¹ I have greatly benefited from discussions with Norman Malcolm and John Cook, even though they do not agree with much of what I say here. Their views prompted me to try to support my own views. I thank each of them for allowing me to see unpublished material and for various references to Wittgenstein's writings that I had not seen. See also my discussion, ‘What if Something Really Unheard-of Happened?’ and Malcolm, 's reply in Philosophical Investigations, 04 1990, 154–168Google Scholar and Cook, 's article, ‘The Metaphysics of Wittgenstein's On Certainty’ in Philosophical Investigations, 04 1985, 81–119.Google Scholar

² I will be citing passages from the following writings and lectures of Wittgenstein, 's: Philosophical Investigations (PI), Anscombe, G. E. M. and Rhees, R. (eds), Anscombe, G. E. M. (trans.) (Oxford: Blackwell, 1953).Google Scholar

On Certainty (OC), Anscombe, G. E. M. and von Wright, G. H. (eds), Paul, D. and von Wright, G. H. (trans.) (Oxford: Blackwell, 1969).Google Scholar

Zettel (Z), Anscombe, G. E. M. and von Wright, G. H. (eds), Anscombe, G. E. M. (trans.) (Oxford: Blackwell, 1967).Google Scholar

Remarks on the Foundations of Mathematics (RFM), von Wright, G. H., Rhees, R. and Anscombe, G. E. M. (eds), Anscombe, G. E. M. (trans.) (Oxford: Blackwell, 1956).Google Scholar

Philosophical Grammar, Rhees, Rush (ed.), Kenny, Anthony (trans.) (Berkeley: University of California Press, 1974).Google Scholar

Remarks on the Philosophy of Psychology (RPP), Anscombe, G. E. M. and von Wright, G. H. (eds), Anscombe, G. E. M. (trans.) (University of Chicago Press, 1980), Vol. I.Google Scholar

The Elite and Brown Books (Oxford: Blackwell, 1958).Google Scholar

Tractatus Logico-Philosophicus, Pears, D. F. and McGuinness, B. F. (trans.) (London: Routledge and Kegan Paul, 1961).Google Scholar

Culture and Value, von Wright, G. H. (ed.), Winch, Peter (trans.) (University of Chicago Press, 1984).Google Scholar

‘Cause and Effect: Intuitive Awareness’ (C and E), Philosophia, R. Rhees (ed.), Peter Winch (trans.), Vol. 6, Nos 3–4, 391–408. Notes taken by students.Google Scholar

‘A Lecture on Freedom of the Will’ (FW), Philosophical Investigations, Vol. 12, No. 2, 85–100Google Scholar. Notes by Yorick Smythies.

Wittgenstein's Lectures: Cambridge, 1930–32 (WL, 30–32), Lee, D. (ed.) (University of Chicago Press, 1982).Google Scholar

Wittgenstein's Lectures: Cambridge, 1932–35 (WL, 32–35), Ambrose, Alice (ed.) (University of Chicago Press, 1982).Google Scholar

Wittgenstein's Lectures on the Foundations of Mathematics (LFM), Diamond, Cora (ed.) (Ithaca, N.Y.: Cornell University Press, 1976).Google Scholar

Wittgenstein and the Vienna Circle (WVC), McGuinness, Brian (ed.), Schulte, Joachim and McGuinness, Brian (trans.) (Oxford: Blackwell, 1979).Google Scholar

³ In ‘Causality and Determinism’, 144–45Google Scholar (from The Collected Philosophical Papers of G. E. M. Anscombe (Oxford: Basil Blackwell, 1981), Vol. IIGoogle Scholar. Anscombe offers an example of a ‘non-necessitating cause’. This is an interesting example because another description of the ‘cause’ will change the cause of the explosion from a non-necessitating cause to a necessitating one. The example is Feynman's: ‘… a bomb is connected with a Geiger counter, so that it will go off if the Geiger counter registers a certain reading; whether it will or not is not determined, for it is so placed near some radioactive material that it may or may not register that reading’. She remarks that the motion of the Geiger counter's needle is caused, ‘and the actual emission is caused, too’ but ‘… all the same the causation itself is, one could say, mere hap.’ But although Anscombe does not point it out, there is a necessitating cause here as well, namely the radiation at a level sufficient to make the Geiger counter set off the bomb. What is ‘mere hap’ on this reading is whether the cause will materialize; i.e., whether the cause and effect—the causation—will materialize. This example could be construed as an illustration of ‘there are no causes in nature’, since the account of a sequence of events changes from indeterminism to determinism depending on how the crucial event is described, or happens to be described.

Common examples of non-necessitating ‘somethings’ (I take it) would be such things as insults and praise. Such a ‘something’ would affect behaviour sometimes as a cause would and sometimes as a reason would. That is, sometimes the reactions to such a thing would be a result of reflection and sometimes not. There are not only causes of anger but reasons for anger in the absence of causes.

⁴ Clearly, the negation of an empirical statement must make sense, but does a proposition that makes sense describe a possible state of affairs or event? In lectures in which Moore took notes, Wittgenstein says that to say that p makes sense means that p is ‘logically possible’. (Moore, , Philosophical Papers, 275)Google Scholar. It makes sense, though it is false, to say that an iron nail scratches glass. Although it is logically possible, we know that it will not.

⁵ In Lectures 1932–1935Google Scholar, there is this puzzling remark: A man struggles against having his hand put in the fire:

If he says he ‘is likely’ to be burnt, this seems to be a new suggestion. The reason why it's likely is that he has-been burnt a thousand times before. But we can omit the ‘likely’; for whatever reason he gives, he still may be burnt or not.

Perhaps the point is that it is not a priori ‘likely’ that he will be burnt. That is, it is not ‘likely’ in the sense in which we say, ‘It's equally likely that I will throw heads or tails.’ (Philosophical Grammar, 224)Google Scholar This, I take it, is what is implied by the remark that follows; ‘Trains of reasoning go on and then something either happens or does not’ (Lectures, 1932–1935, 88)Google Scholar. We don't experience a logical necessity for events to occur or a logical likelihood that they will.

Or, of course, perhaps Wittgenstein did believe in 1932–35 that there is no likelihood, in any sense, that a person would be burned in a fire. But if so, it is obvious that he changed his mind.

⁶ It is interesting to note that in the early book, Philosophical Grammar, in the discussion of probability, the point is made that the probability of a hypothesis can only be determined on the basis of experience, that the calculus of probability is not useful for this purpose. What is interesting in that discussion is that nothing Wittgenstein says indicates that he believes that in tossing a die the number thrown is independent of the factors that (in fact) determine which number is thrown:

It is perfectly compatible with our assumptions [that the die is accurate and regular] for one hundred ones to be thrown in succession, if friction, hand-movements and air-resistance coincide approximately. The experimental fact that this never happens is a fact about those factors, and the hypothesis that the throws will be uniformly distributed is an hypothesis about the operation of those factors.

One might think that if he did hold that nature is indeterministic and that therefore irregularities would occur, he would find such irregularities possible in the results of throwing a die. I don't see that he does. But see Anscombe's discussion of the ‘Gallon board’, in ‘Causality and Determinism’, op. cit., 141. Anscombe's discussion is helpful as an explanation of many of Wittgenstein's remarks.

⁷ One may also suppose that he came to think that ‘steel cog wheels cannot pass through one another’ is not an empirical proposition, but one that ‘stands fast’ for us.

⁸ See Malcolm's comment on my earlier suggestion; ‘Reply to Scheer’, op. cit., 167.Google Scholar

⁹ In addition to the passages discussed below, there are passages such as OC, 162Google Scholar, and other passages in which Wittgenstein says that he believes in the main facts of geography, history and science. See OC, 138Google Scholar and OC, 288, for example.Google Scholar