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

Metaphysics & Epistemology
175

While I understand your arguments against a “foundational” interpretation of the text, I would like to make clear that my comments challenge the view of certainty as equivalent to “taking for granted.” Noting the groundlessness of belief is not presented as something like saying “nothing under-girds our ideas.” What can be said at a minimum is that Moore wants a standpoint beyond belief.

So, while I can appreciate how different theories can take lessons from Wittgenstein’s method, my effort is more strictly exegetical: If we don’t use certainty as a resource as dismissed in #30, phrases like “hinge certainty” do not fit the context of the text.

176

Once again, then don’t.

I understand your frustration with being challenged, but the insults are uncalled for, inaccurate, and rude. Disagreement is an ordinary part of philosophical argument. Accusing someone of trolled is not. You have put me in an awkward position where any further disagreement with you will not be taken in the spirit in which it was given, but the accusation will serve as a form of implicit suppression.

2 Likes
177

All complex languages are abstractions built on a true/false duality. This is provable by the fact that the most advanced computer systems and AI LLMs, which can store and process all epistemic knowledge, are built on the simplest possible language: binary, true (1) versus false (0). It is literally impossible to make a simpler coherent language.

Wittgenstein calls these foundational elements “hinges.” The issue is that he treats them as merely ungrounded “background practices” or grammatical frameworks of a language game, not as pointers toward an actual ontological precondition.

That assertion itself is self-refuting. It makes a universal truth claim about the nature of certainty and knowledge while simultaneously denying any universal Truth.

178

That proof only works if the AI demonstrates that it understands the language to the same extent as any human being. But the AI obviously doesn’t, so the proof isn’t a proof.

179

Thanks for the reply.

Whether the AI understands or not is irrelevant. The point is that all epistemic knowledge can be stored on a computer system that uses binary to operate, literally flipping physical switches in the silicon circuits between on (1) and off (0). This shows that all languages like English or German are just abstractions built on the simplest possible language: binary true (1) vs false (0).

That binary foundation is not merely epistemic. It is the ontological precondition that makes any coherent distinctions or reality possible at all, including Wittgenstein’s hinges. Without a true/false binary no system, natural or man-made, could exist.

180

How do you know this?

181

But that doesn’t prove that all languages are built on a true/false duality. It just shows that all languages can be translated into such. If the translation misses out on understanding aspects of the original, as all translations do, then nothing is proven. In fact, that the opposite is the case, is demonstrated quite strongly.

182

We know because we can directly observe it. Furthermore, your very question exposes the duality directly. You have assumed that there is a distinction between the state of knowing (true) and not knowing (not true). You are already using Logic’s Law of Identity to make that distinction. This is what line 3 in my “Try Denying This Argument” thread is saying:

You wouldn’t even be capable of thought if this wasn’t true.

183

Thanks for the reply, Meta_U.

All languages can be encoded into binary without any loss of meaning. Every symbol, word, or structure is preserved exactly. This is how computers, digitized books, websites, and AI systems work.

Even higher levels of abstraction like music and movies are fully encoded in binary (digital audio and video files) with no loss of the original content.

Can you point to even one language that cannot be encoded this way?

184

I don’t understand why you would separate “hinges” from the rest of his metaphors, which seem to me to be circling the same concept. Whether he calls it a “hinge,” an “axis”, a “river-bed” or “foundation walls” the principle is the same: certain empirical propositions function like logical or mathematical ones (308, 340, 653, 657).

Wittgenstein’s insight that certain empirical propositions function like logical or mathematical ones applies equally to modern man and to hunter gatherers alike. Hunter gatherers and modern man have both had to take some empirical propositions for certain in order to doubt other empirical propositions.

185

I believe that Wittgenstein’s focus in both Philosophical Investigations and On Certainty is strictly on the grammar of language. Therefore, I think what you are referring to here - beliefs and/or actions “apart from language games” - falls outside the scope of his interest.

186

They are related but different pictures and cannot be reduced to a single function.

We trust what we see unless there are reasons not to. Where there is doubt it is not the doubts of the radical skeptic.

  1. “I know” has a primitive meaning similar to and related to “I see” (“wissen”, “videre”).
187

Ambiguity is common in language usage. it’s common in poetry, metaphor and parable. How does a true/false binary system deal with this sort of stuff?

That is your claim isn’t it, that since words can be represented by a binary system, this means that what words represent (meaning) is itself binary? Do you see the problem? It’s like saying that since the sensation of the colour red can be represented by the word “red”, this means that the colour red is experienced as words.

188

You haven’t provided any argument to support this assertion, so I don’t see why they would be different in practice. Whether he calls it a hinge, a river-bed, or a foundation, he’s talking about the same thing: some parts of our world-picture have to stay still so that other parts can move. Why do you see these as separate ideas rather than different ways of describing that same requirement?

I agree, but “seeing” isn’t a neutral starting point. You can’t have a “reason not to trust” what you see unless you already have a massive background of other things that you aren’t questioning. The logic is the same for the hunter-gatherer and the scientist: you have to be certain about the ground under your feet before you can start doubting the map in your hand.

189

Following TPF Guidelines, trolls should be ignored and reported. However, before I provide relevant discussion links, I’d like to respond.

This, to me, is unnecessarily antagonistic and dismissive. It isn’t the first time.

The exchanges about Wittgenstein and ‘On Certainty’ with its uncertainties as to the meaning of its metaphors…where does it lead us? In circles with overlapping pictures and perspectives.

Previously, I offered a tentative response in Sam’s TPF essay regarding Wittgenstein, his hinges and river-beds.
This followed effort, some research and a conversation with a helpful @Fooloso4.
I’m grateful but still not sure I have understood correctly - and I probably never will. So be it.

[TPF Essay] Wittgenstein’s Hinges and Gödel’s Unprovable Statements - The Philosophy Forum Archive

Sam’s essay gained a lot of attention, interest and some criticism.
I had not realised the time, word-spinning and image-painting spent on this. From Sam:

I decided to put my paper in this thread where it belongs. The paper tends to be a bit more precise than my general comments in this thread and elsewhere, which is why it’s important to write down one’s thoughts using more precise language. The area where my paper falls short is in not responding to potential criticisms.

The thread, started by Sam26, 5 years ago - An Analysis of “On Certainty”.

So much personal investment. Not only by Sam but other philosophers continuing to analyse, debate, trying to understand.
A never-ending challenge of patience and care.

190

I don’t think an argument is needed to support it. Wittgenstein’s description is sufficient to see that the movement around an axis, the movement of water on a river-bed, the shift of the bed itself, and the fixity and movement of a hinge are not the same.

In addition, the river-bed of thought and propositions exempt from doubt are two different things. What holds a hinge in place and nothing holding an axis fast are not the same.

Of course there are many things we do not question! Wittgenstein’s remarks are aimed at the radical skeptic for whom everything is doubted.

One issue in contention is whether hinges are, as Wittgenstein says, propositions that are exempt from doubt, or if they are some of the things or all of the things that are not doubted. And if this extends to other animals.

191

Sorry I took so long to respond. I understand your concern, and you’re right that Witt’s primary focus in both PI and OC is on the grammar of language. But I think the text itself pushes beyond that focus in ways Witt recognizes even if he doesn’t fully develop them.

OC 204 says “Giving grounds, however, justifying the evidence, comes to an end; but the end is not certain propositions’ striking us immediately as true, i.e. it is not a kind of seeing on our part; it is our acting, which lies at the bottom of the language-game.” Acting lies at the bottom of the language-game. Not language. Acting. The language-game rests on something that isn’t itself linguistic.

OC 110 makes the same move. The end of giving grounds is not an ungrounded presupposition. It is an ungrounded way of acting. Again, what’s at the bottom isn’t a proposition or a piece of grammar. It’s activity.

OC 7 points to certainty showing itself in action: sitting at someone’s bedside, telling a friend to take a chair. These aren’t linguistic acts. They’re bodily engagements with the world that display certainty without asserting it.

OC 128: “From a child up I learnt to judge like this. This is judging.” The child learns to judge before it can articulate rules. The judging comes first. The grammar comes after.

OC 538: “The child, I should like to say, learns to react in such-and-such a way; and in so reacting it doesn’t so far know anything. Knowing only begins at a later level.”

So Witt himself keeps pointing to something prior to language-games as their foundation. His focus is on grammar, yes. But his own investigation of grammar leads him to something beneath grammar, to acting, to reacting, to engagement with the world that is prior to the linguistic practices he’s describing. He can’t fully develop it because his tools are grammatical and what he’s pointing to is pre-grammatical. But he’s honest enough to point to it anyway.

What I’m doing in distinguishing prelinguistic hinges from linguistic ones is following where Witt points rather than staying only where his method is most comfortable. The beliefs I’m calling prelinguistic, the continuity of experience, the persistence of objects, the existence of other beings, these are the acting that lies at the bottom of the language-game. Witt tells us they’re there. He just doesn’t have the tools to say much about them, because saying is what his method investigates, and these beliefs show themselves in doing rather than saying.

192

Banno, I want to press on two points that come at this from a different direction than what I’ve said before.

First, consider the order in which certainty appears. A newborn grasps a finger. Is the newborn showing the truth of “here is a finger”? The newborn has no language, no concepts, no propositions. But the certainty is there in the grasping. The grasping isn’t waiting for a proposition to give it standing. It’s the original certainty, entirely non-propositional. Years later, when the child acquires language, the proposition “here is a finger” becomes available. But the certainty came first and the propositional expression came after. OC 128 says “From a child up I learnt to judge like this. This is judging.” The judging precedes the propositional capacity. And OC 538 says the child “learns to react in such-and-such a way; and in so reacting it doesn’t so far know anything. Knowing only begins at a later level.”

So what was the child’s certainty before it had language? If you say there was no certainty, you’re denying what’s plainly visible in how every infant engages with the world. If you say the certainty was there but wasn’t real certainty until it could be expressed propositionally, you’re imposing a requirement Witt never imposes. The certainty was fully operative before the child had a single word. The proposition is our description of what the child’s acting already had. Wittgenstein himself points to this priority: what lies at the bottom is acting, not propositions about acting.

Second, consider what your own position requires. If you insist that certainty must be propositional. But your propositions have to mean something. And OC 114 says if you’re not certain of any fact, you can’t be certain of the meaning of your words either. So the certainty that your words mean what they mean is a condition of your propositions having content. But on your view, that certainty would itself have to be propositional. And that proposition would need its words to mean something, which requires certainty about their meaning, which would itself have to be propositional, and so on. Your position generates exactly the kind of regress Witt keeps dissolving. Somewhere the propositional chain bottoms out in something non-propositional, i.e., the trained capacity to use words correctly, which is shown in practice, in acting within the language-game, not in further propositions about the meanings of words. You can describe that capacity propositionally after the fact, but the capacity itself isn’t a proposition. It’s what makes propositions possible.

The question isn’t whether we can express certainties propositionally. Of course we can. The question is whether the propositional expression is the certainty or a description of something that was already certain without it. The newborn settles that question.

193

Most thematic treatments of OC pull out one thread and follow it, which gives you a clean argument but loses the density. When you read OC in order, you see that Witt is working five or six lines simultaneously, viz., the grammar of “I know,” the regress of grounds, the mistake/mental disturbance distinction, the relationship between experience and judgment, the structure of framework disputes, the levels at which certainty operates. And each remark touches several of these lines at once. OC 130 alone connects to the regress problem, to the philosophy of experience, to the relationship between cause and ground, to the question of how practices are acquired, and to the deeper question of what lies at the bottom of all our epistemic activity.

And nobody has written those books. The secondary literature treats OC as a single-thesis text, either about hinges, or about Moore, or about the structure of epistemic practices. But the text is doing all of those things simultaneously, and it’s also pointing to things it can’t fully develop because its own method is grammatical and some of what it’s pointing to is pre-grammatical.

194

You’re treating “language” as if it meant verbal or propositional articulation, and then concluding that Wittgenstein is pointing to something prior to language.

I don’t think Wittgenstein means “non-linguistic” when he says “acting.” I think he means non-representative and non-propositional language, which includes non-verbal gestures and other behavior. A language game already includes actions, gestures, responses, training, and shared practices. Think of how Philosophical Investigations introduces language-games: builders shouting “Slab!” while handing stones, or a child being trained to respond to words. These are not purely linguistic in the narrow sense, they are patterns of coordinated activity in which words are only one element.

For Wittgenstein, grammar is not a set of explicitly formulated rules or propositions. It is the normative structure of our practices, the ways we go on correctly or incorrectly, the patterns of use that give expressions their sense. That structure is already there in the child’s trained responses.

So when Wittgenstein says “From a child up I learnt to judge like this. This is judging,” the point is not first non-linguistic judging , and later: linguistic/grammatical structure.

Rather, the point is first: participation in a practice with implicit norms (grammar-in-use) , and later explicit articulation of those norms in rules or propositions. In other words, what comes later is not grammar itself, but the ability to say what the grammar is.

1 Like