05 / 06

Time, Eternity, And Eschatology

William Lane Craig


In order to explore the interface of time, eternity, and eschatology, we must first come to a differentiated understanding of these three notions.


Time is that dimension of reality whose constituent elements are ordered by relations of earlier than, simultaneous with, and later than and are experienced by us as past, present, and future. This much, at least, is common property among almost all disputants in debates about the nature of time. [1]

Beyond that point, philosophers are deeply divided about the nature of time. The controversy most relevant to the concerns of eschatology is the debate over whether time is tensed or tenseless. We are all familiar with tense as it plays a role in natural languages. But many philosophers hold that tense is an objective feature of reality as well, that things in time are really past, present, or future. Other philosophers regard tense as purely mind-dependent: things in time are no more objectively "now" than things in space are objectively "here." Following J. M. E. McTaggart, philosophers have called these two views of time the A-Theory and the B-Theory respectively.

According to the A-Theory things/events in time are not all on an ontological par: the future does not yet exist and the past no longer exists; only things which are present are real. Temporal becoming is an objective feature of reality: things come into being and go out of being. By contrast, on the B-Theory all events in time are equally real, and temporal becoming is an illusion of human consciousness. Pastness, presentness, and futurity are at most relational properties: relative to the persons living in 2050 the people and events of 2000 are past, but relative to the persons living in 1950 the people and events of 2000 are future. Things and events in time are objectively ordered by the relations earlier than, simultaneous with, and later than, which are tenseless relations that are unchanging and obtain regardless of whether the related events are past, present, or future relative to some observer. B-Theorists typically unify time with space into a four-dimensional, geometrical entity called space-time, all of whose points are equally real and none of which is objectively present.

A pure B-Theory, then, involves a tenseless ontology devoid of temporal becoming; such a view is often called "eternalism." By contrast, a pure A-Theory involves a dynamic ontology exhibiting mind-independent temporal becoming, with tense an objective feature of reality and the present as ontologically privileged; such a view is usually called "presentism."

Between these pure theories some thinkers have sought to locate hybrid A/B-Theories, sometimes by combining objective tenses with a static ontology or by treating the past and present as on an ontological par in distinction from the future (Fig. 1).

Diagram depicting ontologies of Competing Theories of Time.

Fig. 1. Ontologies of Competing Theories of Time. The vertical dimension represents time and the two horizontal dimensions represent the three dimensions of space. The depicted universe begins in a Big Bang and ends in a Big Crunch. On the pure A-Theory only the things existing on the present "slice" of space-time are real. On the pure B-Theory all things and events in space-time are real, and no slice is present. On hybrid theories, there is an objective present, but it is not ontologically privileged: non-present things and events are equally real.

It is doubtful, however, whether these hybrid theories are coherent. As McTaggart himself argued, on these theories some event e must have at one time the property being present and then at a later time the property being past. But then by the Principle of the Indiscernibility of Identicals, e cannot be the same event, since it possesses different properties, which contradicts the hypothesis. The A-Theorist avoids the problem by holding that the only properties an entity has are those it presently has, so that e does not have incompatible properties. The B-Theorist avoids the problem by treating tensed properties as relational rather than monadic properties, so that e is (tenselessly) present relative to certain (simultaneous) events and past relative to other (later) events. But the hybrid A/B-Theorist seems unable to avoid what has come to be known as McTaggart's Paradox.

Thus, our choice seems to be between presentism and eternalism as our ontology and a pure A- or B-Theory as our theory of time. Weighty arguments can be brought to bear on both sides, and the controversy shows no sign of abating. Indeed, an almost one-to-one correspondence could be constructed between A- and B-Theorists of comparable stature among philosophers of time and space. Fortunately, we need not adjudicate that debate here, for our interest is in the comparative eschatological import of competing theories of time and eternity.




Eternity is the property of existing permanently, without beginning and end. Since our concern is theological, we focus on the nature of divine eternity. All disputants concur that divine eternity is the property of God's existing permanently, having neither beginning nor end of existence. But, as with the nature of time, there the agreement ends. Philosophers and theologians alike are deeply divided over the nature of divine eternity, debating whether God's eternity is to be construed as a state of timelessness or of infinite, omnitemporal duration.

The question here is whether God exists temporally or atemporally. An entity exists temporally just in case it has a temporal location and temporal extension. To have a temporal location is to exist at a time, so as to stand in temporal relations of simultaneity with other entities existing at that time and posteriority or priority to any entities that either have existed or will exist relative to that time. Temporal extension is an entity's duration, roughly, how long it exists. Even an entity which exists for only an instant has an extension, the measure of which is, in this case, zero. As such it differs from an atemporal entity, for the category of temporal extension does not even apply to a timeless entity, whereas any temporal entity has an extension, even if the measure of that extension happens to be zero.

An entity exists atemporally just in case it is not temporal. This explication makes it evident that temporality and atemporality are contradictories; one is the negation of the other. On pain of contradiction, then, God cannot be said to be both temporal and atemporal, unless one is able to provide some sort of model whereby these properties are qualified. For example, an entity can be both black and non-black by being black at one time and non-black at another. But such a model is inapplicable in the case of temporality, for it would be incoherent to affirm that God is temporal at one time and atemporal at another time. And, of course, even on such a model God could not at the same time be both temporal and atemporal.

Although some contemporary theologians are wont to speak of God's eternity as being atemporal yet embracing time or as transcending time yet positively related to time, no model is ever offered for these affirmations, so that they remain either vague metaphors or mere restatements of the problem at hand. A few thinkers have interpreted such expressions in terms of God's extra-dimensionality: our temporal dimension is embedded in a divine hyper-time (Fig. 2).

Graph depicting our temporal dimension is embedded in a divine hyper-time.

Fig. 2. Our temporal dimension t is represented by the vertical axis, which is orthogonal to God's embedding hyper-time T represented by the horizontal axis. At successive moments of hyper-time T1, T2, T3, . . . our entire temporal series of moments t1, t2, t3, ... exists. Our whole time line thus has both temporal location and extension in divine hyper-time. Since God exists permanently, hyper-time must extend infinitely in both directions, and our time dimension must be created by God at some hyper-time Tn.

But such a metaphysical extravagance (whose possibility is moot) solves nothing. For if God exists in hyper-time, then He is temporal, not atemporal, albeit it in a different temporal dimension than the one we experience.

Again, there are weighty arguments on both sides of the debate over the nature of divine eternity. Christian tradition has favored divine timelessness, but the consensus has recently turned very markedly in favor of divine omnitemporality. Since our interest is comparative, we may also leave this debate unresolved.

On the basis of the distinctions we have drawn there appear to be four options available with respect to the problem of God and time (Fig 3).

Chart showing options for God and Time.

Fig. 3. Options for God and Time.

The first question that presents itself is whether all of these options are coherent.

Now prima facie options (2) and (3) seem to be coherent. According to (2) the entire four-dimensional space-time manifold exists as a whole. Time is merely an internal dimension of the manifold, ordering events and moments by tenseless temporal relations. But the space-time block itself exists timelessly. It is therefore easy to see how God could co-exist timelessly with the universe. The universe is ontologically dependent upon God, even though it is co-eternal with Him. Such a view may be theologically objectionable to some who regard a timeless person as impossible and religiously unavailable in any case; but this is an objection to divine atemporality per se, not to its coherence with a B-Theory of time.

Similarly, the combination of divine temporality and an A-Theory of time envisioned in option (3) seems perspicuous. Again, the affirmation that God is temporal may strike some theologians as unacceptable, since divine simplicity and immutability must then be sacrificed in view of God's changing knowledge (e.g., of what is happening now) and relations (e.g., causing the Red Sea to part). But that is an objection to divine temporality as such, not to the coherence of divine temporality with the A-Theory. On this view there is an objective present in which God exists, and He experiences the phenomenal flow of time as the world undergoes real, temporal becoming. It has been objected that such a view is incompatible with divine foreknowledge of future contingents. But A-Theorists are by no means committed to the denial of the Principle of Bivalence for future contingent statements (a deeply problematic and, hence, rarely defended position), and there is no good reason to think that the truth of such statements or God's knowledge of them in any way prejudices human freedom. [2] Thus, the combination of divine temporality with an A-Theory of time remains a live option.

On the other hand, options (1) and (4) are dubiously coherent. Option (1) is evidently the more problematic, trying to combine as it does a tensed theory of time with God's timelessness. For if God is really causally related to the world and temporal becoming is real, then it is extraordinarily difficult to see why God would not be constantly changing in His relations with the world, and, hence, temporal. Thomas Aquinas recognized this problem and as a solution appealed to the doctrine that God sustains no real relations to creatures, an expedient which seems as desperate as it is implausible. Some contemporary thinkers have tried to solve the problem by crafting an eternal-temporal simultaneity relation; but this attempt has been widely rejected as vacuous or viciously circular. Moreover, option (1) requires that there be tensed facts such as "Richard the Lionheart led the Third Crusade" which must be known to an omniscient God. But God cannot know such facts without locating Himself temporally (in this case after the Third Crusade), which is incompatible with divine timelessness. At best a timeless deity could know such tenseless facts as "Richard the Lionheart leads the Third Crusade in 1192," which leaves Him ignorant of whether King Richard has yet engaged in his exploits.

Option (4) is more bizarre than incoherent, so that it has been almost universally ignored in discussions of divine eternity. If the four-dimensional space-time manifold exists timelessly, then it seems maladroit to hold that God is not also timeless but is located on the temporal dimension and extended as a one-dimensional "worm" (assuming divine incorporeality) along its length. It might be thought that such a view is incompatible with the world's ontological dependence on God, but there is no evident reason why the space-time manifold could not be produced from within, so to speak, rather than from without. The more serious problem with this view is that if (as seems plausible) the B-theory of time implies that temporally extended entities are composed of temporal parts (on which notion see below), then God is either not a person (if He is the entire one-dimensional worm) or else is not the same person at different times (if He is one segment or temporal part of the worm). Moreover, God, in order to act in a timely fashion, must on such a view hold tensed beliefs like "Christ is presently on the way to Golgotha," and it is hard to reconcile His cognitive perfection with such an illusory grasp of reality.

This analysis suggests that as goes the A- or B-Theory of time so goes divine temporality or timelessness. Divine timelessness is most naturally associated with a B-Theory of time, while an A-Theory of time seems to imply that if God exists, He exists now, i.e., temporally.




Finally, eschatology is the description of the destiny of the spatio-temporal world and its human inhabitants. Although some theologians have used the word "eschatological" to mean little more than "pregnant with theological significance," such a de-temporalized language is etymologically unjustified and fails to connect with traditional usage. As the doctrine of the last things, eschatology is an inherently temporal notion. It deals with the history of mankind and the universe in the later than direction.

Eschatology is no longer exclusively the subject matter of theology but has in the last quarter century or so emerged as a new branch of cosmology (the study of the large-scale structure and history of the universe), being a sort of mirror image of cosmogony, that branch of cosmology that studies the origin of the universe. Not that the future of the universe will resemble its past; far from it. But just as physical cosmogony looks back in time to retrodict the history of the cosmos based on traces of the past and the laws of nature, so physical eschatology looks forward in time to predict the future of the cosmos based on present conditions and laws of nature.

Physical eschatology paints a very bleak and very different picture of the future than that of theological eschatology. The most likely scenario based on present scientific evidence is that the universe will continue to expand forever. As it does, the stars eventually burn out and the galaxies grow dark. Around 1015 years after the Big Bang most of the baryonic mass of the universe will consist of degenerate stellar objects like brown dwarfs, white dwarfs, neutron stars, and black holes. Elementary particle physics suggests that around 1037 years protons will decay into electrons and positrons, filling space with a rarified gas so thin that the distance between two particles will be about the size of the present galaxy. At around 10100 years, the commencement of the so-called Dark Era, black holes themselves may have evaporated. The mass of the universe will be nothing but a cold, thin gas of elementary particles and radiation, growing ever more dilute as it expands into the infinite darkness, a universe in ruins.

A bleak picture, indeed; but as Freeman Dyson has reminded us, the predictions of physical eschatology are subject to the proviso that intelligent agents do not interfere with the envisioned natural processes. [3]  If intelligent beings are able significantly to manipulate natural processes, then the actual future of the cosmos could look quite different than the trajectory predicted on the basis of laws and present conditions. Dyson's own attempt to craft a scenario whereby immanent agents might stave off extinction, is, doubtless, desperate and implausible. [4]  But why should we restrict our attention to immanent agents? Theists believe in the existence of an intelligent being who is the Creator of the space-time universe and transcends the laws that govern the physical creation. On the Christian view God will bring about the end of human history and the present cosmos at such time as He deems fit (Mk. 14.32; Mt. 24.43; 1 Thess. 5.2; Heb. 1.10-12; 2 Pet. 3.10; Rev. 3.3). He will not allow events predicted on the basis of present trends in even the relatively near future, such as the extinction of the human race, to occur, much less events in the unfathomably distant future such as stellar extinction or proton decay. Before these events can take place, God will act to terminate human history and usher in a new heavens and a new Earth (1 Cor. 15.51-52; 1 Thess. 4.15-17; Rev. 21.1).

What this implies is that the findings of physical eschatology are at best projections of the future course of events rather than actual descriptions. They tell us with approximate accuracy what would take place were no intelligent agents to intervene. Thus, the findings of physical eschatology are in no way incompatible with Christian eschatology, since those findings involve implicit ceteris paribus conditions with respect to the actions of intelligent agents, including God.

Of course, physical eschatologists might ask whether there is any reason to take seriously the hypothesis of a transcendent, intelligent agent with requisite power over the course of nature to affect the projected trajectories of physical eschatology. Intriguingly, physical eschatology itself furnishes grounds for taking seriously such a hypothesis. Already in the nineteenth century with the enunciation of the Second Law of Thermodynamics, scientists realized that the application of the Law to the universe as a whole (which, on naturalistic assumptions, is a gigantic closed system, since it is all there is) implied a grim eschatological conclusion: given sufficient time, the universe would eventually come to a state of equilibrium and suffer heat death. But this apparently firm projection raised an even deeper question: if, given sufficient time, the universe will suffer heat death, then why, if it has existed forever, as naturalists assumed, is it not now in a state of heat death? If in a finite amount of time the universe will inevitably come to equilibrium, from which no significant further change is physically possible, then it should already be at equilibrium by now, if it has existed for infinite time.

The advent of relativity theory and its application to cosmology altered the shape of the eschatological scenario predicted on the basis of the Second Law of Thermodynamics but did not materially affect the fundamental question. Assuming that there is no positive cosmological constant fueling the expansion of the universe, that expansion will decelerate over time. Two radically different eschatological scenarios then present themselves. If the density of the universe exceeds a certain critical value, then the internal pull of the universe's own gravity will eventually overcome the force of the expansion and the universe will collapse in upon itself in a fiery Big Crunch. There is no known physics that would permit the universe to bounce back to a new expansion prior to a final singularity or to pass through the singularity into a subsequent state. On the other hand, if the density of the universe is equal to or less than the critical value, then gravity will not overcome the force of the expansion and the universe will expand forever at a progressively slower rate. Because the volume of space constantly increases, the universe will never actually arrive at equilibrium, since there is always more room for entropy production. Nonetheless, the universe will become increasingly cold, dark, dilute, and dead.

Very recent discoveries provide strong evidence that there is effectually a positive cosmological constant which causes the cosmic expansion to accelerate rather than decelerate. Paradoxically, since the volume of space increases exponentially, allowing greater room for more entropy production, the universe actually grows further and further from an equilibrium state as time proceeds. But the acceleration only hastens the universe's disintegration into increasingly isolated material particles no longer causally connected with similarly marooned remnants of the cosmos.

Thus, the same pointed question raised by classical physics persists: why, if the universe has existed forever, is it not now in a cold, dark, dilute, and lifeless state? If one is to avoid the inference that the assumption on which the question is based—viz., that the universe has existed forever—is false, then one must find some scientifically plausible way to overturn the findings of physical eschatology so as to permit the universe to return to its youthful condition.

But no realistic and plausible scenario is forthcoming. Andrei Linde has proposed a future-eternal inflationary model in which inflation begets inflation, ad infinitum, and he once proposed that such a model might be extended infinitely into the past, with the result that the beginning of the universe was averted. But it has been shown that future-eternal inflationary space-times cannot be past-eternal: they must involve initial singularities and so an absolute beginning of the universe. [5] In any case, there is no reason to think that further inflation would subvert the implications of the Second Law or serve to restore the universe's youth. On the contrary the effect of further inflation would be to increasingly isolate stars and galaxies in the universe until they cease to be in causal contact with one another and so disappear from view.

It has been suggested that the universe might undergo quantum tunneling into a radically new state. For example, if the universe were currently in a false vacuum state, then it could eventually tunnel into a lower-energy vacuum state. In going through such a phase transition all the physical constants' values would change and a totally new universe would emerge. Now even if such a transition were to take place, the probability that the values of all the constants would fall into the unimaginably narrow life-permitting range is vanishingly small (a staple of discussions of cosmic fine-tuning). Hence, it is highly improbable that our present life-permitting constellation of physical constants is the chance result of such a phase transition from a higher-level vacuum state about 13 billion years ago. Worse, if there is any non-zero probability that such a meta-stable state would tunnel to a true vacuum state, then given infinite past time it should have already occurred infinitely long ago, not just 13.5 billion years ago. But then it again becomes inexplicable why the universe is not already dead.

Speculations about our universe begetting future "baby universes" have also been floated in eschatological discussions. It is conjectured that black holes may be portals of wormholes through which bubbles of false vacuum energy can tunnel to spawn new expanding baby universes, whose umbilical cords to our universe may eventually snap as the wormholes close up, leaving the baby universe an independently existing spacetime. Could this eschatological scenario be extrapolated into the past, such that our universe is one of the baby universes spawned by the mother universe or by an infinite series of ancestors? It seems not, for while such baby universes appear as black holes to external observers, an observer in the baby universe itself will see the Big Bang as a white hole spewing out energy. But this is in sharp contrast to our observation of the Big Bang as a low-entropy event with a highly constrained geometrical structure. And again, what rescues the infinite sequence of cosmic descendants from the consequences of the Second Law of Thermodynamics is unclear.

If such speculative conjectures fail to commend themselves, then we are left with the hypothesis that the Big Bang represents the absolute beginning of the universe, just as in customary Big Bang models; and the low entropy condition was simply put in as an initial condition.

At this point theories of time become crucial. If we adopt an A-Theory of time, according to which temporal becoming is real, then the conclusion that the universe had a beginning has profound metaphysical significance. On a B-Theory of time the beginning of the universe is no more important than its end, since for the universe to have a beginning (or an end) is simply for the four-dimensional, timelessly existing, space-time block to be finite in a certain direction and to have an edge (if its terminus is a singularity). But on an A-Theory of time the beginning of the universe is the point at which the universe literally came into being. The universe does not transition from nothingness into something; rather it comes into being absolutely. But if anything seems metaphysically impossible, it is that something can come into being absolutely without a cause. Being only comes from being. There must therefore be causally prior (if not temporally prior) to the Big Bang an ultra-mundane cause of the universe.

Such a cause must transcend physical space and time and therefore be immaterial not physical. Since the only immaterial entities of which we know are either minds or abstract objects (like numbers), and since the latter do not stand in causal relations, it is plausible that the cause of the universe is an unembodied mind or person who created the universe. Thus, physical eschatology itself—in combination with an A-Theory of time—provides grounds for believing in the existence of just that sort of agent who is capable of altering the projections of physical eschatology.

The naturalist might insist that we have no good reason for thinking that the personal Creator would intervene in the natural world so as to avert the consequences toward which the universe tends. But Christian eschatology is inextricably bound up with the person of Jesus of Nazareth: his physical resurrection is the harbinger not only of our own eschatological resurrection but also of a sort of cosmic resurrection as well (Rom. 8.19-23). The Christian eschatological hope is therefore based on the historical reality of Jesus' resurrection.

It is perhaps not appreciated outside the field of New Testament studies how impressive the historical credentials for this remarkable event are. [6] Today the majority of New Testament historians who have written on the topic agree that (1) Jesus of Nazareth was executed by crucifixion under Roman authority; (2) Jesus' corpse was then laid in a tomb by Joseph of Arimathea, a member of the Jewish Sanhedrin; (3) On the Sunday morning after the crucifixion Jesus' tomb was found empty by a group of his women followers; (4) Thereafter, different individuals and groups under a variety of circumstances experienced appearances of Jesus alive; and (5) The earliest disciples came suddenly and sincerely to believe that God had raised Jesus from the dead, despite their strong predisposition to the contrary. While these points are by no means uncontested, they nonetheless do represent the mainstream view.

The remaining question is how these facts are best explained. We have seen reason to think that a transcendent, personal Creator of the universe exists. In that light, it can be plausibly argued that, when assessed by such standard criteria as explanatory power, explanatory scope, plausibility, and so forth, the Resurrection Hypothesis (viz., "The God of Israel raised Jesus from the dead") emerges as the best explanation. [7] If that is the case, then the prospect of an eschatological return of Christ to inaugurate fully the Kingdom of God, with a new heavens and a new Earth, cannot be dismissed as mere mythology.

The plausibility of Christian eschatology vis à vis the projections of physical eschatology is therefore inherently bound up with one's ontology. If, as physical eschatology itself intimates, there exists a personal, transcendent agent who created the universe with all its natural laws and boundary conditions, and if that agent has raised from the dead Jesus of Nazareth, who promised his eschatological return, then it is eminently rational to entertain "the blessed hope" of Christian eschatology, while accepting the findings of physical eschatology as more or less accurate projections based on present conditions.


Time, Eternity, and Eschatology


We have already seen that theories of time intersect crucially with theories of divine eternity and with cosmogony. This is also the case with eschatology, in particular with physical eschatology. As a result of the anti-metaphysical prejudices of Positivism, with its verificationist theory of meaning, and the naturalized epistemology of W. V. O. Quine, which takes the deliverances of the physical sciences to be determinative of reality, the almost unquestioned assumption of contemporary physical theorists and philosophers of time has been that time is that quantity which is studied under that name by physics. But in a review of the time concept in various fields of physics, Carlo Rovelli has emphasized how unlike the intuitive notion of time physical time concepts are and how diverse they are when compared among themselves. [8] He lists eight characteristics commonly associated with time:

1. One dimensional: time can be thought of as a collection of instants which can be arranged in a one-dimensional line.

2. Metric: time intervals can be measured such that two intervals can be said to have equal duration.

3. Temporally global: the real variable t which we use to denote the measure of time goes through every real value from -infinity to +infinity.

4. Spatially global: the time variable t can be uniquely defined at all space points.

5. External: the flow of time is independent of the specific dynamics of the objects moving in time.

6. Unique: there are not many times, but just the time.

7. Directional: it is possible to distinguish the past from the future direction of the time-line.

8. Present: there always exists a preferred instant of time, the Now.

Rovelli then provides the following chart to illustrate the diversity of physical time concepts (Fig. 4).

The Notion of Time Used in: has properties:
Ordinary language 1, 2, 3, 4, 5, 6, 7, 8
Thermodynamics 1, 2, 3, 4, 5, 6, 7
Newtonian mechanics 1, 2, 3, 4, 5, 6
STR 1, 2, 3, 4, 5
Cosmology 1, 2, 3, 4
GTR - proper time 1, 2, 3, 5
GTR - coordinate time 1, 3, 4
GTR - clock times 1, 2
Quantum gravity none

Fig. 4: The concept of time in various fields of physics in comparison with the customary concept.

This lack of a unified concept ought to suggest already that in physics that with which we have to do is not time itself, but physical quantities which serve as stand-ins or measures of time and which differ from field to field.

That temporality is not inherently connected to the occurrence of physical events is evident from the fact that a succession of mental events alone is sufficient for a temporal series. Thus, if God (or an angel) were to count down to creation, ". . . , 3, 2, 1, fiat lux!", that sequence of mental events would suffice for a temporal series wholly in the absence of physical space and time and, indeed, literally prior to the inception of physical time. There would in such a case exist a metaphysical time of which physical time is merely the measure. Time itself could thus extend beyond the Big Bang and the Big Crunch, though these mark the beginning and end of physical time respectively. [9]

The question of a beginning and end of time serves to expose a further deficiency of a purely physical eschatology: it has no basis for the asymmetry of time. Time's asymmetry consists of time's anisotropy and orientation. A series can be anisotropic—that is, varying in different directions—without being oriented. For example, a series of balls arranged by size is anisotropic, being ordered by the smaller than and larger than relations; but such a series lacks any orientation or ordinal structure, and thus lacks intrinsic direction. By contrast, time is both anisotropic and oriented, being ordered by the earlier than/later than relations, which not only render moments discernible but also accords to time a direction (the "arrow of time").

Physical eschatology by its very nature is committed to the asymmetry of time, for the eschaton is the state which comes last, not first. Physical eschatology studies our future cosmic development, not our past, which is the study of cosmogony. But what basis is there in purely physical states that justifies such a differentiation? An enormous amount of ink has been spilled in the attempt to ground time's asymmetry in physical processes, such as entropy increase or the expansion of the universe. But all such attempts are question-begging, presupposing what they purport to ground. The very statements that entropy increases in a closed system or that the universe is expanding presuppose that there already exist earlier than and later than directions in which these processes occur. The universe exhibits physical anisotropy in that its physical description differs at different times. But that anisotropy provides no justification for saying that the Big Bang is first and the Big Crunch, for example, is last, rather than the reverse. Appeal to the Second Law of Thermodynamics is nugatory, for there is no physical justification for calling the direction of entropy increase "later" rather than "earlier." The very statement of such laws or de facto physical asymmetries presupposes the asymmetry of time rather than grounds it. To think that the states of the universe are temporally asymmetric on the basis of physical anisotropy is to act like the benighted carpenter who threw away half his nails because he thought the point was at the wrong end!

Thus, if physical eschatology is even to make sense, the discipline must be grounded in the nature of time itself. Here the A- and B-Theories of time provide interestingly different perspectives on eschatology. A-Theorists may ground the asymmetry of time in the objectivity of temporal becoming, which gives time a unique orientation due to the impossibility of backward becoming. Relations of earlier than/later than can be reductively analyzed in terms of the ordinal notions occurring first/occurring second or in terms of the tenses past, present, and future. Thus, the A-Theory of time makes eschatology a meaningful notion.

Eschatology is meaningful as well on the B-Theory, since that theory also affirms temporal asymmetry, though in the absence of tense or temporal becoming. But the question which arises here is whether the B-Theory has the metaphysical resources for the reality of time's asymmetry which it affirms. Given that temporal asymmetry cannot be grounded in physical anisotropies, the B-Theory seems only gratuitously to affirm that the tenseless ordering relations among events are genuinely temporal relations. Most B-Theorists simply take the earlier than/later than relations to be primitives of the theory, without further grounding. But given that the A-theorist can ground those relations in the reality of temporal process, the B-Theorist's affirmation of time's asymmetry appears to involve theft rather than honest toil. Given the ontological parity of all events and the illusory nature of tense according to the B-Theory, it is hard to resist the impression that the B-Theorist's ontology is really timeless. For that reason some eternalists have flatly denied the asymmetry of time, holding that time is wholly isotropic and directionless. Objectively speaking, there is no beginning and end of the universe, but merely termini of the universe, and which one is labeled "initial" and which "final" is utterly arbitrary. On such a view of "time" there is strictly no place for eschatology at all.

There are other ways in which one's theory of time will affect one's eschatology. For example, if an A-Theory of time is correct, then God is plausibly temporal. As such the eternal God literally shares our eschatology, that is to say, though He foreknows the future, He like us anticipates the events involved and will experience them as they happen. By contrast, if the B-Theory implies divine timelessness, then the truth of that theory implies that God Himself has no eschatology, for He transcends temporal process. For Him all events in the four-dimensional space-time manifold are equally real. While God knows that certain events are tenselessly located later than certain other events on our respective world lines, for Him they are just "there," so to speak, and there can be no question of His waiting for them to occur or experiencing them as present.

Some detractors of the B-Theory have alleged that the theory is incompatible with human freedom, since the future is as real as the past and is therefore determined or, at least, fated to occur. Such fatalism would undermine the significance of eschatology, since everything would come to pass necessarily. But this allegation is misconceived. Libertarian freedom requires the absence of causal determinism, sometimes called the liberty of opposites. That this is in no way abrogated by the B-Theory is evident by the fact that the theory is wholly compatible with indeterministic quantum mechanics. The conditions obtaining up to some time t do not determine what actually occurs at t, even if the events at and subsequent to t are as real as events prior to t. Neither are events at t non-causally "fated" to occur (if we can even make sense of such an idea), for the reality of the consequences of one's choices only shows that one will not choose differently, not that one cannot choose differently. If one were to choose differently (as one is perfectly capable of doing), then there would be different events located in space-time as a consequence of those choices. Thus, the B-Theorist cannot be justly accused of undermining human freedom.

Nonetheless, the B-Theory does have other implications for eschatology which are troubling. In general, one gets a less robust eschatology on a B-Theory than on an A-Theory. If there really is no future, then eschatology also takes on the appearance of an illusion. On an eternalist ontology, there is no becoming, no realization, no actualization of things and events. The "not yet," so central to eschatology, is an illusion of human consciousness. Similarly, for the B-Theorist activities such as waiting, dreading, anticipating, and so forth, are all ultimately irrational, since temporal becoming is illusory. Events are fixed in their tenseless temporal locations and not in any sense approaching so as to be differently regarded than past events. Yet these notions seem vital to an existentially meaningful eschatology.

The problem of evil becomes especially acute on a B-Theory of time. On an A-Theory of time, once the eschaton arrives, evil, being part of the past, disappears forever from creation, thanks to the objectivity of temporal becoming. But on an eternalist ontology, evil is never really vanquished from creation. Evil's being destroyed amounts to no more than later portions of the space-time block's being free of evil. But the earlier parts infected with evil exist just as robustly as the later parts. The stain of evil on creation is indelible. What this implies for Christ's crucifixion and resurrection is especially disturbing. In a sense Christ (or at least that temporal part of him) hangs permanently on the cross. The victory of the resurrection seems hollow, since the death and suffering are never really over and done away with. What kind of eschatology is it that cannot erase evil from the universe?

The B-Theory also has problematic implications for divine judgement. The problem here arises as a result of how things are conceived to persist through time. Endurantists believe that objects endure through time, that the same object that exists at t1 exists at t2, that material objects have spatial but not temporal dimensions so that they exist wholly at a time. Perdurantists believe that objects are extended in time, that they have temporal as well as spatial dimensions, that they therefore have temporal parts or slices. On perdurantism the part of a temporally extended object that exists at t1 is obviously not the part that exists at t2, for they have different properties, and the Principle of the Indiscernibility of Identicals therefore requires them to be non-identical.

Perdurantism construes the persons with whom we have intercourse to be mere parts of extended, spatio-temporal "worms" which are not persons, for the whole extended object is neither self-conscious nor endowed with freedom, intentionality, and the like, properties essential to personhood. But then the eschatological doctrine of divine reward or punishment becomes incoherent. For the person who appears before God at the judgement is a different person from the person who did the good deed or committed the crime. But it would be immoral to accord praise or blame to a person who did not do the praise- or blameworthy acts. Alternatively, if one considers a temporal part large enough to have been located both at the time of the action and the time of judgement, then God must also accord praise or blame to an infinite number of other shorter parts who again are present at the judgement but not at the actions, since the extended part can be divided into all the smaller parts, each of which is a person. Thus, perdurantism makes a mockery of moral praise and blame.

The B-Theory, unfortunately, seems to entail perdurantism. For objects which exist at different times have by that very fact discernible properties and therefore cannot be identical. Hence, objects cannot endure through time on an eternalist ontology. Some B-Theorists have tried to resist this conclusion by adopting a doctrine called Adverbialism, which holds that things do not have different properties at different times, but that the way things have their properties is modified by dates, e.g., John F. Kennedy has the property of being assassinated 1963ly. This attempt to avoid perdurantism seems, however, to conflate semantics and ontology. Semantically tenses operate on sentences adverbially; but commitment to tenses ontologically is a peculiar feature of the A-Theory. The B-Theorist has no place for tense in his ontology, so that the semantical function of dates as adverbs is irrelevant to things' actually possessing different properties (tenselessly) at different times. Having a property 1963ly is, on an eternalist ontology, no different from possessing a property in (or at) 1963. So if there were enduring objects on an eternalist ontology, such an object would possess different properties at different times, for example, being elected President in 1960 and being assassinated in 1963. But then the Principle of the Indiscernibility of Identicals requires that the person possessing these properties is not the same person. Hence, on a B-Theory of time objects do not endure through time; rather they perdure.

Perdurantism not only renders moral praise and blame problematic but also raises another related difficulty for eschatology. [10] Suppose that God will permit a person into heaven based on whether that person is in a state of grace either at death or at Christ's return. Take the latter case. Suppose that some person has been in a state of grace for much of his life but later apostasises, so that upon Christ's return he is no longer in a state of grace. On an A-Theory of time, he exists wholly at that moment, and God judges him based upon the condition that he is actually in. But on a perdurantist view the temporal part of the extended object that exists at Christ's return is not in any way privileged over earlier temporal parts that were in a state of grace. Therefore, God's judgement of that individual on the basis of his final temporal part (or, alternatively, God's selectively judging only persons who are the final temporal parts of spacetime worms) is arbitrary and unjust. Thus, the perdurantist entailments of a B-Theory of time seem to undermine various, important aspects of theological eschatology. [11]



Adams, Fred, and Laughlin, Gregory. "A Dying Universe: The Long-term Fate and Evolution of Astrophysical Objects." Reviews of Modern Physics 69 (2) (April 1997): 337-72. A seminal article surveying the predictions of physical eschatology.

Cirkovic, Milan M. "Entropy and Eschatology: a Comment on Kutrovátz's Paper 'Heat Death in Ancient and Modern Thermodynamics'." Open Systems and Information Dynamics 9 (2002): 291-99. Critique and development of Kutrovátz's inquiry into cosmic heat death.

_______. "Physical Eschatology." American Journal of Physics 71 (2) (February 2003): 122-33. A very helpful annotated bibliography on various aspects of physical eschatology.

Craig, William Lane. God, Time and Eternity. Dordrecht: Kluwer Academic Publishers, 2001. A thorough treatment of God's relationship to time.

________. "Naturalism and Cosmology." In Analytic Philosophy without Naturalism. Ed A. Corradini, S. Galvan, and J. Lowe. London: Routledge, 2005. A survey of the evidence of an absolute beginning of the universe in contemporary cosmology.

________. The Tensed Theory of Time: A Critical Examination. Synthese Library 293. Dordrecht: Kluwer Academic Publishers, 2000. A thorough examination of the arguments for and against the A-Theory of time.

________. The Tenseless Theory of Time: A Critical Examination. Synthese Library 294. Dordrecht: Kluwer Academic Publishers, 2000. A thorough examination of the arguments for and against the B-Theory of time.

Ganssle, Gregory, ed. God and Time: Four Views. Downer's Grove, Illinois: Inter-Varsity, 2001. A non-technical discussion of four competing view of divine eternity: absolute timelessness (Paul Helm), relative timelessness (Alan Padgett), simple temporality (Nicholas Wolterstorff), and qualified temporality (William Lane Craig).

Ganssle, Gregory E., and Woodruff, David M., eds. God and Time: Essays on the Divine Nature. Oxford: Oxford University Press, 2002. A collection of original essays by leading contemporary thinkers on divine eternity.

Kutrovátz, Gábor. "Heat Death in Ancient and Modern Thermodynamics." Open Systems and Information Dynamics 8 (2001): 349-59. Discussion of two solutions to the puzzle posed by the prospect of the heat death of the universe.

Mellor, D. H. Real Time. Cambridge: Cambridge University Press, 1981. Second revised edition: London: Routledge, 1998. A brilliant and influential defense of the tenseless theory of time.

Sider, Theodore. Four-Dimensionalism: An Ontology of Persistence and Time. Oxford: Oxford University Press, 2001. An important defense of eternalism by a prominent exponent.

  • [1]

    A few philosophers have argued that advocates of an A-Theory of time (to be explained in the sequel) cannot coherently hold that temporal entities stand in earlier than/later than relations, since a pure A-theorist believes that past and future entities do not exist; but the vast majority of philosophers have not found this argument convincing. In any case, all disputants acknowledge that temporal entities are uniquely experienced by us in such a way as to be anticipated as future, experienced as present, and remembered as past.

  • [2]

    The recently revived dispute over theological fatalism hinges upon the detractors of divine foreknowledge of future contingents' being able to enunciate a sense of temporal necessity which is (i) applicable to statements about God's beliefs about the future, (ii) closed under entailment, such that the same necessity that characterizes the past is mediated to the foreknown event, and (iii) is incompatible with libertarian freedom. No one has succeeded in crafting such an account. So long as we are capable of acting in such a way that, were we to act in that way, God's foreknowledge would have been different, divine foreknowledge presents no threat to future contingency.

  • [3]

    Freeman J. Dyson, "Time without End: Physics and Biology in an Open Universe," Reviews of Modern Physics 51 (1979): 447.

  • [4]

    See Lawrence M. Krauss and Glenn D. Starkman, "Life, the Universe, and Nothing: Life and Death in an Ever-Expanding Universe," Astrophysical Journal 531 (2000): 220-30.

  • [5]

    See A. Borde and A. Vilenkin, "Eternal Inflation and the Initial Singularity," Physical Review Letters 72 (1994): 3305-08; A. Borde, A. Guth, and A. Vilenkin, "Inflation is Not Past Eternal," (1 Oct 2001): 4.

  • [6]

    See William Lane Craig, Assessing the New Testament Evidence for the Historicity of the Resurrection of Jesus (Lewiston, N.Y.: Edwin Mellen, 1989); N. T. Wright, the Resurrection of the Son of God (London: 5PCK, 2003).

  • [7]

    For an illustrative application of three criteria to rival hypotheses see William Lane Craig and Gerd Lüdemann, Jesus' Resurrection: Fact or Figment?, ed. Paul Copan and Ronald Tacelli (Downer's Grove, Ill.: Inter-Varsity Press, 2000).

  • [8]

    Carlo Rovelli, "What Does Present Days [sic] Physics Tell Us about Time and Space?", lecture presented at the Annual Series Lectures of the Center for Philosophy of Science of the University of Pittsburgh, 17 September 1993.

  • [9]

    Most philosophers who regard God as temporal hold that the beginning of physical time did not mark the beginning of God's time, though a few, troubled by the prospect of God's enduring through an infinite time prior to creation, prefer to regard God as timeless sans the universe, even if He is temporal from the moment of creation, which, for simplicity's sake, is identified with the Big Bang. In either case, however, it is possible for time to exist prior to the Big Bang, so that time is not essentially physical time. Similarly, even if physical time were to end in a cosmic Big Crunch, that would not entail the end of time itself. Indeed, on the Christian view God has bestowed on His children "everlasting life" (Jn. 3.16), so that time will never end.

  • [10]

    See Patrick Toner, "Divine Judgment and the Nature of Time," Faith and Philosophy (forthcoming).

  • [11]

    My thanks to Fred Adams and Milan Cirkovic for fascinating discussion of issues raised in this paper.v