1. Aristotles’ Cosmos

Aristotles’ view of the universe is a natural focal point for discussions of ancient cosmologies, whether these belong chronologically before or after Aristotle. (To speak of non-Aristotelian cosmology before Aristotle is of course anachronistic, but it nonetheless makes good sense.) The long-term significance of Aristotle’s picture of the universe can hardly be overrated, not least because the major part of it became integrated in the natural philosophy of the middle ages, where it obtained a nearly doctrinal status. In fact, in Greek antiquity Aristotelian cosmology held considerably less authority that in did in the medieval and renaissance eras. At any rate, no account of Greek-Roman cosmology can avoid relating to the issues discussed so influentially by Aristotle. The books of relevance for cosmology are principally On the Heavens and parts of Physics, and secondarily Meteorology, On Generation and Corruption, and book 12 of Metaphysics (Barnes 1984). A further work, known as On the Universe, is relevant as well, but it is generally believed to be pseudo-Aristotelian, that is, written by a later author. Aristotle’s main works are often known by their Latin titles, in this case De Caelo, Physica, De Meteorologica, De Generatione et Corruptione, Metaphysica Lambda, and De Mundo.

So, what were the defining features of the Aristotelian universe? Briefly put, it can be described as a two-region universe in a steady state. According to Aristotle, the universe at large consisted of two essentially different realms, the sublunar and the superlunar world. The first region, covering the Earth and the air up to the Moon, was composed of bodies made up of the four terrestrial elements with their rectilinear motions, either towards the centre of the Earth (earth, water) or away from it (air, fire). Beyond the Moon, the bodies moved naturally in eternal, uniform circular motions. The stars, planets and celestial spheres were composed of an entirely different kind of matter, an ethereal semi - divine substance or fifth element, what Aristotle called the aither but is better known by the later Latin name quinta essentia. The aither resided exclusively above the Moon and its natural motion was, contrary to that of the terrestrial elements, circular. In the Meteorology Aristotle said that the purity of the aither increased in proportion to its distance from the sublunar world. Whether in the sublunar or superlunar region, a void could not possibly exist, and hence the universe was a plenum.

Aristotle’s cosmos was in a steady state in so far that it was eternal and local changes were restricted to the sublunar world. He argued that the universe as a whole was ungenerated as well as indestructible. Among several arguments against cosmic creation, Aristotle referred to what is known as the ‚why not sooner?' argument: If the universe came into being a finite time ago, what reason could there possibly be for just this time rather than some other time? He also denied that the universe (or any other body) could be created out of nothing, because that would require a pre-existing void, which he considered an impossible notion. A spatially infinite world was another impossibility, for by its very nature the world – meaning the heavens – revolved in a circle, and Aristotle pointed out that such motion was impossible as it would lead to an infinite velocity. What was enclosed by the outermost sphere comprised everything. In summary, Aristotle maintained that the universe was unique, eternal, and all inclusive:

It was this conception of the universe that was incorporated into the medieval world view, except for the controversial and most un-Christian claim of the universe being past eternal.

2. Cyclic Conceptions

Ideas of cosmic cycles were well known in ancient Greece, both before and after Aristotle. Empedocles from Acragas in Sicily famously postulated four basic elements (earth, air, water, fire) which permanently kept their character and the arrangements and rearrangements of which accounted for what appeared to be generation and destruction in nature. He described the cosmos as a self-contained sphere passing through cycles of rest and change, with recurring stages of conflict which produced forms of life. In a characteristic vitalistic terminology Empedocles called the polar opposing forces or principles for philia ('Love') and neikos ('Strife'). The changes between dominance by Love and Strife proceeded eternally, corresponding to continual creations and destructions of the world. However, the two forces were not simply creative and destructive, for the conditions of life demanded a certain balance between them. When Love dominated, the elements were mixed up into a uniform mass, while at the time of Strife’s complete dominance they were fully separated from one another and arranged in concentric spheres. Only in between the two extremes was the universe hospitable to processes generating life.

Empedocles’ cycles were symmetric, so that the events in one phase were repeated in the opposite phase, but in reverse time order. Thus a process from birth to death would be followed by one from death to birth. Without requiring an identical repetition, Empedocles posited a cosmos without beginning or end. In one of his enigmatic fragments he speaks of a 'double birth' and a 'double passing away,' for 'the unity of all things brings one generation into being and destroys it, and the other is reared and scattered as they [the elements] are again being divided' (Wright 1995, p. 142; O’Brien 1969). The periods of the cosmic cycles were said to be very long, but Empedocles did not specify their length.

The later Stoic philosophers adopted the idea of temporally multiple universes, which they associated with thermal phenomena. What Empedocles poetically had named Love and Strife was now conceived more naturalistically or mechanically, namely as condensation and rarefaction. In what has been called "a first tentative approach to the conception of thermodynamic processes in the inorganic world" the Stoics assigned a dynamic role to fire in all areas of natural phenomena, indeed to the cosmos itself (Sambursky 1959, 1963, p. 133). Fire was the agent that caused change and decay in the universe, eventually to lead to its conflagration – evidently a most un- Aristotelian notion. However, the conflagration would not be the absolute end of the universe, for it was thought to be reborn and from the primeval fire to return identically to its former state. In a lost book known as On the Cosmos, Chrysippus from Cicilia is to have said that "after the conflagration of the cosmos everything will again come to be in numerical order, until every specific quality too will return to its original state, just as it was before and came to be in that cosmos" (Sambursky 1963, p. 202).

Assuming the physical world to be placed in an infinite non-physical void, the Stoics conceived the cosmos as a gigantic sphere oscillating through cycles of expansion and contraction in the void surrounding it. The agent responsible for the cyclic changes was ultimately the fire element. According to a Stoic source, "the material world preserves itself by an immense force, alternately contracting and expanding into a void following its physical transmutations, at one time consumed by fire, at another beginning again the creation of the cosmos" (Sambursky 1963, p. 203). For the idea of cosmic conflagration Zeno of Citium and later Stoic philosophers used the term ekpyrosis, meaning 'out of fire.' According to Plutarch, "When ekpyrosis takes place, [Chrysippus] says that the universe is totally alive and is a living being, but thereafter, as it is quenched and becomes concentrated, it turns into water and earth and things substantial" (Lapidge 1978, p. 183). Another Roman author, Marcus Tullius Cicero, adopted a version of the Stoic universe, such as appears from his treatise On the Nature of the Gods. We Stoics, he said, conclude that

It is tempting to think of the ekpyrotic state as a violent conflagration, perhaps a kind of 'big crunch' in which all matter collapses and is turned into hot radiation energy. But the Greek texts mostly describe the decay process as a very slow, almost imperceptible combustion. Some of them liken the conflagration of the world to the gradual transformation of swamps into dry grounds.

More than two thousand years later, the name 'ekpyrosis' reappeared in cosmology, now in a cosmological model based on concepts of string theory proposed by Paul Steinhardt, Neil Turok, Bert Ovrut and Justin Khoury in 2001. As they explained in their paper in Physical Review, "We refer to our proposal as the ‘ekpyrotic universe,’ a term drawn from the Stoic model of cosmic evolution in which the universe is consumed by fire at regular intervals and reconstituted out of this fire" (Khoury et al. 2001). The new ekpyrotic model depicted a kind of phoenix universe and was subsequently transformed by Steinhardt and Turok into an eternally cyclic model of the universe intended as an alternative to the standard inflationary scenario of the big bang. According to Steinhardt and Turok, the association to Stoic cosmology was suggested by two classics scholars, Joshua Katz from Princeton University and Katharina Volk from Columbia University (Steinhardt & Turok, 2007, p. 149).

Other modern authors have found a different kind of inspiration in Empedoclean and Stoic cosmology, seeing them as similar to the much later cosmological views of the twentieth and twenty first century, such as the steady-state theory of Hoyle and colleagues (2000), the quantum theory of an infinite universe which continually recycles itself (Joseph 2010a,b), or the relativistic theory of a closed cyclic universe with initial and final singularity (Ćirković 2003).

3. Cosmic Beginnings and Multiple Worlds

The cyclic universe is not the only modern cosmological idea that with some justification can be traced back to Greek-Roman antiquity, if of course only in a qualitative sense. The same is perhaps the case with the controversial idea of many universes – presently known as the multiverse – and its associated notion of the anthropic principle (Carr 2007). The cyclic universe of the Stoics thinkers constitutes one kind of multiverse, but in a temporal version only.

In sharp contrast to Aristotle, Epicurus (ca. 342-271 BC) advocated an original version of atomism according to which atoms moved ceaselessly in an infinite void, constantly forming and reforming the world. And not only that, for he also disagreed with Aristotle’s conclusion of the uniqueness of the world. Boldly stating that "there are infinite worlds both like and unlike this world of ours," Epicurus argued as follows: "For the atoms being infinite in number... have not been used up either on one world or on a limited number of worlds, nor on all the worlds which are alike, or on those which are different from these. So that there nowhere exists an obstacle to the infinite number of worlds." He further stated that "we must believe that in all worlds there are living creatures and plants and other things we see in this world" (Crowe 1999, p. 3; Trimble 2009).

Epicurus’ atomistic ideas greatly influenced the Roman poet Titus Lucretius Carus who about 50 BC composed his famous work On the Nature of Things (Latin: De Rerum Natura), one of the masterpieces of Greek-Roman natural philosophy. After having affirmed that the universe is spatially infinite--"All that exists... is bounded in no direction" --Lucretius proceeded with arguing for an infinity of inhabited worlds, of "other orbs of Earth in other regions of space, and various races of men and generations of beasts." He further explained that although the cosmos is infinite in space, it is of finite age and "there will be an end to the heaven and the Earth." He thus opted for a combination of space -time finitude-infinitude that was opposite to the one argued by Aristotle. Lucretius based his argument on the shortness of human history, which he found to be inexplicable if the world had existed eternally in the past:

According to Lucretius, not only did the universe have a beginning, it was also decaying towards a final end. "It is vain to believe that this frame of the world will last for ever," he wrote, "for neither do its veins, so to speak, submit to receive what is sufficient for its maintenance, nor does nature minister as much aliment as is needed." Here we have an early statement of what anachronistically can be considered the universal principle of dissipation or what in the nineteenth century became known as the heat death of the universe, supposed to be a consequence of the law of entropy increase.

The problem of the eternity of the world was raised by Stoic philosophers long before Lucretius. For example, Theophrastus reported how Zeno of Citium used the observed surface of the Earth, characterised as it is by mountains, valleys and plains, to argue that it could not have existed in an infinity of time (Freudenthal 1991, p. 50). This may be the first instance of a general line of reasoning that in the late nineteenth century was dicussed as the "entropic creation argument": From the existence of unidirectional natural processes, such as the steady increase of entropy, it can be inferred that the world is not eternal in the past (Kragh 2008).

Lucretius' reasoning has been read as "an almost modern formulation of the anthropic argument against the past temporal infinity" (Ćirković 2003, p. 883), which is a reference to what is sometimes known as the Davies-Tipler argument. The essence of this argument, due to Paul Davies and Frank Tipler, is that in an infinitely old universe one would expect colonization all over by technologically advanced civilizations. Since this is contradicted by observation, the world cannot have existed in an infinity of time (Barrow & Tipler, 1986, pp. 601-608). However, the Davies-Tipler argument is far from unproblematic and it is questionable if it qualifies as an anthropic prediction. At any rate, Lucretius' anticipation of the argument was not the only one of its kind and it is unjustified to speak of "the historical blindness of subsequent generations' to this form of argumentation (Ćirković 2003). In fact, from late antiquity over the middle ages and the renaissance to the modern period many philosophers and scientists have argued in similar ways for a universe of limited age (Kragh 2008).