My reply on a blog

Here's my reply to a blog's post about Wittgenstein's On Certainty.

I think it quite a naive reading of W’s On Certainty to say that the Moon Landing was a refutation of his views. Keep in mind that, as a young man, W was a year away from his doctorate in aeronautical engineering and he probably was well aware that someday, maybe even in the near future, people would walk on the moon.

Let me just say that I believe W’s work on doubt to be very insightful and even correct. I may be taking some exegetical liberties in my interpretations but such liberties are justified when dealing with W because he was such a horrendous writer. Horrendous writer may be but I still find many of his ideas quite profound though, as with you, I also think that much of his esteem among many philosophers to be from his “bewitching” personality as opposed to the quality of his views.

This is how I interpret those passages. W was making an analogous move to Gettier in his seminal work on knowledge. As Gettier showed that justified true belief (JTB) is not necessarily a well defined and philosophically robust definition of knowledge but may be loosely termed such, W showed that just by raising possibilities counter to some forms of common sense is not a philosophically relevant form of doubting.

Gettier showed that it takes more than just JTB because the justification criterion must occur in a certain way. It can’t just be any ol’ sort of justification because one can be justified in some belief coincidentally. W’s example of being correct about dreaming and being on the moon, I believe, is meant to show that same point as it relates to doubt.

Mere raising of possibilities such as the possibility that one is now dreaming or being on the moon yesterday when one doesn’t believe oneself to be does not destroy one’s belief or even certainty in such beliefs. W’s reasoning for this conclusion as is detailed is by several examples and arguments.

He argues that raising odd possibilities tend to cancel out. For example, the possibility of someone being on the moon in 1949 (when On Certainty was written) when he tells you that he was is more than canceled out by the possibility of his being crazy or by him using those words in a way that you have misunderstood him. He could be in a different linguistic context than what you presume him to be in of speaking truthfully (such as sarcasm, telling a fictional story, etc) or he could be speaking in a language you don’t even understand. These unlikely possibilities cancel out any such *even more* unlikely possibilities of someone being on the moon in 1949 (which was impossible given the technology of the time). Raising of mere (logical) counter possibilities does not entail one should justifiably lower the probability of any belief as doubt would require. This was, I think, W’s point.

I think W was also making some kind of distinguishing remark about raising a counter possibility and the *feeling* of being doubtful and being doubtful itself. Think of this example (mine, not W’s but I think captures what he meant). Imagine a drug that is able to induce a *feeling* of doubt even in your most confidently held beliefs (maybe in logic or math). Now just because you feel doubtful doesn’t make it’s doubtful that that belief is false. Consider W’s quote:

“From its seeming to me – or to everyone – to be so, it doesn’t follow that it is so.”

I think W was making the point that true doubt requires justified reasons (and not just coincidentally justified) and not just a feeling of doubt or a mere possibility. That may have been an analytic truth to him as Gettier’s point that it is just an analytic truth (“built in the very meaning”)that knowledge requires more than JTB. Hence someone who claimed to doubt based on some mere possibility and a “doubtful feeling” simply don’t know what to doubt means (though they may in a philosophically irrelevant sense be correct in that usage much as JTB may be correct in some common usage).

To be doubtful you would have to be justified in the doubt in just the right way (just like you would have to be justified in just the right way for a belief to be knowledge). And that would entail you being able to answer a host of questions to which all our common sense beliefs may be connected in some way to that belief. See W’s quote:

“For this demands answers to the questions “How did he overcome the force of gravity?” “How could he live without an atmosphere?” and a thousand others which could not be answered.”

The more certain and “common sense” the belief, the more questions you’d have to answer which are connected to it. Compare this view with Quine’s web of beliefs and I think there’s quite a similarity. With Quine’s web, the closer it is to the center of the web, the more it will disturb the outer edges of that web and more of the whole web will have to be replaced by a more justified system or web of coherent beliefs.

W also notes that normal doubt requires some standard of weight to measure certitude or confidence. We doubt some belief X by using a stronger standard Y. Y provides the standards to put X into question. Even if X ultimately ends up being correct, if we had a stronger standard in a “common sense” Y to justifiably put X into doubt, X would only end up being correct *by coincidence* much like someone who says in his dream “it is raining” may be only coincidentally correct about the status of the non dream world.

Someone who doubted Y without good reason to by holding fast to some contradictory belief X would not be in fact correct to doubt Y in favor of X (compare with the Matrix where we are told that we don’t live in a real world but a dream world where evil robots are controlling our dream environment). He would not have an epistemic right to doubt Y because he could not provide a stronger standard to do so (which would entail coherently and justifiably answering a host of questions related to common sense Y).

I am now convinced that raising mere possibilities cannot justify loss in certainty in some belief as W claimed for much of the same reasons he argued. Doing so would be more akin to artificially inducing a feeling of doubt rather than real doubting. Lowering of confidence in our common sense beliefs (or any belief) has to occur in just the right way much as justification has to occur in just the right way for knowledge. This would still allow for the normative claim that it is still good practice to question all our beliefs but just that when that is done without good justification, it is not real doubt.