The illusion of certainty

4 July, 2018 at 22:37 | Posted in Statistics & Econometrics | 3 Comments



  1. Ludwig Wittgenstein
    On Certainty (Uber Gewissheit)
    ed. G.E.M.Anscombe and G.H.von Wright

    • I haven’t read it all, but it seems that Wittgenstein argues, more or less persuasively, that it is pointless to say ‘I think’ before stating beliefs that are as well grounded as any are in everyday life. That is, in so far as we are involved in a language game, it is redundant to express a certain minimum irreducible uncertainty. But does Wittgenstein address the problem of updating beliefs?

      According to Boole, we hardly ever falsify ‘atomic’ beliefs, but only compound ones, and so we are normally at liberty to choose which components we suspect. We would ordinarily only suspect those components that we are uncertain about. What would Wittgenstein have us do when we find a compound statement that is logically deducible from certain facts falsified? It seems to me that this if we accept Wittgenstein’s characterisation then scientists actively seek this kind of dilemma, and sometimes find it.

      Most of his examples seem unfortunate, in that modern technology now gives us ample reason to doubt what he often thinks certain. Thus with hindsight we see that if Wittgenstein had been a fan of science fiction he might have not been certain. Does he give any examples that are not technology dependent, so that one could not possibly conceive of any circumstances in which he might have been wrong?

      On a broader point, Wittgenstein mostly argues as if language is used within a fixed context, as if there were never any innovation. So I hope that he was wrong! (I prefer Russell et al, as on my blog.)

Sorry, the comment form is closed at this time.

Blog at
Entries and comments feeds.