Wittgenstein on Certainty: Should We Be Skeptics?

SUMMARY

• Wittgenstein’s “On Certainty” was a response to G. E. Moore’s essays which aimed to identify propositions that are beyond skepticism.
• Wittgenstein examined the idea that certain propositions serve as the bedrock or foundation for other empirical statements. He likened these foundational statements to a riverbed that must remain stable for the river to flow.
• For Wittgenstein, the certainty we feel about some propositions stems from their deep integration into our daily activities or “forms of life.”

An important part of Wittgenstein’s posthumous work is devoted to ideas about certainty, the things we seem to be certain of, and how those things interact with knowledge and skepticism. Much of this has been gathered under the title On Certainty (1969), a winding text full of examples, self-doubt, and conflicting directions, worked on until the last days of Wittgenstein’s life and left decidedly unfinished. Nonetheless, even in its incomplete state, On Certainty poses serious problems for the philosophy of language, for epistemology at large, and particularly for skepticism.

Ludwig Wittgenstein notoriously published only one book in his lifetime, the Tractatus Logico-Philosophicus (1921). Everything else now studied as his work was posthumously published, including On Certainty.

The notes that comprise On Certainty respond to G. E. Moore’s “A Defence of Common Sense” (1925) and “Proof of an External World” (1939). In these two essays, Moore tries to establish the existence of a set of propositions that are beyond the scope of skepticism—in short, of things that are not tautologies, but that we nonetheless cannot doubt. For Moore, we cannot doubt that we know them, but Wittgenstein treats these “Moorean propositions” quite differently, and seeks to establish them as a category of their own, one which is ultimately non-epistemic.

The most famous example, given in “Proof of an External World”, is “Here is a hand,” but Moore gives many other examples, including “The earth existed for a long time before my birth” and “I have never been far from the earth’s surface.”

Wittgenstein is fascinated by these propositions because they are so structurally different from the kinds of things philosophers tend to talk about as being invulnerable to skepticism. Moorean propositions are not at all the kind of thing that would have satisfied Descartes, but nonetheless—Wittgenstein thinks—expose the limits of skepticism.

Hinge Propositions and Linguistic Frameworks

Part of what is so counter-intuitive about the way that Wittgenstein speaks about Moorean propositions is that he introduces the possibility that over time, and in different contexts, the same proposition can move between the certain and the uncertain.

Indeed, many of the examples given by both Moore and Wittgenstein were at some point far from certain, tentative empirical observations in need of extensive verification and investigation. That water boils at 100°C (at 1 atmosphere of pressure) might now be one of those things I simply have to know in order to investigate the world more generally, but at some point, it was an observation or a discovery.

Because of this, Moorean propositions look like empirical ones but fail to behave like them. In asserting this, Wittgenstein diverges from Moore’s own view, which treats these propositions as still basically empirical, albeit exhaustively evidenced. Instead, these propositions become the background or bedrock for actual empirical propositions, and for all manner of other linguistic activity. Wittgenstein himself uses the analogy of a riverbed: the riverbed is not absolutely fixed, and is in large part composed of what was once flowing in the river itself, but it must remain in place in order for the river to flow.

What Wittgenstein describes is a process of sedimentation of propositions, whereby what were once empirical propositions become indubitable, and come to take part in frameworks that we use for other purposes, or to investigate new observations. This sedimentation is not always diachronic, either. There are propositions that in one context, or framework, have to stand fast in order for the rest of the language game to work, but that in another context are subject to doubt or verification, even if the referent of that proposition doesn’t seem to change.

What cannot happen, however, is that some context or language game arises wherein all the propositions or beliefs of the interlocutors are subject to meaningful doubt all at once. These propositions are like “hinges” for the rest of language, points that have to remain fixed in order for other things to move around them.

Though none of the questions Wittgenstein sets out to answer are really settled at the end of On Certainty, the idea that some propositions are “held fast” by others in the course of linguistic activity is one that keeps cropping up. Wittgenstein writes:

“The child learns to believe a host of things. I.e. it learns to act according to these beliefs. Bit by bit there forms a system of what is believed, and in that system some things stand unshakeably fast and some are more or less liable to shift. What stands fast does so, not because it is intrinsically obvious or convincing; it is rather held fast by what lies around it.”
On Certainty, §144

As plausible as this account of Moorean propositions and their special epistemological status sounds, it still requires an explanation of why certain propositions, and not others, gain this quality of certainty in a given context.

Wittgenstein entertains a number of possible answers to this question, one of which is that our certainty in some of these propositions derives from their enmeshment in our activity, or in a “form of life” (OC §358). It’s a phrase that is picked up often in scholarly commentary on On Certainty, particularly in Avrum Stroll’s writing on Wittgenstein, but—like so much of the material in On Certainty—raises many more questions than it answers.

By “form of life,” Wittgenstein points to the ways in which Moorean propositions seem to be non-epistemic, contrary to Moore’s own assessment of them. Wittgenstein does not, unlike Moore, think that we know things like “the earth is very old,” our relationship to them is instead a kind of non-epistemic, or perhaps pre-epistemic certainty.

For Wittgenstein, it is a fundamental philosophical misstep to abstract all forms of sureness or behavioral reliance to knowledge. In §477 of On Certainty he asks, “why should the language-game rest on some kind of knowledge?” Why shouldn’t certainty be a thing unto itself? More than this, however, he struggles to say. He describes the process of interrogating Moore’s ideas as being like seeing something clearly from a long way off, and then getting closer and finding the picture blurred and confusing. By the end of On Certainty, Wittgenstein remains far from any concrete ideas this special non-epistemic category that characterizes certainty.

We are sure of the Moorean propositions in the way that we are sure we can rely on objects in our environment to behave in certain ways before we have any conscious knowledge of them or their properties. Moore thinks this sureness is the result of a huge bank of empirical evidence, but Wittgenstein entertains the idea that it’s something else altogether, sometimes described in terms of context and culture, but in §359 of On Certainty described as “something animal.”

Elsewhere, however, Wittgenstein rejects the idea that Moorean propositions are in some way innate to humans as animals. Instead, he repeatedly returns to the idea that these propositions are held fast by the games and activities they are used for—that they are in some way contingently certain, but certain all the same. Wittgenstein writes:

“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.”
(OC §248)

It is important to stress that the kind of indubitability Wittgenstein is talking about is not, for him, a function of psychological limitations. As Andy Hamilton puts it in his guide to On Certainty:

“For Wittgenstein, being ‘exempt from doubt,’ in the case of Moorean propositions, is not a psychological phenomenon, like someone being unable to doubt their son’s innocence. It is a matter of logic, in Wittgenstein’s broad sense, not one of psychology.”
Andy Hamilton, Wittgenstein and On Certainty, 2014

That Wittgenstein conceives of doubt and its limits this way, however, is itself a consequence of his critique of skepticism. This critique, which appears in several guises throughout On Certainty, is foundational to it as a text and hinges on the idea that skepticism is often nonsensical.

By nonsensical, he usually means that expressions of apparent doubt are either liable to bring everything else crashing down with them, including the framework of their own expression (making them self-defeating), or that they are simply meaningless.

Part of what makes the idea of psychological versus logical doubt-limitations so knotty is that this latter critique relies on the idea that meaningful doubt consists in imagining a counterfactual state of affairs, and imagining at least looks like a psychological activity. Indeed, we might wonder whether the limits of imagination are exactly what prevents a parent from doubting their child’s innocence.

Additionally, there are times when Wittgenstein’s approach appears psychologically descriptive, seeking to uphold certainty by observing the fact that we perform actions without doubting the physical facts upon which our actions depend.

When Wittgenstein gives the example, “Why do I not satisfy myself that I have two feet when I want to get up from a chair? There is no why. I simply don’t. This is how I act.” (OC §148), it is not a function of logic that I do not question my feet. We can readily imagine a person whose feet have been recently amputated needing to ask themselves precisely such a question; what is being highlighted here is surely a kind of psychological habit.

What Wittgenstein seems to mean by the logical nature of certainty with respect to Moorean propositions is that these propositions function as something like rules in a given language game. On this view, when we “break” rules, we really just enter into a different language game, one in which the proposition we are doubting is not a hinge proposition.

It becomes clearer, on this view, how Moorean propositions and the things that we rely on certainty in them for might be interdependent. If propositions act like rules for certain kinds of conversation or activity, we can both not know them and nonetheless be sure that they are the rules relevant to the present game.

It’s a conception of language that radically reduces the power of skepticism, confining its activity to a particular kind of language game, while everywhere else we really are sure of the things we feel sure of. For Wittgenstein, it is not just that we do not ordinarily doubt things like the existence of the chair over there, or the fact that my name refers to me; it is that we cannot—ordinarily—doubt these things. The course of life, and our use of language (as well as our non-linguistic interactions with our environment), relies on the certainty of propositions we would never think to even articulate, never mind doubt.