² Originally set out in Hempel, C. G. and Oppenheim, Paul, ‘Studies in the Logic of Explanation’, Philosophy of Science 15 (1948), 135–75.CrossRefGoogle Scholar

³ See Hempel, C. G., ‘Deductive-nomological vs Statistical Explanation’ in Feigl, H. and Maxwell, G. (eds.) Minnesota Studies in the Philosophy of Science, Vol. III (University of Minnesota Press, 1962)Google Scholar; and Hempel's final account of explanation in ‘Aspects of Scientific Explanation’, in his Aspects of Scientific Explanation (New York: Free Press, 1965)Google Scholar. For an elementary statement of Hempel's account of explanation, see his Philosophy of Natural Science (Englewood Cliffs, N.J.: Prentice Hall, 1966), ch. 5.Google Scholar

⁴ Salmon, W. C., Statistical Explanation and Statistical Relevance, (University of Pittsburgh Press, 1971).CrossRefGoogle Scholar

⁵ See, e.g. Hawking, Stephen W., A Brief History of Time (London: Bantam Press, 1988), ch. 8.Google Scholar

⁶ Confessions 11.30.

⁷ Goodman, N., Fact, Fiction and Forecast, 2nd edn (Indianapolis: Bobbs-Merrill, 1965) chs. 3 and 4.Google Scholar

⁸ See especially my The Evolution of the Soul, ch. 10.

⁹ I write, for the sake of simplicity of exposition, in my definition and subsequent use of the notion of ‘complete explanation’, of factors acting ‘at the time of’ or ‘simultaneously with’ or ‘contemporaneously with’ their effect. It may be that there are logical problems with the idea of some cause C bringing about an effect E literally simultaneously with it. In that case talk about C bringing about E ‘simultaneously’ is to be read as C bringing about E in such a way that there is no temporal interval between the end of C and the beginning of E.

¹⁰ Except in so far as there is a certain amount of indeterminancy within the processes of the natural world (e.g. the fundamental laws of quantum theory being merely probabilistic). There is in that case no full, complete or ultimate explanation of how they operate on a particular occasion, but my claim is that there is such explanation of why there is such indeterminancy in the natural world.

¹¹ For the main idea of this account, I am of course indebted to Harré, R. and Madden, E. H., Causal Powers (Oxford: Blackwell, 1975).Google Scholar

¹² Any explanation of the present existence of the universe by the action of God is not merely complete, but ultimate, because in bringing about the laws of nature (i.e. the powers and liabilities of objects) what God brings about is not just laws that, as a matter of fact are such that, given a past state of the universe, there is a present state (i.e. matter is conserved); but that the laws of nature are such as to yield the present state (whether by making something out of a previous nothing, or by conserving in existence previous matter). Thus his action by itself suffices to explain the present existence of the universe, independently of any earlier factors (God, his purpose and capacities) which have no explanation in terms of anything else.

¹³ The hypothesis that light has infinite velocity was held by the most influential writers before Romer's discovery of its finite velocity, such as Aristotle and Descartes who held that the transmission of light was the transmission of the state of a medium, rather than of a particle. Those who adopted a particle theory were moved by considerations of ‘fit’ with the behaviour of other particles to postulate a finite velocity, i.e. considerations of the simplicity of an overall theory able to explain other data as well as those concerned with light.