The issue most strongly associated with the history of philosophy in the Islamic world is the eternity of the universe. This is rather surprising. It’s not exactly a pressing topic in philosophical discussions nowadays, after all. Besides, you might expect Muslim, Jewish, and Christian philosophers simply to assume that the universe is not eternal, but rather created. Yet the question of eternity deserves its prominence. It features in several key works from the Islamic world—notably al-Kindı’s On First Philosophy, al-Ghazālı’s Incoherence of the Philosophers, and Maimonides’ Guide for the Perplexed—and relates to several other philosophical problems that are indeed pressing issues for us today, such as time, infinity, modality, and causation.
What does it mean for something to be ‘eternal’? A first approximation might be that something is eternal if it exists at all times. To call something ‘eternal’ in this sense is both an observation about past history and a prediction about the future: an eternal thing has already existed at all previous times, exists now, and will exist at all future times. But there is an ambiguity here. Suppose that time began only 100 years ago. In that case, something one hundred years old that will continue existing indefinitely into the future would be ‘eternal’ in the sense just suggested, since it would exist ‘at all times’. But we might hesitate to call such a thing eternal, and instead insist that genuine eternity requires time to be infinite. With this presupposition in hand, something is eternal only if it has already existed for an infinity of time, exists now, and will continue to exist for an infinity of future time.
Both conceptions of eternity have their problems. To suppose that time began at some point in the past confronts us with the paradox of how time itself can begin. It seems irresistible to think that, beforehand, there would be earlier moments at which time did not yet exist. But this looks like a contradiction: how can there be moments without time? Unfortunately, assuming an infinite time yields an equally unappealing paradox. For we would need to suppose that an infinite past time has already elapsed in order that we may have reached the present moment. But how can an infinity of time elapse? Both puzzles were raised in antiquity. For Aristotle, the paradox to avoid was rather the first one of admitting that time began. Nor did Aristotle think it was possible that motion ever began. For any motion requires another motion to trigger it. In his technical language, the potentiality for any motion must be realized through some previous actual motion. Thus any putative first motion would need a prior motion that causes it.
These two points, concerning the infinity of past time and the impossibility of a first motion, are intimately connected. Aristotle thought that there can be no time without motion. According to his much-discussed definition, time is ‘the number of motion in respect of priority and posteriority’. The impossibility of a first motion gives us a further reason to think that past time is infinite. If there has always been motion, then there has always been time measuring or numbering motion. Many representatives of the Abrahamic faiths would find Aristotle’s eternalism incompatible with the idea of a Creator God. But it’s worth noting that he used his claims about eternity to prove the existence of God. An immaterial cause, he reasoned, would be needed to give rise to the infinite motion required for such an eternal universe. Of course, Aristotle’s God was not a creator, at least not in the sense of making the universe exist after previously not existing. Rather, his God was a first cause of motion, in particular of heavenly motion, which then gives rise to the motions in our lower world below the heavens (see Box 12).
In late antiquity, the Christian John Philoponus argued vigorously against Aristotle’s eternalism. In this context he invoked our second paradox: that an infinite past time cannot finish elapsing in order to reach the present. His anti-eternity works were known in the Arabic world, and were used by al-Kindı and Saadia Gaon. Taking up Philoponus’ polemic, they argued that the physical universe cannot be eternal if, as Aristotle believed, it is finite in size. For Aristotle, the cosmos has an outermost limiting body which is spherical in form. In this outermost sphere are embedded the so-called ‘fixed stars’, while the planets are seated upon lower spheres arranged concentrically around the earth. Al-Kindı and Saadia broadly accepted this Aristotelian description of the cosmos, but turned it against Aristotle himself. Nothing finite can have properties that are infinite, they argued, so a universe of limited size cannot have a power to exist for an unlimited time.
Aristotle’s idea that God directly causes celestial motion, and only indirectly affects our lower world, was taken on in a surprising way in the Islamic world. It became a rationale for the science of astrology, which could now be presented as the study of God’s providential care for the universe. When we study the motions of the stars, we are examining the instruments through which God brings about His cosmic plan. This can already be found in al-KindI and his colleague Abī Maʿshar, a major figure in the history of astrology. By associating the Aristotelian theory with astrology, they departed from the view of earlier exegetes like Alexander of Aphrodisias. Alexander too thought that God influences our world through heavenly motion, but only at the level of species. In other words, heavenly motion providentially ensures that humans, giraffes, and sunflowers are propagated from generation to generation, without being designed to cause particular events involving this or that individual. Yet precisely these sorts of events were at stake in astrological prediction. Philosophical debate would continue in subsequent centuries, for instance among Jews in Andalusia. Here Abraham ibn Ezra was a strong proponent of astrology, who even explained the misfortunes of the Jewish people with reference to the malign influence of Saturn. Similarly Abraham ibn Daud wrote of the stars as the ‘servants of God’s decree’. Not long after, though, Maimonides was highly critical of astrology. In a letter on the topic written to a Jewish community in Provence, Maimonides went so far as to blame the destruction of the Jewish temple on the fact that Jews were distracted by astrological pursuits. He preferred Alexander’s position that heavenly motion gives rise to the continuity of species, rather than individual events.
Ironically, it was precisely on this basis that Aristotle had proved the existence of God. Since the celestial motion is infinite in time, it cannot be explained with reference to the finite cosmos itself, but requires an immaterial cause. The upshot is that there were two, mutually incompatible paths from the finite size of the universe to the existence of God. One could, with Aristotle, say that this finite cosmos exists eternally, and requires God to make it move for an infinite time. Or with Philoponus, al-Kindı, and Saadia, one could say that a finite cosmos cannot be eternal, and therefore must be created—thus proving that there is a Creator. That God’s existence can be proven either on the assumption that the universe is eternal, or on the assumption that is not, did not escape the notice of philosophers in the Islamic world. The point was made by Ibn Ṭufayl, who had his desert island protagonist Ḥayy ibn Yaqẓān use both methods of proof.
Another philosopher of Andalusia, Maimonides, provided a more involved discussion, contending in his Guide for the Perplexed that none of the arguments for or against the eternity of the universe is decisive. He assumed nonetheless that the universe is indeed eternal, for the sake of argument—the argument being one for God’s existence, based closely on the reasoning used by Aristotle. This has occasioned some suspicion that Maimonides was in fact a convinced eternalist, but reluctant to say so. More likely, he saw the same point made by Ibn Ṭufayl, that God’s existence can be proved on either assumption. Admittedly, Maimonides lavished more attention on the path from eternity to God than the one from non-eternity to God. But this was a reasonable strategy, since the latter path looks to be an easier one: if the universe is not eternal but created, then obviously it has a Creator.
Maimonides’ nuanced approach did not please everyone. The Guide’s elaborate Aristotelian argument for God’s existence was subjected to a searching criticism by the later Jewish thinker Ḥasdai Crescas. In the course of his critique, Crescas put forth a number of innovative ideas concerning physics and infinity. (Some of these ideas can also be found in the earlier Abī l-Barakāt al-Baghdādı, but it is unclear whether there is a connection between the two.) Aristotelians had always drawn a distinction between the sort of infinity involved in an unlimited physical magnitude and the one involved in eternal past time. A body cannot be infinitely big because it would be actually infinite. In other words, it would be an unlimited magnitude that is actually present in its entirety. But past time can be eternal, because it is only potentially infinite. This infinity is like the one involved in counting up through the numbers. When we count up, we need never come to an end, since we never reach a highest number. The same applies to the number of years that the universe has already existed.
Philoponus, al-Kindı, and Saadia thought that a past eternity would rather be an actual infinity, because an actually infinite number of moments, hours, or years must already have elapsed. Crescas took a startling and different tack, arguing that actual infinities are not absurd anyway. This is not to say that he actually believed the universe to be infinitely large. Rather, he was out to show that the Aristotelians have failed to rule out the possibility, along with several other hypotheses concerning the physical universe. Other examples included the possibility of an infinite void surrounding a finite cosmos, or of a succession of universes that come into and out of existence. Crescas thus rejected all proofs of God based on Aristotelian physics.
Aristotle’s influential definition of time as a measure or number of motion left the metaphysical status of time somewhat unclear. Time might be objectively existent, as a quantity that depends on moving bodies for its existence. Or it might be subjective, and need somebody to do some actual measuring of motion. (At one point Aristotle implied this, remarking that there would be no time without soul.) Though the subjectivist approach was not unknown in antiquity—Augustine is a famous example—most Aristotelians thought that time is objectively real. Its reality would however depend on the more fundamental reality of bodies, since there can be no motion without bodies. Likewise, it would only be through motion and change that we come to grasp time. Taking on a brief suggestion found in Aristotle himself, later Aristotelians gave celestial motion primacy in their account of time. Since this is the fastest motion, and is regular and unceasing, the temporal measure of this motion can serve as a standard for coordinating and measuring all other motions.
A rival view was set forth by Abī Bakr al-Rāzı, whose notorious theory of ‘five eternals’ makes time one of the fundamental, uncaused principles of the universe along with God, soul, matter, and place. For al-Rāzı, temporal duration is not dependent on motion or bodies, but is a ‘self-subsisting substance’ in its own right. Nor do we need to experience motion to grasp that there is time. Rather, it is something to which we have immediate access. Contesting the rival Aristotelian theory, he invoked the intuition that time would continue passing even if there were no celestial sphere at all, and the fact that someone can be aware of the passage of time without being aware of motion. He also argued that time cannot be created by God, because that creation would have to occur at some time; thus time would need to be present before being brought into existence, which is absurd.
Al-Rāzı called his causally independent temporal duration ‘absolute time’ or ‘eternity’. He contrasted it with the sort of time envisioned by Aristotle, which is at stake when we measure particular motions. Consider the way that, according to pre-Copernican cosmology, the sun revolves around the earth. One such revolution is measured by us as a single day. For al-Rāzı, this day is nothing but a demarcated segment of absolute time, which has always been and always will be elapsing, independent of the motion of the sun or anything else. It provides a background framework against which we can say that one motion or event is before, after, or simultaneous with another. So when al-Rāzı spoke of ‘eternity’, he did not have in mind timelessness, but infinite duration. The notion that something might exist beyond time had, however, already been explored in antiquity. For instance the late ancient Latin Christian author Boethius spoke of God as surveying past, present, and future from a vantage point beyond the passage of time—like someone seeing a whole landscape from a high mountain. Boethius was not known in the Arabic tradition, but he was inspired by earlier Platonist works that were indeed known. Both Plotinus and Proclus ascribed unchanging eternity to the universal Intellect envisioned in their metaphysical hierarchies, while time was assigned to the soul and the physical universe.
Drawing on this tradition, numerous philosophers of the Islamic world described God as being beyond time (for kalām views see Box 13). Al-Kindı gives us our earliest example, albeit only in a passing remark from his On First Philosophy: ‘the First Cause is first in respect of time, since it is the cause of time’. This is rather cryptic, but does express an obvious rationale for ascribing timelessness to God. If time is linked to or dependent on created things—as implied by the Aristotelian view that time is the measure of bodily motion—then time itself comes into existence along with the created universe. So time itself is created, and God must be above time, given that He created it. For Abī Bakr al-Rāzı this argument would fall at the first hurdle, since he did not make time dependent on bodies, motions, or changes, and in fact insisted that God must create anything at a time. For him, creation thus presupposes time rather than bringing time into existence.
An alternative to the Aristotelian view of time was offered by Islamic theologians. A commitment to atomism spanned the Muʿtazilite–Ashʿarite divide, with representatives of both schools affirming that there are smallest, indivisible parts of matter. These theologians also apparently thought of time as being composed of atoms, that is, smallest indivisible units. That allowed them to explain differences in speed, by saying that slower motions include a higher number of imperceptible pauses or ‘rests’ at indivisible times. All of this contradicted the Aristotelian physics accepted by most falāsifa, according to which matter, space, and time are continuous, that is, indefinitely divisible. The kalām authors also differed from the Aristotelians when it came to the definition of time. The Muʿtazilite thinker Abī ʿAlI al-JubbāʾI (d. 915) proposed that time can be thought of as something ‘stipulated’ for the coordination of multiple events. For instance if I say, ‘Zayd will come when the sun rises’, then Zayd’s arrival and the sunrise are being coordinated in virtue of a third thing, namely the moment or time at which both will occur. Though this is completely different from the Aristotelians’ definition of time as a measure of motion, al-JubbāʾI agreed with them that celestial motion has some claim to primacy in our understanding of time. According to al-AshʿarI, al-JubbāʾI stated that ‘moments are the motions of the celestial sphere, because God stipulated them for things’. The example of Zayd and the sunrise would illustrate this: God has providentially assigned time to the regular and predictable celestial motions, and we can coordinate other events with those motions.
Asserting God’s timelessness could help to solve a standard argument against the eternity of the universe. Consider the situation prior to the moment of the universe’s creation. It seems natural to imagine that there was an infinity of time prior to that moment, during which God was not yet creating. So, what was He doing? (This puzzle too was already discussed in antiquity, for instance by Augustine, who was tempted to answer: ‘creating hell for people who ask such impertinent questions’.) A more rigorous way of putting the problem is this: how can God choose the moment for creating the world? All the moments in this infinity of time are equally suitable, so it looks like He would have to choose arbitrarily, or at random. But such arbitrary choice seems unfitting for a perfectly wise Creator. Al-Rāzı found this argument persuasive. One of his reasons for positing Soul as a eternal principle was that unlike God, Soul began in a state of ignorance, and so could unwisely and arbitrarily select a moment for the universe to begin. But by saying that God is timeless, one could avoid the argument entirely. There would be no infinity of moments prior to creation. Rather, moments and time in general would only exist once God has created them. This point was made by al-Ghazālı in his Incoherence, though as we’ll see shortly, he believed that the argument can also be defeated without invoking God’s timelessness.
Of course the idea raises problems too. The gravest difficulty is God’s relationship to created things. How can a timeless God create one thing on Monday and another on Tuesday? How can He know that you are reading this book now, but were not reading it one hundred years ago? One answer was proposed by the Safavid era Iranian thinker Mır Dāmād, one of the teachers of Mullā Ṣadrā. He took inspiration from Avicenna, who had distinguished between time (zamān), perpetuity (dahr), and eternity (sarmad). For Avicenna these words are appropriate to relations between different sorts of entity: ‘time’ in the proper sense applies only to relations between changing things, as one would expect given the Aristotelian definition of time. ‘Eternity’, by contrast, applies to relations between unchanging things. Between them we have ‘perpetuity’, which applies to the relations that unchanging things bear to changing things.
Expanding on this, Mır Dāmād put forward his theory of ‘perpetual creation (ḥudīth dahrı)’. God in Himself remains above time, since He is unchanging. Yet He can bear relations to created, changing things through a creative act which is not ‘eternal’ but ‘perpetual’. Through this act, God brings forth things that were hidden within His eternal power. The creation comes ‘after’ God in the sense that it is causally posterior, but this has nothing to do with time. Temporal beforeness and afterness apply only to the created things themselves, here in our physical universe. So we can think of things as existing in three ways: as eternally hidden when they are ‘with God’, as contained within a perpetual creative act, and as fully manifest or ‘existent’ in time. All this helps to resolve our puzzle. We can now say that the relations God bears to created things actually belong to the perpetual act which is distinct from God Himself, who is timelessly eternal. Mır Dāmād’s account attracted criticism from a philosopher of Mughal India, Maḥmīd Jawnpīrı (d. 1652). He complained that Mır Dāmād was basically imagining that the same thing is created twice, first perpetually in God’s act, and then temporally in the world. If that were the case, then one and the same object would exist ‘before’ itself.
Mır Dāmād’s theory of ‘perpetual creation’ finds a third way between two options that seem mutually exclusive. Intuitively, it seems that the universe must be either created or eternal, in the sense of having no beginning in time. This opposition was presupposed in the early arguments inspired by Philoponus and given by al-Kindı and Saadia. For both of them, asserting that the universe is eternal would immediately imply that God did not create it. Conversely, it would be incoherent to imagine a created but eternal universe. The same opposition was assumed in other early debates, for instance the one over the createdness of the Qurʾān, in which opponents of this thesis were routinely said to believe that the Qurʾān is ‘eternal’. Mır Dāmād was working much later, and no longer took it for granted that something created must have a beginning in time. What changed in the meantime?
The short answer, as so often, is Avicenna. His understanding of causation entailed that something could be caused to exist, despite existing for an infinite time. That may seem pretty obvious. We need only imagine a cause that is producing its effect forever. Al-Ghazālı, trying to illustrate this notion on Avicenna’s behalf, gave the example of a finger that is eternally stirring water. In such a case we would presumably have no hesitation in identifying the finger as the cause and the water’s motion as effect, without demanding that the motion must have begun at some point. But there was a significant philosophical obstacle to such a conception, namely that eternity had long been supposed to imply necessity. In fact, Aristotle argued for this point explicitly in his On the Heavens. The idea is a seductive one. Clearly, impossible things are things that never exist. Why not suppose that, conversely, necessary things are just those things that always exist?
In a sense, Avicenna would agree with this, but he would also say that it is put too simply. He distinguished between two kinds of necessity: that which is ‘necessary in itself’ and that which is ‘necessary through another’. As we know, he conceived of God as the only thing that is necessary through itself. Of course this means that God does not come into existence. He is, rather, eternal and unchanging. From his key tenet that God is necessary in every respect, Avicenna inferred that God is not only necessarily existent in Himself, but necessarily causes all other things to exist. He has only one direct effect, which is an intellect associated with the outermost celestial sphere of the universe. This intellect, and further celestial intellects caused in a chain descending from God, are ‘necessary through another’. In themselves, they would neither exist nor not exist, because they are contingent entities. But they are necessarily made to exist, whether directly or indirectly, by God.
As a result, whenever God exists—which is always—His effects will also exist. We might say that they are ‘eternal through another’: they acquire their necessity from a cause, and acquire their eternal existence along with that necessity. How then could Avicenna account for the fact that some things are not eternal? Since you and I, just as much as the celestial intellects and spheres, are the result of a causal chain proceeding necessarily from God, why don’t we live forever? Avicenna’s answer was that change, including generative and destructive change, is introduced by heavenly motions. Though those motions are eternal, they produce varying effects in the world below. For instance, the sun will at one point be further away from a given object, at another point nearer, so the object will be first colder and then warmer. The accumulation of such effects yields the things we see in the earthly realm, as complex substances are built up out of simpler material constituents. Sadly you and I are such complex earthly substances, and are thus subject to bodily destruction.
Avicenna’s distinction between the necessary-in-itself and necessary-through-another was a conceptual breakthrough. It was gladly used by later thinkers, even those who were highly critical of him. Much of the criticism came from theologians of the Ashʿarite tradition, most prominently al-Ghazālı. This was only to be expected. Where Avicenna made God necessary in all respects, the Ashʿarites always sought to emphasize God’s untrammelled freedom. Originally, this had formed a part of their critique of Muʿtazilism. The Ashʿarites denied that humans can use their natural powers of reason to discern any requirements or limits on God, for instance with respect to morality. If God chose to reward sinners and punish believers, that would be His prerogative; we know that He will do the reverse only because of His revealed promises to humankind.
Furthermore, mutakallimīn from both schools held that God constantly creates all things, sustaining them in existence and also creating their properties (or ‘accidents’) and actions. On the standard kalām view, an apple sitting quietly on a table needs God to keep creating it at every moment; God will also be creating its scent, colour, and so on. These creative acts are subject to God’s will. For the Ashʿarites this meant that absolutely everything in the created world is subject to God’s free choice. This is not just because God could in theory intervene to change things, for instance by miraculously turning the apple into a pear. Rather, God must actively choose at every moment to (re-)create things the way they have been. If we see stability and constancy in the world, it is simply because God has graciously and providentially chosen to create more or less the same things from moment to moment. The theologians connected this to the Qurʾānic statement ‘you will find no variation in [God’s] custom (sunna)’ (17:77). Their position is often referred to as ‘occasionalist’, borrowing a term applied to similar theories put forward by modern European philosophers, such as Malebranche.
Avicenna’s ‘hands off’ God, who causes necessarily and has only one direct effect, could hardly be less like the Ashʿarite God. The clash occupies centre stage in al-Ghazālı’s trenchant refutation of Avicenna, The Incoherence of the Philosophers. In al-Ghazālı’s eyes, Avicenna’s necessitarian beliefs amounted to apostasy from Islam, a crime that is in theory punishable by death. The Incoherence begins with a thorough discussion of the eternity debate, going over several of the arguments we have already considered in this chapter. Since his aim was to show that Avicenna’s case remains unproven, al-Ghazālı concentrated on exposing the weaknesses in the ‘philosophical’ position rather than developing a positive theory of his own. Still, al-Ghazālı’s Ashʿarite sympathies are frequently evident, as he emphasizes God’s unrestricted freedom and power. Consider, for example, his treatment of the argument that if God were to create the world with a first moment, He would have to make an arbitrary choice between equivalent moments. We’ve seen that al-Ghazālı had an easy solution available to him: there is no time prior to creation, and so no moments among which God would be choosing. Yet he chose a different response, insisting that making arbitrary choices is precisely the sort of thing God can do, given that He acts freely and not necessarily.
Behind this point lies a radically different conception of agency from the one adopted by Avicenna. Avicenna too thought that God possesses ‘will’, in the sense that He acts while Himself remaining uncaused. He does not even have a final end or goal that He seeks to achieve, since any final end would be a sort of cause for Him. Of course al-Ghazālı agreed that God’s creative agency has no external cause. What he found missing in Avicenna’s account, though, was the presence of alternative possibilities. If God cannot choose between multiple, genuinely open courses of action, then He is not free and in fact does not even count as a real ‘agent’. He would be more like fire, automatically and necessarily giving rise to warmth. As an example of the sort of freedom he has in mind, al-Ghazālı asks us to imagine someone choosing between two identically appealing dates (fruits, not romantic assignations). It is possible to choose just one arbitrarily, and also rational to do so. In the same way, if God really did need to choose one from an infinity of equally appropriate moments for the creation of the world, there would be no obstacle to His doing so, nor would doing so besmirch His perfect wisdom. God likewise has the freedom to perform miracles, so that the supposedly necessary laws discovered in natural philosophy turn out not to be necessary after all (see Box 14).
Al-Ghazālı turns the tables on ‘the philosophers’ by contending that they, too, must admit that the universe has arbitrary features. This is one of several contexts in which he invokes the analogy between spatial and temporal extension. Just as God could have created the universe five minutes earlier, He could have made the universe five metres wider in diameter. Given the vast size of the universe this would make no practical difference. Thus the actual finite dimensions of the universe, no less than the specific finite duration of the universe, are a matter of arbitrary choice. In the same way, God could have chosen that the heavens rotate in the opposite direction, or selected different locations for the poles of that rotation.
A famous section of al-GhazālI’s Incoherence concerns the possibility of miracles. According to al-GhazālI, this is something the philosophers deny, because of their understanding of causation. For them, a genuine cause necessarily gives rise to its effect. If contact with fire really causes cotton to burn, then the burning is a necessary consequence of that contact. Hence it would be impossible for God to intervene and miraculously save the cotton from burning—presumably not a casually chosen example, since the Qurʾān speaks of God miraculously saving Abraham when he was cast into a fire. Al-GhazālI argued that our assumption of a necessary connection between cause and effect is due to a habitual expectation derived from past experience. We have previously seen fire touch and burn cotton, or things like cotton, and assume that burning will continue to occur in similar cases. There has been extensive debate about what al-GhazālI meant to argue here. Some believe that he sought to uphold the occasionalist position of other Ashʿarites: all causation is exercised by God, so that even when there is no miracle it is God who is active. On this interpretation, when cotton does burn upon contact with a flame it is God who causes the burning, not the fire. A different interpretation holds that al-GhazālI accepted the efficacy of what are sometimes called ‘secondary causes’, in this case the fire. On this reading, he wanted to say that fire does burn things but does not necessitate the burning, since that effect can always be thwarted if God intervenes. Since God has the power to trump any other cause, nothing apart from Him has the power to necessitate. In a sense, God would still be involved in every case of causation, but in non-miraculous cases only tacitly, in that He would need to refrain from intervening. It may well be that al-GhazālI wanted to leave both options open, at least in this context: whether God does everything directly, or allows for secondary causation which He can override, the possibility of miracles is safely established.
This is one of the many arguments found wanting in Averroes’ response to al-Ghazālı, The Incoherence of the Incoherence. Averroes accused al-Ghazālı of tackling difficult issues with merely dialectical arguments, which might confuse readers incapable of reaching demonstrative understanding. Furthermore, al-Ghazālı was equating the views of ‘the philosophers’ with the views of Avicenna, when in fact Avicenna himself had departed from correct Aristotelian doctrine on many points. On this more specific point concerning apparently arbitrary features of the universe, like its exact size, Averroes simply denied al-Ghazālı’s initial premise. The heavens affect our earthly world through their motions, and even a slight change to the physical arrangement of the whole system would yield dramatically different results. By contrast, Averroes’ contemporary Maimonides was impressed with the apparent contingency of the universe as we see it. He gave the example of the number of fixed stars. These are so numerous that one star more or less would presumably make no difference. One cannot definitively rule out that the stars must be exactly as many as they are, for some reason known only to God. But it seems unlikely. The universe is, then, probably the product of contingent choice. From this, we can infer (though not demonstrate) that the universe was most likely created with a first moment in time. For, as the Aristotelians themselves had always claimed, eternity is linked to necessity, and contingency to that which is not eternal.
In a later resumption of these disputes, the Ottoman sultan Meḥmed II invited two scholars to write competing assessments of al-Ghazālı’s Incoherence. The winner, Khojazāda (d. 1488), broadly agreed with al-Ghazālı’s stance. He too thought that Avicenna had strayed from acceptable teaching and the bounds of orthodoxy, and rejected the eternity of the universe. However, he pronounced himself unimpressed by al-Ghazālı’s arguments. For example, he agreed with the Aristotelians that a past eternity would not constitute an actual infinity, but only a potential one. Thus Philoponus’ claim that the present moment cannot have been reached without getting through an actual infinity has no force. Khojazāda proposed a clever improvement to the argument to make it more effective. The flaw in the argument is that past times are never all existent simultaneously, so the eternalist cannot be compelled to accept that past eternity yields an actual infinity. But what of the past times existing in God’s knowledge? Since God knows the entire history of the world, all these times will be simultaneously present in His mind. And this will be an actual infinity. As Khojazāda himself admits, these past times will be only mentally existent, but that is enough to yield the needed absurdity.