SOURCE: William K. C. Guthrie, "Anaximander," in A History of Greek Philosophy, Vol. I, Cambridge at the University Press, 1962, pp. 72-145.

[In the excerpt that follows, Guthrie provides a historical framework for Anaximander's cosmology and cosmogony, contending that Anaximander made original and significant contributions to scientific thought.]

(1) Date, writings, interests

Anaximander was a younger friend and fellow-citizen of Thales…. Apollodorus says with unusual precision that he was sixty-four in the year 547/6 ([Diogenes Laertius, Lives of the Philosophers; hereafter D.L.] II, 2).¹ Following the tradition that Thales wrote nothing, Themistius described him as 'the first of the Greeks, to our knowledge, who was bold enough to publish a treatise on nature'. Certain it is that he wrote a book, which seems to have come into the hands of Apollodorus the chronologist, and we may feel some confidence that it was in the library of the Lyceum under Aristotle and Theophrastus. Yet it is perhaps worth remarking that neither Anaximander nor Anaximenes is mentioned by any writer before Aristotle. Plato, though he tells a story about Thales, and quotes the dictum elsewhere attributed to him that all things are full of gods, nowhere mentions the other two Milesians, nor makes any certain reference to their doctrines. This remarkable fact has led the Swiss scholar Gigon to suppose that Aristotle, with his deep interest in the historical aspect of his subject, must have sought out the works of these two and discovered copies which up to his time had been lost.

The Suda lists as titles of works by Anaximander: On Nature, Description of the Earth, The Fixed Stars, Sphere, 'and a few more'. These probably come from the catalogue of the Alexandrian library and represent divisions of a single work which Anaximander himself would almost certainly, in accordance with the custom of his time, have left unnamed, and, on the assumption that the titles are in fact sub-titles, the lists may well have varied. Throughout antiquity the title 'On Nature' … was given indiscriminately to the writings of the Presocratics, who from the main bent of their interests were known as 'the natural philosophers' or 'physiologers'…. The phrase was already in use as a title in the fifth and fourth centuries B.C., though this fact is not indeed proved by the passages commonly cited to support it, in which a Hippocratic writer or Plato refers to 'those who write on nature': this and similar phrases mark them off as a recognized group, but cannot be said to indicate anything so definite as a title.² More certain proof comes from something which does not seem to have impressed scholars in this connexion, namely the statement that Gorgias the fifth-century Sophist wrote a book which he called 'On the Non-existent or On Nature'.³ One cannot doubt that the deliberately provocative title was chosen by Gorgias himself, nor that it was intended as a parody of titles already extant. He may have had particularly in mind his contemporary Melissus, whose book according to Simplicius (Phys. 70. 16, De Caelo, 557. 10; DK, 30A4) was called 'On Nature or the Existent'.

The classification of his writings in the Suda may be fairly taken to represent the scope of Anaximander's interests. Coupled with the well-authenticated fact that he drew a map of the known world, it suggested to Heidel that he was more of a geographer than a philosopher and that the limited interest of the Peripatetics who were responsible for the doxographic tradition has therefore given a somewhat distorted picture of his achievements as a whole. The reports of his map go back to the great Alexandrian geographer and librarian Eratosthenes, e.g. that of Strabo who in claiming that geography is a study worthy to be called philosophical says, after giving pride of place to Homer (I, I, 11, DK, 12A6):

Those who followed him were clearly notable men and at home in philosophy, of whom Eratosthenes says that the first after Homer were two, Anaximander the acquaintance and fellow-citizen of Thales, and Hecataeus of Miletus. The one was the first to publish a geographical tablet [map of the earth], whereas Hecataeus left a treatise which is authenticated as his from the rest of his writings.⁴

Anaximander was also noted for his astronomical achievements, a natural accompaniment to his interest in the cosmos as a whole. He is said (D.L. II, 2) to have constructed a sphere, that is, some sort of model of the heavens, but unfortunately we have no details of this, and we are still at the cloudy stage of history when the attribution of particular actions or discoveries to an individual is almost impossible of verification. We read in Cicero that the first celestial sphere was fashioned by Thales (eam a Thalete Milesio primum esse tornatam, De Rep. I, 14, 22). Heidel mentions that according to Pliny (II, 31, DK, 12A5) Anaximander discovered the obliquity of the zodiac, but does not here note that Eudemus in his Astronomical History credited this to Oenopides in the fifth century (DK, 41.7). Like Thales, Anaximander was said to have invented, or introduced, the dial with upright rod (gnomon), and to have shown by its aid the 'solstices, times, seasons, and equinoxes' (Eusebius, DK, A 4, cf. D.L. II, 1). Herodotus, as we have seen…, regarded this as an importation from Babylonia, and the different words used by our authorities indicate at least some doubt as to the extent of Anaximander's achievement here. This dial, according to Favorinus (ap. D.L. II, 1), he set up at Sparta, a city with which he is further connected by a story in Cicero (De Div. 1, 50, 112) that he was responsible for a considerable saving of life by warning the Spartans of an impending earthquake and persuading them to spend the night in the open.⁵ Thus, as one would expect from his geographical interests, he evidently had the Ionian taste for travel, and Aelian (c. A.D. 200) says that he led the expedition to found one of the numerous colonies of Miletus, that at Apollonia on the Black Sea coast (V.H. III, 7, DK, A 3). No doubt like Thales he took a full part in the public life of his city, even if we may no longer accept the sixth-century statue bearing the name of Anaximander, the lower part of which has been discovered in the bouleuterion of Miletus, as having been erected in honour of the philosopher.⁶

Heidel's minute examination of the evidence from non-Peripatetic sources led him to the conclusion that Anaximander's book was, in short and summary form, a universal history and geography, 'purporting to sketch the life-history of the cosmos from the moment of its emergence from infinitude to the author's own time'. Carrying this tendency even further, Cherniss says: 'Anaximander's purpose was to give a description of the inhabited earth, geographical, ethnological and cultural, and the way in which it had come to be what it was.' This would mean that the only part of Anaximander's doctrine on which we have anything but the smallest and most doubtful bits of information, namely his cosmogony, was to him only incidental or preparatory to the main purpose of his work. We may admit the likelihood that Aristotle and his followers were silent about parts of the book that did not interest them, but to go so far in the opposite direction is to outrun the evidence.

Here our main purpose must be to attempt a reconstruction of Anaximander's cosmogonical views, and in this, as we have seen, we are better situated than we were with Thales. None of our informants, or their sources, had knowledge of a book by Thales. They were dependent on anecdotes or a few apophthegms, the authenticity of which was doubtful or worse. The treatise of Anaximander could be quoted, and its style criticized, by Theophrastus, and we are told that Apollodorus saw a copy in the second century B.C. Whatever we may think of their interpretations, it is safest to assume that Aristotle and Theophrastus both had the work, and to be correspondingly cautious in criticizing what they say from the standpoint of our own comparative ignorance.

(2) The Unlimited as 'arche '

The best starting-point will be the account which Simplicius gives, in large part from Theophrastus (Phys. 24.13, DK, A9 and Bl):

Anaximander named the arche … and element of existing things 'the boundless', being the first to introduce this name for the arche. He says that it is neither water nor any other of the so-called elements, but a different substance which is boundless, from which there come into being all the heavens and the worlds within them. Things perish into those things out of which they have their being, as is due; for they make just recompense to one another for their injustice according to the ordinance [or perhaps 'assessment'] of time—so he puts it in somewhat poetical terms.

Having thus paraphrased and in part quoted Anaximander's words, Simplicius, with Aristotle and Theophrastus before him, proceeds to interpret them:

It is clear that when he observed how the four elements change into each other, he did not think it reasonable to conceive of one of these as underlying the rest, but posited something else. Moreover he does not account for genesis by a qualitative alteration of the element, but by a separation of the opposites caused by the eternal motion.

Few passages descriptive of Presocratic doctrine have escaped a thorough mauling from many modern commentators. The above is no exception, and many difficulties have been discovered, if not created, in it. The casual aside, that Anaximander's language here is rather poetical, gives us the valuable information that the previous sentence, though cast in indirect speech in the Greek, preserves some of his actual words. At a minimum, the criticism must refer to the clause: 'for they make just recompense to one another for their injustice according to the ordinance of time', and this is sufficient guarantee that the preceding clause is a true representation of Anaximander's thought.⁷ The statement in Simplicius's explanation that Anaximander accounts for the origin of things 'by a separation of the opposites', etc., depends no doubt on Aristotle, who writes (Phys. 1, 4, 187a20): 'Others teach that the opposites are in the one and are separated out, as Anaximander says.'⁸

It is clear (though a different view has been taken) that the first sentence in the passage from Simplicius means that Anaximander was the first to give the name apeiron (boundless, unlimited) to the arche. That he was also the first to use arche for that which writers from Aristotle onwards, with rather different ideas in their heads, called 'the substratum' appears not from this but from another passage of Simplicius (Phys. 150.22): 'Anaximander says that the opposites were in the substratum, which was a boundless body, and were separated out: he was the first to name the substratum arche.' We notice also that Theophrastus deemed it necessary to explain the archaic word by adding the Aristotelian term … (element).⁹

With Anaximander physical theory takes a momentous step, to a notion from which it has retreated many times before its reappearance in very different form in the modern world: the notion of the non-perceptible. 'The physical view of the world', writes the physicist von Weizsäcker, 'has always had a tendency towards the nonperceptible. This stems immediately from the endeavour of physics to achieve a unified world-view. We do not accept appearances in their many-coloured fulness, but we want to explain them, that is, we want to reduce one fact to another. In this process what is perceptible is often explained by what is not perceptible.'¹⁰

Anaximander then rejected the idea that water, or any of the popularly (and later philosophically) recognized elemental masses visible in the world of today, could have served as a basis for all the rest. Instead he posited an unnamed substance behind them all, less definite in character, which he described as apeiron (from a privative, indicating absence, and peras = limit or boundary). There was no reason for regarding water, earth, fire or any such familiar, sensibly manifest phenomenon as prior to the rest. The original matrix of the universe must be something more primitive and ultimate than any of them, of which they are all alike secondary manifestations or modifications, obtained by a process of 'separating out'.

The following questions therefore suggest themselves: Why did he thus go behind the phenomena? What did he mean by apeiron? What were the 'opposites', and in what sense 'in the one and separated out'?

(3) The opposites

The assumption of an imperceptible reality behind the perceptible was, for one seeking a unity behind the multiplicity of phenomena, on general grounds a reasonable one, as von Weizsäcker has confirmed from the scientist's standpoint. Anaximander had also a more specific reason for adopting it, and this introduces a fundamental feature of Greek thought with a long and influential history, namely the notion of the primary opposites. Later, when substance and attribute had been clearly distinguished by Plato and Aristotle, it was said that the elements—earth, water, air and fire—were characterized by one or more of a set of contrary qualities, hot, cold, wet and dry, and because of their contrary attributes were always in a state of conflict. European literature attests the vitality of this semi-anthropomorphic notion. From Ovid—

Frigida pugnabant calidis, umentia siccis—

we pass to Spenser—

The earth the air the water and the fire
Then gan to range themselves in huge array,
And with contrary forces to conspire
Each against other by all means they may—

and Milton—

Hot, cold, wet and dry, four champions fierce
Strive here for maistery.

When Anaximander first tried to give philosophical expression to the idea, no clear distinction was possible between substance and attribute. Just as he spoke of 'the boundless', so also he designated the opposites by article and adjective as the hot, the cold, the wet and the dry.¹¹ These, as Cornford has said, are for Anaximander not qualities but things. "'The hot" was not warmth, considered as an adjectival property of some substance which is warm. It is a substantive thing, and "the cold", its contrary, is another thing. Hence it was possible to think of the hot and the cold as two opposed things which might be fused together in an indistinct condition, like a mixture of wine and water' (Princ. Sap. 162).

The conflict of the opposites is an undeniable fact of nature. Water for instance, whose nature it is to quench fire whenever it meets it, can hardly be the original substance out of which fire, along with all the other forms of material existence, had its being. Aristotle puts the argument thus, though without mentioning Anaximander by name:

Some thinkers make this [sc. a substance other than the elements, out of which they have evolved] the unlimited, not air or water, to prevent their destruction by that one of them which is unlimited; for they are marked by mutual opposition—e.g. air is cold, water wet, fire hot—so that if one of them were unlimited, the others would have perished. As it is, they say, it is something else, out of which the known elements come (Phys. III, 204b 24).

The conflict is referred to by Anaximander himself in the only well-attested fragment of his writings: 'They make just recompense to one another for their injustice according to the ordinance of time.'

To avoid misunderstanding, a distinction must be observed here which may at first sight seem rather subtle. There is a sense in which water (the cold and wet) can and does give birth to its opposite, fire (the hot and dry). No other meaning can be attached to Anaximander's sentence than that the 'injustice' which they commit consists in an encroachment, say, of fire by swallowing up some of its rival water, and vice versa. It was in fact a common Greek belief, which emerges still more clearly in Anaximenes, that the fiery heat at the circumference of the universe (that is, in the present world-order the sun) not only vaporized the moisture of earth and sea, thus turning it into mist or air, but finally ignited it and transformed it into fire. The process was actually spoken of as the 'nourishment' of the sun by water or moisture, as we saw in connexion with Thales….

In this sense fire can be created out of water, but only because of the simultaneous existence of both, and, as Anaximander says, their balance is always being redressed: the encroachment of one opposite is followed by a retribution in which the other regains the lost ground. Fire becomes cooled into cloud, cloud into rain which once more replenishes the moisture on earth. This alternate advance and retreat of the hot and the dry, the cold and the wet, is an obvious expression of the annual variation of the seasons.¹² It in no way contradicts the observation which led to the abandon ment of one of the opposites as primal arche, for it remains as true as ever that in a universe which was all water, like that of the Mesopotamian and Egyptian myths which Thales perhaps rationalized a little too precipitately, no fire could ever have come into being. For water to turn into fire requires the action of fire already existing.

Thus whether the sentence 'things perish into those things out of which they have their being, according to necessity' is also Anaximander's or is a paraphrase by Theophrastus or Simplicius, it cannot refer (as it has frequently been thought to do) to the primal generation of the opposites out of, and final reabsorption into, the ultimate apeiron, but only to their mutual transformations in the present order. Otherwise its connexion with the quotation which follows would make no sense.¹³

To sum up, Anaximander had noticed that it is the natural tendency of each of the elements to swallow up its opposite. Fire and water must inevitably be in conflict. When they meet they struggle until one or the other prevails, and either the fire is put out and noth ing but the water remains, or else the water is dried up and fire remains in sole possession of the field. Conversely this may be described, in Simplicius's words, as the conversion of water into fire and vice versa. There is of course an intermediate stage, clearly visible to observation, of the conversion of water into steam or vapour, which for the Greeks are included in the term aer. In the world as a whole, complete and final victory is never granted to one or the other of the opposing forces (or litigants, as Anaximander imagined them): the balance between them is always being restored or maintained. If one gains a local advantage, the other is encroaching elsewhere.

Now if the world is evolved from a single substance, there must be at least enough of this substance to make the whole world, and probably a good deal more besides. But if fire existed in that quantity, it would inevitably enjoy a permanent victory over its potential rivals, none of which could be allowed to come into existence; or if the arche and physis of the world were water, there could never be fire. This remains true whether we take Anaximander's word apeiron, which he applied to his primary state of matter, to mean strictly infinite in extent, as Aristotle did, or simply of an indefinite quantity large enough to serve as source or reservoir¹⁴ from which all that exists has been drawn. What exactly he did mean by the word has been matter of considerable controversy, and is now due for consideration.

It was long customary to regard the Milesians as interested only in the question 'What is the world made of?' They assumed it to be made of one material substance, and asked only whether that substance was water, air or something else. This was Aristotle's view, because when he approached them from the standpoint of his own fourfold scheme, seeking only, as he tells us, for anticipations of the material, efficient, formal and final causes as he conceived them, they appeared to be concerned only with 'principles of a material kind'…. But by thus limiting the scope and purpose of his review, he has undoubtedly misled those who, ignoring his own explicit declaration of intentions, supposed him to be writing a history of philosophy. Not 'matter' (for which they had no word, since they knew of no other form of existence) but rather 'nature' (physis) is the correct keyword. It may be that no certain instance of this word occurs in the scanty fragments of the philosophers before Heraclitus, but we have it in a very similar sense in Homer,¹⁵ and this with the universal consensus of antiquity is enough to justify the claim of Pohlenz that 'the concept of physis is a creation of Ionian science, in which they summed up their new understanding of the world'.¹⁶ Most commonly it meant the real constitution or character of things, including the way they behave, though it could also mean 'birth' or 'growth' (e.g. in Empedocles, fr. 8). The two are not unconnected, since, as Aristotle said (Phys. 193b12), 'Physis in the sense of comingto-be is the path to physis' (in the sense of state or structure finally reached). Physis could be both process and constitution or developed form, and the Milesians were interested in both aspects, though the evidence, such as it is, suggests that the latter sense (which it has in the Odyssey) is likely to have predominated in the sixth century.

The 'new understanding of the world' consisted in the substitution of natural for mythological causes, that is, of internal development for external compulsion. This, as Pohlenz says, is well expressed by the generalized use of physis,¹⁷ which is something essentially internal and intrinsic to the world, the principle of its growth and present organization, identified at this early stage with its material constituent. The primary assumption is not simply that it consists of a single material substance, but that the diversity of its present order is not from eternity, but has evolved from something radically simpler at a particular point in time.

(4) The meaning of 'apeiron '

To this initial simple state or arche Anaximander gave the name of the Boundless, and the process by which a world-order emerged from it he described as a 'separating-off. To consider first the initial state itself, how did he conceive it and why did he call it apeiron? Aristotle (Phys. 203b 15; DK, A15), for whom the word had the strict sense of infinite, mentions five considerations as leading to the belief that something is apeiron. We may take it that they include all the traditional aspects of the word up to and including his time. In the first of these aspects, the temporal, the apeiron of Anaximander certainly deserves to be called infinite. The notion of temporal infinity was familiar to the Greek mind from remote antiquity in the religious conception of immortality, and Anaximander's description was in terms appropriate to this conception, for like many of his successors, says Aristotle (Phys. 203b 13; DK, B3), he called his arche 'deathless and imperishable'. According to Hippolytus he also applied to it the words 'eternal and ageless' (Ref. 1, 6, 1; DK, B2). This marks it off as something of a different order from anything recognizable in the present world, and also illustrates the meaning of arche as both the original state of things—for it has existed from all time—and the permanent ground of their being. The arche of all things cannot itself have an arche—a beginning—because then not it but that further arche would be the ultimate one. And what has no arche, and also no ending, is apeiron, for an arche would be a limit. So, in effect, says Aristotle (Phys. 203b7), using an argument which seemed to Cornford to have 'an archaic ring'.¹⁸

Apart from the temporal sense of 'everlasting', apeiron has two main meanings, according as the 'boundaries' (perata) which it lacks are thought of as external or internal. If a body is limited externally, this can only be because it comes up against something else, or so at least it seemed to Aristotle and later writers.¹⁹ Beyond its limit there must be something other than itself. Conversely, then, a body which is unlimited in this sense must continue infinitely, or at least indefinitely, in space. In the Placita of Aëtius (DK, A14) we are told that Anaximander regarded the apeiron as infinite in this quantitative sense 'in order that becoming might not fail'. The extant 'opposites', as we have seen, and more complex bodies composed of them, are continually perishing. Consequently, it seems to be argued, if the supply of them is to be kept up—as it is and has been for time out of mind—the reservoir out of which new supplies come—that is, the apeiron—must be inexhaustible and therefore infinite.

It seems doubtful, however, whether Anaximander used this argument himself.²⁰ It looks rather as if the author had drawn that inference from a sentence in which Aristotle denies that this is a valid reason for supposing the existence of an infinite body, but without suggesting that Anaximander did so. 'Nor is it necessary', writes Aristotle (Phys. 208a8; DK, A14), 'for an infinite sensible body to exist in actuality in order that becoming may not fail; for the destruction of one thing may be the genesis of another, while the whole sum remains finite.' What Aristotle says is clearly right. The process of becoming and perishing is circular. Perishing does not mean vanishing into nothingness, but changing into a different form of matter. This circularity, symbolized by anaximander as the alternation of 'injustice' and 'reparation', seems to have been central in his thought. If he did not see that it does away with the necessity for an inexhaustible reservoir of primal matter which is everlastingly being drawn upon to form new creatures and other things in the world, then his mind was less acute than the evidence suggests it to have been.

It is unlikely that Anaximander was capable of grasping the notion of strict spatial or quantitative infinity, which came with further advances in mathematics. It is indeed purely conceptual, and has no meaning in the world of immediate sensible experience. As one of the five reasons for believing in an infinite, Aristotle says that number, mathematical magnitude and the space beyond the sky are thought to be infinite 'because they never give out in our thought'. It is hardly credible that Anaximander reasoned like this. He certainly regarded the apeiron as an enormous mass surrounding … the whole of our world, but it may even have presented itself to his mind, as Cornford suggested, as a vast sphere. The word was in use in Greek to describe both spherical and circular shape, and, in an age without any sciences of grammar, semantics or logic, men were at the mercy of words to an extent which it is not always easy to realize. A word was more like a single whole entity, and its various meanings, which we without difficulty analyse and separate, could only appear as different aspects or facets of a single meaning.

It is right therefore to take into account the fact that apeiron was used of spheres and rings, to indicate no doubt that one can go on and on around them without ever coming to a bounding line. This comes out particularly clearly when Aristotle says (Phys. 207a2) that finger-rings are called unlimited if they have no gem-socket. Empedocles (fr. 28) speaks of an unlimited sphere, and the word is also applied to a seamless robe and a circular band of worshippers round an altar.²¹

Secondly apeiron was used with internal perata chiefly in mind, to indicate that no line could be drawn between part and part within the whole. In this way it approximates to the notion of indeterminacy. A body unlimited in this sense may be made up of different sorts of matter, but they are fused into an indistinguishable mass. Standing on the shore, we can see clearly where sea and earth and air begin and end. The world is not apeiron in the sense we are considering. But we can imagine some cataclysm occurring which would destroy those boundaries, just as we can imagine an initial state of chaos before the main divisions of the world were so cleanly distinguished as they are now. If earth, sea and sky were fused in one heaving, molten mass, the world might be described as a boundless, or unlimited, mixture…, meaning that the boundaries between its various components were non-existent and they were inextricably confused. The extent of the world's own boundaries is not in question.

Let me repeat that we are not at a stage of thought when clear distinctions between different uses of the same word are possible. Some inheritance of the magical idea that a word or name has an independent existence and essence of its own, and can only therefore be one thing, persisted until later times than this, and influenced even the thought of the most enlightened, however far it may have receded into the subconscious. Of that the Cratylus of Plato is ample evidence. There is no question then of deciding in which of several senses Anaximander intended us to take his word, but only which sense was upper-most in his mind. This is likely to have been the notion of internal indeterminacy rather than of spatial infinity, since the former offered a solution to the problem that he was trying to solve. He was impressed, as we have seen…, with the difficulty of supposing the single primary element to be water, or 'the wet', as Thales had done, or any of the actual opposites with their determined characteristics. Owing to his belief in the inherent hostility and 'injustice' of these, any single one of them, far from serving as source of being to the rest, would prevent it altogether. A primitive stuff must be, so to speak, a neutral in these hostilities, and must therefore have no definite characteristics of its own. It must hold, inactive in the first place and suspended as it were in solution, the characteristics of all the future opposites which in due course were to be, in the significant word which was probably his own, 'separated off' (or 'out') from it. Here we may find, in all probability, the chief reason why he called his arche simply 'the apeiron'. There were no perata in it between the hot, the cold, the wet and the dry. Before the formation of a cosmos, the opposites as such could be said to be as yet non-existent, because they were indistinguishably mingled. At the same time (to use a resource of language that was not at Anaximander's disposal) they were present in a potential state, so that their subsequent emergence into actual and active being was always a possibility. The difficulties of this conception, at least as it was expressed in the crude language of his time, were not immediately apparent. To bring them out fully required the uncompromising clarity of a Parmenides. If the opposites could be separated out from the arche, we may say, it must have contained them all the time and therefore could not be described as a unity. In applying the ancient formula 'everything came into being out of one thing', Anaximander virtually cheated.²² But to make this criticism belonged to a more advanced stage of thought, a necessary stage between the naive monism of the Milesians and the Aristotelian distinction between various modes of being.²³

(5) The 'apeiron' divine

There is a little more to be said about Anaximander's arche, based on the words of Aristotle in Phys. 203b6 (DK, A15):

Everything either is an origin or has an origin: the unlimited has no origin, for that would be a limit of it. Moreover, being an origin [or source or principle: arche], it is ungenerated and imperishable…. Therefore, as I say, there is no origin for it, but it appears to be the origin of other things and to encompass all things and direct all things, as those philosophers say who do not posit besides the unlimited other causes such as Mind or Love; and this they say is the divine, for it is immortal and imperishable, as Anaximander and most of the writers on nature call it.

Aristotle is here distinguishing later thinkers, to whom the belief in an animate self-moving stuff was beginning to seem unsatisfactory so that like Empedocles and Anaxagoras they moved towards the notion of a separate moving force, from those who like the Milesians were still at the hylozoist stage. For these a single arche filled the dual role; it included or surrounded all things, and was also the directive force. This verb…, literally 'to steer', was applied in the fifth century by Diogenes of Apollonia (fr. 5) to air, which he adopted as the arche from Anaximander's successor Anaximenés. Elsewhere among the Presocratics we find it in Heraclitus (whatever the correct reading and interpretation of fr. 41) and Parmenides (fr. 12, v. 3). In all probability this word and the rest of the language here quoted from 'philosophers of the unlimited' go back to Anaximander as well as the two epithets explicitly vouched for by Aristotle as his.²⁴

These words, as Aristotle says a little later (207a18), impart a certain loftiness of tone to the pronouncements of early philosophers on the apeiron. Indeed the attribution to the arche not only of life but of directive powers immediately suggests divine status. The same verb (to steer…) is of course applied to divinities in non-philosophical contexts. It is therefore no surprise when Aristotle goes on to ascribe divinity explicitly to the arche of Anaximander and those who thought like him. For a Greek indeed, as he indicates in the next clause, it follows directly from the fact of immortality. If it includes directive or governing power it also implies at least some form of consciousness. For Anaximander we have no further evidence on this point, but later monist philosophers ascribe consciousness and intelligence explicitly to their single material arche. This is the beginning of the road which will lead ultimately to the separation of matter and moving cause, that is of matter and spirit, as the difficulty of their identification becomes more apparent; but that is still in the future. At present the very word 'matter' is an anachronism.²⁵

(6) Cosmogony and cosmology

From the primal state, or original source of all things, we turn to the process by which a world-order comes into being. This is described as being, in general terms, one of 'separating-out', caused by an 'eternal motion' in the apeiron. In Aristotle's words (Phys. 187a20), 'the opposites are in the one and are separated out'. This statement of the process follows well on our description of the primary nature of the apeiron as an initial indeterminate fusion of all the opposites.²⁶ But we are not confined to the general term ekkrisis (or apokrisis) for our knowledge of how Anaximander supposed a world to be formed from the apeiron. Part of a description of his cosmogony, taken by Eusebius from the compilation called Stromateis and originating in Theophrastus, reads thus (DK, A10):

He says that at the birth of this cosmos a²⁷ germ of hot and cold was separated off from the eternal substance, and out of this a sphere of flame grew about the vapour surrounding the earth like the bark round a tree. When this was torn away and shut off in certain rings, the sun, moon and stars came into existence.

The last sentence can be better understood by comparison with the following….

(a) (Aët. 11, 13, 7, DK, A18) Anaximander says that the stars are wheel-shaped concentrations (lit. 'feltings') of mist filled with fire, breathing out flames through openings in a certain quarter.

(b) (Hippolytus, Ref. 1, 6, 4, DK, All) The stars come into being as a circle of fire, separated off from the fire that pervades the cosmos and surrounded by mist. There are breathing-places, certain pipe-like passages,²⁸ through which the stars appear. When these are blocked, eclipses occur.

The word … here translated 'germ', is an adjective meaning generative, fertile, able to bring to birth, and is used of eggs and seed. It is used again by Theophrastus in De Igne, 44 in relation to the life of animals and plants only. We can never know whether it is the actual word used by Anaximander, but it is in keeping with the language of organic generation which seems to pervade the passage and, as we saw in discussing Aristotle's conjecture about Thales, is a likely colour for the thought of these early speculators to have taken…. The whole sentence strongly suggests, as Professor Baldry has well brought out,²⁹ that Anaximander conceived his cosmogony on the analogy of early views concerning the seed of animals and the development of the embryo. The mythical world-egg of Orphic and other cosmogonies shows how primitive such a notion could be, and the 'separation' … of the seed in the womb, the part played by hot and cold…, and the 'detachment' … of the new organism from the parent body, are all familiar from Greek medical writers as well as finding their place in the present account. As to … (the word translated 'bark' above), one may note with Baldry that it means 'any skin that forms round a growing organism, whether plant or animal'. Aristotle (H.A. v, 558a28) uses it of the membrane round an egg, and Anaximander himself is said to have applied it to the prickly skin which on his theory surrounded the earliest forms of animal life. It looks as if Anaximander saw the outer 'skin' of the embryo world, separating it from the womb of the 'Boundless' in which it was formed, as a parallel phenomenon to this membrane which developed round eggs, animal embryos and trees alike. Since the world's skin is spherical, the reference to trees (which may have been added by Theophrastus or even later) is obviously not intended to be pressed.³⁰

That cosmogony should be described in terms of or ganic life is appropriate to the mentality of these intellectual pioneers. The arche of Anaximander, the doxographers tell us,³¹ was in eternal motion. The reason for this is nowhere explained, an omission censured by Aristotle but no doubt due to persistence of the belief that the arche is eternally alive. Since for the Greek the very notion of life involves self-caused motion, no external cause was conceivable, let alone demanded. Anaximander has rejected the anthropomorphic imagery of sexual mating which formed the basis of mythical cosmogonies, but for him it is still natural and rational to regard the matrix of the world as animate and its origin as taking place from a kind of seed or egg.

This fertile nucleus, pregnant with the opposites, becomes detached from the Boundless and develops into a sphere of fire enclosing a cold, moist mass. Between the two is dark mist…. At this stage only two primary opposites can be said to be separated, hot including dry and cold including wet. The mist arises from the action of the hot periphery on the cold-wet centre, and, under the same action of heat, wet and dry become in the end more completely separated, producing land and sea. So Aëtius (DK, A27 fin.):

Anaximander says that the sea is a relic of the primal moisture, the greater part of which has been dried up by the fire.

Anaximander was among those whose accounts of the origin of the sea are mentioned by Aristotle in the Meteorologica (353b5, trans. Lee):

Those who were more versed in secular philosophy [as opposed to the ancient theological poets] suppose it to have had a beginning. They say that at first the whole region about the earth was wet, and that as it dried up the water that evaporated became the cause of winds and the turnings of the sun and moon, while what was left is the sea: consequently they believe that the sea is still drying up and becoming less, and that in the end a time will come when it is all dried up.

He is mentioned by name in the commentary of Alexander of Aphrodisias on the passage (DK, A27):

Some of them say that the sea is a residue of the primal moisture. The region of the earth was moist, and subsequently part of the moisture was vaporized by the sun … but part of it left behind in the hollows of the earth forms the sea. Hence it is continually becoming less as the sun dries it up, and eventually it will be dry. Of this opinion, according to Theophrastus, were Anaximander and Diogenes.

It is characteristic of Milesian thought that once the separation of the mutually hostile opposites has begun, the process of cosmogony is continued by the natural exercise of their respective powers: heat dries up moisture and so on. Interesting also, after Aristotle's and our own conjectures about Thales, is the immediate prominence of moisture and heat as soon as fertilization and generation are to take place. Heat especially has an important part to play as a first agent of genesis, and at a later stage it is the action of heat on moisture which produces animal life. These features of Anaximander's system strengthen the case against those who have disparaged Aristotle's conjecture as arising out of the later progress of physiological and medical knowledge in Greece. We find rather, as might be expected, certain points of contact between the two who were fellow-citizens and fellow-workers in their field.

The next stage in the same continuous process explains the formation of the heavenly bodies. In addition to evidence already quoted, we have the following:³²

(a) Hippolytus, after the words quoted…, adds: 'and the moon is seen to wax and wane according as the passages close or open'. There follows a sentence in which some words have probably dropped out of the manuscripts, but which seems to say that the circle of the sun is twenty-seven times the diameter of the earth³³ and that of the moon eighteen times, and adds that the sun is the highest of the heavenly bodies, and the stars are the lowest.

(b) Simpl. De Caelo 417.4 (DK, A19, speaking of the planets): Anaximander was the first to discuss their sizes and distances, according to Eudemus, who attributes the first determination of their order to the Pythagoreans. The sizes and distances of the sun and moon are reckoned to this day by taking eclipses as the starting-point of our knowledge, and we may reasonably suppose that this too was Anaximander's discovery.'

(c) Aët. 11, 15, 6 (A18): 'Anaximander, Metrodorus of Chios and Crates held that the sun was situated highest of all, next the moon, and beneath them the fixed stars and planets.'

(d) Ibid. 20, 1 (A21): 'According to Anaximander, the sun is essentially a circle twenty-eight times the size of the earth, shaped like a cartwheel. The rim is hollow and full of fire, and at a certain point allows the fire to be seen through an orifice like the nozzle of a bellows: this is the sun.'

(e) Ibid. 21, 1 (A21): 'Anaximander says that the sun is the same size as the earth, but the circle in which is its blowhole, and by which it is carried round, is twenty-seven times the size of the earth.'

(f) Ibid. 24, 2 (A21): 'According to Anaximander the sun is eclipsed when the orifice through which the fire escapes is shut up.'

(g) Ibid. 25, 1 (A22): 'According to Anaximander, the moon is essentially a circle nineteen times the size of the earth, resembling a cartwheel with the rim hollow and full of fire like that of the sun, lying obliquely as does the sun's and having a single blowhole like the nozzle of a bellows. It is eclipsed according to the turnings of the wheel.'³⁴

(h) Ibid. 29, 1 (A22): 'Anaximander says that the moon is eclipsed when the orifice in the wheel becomes blocked.'

(i) Ibid. 28, 1 (A22 omits last phrase): 'Anaximander, Xenophanes and Berosus say that the moon has its own light, in some way rarer [than the sun's].'³⁵

In spite of minor discrepancies, we may accept the following account as probable. The fiery, spherical membrane about the new-born cosmos parted (doubtless under increasing pressure from the mist or steam caused by its own action in evaporating the watery centre) into separate circles, around each of which the dense mist surged and closed. Where there are apertures in this surrounding envelope, we see the heavenly bodies themselves. Thus the sun and moon are really rotating wheels of fire going right round the earth, but encased in tubes of mist except at one point where there is a hole, through which the fire streams like an ignited jet of gas through a leak in its pipe. (The modern simile is closer than the Greek one of air escaping through the nozzle of a pair of bellows.) The circles of the stars are not so easy to visualize from our fragmentary authorities, but one would suppose that each contained many holes.³⁶ Mention of the Milky Way, as by some modern authorities, hardly gives an adequate explanation, though its appearance may possibly have helped to put the idea of the wheels into Anaximander's head. They were evidently all regarded as lying in the same spherical plane, nor are the planets yet distinguished in this respect from the fixed stars. (Eudemus, in passage (b) above, attributes the deter mination of the planetary orbits to the Pythagoreans. Simplicius's introduction of Anaximander in the context of the planets is confusing, as is his apparent suggestion that Anaximander himself might have calculated the sizes and distances of the sun and moon from the observation of eclipses.)

To suppose that the stars are nearer the earth than either sun or moon is contrary to later Greek astronomy, according to which the fixed stars are—as seems most natural—in the plane of the outermost circumference of the spherical cosmos, and the sun, moon and planets revolve in different orbits beneath them. Anaximander's order raises the question in a modern mind how the rings of the stars avoid obstructing, at least at times, the light of the sun and moon, but it is very doubtful whether this consciously troubled him.³⁷

We may assume that the rings are one earth-diameter thick. The variants in the reported sizes (diameters) of the rings of sun and moon (27 and 18 or 28 and 19 times the size of the earth) were, since Burnet's time (EGP, 68), accounted for as measurements to the inner or outer surface of the rings, until Kirk pointed out the simple fact that this requires a difference of two earth-diameters, not one. He suggests that 'the larger figure might represent the diameter from outer edge to outer edge, the smaller one that from points half-way between the inner and outer edges of the actual felloe of air' (KR, 136). In any case the larger figures are likely to have been some commentator's refinement on the simple scheme of Anaximander expressed in multiples of three. No statement of the size of the star-rings is preserved, but since the diameter of the earth is said to be three times its height…, it looks as if these numbers have a conventional or sacred origin which Anaximander has not outgrown; in which case the missing number seems to be nine.³⁸

The statement that the visible sun is the same size (of the same diameter) as the earth is, for Anaximander's time, most remarkable. (In the next century Anaxagoras could be prosecuted for saying that it was an incandescent stone larger than the Peloponnese.) It also causes a difficulty if we try to correlate it strictly with the distance of the sun from the earth, i.e. the diameter of its wheel.³⁹ This need not have presented itself forcibly to his mind, and all the evidence confirms that he was a fearless and original thinker. Perhaps, however, the possibility that the statement is not authentic cannot be altogether excluded.

The well-attested explanation of eclipses, and of the phases of the moon, as due to alternate contracting and opening of the holes in the tubes of mist through which the heavenly bodies are seen, is another indication of the inchoate state of Anaximander's astronomy, and puts out of court the charitable guess of Simplicius that he might already have been capable of using these phenomena to calculate the sizes and distances of the sun and moon.

One can hardly extract further detail on this part of his system with any approach to certainty. Aëtius speaks of the circles of the sun and moon as 'lying obliquely', presumably to the celestial equator, and the phrase is no doubt, as Heath says, an attempt to explain the annual movement of the sun and monthly movement of the moon. Ingenious ways have been suggested in which Anaximander may have intended to explain the solstices,⁴⁰ but all are conjectural. It is not even certain whether the word 'turnings'…, occurring in passages which are evidently meant to apply to Anaximander among others, refers to the solstices⁴¹ or simply, as it sometimes does, to the revolutions of the heavenly bodies.⁴² In [a] passage of the Meteorologica…, Aristotle states that the action of the heavenly fire in drying up the water caused 'winds and the turnings of the sun and moon'. Commenting on this Alexander says … :

From it [that is, that part of the original moisture which was vaporized by the sun] arose winds and turnings of the sun and moon, the notion being that the turnings (revolutions?) of those bodies too are accounted for by these vapours and exhalations, since they turn in those regions where they receive a plentiful supply of the moisture.

Here is a clear reference to the early idea that the cosmic fires, or heavenly bodies, are 'nourished' by moisture….⁴³ Further than that this second- or third-hand description will hardly allow us to go. Anaximander may have been supposing the limits of the sun's path in the ecliptic to be fixed by the abundance, in a certain region of the sky, of the moisture on which it depended for its existence; or he may have been trying to produce a theory to account for the whole fact of the cosmic revolutions, suggesting that the motion was started and maintained by these currents of air which the evaporating process somehow set up. We are not offered any other explanation of the revolving motion of the cosmic circles, and the only alternative is to suppose that the movement was somehow implanted by the 'eternal motion' of the Boundless, the nature of which is not specified. In 'giving birth' to its 'offspring' the cosmos, it produced no stillborn child. Language like this has been shown to be appropriate to Anaximander's thoughts in no merely metaphorical sense. Ingenuous as it sounds, this explanation is on the whole the more likely. Enough remains to show that astronomy was still in its infancy among the Greeks. The strength of someone like Anaximander lay in the bold flight of imaginative reason with which he sketched the outlines of a cosmos, and we may agree with Dreyer that 'probably the system never advanced beyond a mere sketch and was not worked out in detail'.

The evidence for the shape and position of the earth is as follows:

(a) [Plut.] Strom. (A 10…): 'And he says that the earth is cylindrical in shape, with a depth three times its breadth.'

(b) Hippolytus, Ref. 1, 6, 3 (A11): 'The earth hangs freely, not by the compulsion of any force but remaining where it is owing to its equal distance from everything. In shape it is rounded [see below for this word], circular, like the drum of a column; of its surfaces one is that on which we stand, and there is another opposite.'⁴⁴ Aëtius (A25) repeats that the earth 'resembles the drum of a column'.

(c) The reason why the earth remains at the centre had previously been more fully given by Aristotle (De Caelo, 295b10, A26): 'But there are some who name its "indifference"⁴⁵ as the cause of its remaining at rest, e.g. among the ancients Anaximander. These argue that that which is situated at the centre and is equably related to the extremes has no impulse to move in one direction—either upwards or downwards or sideways—rather than in another; and since it is impossible for it to accomplish movement in opposite directions at once, it necessarily remains at rest.'

(d) Eudemus, Astronomy, quoted by Theo (p. 198.18 Hiller, A26) via Dercyllides: 'Anaximander says that the earth is freely suspended and moves around the centre of the universe.'

The exact meaning of the word γυρᾳν (translated by 'rounded' in passage (b); it is a correction for the impossible ύγρᾳν of the manuscripts) is difficult to determine. The lexica gloss it as both 'round' and 'convex', and it is used of a round-shouldered person in the Odyssey (XIX, 246). Anaximander, if he used the word, may have meant that the surfaces of the earth are not flat but convex, as observation might suggest, though this would make the comparison to 'the stone of a column' less appropriate. The corresponding noun (γύρᾳς) is used of something ring-shaped, as for example a trench dug round a tree, and another possibility is that Anaximander meant to indicate that the earth had a hole at the centre, thus bringing its shape into line with the circles of the heavenly bodies around it. Column-drums often had such a hole.⁴⁶

The statement quoted from Eudemus in passage (d), that the earth is in motion, need not be taken too seriously. In the same passage Eudemus is credited with a probably exaggerated account of Thales's astronomical knowledge and with saying that Anaximenes discovered the cause of eclipses of the moon and the fact that its light is derived from the sun. As Zeller suggests, and Alexander seems to have suspected, there has probably been a misunderstanding of the words in which Anaximander expressed his highly original notion that the earth floats freely in space with nothing to keep it stationary.⁴⁷

Anaximander's most striking contribution to cosmological theory was undoubtedly to emancipate himself from the idea that the earth needed a support. The belief that it floated on water was, as we saw, an inheritance from mythology perpetuated by Thales, and intellectually it was a leap forward when the argument from 'indifference' was invoked in favour of the view that it remained unsupported at the centre of a spherical universe, and that the heavenly bodies revolved in complete circles below as well as above it. Nothing shows more clearly the independent quality of Anaximander's mind, and, as we shall see, the advance was too rapid for some of his successors. Nearly two centuries later, Plato paid him the compliment of making Socrates adopt his view, when he said in the Phaedo (108E, trans. A. J. Church):

In the first place then, I believe that the earth is a spherical body placed in the centre of the heavens, and that therefore it has no need of air or any other force to support it: the equiformity of the heavens in all their parts, and the equipoise of the earth itself, are sufficient to hold it up. A thing in equipoise placed at the centre of what is equiform cannot incline in any direction, either more or less: it will remain unmoved and in perfect balance.

Clearly the recognition of the earth's sphericity could not be long delayed, but it did not appear first in the Milesian tradition, and the mention of air is a reminder that later Ionians went back to the more simple-minded notion that the earth needed material support, for they supposed it to be buoyed up by air.

We are told (and might in any case have assumed) that just as the world-order had a beginning out of the apeiron, so also it will have an end, fading back, as it were, into the formless state from which it came. Only the apeiron itself is 'eternal and ageless', 'immortal and indestructible'. So Aëtius (A14):

Anaximander of Miletus, son of Praxiades, says that the first principle of existing things is the Boundless; for from this all come into being and into it all perish. Wherefore innumerable worlds are both brought to birth and again dissolved into that out of which they came.

But our sources nowhere explain how this will occur. It looks as if Anaximander were less interested in the end of a world than in its beginning. The one sentence of his which we possess (if indeed this first part of the sentence is his) has been commonly held to refer to it in the words: 'Things perish into that out of which they have their being': but in fact this obviously describes the transformation of the elements into one another, which, far from signifying the destruction of the world, is the process by which it is maintained.⁴⁸

What seems more relevant is the mention of a time when there will be no more water left, since fire, its opposite, will have prevailed completely and dried it all up.⁴⁹ This will clearly upset the balance of the universe which is maintained by the alternate and mutual encroachment of the opposites on each other, followed by their recession as 'penalty' for their 'injustice'. One cannot suppose this cyclic process, taking place as Anaximander says 'according to the ordinance (or assessment) of time', to be anything other than the annual alternation of the seasons. The permanent victory of the hot and dry would obviously disorganize the whole world-order. Cornford connected this possibility with the archaic idea of a 'great summer' and 'great winter', and assumed 'alternate destructions of the world by the Hot and by the Cold moisture'. Our world will be ultimately destroyed by fire, the next by flood.⁵⁰ This may have been what Anaximander meant, but if so, it is something different from reabsorption into the apeiron, and it is difficult to see how the Hot, having once been allowed to gain the supreme victory—or commit the supreme injustice—could ever be forced to give up its ill-gotten gains. A cosmos starts from a neutral state, not from an extreme. If that is not the thought from which Anaximander started out, which impressed on him the need for an apeiron as the arche rather than water or anything else, then we have indeed failed in our interpretation of him and there is little chance of success.

(7) Origin of animal and human life

After the formation of a world-order by the separation of the opposites, or elements, into their proper stations, the next stage is the emergence of animal life. This is explained with remarkable consistency (and complete disregard for religious or mythological modes of thought) as due to a continuation of the same process of 'separating-out' through the action of the hot and dry on the cold and moist: for life arose in the moist element through the action on it of the sun's warmth. This theory was probably connected with the persistent belief that even in the present world life is generated 'spontaneously' from the warmth of putrescent matter, a belief doubtless based on observation—'an observation', as Dr W. P. D. Wightman has remarked, 'which must have been only too familiar, though misinterpreted, in a warm climate'.⁵¹

The testimonies are as follows:

(a) Hippolytus, Ref. 1, 6, 6 (A11): 'He said that living creatures arose from the evaporation of the moist element by the sun; and that man originally resembled another creature, namely a fish.'

(b) Aëtius V, 19, 4 (A30): 'Anaximander said that the first animals were born in moisture and surrounded by prickly integuments, but that as they grew older they emerged on to the drier part, the integument split off, and they lived on⁵² for a short time.'

(c) [Plut.] Strom. (A10…): 'He says moreover that originally man was born from creatures of a different species, on the grounds that whereas other creatures quickly find food for themselves, man alone needs a long period of suckling; hence if he had been originally what he is now he could never have survived.'

The references to the origin of mankind are naturally of particular interest. So far we have nothing inconsistent with the supposition that Anaximander was describing its gradual evolution, on Darwinian lines, from some marine species. Indeed the statement of Hippolytus, that man 'originally resembled another creature, namely a fish', would, by itself, hardly allow a different interpretation. Yet this does not seem to have been in fact what he meant. Plutarch in his Quaestiones Conviviales (730E, A30) says that at first men were born in fish, and makes this meaning clearer by contrasting it with the more plausible view that they are related to them. The guests are discussing the custom of abstaining from fish on religious grounds. One of them mentions examples of people who do this because they worship Poseidon as Fosterer and Ancestor, believing, like the Syrians, that man arose from the wet element. 'For this reason,' he continues, 'they reverence the fish as kindred and foster-brother, displaying a more reasonable philosophy than Anaximander; for he does not class fish and men together, but declares that men were first born in fish, and having been nurtured in the manner of galei and become capable of looking after themselves, they emerged and occupied the land. And so just as fire devours the matter in which it was kindled and which is father and mother to it (as the writer said who interpolated the wedding of Ceyx in Hesiod), so Anaximander, having shown the fish to be the common father and mother of men, put us off eating it.'⁵³

The Latin writer Censorinus gives an even clearer account to the same effect (IV, 7, A30):

Anaximander of Miletus said that in his opinion there arose out of water and earth, when warmed, either fish or creatures resembling fish. In these creatures men were formed, and the young were retained within until the time of puberty; then at last the creatures were broken open and men and women emerged already capable of finding their own nourishment.

The theory of Anaximander seems then to have been that human embryos grew inside the bodies of the early fish-like creatures, and later emerged as fully-formed men and women. His account proceeds in the first place by deduction from the hypothesis that all life had its origin in moist slime acted on by the heat of the sun, this being in its turn only a particular stage in the evolution of the cosmos by the interaction of the opposites. It would acquire seeming confirmation either from observation or from the lore of Egyptians or Orientals. The first living creatures must therefore have been of a kind suited to a moist habitat, perhaps rather like the prickly sea-urchin. A human infant could hardly have survived under these conditions unless some special protection were devised, and here the example of the galeus came to his mind as a possible solution. This name was applied to dogfish or sharks, and Plutarch, commenting on the parental affection of galei, says (De Soll. Anim. 982C): 'They produce an egg, and then the creature itself, not outside, but within their own bodies, and nurse it there and carry it as if there had been a second birth. Then when they have grown larger they put them forth'; and more clearly in De Amove Prolis 494C: 'The galei in particular reproduce viviparously, and allow their young to issue forth and feed, then take them back and enfold them in the womb to rest.'

The species that Plutarch has in mind is no doubt the smooth dogfish (mustelus levis…) a viviparous variety which forms 'the subject of one of Aristotle's most celebrated descriptions, and a famous example of his anatomical erudition'.⁵⁴ Aristotle (HA, 565b1) refers to the remarkable peculiarity that 'the young develop with the navel-string attached to the womb, so that, as the egg-substance gets used up, the embryo is sustained, to all appearances, just as in the case of quadrupeds. The navel-string is long, and adheres to the under part of the womb (each navel-string being attached as it were by a sucker), and also to the middle of the embryo where the liver lies.' He also associates himself with the common belief that 'galei in general can extrude their young and take them back again' (565b24), a belief which persisted in the middle ages. Burnet (EGP, 71, n. 2) thinks that Anaximander's comparison is sufficiently accounted for by the anatomical details of the placenta and umbilical cord, and that there is no need to associate him with the other belief. Much as one would like to discover such faithfulness to observed fact in the first youth of Greek natural philosophy, it seems hardly likely that Anaximander disowned a belief which was still seriously held by Aristotle, and which undoubtedly provides the best illustration for his purpose.⁵⁵

(8) Meteorology

Anaximander's reported views on meteorological phenomena provide further illustration of his principle of consistency, that events in the present world must be attributed to the continued operation of the same forces and processes that brought about its formation in the beginning. This is especially obvious in his explana tion of wind, which he regarded as a flow of air, or as air in motion.

(a) Aëtius (A24): 'Wind is a flow of air, occurring when the finest [and most moist] elements in it are set in motion [or liquefied] by the sun.' …

(b) Hippolytus, Ref. 1, 6, 7 (All): 'Winds are produced when the finest vapours of the air are separated off, and being gathered together are set in motion; rain out of the evaporation given off from the earth by the sun's action.'⁵⁶

As O. Gilbert remarked (Meteor. Theorien, 512), the brief note about Anaximander inserted by Aëtius in his section on winds seems to have conflated Theophrastus's reports of his explanation of winds on the one hand and rain on the other. Comparison with Hippolytus suggests that the cosmogonic process of apokrisis is still at work. After water had been separated from earth, the sun drew vapour up from the water to form the atmosphere. This in its turn, as the 'separating-out' continues, divides into two substances, a lighter (finer, drier) and a heavier (wetter). The former is set in motion as wind, the latter precipitated as rain. It is all part of the same operation of peripheral heat on the moist centre which in due course was responsible for the emergence of life.⁵⁷

Once the air has been separated into wind (the light and dry part) and rain-cloud (the heavy and wet), these, and in particular the wind, are made to account for thunder and lightning. Thus Aëtius (A23), in his section on thunder, lightning, meteorites, waterspouts and whirlwinds:

Anaximander says that all these are caused by wind. When it is imprisoned in thick cloud and forces a way out by reason of its fine texture and lightness, then the tearing makes the noise, and the contrast with the blackness of the cloud produces the flash.⁵⁸

It would appear that, in the process of 'separating-out' of the air into wind and cloud, some of the lighter and finer sort may find itself so completely surrounded by the denser that it cannot easily complete the process of 'gathering together' with its like. The result is a violent explosion of the cloud, perceived by us as thunder and lightning.

Notes

¹ On the date of Anaximander cf. Heidel in Proc. Am. Ac. 1921, 253f.

² Hippocr. De Vet. Med. 20 (1,620 Littré), Plato, Lys. 214B, Phaedo, 96A, Eur. fr. 910 Nauck, Xen. Mem. I, I, 11, Ar. Gen. et Corr. 333b18, Phys. 185a18 (quoted by Verdenius, loc. cit. below). On the strength of some of these, Heidel (Proc. Am. Acad XLV (1910), 81) said: 'It is reasonably certain that philosophical works were familiarly quoted as bearing the title [On Nature] some time before the close of the fifth century', and Verdenius (Mnemos. 1947, 272—3): 'In the fifth and fourth centuries [On Nature] was obviously regarded as the authentic title of early philosophical works.'

³ Sext. Adv. Math. VII, 65 ([Diels-Kranz, Fragmente der Vorsokratiker; hereafter DK], 82B3)

⁴ For supporting passages in the Greek geographical tradition (D.L. II, 2, Agathemerus in DK, 12A6, etc.) see Heidel, op. cit. 247; and for conjectural details about the nature of Anaximander's map, Kahn, Anaximander, 82-4.

⁵ Cicero denies that this was an act of divination, and compares it to the forecasts made by doctors, seamen and farmers by reason of their special skill and experience, calling Anaximander 'physicus'. It would be interesting to know how Anaximander did it: perhaps by observing the behaviour of the storks, like the inhabitants of the Larissa neighbourhood in the earth-quakes of 1954. (The Times, 3 May 1954: "We have watched the storks all day; it is the best way to know when it is coming.')

⁶ Burnet had no doubts ([Early Greek Philosophy, hereafter EGP] 52), but see now W. Darsow in Jb. D.A.I. 1954, 101-17: the statue, it appears, is female, and the name must be that of the donor or dedicator.

⁷ In spite of McDiarmid, Theophr. on Presoc. Causes, 96ff. McDiarmid is of course right in saying that 'recompense to one another for injustice' can have nothing to do with the relation of generated things to the apeiron.

⁸ … Aristotle here groups Anaximander's 'boundless' with the 'mixture' of Empedocles and Anaxagoras, whose conceptions were in fact different, since they represented conscious attempts to escape the dilemma posed by Parmenides. After him, philosophers were conscious of distinctions and difficulties of which Anaximander had no inkling….

⁹ Cf. Heidel, [Classical Philology; hereafter CP], 1912, 215-16. McDiarmid (Theophr. on the Presoc. Causes, 138ff.) has cast legitimate doubt on this final point…. See also on this point Jaeger, [Theology of the Early Greek Philosophers; hereafter TEGP], 26f., Kahn, Anaximander, 30-2.

¹⁰The World-View of Physics, 30.

¹¹ Simpl. 150. 24, DK, A9. We may note that although by Plato's time the abstract nouns 'heat' or 'dryness' are currently distinguished from 'hot' and 'dry', he still has to apologize for the general term 'quality' … as an uncouth neologism (Theaet. 182A).

I cannot agree with the reasoning of Hölscher (following Reinhardt; see Hermes, 1953, 266) that the opposites enumerated as Anaximander's by Simplicius are not 'anaximandrisch' …. 'Because', says Hölscher, 'it stands not for a quality (like the hot), but a phenomenon like fire.' But the hot was also for Anaximander a material phenomenon.

¹² So Heidel, 'On Anaximander', CP, 1912, 233-4, and Proc. Am. Acad. 1913, 684-5; Vlastos, CP, 1947, 172; Cornford, Princ. Sap. 168.

¹³ Vlastos (CP, 1947, 170) thinks the plural … 'is strange, for the reference is obviously to the Boundless', but concludes that 'the Boundless is explicitly thought of as a plurality'. This is much less probable than that the reference is not to the Boundless at all. The view of H. Fränkel (Dichtung u. Philos. 345-7) is subtle and interesting, but as Woodbury says (CP, 1955, 154f.) it credits Anaximander with a more developed sense of the distinction between possible and actual than he is likely to have possessed. The view here taken is now supported by Kahn, Anaximander, 167f., 195f.

¹⁴ Another meaning of arche, as Heidel has illustrated in CP, 1912, 219ff.

¹⁵Od. x, 303, the 'bodily form' of a plant. See Kahn, Anaximander, 4, n. 1 and 201, n. 2.

¹⁶ 'Nomos und Physis', Hermes, 1953, 426. For a good discussion of the meaning of the word see [G.S. Kirk, Heraclitus: the Cosmic Fragments; hereafter HCF] 42-3, 228-31.

¹⁷ Very possibly at this stage with a limiting genitive … though Heraclitus (fr. 123 DK) already uses it absolutely.

¹⁸ For reasons in favour of supposing that the whole argument goes back to Anaximander himself see C. H. Kahn in Festschr. Kapp, 1958, 19-29.

¹⁹Phys. 203b20….

²⁰ What follows goes against the opinion of Burnet (EGP, 57), Cornford (Princ. Sap. 173), [H.] Cherniss ([Aristotle's Criticism of Presocratic Philosophy; hereafter ACP] 379) and others.

²¹ Eur. Or, 25 (cf. Aesch. Ag. 1382), Aesch. fr. 379 Nauck. (These and other examples are cited by Cornford, Princ. Sap. 176f.)

²² The fact that this is an ancient formula, going back beyond the beginning of philosophy, is our best guarantee that in calling the earliest philosophers monistic in intention we are not (as some modern interpreters have argued) foisting on them the misconceptions that we have absorbed from Aristotle….

²³ It is no wonder that later writers, both ancient and modern, have been puzzled to know whether Anaximander's apeiron is a single substance or a mixture. (Cf. Cherniss, ACP, 375 ff., McDiarmid, Theophr. on Presoc. Causes, 100.) Probably the explanation given above comes closer to the mind of Anaximander than an outright denial of Aristotle's supposition that the opposites were in the apeiron, which was therefore a mixture. He had not faced the question. The distinction which some have emphasized between separating out and separating off … seems to me of little significance in this connexion. (For Hölscher's contrary view see [G. S. Kirk and J. E. Raven, The Presocratic Philosophers; hereafter KR] 130).

Perhaps the explanation which shows most insight is that of Kahn (Anaximander, 236)….

²⁴ Cf. Jaeger, TEGP, 29 ff.

²⁵ It must be stated in fairness that Prof. G. Vlastos has written ([Philosophical Quarterly; hereafter PQ] 1952, 113): 'There is no good conclusive evidence that either Anaximander or Anaxagoras called their cosmogonic principle "god" or even "divine".' I can only say that for me the evidence of Aristotle makes it much more probable than not….

But whether or not Anaximander called his principle 'divine', it is of course true and important (and this is Vlastos's main point) that it had nothing whatever to do with the gods or cults of popular religion.

²⁶ Aristotle is in the context drawing a distinction, from his own point of view, between two kinds of early physical theory, those involving an alteration in the nature of the primitive stuff…, and those—of which Anaximander's was the first—which speak only of a separating-out of what was there all the time. Thales he leaves out of the account, probably on the grounds that too little was known about him. Following him Simplicius says (Phys. 150.20): 'Another way is not to adduce a change of matter as the cause, nor to account for the generation of things by the alteration of the substratum, but by separation…. Thus Anaximander says that the opposites were in the substratum, which was an indeterminate … body, and are separated out.' …

²⁷ Or 'the'; but cf. Diels, Dox. 579, crit. n.

²⁸ Perhaps the simile is intended to compare the breathing-holes to the holes in a (musical) pipe. This would be appropriate, but cannot be said to be a certain translation of the Greek.

²⁹ [Classical Quarterly; hereafter CQ] 1932, 29f. There is admittedly an element of speculation in this, and for a more cautious view the reader is referred to Kirk in KR, 132f., but I should certainly not go further in that direction than to agree with Kahn that though the phrasing may be more recent, nevertheless the idea is old (Anaximander, 57).

³⁰ The word meant sometimes the soft inner rind rather than the outer bark…. In Hellenistic times Nicander uses it for the skin of Marsyas (Al. 302) and of serpents (Th. 355, 392).

³¹ A11 (Hippolytus), 12 (Hermeias).

³² I omit (a) the passage from Achilles (DK, A21), as being obviously an unintelligently garbled version of what is described more clearly by Aëtius, (b) Aët. 11, 16, 5 (A18), which as Kahn has seen (Anaximander, 59) is only an accidental repetition of the preceding reference to Aristotle.

³³ Though Dreyer (Planetary Systems, 15, n. 1) would take the text as it stands.

³⁴ The last sentence, which occurs in Stobaeus but not in Plutarch's Epitome (Dox. 355), is obscure (and perhaps corrupt: Kahn, Anaximander, 60), but cannot be held to be a valid contradiction of the next passage quoted.

³⁵ This must be preferred to the statement of D.L. 11, 1 (DK, Al) that it gets its light from the sun. The correct view was in later antiquity attributed even to Thales…, and also to Anaximenes, in whose somewhat fantastic astronomy it can scarcely have found a place. It seems to be first clearly attested in Parmenides (fr. 14), but Health (Aristarchus, 75 f.) is sceptical about this line and would credit the discovery to Anaxagoras….

³⁶ For Burnet's suggestion that there is only one 'wheel of the stars', and that it is intended to explain the motions of the morning and evening stars alone (not yet recognized as one), see EGP 69 and Taylor, Timaeus, 160, n. 1. Even though this would explain why the 'wheel of the stars' was smaller than those of sun and moon, it does not seem to be supported by our texts.

³⁷ See on this point Health, Aristarchus, 31, Burnet, EGP, 68, Kahn, Anaximander, 89f. Burnet suggests, referring to Homer, that in early Greek thought aer could be seen through, although it had the property of rendering invisible anything enclosed in it. Dreyer (Planetary Systems, 14) remarks that astronomical observation must have been still so backward that Anaximander had never noticed the frequent occultation of a bright star by the moon. According to the doxography (D.L. IX, 33), Leucippus also placed the path of the sun furthest from the earth, but with the stars between it and the moon. A single statement in the Placita (DK, 28 A 40 a) seems to credit Parmenides with having placed the fixed stars nearest the earth….

³⁸ So, e.g., Tannery, Burnet, Health, Cornford. In a predominantly sceptical period of scholarship, there is some pleasure in recording the contrary view of R. Baccou (Hist, de la sc. gr. 77): 'Quelle impossibilité y at-il à imaginer qu'il a mesuré, de manière plus ou moins approximative, l'angle du diamètre apparent du soleil, et que, d'après l'idée qu'il se faisait de la grandeur de la terre—idée naturellement restreinte à l'oikumene—il en a déduit les chiffres plus haut cités?' H. Gomperz, in an interesting discussion of the various types of analogy employed by the Presocratics, connects it rather with the sense of fitness and proportion exhibited by a Greek architect or planner in designing a city or a temple (Journ. Hist. Ideas, 1943, 166-7). Cf. also W. I. Maison, Rev. Metaph. 1954-5, 447: 'Prima facie, however, we have here an early example of the insistence that Nature must conform to reason, i.e. a sort of embryonic metaphysics of the mathematizing sort. One is presumably supposed to accept the figures because of their inherent reasonableness (v. the Pythagorean harmonies)…. Moreover, we must not overlook the fact that these figures occur in the context of an astonishingly rational account of the nature of things, which is by no means devoid of references to observation, as Cornford admits.' (The reference is to Cornford, Princ. Sap. 165 and 170.) Kahn (Anaximander, 94-7) emphasizes the rational element in Anaximander's scheme.

³⁹ The question is discussed by G. B. Burch in an article on Anaximander (Rev. Metaph. 1949-50, 137-60), though not all of his ideas are acceptable.

⁴⁰ For which see Heath, Aristarchus, 32 ff. Heidel (CP, 1912, 233, n. 4) thought it very probable that the 'ordinance of time' in the one extant fragment of Anaximander refers to the obliquity of the ecliptic, which, he says, Anaximander is said to have discovered. He notes how well this would fit with the designation of the litigants as the opposites—hot and cold, wet and dry.

⁴¹ And to a parallel phenomenon of the moon, of which, however, Zeller considered that it was most unlikely that Anaximander would have been aware. Dreyer Planetary Systems, 17, n. 1) disagrees.

⁴² Arist. Meteor. 353b8 (quoted above, p. 92), 355 a 25. In the latter passage Zeller pointed out that according to the most natural meaning … Aristotle is speaking of the 'turnings' of the heaven, not of the sun. ([E. Zeller, Die Philosophie der Griechen, ed. W. Nestle; hereafter ZN], 298, n. 4, Health, Aristarchus, 33, n. 3.) For the contrary view see Chemiss, ACP, 135, n. 544.

⁴³ Cherniss, op. cit. 135, n. 544, disagrees, mainly because Meteorol. 355 a 24-5 'shows definitely that it is air and not moisture which causes the turnings'….

⁴⁴ This translation depends on several corrections of the received text, for which see Diels's apparatus, and cf. Cornford, Princ. Sap. 166, n. 2.

⁴⁵ So Burnet and Stocks translate … ; see Stocks's note ad loc. in the Oxford translation. The context makes the meaning clear.

⁴⁶ It is interesting that the Babylonian map illustrated by Kahn as a probable prototype of Anaximander's (Anaximander, pl. 1) not only shows the world as circular but has a round hole in the middle. This is explained as 'probably left by the scribe's compass', but only because 'there is at any rate no other good explanation' (op. cit. 83).

⁴⁷ ZN, 303, Alex. ap. Simpl. De Caelo, 532. 6ff. Burnet's contrary view (EGP, 66) is bound up with certain other preconceptions which are not necessarily correct. We need not avail ourselves of the emendation of Montucla….

⁴⁸ Heidel saw this, CP, 1912, 233-4. As to the destruction of the world, Heidel says (234, n. 3): 'No doubt Anaximander believed in the destruction of the world, and so of the opposites also; but he doubtless thought of this as a question of nutrition.' This is very possible, but we are not told.

⁴⁹ Ar. Meteor. 353b9….

⁵⁰Princ. Sap. 183f. Certainly, as Cornford says, 'the notion of alternate destruction of at least a great part of mankind by fire and flood was deeply rooted in Greek thought'….

⁵¹Growth of Scientific Ideas, 14. Spontaneous generation seemed an incontrovertible fact to Aristotle (unfortunately for him, since it made an awakward exception to his general theory of the workings of nature), and the belief lingered on in Europe until the nineteenth century. See Guthrie, In the Beginning, 41f. J. A. Wilson in Before Philosophy, 59, says that the modern Egyptian peasant still believes in the life-giving power of the mud left behind by the retreating Nile. (Both these last writers quote further illustrations of the belief.)

⁵² Or 'lived a different life' (i.e. on land). See KR, 141, 142.

⁵³ The last sentence is troublesome. Its logic seems to require, if Anaximander acted 'just like' the fire, that he did eat fish, or approve of eating it. This would also be a satisfactory reason why his philosophy was less … ('humane') than that of the Syrians and others. Plutarch no doubt knew nothing of Anaximander's actual habits of diet. But again, if this were so, he would be more likely to assume that like most ordinary men he ate fish than that he preached an abstention for which there is no other evidence at all. Yet the negative sense … seems undoubted, however much one would like it to mean 'he mistreated as food'…. If the text is sound, it must be intended to convey that Anaximander deprecated the eating of fish because it resembled the action of fire in devouring parents, and the 'unreasonableness' of his philosophy consists simply in the fact that he justified the ban by his queer idea of men coming out of fish …. But if so, it is not very well expressed.

⁵⁴ D'Arcy Thompson, Glossary of Greek Fishes, 41….

⁵⁵ After all this discussion, it must be pointed out that the appearance of the [galei] in Plutarch's reference to Anaximander depends on an emendation of the MS. text…. Kirk however believes (KR, 142) that the comparison may not be Anaximander's, but put in by Plutarch as throwing light on Anaximander's theory. This is of course possible, but I do not agree with Kirk that the knowledge which it displays is 'unlikely' for Anaximander. Inhabitants of an ancient seaport probably knew more about the facts of life among fishes than do the unscientific among ourselves.

⁵⁶ Reading uncertain….

⁵⁷ The theory bears a superficial resemblance to Aristotle's, and might therefore come under suspicion of having been brought into conformity by our sources under Peripatetic influence. Starting from his assumption of two sorts of exhalation, a dry and a wet, Aristotle continues (Meteor, 360a11): 'Of these the exhalation containing the greater quantity of moisture is the origin of rainwater, whereas the dry one is the origin and substance of winds.' But he goes on to emphasize that, since the two exhalations are specifically different, the natural substances of wind and rain are also different, and from that to criticize those who claim that the same substance, air, becomes wind when set in motion … and rain when condensed…. This was inevitably the view of the monist Anaximander….

⁵⁸ Briefer statements are found in Hippolytus (A11) and Seneca (A23). Kahn (Anaximander, 108) has pointed out how authentically this theory is reproduced by Aristophanes in the Clouds (404-7).