^{page 19 note 1} Augustine, St, Enchiridion, XI.Google Scholar

^{page 19 note 2} Aquinas, St Thomas, Summa Theologiae, Ia, ii, 3, ad 2um.Google Scholar

^{page 19 note 3} Geach, P. T., Providence and Evil (Cambridge, 1977), p. 66.Google Scholar

^{page 21 note 1} Hume, D., Dialogues Concerning Natural Religion, part xGoogle Scholar (quoted in Ahern, M. B., The Problem of Evil (London, 1971), p. 3).Google Scholar

^{page 21 note 2} Quoted in Ahern, , op. cit. p. 3.Google Scholar

^{page 22 note 1} Ahern, , op. cit. pp. 7–8.Google Scholar

^{page 23 note 1} Ahern, , op. cit. p. 28.Google Scholar

^{page 23 note 2} Ahern, , op. cit. p. 29.Google Scholar

^{page 23 note 3} This definition of omnipotence raises several issues which cannot be adequately treated in this paper. A couple of remarks should, however, be made on aspects of the definition which may be controversial.

(I) The strictures of Professor Geach (in Providence and Evil and elsewhere) against the notion of omnipotence seem to be based on the fact that God cannot be said to do certain things (particularly evil acts such as breaking promises) which are nevertheless intrinsically possible. But if we analyse ‘God can do X’ as ‘If God chooses to do X, He will do X’, then we can say that God is, after all, able to do these things (i.e., he has the power to do them if he chooses) but that because he is infinitely good he will never, in fact, choose to do them. In any case, even if the notion of omnipotence does turn out to be incoherent, the rival notion of almightiness, which Geach accepts, is close enough to that of omnipotence to be equally serviceable in the formulation of the problem of evil (cf. Providence and Evil, pp. 4–5). (2) The above definition of omnipotence is formulated in terms of absolute (rather than logical) necessity because I reject the view that logical necessity is the only kind of necessity. On the contrary, in my view, the notion of absolute necessity encompasses necessity of both the analytic and the synthetic a priori kinds. Obviously, if there are synthetic a priori necessities, such as that no object can simultaneously have two different colours over its entire surface, then God can no more change these truths to falsehoods than he can suspend the law of non-contradiction or bring it about that some bachelors are married. Acceptance of non-logical necessity is doubtless a controversial move, but attempts in recent times to do away with the synthetic a priori have, I believe, consistently failed.

^{page 24 note 1} Ahern, , op. cit. p. 32.Google Scholar

^{page 25 note 1} Job 42: 1–4 (RSV edition).

^{page 26 note 1} S. Theol., la, iii, prologue; cf also 1a, i, 7, ad 1um; 1a, xii, 13, ad 1um; In Boethii de Trinitate, i, ii, ad 1um.

^{page 26 note 2} S. Theol., la, xii, 12.

^{page 26 note 3} Cf. Lossky, V., The Mystical Theology of the Eastern Church (London, 1957), chapter 4.Google Scholar

^{page 28 note 1} Ahern, , op. cit. p. 50.Google Scholar

^{page 31 note 1} Ahern, , op. cit. p. 57.Google Scholar

^{page 31 note 2} Flew, A. G. N., God and Philosophy (London, 1966), pp. 54–5.Google Scholar

^{page 32 note 1} Rowe, L. L., ‘The Problem of Evil and some Varieties of Atheism’, American Philosophical Quarterly, xvi (1979), 337.Google Scholar

^{page 32 note 2} Ibid.

^{page 32 note 3} Geach, , op. cit. pp. 79–80.Google Scholar

^{page 33 note 1} Lewis, C. S., The Problem of Pain (London, 1966), pp. 124–30.Google Scholar

^{page 33 note 2} Darcy, M. C., The Problem of Evil (London, 1971), pp. 13–14.Google Scholar

^{page 35 note 1} Mind, LXXXVII (1978), 429–32.CrossRefGoogle Scholar

^{page 36 note 1} Martin, , art. cit. p. 430.Google Scholar

^{page 37 note 1} Hare, P. H. and Madden, E. H., ‘Evil and Inconclusiveness’, Sophia, IX (1972), 11.Google Scholar