There are other more general considerations which point to the same conclusion: for what sudden motive can we conceive which should make the Gods abandon their state of eternal repose, and set to the work of creation, and how, with no model before them, could they know what to make or how to make it; again, how can we possibly believe that any being should be powerful enough to administer, not to say to create, the infinity of nature? It is equally impossible to ascribe to the Gods such weakness and pettiness of mind as to feel anger or be propitiated with gifts, or to take a fussy interest in the affairs of men. They enjoy undisturbed tranquillity in some region far removed from our troubled world.

This tranquil region Epicurus found in the intermundia, the spaces between his countless worlds. He seems to have borrowed the suggestion from Aristotle, who transformed the heaven of the poets into the supra-celestial region where space and time are not, but 'where the things outside enjoy through all eternity a perfect life of absolute joy and peace'.' But the unchangeableness which belongs naturally to Aristotle's solitary world is altogether out of place in the countless perishable worlds of Epicurus. For successive worlds need not occupy the same point in space nor be made up of the same materials; new worlds are formed καὶ ἐν κόσμῳ καὶ ἐν μετακοσμίῳ, and their materials may have been either already made use of for the formation of a world or they may be floating loose in an intermundium3. Moreover, during the existence of each world, it is constantly either 1 Arist. De Caelo 1 9.

2 Diog. x 89.

receiving an accession of atoms from the intermundia or, in its later stages, giving them back again. It is plain therefore that Epicurus has failed to find a safe retreat for his Gods in the intermundia and that they are quite as much exposed to the metus ruinarum there as they would have been within the world'.

Again, the Gods, like every other existing thing, are made up of atoms and void; but every compound is liable to dissolution; how is this compatible with immortality? One answer given was that the destructive and conservative forces in the universe balance one another, but in this world the destructive forces have the upper hand, therefore elsewhere, probably in the intermundia, the conservative forces must prevail. Another reason was that the atoms of which the Gods are composed, were so fine and delicate as to evade the blows of the coarser atoms". This idea of the extreme tenuity of the divine corporeity was doubtless suggested partly by the Homeric description of the Gods 'who are bloodless and immortal' (II. v 340) and partly by the shadowy idola of the dead, which escape the grasp of their living friends. We find yet another reason assigned, not so much perhaps for the actual immortality of the Gods, as for our belief in it, in the alleged fact of an incessant stream of divine images (eldwλa), too subtle to impinge on the bodily senses, but

1 Compare Cic. Divin. 11 40, N. D. 1 18, 53, 114, Diog. x 89, Lucr. II 1105-1174.

2 Cic. N. D. I 50, with my note.

8 See Cic. N. D. 1 68—71, and the passage from Herculanensia, Vol. VI. pt. 2 p. 35, quoted in my note on § 71 'no object which is perceptible to the senses is immortal, for its density makes it liable to severe shocks.'.

perceptible by the kindred atoms of mind'. Evidently this incessant never-ending influx of divine images is not a thing which can be directly vouched for by any human experience. We are not directly conscious even of the stream of images. All that an Epicurean could say is that we seem from time to time to behold the same glorified form, and that there is some ground for supposing similar appearances during past ages; that we can only account for such appearances by the supposition of an uninterrupted succession of images continued from a very remote period. But this of course is no proof of immortality if it were so, we must a fortiori believe the immortality of the sun, or indeed, as the Ciceronian Cotta remarks (N. D. 1 109), of any common object, since our ordinary perceptions are due to such an uninterrupted stream of images. If it is said that we cannot help attributing in our thought a permanent unchanging existence. to the divine nature, and that this law of thought is only explicable, on the Epicurean hypothesis, by the supposition of an endless stream of images actuating our mind, then the belief in the divine immortality is made the

1 Lucretius (v 1161 foll.) describing how the belief in the gods originated in visions, tells us that they were thought to be immortal, partly because they seemed to be too mighty to be overcome by any force, and partly quia semper eorum subpeditabatur facies et forma manebat, one image constantly succeeded another giving the impression of a permanent form. There is a similar use of the verb suppedito in IV 776, (where he explains the apparent movements in dreams by the rapid succession of particles, tanta est copia particularum ut possit suppeditare) and in Cic. N. D. 1 109 (referring to the divine images) innumerabilitas suppeditat atomorum. See for a general discussion on the subject my notes on N. D. 1 49.

2 See Lucr. IV 26 foll., Diog. X 48.

ground of our belief in the interminableness of images, not vice versa. When we further remember that these countless images are supposed to travel intact all the way from the intermundia, (see Cic. N. D. 1 114 ex ipso (deo) imagines semper affluant, and Lucr. vi 76 de corpore quae sancto simulacra feruntur in mentes hominum divinae nuntia formae,) and to be incessantly thrown off from bodies which were themselves scarcely more than images, we shall not wonder that some of the Epicureans failed to rise to the height of the credo quia impossibile which their system demanded, and fell back on the easier doctrine of Democritus, asserting the divinity of the images themselves, and deriving them not from the deities of the intermundia, but from the combinations of etherial atoms floating in the surrounding air'.

1 This seems to me to be the easiest explanation of the much disputed words of Diogenes x 139, ἐν ἄλλοις δέ φησι τοὺς θεοὺς λόγῳ θεωρητούς, οὓς μὲν κατ' ἀριθμὸν ὑφεστῶτας, οὓς δὲ καθ ̓ ὁμοειδίαν ἐκ τῆς συνεχοῦς ἐπιῤῥύσεως τῶν ὁμοίων εἰδώλων ἐπὶ τὸ αὐτὸ ἀποτετελεσμένων άv0рwπoe‹dŵs. Hirzel in his Untersuchungen zu Cicero's philosophischen schriften, pp. 46-90, whom Zeller follows in his last edition, p. 431, has shown, in opposition to Schömann (De Epicuri Theologia, contained in the 4th vol. of his Opuscula), that there is no reason for altering the text, and that we must accept it as a fact that there were two classes of gods recognized in the Epicurean school, one possessed of a separate individuality and having their abode in the intermundia, the other existing only in virtue of a continuous stream of undistinguishable images which in their combination produce on our minds the impression of a human form. Zeller thinks that the latter are meant for the unreal gods of the popular mythology, which, like the centaur and every other human imagination, must have their origin in some corresponding image; but the words of Diogenes seem to me to be less appropriate to the very concrete deities of the Greek pantheon than to some vague feeling of a divine presence such

Leaving the question of immortality, we pass on to speak of the Epicurean belief as to the shape of the Gods. They derided the spherical mundane God of the Stoics, and held that the direct evidence of visions, no less than the general belief of mankind, testified that the Gods were in the likeness of men. But this might also be proved by reasoning, for experience showed that rationality was only found in human form; and besides, the human, being the most perfect form, must be that of the most perfect being. Some of the later Epicureans went on to describe in detail the manner of life of their Intermundian Gods. They lived in houses, ate and drank celestial food, needed no sleep, for they were never weary; their chief enjoyment was conversation, which probably went on in Greek or something very like it: in fact they were in heaven what the Epicurean brotherhood was, or strove to be, on earth'. Such Gods were worthy of our reverence and imitation, but they were not objects of fear, as they neither could nor would do us harm2.

While Epicurus agrees with Aristippus in making pleasure the sole natural end of life, the standard of good, as sensation is of truth, he differs from him in attaching more value to permanent tranquillity than to as might be caused by the idola of Democritus. Compare also the parallel passage in Cic. N. D. I 49.

1 See Philodemus, quoted by Zeller, p. 434 foll.

2 Some of the Epicureans seem to have allowed to their Gods a certain influence over the happiness of men; see the passages quoted from Philodemus πepì evσeßeías in my note on Cic. N. D. I 45, especially pp. 86—89 (Gompertz) 'the Stoics deny that the Gods are the authors of evil to men and thus take away all restraint on iniquity, while we say that punishment comes to some from the gods and the greatest of good to others.' See too Lucr, vi 70.