In their joint work, Theism, Atheism, and Big Bang Cosmology, Quentin Smith and William Craig debate the consequences of the new cosmology for belief in God. Smith claims that Big Bang cosmology effectively disproves the existence of a Creator. One of his arguments runs like this. According to Big Bang theory, the universe began from a singularity. Such singularities are inherently chaotic and unpredictable. No physical laws connect them to later states; in fact, the form taken by later physical laws, the relative magnitude of the four fundamental forces, for instance, is not determined by the initial singularity and is, in principle therefore, a random outcome. Thus, “God has no basis on which to compute what will emerge from the singularity.” God (if there is a God) could not, then, have had as a purpose that the created universe should contain animate creatures, since a universe in which such creatures could never appear could equally well have developed from the initial singularity (indeed, according to recent “anthropic” arguments, it would have been far more likely to do so). This, therefore, according to Smith, constitutes an argument against the sort of Creator that religious people normally believe in, namely one whose purposes in creating the universe included the bringing to be of the human race./390/

Smith assumes, reasonably enough, that for an action to count as purposive, the agent must have some degree of knowledge of the likely outcome of the action. But if the physical processes involved are such that the agent cannot tell in advance how to act in order to achieve the desired end, then purposive action is effectively blocked. Extending this plausible premise from human agency to the agency of a Creator may not, however, be as simple a matter as Smith is assuming. Does God anticipate the future by extrapolating from a knowledge of the present, as we do? Does God have to rely on the predictability of a particular physical process in order to make use of that process to achieve a divine end? Is God's relationship to past, present, and future sufficiently like ours for these analogies to go through?

This is the topic I want to address, not in the context of early‐universe cosmology where the processes involved in the first primitive stages of the universe's expansion are barely understood, but rather in the context of evolutionary accounts of human origins where the fortuitous character of the outcome has been debated since Darwin first proposed his theory of natural selection a century and a half ago. The role played by chance in that theory seemed from the beginning to call into question the teleological character of the overall evolutionary process. Yet if this be so, how is one to reconcile it with the traditional biblical understanding of human origins? Even if one were to interpret the Genesis account of the Creator's breathing of life into the dust of the earth to mean that God relied on the natural processes of the created world to form the first humans, could these processes really be as contingent in their working as the new theory seems to make them and yet still serve as the agencies of God's purpose in ensuring the appearance of beings made in God's own image?

One possible response is to point to the overall directionality of the evolutionary process, its apparent tendency to lead to greater and greater degrees of complexity, however that elusive term be defined. Even if the multitude of individual mutations and crossings of causal lines that compose evolutionary process were to be inescapably random in nature, the long‐term direction of the process itself could then be relied on to lead to the sort of superior adaptability associated with intelligence. And so evolution could once again be construed as God's instrument in achieving a definite end: the appearance of human beings somewhere in the cosmic vastness. Christians sympathetic to Darwin have been inclined to orthogenetic interpretations of the evolutionary process of this sort in which chance was in the end overcome by something like necessity.

This essay falls into two main parts. First, I want to sketch two opposing trends in the understanding of the evolutionary process, one which represents it as more or less predictable, and the other that stresses its contingency. In recent years, thanks in particular to the writings of Stephen Jay /391/ Gould, the latter trend has attained considerable prominence. In the second part, I will explore the implications of the traditional religious view of the cosmos as the work of a Creator for whom the existence of human beings could be construed as in some appropriate sense fulfilling divine purpose. One way in which the contingency of the evolutionary account of natural selection might be countered from the scientific point of view would be to suppose that God acted in some sort of “special” way to bring about the appearance of humanity. And this has, indeed, been a fairly standard response on the part of Christian writers. But there is another alternative. If the Creator escapes from the limits imposed on the creature by temporality, as indeed the traditional Augustinian account supposes, might not the effect of contingency be blunted since the Creator would no longer be dependent on a knowledge of the present for an anticipation of the future?

Predicting Evolution

A common view of biological evolution is that, given the right environment, it will necessarily occur and in the course of time necessarily give rise to intelligence. Textbook presentations of Darwinian theory often make a progressive process of this sort seem like a simple consequence of natural selection in operation. Heritable variations that favor differential survival of descendants will necessarily tend to spread in the population. There may be additional complications involving geographical isolation, environmental change and the like, but the impression is of a gradual but steady drift towards greater complexity. Organic structures become more complex as new organs develop and old ones find new uses. Intelligence itself, with the enormous advantage it confers in terms of survival and propagation, may then seem an almost inevitable development, if the time‐scale be generous enough.

This “upward and onward” view of the action of evolution finds some support in the text of The Origin of Species itself.

Natural selection acts, as we have seen, exclusively by the presentation and accumulation of variations which are beneficial under the organic and inorganic conditions of life to which each creature is at each successive period exposed. The ultimate result will be that each creature will tend to become more and more improved in relation to its conditions of life. This improvement will, I think, inevitably lead to the gradual advancement of the organization of the greater number of living beings throughout the world.

Elsewhere, it should be said, Darwin sometimes shows himself much less confident about the inevitability of this sort of progress.

One recent strong supporter of the inevitability thesis is Nobel‐prize biochemist, Christian de Duve. In his book, Vital Dust, he argues that, as we have come to understand the complex processes of the living cell, we have been able to give a more and more satisfactory account of the developments that could have led up to the appearance of the first cell. He /392/ maintains that: “Most of the steps involved must have had a very high likelihood of taking place under the prevailing conditions”. The universe, he concludes, “is pregnant with life”:

In this organic cloud, which pervades the universe, life is almost bound to arise, in a molecular form not very different from its form on Earth, wherever physical conditions are similar to those that prevailed on our planet some four billion years ago. This conclusion seems to me inescapable. Those who claim that life is a highly improbable event, possibly unique, have not looked closely enough at the chemical realities underlying the origin of life. Life is either a reproducible, almost commonplace manifestation of matter, given certain conditions, or a miracle. Too many steps are involved to allow for something in between.

And he extends this argument from the development of the first cell to the appearance on earth of intelligence; he “sees at work throughout animal evolution a strong selective pressure favoring the creation of neuronal networks of increasing complexity.” And again: “The drive toward larger brains and, therefore, toward more consciousness, intelligence, and communication ability dominates the animal limb of the tree of life on earth. His conclusion is directed specifically against Stephen Jay Gould: “The history of life on earth allows less leeway to contingency and unpredictability than current fashion claims.”

It was among philosophers, perhaps, that the progressivist view of the development of life found warmest welcome originally, especially among those who regarded evolution as the key to their cosmology and to their philosophy generally. Herbert Spencer formulated a “law” of evolution that would, he believed, hold not only for living things but for the physical world generally. Organic structure, he claimed, tends to become more and more differentiated over time, with new forms of integration constantly appearing. Following Lamarck, he maintained that the use or disuse of an organ could lead to hereditable changes of function. Later philosophers like Lloyd Morgan, Samuel Alexander, and Henri Bergson proposed theories of evolution that departed even more from the Darwinian norm than did Spencer's, while agreeing that evolution is a relatively steady and progressive process.

It is notable that those philosophers who have represented evolution in strongly progressivist terms have as a rule (Spencer would be an obvious exception) seen evolution as God's mode of action in the world. This /393/ conjunction finds its most striking expression, perhaps, in the work of Pierre Teilhard de Chardin. He sought an explanation for the steady “complexification” he found in the fossil record of life, in a “psychic” or “radial” energy that operated directively, unlike the “tangential” energies treated in physics and chemistry. Though he allows for a degree of “groping” along the way, evolution is for him “a grand orthogenesis of everything living toward a higher degree of immanent spontaneity,” “a spiral which springs upwards as it turns. From one zoological layer to another, something is carried over: it grows, jerkily, but ceaselessly and in a constant direction.” So steady, indeed, has the upward curve been in his view that he felt entitled to extend it into the far future to an Omega Point where consciousness will finally be fully realized, a Final Cause in which an explanation will be found for the entire course of evolution that inexorably led in its direction.

Few other evolutionary philosophers were quite so confidently orthogenetic in their understanding of the evolutionary process. But philosophers have been on the whole more likely than biologists to see the operation of evolution in terms of law, of a force analogous to Newtonian gravity that relentlessly alters the composition of the gene‐pool to create more and more complex organisms.

In recent years, the “progressivist” view of evolution is even more likely to be found among physical scientists, especially those who speculate about the existence of intelligent life elsewhere in the cosmos. If one is to devote expensive resources to SETI, as the search for extraterrestrial intelligence has been dubbed, it is important to have at least some sort of estimate of the likelihood of success. Frank Drake, a highly‐regarded radio‐astronomer eager to direct his Arecibo radio‐telescope to the search, formulated in 1961 an “extraterrestrial civilization equation” involving seven factors that should yield such an estimate if the magnitude of each factor could be roughly estimated. The number of extraterrestrial civilizations in our galaxy with both the capacity for, and an interest in, interstellar communication would be given by the product of these factors, two of them intended to specify the likely number of planetary systems in the galaxy, the others dependent on the likelihoods of various outcomes: how likely it would be for a planetary system to be hospitable to life, how likely it would be that life would actually develop in such a case, how likely it would be that such life would evolve into intelligence, how likely it would be that intelligent life would develop the means for interstellar communication, and how long a civilization with these capacities would be likely to last./394/

There is much one could say about the hazards of venturing even the roughest guesses in such a context; undeterred by such hazards, Carl Sagan ventured 10⁶ as a reasonable estimate of the number of potential partners in an intra‐galactic dialogue. My interest here lies in Sagan's understanding of biological evolution (one shared with the great majority of those who favor SETI expectations) that given the “right” planetary environment and enough time, biological evolution will necessarily occur, and that in the course of time this evolution will necessarily progress towards higher and higher levels of intelligence. Darwinian natural selection is taken to operate in a steady way, favoring the spread of hereditable variations that promote differential survival of descendants. Though the potential role of environmental factors and physiological constraints in inhibiting this process is not denied, the tendency is to minimize these other aspects of neo‐Darwinian theory and to regard the central role of natural selection as warranting a confidently progressive outcome, “given sufficient time.” Thus, evolutionary theory becomes a predictive resource, not just an explanation for the radiation of living forms in times past.

The Contingency of Evolution

Those who shaped the “new synthesis” in evolutionary biology over the past half century were never comfortable with the predictive uses of evolutionary theory by exobiologists and others and were flatly opposed to orthogenesis in any shape or form. Ernst Mayr and Theodosius Dobzhansky were among those who expressed their skepticism about this way of understanding evolutionary modes of explanation. The most outspoken critic was, perhaps, George Gaylord Simpson, who, in This View of Life, developed an extended polemic against the assumptions underlying the predictivist account. He emphasized, in particular, the fundamental differences between such non‐historical natural sciences as physics and chemistry and the historical sciences of geology, paleontology, and evolutionary biology. The latter deal with unique events for which the notions of law applicable in physics, on which people like Drake and Sagan draw, simply do not work. The complexity of the interactions between environment and gene‐change is so great that any attempt to abstract “trends” or “tendencies” is bound to fail. “There is direction, but it wavers, and apparently random effects also occur.”

In Chance and Necessity, Jacques Monod celebrated the decisive role of chance in evolution. Since mutations in DNA:

constitute the only possible source of modifications in the genetic text, itself the sole repository of the organism's hereditary structures, it necessarily follows that chance alone is at the source of every innovation, of all creation in the biosphere. Pure chance, absolutely free but blind, at the very root of the stupendous edifice of evolution: this central concept of modern biology is no /395/ longer one among other possible or even conceivable hypotheses. It is today the sole conceivable hypothesis.

Mutations are “chance” events for him in two different senses. First, they represent the convergence of previously unrelated causal chains; second, they are quantum events and hence essentially unpredictable. The ‘course of evolution is thus itself unpredictable in detail. Yet despite the far‐reaching consequences that Monod draws from the primacy of chance in the story of evolution (losing our “necessary place in nature's scheme” condemns us to “a frozen universe of solitude”), he is still willing to allow that evolution follows a “generally progressive course,” that its general direction is “upward,” that an initial commitment in particular groups to a certain kind of behavior “commits the species irrevocably in the direction of a continuous perfecting of the structures and performances this behavior needs for its support.”. So the operation of natural selection seems to restore a fair degree of directionality, and even of progress, to the course of evolution after all.

Stephen Jay Gould takes a much stronger line regarding the contingency of evolutionary change. He will have no truck with “upward courses” or “trends,” or with predictability of even the most modest kind. And his emphasis is not on the randomness of the mutations that afford the material for natural selection nor that of the genetic drift in founder populations. Rather, it is on the lack, in general, of connection between the multiple lines of causality that affect singular historical events, such as changes in the gene‐composition of a population.

In his popular essays, he returns over and again to the flexibility of the evolutionary process that makes it something other than simple selectionist accounts would lead one to expect. In the title essay of Eight Little Piggies, he argues that the pentadactyl limb we share with so many other mammalian species “just happens to be.” It ought not necessarily be taken to testify to some intrinsic adaptive advantage of five, as against some other number, of digits; the earliest tetrapods, in fact, had seven or eight digits. Rather, the number may derive from:

the complex, unrepeatable, and unpredictable events of history. We are trained to think that the “hard science” models of quantification, experimentation, and replication are inherently superior and exclusively canonical, so that any other set of techniques can pale by comparison. But historical science proceeds by reconstructing a set of contingent events, explaining in retrospect what could not have been predicted beforehand. . . . Contingency is rich and fascinating; it embodies an exquisite tension between the power of individuals to modify history and the intelligible limits set by laws of nature. The details of individual and species’ lives are not mere frills, without power to shape the large‐scale course of events, but particulars that can alter entire futures, profoundly and forever./396/

The nature of history and of historical science is the theme around which Wonderful Life, his lively account of the successive and conflicting interpretations of the Cambrian fauna found in the Burgess shale, is organized. He has long been a critic of the gradualism of the traditional Darwinian account of the operation of natural selection, urging instead a “punctuated equilibrium” in which long periods of stasis, when species remain more or less unchanged, are interspersed with moments of relatively sudden speciation. In this ambitious work, he reconstructs the extraordinary original flowering of the major phyla of nearly all modern animal groups within a geologically (and biologically) brief interval of a few million years during the Cambrian period, beginning around 570 million years ago. What excites Gould most about the “Cambrian explosion,” as it has been called, is not just the fact that the phyla appeared over such a relatively brief time nor that no new phyla have appeared since, but that the vast majority of the arthropod “ground‐plans’’ found in the Burgess shale have no modem representatives.

Put in another way, of the twenty‐five or so diverse anatomical designs found in the shale, any one of which could, in Gould's view, have served as ancestor for a distinct phylum, only four survived the Cambrian period and gave rise to the modern animal phyla. It is this decimation of phylum‐candidates, this “lottery” as he terms it, that Gould sees as testimony to the effects of historical contingency. The conventional response, of course, would be that the four surviving phyla were in some way better adapted for changing environmental conditions. Gould regards this as implausible. But even if this were to have been the case, he claims that under a different environmental scenario the list of survivors would have been quite different. And everything that came later would then have taken a quite different direction.

Gould's emphasis on extinctions, particularly the great extinctions of life that marked the end of the Permian period, when up to 96% of marine species died off, and of the Cretaceous, when the dinosaurs vanished, is in some ways reminiscent of the catastrophism that enlivened geological debate two centuries ago. His claim is that in such episodes natural selection of the usual sort would cease to operate; it would in large measure be a matter of luck which among all the existing species would survive to propagate themselves in a depopulated world. Furthermore, the causes of such massive extinctions are a matter of chance, relative to the prior history of the affected populations. And so he concludes:

Since dinosaurs were not moving toward markedly larger brains, and since such a prospect may lie outside the capabilities of reptilian design, we must /397/ assume that consciousness would not have evolved on our planet if a cosmic catastrophe had not claimed the dinosaurs as victims.

The strength of Gould's case lies in his insistence on the importance of the web of necessary conditions in any explanation of a complex historical event, conditions, that is, in whose absence the outcome would have been different, perhaps altogether different. One specific source of contingency to which he often returns is the constraint set on possible adaptive lines of development in a particular population by the availability in some corner of that population, for quite other reasons, of the appropriate anatomical framework for that development. Thus, one obscure group (lungfish/ coelocanth) belonging to the vast domain of fish species in the Devonian period happened to have the sort of skeleton that would permit the development of limbs, thus allowing locomotion on land. Had those species not been present, as they might well not have been, Gould remarks, amphibians could not have invaded the land, which in that event might still be inhabited by insects only.

Few have pushed the theme of contingency as far as Gould has; others have found his emphasis much overdone. He is, of course, right about the overall contingency of the evolutionary path actually followed. But “contingency” is ambiguous in this context: it may refer to the chance character of the particular outcome, or to the unlikelihood of an outcome of this general sort. Accepting the first by no means commits one to the second. The question remains: How does one know what would have happened if life had taken a different fork along the way? Or more exactly: How likely was it that life on land would not have developed if lungfish had not been around at the right time or that consciousness would not have developed if an asteroid had not hit or if climate change had covered Africa in forest three or four million years ago? The massive evidence for parallel evolution of such organs as the eye or of physiologically very similar species ought to give one pause in making such claims of unlikelihood. It seems as though contingency has in many instances been overridden by strong selective advantage.

There appears, then, to be a considerable risk involved in adopting either of the extremes above, the appeal to laws or tendencies that would allow one to assert that life on land or the advent of consciousness on this planet would assuredly have come about anyway, or the emphasis on radical contingency that allows Gould to conclude that Homo sapiens is a “tiny twig on an improbable branch of a contingent limb on a fortunate tree.” “Replay the tape a million times from a Burgess beginning,” he remarks, “and I doubt that anything like Homo sapiens would ever evolve again.”/398/ How can we be so sure either of the inevitability or the improbability of the advent of consciousness?

Most evolutionary biologists and philosophers of biology seem to adopt a middle course somewhere between these extremes, but this still allows for a lot of latitude. Dobzhansky, for example, disagrees with what he regards as an overemphasis on chance on the part of Monod. On the contrary, he remarks: “Viewing evolution of the living world as a whole, from the hypothetical primeval self‐reproducing substance to higher plants, animals, and man, one cannot avoid the recognition that progress or advancement, or rise, or ennoblement, has occurred.” Though chance predominates in mutation and recombination, he goes on, natural selection serves to counterbalance this as an “anti‐chance’’ factor. Thus, though the course of evolution cannot be predicted, “it does not follow that the human species arose by a lucky throw of some evolutionary or celestial dice.” In a recent assessment of the issue, Elliott Sober is more cautious. He is skeptical of the suggestion that the evolutionary process has in the past displayed progress or even direction. Though there may have been directional trends within specific lineages, all that the theory of natural selection allows one to conclude is that such trends are possible. It does not, however, allow one to anticipate them in advance; the multiple sources of contingency exclude this.

What may we conclude from this rapid survey? Macro‐evolution is an irregular process, admitting of breaks, reversals, large‐scale extinctions. Its course can, in principle at least, be explained after the fact, but it cannot be anticipated by us. The last billion years has seen an enormous growth in the variety and number of species. There has been a concomitant growth in the complexity of organisms that (according to some) can be construed as a form of progress; it has, however, proved difficult in practice to find an agreed definition of what “complexity” and “progress” should be taken to /399/ mean in this context. Nevertheless, as the palaeontological and geological records come under closer scrutiny and genetic mechanisms come to be better understood, the fragile character of the causal skein leading up to the first appearance of humans becomes ever more evident.

What are the theological implications of this emphasis on contingency, whether at the evolutionary or the cosmological levels? Belief in a Creator has usually gone hand in hand with a conviction that the human race has a special role to play in the story of the Creation: fashioned in the Creator's image, they are the only creatures so far known to us that are able freely to offer or to deny the Creator their love. Jews, Christians, and Moslems would be at one in supposing that insofar as we can speak of God's plans at all, we can assume that humans have a significant part in at least one corner of them. It would seem to follow, then, that the appearance of the human species would not, as it were, have been left to chance. If it was part of the Creator's purpose that humans should eventually make their entrance on planet Earth after a fifteen‐billion year preparation, can the story of that long prelude be as shot through with contingency as it seems to be? Conversely, if the contingency thesis be accepted, even if not in as radical a form as Gould or Smith propose, does this not cast doubt on the belief that the Creator intended the cosmos to bring forth human beings? And if it does, would it not also call in question the whole notion of an omnipotent Creator whose purposes give meaning to a universe that would otherwise be pointless?

The frank anthropocentrism of the line of inquiry these questions open up runs counter, of course, to the instincts of scientists who sometimes call on a “Copernican principle” to justify their refusal to grant any form of privilege to humans. But Western theology is of its nature anthropocentric; it is concerned centrally with human destiny. When theologians hurdle the eons of evolutionary time in order to concentrate on the relationship between human beings and God, the form their inquiry takes will necessarily appear alien to scientists who look on humans as one node, admittedly a particularly intricate node, in a vast network of living kinds. But if scientists ought to be careful not to rush too rapidly to judgment when their theological colleagues focus on human destiny, theologians have to take seriously what the sciences have to say about how human beings came to be on this planet in the first place.

Contingency and Cosmic Purpose

What is at issue is not whether purpose can be discerned in the evolutionary sequence, allowing inference from this to a Purposer. That would resemble the traditional design argument. From the standpoint of the synthetic theory of evolution, evidence of design, of the work of conscious purpose, is not to be found in evolutionary process. Rather, what we are /400/ asking here is the more oblique question: Might not this process, despite its contingency, still be consonant with the achievement of purpose on the part of a creative Agent? Within the religious traditions of the West, the assumption has always been that human beings play a special role in the creation. Surely their appearance, then, must be more than a cosmic accident from the standpoint of the Creator?

It is true that some Christian theologians have been developing the relatively non‐traditional view that God's efficacy is limited in regard to the achievement of ends within the created order. W. H. Vanstone bases his argument on the nature of loving relationship:

The activity of God in creation must be precarious. It must proceed by no assured programme. Its progress, like every progress of love, must be an angular progress‐in which each step is a precarious step into the unknown. . . . If the creation is the work of love, then its shape cannot be predetermined by the Creator, nor its triumph foreknown.

Keith Ward offers a nuanced argument, drawn in part from metaphysics, in part from biblical theology. “God's omniscience is the capacity to know everything that becomes actual, whenever it does so.” God's knowledge of the future is thus of the possible only and changes as the set of actual existents changes. “What is possible is determined by God, but what becomes actual may be determined by creature.” Ward challenges the traditional view of God as atemporal: “That which is wholly necessary can only produce that which is necessary. A contingent universe can only be accounted for if one makes free creativity a characteristic of the First Mover, which entails placing change and contingency within the First Mover itself.” A Creator whose knowledge of the future is only of the possible could presumably see his purposes defeated by the contingency of the creation itself.

John Polkinghorne agrees with Ward in holding that God is temporal. But he maintains that limitations on God's knowledge and power are freely chosen on God's part, a kenosis, or emptying, freely undergone, akin to the kenosis that Christians already find in the doctrine of the Incarnation: “God has permitted a kenosis of his omniscience, parallel to the kenosis of his omnipotence. Even he does not know the unformed future, and that is no /401/ imperfection in the divine nature, for that future is not yet there to be known.”

The vulnerability that these authors, and others who hold similar views, speak of bears to a large extent on God's relationship with creatures who can exercise free choice. They do not ask whether the existence of such creatures was in the first place defeasible. Ward remarks that God causally determines “those outcomes which are necessary to the fulfillment of the dynamic purpose.” Presumably the advent of human beings would qualify as one of those outcomes. But how does God causally determine the outcome of evolution? This brings us right back to the question from which we began. Those who maintain that God's purposes are limited by God's own nature or the nature of God's relationship with the world might still be persuaded to allow that the advent of human beings was something God knew from the beginning would happen. We would not have to suppose, then, that the Creator uttered a metaphorical sigh of relief when human beings finally made it!

That still leaves us groping for a means by which the real contingency of the evolutionary process might be overcome by the Creator. Several possibilities suggest themselves. The most obvious, perhaps, might be introduced by recalling an objection posed by Simplicio, the Aristotelian, to Salviati, Galileo's spokesman, in the Dialogue Concerning Two Chief World Systems. If Copernicus was right about the earth's motion round the sun, a parallax shift ought to be noticeable in the relative positions of the stars. Yet none is seen. The alternative is that the stars are at an enormous distance from us. But then, Simplicio asks, to what purpose are these great spaces? Are they not “superfluous and vain”? To which Salviati replies that God may well have other plans in mind besides the care of the human race. And, in any event, “it is brash for our feebleness to attempt to judge the reasons for God's actions.” Good advice still!

But suppose we were to put this objection again today. Our universe, as we now know, is far, far greater than Copernicus could ever have dreamed; space and time stretch out to the limits of human imagination. Does this not strengthen Simplicio's objection? Perhaps not. Might it not be said that such great spaces, populated by billions of galaxies that have developed /402/ over billions of years, may have been needed in order that in a natural way the cosmos might give birth somewhere within it to human life, one or maybe a multiplicity of times? The contingency of the single evolutionary line might thus be overcome by the immensity of the cosmic scale. Evolutionary biologists are divided, as we have seen, as to whether, on general evolutionary grounds, life of a broadly human type would be bound to originate somewhere in all those myriad planetary systems. But assuming for the moment a positive answer to this question, the enormous space of evolutionary possibilities would then make it possible to maintain that there could be a cosmic purpose at work here on the part of a Creator, a purpose that the contingency of a particular evolutionary line like ours would not defeat. If God be conceived as a time‐bound Creator whose knowledge of the future depends on a knowledge of the present, this way of swamping contingency in order to achieve a distant end might seem plausible. Such a universe could, of course, give rise not just to one evolutionary line but to many, something that those Christians who emphasize the uniqueness of Christ's incarnation on earth might find unwelcome. More seriously, it could, in principle, give rise to none at all, or so it would seem. Could a time‐bound Creator rely on this as a way of achieving purpose?

Before we look for other ways than this that the Creator might relate to contingency in the evolutionary process, it may be helpful to look in somewhat more detail than we have done so far at the main sources of contingency that writers on evolution have claimed. The first of these is the interweaving of independent causal lines, the source of “chance” (tychē), as Aristotle defines it in Physics II, the sort that defeats teleology, in his definition of that term. (Pig eats acorn, which never, in consequence, develops into an oak.) These causal lines may be separately deterministic, so that a comprehensive knowledge of the situation would allow its possessor to anticipate the “chance” outcome. (Meteor hits earth, altering the course of evolution.) It is lack of knowledge that leads us to regard such an outcome as chance. (For Aristotle, it would, rather, be the lack of teleological explanation.) Were the universe to be entirely deterministic, a time‐bound Creator, knowing only past and present, could still anticipate the future with absolute assurance. When Gould points to catastrophe as a major, perhaps the major, source of evolutionary contingency, it is chance in the Aristotelian sense that he has in mind, that is, the crossing of causal lines whose intrinsic teleologies are unrelated to one another. But some sort of causal indeterminism might also intrude.

What kind of indeterminism? The most obvious candidates are quantum indeterminism and the indeterminateness (not indeterminism, strictly speaking) associated with human choice. The latter can be left aside as irrelevant to the process leading to the first appearance of humanity. Peter van Inwagen points to the challenge of physical indeterminism:

Since the actual physical world seems, in fact, to be indeterministic, it is plausible to suppose that there are a great many states of affairs that are not part of God's plan and which, moreover, cannot be traced to the free decisions /403/ of created beings. I very much doubt that when the universe was (say) seconds old, it was then physically inevitable that the earth, or even the Milky Way Galaxy, should exist. Thus, these objects, so important from the human point of view, are no part of God's plan – or at least not unless their creation was due to God's miraculous intervention into the course of development of the physical world at a relatively late stage.

In his view, then, for something to be part of God's plan, it would have to be either “physically inevitable” or else the product of a “special” action on God's part. Given the contingency of the evolutionary record, van Inwagen concludes: “I see no reason as a theist, or as a Christian, to believe that the existence of human beings [i.e., our terrestrial human race] is part of God's plan.”

If physical inevitability is lacking, only a “special” action of some sort on God's part would ensure the realization of long‐term purpose. (We are assuming, remember, that God is time‐bound.) Christians have a name for such special action: ‘‘miracle.” However, those who call on miracle in the context of explaining human origins are no admirers of the theory of evolution as a rule. The defenders of so‐called “creation science” see this theory as a rival to their own preferred account of cosmic origins, which is derived from a more or less literal reading of the first chapters of Genesis. A more nuanced view defended by some Christian philosophers of religion is prompted not so much by the Genesis story, literally or partly literally taken, as by an intense irritation with evolutionists like Richard Dawkins whose dogmatic‐seeming confidence in the all‐sufficiency of evolutionary explanation ultimately depends (they charge) on presupposing an atheism that would make evolution the only possible explanation of origins.

Alvin Plantinga, the main proponent of this view, argues for the insufficiency of current evolutionary theory to account for various stages in /404/ the development of life, beginning from the appearance of the first life and progressing through the appearance in the Cambrian period of the main phyla. From this, he concludes the greater likelihood—from the Christian standpoint (as the scientific evidence currently stands)—of a special creation of a straightforwardly miraculous sort to account for these transitions. Since he is prepared to invoke miracle in these gaps in the orthodox evolutionary reconstruction of the history of life, he would presumably be willing to deal in the same way with the factor of contingency in those parts of the evolutionary account that he accepts. Since he holds God to be a temporal being, the reconciliation of contingency at the microlevel with divine purpose is a topic he has then to address. The solution, miracle, has the merit of simplicity. My hesitations in its regard I have laid out elsewhere.

“Special action” is, however, a capacious category, and some scientist‐theologians have recently been exploring its limits. Arthur Peacocke and John Polkinghorne are in agreement that invoking miracle to accomplish what might be called God's “basic” purposes in nature is unacceptable. They agree, further, that a deterministic universe of the sort described by Newton's physics would leave no room for any sort of “special” action on God's part, miracle aside. But because Newtonian determinism has been multiply challenged by twentieth‐century physics (the candidates here are quantum mechanics, non‐linear dynamics, and chaos theory), there are now points of entrance, as it were, for the Creator's “special” action, without in any way compromising the integrity of the laws that the Creator has implanted in nature.

Some such action on God's part is called for because of another consequence of this sort of cosmic indeterminism. Since God is a temporal being (another point of agreement between the protagonists of this line of argument), such “looseness” in cosmic process not only severely limits God's knowledge of the future but also threatens the realization of the divine purposes. However, if God can somehow “influence” cosmic process within the limits set by the natural indeterminism of the physical order, the ends of Providence may still be achieved. When it comes to the /405/ detail of how exactly this is to be done, the two authors diverge, as they struggle to formulate the elusive notion of an “influence” that is something other than an “intervention.”

Each sets great store by the model offered by “top‐down’’ causality, the action of the whole upon its parts. Polkinghorne looks to chaotic behavior for inspiration in that regard favoring Prigogine's controversial suggestion that such behavior may be at bottom indeterministic. For God to influence such a system, leading it to take one among a number of energy‐equivalent paths, would be to communicate “information” in a “top‐down” way, he suggests, but not to affect the energy of the system, and hence not to “intervene,” or overrule the laws of nature. God's action would thus be permanently “cloaked from view.” Earlier proponents of the theological implications of the breakdown of Newtonian determinism, like Arthur Compton and William Pollard, had relied on quantum indeterminism as their mainstay. Polkinghorne is skeptical: “It does not seem that the proponents of divine action through quantum events have been able to articulate a clear account of how this could actually be conceived as the effective locus of providential interaction.”

Polkinghorne's ingenious deployment of chaos theory in the interests of Christian theology faces a number of challenges. For example, is chaos theory really indeterministic? Can “information” be imparted without exchange of energy? I do not propose to deal with these here. But Polkinghorne's exclusion of quantum indeterminism from the scope of his solution is significant for the issue of evolutionary contingency. This latter is usually held to be due, to a significant extent, to quantum contingency at the level of mutation and possibly recombination. If the “special” action envisaged by Polkinghorne does not bear on this sort of process, it appears that it may not be of much help in explaining how the Creator's purpose might still prevail at the micro‐level. /406/

The line of argument followed by Peacocke is quite different. Though he emphasizes the importance of the breakdown of Newtonian assumptions about predictability and determinism to the issue of God's action in the world, he views these not as facilitating “special” action on God's part but as presenting, if anything, a barrier to it. If the future behavior of a physical system is in principle unpredictable, then God, a temporal being, cannot know how it will behave. Its “manipulation by God would then be impossible.” God can only will what God can foresee. (Peacocke excludes the alternative that God might simply intervene to make the system go one “open” way rather than another; this would still in his view be to overrule nature.) How, then, in a universe where unpredictability looms everywhere,are the divine purposes to be accomplished?

To this, he offers two quite different answers. First is that the Creator has implanted in the world certain propensities, notably the propensity for increase in complexity that is continually manifested in the evolutionary process:

So that, although we have had to infer that God cannot predict in detail the outcome of in‐principle unpredictable situations, this does not derogate from his having purposes which are being implemented through the inbuilt propensities that load, as it were, the dice the throws of which shape the course of natural events.

One would want to know more about how those propensities “load the dice,” that is, overcome the contingency that to others seem such a challenge.

But there is a second line of defense. Drawing heavily on the notion of top‐down causality, Peacocke suggests that God acts upon the “world‐as‐a whole” which in turn constrains what happens at all lower physical levels, including that of quantum interaction. The exemplar he most relies on here is the causal relation of brain/mind and the bodily states it directs, including the myriad quantum processes that power organic life. Since the world‐as‐a‐whole “may be regarded as in God, though ontologically /407/ distinct from God,” God may be presumed to interact with it in its totality. Thus:

God, by affecting the state of the world‐as‐a‐whole, could, on the model of whole‐part constraint relationships in complex systems, be envisaged as able to exercise constraints upon events in the myriad sub‐levels of existence that constitute that “world” without abrogating the laws and regularities that specifically pertain to them—and this without “intervening” within the unpredictabilities we have noted.

Peacocke has, at this point, two daunting hurdles to clear, corresponding to the two stages of his top‐down descent. First, how is God supposed to “interact” with the world‐as‐a‐whole? The world is not God's body, as Peacocke agrees. But if it is not, how is the top‐down metaphor to work, exactly? If God and the world are, as he says, ontologically distinct, does not interaction reduce to the forbidden sort of intervention? Second, is it really the case that the world‐as‐a‐whole has a top‐down relationship to everything below it, after the model of brain and body? If, as Peacocke emphasizes, quantum indeterminism is an absolute barrier to God's knowledge of the future, how exactly does God's action on the world‐as‐a‐whole succeed in overcoming that barrier in order to achieve the divine purpose?

Atemporality and Teleology

It will not have escaped the reader that the authors we have been discussing hold in common the view that the Creator is a temporal being, facing something, at least, of the limitations and vicissitudes that beset the creature. Nor will the reader have missed the fact that it is in consequence of this limitation that the contingency of the evolutionary process becomes such a challenge to the strongly‐held religious belief that this process fulfilled the Creator's purpose when it gave rise to rational free creatures. This leads to a fairly obvious question: Might there not be something awry about the starting‐point? What if we were to suppose that God is not, after all, a temporal being?

This would not be a radical departure. The belief that the Creator stands outside temporal process entirely has, indeed, been the dominant one within the Christian tradition from Augustine's day onwards. It is true that /408/ it has been strongly challenged in recent times, not least by scientist‐theologians who seek a closer affinity between Creator and creature than the traditional account admits. First, then, a quick, and necessarily superficial outline of this view.

Augustine saw God not as a Demiurge shaping an independently existing matter nor as a First Mover responsible for the motions of a world whose natures were not of the Mover's fashioning, but as a Creator in the fullest sense, a Being from whom the existence of all things derives. Such a Being cannot operate under constraints, as the God of Greek philosophers did. Temporality is the first and most obvious constraint of the created world, a mark of its dependent status. A temporal being exists only in the present moment, without secure access either to its past or its future. Its past is no longer; its future is not yet. So even though both past and future are somehow constitutive of what “it” is, in a real sense, they do not in the temporal sense exist. Such a being is evidently lacking, incomplete.

The Creator on whom the universe depends for its existence cannot be limited in this way. Time is a condition of the creature, an evident sign of dependence. It is created with the creature; by bringing a changing world to be, God brings time, the condition of change, to be. The act of creation is a single one, in which what is past, present, or future from the perspective of the creature issues as a single whole from the Creator. God is not part of the temporal sequence that the act of creation brings to be; God is not one more temporal thing among other things. The Creator is “outside” time created, though the metaphor is an imperfect one. Calling God “eternal” is not a way of saying that God is without beginning or end, like Aristotle's universe. “Eternal,” as applied to God, does not mean unending duration; it means that temporal notions simply do not apply to the Creator as /409/ Creator. “Atemporal” might be less misleading. Nor is God static or frozen, as nineteenth‐century critics charged. In a famous formula, Boethius expressed the doctrine in lapidary terms: “Eternity is the whole, simultaneous, and perfect possession of boundless life.” God's life transcends the sort of dispersal that is the first characteristic of the creature; it is not subject to the kind of division that time‐marking would require.

Creation and conservation blend together in this view, as do transcendence and immanence. Creation was not just a moment of cosmic origination a long time in the past, though we often speak of it that way, since the first moment seems to call in a special way for a transcendent cause. Creation continues at every moment, and each moment has the same relation of dependency on the Creator. God transcends the world; the Divine Being in no way depends on the world for existence nor requires it as complement. Yet the Creator is also immanent in every existent at every moment, sustaining it in being. God knows the world in the act of creating it, and thus knows the cosmic past, present, and future in a single unmediated grasp. God knows the past and the future of each creature, not by memory or by foretelling, then, as a creature might, but in the same direct way that God knows the creature's present. When we speak of God's “foreknowledge,” the temporal “fore” has reference to our created reference‐frame, within which the distinctions between past, present, and future are real. From God's side, however, there is only knowledge, the knowledge proper to a maker who is not bound by these distinctions.

Aquinas enters a formal defense of the thesis that God knows future contingent things. Such things are contingent relative to their antecedent physical causes, which is why temporal creatures like us, whose assessments of the future depend on a knowledge of such prior causes, can only conjecture about their contingent outcomes. But God knows these outcomes directly in their presentness as their Creator; the act of bringing them to be has no temporal divisions within it. Some of the analogies Aquinas draws on here need careful construal: “He who sees the whole road from a height sees at once all those traveling on it” (a.13, ad 3); “His gaze is carried from eternity over all things as they are in their presentness” (a.13, c.). Such analogies might suggest that our inability to predict a contingent outcome is simply due to our lack of proper vantage‐point: The various events taking place at this moment on the road just happen to be out of our sight. This in turn might be taken to imply that the future is already set, that it is only our powers of knowing that are unequal to the task of grasping it. But contingency is real, as Aquinas elsewhere makes clear. God knows contingent things that are future to us not as a viewer would /410/ [see] features of a landscape already determinate but rather as a maker might, a unique sort of a maker who respects contingency in the cross‐causal connections between the things made.

This is familiar, of course. It is all very conceptual, as philosophers’ talk of God inevitably is. It is no more than an exploration of an initial postulate concerning the act of creation, when that act is understood as a bringing into existence and a holding in existence, both entirely outside the range of our experience. How would such an account be supported? How does it meet the two major objections that Augustine already anticipated: Can this way of construing the work of creation be made compatible with the reality of human freedom? Does it not saddle the Creator with responsibility for all of the manifest evils of cosmic history? I am leaving these familiar and troubling questions aside in order to focus on a limited but perhaps more tractable issue: How does the apparent defeasibility of the evolutionary line leading to the emergence of Homo sapiens fit with the view that the act of creation is a single atemporal action on the part of God?

What I want to argue is that both Christian evolutionists who have assumed that the purposes of the Creator can be realized only through lawlike, and more or less predictable, processes as well as those who on the contrary infer from the contingency of the evolutionary process to the lack of purpose and meaning in the universe generally, are mistaken from the perspective of the traditional doctrine of God's atemporality. Our notions of teleology, of purpose, of plan, are conditioned by the temporality of our world, in which plans gradually unfold and processes regularly come to term. In such a world, purpose depends on foreknowledge, and foreknowledge in turn depends on the predictability of the processes involved. Lacking such predictability, there cannot be reliable foreknowledge, and without foreknowledge purpose is ineffective. But a Creator who brings everything to be in a single action from which the entirety of temporal process issues, does not rely on the regularity of process to know the future condition of the creature or to attain ends.

The notion of “purpose” must itself be reinterpreted in such a case. God's knowledge of how a situation will develop over time is not discursive; God does not infer from a prior knowledge of how situations of the sort ordinarily work out. It makes no difference, therefore, whether the appearance of Homo sapiens is the inevitable result of a steady process of complexification stretching over billions of years, or whether, on the contrary, it comes about through a series of coincidences that would have made it entirely unpredictable from the (causal) human standpoint. Either /411/ way, the outcome is of God's making, and from the biblical standpoint may appear as part of God's plan.

Terms like “plan” and “purpose” obviously shift meaning when the element of time is absent. For God to plan is for the outcome to occur. There is no interval between decision and completion. Thus the character of the process that, from our perspective, separates initiation and accomplishment is of no relevance to whether or not a plan or purpose on the part of the Creator is involved. Reference to “cosmic purpose” in the evolutionary context need not, however, involve design in the traditional sense. That is, it does not point to features of the process or the outcome that specifically demonstrate the intervention of mind and that would, therefore, allow one to infer to the agency of a cosmic Planner. There is nothing about the evolutionary process in itself that would lead one to recognize in it the deliberate action of such a Planner. It does not look like the kind of process human designers would use to accomplish their ends. When critics of the Christian understanding of cosmic history conclude, in consequence, that we live in a universe lacking in purpose, they are pointing to the lack of independently recognizable design in evolutionary change. But the Creator may not be a designer in this time‐bound sense.

The type of contingency we have been discussing bears on the outcome of a process, the evolutionary process, so it involves time in an essential way. But contingency can take other forms, and at least one of these does not involve time directly. Discussion of the so‐called anthropic principle begin from the claim that the application of quantum theory to the first moments of the cosmic expansion postulated in the Big Bang cosmological model shows how extraordinarily contingent the initial conditions were that permitted a universe in which heavy elements, planets, and ultimately complex life, could develop. It has been assumed in cosmological discussions from Descartes’ time onwards that no special setting of the initial conditions would be needed in order that a universe of the kind we know should come to be. The shock of discovering that this was apparently not the case led to various attempts in the 1970s to explain how such an “unlikely” universe might have originated. /412/

Two types of “anthropic” explanation were proposed. Collins and Hawking suggested that if there were a vast number of independent universes, we would naturally find ourselves in one where human life could evolve, so there would be nothing surprising about the fact that the universe seems “fine‐tuned” for life. A very different sort of explanation is theological in inspiration: The universe is the work of a Creator whose purposes in creating include humanity. God simply chooses the initial energy‐density and the laws of force that would allow that purpose to be realized. In the evolutionary context that we have been discussing, contingency has been seen as a challenge to the possibility of cosmic purpose. In the cosmological case, on the contrary, contingency is regarded as a sign of cosmic purpose.

The attribution of cosmic purpose in the two cases must be carefully distinguished. Proponents of the anthropic argument claim to find recognizable signs of design in the initial cosmic configuration, so that the cosmological evidence is urged as independent evidence of cosmic purpose and even of the existence of a Designer. Whereas in the evolutionary context, the claim that the advent of human life displays purpose, despite the contingency of the process leading to it, is based not on the scientific evidence but on considerations that are either theological or metaphysical in nature.

Yet there is an important similarity between the two contexts also. In the anthropic case, the Creator achieves purpose by choosing the kind of universe whose laws will permit the accomplishment of that purpose in a fully “natural” way. In the evolutionary case, something similar is true, though here God is choosing not between different kinds of universes, universes with different laws, but between different instantiations of the /413/ same laws. It will make a difference here whether the universe is regarded as deterministic or not. If it is, then it will be a matter of God's choosing the appropriate initial cosmic conditions to bring about eventually the appropriate outcome of the particular contingent evolutionary process. If the universe is indeterministic, God will choose among the range of possible outcomes of that causal component of the particular evolutionary process. This can be done without violating the probability distribution among these outcomes over the long run. Obviously, the indeterministic alternative is much the simpler to envision.

The difference between this sort of action on God's part and the sorts of “special action” hypothesized by those who, for other reasons, prefer to suppose the Creator to be dependent on knowledge of the present for planning a future not otherwise accessible is that God does not have to “influence” a process that in the former scenario is already in progress. Rather, God's action is simply the timeless one of choice, in which the universe comes whole from the Creator's hands. There is no need for a top‐down causality akin to that of mind on body or of the whole upon the part in chaotic process.

We can conclude, then, on the basis of the traditional doctrine of God's atemporality that the contingency or otherwise of the evolutionary sequence does not bear on whether the created universe embodies purpose or not. Asserting the reality of cosmic purpose in this context takes for granted that we already believe that the universe depends for its existence on an omniscient Creator whose action is sufficiently like ours to allow us to call it purposive, in an admittedly analogical sense. It does not mean that we are privy to that purpose, though the traditions of the Torah, the Bible, and the Koran would imply a recognition of at least a part of it. Only to the extent that such a prior recognition were possible could one allow cosmic purpose to constitute a special form of teleology (recalling that “teleology” refers to a specific mode of explanation). When in the Confessions Augustine looks back over his life and finally recognizes a Providence at work through all the contingency, it is to teleology of this sort that he is appealing.

Linking plan to Providence in this way gives rise to many other questions, of course. One would need, in particular, to be allowed to distinguish between God's intending and permitting something to occur. But the answers to those questions, important—indeed crucial—though /414/ they are, do not affect the contention of this essay: If one maintains the age‐old doctrine of God's atemporality, the contingency of the evolutionary process leading to the appearance of Homo sapiens makes no difference to the Christian belief in a special destiny for humankind.