¹ Aquinas, , for instance, states that ‘God knows all enunciations that can be formed’.Google Scholar Cf. Summa Theologiae (henceforth ST) Ia 14, 14–though we will see below that his arguments actually support a somewhat different conclusion.

² Cf. e.g. Aquinas, , ST Ia 10.Google Scholar

³ Prior, Arthur, ‘The Formalities of Omniscience’, Philosophy, XXXVII (1962), 114–29;CrossRefGoogle ScholarCoburn, Robert, ‘Professor Malcolm on God’, Australasian Journal of Philosophy, XLI (1963), 155–6;Google ScholarKretzmann, Norman, ‘Omniscience and Immutability’, Journal of Philosophy, LXIII (1966), 409–21;CrossRefGoogle ScholarWolterstorff, Nicholas, ‘God Everlasting’, in Cahn, Steven, ed., Contemporary Philosophy of Religion (London: Oxford University Press, 1983), pp. 77–98;Google ScholarGrim, Patrick, ‘Against Omniscience: the Case from Essential Indexicals’, Nous, XIX (1985), 151–80;CrossRefGoogle ScholarGale, Richard, ‘Omniscience-Immutability Arguments’, American Philosophical Quarterly, XXII (1986), 319–35.Google Scholar This argument and its metaphysical congener have a long history; for their loci in medieval Arabic philosophy and in Aquinas, cf. Sorabji, Richard, Time, Creation and the Continuum (Ithaca, NY: Cornell University Press, 1983), pp. 260–1, nn. 28–31.Google Scholar

⁴ Richard Gale emphasizes this; cf. op. cit., p. 322.Google Scholar

⁵ Clarke, Bowman, Language and Natural Theology (The Hague: Nijhoff, 1966), p. 116;Google ScholarSutherland, Stewart, ‘Truth and God’, Proceedings of the Aristotelian Society, LVI (1982), 109;Google ScholarLewis, Delmas, ‘Eternity Again’, International Journal for Philosophy of Religion, XV (1984), 78, 79;Google ScholarCraig, William L., ‘Was Aquinas a B-Theorist of Time?’, The New Scholasticism, LIX (1985), pp. 475–83;CrossRefGoogle ScholarCreel, Richard, Divine Impassibility (NY: Cambridge University Press, 1986), pp. 96–7, III;Google ScholarHasker, William, ‘Yes, God Has Beliefs’, Religious Studies, XXIV (1988), 385–394.CrossRefGoogle Scholar Cf. Lewis, Delmas, ‘Eternity, Time and Tenselessness’, Faith and Philosophy, V (1988), 82–3.Google Scholar

⁶ Kretzmann, , op. cit.Google Scholar

⁷ Cf e.g. Wolterstorff, , op. cit.Google Scholar

⁸ Cf. Perry, John, ‘The Problem of the Essential Indexical’, Nous, XIII (1979), 3–21.CrossRefGoogle Scholar

⁹ Aquinas, , ST la 14, 13 and ad 3.Google Scholar

¹⁰ This should not be glossed by saying that these occur ‘at the same time’. They do not occur in time at all. For defence of the legitimacy of talk of simultaneity here, cf. Stump, Eleonore and Kretzmann, Norman, ‘Eternity’, journal of Philosophy, LXXIX (1981), 429–58.CrossRefGoogle Scholar

¹¹ ST la 14, 9.Google Scholar

¹² Lewis, Delmas, ‘Eternity Again’, International Journal for Philosophy of Religion, XV (1984), 78, 79.Google Scholar

¹³ Aquinas, Thomas, QD de Potentia Dei 3, 15 ad 4.Google Scholar

¹⁴ Cf. ST la 3, 1 and 4, 1.

¹⁵ Given that nothing can be both active and passive in the same respect at once. Thomas in effect insists on this at ST Ia 2, 2, when he asserts that nothing can be in act and in potency in the same respect at the same time. For in Thomas' eyes, something is ‘in potency’ with respect to F if and only if it can be acted upon by or passive in relation to an agent which can actualize its ‘potency’ to being F.

¹⁶ Moreover, if Lewis et al. assume (uncharitably) that Thomas is inconsistent, it is still incumbent on them to explain why we ought to take the perceptual model of God's knowledge rather than the claim that God is purely actual as the commitment Thomas would have chosen to keep had he become aware of his inconsistency. Arguably Thomas finds the claim that God is pure act far more important than the perceptual model.

¹⁷ Cf. e.g. Augustine, , Confessions, Bk. 11, cc. 14–15;Google ScholarBoethius, , De Trinitate, c. 4;Google ScholarAnselm, , De Concordia I, 2Google Scholar, Monologion cc. 22, 28;Google ScholarAquinas, , ST Ia 10, 4 ad 2Google Scholar, In IV Physica 1. 17,572–576; Scotus, Duns, Ordinatio I, d. 39, q. un., a. 5.Google Scholar

¹⁸ ST Ia 14, 14.

¹⁹ Cf. Alston, William, ‘Does God Have Beliefs?’, Religious Studies, XXII (1986), 287–306.CrossRefGoogle Scholar

²⁰ Ibid. p. 305.

²¹ Cf. Kretzmann, Norman, op. cit. p. 376.Google Scholar This claim is contested in Kvanvig, Jonathan, op. cit. pp. 47–71.Google Scholar Kvanvig also cites a number of recent defences of private truths.

²² My argument here depends on Perry, , op. cit.Google Scholar

²³ Were there no other hard cases for propositional omniscience lurking about, it would seem that had God not created other persons, He would be propositionally omniscient. But it is also true that if God has created time, then if essentially tensed facts are a problem for His omniscience, they would not be so had God not created time. (I owe this point to William Alston.) Were this true, however, it would not entail that creating changes God. One could say rather that God always timelessly knows just what He in fact timelessly knows, but whether this knowledge counts as propositional omniscience depends on whether there are truths in addition to the truths God always timelessly knows.

²⁴ Cf. Grim, Patrick, ‘Logic and Limits of Knowledge and Truth’, Nous, XXII (1988), 341–67CrossRefGoogle Scholar, and ‘Truth, Omniscience and the Knower’, Philosophical Studies, LIV (1988), 9–41.Google Scholar

²⁵ Since Herman exists and it is not possible that God = Herman.

²⁶ Thus a timeless, non-omniscient God could be in living dialogue with believers (cf. Alston, Williams, ‘Divine-Human Dialogue and the Nature of God’, Faith and Philosophy, II (1985), 5–20).CrossRefGoogle ScholarWolterstorff, , op. cit.Google Scholar, argues that divine timelessness is not compatible with providential activity. While I do not explicitly address Wolterstorff, this suggestion in fact undercuts his main argument.

²⁷ ‘Because’ does not imply that God acts after hearing me, but only that had it not timelessly been the case that God hears me, it would not have timelessly been the case that He saves me.

²⁸ That God eternally responds to a prayer which takes place only on 7 July, 1990 may sound paradoxical. For argument that it is not, cf. my Time and Eternity (Ithaca, N.Y.: Cornell University Press, forthcoming).Google Scholar

²⁹ That God cannot do these things does not entail the falsity of the Christian doctrine of the Incarnation. If God is able to become incarnate, He is able to be able to walk, breathe etc., because He is able to peform an action which will acquire for Him these abilities. But the logic of ability is not such that ‘able to be able’ entails ‘able simpliciter’. Socks are able to fit snugly around my feet merely by expanding in a shape they already possess. A ball of wool is able to be knitted into a pair of socks. Hence it is able to be able to fit snugly around my feet merely by expanding in a shape it already possesses. But it is not at present able to fit snugly around my feet etc. A ball of wool does not fit my feet at all. Again, a child with great natural talent may be able to be able to play all Beethoven piano sonatas. But if he has not yet had piano lessons, he is not yet able to do so.

Now some might think that a believer in both divine timelessness and the Incarnation (if these beliefs are co-tenable) cannot say that at any time God is able to be able to walk but not able to walk. For the transition from ability to be able to actually being able constitutes a real change, and if God is timeless, He is unable to undergo real change: so if God is ever able to walk, He is always so. There is a truth in this. If God is timeless and at some time becomes incarnate, then since the incarnation cannot constitute a change in Him, He is from all eternity incarnate, notwithstanding that in time He becomes incarnate only at a certain date. (There is no paradox here; cf. my Time and Eternity.) But it remains true that God qua nonincarnate remains eternally able to be able but not able. This being so, my earlier point about God's disabilities stands, suitably qualified, even for a believer in the Incarnation. Moreover, even if the Incarnation be held to eliminate inabilities to walk, breathe etc. tout court, there wi11 still remain God's inability to do moral evil, at least according to such traditional theists as Anselm (cf. Proslogion, c. 7) and Aquinas (cf. ST Ia 25, 3 ad 2).

³⁰ This does not entail that there is some distinct first-order power to do evil which we have and God lacks, as there is a distinct power to walk, breathe etc. (cf. Morris, Thomas, ‘Perfection and Power’, International Journal for Philosophy of Religion, XX (1986), 167–8).Google Scholar An act of doing so-and-so is also an act of evil-doing, in appropriate situations; its moral character supervenes on its other traits. If God is necessarily morally perfect, then in any circumstance C, He cannot perform any action A such that A would be an evil-doing in C. What we have and a necessarily morally perfect God lacks, then, is a second-order power, the power to employ all one's first-order powers in all situations without regard for their moral character.

³¹ Thus limits on omniscience and limits on omnipotence seem fully parallel, counter to Kvanvig, Jonathan, The Possibility of an All-Knowing God (NY: St. Martin's Press, 1986), pp. 23–4.CrossRefGoogle Scholar

³² Cf. Nagel, Thomas, ‘What is it Like to be a Bat?’, The Philosophical Review, LXXXIII (1974), 435–50.CrossRefGoogle Scholar

³³ If God logically cannot fail, He cannot even try to grasp this proposition, for if He tried, He would fail.

³⁴ The entailment between limits on God's power and limits on His knowledge runs in both directions. Suppose that we know and God does not know the proposition P. If we know P, we have an ability to express a proposition one knows by tokening ‘P’. If God does not know P, God lacks this ability.

³⁵ Again, this does not entail that there is some first-order ability to fail, which we have and God lacks. Rather, what we have and God lacks is a second-order ability, the ability to employ one's first-order abilities defectively.

³⁶ ‘Process’ philosophers can accept this, by interpreting ‘being’ in a way reflecting a ‘process’ metaphysics.

³⁷ Cf. Anselm, , Monologion, c. 15; Proslogion, c. 5.Google Scholar

³⁸ Note that this procedure does not guarantee that just one such set of attributes will emerge. It may be that we are not that lucky. It would guarantee this if we legislated that if sets of attributes S and S* each lend their possessors the same degree of perfection, and no other set of attributes lends its possessors a greater perfection, the perfect being has neither S nor S* – for then in this case, the sole set of attributes the procedure would warrant would be the null set.

³⁹ Gale, Richard, op. cit. p. 332.Google Scholar

⁴⁰ If one called cognitively perfect any being which has all the knowledge it can have, regardless of the range of that knowledge, then if there were a being which could know only one proposition, this being would count as cognitively perfect in virtue of knowing just one truth – which is absurd.

⁴¹ I owe the device of using the relativity of simultaneity to express the doctrine of God's timelessness to Stump, and Kretzmann, , op. cit.Google Scholar

⁴² Given the definition which follows, one must pick out a first inertial frame of reference: to be told that an inertial reference-frame is a system of physical objects which move at uniform velocity relative to other inertial frames of reference tells one nothing unless one has at least one such reference-frame in hand, so that relative to it one can discriminate other inertial frames. The mention of our own rest-frame is just an arbitrary choice for this role; nothing hangs on it.

⁴³ Cf. Stump, and Kretzmann, , op. cit.Google Scholar, and my response ‘Eternity and Simultaneity’, Faith and Philosophy 8, I (01 1991).Google Scholar

⁴⁴ Op. cit. pp. 434–44.Google Scholar I have criticized the Stump-Kretzmann account of ET-simultaneity in ‘The Roots of Eternity’ (Religious Studies XXIV (1988), 189–212)Google Scholar and ‘Eternity and Simultaneity’ (Faith and Philosophy 8, I (01 1991)).Google Scholar But in neither place do I claim to show that no concept of ET-simultaneity is possible.

⁴⁵ According to special relativity, that some x occurs simultaneously with some y in some frame of reference and that y occurs simultaneously with some z in some frame of reference does not entail that x and z are simultaneous in any frame of reference. If it did, simultaneity would not be framework-relative.

⁴⁶ Op. cit. pp. 434–5.Google Scholar

⁴⁷ That generic simultaneity is not transitive blocks the argument that since God's seeing of event E is ET-simultaneous with E, God's seeing of later event E* is ET-simultaneous with E*, and God's seeings are eternally simultaneous, E and E* are in some manner simultaneous.

⁴⁸ At least later in life. For supporting exegesis, cf my ‘Aquinas on Time and Eternity’, The New Scholasticism, forthcoming. For a fuller development of the ideas about simultaneity, time and eternity sketched here, cf. my ‘Eternity and Simultaneity’.

⁴⁹ ‘Eternity and Simultaneity’.