Divine Timelessness Defined and Defended
This paper concerns itself with objections to the doctrine of divine timelessness. Since, however, the doctrine needs careful articulation before such objections can be considered, I begin by briefly defining it. In so doing, I elucidate three distinct positions that its advocates may take. Finally, I consider objections to the doctrine which have appeared in the contemporary literature, concluding that none succeeds.
The Doctrine Defined
Some theists maintain that being eternal amounts to existing everlastingly in time. In their view, God enjoys a beginningless and endless temporal existence. Thus, they affirm that
(1) For every time t, God exists at t.
Such theists I refer to as temporalists; their view I refer to as temporalism. Not all theists are temporalists; some deny (1). They do so not because they think God exists at some but not all times; instead, they maintain that God’s existence is not temporal, that he is timeless.
Nelson Pike separates the view that God is timeless into two distinct claims. “First,” he states, “if God is timeless, He has no duration, i.e., He lacks temporal extension…Secondly, if God is timeless, God also lacks temporal location.” Now while it might be possible for one to be temporally located without being temporally extended, it is not possible for one to be temporally extended without being temporally located. To be temporally extended is to exist for the length of some temporal interval. But since a temporal interval amounts to nothing more than an unbroken series of temporal locations, to exist for the length of such an interval is to exist during a series of such locations. Thus, one cannot be temporally extended without being temporally located. For this reason, the claim that God is timeless can be reduced simply to the claim that he lacks temporal location. Or so it can on Pike’s account of it. On his account, then, advocates of divine timelessness are theists who affirm that
(2) For every time t, it is not the case that God exists at t.
Such theists I refer to as atemporalists; their view I refer to as atemporalism.
Richard Swinburne attributes to atemporalists the view that “[t]here is no temporal succession of states in God.” On its face, this attribution seems well-founded. Since having temporally successive states involves having some state s1 at some time t1 and having some other state s2 at a later time t2, one has temporally successive states only if one exists at both t1 and t2. But if one exists at both t1 and t2, one is temporally located. Thus, if one has temporally successive states, one is not timeless. Or, at least, not on Pike’s account.
So, on that account, the doctrine of divine timelessness entails the absence of successive states in God. That this follows from (2) indicates a serious inadequacy in Pike’s understanding of that doctrine. For, however surprising it might be, one can coherently maintain both that God is timeless (in a significant sense of ‘timeless’) and that states of the divine mind are successively ordered. This follows from the fact that God might be located at a time (or times) not temporally related to the present moment.
Here I need to introduce a notion which features prominently below—the notion of a temporal array. Let a time t be temporally related to a time t* just in case t is earlier than t*, simultaneous with t*, or later than t*. Roughly put, a temporal array is a set of times temporally related to one another. The present moment happens to be a member of infinitely many different temporal arrays; and among the temporal arrays having that moment as a member is one which has no members temporally related to times not belonging to it. Such a temporal array I refer to as a complete temporal array. More formally, these points can be put as follows:
(3) For any set of times T, T is a temporal array if and only if every member of T is temporally related to every other member of T.
(4) For any set of times T, T is a complete temporal array if and only if (i) T is a temporal array, and (ii) no time which is a member of T bears a temporal relation to a time which is not a member of T.
Let the complete temporal array to which the present moment belongs be t. Atemporalists affirm that
(5) For every time t, if t is a member of t, it is not the case that God exists at t.
From (5), (2) follows only if every time is a member of t. Is every time a member of t?
Since the existence of a time which is not a member of t entails the existence of at least one complete temporal array other than t, the question of whether such a time exists amounts to the question of whether t is the only complete temporal array. Is it? Perhaps, but perhaps not. For while t might be the only complete temporal array, it certainly seems possible that there be others.
Atemporalists who believe it possible that complete temporal arrays other than t exist might characterize eternity as just such an array. Such atemporalists might view eternity as a complete temporal array having only one member, namely, the divine present. Or they might view it as a complete temporal array having, say, infinitely many members. On the latter view, states of the divine mind might be successively ordered, even though God is not temporally located in t. Hence, atemporalists holding this view could affirm both that there is succession in the divine mind and that God is in a significant sense timeless. On such a view, states of the divine mind would be successively ordered, but God would not exist at any time temporally related to the present moment. Advocates of this view would affirm (5) but deny (2). So also advocates of the view that eternity is a complete temporal array which has only one member—namely, the divine present—would affirm (5) but deny (2).
Hereafter I refer to atemporalists who affirm (5) but deny (2) as relative atemporalists and to their view as relative atemporalism. I refer to relative atemporalists who view eternity as a complete temporal array having more than one member as extrinsic atemporalists and to their view as extrinsic atemporalism; I refer to those who view eternity as a complete temporal array having only one member as intrinsic atemporalists and to their view as intrinsic atemporalism. Not all atemporalists deny (2). I refer to theists who affirm not only (5) but also (2) as absolute atemporalists and to their view as absolute atemporalism. According to absolute atemporalists, eternity is notproperly characterized as a complete temporal array. Elsewhere I argue not only for atemporalism broadly construed but for absolute atemporalism. What follows, however, concerns itself not with any specific version of atemporalism but rather with atemporalism per se.
The Doctrine Defended
An examination of objections to the doctrine of divine timelessness is worth undertaking for at least two reasons. First, if any of these objections were decisive, whether God exists temporally would be a settled issue. Second, even if none of these objections were to settle the issue, they might nonetheless be instructive for its advocates, providing constraints for atemporalists and thus establishing parameters for delineating their view. Determining whether they in fact provide constraints and set parameters for atemporalists requires considering them carefully. So, then, the remainder of this essay comprises an examination of the more prominent objections to the doctrine of divine timelessness in the contemporary literature.
The Objection from Simultaneity
Perhaps the most prosaic objection to the doctrine of divine timelessness is one which arises from the transitivity of simultaneity. This objection finds eloquent expression in the work of Anthony Kenny, who takes the doctrine’s incoherence to follow from a straightforward understanding of simultaneity. His argument goes as follows:
Indeed, the whole concept of a timeless eternity, the whole of which is simultaneous with every part of time, seems to be radically incoherent. For simultaneity as ordinarily understood is a transitive relation. If A happens at the same time as B, and B happens at the same time as C, then A happens at the same time as C. . . . But, on [the atemporalist’s] view, my typing of this paper is simultaneous with the whole of eternity. Again, on [this] view, the great fire of Rome is simultaneous with the whole of eternity. Therefore, while I type these very words, Nero fiddles heartlessly on.
Kenny thus sees the transitivity of simultaneity as a serious threat to atemporalism.
Is he right? Well, if it follows from atemporalism that the burning of Rome is simultaneous with the typing of these words, then atemporalism is indeed untenable. So if Kenny’s argument turns out to be sound, atemporalists face insuperable difficulties. Fortunately for atemporalists, however, his argument does not turn out to be sound. To see this, consider the argument more closely. It goes as follows:
(6) For any times t, t*, and t**, if t is simultaneous with t* and t* is simultaneous with t**, then t is simultaneous with t**.
(7) If the doctrine of divine timelessness is true, then eternity is simultaneous with every moment in history.
(8) If eternity is simultaneous with every moment in history, then every moment in history is simultaneous with every other moment in history.
(9) But every moment in history is not simultaneous with every other moment in history.
(10) Therefore, the doctrine of divine timelessness is false.
The conclusion, (10), follows from (7), (8), and (9). (8) follows from (6). As nothing more than a statement of the transitivity of simultaneity, (6) is beyond reproach. (9) is surely true. So if (7) turns out to be true, the argument succeeds. As it turns out, however, (7) is not true. According to the doctrine of divine timelessness, eternity is not simultaneous with every moment in history. Indeed, it is not simultaneous with any moment in history. Since the doctrine stipulates that God is not located at any moment temporally related to the present one, eternity cannot be simultaneous with any such moment according to atemporalists. Hence, on their view, it cannot be simultaneous with every such moment.
In attributing (7) to the atemporalist, Kenny takes too literally atemporalist attempts to characterize God’s uninhibited access to every moment in history. Let us say that one has ‘uninhibited access’ to an event e if and only if there is no temporal distance between the occurrence of e and one’s access to e. While a temporal being might enjoy uninhibited access to certain events which occur in 2014 and certain other events which occur in 2017, its uninhibited access to events occurring in 2017 comes at the expense of its uninhibited access to events occurring in 2014. To be sure, such a being might have uninhibited access to each set of events. Even so, its uninhibited access to the first set must be temporally distant from its uninhibited access to the second set. But the timeless God suffers no such indignity. While he enjoys uninhibited access to events occurring in 2014 as well as uninhibited access to events occurring in 2017, no temporal distance stands between his uninhibited access to the first set and his uninhibited access to the second set. If atemporalists appropriate the language of simultaneity in order to make this point, they clearly do not use it in its ordinary sense. William Hasker is correct: “the statement about simultaneity is simply a metaphorical way of putting the point that all of time is ‘present’ in the ‘now’ of eternity.” So for the objection from simultaneity to succeed, atemporalist claims that eternity is simultaneous with every moment in time must be understood in a flat-footedly literal sense. Given their commitment to the doctrine of divine timelessness, however, atemporalists clearly do not intend such claims to be understood in this way. While the ordinary understanding of ‘simultaneous’ is ‘at the same time as,’ atemporalists use it to mean something like ‘at no temporal distance from.’ By thus appropriating ‘simultaneous,’ they make the point about God’s uninhibited access to all temporal events mentioned above. So, then, the objection from simultaneity fails.
The Objection from Agency
Theists hold that God created the universe and sustains it in being. Moreover, they claim that he plays an active role in the course of temporal events. He speaks to Moses from a burning bush, protects Daniel by shutting the mouths of lions, hears Jonah’s prayer and delivers him from a great fish. According to Robert Cook, however, “it is difficult to fathom how an atemporal God could do anything at all.” Being active, J. R. Lucas claims, presupposes being temporal. “To act purposefully,” William Kneale tells us, “is to act with thought of what will come about after the beginning of the action.” Lucas and Kneale thus suggest the notion of atemporal agency turns out to be incoherent. Unfortunately, however, they offer little by way of argument to support their suggestion.
Still, other opponents of atemporalism have offered arguments for eschewing atemporal agency. Richard Gale argues that
Our ordinary concept of causation involves some sort of temporal relation, which can be that of simultaneity, between cause and effect. This holds even for the notion of agent causation in which the cause is not an event but a person. God’s timeless causation is a species of such agent causation but one that has no temporal relation to its temporal effect. Our ordinary concept of causation does not make room for timeless causation…
If one then employs the “ordinary concept of causation,” the notion of atemporal agency makes no sense. Hence, if one wants to save that notion, one must “give a mystical interpretation” to the doctrine of divine timelessness. “But,” Gale argues, “the theist must pay a significant price for going this mystical route, namely, he winds up with a God who is a nonperson.” Now whether the ‘mystical route’ leads to a God who is not a person is an issue which need not detain us. For given the notorious problems associated with the concept of causation, I see no reason for thinking Gale correct that only such a route can save the notion of atemporal agency from incoherence. He assumes that all causes either precede or are simultaneous with their effects. Since an atemporal agent’s actions would neither precede nor be simultaneous with its effects, he concludes that no such agent is possible. But what atemporalist would grant the assumption upon which Gale rests this conclusion?
Even if effects cannot precede their causes (and this is a matter of debate), why can the atemporalist not claim that
(11) Necessarily, for any cause C and effect E, if C causes E, then it is not the case that E precedes C, rather than
(12) Necessarily, for any cause C and effect E, if C causes E, then either (C precedes E) or (C is simultaneous with E), is the appropriate principle here? Both (11) and (12) seem consistent with our experience of ‘ordinary causation.’ But unlike (12), (11) is also consistent with timeless agency. Since Gale’s argument hinges on eschewing (11) in favor of (12), he needs to argue for the truth of (12) over (11). Otherwise, he begs the question against the atemporalist. Unfortunately, he offers no such argument. Thus, Gale’s argument against atemporal agency fails.
The Objection from Omniscience
While those who raise the objection from agency claim that the doctrine of divine timelessness is inconsistent with the activities which theists traditionally ascribe to God, those who raise the objection from omniscience claim that it is inconsistent with the knowledge which theists traditionally ascribe to God. Perhaps the best-known statement of this objection comes in the following passage by Arthur Prior.
I want to argue against [divine timelessness], on the ground that its final effect is to restrict what God knows to those truths, if any, which are themselves timeless. For example, God could not, on the view I am considering, know that the 1960 final examinations at Manchester are now over; for this isn’t something that He or anyone could know timelessly, because it just isn’t true timelessly. It’s true now, but it wasn’t true a year ago (I write this on 29th August 1960) and so far as I can see all that can be said on this subject timelessly is that the finishing-date of the 1960 final examinations is an earlier one than 29th August, and this is not the thing we know when we know that those examinations are over.
Since, then, a timeless being could not know such propositions as
(13) The 1960 final examinations at Manchester are now over,
such a being could not be omniscient.
What Prior claims a timeless being could not know are tensed propositions. It might seem that such propositions can be reduced to dated, tenseless ones. So, for instance, one might think that
(13) The 1960 final examinations at Manchester are now over,
expressed on August 29, 1960, can be reduced to
(14) The 1960 final examinations at Manchester are (tenselessly) over on (or before) August 29, 1960.
If such a reduction were plausible, atemporalists could offer as a plausible response to Prior the argument that a timeless being could know propositions such as (14) and thus know ones such as (13) as well.
According to Prior, however, such a reduction is not plausible. Here, he states,
I cannot think of any better way of showing this than one I’ve used before, namely, the argument that what we know when we know that the 1960 final examinations are over can’t be just a timeless relation between dates because this isn’t the thing we’re pleased about when we’re pleased that the examinations are over.
So, on Prior’s view, propositions such as (13) are essentially tensed and cannot be reduced to tenseless ones. If propositions such as (13) turn out to be essentially tensed, it seems that no timeless being could so much as entertain them; and propositions one cannot entertain are also propositions one cannot know.
Although Prior’s claim that tensed propositions cannot be reduced to dated, tenseless ones is far from true, let us suppose it to be correct. Given this supposition, what plausible response can atemporalists make to the objection from omniscience? Well, as Patrick Grim points out, being omniscient seems no more compatible with the truth of essentially indexical statements such as ‘I am making a mess’ than it is with those such as ‘the 1960 final examinations at Manchester are now over.’ It thus follows that first-person knowledge creates for temporalists whatever problems knowledge of the present creates for atemporalists. So if arguments involving essentially indexical statements raise problems for atemporalists, they do so for temporalists as well. Or, to put the point differently, whatever force the objection from omniscience has against atemporalists, similar objections have the same force against theists more generally.
Given these considerations, then, all theists have good reason to conceive of divine knowledge in a way that does not fall prey to objections involving essentially indexed statements. This can be accomplished by understanding omniscience not in terms of what an omnipotent being knows but rather in terms of the cognitive power possessed by such a being. On such an understanding, being omniscient amounts to having unsurpassable cognitive power. Since one’s having unsurpassable cognitive power does not preclude the possibility that there are truths which one could not know, such an account provides atemporalists who agree with Prior about the irreducibility of tensed statements to tenseless ones with an obvious response to the objection from omniscience. Even if a timeless being could not know what time it is now, such a being may nonetheless possess unsurpassable cognitive power.
The Objection from Personhood
Stewart Sutherland claims timeless beings cannot utter, represent to themselves, create, deliberate, reflect, anticipate, intend, remember, suspect, confirm, or love. “The question which inevitably arises,” he tells us, “concerns what sense, if any, could be attached to the claim that a God who cannot do any of these could be regarded as a person, or even as ‘personal’.” He thus argues against the possibility of a timeless person. Since God’s personhood is an essential tenet of theism, a successful argument against the possibility of a timeless person would thoroughly undermine atemporalism.
Since this objection involves claiming both that there are certain activities in which a timeless being could not engage and that a necessary condition of being a person is being able to engage in those activities, atemporalists have two available responses. First, they can deny that being able to engage in the activities in question is necessary for being a person. Given the notorious difficulty of delineating an adequate conception of personhood, atemporalists can simply maintain that any conception of personhood according to which no timeless being could be a person ought ipso facto to be rejected. If Sutherland expects the objection from personhood to be persuasive, he needs to provide compelling reasons for thinking that the conception of personhood underlying it ought to be accepted; he provides no such reasons.
Second, atemporalists can argue that a timeless being could engage in at least some of the activities in question and thus count as a person. Consider, for instance, Moses’ request that God show him his ways. This request pleases God and Moses receives a favorable answer to it. If God is timeless, his being pleased does not occur after Moses makes the request. Moreover, Moses’ making the request does not occur before God answers it. It nonetheless remains true that both God’s being pleased and his answering Moses’ request result from Moses’ making it. If Moses had not made the request, God would neither have been pleased about his doing so nor have answered it favorably. In such a case, then, it makes perfectly good sense to maintain that Moses’ making the request affects God and that God’s answering it amounts to his responding to Moses. As far as I see, Sutherland does nothing to show such a case to be impossible. Thus, he fails to show that the conception of personhood he employs precludes the possibility of a timeless person.
The Objection from Theological Inadequacy
Although coherent, William Hasker maintains, the doctrine of divine timelessness turns out to be religiously inadequate. This follows, he tells us, from the account of divine knowledge to which the doctrine commits its advocates. On this, he states,
It seems inescapable . . . that if God is eternal, he knows us only by contemplating in eternity his own unchangeable “similitudes,” “images,” or representations of us. But I find this extremely difficult to accept as the truth of the matter. I can tell myself that an eternal God can still cause there to exist in time all of the events that we experience as his historical interventions, as his gracious presence in our lives, and the like. But that God in very truth knows us, and relates to us, only as the eternal representations in his own essence—this is a hard doctrine.
So in the end, Hasker concludes, the doctrine of divine timelessness “leaves too great a distance between the God who is affirmed theologically and the God who is known through Scripture and experience.”
Let us regard as too hard to accept a doctrine according to which God knows and relates to us only as “eternal representations.” Still, the question remains, why think the doctrine of divine timelessness is such a doctrine? Hasker gives his reasons for thinking this in the following passage.
One can be immediately aware only of what is present for one to be aware of; what else, after all, can “immediate” mean? If God is timeless, he can be immediately aware of (supposedly) temporal facts only if these facts really are timeless after all. If, on the other hand, the world really is temporal, only a temporal God can be immediately aware of it—and then only of its present, not of its past or future.
From the claim that objects of immediate awareness must be present to their knowers, Hasker thus concludes that a timeless God could not be immediately aware of temporal beings.
Put more formally, his argument goes as follows.
(15) For any knower K and object of knowledge O, if K is immediately aware of O, then O is present to K.
(16) Thus, if God is timeless, then God is not immediately aware of temporal beings.
Now, as it stands, this argument is invalid. In order to deduce (16) from (15), Hasker needs something like
(17) For any knower K and object of knowledge O, if K is timeless and O is temporal, then O is not present to K,
to be true. Of course, from (15) and (17) it follows that
(18) Thus, for any knower K and object of knowledge, if K is timeless and O is temporal, then K is not immediately aware of O.
And, moreover, (16) follows from (18).
In order for Hasker’s argument to succeed, then, both (15) and (17) must be true. In support of (15), Hasker simply asks what else ‘immediate’ can mean. But one is immediately aware of something if both one is aware of it and one’s awareness of it is not mediated by something else. That (15) follows from this is far from obvious. In fact, as I see it, such an understanding of ‘immediate’ provides no reason whatsoever for accepting (15). Moreover, while Hasker claims that a timeless being’s awareness of temporal beings would be mediated, he does not tell us what would mediate it.
Let us suppose that a timeless being is indirectly aware of some temporal being. What is mediating its awareness? The most plausible answer to this question is that those ‘eternal representations’ to which Hasker refers are mediating its awareness. Such representations amount to ideas (i.e., ‘images’ or ‘similitudes’) within a timeless being’s mind which correspond to the temporal beings known by such a being. Were he so to answer, however, Hasker would undercut his argument against the possibility of a timeless being’s immediately knowing temporal beings. For if such representations were mediating a timeless being’s knowledge of temporal beings, it seems that they would be directly related in some way to the temporal beings which they represent. But if they could be so related, why could not the timeless being itself be so related? So, then, Hasker’s admission that a timeless being could know temporal beings indirectly undercuts his argument that such a being could not know them directly.
Here it also seems worth noting that what it means for an object of knowledge to be ‘present’ to its knower is far from clear; and as long as this remains the case, determining what (15) means will be a difficult task. Still more difficult will be the task of determining whether it is true. Of course, if its meaning were perspicuous, perhaps the truth of (15) would be evident. Even so, Hasker will find little comfort in this fact. For, even if (15) were to turn out to be true, it also might turn out that temporal beings can be present to a timeless being in the sense of ‘present’ at issue. Thus, even if (15) is true, (17) might nonetheless be false.
For (17)’s truth, Hasker offers no argument. Since he himself does not invoke (17), this is hardly surprising. Still, if his argument is to succeed, he needs (17) (or something like it) to be true. Unfortunately, as Eleonore Stump and Norman Kretzmann point out, he seems simply to assume the truth of something like
(19) x can be directly aware of or epistemically present to y only if x and y share the same mode of existence.
Such an assumption, however, has little to commend it. Few theists would accept (19) with respect to space. In fact, since theists have traditionally maintained that God is spaceless, no traditional theist would accept (19). “And,” Stump and Kretzmann argue, “if traditional theists cannot accept [(19)] as applied to space, they cannot reasonably apply it to time.” So, then, traditional theists—both temporalists and atemporalists alike—have compelling reasons to reject (19). Moreover, given the apparent incompatibility between (17) and a timeless God’s direct awareness of temporal beings, few committed atemporalists are likely to grant the truth of (17). If then Hasker expects his argument to succeed, he needs to offer compelling arguments for affirming (17) (or something like it). Since he fails to provide such arguments and since the general principle which underlies (17) is inconsistent with traditional theism’s view of omnipresence, I conclude that the argument for (16) from (15) and (17) fails.
Having weighted five of the most prominent objections to atemporalism in the recent literature, I find them wanting. Of course, the failure of objections to the doctrine of divine timelessness does not entail its truth. So whether some version of atemporalism succeeds remains an open question. As mentioned above, I argue elsewhere for absolute atemporalism. Here I have attempted to show that atemporalism per se is neither indefensible nor one-dimensional.