Wittgenstein: On Certainty

(Note: This was written as the third and final  in a series of three posts devoted to his late work, particularly Philosophical Investigations. The previous two posts are below in archives, though this can be read independently of them as well.)


After Philosophical Investigations was published posthumously in 1953, one last book, entitled On Certainty appeared. The book takes the form of a response, or series of responses to the common sense realist GE Moore. But the main line of Wittgenstein's response to Moore gives great insight into the later W's overall appraisal of philosophy and its quest for foundational knowledge.

The problem that faced GE Moore in the early 20th century was that of Idealists who said that there is no way to know that we are not actually immaterial beings in the mind of God, or something of that kind, despite all the appearances to the contrary. Thus we could not really know that our bodies exist; they could be mere images we see when God causes us to see them (or in this era a scenario like that in the Matrix might enable the same type of skepticism such that you think you are moving or eating, but you are really in a box with your brain connected to an elaborate virtual reality program). GE Moore thought that it was sheer nonsense to doubt such "self-evident knowledge" as being in possession of your own body. In sometimes-heated debates with skeptics and idealists Moore would raise his hand up, then raise his voice and exclaim "I KNOW that this is my hand." He was trying to say that knowledge of the real world does not really lead to an infinite regress (e.g. "How do you KNOW that is your hand, Prof. Moore?") but rather reaches a stopping point or foundation which is common-sense self-evident knowledge. "I know I am typing now" would be another such statement.
Why are these sorts of statements self-evidently the case? Because we cannot function in the everyday world if we seriously and sincerely start to doubt these basic articles of common sense facts, according to Moore. Only philosophers interested in abstract speculation sincerely argue about these things with counter-examples. In everyday life it would never come up. These common sense foundations are, said Moore, indubitable.

Moore's realism came to be seen as, and indeed called "Naive Realism." But Wittgenstein brought Moore's defense of realism back in a very interesting way. He said, in On Certainty, that Moore was correct to say that there was no reason to doubt that "This is my hand" while holding it up. Why? Because there is no language game in which such a doubt ever arises accept philosophy. This only shows how peculiar and detached from reality philosophy actually is, says W. So he agrees that it is "nonsense" to doubt whether or not the hand you are waving is really "your hand." But crucially it is also nonsense to say "I know this is my hand" while holding it up. Why? Because knowledge claims make no sense unless they can be doubted. If you cannot doubt something then the question of knowledge doesn't arise. The point of knowledge is to remove doubt. This was implicit in the PI when discussing sensations and pain. If I cannot doubt that I am in pain, then it makes no sense to say "I know I'm in pain." (Unless someone like a doctor doubts me, but that is not the same as me doubting my own sensations). What, would saying "I KNOW THAT I am in pain" add to the sentence, "I am in pain?" To understand this rhetorical question is to understand why W thinks philosophy creates psuedo-problems rather than clarifying things.

Wittgenstein holds that a person who doesn't believe they are in pain when they are feeling pain; or doesn't believe they have a hand when they are using one; or doesn't believe it is raining while they are walking through the rain with an umbrella; such people do not need training in the intricacies of epistemology, but rather something more like "a slap in the face, or a conversion experience." His image of a situation in which someone acts like the skeptic or idealist in these situations is one where the listener grabs the doubter and shakes him while saying something like,"For God's sake, come to your senses! Snap out of it. You're acting crazy." The only exceptions would involve idle speculation among philosophers or creative imaginings for the production of fantasy novels or movies. In everyday life, it doesn't come up. His view is that epistemology and doubt go together. The quest for certainty tends to generate doubts that otherwise do not arise. In ordinary circumstances we settle for what W calls riverbeds which are not beliefs or doubts, but the solid ground of pragmatic givens in everyday experience. I am here and you are there, and there's no problem, no intellectual issue at all. This is a lot like American Pragmatism but without any desire to convert it into a "theory" of experience. The experience is just fine as it is and requires no no theory, thinks W.  Any theory devised to reassure me that "my arm is 'really' my arm," or "I am in pain at T" or "It is raining" is absurdly superfluous, for W. There is no need to gild the lily of non-problematic experience with theories designed to "prove" that there are no problems. Anyone who is caught up in trying to prove that they are feeling pain while they are feeling it, for example, needs psychological help and not philosophy. They need the "Philosophical Therapy" of the Philosophical Investigations which, as covered in previous posts, is designed to convince the suffering philosopher that philosophy itself is a kind of illness from which s/he can only be cured by ceasing to engage in the activity. Perhaps, W saw himself as something like a "a philosopher in recovery!" And here we come to the end of the line of W's later work.
_____________________________________________________________________
 
Relevant Readings:

-On Certainty: Ludwig Wittgenstein, trans. Dennis Paul & G.E.M. Anscombe, Oxford Basil Blackwell, 1969

- Letures & Conversations on Aesthetics, Psychology and Religious Belief: Ludwig Wittgensetein, ed. Cyril Barrett, Univ. of Cal. Press

-An Introduction to Wittgenstein's Philosophy of Religion: Brian R. Clack, Edinburgh Univ. Press, 1999
-------------------------------------------------------------------------------------------------------------

Below is a humorous snippet from Derek Jarman's film, Wittgenstein, which deals with just the issues of this post. He has both Russell and Keynes as part of his audience, and storms out of the room in a state of vexation. (He did do this during his lectures at times).

Following the original OP from 2016 (reposted above) was  a dialogue between  a reader/commenter  (who now shows up only as "Guest" in the Disqus comments) and myself. I think it's worth including here, as the content and meaning of W's On Certainty, was made more elaborate. Also, the possible relation of W's post-1929 thought to the earlier thought of American Pragmatists like Peirce was discussed.


Guest:

Maybe I missed it, but it seems we're leaving out sort of the most crucial part of Wittgenstein's view here. For Wittgenstein, radical doubt/skepticism is structurally incoherent. When we claim we know or doubt some proposition, we are submitting that proposition to an evaluation: we are testing it to see if it can hold up to the evidence. But in order to test or evaluate a proposition, we need to do so in terms of things which are taken for granted: if I'm testing whether there is an elephant on my couch by looking around my living room, I am taking it for granted, for instance, that my eyes are working properly. Without taking this for granted, testing "there is an elephant on my couch" by looking at my couch would not work. In order to evaluate some proposition, we must take other propositions for granted: else we don't have any standard or criteria to compare them to. Something must hold firm. And the things we take for granted, like that our eyes work properly under normal circumstances, that we have hands, that our body has never left the Earth, and so on, forms our epistemic bedrock- the background framework of beliefs and assumptions against which we test epistemic claims.
On the other hand, the whole point of radical, universal doubt is that nothing is taken for granted. Universal doubt removes the very thing it requires to function, if Witt's epistemological picture here is correct: with no epistemic bedrock, we have no criteria or standard for comparison, and no basis for accepting or doubting anything. So universal doubt is incoherent and ultimately self-defeating, for Witt. And Moore is mistaken in saying that he knows he has two hands, for the same reason that the skeptic is mistaken in doubting he has hands- that we have hands is part of the background framework we use to evaluate knowledge and doubt claims, it cannot coherently be itself subjected to that framework. We don't know that we have hands- we take it for granted that we do, and such foundational assumptions provide the necessary starting point for epistemic evaluation, such as knowledge and doubt.

PD:

I don't think the points you raise were missed, at least I hope not as I did my best to summarize them (and other points) in a short and accessible way. Anyway, your reading sounds right to me. The "background framework" of assumptions against which we check knowledge claims whenever we do is a good description of the "riverbeds", "banks", "scaffolds" etc., which metaphors W uses for these presuppositions you are discussing. Importantly, they are internal to, and embedded in cultures (forms of life and associated language games). He uses metaphors like "beds and banks" to avoid any misinterpretation of him setting up yet another theory of knowledge, as I point out in the post. I thought the ideas you are mentioning were conveyed in passages like this one above:

" In ordinary circumstances we settle for what W calls "riverbeds"
which are not beliefs or doubts, but the solid ground of pragmatic
givens in everyday experience. I am here and you are there, and there's
no problem, no intellectual issue at all. This is a lot like American
Pragmatism but without any desire to convert it into a "theory" of
experience. The experience is fine as it is and needs no "theory." Any
theory devised to reassure me that "my arm is 'really' my arm," or "I am
in pain at T" or "It is raining" is absurdly superfluous, for W. There
is no need to gild the lily of non-problematic experience with theories
designed to "prove" that there are no problems."

Statements like "The Earth was here before I was" and "I know I have a hand before me" may be bedrock, but I 'm not sure I interpret them as *epistemic* bedrock, if by that there is any implication of foundationalism. I think he sidesteps (like Pragmatists) epistemology and ontology, predicating knowledge on shared norms within cultural-linguistic communities. I think you may have the same idea in mind, but only because you used the phrase *epistemic bedrock* I wanted check whether or not we can agree that if this is epistemology then it is social epistemology at bottom.

The question of how satisfactory such an account (and, of course, there's much more to On Certainty than either of us can encapsulate here) may or may not be is one that divides interpreters into myriad camps. There are a lot of different answers out there. But as far as the overview skipping your point, I don't think that happened. If it did, I'll edit in some stronger statement of the notion of presuppositions that allow language games to "go on."

Guest:

I don't say "epistemic" here because Witt was a foundationalist, but because the issue he's concerned with is justification (including of epistemic attitudes like knowing, doubting, believing, etc.) and the problem of regress. On the other hand, Witt's solution does share one striking feature with foundationalism: like foundationalists, Witt holds that the regress of justification eventually terminates- when we reach those framework assumptions which we do not, and even cannot, subject to evaluation, the chain of justification ends. But unlike in foundationalism, these terminal judgments are not intrinsically justified (due to self-evidence or immediate presence to our awareness or what have you)- being part of the background framework required for justification in general, they are themselves outside the scope of justification, and in this sense are non-epistemic. In any case I don't see anything wrong with construing this as "social" epistemology, given Witt's somewhat novel (at the time at least) emphasis on the communal nature of language and even judgment.

And though your quoted passage is perfectly consistent with what I've said, I'm still not sure that someone who wasn't already familiar with Witt's view on the matter would come away understanding why there are "no intellectual issues" with these framework judgments (i.e. because of their role in the structure of judgment and evaluation). But either way I don't think you need to edit your OP at this point: I imagine that if the point wasn't clear enough to begin with, it should be fairly clear with these additional comments.

PD:

Thanks. It's hard for me to take on the perspective of an outsider, maybe I'll add a sentence or two, though I think that after 15 days, and with comments like these, your probably right that it's unnecessary.

Also, I pretty much figured you meant what you just explained when using the term "epistemic."
I'm not sure how new the communal conception of knowledge was to be honest. As I wrote in the OP, I think he ends up in a very similar position as some of the American Pragmatists like Dewey and GH Mead, though they weren't against systems and theories, nor did they think philosophy was always misleading; rather they looked at the history of phil piecemeal.

Guest:

Sure, and I only say that Witt's emphasis on the social nature of knowledge/judgment was "somewhat novel", i.e. in the sense that it was waaaaaay outside the norm (which was to consider individual knowers in isolation from their environment, including/especially their social environment), and particularly so for the hardcore logical/analytic tradition Witt found himself in at Cambridge... not because Witt was the first to consider it. Maybe "unusual" would've been a better word for it.

PD:

Absolutely. I didn't mean to nitpick. I'm interested in the history of ideas, and it turns out that W's friend Frank Ramsey--before he died-- discussed pragmatism with Wittgenstein. Ramsey was doing work on Peirce and was familiar with Dewey et al. What is interesting about all this is that you're right in suggesting that Wittgenstein was out of step with the analytic world. He was seen by some of his peers as having gone astray. Russell was particularly scornful saying that W's later work was "invented to save W the trouble of thinking" or words very close to those.
Russell had heard some of these ideas before though. His debates with John Dewey were sustained and the divisions sharp ( in terms of theories; they met and actually became friends who disagreed about philosophical matters). Point is, there's been all this talk since Rorty about "neo-pragmatism" and I think in many ways Wittgenstein was already there in the mid- late 30s when he arrived at most of what ended up in PI. Interestingly it spawned a minor, if forgotten, revival of pragmatism at the time, and a very similar field in linguistics with Paul Grice and "Pragmatics." This leitmotif of finding meanings in the various uses of words in social contexts is definitely part of Mead and Dewey, who had also railed against the picture theory approach, the residual effects of Platonism (essential definitions etc.) and Cartesian dualism etc. I'd say there's a good bit of "family resemblance." But maybe I'm biased as I did a lot of work on the pragmatists in the past.
Of course none of this means that either pragmatists or Witt are right about many of the things argued. And indeed, as you point out on another thread, there are many interpretations of W (and for that matter Dewey). Still, the general contours are strikingly similar imo.
Though I do not like this author's interpretations of Peirce, James or Dewey, philosopher, Cheryl Misak wrote on the historical connections between the pragmatists and Ramsey---> Wittgenstein. She may overstate the case, but examines letters and dusty manuscripts from Peirce, James, Dewey, Moore, Ramsey, Wittgenstein and others. Anyway, it's an interesting historical connection.https://global.oup.com/academic/pro...

PD: (Following up on last comment about Witt. and Pragmatism)

The following passage is from W's notebooks of 1930 which Misak published for the first time in its entirety, and is one of several intriguing documents re: the prag-Witt connection. It is Wittgenstein MS 107 fr. 1930 (2 days after Ramsey's death) and contains an explicit reference to "meaning as use" as an idea coming from pragmatism. 1930 was the year of Wittgenstein's so-called "U-turn", and these are among the notes that represent a decisive change of direction which culminates in PI and On Certainty.

-From Cambridge Pragmatism pp 239-40. Wittgenstein: MS 107-

"Sentences--what we ordinarily call sentences in everyday use-- seem to me to work differently from what in logic is meant by propositions, if there are such things at all. And this is due to their hypothetical character. Events do not seem to verify or falsify them in the sense I had originally intended--rather there is still, as it were, a door left open. Verification and its opposite are not the last word.

Is it possible that everything I believe to know for sure--such as that I had parents, I have brothers and sisters, that I am in England--that all of this should prove to be false? That is, could I ever acknowledge any evidence as sufficient to show this? And then could there be even more reliable evidence to show that the first kind of evidence was deceptive?

When I say, "There is a chair over there," this sentence refers to a series of expectations. I believe I could go over there, perceive the chair and sit on it, I believe it is made of wood, and I expect it to have a certain hardness, inflammability etc. If some of those expectations are disappointed, I will see it as proof for retaining that there was no chair there. [This sounds almost like a quick summary of the main arg. in Peirce's How To Make Our Ideas Clear.]

Here one sees how one may arrive at the pragmatist conception of true and false: a sentence is true as long as it proves to be useful. Every sentence we utter in everyday life appears to have the character of a hypothesis. A hypothesis is a logical structure. That is, a symbol for which certain rules of representation hold. The point of talking of sense-data and immediate experience is that we are looking for a non-hypothetical representation.

But now it seems the *representation* [sic] loses all its value if the hypothetical element is dropped, because then the proposition does not point to the future any more, but it is, as it were, self-satisfied, and hence without any value. Experience says something like, "It is nice elsewhere too, and this is where I am anyway." And it is through the telescope of expectation that we look into the future.
It makes no sense to speak of sentences if they have no instrumental value. The sense of a sentence is its purpose.

When I tell someone, "There is a chair over there," I want to produce in him certain expectations and ways of acting. It is terribly hard here not to get lost in questions that do not concern
logic. Or, rather, it is terribly hard to find one's way out of this tangle of questions, in order to contemplate it whole, from the outside." -Ludwig Wittgenstein (Fr. 1930 Notebooks of Wittgenstein as reprinted in Cheryl Misak's book, Cambridge Pragmatism: pp. 229-30)

That's MS 107 from start to finish. I find all of this--though academic-- just fascinating. The official story is that Pragmatism died with Dewey himself, if not sooner (around 1950). "Sloppy" and "crude" Pragmatism was replaced by the icy logic and rigor of Analytic Philosophy. But PI was published in 1953, and focused anglo/analytic philosophy on decidedly social and pragmatic ideas and themes, leading to a cross-disciplinary set of movements emphasizing everyday or "ordinary" language, speech act theory, linguistic analysis of "pragmatics" or language *use*, and Wittgenstein-inspired social science, driven largely by his student, Peter Winch, among others. This is all years before Rorty's Consequences of Pragmatism which brought the pragmatists back into dialogue with Wittgenstein, Heidegger, Quine, Goodman, et al. But had the divide ever been as sharp as many mainstream analytic philosophers had imagined? I wonder.

Any thoughts?

Guest:

Yeah that is really interesting- obviously one can find areas of overlap and agreement between pragmatism and the PI, but not quite like this: parts of this passage sound, as you note, as if it could have been written by Peirce himself! And for my part, I tend to think that these historical narratives we get about AP [i.e. Analytic Philosophy- ed.] killing off pragmatism, or the continental/analytic split, tend to be overly simplified and often just misleading- useful if you're trying to give a "history of philosophy in 60 seconds or less", but not that useful otherwise. There was alot more common ground than is often supposed; one can even find hints of pragmatism in the positivists, and in some ways LP represents analytic philosophy taken to its logical extreme. I suspect if you looked at everyone's manuscripts, you'd find alot more of this behind-the-scenes dialogue going on, like what we see here with Wittgenstein.

And I guess its sort of fitting that Witt could be viewed as a kind of bridge between pragmatism and AP/OLP,  [OLP = Ordinary Language Philosophy - ed.]the same way he is sometimes viewed as a bridge between AP and continental philosophy- I remember Walter Kaufmann wrote about GE Moore and Soren Kierkegaard each being "half a Socrates", and Witt being the closest we get since Socrates to fusing these two disparate tendencies or philosophical spirits. The fact that this buys into/depends on the simplistic picture of the AP/continental split is sort of beside the point: Wittgenstein displayed qualities, and was concerned with topics, that were all over the philosophical map, and so is natural to view as a bridge between philosophers who were a little more domain-specific. And I suppose it shouldn't be overly surprising, when we consider how much of an odd duck Witt was to begin with: a Vienesse occupying a chair at Cambridge, writing in German about characteristically British philosophical concerns!