God, Time and Eternity

William Lane Craig

Is God's eternity to be construed as timeless or temporal? Given that the universe began to exist, a relational view of time suggests that time also began to exist. God's existence "prior to" or sans creation would not entail the existence of time if God in such a state is changeless. But if God sustains real relations with the world, the co-existence of God and the world imply that God is temporal subsequent to the moment of creation. Given the superiority of a relational over a non-relational (Newtonian) view of time, God ought to be considered as timeless sans creation and temporal subsequent to creation.

Source: "God, Time, and Eternity." Religious Studies 14 (1979): 497-503.

God is the 'high and lofty One who inhabits eternity',{1} declared the prophet Isaiah, but exactly how we are to understand the notion of eternity is not clear. Traditionally, the Christian church has taken it to mean 'timeless'. But in his classic work on this subject, Oscar Cullmann has contended that the New Testament 'does not make a philosophical, qualitative distinction between time and eternity. It knows linear time only…'{2} He maintains, 'Primitive Christianity knows nothing of a timeless God. The "eternal" God is he who was in the beginning, is now, and will be in all the future, "who is, who was, and who will be" (Rev. 1:4).'{3 } As a result, God's eternity, says Cullmann, must be expressed in terms of endless time.

When we speak of God as eternal, then, we may mean either 'timeless' or simply 'everlasting'. The question is: which understanding of God's relationship to time is to be preferred? Taking sharp issue with Cullmann's study, James Barr has shown that the biblical data are not determinative. He argues that Cullmann's study is based too heavily upon etymology and vocabulary studies, and these cannot be determinative in deciding the meaning of a term apart from use.{4} Barr thinks that Genesis may very well teach that time was created along with the universe, and that God may be thought of as timeless.{5} Barr's basic contention is that, 'A valid biblical theology can be built only upon the statements of the Bible, and not on the words of the Bible.'{6} When this is done, the biblical data are inconclusive: '. ..if such a thing as a Christian doctrine of time has to be developed, the work of discussing it and developing it must belong not to biblical but to philosophical theology'.{7}

Therefore, the issue lies in the lap of the philosopher, not the theologian. Are there, then, good philosophical arguments for preferring one of these competing notions of God's eternity to the other? I think that there are.

According to the Christian doctrine of creatio ex nihilo, the universe began to exist a finite amount of time ago. This doctrine receives philosophical confirmation from arguments demonstrating the absurdity of an infinite temporal regress of events{8} and empirical confirmation from the evidence for the so-called 'big-bang' model of the universe.{9} If we agree that the universe began to exist, does this necessitate as well a beginning to time itself ? The answer is: it all depends. If a person believes that time exists apart from events such that if there were no events there would still be time, then our argument does not entail prima facie a beginning to time. On the other hand, if one accepts that time cannot exist apart from events, then a beginning of events would entail a beginning of time as well.

There are a few modern authors who hold to the independent status of time apart from events, and they are thus the heirs of the Newtonian conception of absolute time. Swinburne argues that time, like space, is of logical necessity unbounded.{10} For every instant of time must be preceded and succeeded by another instant of time. The physical universe itself may have had a beginning – but this can only be true if there is a period of time before the beginning during which the universe did not exist. Since time is unbounded, it is of logical necessity infinite. Since prior to and after every period of time there is more time and since the same instant of time never recurs, time must have gone on and will go on forever. Although space would not exist without physical objects, time would. But, he adds, without physical objects, time could not be measured: one could not distinguish an hour from a day in a period of time without objects.{11} Therefore, Newton's claims about Absolute Time were correct.{12} To say that the universe began to exist on such a time scale would simply be to say that a finite time ago there were no physical objects.{13}

J. R. Lucas also contends that time could neither begin nor end. {14} He notes that if time is defined in a relational manner, then if there was an absolutely stationary universe prior to the first event, we would have to say that time did not exist until the first event occurred. At the beginning of time no past tense statements could be made, since there was no past. Yet it is obvious that certain statements, such as, 'The stars were moving', is a meaningful, though in this case false, statement that could be made about the state of the universe prior to the first event. Lucas does not deny that the universe may have had a beginning, but he, like Swinburne, argues that in such a case time would precede the beginning of the universe and that it would be undifferentiated.{15} Without a world there would be no metric to impose upon time.

A variant on the above view is expressed by Lawrence Sklar, whose theories of time are heavily influenced by relativity theory. He interprets Minkowski spacetime in a literalistic way, asserting that future events 'have determinate reality' and future objects are 'real existents'.{16} Accordingly, he regards time and space as inextricably bound up together in spacetime. {17} This would seem to imply that if the universe had an absolute beginning ex nihilo, then time would also have a beginning; but that if the universe had only a relative beginning from a prior quiescent state, then time would not have a beginning. Ian Hinckfuss also argues that if the universe were frozen into immobility, there would still be time because temporal duration and measurement are not dependent upon the continuous operation of a clock throughout that time.{18}

Presumably to such thinkers the beginning of the temporal series of events would not entail a beginning to time itself. On the other hand, those who adhere to a relational view of time generally take the beginning of events to be synonymous with the beginning of time itself. Zwart, for example, asserts,

According to the relational theory the passage of time consists in the happening of events. So the question whether time is finite or infinite may be reduced to the question whether the series of events is finite or infinite.{19}
It might be asserted that even on the relational view of time there can be time prior to the first event because one may abstract from individual events to consider the whole universe as a sort of event which occurs at its creation. There would thus be a before and an after with regard to this event: no universe/ universe. And a relation of before and after is the primitive relation of which time consists.{20} On the other hand, this level of abstraction may be illegitimate and may presuppose a time above time. For prior to the universe's beginning, if there was nothing at all, not even space, then it would certainly seem to be true that there was no time either. For suppose the universe never came to exist - would there still be time? But if the universe does come to exist ex nihilo, how could we say this first event has an effect on reality (but of course there was no reality!) before it ever occurred, especially when its occurrence is a contingent matter? We might want to say that time does not exist until an event occurs, but when the event does occur, there is a sort of retroactive effect causing past time to spring into being. But this seems to confuse our mental ability to think back in time with the progressive, unidirectional nature of time itself. Though we can, after creation, think of nothingness one hour before the first event, in terms of reality, there was no such moment. For there was just nothing, and Creation was only a future contingent. When the first event occurred, the first moment of time began.

These are difficult conundrums, and it is at least an open question as to whether a beginning of events necessitates a beginning of time. Therefore, we need to ask whether there is any absurdity in supposing that time had a beginning. Some philosophers have argued that time cannot have a beginning because every instant of time implies a prior instant. Thus, there could be no first instant of time. Within a Newtonian understanding of time this argument, even if valid, would only imply that the universe had a beginning in time instead of with time. But, in fact, it does seem plausible to contend on a relational view of time that a first instant could exist, since apart from events no time exists. Stuart Hackett argues,

Time is merely a relation among objects that are apprehended in an order of succession or that objectively exist in such an order: time is a form of perceptual experience and of objective processes in the external (to the mind) world. Thus the fact that time is a relation among objects or experiences of a successive character voids the objection that the beginning of the world implies an antecedent void time: for time, as such a relation of succession among experiences or objective processes, has no existence whatever apart from these experiences or processes themselves.{21}
Therefore, if nothing existed and then something existed, there is no absurdity in speaking of this as the first moment of time. Brian Ellis notes that because we speak of 'before creation' or 'prior to the first event', we tend to think that a beginning of time is impossible.{22 } But Ellis draws a very instructive analogy between this sort of speech and talk of temperatures below absolute zero. When a physicist says there are no temperatures lower than absolute zero, the use of 'lower than' does not presuppose there actually are such temperatures, but only that we can conceive it in our minds. In the same way, to say there was a time when the universe did not exist does not imply there was such a time, but only that we can mentally conceive of such a time. To say there is no time before the first event is like saying there is no temperature -273 C. Both express limits beyond which only the mind can travel. Whitrow remarks in this connection that many people have difficulty imagining a beginning to time because they think of it as a boundary similar to a boundary of space.{23} We reject the latter because we could presumably cross the boundary and find space on the other side. But the case with time is different because we cannot travel freely in time as in space. If time coexists with events, then an origin of time merely implies a beginning of the universe. The first moment of time is not a self-contradictory concept.

There does not appear to be, therefore, any absurdity in the notion of a beginning of time. The idea of a 'time before time' is a mental construction only, a product of the imagination. In reality there seems to be no impossibility in having time arise concommitantly with the universe ex nihilo. Thus, on a Newtonian view of time, the universe arises in an absolute, undifferentiated time, while on a relational view of time, it comes into existence with time.

But, of course, prior to creation was not simply nothing, but God. Would his existence necessitate the presence of time prior to creation? Lucas argues that a personal God could not be timeless and that if God is eternal, then time must be infinite as well. {24} But Hackett argues convincingly that a personal God need not experience a temporal succession of mental states. He could apprehend the whole content of the temporal series in a single eternal intuition, just as I analogously apprehend all the parts of a circle in a single sensory intuition. God could know the content of all knowledge - past, present, and future - in a simultaneous and eternal intuition.{25} Therefore, the fact that the creator is personal does not necessitate the presence of time prior to creation. Sturch argues that in order to avoid an infinite temporal regress of states of consciousness, God's knowledge must be timeless.{26} On a Newtonian view of time, God would exist changelessly in an undifferentiated time prior to creation. On a relational view of time, God would exist changelessly and timelessly prior to the first event, creation, which marks the beginning of time.

But what about subsequent to the first event? If God sustains any relations to the world, does not this imply that he exists in time? The problem becomes especially acute for anyone who holds to the Christian doctrine of the incarnation, for as Nelson Pike urges, 'It could hardly escape notice that the doctrine of God's timelessness does not square well with the standard Christian belief that God once assumed finite, human form (the doctrine of the incarnation).' {27} Soren Kierkegaard called this the Absolute Paradox; this is the contradiction of existence: the presence of the Eternal in time, how God can enter the space-time world without ceasing to be the Eternal.{28}

Thomas Aquinas attempted to solve this problem by arguing that while creatures are really related to God, God sustains no real relation to creatures.{29} Hence, God exists timelessly, unrelated to creatures, while creatures in time change in their relations to him. In the incarnation, a human nature becomes related in a new way to the second person of the Trinity, but that person does not sustain a real relation with that human nature. But this doctrine is singularly unconvincing. It is system-dependent upon regarding relation as an accident inhering in a substance. Because God is absolutely simple, he has no accidents and, hence, no real relations. But if we reject the Aristotelian metaphysical doctrine of substance and accidents, then it seems foolish to say God is not really related to the world as Creator to creature.

If God is really related to the world, then it seems most reasonable to maintain that God is in time subsequent to creation. This also removes Kierkegaard's Absolute Paradox concerning the incarnation, for God would be in time prior to his assuming a human nature. This understanding does not involve any change in God; rather he is simply related to changing things. As Swinburne explains,

...since God coexists with the world and in the world there is change, surely there is a case for saying that God continues to exist for an endless time, rather than that he is timeless. In general that which remains the same while other things change is not said to be outside time, but to continue through time.{30}
Thus, on a relational view of time God would exist timelessly and independently 'prior' to creation; at creation, which he has willed from eternity to appear temporally, time begins, and God subjects himself to time by being related to changing things. On the other hand, the Newtonian would say God exists in absolute time changelessly and independently prior to creation and that creation simply marks the first event in time.{31}

These, then, are the alternatives. A relational view of time seems superior to a Newtonian view because (1) it is difficult to see how time could exist apart from events and (2) the Newtonian objection that every instant of time implies a prior instant is adequately answered by the relational view. Thus, the proper understanding of God, time, and eternity would be that God exists changelessly and timelessly prior to creation and in time after creation.


{1} Isaiah 57:15 (RSV).

{2} Oscar Cullmann, Christ and Time (London: SCM Press, 1962), p. xxvi.

{3} Ibid. p.63.

{4} James Barr, Biblical Words for Time (London: SCM Press, 1962), p.80.

{5} Ibid. pp. 145- 7.

{6} Ibid. p. 147.

{7} Ibid. p. 149.

{8} Historically, this argument has been defended by Al-Kindi, Al-Kindi's Metaphysics: A Translation of ˇa' qub ibn Ishaq al-Kindi's Treatise ' on first Philosophy', with an Introduction and Commentary by Alfred L. Ivry (Albany, N.Y.: State University of New York Press, 1974); Al-Ghazali, Tahafut al- Falasifah (Incoherence of the Philosophers), trans. Sabid Ahmad Kamali (Lahore: Pakistan Philosophical Congress, 1958); Saadia Gaon, The Book of Beliefs and Opinions, trans. Samuel Rosenblatt (New Haven, Conn.: Yale University Press, 1948); Bonaventure, 2 Sentences Modern defenders of the argument include Stuart C. Hackett, The Resurrection of Theism (Chicago: Moody Press, 1957); G. J. Whitrow, The Natural Philosophy of Time ( London and Edinburgh: Thomas Nelson and Sons, 1961); Pamela M. Huby, 'Kant or Cantor? That the Universe, if real, must be finite in Both Space and Time', Philosophy, XLVI (1971),121-32. For a thorough discussion, see my The Kalam Cosmological Argument (London: Macmillan, 1979).

{9} On the big-bang model see P. J. E. Peebles, Physical Cosmology (Princeton, N. J.: Princeton University Press, 1971); S. Weinberg, Gravitation and Cosmology (New York, Wiley, 1972). That this model requires creatio ex nihilo is explained by Fred Hoyle, Astronomy and Cosmology (San Francisco: W. H. Freeman & Co., 1975), p. 658. See also my book mentioned in the above note.

{10} R. G. Swinburne, Space and Time (London: Macmillan, 1968), pp.207-8.

{11} Ibid. p. 209.

{12} Ibid. p. 245.

{13} Ibid. p. 296.

{14} J. R. Lucas, A Treatise on Time and Space (London: Methuen & Co., 1973), pp. 10-11

{15} Ibid. pp. 311-12.

{16} Lawrence Sklar, Space, Time, and Spacetime (Berkley and Los Angeles: University of California Press, 1974), p. 274.

{17} Ibid. p. 297.

{18} Ian Hinckfuss, The Existence of Space and Time (Oxford: Clarendon Press, 1975), pp. 72-3.

{19} P. J. Zwart, About Time (Amsterdam and Oxford: North Holland Publishing Co., 1976), p. 237.

{20} Ibid. p. 36. '. . . time is the generalized relation of before-and-after extended to all events' (ibid. p. 43).

{21} Hackett, Theism, p. 263.

{22} Brian Ellis, 'Has the universe a beginning in time?', Australasian Journal of Philosophy XXXIII (1955), 33.

{23} G. J. Whitrow, What is Time? (London: Thames & Hudson, 1972), pp. 146-7.

{24} Lucas, Treatise, pp. 3, 309.

{25} Hackett, Theism, pp. 286-7. I think that it is within the context of Trinitarian theology that the personhood and timelessness of God may be the most satisfactorily understood. For in the eternal and changeless love relationship between the persons of the Trinity, we see how a truly personal God could exist timelessly, entirely sufficient within himself. Most writers who object to a timeless, personal God consider God only subsequent to creation as he is related to human persons, but fail to consider God prior to creation (e.g. Nelson Pike, God and Timelessness [London: Routledge & Kegan Paul, 1970], pp. 121-9). The former would appear to involve God in time, but the latter would not, for if God is tri-personal he has no need of temporal persons with whom to relate in order to enjoy personal relationships-- the three persons of the Godhead would experience perfect and eternal communion and love with no necessity to create other persons. Thus the answer to the question, 'What was God doing prior to creation?' is not the old gibe noted by Augustine: 'He was preparing hell for those who pry into mysteries' , but rather, 'He was enjoying the fullness of divine personal relationships, with an eternal determination for the temporal creation and salvation of human persons.' Why did God so determine? Perhaps to share the joy and love of divine fellowship with persons outside himself and so glorify himself: on the other hand, perhaps we lack sufficient information to answer this question. Once temporal persons were created, God would then begin to experience temporal personal relationships with them.

{26} R. L. Sturch, 'The Problem of Divine Eternity', Religious Studies X (1974), 492.

{27} Pike, God, p. 172.

{28} Soren Kierkegaard, Philosophical Fragments, trans. David Swenson (Princeton, N. J.: Princeton University Press, 1936), pp. xii, 72.

{29} Thomas Aquinas Summa Theologiae Ia. 13.7. See also John Donnelly, 'Creatio ex nihilo', in Logical Analysis and Contemporary Theism, ed. John Donnelly (New York: Fordham University Press, 1972), pp. 210-11; Peter Geach, (God's Relation to the World', Sophia VII 1969), 1-9.

{30} R. G. Swinburne, 'The Timelessness of God', Church Quarterly Review CLXVI (1965), 331.

{31} This serves effectively to rebut the objection of Julian Wolfe to the kalam cosmological argument 'Infinite Regress and the Cosmological Argument', International Journal for Philosophy of Religion II (1971), 246-9. The crucial premise is, in Wolfe's opinion, that an infinite time cannot elapse. He argues that this is incorrect because prior to causing the first effect, the uncaused cause existed for infinite time. Since the first event did not occur, then an infinite time must have elapsed. But in the first place, Wolfe's formulation of the argument is defective, for the contention is that an infinite number of events cannot elapse, not that an infinite time cannot elapse. Because the argument concerns events, not time, Wolfe's analysis is inapplicable, since prior to creation there were no events at all. Second, if the relationalist is correct, then an infinite time does not elapse prior to creation, because time begins at creation. God is simply timeless before the first event.