language of formal propositional justification is an informal, non-representational language of situational coping. The former is an abstraction secondary to and derived from the latter. Central to situational coping is the meeting between schematized embodied expectations projected onto the world and the feedback from the world, forming actual linguistic as well as perceptual experience. The schematized nature of perception in humans and other animals generates normativity, involving the determinations of correctness and incorrectness. So already in the normative self-organizing dynamics of simple living systems the elements of hinges, grammar and justification are in place in a primordial form.
What is needed for an animal to execute an action like reaching for an object is a normative embodied scheme, the animal’s perceptual ‘description’ of the anticipated sense of the event. What is not needed is a pre-schematic condition of possibility underlying action. The real doesn’t lie hidden beneath layers of norms constructed on top of it. It can only be located, whether in human language or animal perception, as an element of contingent normative structures.
There is always a ‘before’ to be located, a before formal propositional justification , a before skilled situational linguistic coping, a before animal perception. But these ‘befores’ dont act as one-way causes. They are changed by what they enter into, by what is being done with them. The before is repurposed by what comes after. That’s why we can only understand and locate a hinge by what it is doing in the current context. The ‘what was before’ only has its existence in how it is being taken up and repurposed here and now.
The term ‘ground’ as it used at 205 is not about pre-linguistic acting. It refers to giving grounds, that is, justification and evidence for what we say (204). He does not say that the ground is not true or false, what he says is that if the true is what is grounded, that is, if what is true is what we have justified, then the justification itself is not true or false.
200. Really “The proposition is either true or false” only means that it must be possible to decide for or against it. But this does not say what the ground for such a decision is like.
Deciding for or against a proposition is not a matter of grounding it on a “pre-linguistic hinge” or action. In some cases we cannot give a ground for it, that is, we cannot justify it. In such cases we act without grounds.
I hope you’ll forgive my having only read some of the posts here. I hope also that there isn’t something in them that I should have taken into account.
The question what there is/was before language, or independently of language in the context of philosophy is very difficult. But I think you misunderstand it - not quite in the same way that Braver does, but nonetheless.
As soon as we ask, or even think, about how to characterize what there is without or before language, we need to face the limits of language. There is no way other than language to characterize that reality. That’s inherent in the concept of characterizing anything. Adding “pre-linguistic” or “non-linguistic” doesn’t resolve the problem.
Wittgenstein does seem to be sure that the floor is there. I am very puzzled about that. A floor can be part of the building, but in another sense the floor is what the building rests on - and that means that it is not part of the building, but something of a distinct kind, with a distinct manner of existence. But does there have to be a floor at all? There are such things as self-supporting structures. Not everything needs a foundation at all.
I’m afraid that doesn’t help me at all. The talk of structural roles and prelinguistic levels takes for granted what ought to be in question - the model of foundation and superstructure - and looming in the background are that concepts, if they deserve such a name, of noumena and Being-In-Itself.
Surely, if we are to be consistent, we need to do something different from talking. Perhaps all that can be done is to pay attention to what you started with:-
Perhaps that’s all there is, or needs to be. Perhaps Witt never really forgot that there comes a point when speech no longer serves a purpose. For me, paradoxically, that’s where all his talk of foundations leaves me.
You think Witt is sure there is a floor there? I’m partial to the view that he avoids dealing with the question directly. In that regard, both Sam and I believe ourselves to be venturing into territory that is left fuzzy in Witt’s texts.
You read “ground” in OC 205 as referring to giving grounds, i.e., justification. On this reading, Witt is saying that if what’s true is what we’ve justified, then the justification itself is not true or false. The remark would be about the limits of justification within the practice.
But look at OC 204 and 205 together, which is more consistent, since 205 follows directly from 204.
OC 204 says giving grounds comes to an end, and the end is not propositions striking us as true but our acting. OC 205 says if the true is what is grounded, then the ground is not true, nor yet false.
On your reading, “ground” shifts meaning between 204 and 205. In 204, the end of giving grounds is acting. In 205, on your reading, “ground” goes back to meaning justification. But if we hold the meaning steady, then 205 is saying something about the acting that 204 just identified as the end. The acting is the ground. And the ground is not true, nor yet false. This isn’t a remark about justification being neither true nor false. It’s a remark about acting being neither true nor false. And that makes sense, because acting isn’t the kind of thing that has a truth value. Propositions are true or false. Acting is neither.
You say “in such cases we act without grounds.” But OC 204 doesn’t say we sometimes act without grounds. It says acting lies at the bottom of the language-game. Not “in some cases.” At the bottom. As in, always, beneath everything. The language-game rests on acting the way a building rests on the ground. This isn’t an observation about occasional gaps in justification. It’s a claim about the structure of the whole enterprise.
I’ll grant that “ground” is doing double duty in this neighborhood of OC, and that Witt himself doesn’t draw a clean line between ground-as-justification and ground-as-what-lies-beneath. That ambiguity is real and worth thinking about. But reading 204 and 205 as a pair, with 204 identifying acting as the end and 205 characterizing that end as neither true nor false, seems to me the more natural reading. The alternative requires “ground” to shift its reference between consecutive remarks, which is possible but costs more interpretively than holding it steady.
This is not how Wittgenstein is using the term ‘ground’.
200. Really “The proposition is either true or false” only means that it must be possible to decide for or against it. But this does not say what the ground for such a decision is like.
Acting is not the ground for the decision.
204. Giving grounds, however, justifying the evidence, comes to an end
There are no further ground for the decision. We must act but we have reached the end of giving grounds for that decision.
There is no ambiguity. Giving grounds comes to an end. What lies at the bottom of the language-game is not a ground of a different kind with a different meaning.