^{page 47 note 1} In a conversation with Karl Britton. See ‘Portrait of a Philosopher’, in Ludwig Wittgenstein, The Man and his PhilosophyGoogle Scholar, ed. Fann, K. T., p. 61.Google Scholar

^{page 47 note 2} Hacker, P. M. S., Insight and Illusion, p. 218.Google Scholar

^{page 48 note 1} Mind (1905).Google Scholar See also his book, The Philosophy of David Hume. I have made much use of these two works. I have also found very useful Popkin, Richard H.'s essay ‘David Hume: His Pyrrhonism and His Critique of Pyrrhonism’, Philosophical Quarterly (1951)Google Scholar reprinted in Hume, ed. V. C. Chappell.

^{page 49 note 1} Selby-Bigge edition of Hume, 's Treatise, p. 183.Google Scholar Italics added.

^{page 49 note 2} Treatise, p. 184.Google Scholar

^{page 49 note 3} Treatise, p. 185.Google Scholar

^{page 49 note 4} The Blue and Brown Books, p. 48.Google Scholar

^{page 49 note 5} Ibid., p. 46.

^{page 49 note 6} Ibid., p. 48.

^{page 50 note 1} ‘An innumerable variety of cases can be thought of in which we should say that someone has pains in another person's body’, p. 50. On pp. 52–3 he considers an example of feeling pain in another person's tooth, and points out that this is not the kind of thing that the solipsist is denying.

^{page 50 note 2} In Hume's moral philosophy the argument runs: If moral judgements were founded on reason, then they would be vulnerable to the moral sceptic; for “‘tis not contrary to reason to prefer the destruction of the whole world to the scratching of my finger” (Treatise, 416)Google Scholar; therefore they are not founded on reason.

^{page 50 note 3} Selby-Bigge edition of Hume, 's Enquiries, p. 155.Google Scholar

^{page 51 note 1} Treatise, p. 183Google Scholar; from the section ‘Of scepticism with regard to reason’. Although Hume is better known for his concern about the foundations of inductive inference, his ‘Pyrrhonian scepticism’ is first directed against the ‘demonstrative sciences’. Wittgenstein also is concerned with the foundations of both types of reasoning, and in both cases appeals to facts of nature as ultimate. (See, for example, Remarks on the Foundation of Mathematics, p. 63.)Google Scholar I have not thought it useful to keep the two types sharply distinct for the purpose of my discussion.

^{page 51 note 2} Enquiries, pp. 46–7.Google Scholar

^{page 52 note 1} The German Feststellungen does not have the empirical connotations of ‘observations’.

^{page 52 note 2} Philosophical Investigations, 415. Cf. 109.Google Scholar

^{page 52 note 3} Blue and Brown Books, 125.Google Scholar

^{page 52 note 4} On Certainty, 359.Google Scholar

^{page 52 note 5} Ibid., 172; my translation.

^{page 53 note 1} PI, 466–7.Google Scholar

^{page 53 note 2} Treatise, p. 183.Google Scholar

^{page 53 note 3} PI, 471.Google Scholar

^{page 53 note 4} PI, 468.Google Scholar

^{page 53 note 5} Previously quoted on pp. 50–1.

^{page 54 note 1} On Certainty, 287.Google Scholar

^{page 54 note 2} Enquiries, p. 39.Google Scholar

^{page 54 note 3} Philosophical Investigations, 472 ff.Google Scholar

^{page 54 note 4} See, for example, Treatise, p. 624Google Scholar and Enquiries, p. 48.Google Scholar

^{page 55 note 1} PI, 480.Google Scholar

^{page 55 note 2} PI, 325–6.Google Scholar

^{page 55 note 3} PI, 473.Google Scholar

^{page 55 note 4} PI, 474.Google Scholar

^{page 55 note 5} Treatise, p. 88.Google Scholar

^{page 56 note 1} On Certainty, 253.Google Scholar

^{page 56 note 2} PI, 326Google Scholar; Zettel, 301.Google Scholar

^{page 56 note 3} On Certainty, 192.Google Scholar

^{page 56 note 4} Treatise, p. 83.Google Scholar

^{page 56 note 5} Enquiries, p. 293.Google Scholar

^{page 56 note 6} PI, p. 226.Google Scholar

^{page 56 note 7} On Certainty, 559.Google Scholar

^{page 57 note 1} PI, p. 174.Google Scholar

^{page 57 note 2} In this paper I cannot do more than sketch the difference. I have not tried to do justice either to the empiricist notion of ‘simple ideas’ or to Wittgenstein's notions of a language game and a form of life. For a recent comparison of Hume's and Wittgenstein's ideas about ultimate certainties, see Anscombe, G. E. M., ‘Hume and Julius Caesar’, Analysis (1973).Google Scholar

^{page 58 note 1} See his two definitions of “cause”, Treatise p. 170.Google Scholar

^{page 58 note 2} Zettel, 357.Google Scholar

^{page 58 note 3} In 358 he continues in a similar vein: ‘Then there is something arbitrary about this system? Yes and no. It is akin both to what is arbitrary and to what is nonarbitrary.’

^{page 58 note 4} He reaches a similar conclusion in connection with arithmetic, and in connection with moral and aesthetic concepts. There is a comparison of causality with arithmetic in the Treatise, p. 166Google Scholar: ‘Thus as the necessity, which makes two times two equal to four, or three angles of a triangle equal to two right ones, lies only in the act of the understanding, by which we consider and compare these ideas; in like manner the necessity or power, which unites causes and effects, lies in the determination of the mind to pass from the one to the other.’

^{page 59 note 1} Treatise, p. 155.Google Scholar

^{page 59 note 2} Ibid., p. 165.

^{page 59 note 3} ‘Determination’ is here used in a passive sense, i.e. the mind being determined by the conjoined objects.

^{page 59 note 4} Braithwaite, R. B., in his Scientific ExplanationGoogle Scholar, writes: ‘The assertion of a subjunctive conditional may be regarded as a summary statement of this whole situation’ — a situation, described by Braithwaite, in which the person making the assertion has come to hold certain beliefs in a certain way (p. 297). Braithwaite, who sees his account as a development of Hume's ‘constant conjunction’ view, explains our causal notions in terms of the circumstances in which we use causal language.

^{page 59 note 5} PI, 66.Google Scholar

^{page 60 note 1} Treatise, p. 159.Google Scholar

^{page 60 note 2} Ibid., p. 168.

^{page 60 note 3} Enquiries, pp. 64–73.Google Scholar

^{page 60 note 4} Treatise, pp. 219–24, 259.Google Scholar

^{page 61 note 1} Zettel, 610.Google Scholar

^{page 61 note 2} Ibid. 580;my translation.

^{page 61 note 3} Ibid., 611, 613.

^{page 61 note 4} I am awaie that one of the conditions that Hume laid down for the causal relation was that of contiguity; and this might seem to favour the ‘cogwheel’ view. However, I believe that Hume's introduction of contiguity stems from other aspects of his philosophy, and is not an essential part of the ‘constant conjunction’ view of causality for which he is known. John Passmore (Hume's Intentions, p. 30)Google Scholar remarks that ‘spatial contiguity fades away’ from Hume's account of causality. See Passmore's book for a discussion of Hume's problems concerning spatial relations.

^{page 62 note 1} Some degree of simplification is unavoidable in a broad discussion of this kind. Thus (a) I tend to run together the notions of explanation and justification, both of them being opposed to ‘description’, and (b) I do not attempt to do justice to the variety of contexts in which the opposition is made.

^{page 62 note 2} Treatise, p. 13.Google Scholar

^{page 62 note 3} Enquiries, pp. 46–7Google Scholar; previously quoted on p. 51. The passage is used in a criticism of Hume by Flew, A., Hume's Philosophy of Belief, pp. 96–9.Google Scholar

^{page 63 note 1} Treatise, 218.Google Scholar

^{page 2 note 2} On Certainty, 25 3; my translation.Google Scholar

^{page 63 note 3} On Certainty, 166.Google Scholar

^{page 64 note 1} Zettel, 314.Google Scholar

^{page 64 note 2} PI, p. 200.Google Scholar

^{page 64 note 3} On Certainty, 359Google Scholar; previously quoted on p. 52.

^{page 64 note 4} Treatise, p. 187.Google Scholar

^{page 64 note 5} ‘If I have exhausted the justifications I have reached bedrock, and my spade is turned. Then I am inclined to say: “This is simply what I do.”’ PI, p. 217.Google Scholar