From Chapter III., Herakleitos of Ephesos
78. Did Herakleitos Teach a General Conflagration?
Nor is there anything in Aristotle to set against Plato's statement. We have seen that the passage in which he speaks of him along with Empedokles as holding that the heavens were alternately in one condition and in another refers not to the world, but to fire, which Aristotle identified with the substance of his own "first heaven."87 It is also quite consistent with our interpretation when he says that all things at one time or another become fire. This need not mean that they all become fire at the same time, but may be merely a statement of the undoubted Herakleitean doctrine of the upward and downward path.88
The earliest statements to the effect that Herakleitos taught the doctrine of a general conflagration are found in Stoic writers. The Christian apologists too were interested in the idea of a final conflagration, and reproduce the Stoic view. The curious thing, however, is that there was a difference of opinion on the subject even among the Stoics. In one place, Marcus Aurelius says: "So that all these things are taken up into the Reason of the universe, whether by a periodical conflagration or a renovation effected by eternal exchanges."89 Indeed, there were some who said there was no general conflagration at all in Herakleitos. "I hear all that," Plutarch makes one of his personages say, "from many people, and I see the Stoic conflagration spreading over the poems of Hesiod, just as it does over the writings of Herakleitos and the verses of Orpheus."90 We see from this that the question was debated, and we should therefore expect any statement of Herakleitos which could settle it to be quoted over and over again. It is highly significant that not a single quotation of the kind can be produced.91
On the contrary, the absence of anything to show that Herakleitos spoke of a general conflagration only becomes more patent when we turn to the few fragments which are supposed to prove it. The favourite is fr. 24, where we are told that Herakleitos said Fire was Want and Surfeit. That is just in his manner, and it has a perfectly intelligible meaning on our interpretation, which is further confirmed by fr. 36. The next is fr. 26, where we read that fire in its advance will judge and convict all things. There is nothing in this, however, to suggest that fire will judge all things at once rather than in turn, and, indeed, the phraseology reminds us of the advance of fire and water which we have seen reason for attributing to Herakleitos, but which is expressly said to be limited to a certain maximum.92 These appear to be the only passages which the Stoics and the Christian apologists could discover, and, whether our interpretation of them is right or wrong, it is surely clear that they cannot bear the weight of their conclusion, and that there was nothing more definite to be found.
It is much easier to find fragments which are inconsistent with a general conflagration. The "measures" of fr. 20 and fr. 29 must be the same thing, and they must be interpreted in the light of fr. 23. If this be so, fr. 20, and more especially fr. 29, directly contradict the idea of a general conflagration. "The sun will not overstep his measures."93 Secondly, the metaphor of "exchange," which is applied to the transformations of fire in fr. 22, points in the same direction. When gold is given in exchange for wares and wares for gold, the sum or "measure" of each remains constant, though they change owners. All the wares and gold do not come into the same hands. In the same way, when anything becomes fire, something of equal amount must cease to be fire, if the "exchange" is to be a just one; and that it will be just, we are assured by the watchfulness of the Erinyes (fr. 29), who see to it that the sun does not take more than he gives. Of course there is a certain variation, as we saw; but it is strictly confined within limits, and is compensated in the long run by a variation in the other direction. Thirdly, fr. 43, in which Herakleitos blames Homer for desiring the cessation of strife, is very conclusive. The cessation of strife, would mean that all things should take the upward or downward path at the same time, and cease to "run in opposite directions." If they all took the upward path, we should have a general conflagration. Now, if Herakleitos had himself held this to be the appointment of fate, would he have been likely to upbraid Homer for desiring so necessary a consummation?94 Fourthly, we note that in fr. 20 it is this world,95 and not merely the "ever-living fire," which is said to be eternal; and it appears also that its eternity depends on the fact that it is always kindling and always going out in the same "measures," or that an encroachment in one direction is compensated by a subsequent encroachment in the other. Lastly, Lassalle's argument from the concluding sentence of the passage from the Περὶ διαίτης quoted above, is really untouched by Zeller's objection, that it cannot be Herakleitean because it implies that all things are fire and water. It does not imply this, but only that man, like the heavenly bodies, oscillates between fire and water; and that is just what Herakleitos taught. Now, in this passage we read that neither fire nor water can prevail completely, and a very good reason is given for this, a reason too which is in striking agreement with the other views of Herakleitos.96 And, indeed, it is not easy to see how, in accordance with these views, the world could ever recover from a general conflagration if such a thing were to take place. The whole process depends on the fact that Surfeit is also Want, or, in other words, that an advance of fire increases the moist exhalation, while an advance of water deprives the fire of its power to cause evaporation. The conflagration, though it lasted but for a moment,97 would destroy the opposite tension on which the rise of a new world depends, and then motion would become impossible.
85. Schleiermacher and Lassalle are notable exceptions. Zeller, Diels, and Gomperz are all positive that Herakleitos believed in the ἐκπύρωσις.
86. In his fifth edition (p. 699) Zeller seems to have felt this last difficulty; for he said there: "It is a contradiction which he, and which probably Plato too (und den wahrscheinlich auch Plato) has not observed." This seems to me still less arguable. Plato may or may not be mistaken; but he makes the perfectly definite statement that Herakleitos says ἀεί, while Empedokles says ἐν μέρει. The Ionian Muses are called συντονώτεραι and the Sicilian μαλακώτεραι just because the latter "lowered the pitch" (ἐχάλασαν) of the doctrine that this is always so (τὸ ἀεὶ ταῦτα οὕτως ἔχειν).
88. Phys. Γ, 205 a 3 (Met. K, 10. 1067 a 4), ὥσπερ Ἡράκλειτός φησιν ἅπαντα γίνεσθαί ποτε πῦρ. Zeller translates this es werde alles dereinst zu Feuer werden; but that would require γενήσεσθαι. Nor is there anything in his suggestion that ἅπαντα ("not merely πάντα") implies that all things become fire at once. In Aristotle's day, there was no distinction of meaning between πᾶς and ἅπας. Of course, as Diels says, the present tense might be used of a "constant alternation of epochs" (Vors. 12 A 10 n.); but for the purpose of Zeller's argument, we want something which not only may but must mean that.
89. Marcus Aurelius, x. 7, ὥστε καὶ ταῦτα ἀναληφθῆναι εἰς τὸν τοῦ ὅλου λόγον, εἴτε κατὰ περίοδον ἐκπυρουμένου, εἴτε ἀιδίοις ἀμοιβαῖς ἀνανεουμένου. The ἀμοιβαί are specifically Herakleitean, and the statement is the more remarkable as Marcus elsewhere follows the usual Stoic interpretation.
90. Plut. De def. orac. 415 f., καὶ ὁ Κλεόμβροτος, Ἀκούω ταῦτ' ἔφη, πολλῶν καὶ ὁρῶ τὴν Στωικὴν ἐκπύρωσιν, ὥσπερ τὰ Ἡρακλείτου καὶ Ὀρφέως ἐπινεμομένην ἔπη οὕτω καὶ τὰ Ἡσιόδου καὶ συνεξάπτουσαν. As Zeller admits (p. 693 n.), this proves that some opponents of the Stoic ἐκπύρωσις tried to withdraw the support of Herakleitos from it.
91. This has been called a mere argumentum ex silentio; but, in such cases, the argumentum ex silentio is stronger than any other. Positive statements may be misinterpreted; but, when we know that a subject was keenly debated, and when we find that neither party can produce an unambiguous text in support of its view, the conclusion that none such existed becomes irresistible. The same remark applies to modern pronouncements on the subject. Diels briefly says that my view "is wrong" (ist irrig), but he does not adduce any fresh reason for saying so. The conclusion is that he knows of none.
92. Περὶ διαίτης i.3 ἐν μέρει δὲ ἑκάτερον κρατεῖ καὶ κρατεῖται ἐς τὸ μήκιστον καὶ ἐλάχιστον ὡς ἀνυστόν.
94. This is just the argument which Plato uses in the Phaedo (72 c) to prove the necessity of ἀνταπόδοσις, and the whole series of arguments in that passage is distinctly Herakleitean in character.
95. However we understand κόσμος here, the meaning is the same. Indeed, if we suppose with Bernays that it means "order," the argument will be all the stronger. In no sense of the word could a κόσμος survive the ἐκπύρωσις, and the Stoics accordingly said the κόσμος was φθαρτός, though Herakleitos had declared it to be everlasting.
96. Περὶ διαίτης, i. 3 (see above, p. 150, n. 2), οὐδέτερον γὰρ κρατῆσαι παντελῶς δύναται διὰ τάδε· τό <τε> πῦρ ἐπεξιὸν ἐπὶ τὸ ἔσχατον τοῦ ὕδατος ἐπιλείπει ἡ τροφή· ἀποτρέπεται οὖν ὅθεν μέλλει τρέφεσθαι· τὸ ὕδωρ τε ἐπεξιὸν τοῦ πυρὸς ἐπὶ τὸ ἔσχατον, ἐπιλείπει ἡ κίνησις· ἵσταται οὖν ἐν τούτῳ, ὅταν δὲ στῇ, οὐκέτι ἐγκρατές ἐστιν, ἀλλ' ἤδη τῷ ἐμπίπτοντι πυρὶ ἐς τὴν τροφὴν καταναλίσκεται· οὐδέτερον δὲ διὰ ταῦτα δύναται κρατῆσαι παντελῶς, εἰ δέ ποτε κρατηθείη καὶ ὁπότερον, οὐδὲν ἂν εἴη τῶν νῦν ἐόντων ὥσπερ ἔχει νῦν· οὕτω δὲ ἐχόντων ἀεὶ ἔσται τὰ αὐτὰ καὶ οὐδέτερον οὐδαμὰ ἐπιλείψει.
97. In his note on fr. 66 (= 26 Byw.) Diels seeks to minimise the difficulty of the ἐκπύρωσις by saying that it is only a little one, and can last but a moment; but the contradiction remains. Diels holds that Herakleitos was "dark only in form," and that "he himself was perfectly clear as to the sense and scope of his ideas" (Herakleitos, p. i.). To which I would add that he was probably called "the Dark" just because the Stoics sometimes found it hard to read their own ideas into his words.
Created for Peithô's Web from Early Greek Philosophy by John Burnet, 3rd edition (1920). London: A & C Black Ltd. Burnet's footnotes have been converted to chapter endnotes. Greek unicode text entered with Peithô's Younicoder.
Web design by Larry Clark and RSBoyes (Agathon). Peithô's Web gratefully acknowledges the assistance of Anthony Beavers in the creation of this web edition of Burnet. Please send comments to: