Anaximenes of Miletos, son of Eurystratos, was, according to Theophrastos, an “associate” of Anaximander.104 Apollodoros said, it appears, that he “flourished” about the time of the fall of Sardeis (546/5 B.C.), and died in Ol. LXIII. (528/525 B.C.).105 In other words, he was born when Thales “flourished,” and “flourished” when Thales died, and this means that Apollodoros had no definite information about his date. He perhaps made him die in the sixty-third Olympiad because that gives just three generations for the Milesian school.106 We cannot therefore say anything positive as to his date, except that he must have been younger than Anaximander.

Anaximenes wrote a book which survived until the age of literary criticism; for we are told that he used a simple and unpretentious Ionic,107 very different, we may suppose, from the poetical prose of Anaximander.108 The speculations of Anaximander were distinguished for their hardihood and breadth; those of Anaximenes are marked by the opposite quality. He appears to have thought out his system carefully, but he rejects the more audacious theories of his predecessor. The result is that, while his view of the world is less like the truth than Anaximander's, it is perhaps more fruitful in ideas that were destined to hold their ground.

Anaximenes is one of the philosophers on whom Theophrastos wrote a special monograph;109 and this gives us an additional guarantee for the trustworthiness of the tradition. The following110 are the passages which contain the fullest account of the central feature of his system:

Anaximenes of Miletos, son of Eurystratos, who had been an associate of Anaximander, said, like him, that the underlying substance was one and infinite. He did not, however, say it was indeterminate, like Anaximander, but determinate; for he said it was Air.—Phys. Op. fr. 2 (R. P. 26).

From it, he said, the things that are, and have been, and shall be, the gods and things divine, took their rise, while other things come from its offspring.—Hipp. Ref. i. 7 (R. P. 28).

“Just as,” he said, “our soul, being air, holds us together, so do breath and air encompass the whole world.”—Aet. i. 3, 4 (R. P. 24).

And the form of the air is as follows. Where it is most even, it is invisible to our sight; but cold and heat, moisture and motion, make it visible. It is always in motion; for, if it were not, it would not change so much as it does.—Hipp. Ref. i. 7 (R. P. 28).

It differs in different substances in virtue of its rarefaction and condensation.—Phys. Op. fr. 2 (R. P. 26).

When it is dilated so as to be rarer, it becomes fire; while winds, on the other hand, are condensed Air. Cloud is formed from Air by felting;111 and this, still further condensed, becomes water. Water, condensed still more, turns to earth; and when condensed as much as it can be, to stones.—Hipp. Ref. i. 7 (R. P. 28).

At first, this looks like a falling off from the more refined doctrine of Anaximander to a cruder view; but this is not really the case. On the contrary, the introduction of rarefaction and condensation into the theory is a notable advance.112 In fact, it makes the Milesian cosmology consistent for the first time; since a theory which explains everything as a form of a single substance is clearly bound to regard all differences as quantitative. The only way to save the unity of the primary substance is to say that all diversities are due to the presence of more or less of it in a given space. And when once this step has been taken, it is no longer necessary to make the primary substance something “distinct from the elements,” to use Aristotle's inaccurate but convenient phrase; it may just as well be one of them.

The air Anaximenes speaks of includes a good deal that we should not call by the name. In its normal condition, when most evenly distributed, it is invisible, and it then corresponds to our “air”; it is the breath we inhale and the wind that blows. That is why he called it πνεῦμα. On the other hand, the old idea that mist or vapour is condensed air, is still accepted without question. It was Empedokles, we shall see, who first discovered that what we call air was a distinct corporeal substance, and not identical either with vapour or with empty space. In the earlier cosmologists “air” is always a form of vapour, and even darkness is a form of “air.” It was Empedokles who cleared up this point too by showing that darkness is a shadow.113

It was natural for Anaximenes to fix upon “air” as the primary substance; for, in the system of Anaximander, it occupied an intermediate place between the two fundamental opposites, the ring of flame and the cold, moist mass within it (§ 19). We know from Plutarch that he fancied air became warmer when rarefied, and colder when condensed. Of this he satisfied himself by a curious experimental proof. When we breathe with our mouths open, the air is warm; when our lips are closed, it is cold.114

This argument brings us to an important point in the theory, which is attested by the single fragment that has come down to us.115 “Just as our soul, being air, holds us together, so do breath and air encompass the whole world.” The primary substance bears the same relation to the life of the world as to that of man. Now this was the Pythagorean view;116 and it is also an early instance of the argument from the microcosm to the macrocosm, and so marks the beginning of an interest in physiological matters.

We turn now to the doxographical tradition concerning the formation of the world and its parts:

He says that, as the air was felted, the earth first came into being. It is very broad and is accordingly supported by the air.— Ps.-Plut. Strom. fr. 3 (R. P. 25).

In the same way the sun and the moon and the other heavenly bodies, which are of a fiery nature, are supported by the air because of their breadth. The heavenly bodies were produced from the earth by moisture rising from it. When this is rarefied, fire comes into being, and the stars are composed of the fire thus raised aloft. There were also bodies of earthy substance in the region of the stars, revolving along with them. And he says that the heavenly bodies do not move under the earth, as others suppose, but round it, as a cap turns round our head. The sun is hidden from sight, not because it goes under the earth, but because it is concealed by the higher parts of the earth, and because its distance from us becomes greater. The stars give no heat because of the greatness of their distance.—Hipp. Ref. i. 7, 4-6 (R. P. 28).

Winds are produced when air is condensed and rushes along under propulsion; but when it is concentrated and thickened still more, clouds are generated; and, lastly, it turns to water.117 -Hipp. Ref. i. 7, 7 (Dox. p. 561).

The stars [are fixed like nails in the crystalline vault of the heavens, but some say they] are fiery leaves, like paintings.118—Aet. ii. 14, 3 (Dox. p. 344).

They do not go under the earth, but turn round it.—Ib. 16, 6 (Dox. p. 348).

The sun is fiery.—Ib. 20, 2 (Dox. p. 348).

It is broad like a leaf.—Ib. 22, 1 (Dox. p. 352).

The heavenly bodies turn back in their courses119 owing to the resistance of compressed air.—Ib. 23, 1 (Dox. p. 352).

The moon is of fire.—Ib. 25, 2 (Dox. p. 356).

Anaximenes explained lightning like Anaximander, adding as an illustration what happens in the case of the sea, which flashes when divided by the oars—Ib. iii. 3, 2 (Dox. p. 368).

Hail is produced when water freezes in falling; snow, when there is some air imprisoned in the water.—Aet. iii. 4, 1 (Dox. p. 370).

The rainbow is produced when the beams of the sun fall on thick condensed air. Hence the anterior part of it seems red, being burnt by the sun's rays, while the other part is dark, owing to the predominance of moisture. And he says that a rainbow is produced at night by the moon, but not often, because there is not constantly a full moon, and because the moon's light is weaker than that of the sun.—Schol,. Arat.120 (Dox. p. 231).

The earth was like a table in shape.—Aet. iii. 10, 3 (Dox. p. 377).

The cause of earthquakes was the dryness and moisture of the earth, occasioned by droughts and heavy rains respectively. —Ib. 15, 3 (Dox. p. 379).

We have seen that Anaximenes was justified in going back to Thales in regard to the nature of primary substance; but the effect upon the details of his cosmology was unfortunate. The earth is once more imagined as a table-like disc floating on the air. The sun, moon, and stars are also fiery discs which float on the air “like leaves”; an idea naturally suggested by the “eddy” (δίνη). It follows that the heavenly bodies cannot go under the earth at night, as Anaximander must have held, but only round it laterally like a cap or a millstone.121 This view is also mentioned in Aristotle's Meteorology,122 where the elevation of the northern parts of the earth, which makes it possible for the heavenly bodies to be hidden from sight, is referred to. This is only meant to explain why the stars outside the Arctic circle appear to rise and set, and the explanation is fairly adequate if we remember that the world is regarded as rotating in a plane. It is quite inconsistent with the theory of a celestial sphere.123

The earthy bodies, which circulate among the planets, are doubtless intended to account for eclipses and the phases of the moon.124

As might be expected, there is much the same difficulty about the “innumerable worlds” ascribed to Anaximenes as there is about those of Anaximander. The evidence, however, is far less satisfactory. Cicero says that Anaximenes regarded air as a god, and adds that it came into being.125 That cannot be right. Air, as the primary substance, is certainly eternal, and it is quite likely that Anaximenes called it “divine,” as Anaximander did the Boundless; but it is certain that he also spoke of gods who came into being and passed away. These arose, he said, from the air. This is expressly stated by Hippolytos,126 and also by St. Augustine.127 These gods are probably to be explained like Anaximander's. Simplicius, indeed, takes another view; but he may have been misled by a Stoic authority.128

It is not easy for us to realise that, in the eyes of his contemporaries, and for long after, Anaximenes was a much more important figure than Anaximander. And yet the fact is certain. We shall see that Pythagoras, though he followed Anaximander in his account of the heavenly bodies, was far more indebted to Anaximenes for his general theory of the world (§ 53). We shall see further that when, at a later date, science revived once more in Ionia, it was “the philosophy of Anaximenes” to which it attached itself (§ 122). Anaxagoras adopted many of his most characteristic views (§ 135), and so did the Atomists.129 Diogenes of Apollonia went back to the central doctrine of Anaximenes, and made Air the primary substance, though he also tried to combine it with the theories of Anaxagoras (§ 188). We shall come to all this later; but it seemed desirable to point out at once that Anaximenes marks the culminating point of the line of thought which started with Thales, and to show how the “philosophy of Anaximenes” came to mean the Milesian doctrine as a whole. This it can only have done because it was really the work of a school, of which Anaximenes was the last distinguished representative, and because his contribution to it was one that completed the system he had inherited from his predecessors. That the theory of rarefaction and condensation was really such a completion of the Milesian system, we have seen (§ 26), and it need only be added that a clear realisation of this fact will be the best clue at once to the understanding of the Milesian cosmology itself and to that of the systems which followed it. In the main, it is from Anaximenes they all start.

104. Theophr. Phys. Op. fr. 2 (R. P. 26).

105. This follows from a comparison of Diog. ii. 3 with Hipp. Ref. i. 7 (R. P. 23) and Souidas (sv.). In Hippolytos we must, however, read τρίτον for πρῶτον with Diels. The suggestion in R. P. 23 a that Apollodoros mentioned the Olympiad without giving the number of the year is inadequate; for Apollodoros did not reckon by Olympiads, but Athenian archons.

106. Jacoby (p. 194) brings the date into connexion with the floruit of Pythagoras, which seems to me less probable.

107. Diog. ii. 3 (R. P. 23).

108. Cf. the statement of Theophrastos above, § 13.

109. On these monographs, see Dox. p. 103.

110. See the conspectus of extracts from Theophrastos given in Dox. p. 135.

111. “Felting” (πίλησις) is the regular term for this process with all the early cosmologists, from whom Plato has taken it (Tim. 58 b 4; 76 c 3).

112. Simplicius, Phys. p. 149, 32 (R. P. 26 b), says that Theophrastos spoke of rarefaction and condensation in the case of Anaximenes alone. It should be noted, however, that Aristotle, Phys. A, 4. 187 a 12, seems to imply that Anaximander too had spoken of rarefaction and condensation, especially if ὅ ἐστι πυρὸς μὲν πυκνότερον ἀέρος δὲ λεπτότερον is referred to him. On the other hand, at 20, οἱ δ' ἐκ τοῦ ἑνὸς ἐνούσας τὰς ἐναντιότητας ἐκκρίνεσθαι, ὥσπερ Ἀναξίμανδρός φησι seems to be opposed to a 12, οἱ μὲν κτλ. As I have indicated already, it looks as if we were dealing here with Aristotle's own inferences and interpretations, which are far from clear. They are outweighed by the definite statement quoted by Simplicius from Theophrastos, though Simplicius himself adds δῆλον δὲ ὡς καὶ οἱ ἄλλοι τῇ μανότητι καὶ πυκνότητι ἐχρῶντο. That, however, is only his own inference from Aristotle's somewhat confused statement.

113. For the meaning of ἀήρ in Homer, cf. e.g.. Od. viii. 1, ἠέρι καὶ νεφέλῃ κεκαλυμμέναι; and for its survival in Ionic prose, Hippokrates, Περὶ ἀέρων, ὑδάτων, τόπων, 15, ἀήρ τε πολὺς κατέχει τὴν χώρην ἀπὸ τῶν ὑδάτων.. Plato is still conscious of the old meaning; for he makes Timaios say ἀέρος (γένη) τὸ μὲν εὐαγέστατον ἐπίκλην αἰθὴρ καλούμενος, ὁ δὲ θολερώτατος ὁμίχλη καὶ σκότος (Tim. 58 d). For the identification of ἀήρ with darkness, cf. Plut. De prim. frig. 948 e, ὅτι δ' ἀὴρ τὸ πρώτως σκοτεινόν ἐστιν οὐδὲ τοὺς ποιητὰς λέληθεν· ἀέρα γὰρ τὸ σκότος καλοῦσιν. My view has been criticised by Tannery, “Une nouvelle hypothèse sur Anaximandre” (Arch. viii. pp. 443 sqq.), and I have slightly altered my expression of it to meet these criticisms. The point is of fundamental importance for the interpretation of Pythagoreanism.

114. Plut. De prim. frig. 947 f (R. P. 27), where we are told that he used the term τὸ χαλαρόν for the rarefied air.

115. Aet. i. 3, 4 (R. P. 24).

116. See Chap. II. § 53.

117. The text is very corrupt here. I retain ἐκπεπυκνωμένος, because we are told above that winds are condensed air.

118. See below, p. 77, n. 4.

119. This can only refer to the τροπαί of the sun, though it is loosely stated of τὰ ἄστρα generally. It occurs in the chapter Περὶ τροπῶν ἡλίου, and we cannot interpret it as if it were a detached statement.

120. The source of this is Poseidonios, who used Theophrastos. Dox. p. 231.

121. Theodoret (iv. 16) speaks of those who believe in a revolution like that of a millstone, as contrasted with one like that of a wheel. Diels (Dox. p. 46) refers these similes to Anaximenes and Anaximander respectively. They come, of course, from Aetios (Note on Sources, § 10), though they are given neither by Stobaios nor in the Placita.

122. B, 1. 354 a 28 (R. P. 28 c).

123. For this reason, I now reject the statement of Aetios, ii. 14, 3 (p. 76), Ἀναξιμένης ἥλων δίκην καταπεπηγέναι τῷ κρυσταλλοειδεῖ. That there is some confusion of names here is strongly suggested by the words which immediately follow, ἔνιοι δὲ πέταλα εἶναι πύρινα ὥσπερ τὰ ζωγραφήματα, which is surely the genuine doctrine of Anaximenes. I understand ζωγραφήματα of the constellations (cf. Plato, Tim. 55c). To regard the stars as fixed to a crystalline sphere is quite inconsistent with the far better attested doctrine that they do not go under the earth.

124. See Tannery, Science hellène, p. 153. For the precisely similar bodies assumed by Anaxagoras, see below, Chap. VI. § 135. See further Chap. VII. § 151.

125. Cic. De nat. d. i. 26 (R. P. 28 b).

126. Hipp. Ref. i. 7, 1 (R. P. 28).

127. Aug. De civ. D. viii. 2: “Anaximenes omnes rerum causas infinito aëri dedit: nec deos negavit aut tacuit; non tamen ab ipsis aërem factum, sed ipsos ex aëre ortos credidit” (R. P. 28 b).

128. Simpl. Phys. p. 1121, 12 (R. P. 28 a). The passage from the Placita is of higher authority than this from Simplicius. It is only to Anaximenes, Herakleitos, and Diogenes that successive worlds are ascribed even here. For the Stoic view of Herakleitos, see Chap. III. § 78; and for Diogenes, Chap.X. §188. That Simplicius is following a Stoic authority is suggested by the words καὶ ὕστερον οἱ ἀπὸ τῆς Στοᾶς.

129. In particular, both Leukippos and Demokritos adhered to his theory of a flat earth. Cf. Aet. iii. 10, 3-5 (Περὶ σχήματος γῆς), Ἀναξιμένης τραπεζοειδῆ (τὴν γῆν). Λεύκιππος τυμπανοειδῆ. Δημόκριτος δισκοειδῆ μὲν τῷ πλάτει, κοίλην δὲ τῷ μέσῳ. And yet the spherical form of the earth was already a commonplace in circles affected by Pythagoreanism.