^{page 28 note 1} See Gregory, R. L., Eye and Brain (London: World University Library, 1966), 203–206.Google Scholar

^{page 29 note 2} See Polanyi, Michael, Personal Knowledge (London: Routledge, 1958), 55–56.Google Scholar

^{page 29 note 3} This phrase and this notion of ‘critical reflection’ come from Russell, Bertrand, Our Knowledge of the External World, 2nd edn (New York: Mentor Paperbacks, 1929), 60.Google Scholar References to this book will hereafter be given in the text and marked by the abbreviation ‘OKEW’ followed by the page number.

^{page 29 note 4} The reason I say ‘appear’ here and below is that it has been argued that no perceptual experience is without expectation. C. S. Peirce considering the following challenge presented by the suggestion that simple colour experiences provide a counter-example: ‘I lay down a wafer before me. I look at it, and say to myself that wafer looks red. What element of expectation is there in the belief that the wafer looks red at this moment?’ But judgment (or the act of asserting) takes time, and its reference is to the state of the percept at the time it begins to be made. By the time the judgment (or assertion) has been made, it is already about the past. ‘The judgment, then, can only mean so far as the character of the percept can ever be ascertained, it will be ascertained that the wafer looked red.’ The Collected Papers of C. S. Peirce, Hartshorne, Charles and Weiss, Paul (eds), 8 vols (Harvard University Press, 1931–1935), 5. paras 542–544.Google Scholar

^{page 29 note 5} The alternative kind of justification is what Kant would have called ‘a transcendental deduction’. See Immanuel Kant's Critique of Pure Reason, trans, by Smith, Norman Kemp (London: Macmillan, 1933, second impression), 120 Google Scholar, for this use of ‘deduction’ to label the demonstration of a right. The argument, which will be used here in response to Russell's problem, is clearly inspired by Kant, although it does not draw directly on the arguments which Kant himself sets out.

^{page 29 note 6} That Russell allows himself particularly these correlations between different sensory modes was regarded by at least one of his critics, John Dewey, as begging the question in favour of the inferences, which the question was supposed to challenge. And if we are going to assume this much of an external world, then the only question remaining is what are we justified in believing about the world. What Russell assumed, ‘may not be a very big external world, but having begged a small external world, I do not see why one should be too squeamish about extending it over the edges’. Dewey, John, Essays in Experimental Logic (New York: Dover, no date, original 1916), 292.CrossRefGoogle Scholar

^{page 29 note 7} See Putnam, Hilary, Reason, Truth and History (Cambridge University Press, 1981)CrossRefGoogle Scholar, ch. 1, and Nozick, Robert, Philosophical Explanations (Oxford: Clarendon Press, 1981), ch. 3.Google Scholar

^{page 29 note 8} There is a similar insufficiently explored difficulty with the temporal version of the problem of the external world, viz. is there any reality to our beliefs about history? Russell formulated a version of this problem by means of the hypothesis (sometimes known as ‘Russell's Five Minute Hypothesis’) that God created the world five minutes ago complete with all the apparent evidence of possessing a lengthy past, including our memories, historical documents and archaeological and geological phenomena. What evidence could we ever have that this was the case and why, if we had any evidence should we take it seriously? What would make it convincing evidence? On the five minute hypothesis, see Russell, Bertrand, The Analysis of Mind (London: George Allen and Unwin, 1921), Lecture IX.Google Scholar

^{page 29 note 9} Op. cit. above note 7, 14ff.