Notes on Ludwig Wittgenstein, Philosophical Investigations

I’ve been thinking about the topic of cultural change and recently published a  summary  on this Blog (11 September 2018). It struck me as I read through it  that there were things which led back to Wittgenstein, like my use of the idea of a “sample book”.  But I hadn’t read any Wittgenstein-related material let alone Wittgenstein himself since the 1980s. I thought I should retrace my steps and so I bought a copy of the Philosophical Investigations (PI) in the scrupulous modern edition (as a student in the 1960s and 1970s, I used the second edition of 1958 and later sold it when I felt I had no more use).

I have mixed feelings. I’m still not convinced. Here are some thoughts.

  Wittgenstein approaches his work in a territorial spirit. He was of course a Professor of Philosophy in a feudal university, so nothing new there. There is Philosophy and it isn’t Science. The sciences which are mentioned in PI - physiology, psychology, maybe psychoanalysis - look for causes whether in the present or the past. They are sometimes interesting, sometimes not. True, psychology as Wittgenstein knew it (behaviourism, fanciful “experiments ” using undergraduates) had very little going for it (See PI, #307; Part II, #371). Philosophy, in contrast, is concerned with concepts and grammar - though the PI locates those within “forms of life” and so sometimes reads like a theoretical or philosophical anthropology or sociology - and has been read that way. Three thoughts.

   As people took on board his work, they developed it in two directions. First, there was “conceptual analysis” which had mixed success. Extended into areas like ethics or politics, it read as a rather complacent and conservative rendering of what we think, disguised into what “we say” - as if the student needed to learn etiquette. “When we use the word “democracy” we mean …”. As a result, there is a lot of work in old issues of  philosophical journals which is no longer read by anyone. [An aside: in the late 1960s I attended Peter Winch’s seminars at King’s College, London. An old man sat in the corner, Rush Rhees [1905 - 1989], who had been one of Wittgenstein’s acolytes. Occasionally, he made a remark and each time I felt that, well, it wasn’t quite up to what one would expect in a graduate seminar. The remarks were banal. I put it down to his advanced age; but looking now at his dates, that can’t have been the explanation. ]

  The “grammar” when linked into the idea of “forms of life” fairly quickly yielded a sharp distinction between “sentence” and “utterance”, semantics and pragmatics. As early as 1960, J L Austin moved things on with How To Do Things With Words which gets us thinking about situated utterances and the many things they are used to do. In one direction, this helps clarify Wittgensteinian ideas about (say) the way in which understanding might be an achievement rather than a state. We say, “Now I understand!” when we feel we’ve achieved something, that we’ve now got it, and can carry on unaided. We are not reporting a state of mind. In another direction, Austin’s and then Grice’s work turned into contemporary and highly technical linguistic pragmatics.

From (say) the 1970s, the development of artificial intelligence, cognitive science, cognitive theory has often proceeded in ways which has ignored territorial boundaries. Many Wittgensteinians have been appalled. The proof of the pudding is in the eating. There have been extraordinary developments in both what we can do (using machines, implants, what have you) and how we investigate what “forms of life” might be natural to human beings. All this has come about by getting over the prissiness of Wittgenstein’s territorial approach and also past his unreflected thinking about "children", “training” and “ instruction”. (As a young man and a good Austrian Catholic, he clearly believed that error could be beaten out of children: go to Wikipedia for the details).
 Wittgenstein does give the impression that he thinks that it would make no difference if we had sawdust rather than brains between our ears. (see # 282 on the “babble of a baby” as nonsense; #293 “the box might even be empty”; #376;Part II, # 148 “The oddity of children’s drawings”). He also is at risk of turning local empirical truths into necessary truths, a trap into which later Wittgensteinains have fallen. Conversely, necessary truths are sometimes missed:

  # 249 “Lying is a language-game that needs to be learned like any other one”. And truth-telling?  There are plenty of theorists who think that truth-telling isn’t learnt; it’s what comes naturally to us and makes other language games possible.

   Wittgenstein insists that his investigations are not exercises in introspection or phenomenological description (I don’t think the word “phenomenology actually occurs in PI; it isn’t in the Index either). See for example # 232. But they often read as if they are those things and when they do it’s often boring, though there are those who have annexed Wittgenstein for  phenomenology or common sense. There are two specific shortcomings when he is working in this introspective / phenomenological way.

 I want to make a sharper use than does Wittgenstein of the distinction between judgement and intuition. We offer judgements as guidance for other people - You’ve added that up correctly; in English, the plural of sheep is sheep - but intuitions are simply reports of how things strike us, as in the Müller-Lyer illusion where we report that the lines look of unequal length.

 I also want to distinguish two sense of agreement. It is a curious fact that everyone agrees that the Müller-Lyer lines are unequal in length. But they agree this distributively, without discussion or participation in a “form of life” other than a natural one. The Müller-Lyer lines pick up a fact about human vision, not about human culture. In contrast, there are things we agree collectively. That doesn’t mean that we vote or even discuss (very much). All it means is that when we judge (for someone else’s benefit) that the plural of sheep is sheep we are confident enough about what other people think / judge to use our own judgement to guide (say) someone learning English. (Compare PI, # 234, 241; Part II,  #346, #351). Of course, we may get it wrong especially when we are over-reliant on local experience and think that everyone shares our dialect.