page 601 note 1 Some odd points may be dealt with in a note. (1) In the Upanisadie doctrine of immortality the moon is mentioned as the dwelling-place of spirits, as in the Kauṣīalci Upaniṣad and in the Pañcāgnividyā, for which see Deussen, , Phil, of the Upanishads, pp. 328 seq.Google Scholar; Windisch, ,Buddha's Geburt, pp. 71Google Scholar seq. Pythagoras is asserted by Iamblichus, Vit. Pyth., 82,Google Scholar to have said that the islands of the blest were the sun and moon, but the idea is already Orphic (Zeller, , op. cit,, p. 457;Google ScholarRohde, , op. cit., p. 423, n. 4Google Scholar). (2) In Satapatha Brāhmana, i, 9, 3, 10,Google Scholar it is said that the rays of the sun are the good; in Aristotle, , de An., i, 2,Google Scholar that some thought the soul was the dust in the air (cf. Rohde, , p. 453, n. 5Google Scholar; Gomperz, , i, 138Google Scholar). Von Schroeder identifies (p. 25) these views by assuming, what is quite impossible, that the reference in Aristotle refers to the souls of the good, a view which turns the passage into nonsense, as it is a definition of the soul as such, and by converting the good into the souls of the good, which is legitimate. (3) The Pythagorean 
is quoted not only by von Schroeder (p. 39), but also by Garbe (op. cit., p. 43Google Scholar) and by Hopkins (op. cit., p. 559Google Scholar), as Indian because it occurs in the Atharvaveda, xiii, i, 6.Google Scholar But, as Weber and Lanman (Sanskrit Reader, p. 349Google Scholar) have pointed out, the expression is already found in Hesiod, , Op., 727Google Scholar, and no sane criticism will imagine that a piece of folklore like that and the numerous other examples in the preceding and following verses came to Hesiod (eighth century B.C.; cf. Mair, , Hesiod, p. 134Google Scholar) from India. See also Berthelot, , Trans. Third Inter. Congress of Relig., i, 274.Google Scholar (4) It is suggested that the musical theories of Pythagoras may have been due to the Vedic Śiksā. It is perhaps sufficient to say that the seven notes of the Vedic scale do not correspond to the Greek notes (of. Burnell, , Samhitopanisad Brāhmana, pp. vi seq.;Google ScholarCaland, & Henry, , L'Agniṣṭoma, App. iiGoogle Scholar), that the Śikṣās have nothing to do with the theory of music such as Pythagoras developed it, and that the Śikṣās are all late works and of no use as evidence for the sixth century B.C.; cf. Lüders, , Vyāsaśrkṣā, pp. 2Google Scholar seq. (5) It is also suggested that the medical art of Pythagoras, which seems to have been accompanied by the use of spells, music, and song, is Indian. The answer is, of course, that it is ethnic (cf. e.g. the famous spells of the Atharvaveda, Kuhn, , K.Z., xiii, 49Google Scholar seq., 113 seq.) and is earlier proved to exist in Greece, though no doubt it equally early existed in India. (6) Colebrooke and von Schroeder lay stress on the fact that the Pythagoreans are said to have believed in a threefold division of the world into Olympos, Kosmos, and Ouranos, and with this division they compare the three worlds of the Vedic mythology—earth, air, and heaven (see Macdonell, , Vedic Mythology, p. 7Google Scholar). It is sufficient to say that the names are not merely not attributed to Pythagoras himself, but expressly (Stob., Ecl., i, 488Google Scholar) to Philolaos, and even if as applied to him they are genuine, the division has nothing to do with the Indian one, for the upper region contains , the middle the seven planets and the sun and moon, the third the sublunar and terrestrial region. Nor is there any force in the argument that Pythagoras ascribed to spirits the middle region: there is no evidence for any such formal ascription. (7) The arguments of Colebrooke, derived from a distinction between øρήν and θνμός similar to that between jīvātman and manas, and from a distinction between the coverings of the soul analogous to those between the sūkṣma and sthūla śarīra, are not borne out by any Pythagorean writings, and probably refer to Neo-Pythagoreanism. For the real position of θμός in early Greek thought, see Gomperz, , i, 248Google Scholar seq. It may be added that the Upanisads have not yet learned to distinguish sharply mamas and jīvātman (Deussen, , op. cit., p. 271Google Scholar), or the sūkṣma and sthūla śarīra (ibid., pp. 280 seq.). (8) Hopkins (p. 559) mentions the vow of silence and compares it to the vow taken by the Indian muni. The evidence as to Pythagoreanism is late (Zeller, , p. 342Google Scholar); Aristoxenos probably invented the “mystic silence” to explain the absence of philosophical doctrines proper before Philolaos; Burnet, , op. cit., p. 96;Google Scholar in any case ritual silence is ethnic; cf. a curious example in Frazer, , Trans. Third Inter. Congress of Relig., i,Google Scholar 256 seq. (9) The difficulty felt by von Schroeder at the idea of an independent origin of metempsychosis is exaggerated (see Dieterich, , Nekyia, p. 90;Google ScholarZeller, , p. 73;Google ScholarArchiv für Relig., viii,Google Scholar 29 seq.). It existed among the Druids in Caesar's time—his evidence is quite clear (B. G., vi, 14, 5;Google Scholar cf. Diodorus, , v, 28, 6;Google Scholar Ammian. Marcell, ., xv, 9, 8Google Scholar). Moreover, as Rohde (p. 427, n. 3) points out, the Greek mind was familiar with the transference of the soul from one body into another, as in Ovid's Metamorphoses; cf. Kirke and the comrades of Odysseus. Cf. also the case of the Druses and other peoples set forth in Tylor, , Prim. Cult., ii, 3Google Scholar seq., and in Berthelot, Seelenwanderung. (10) It is impossible, as von Schroeder and Gomperz, i, 124Google Scholar seq., try to do, to deduce all Hellenic knowledge of transmigration in early days from Pythagoras and his influence. Plato (Phœd., 62 B; Cratyl., 400 B) clearly refers the belief to the Orphics, and Pindar (who in Thebes could hardly be moved by a mere South Italian belief, Zeller, , p. 71Google Scholar) must here follow the Orphics. Herakleitos (c. 504 B.C.) knew of it (cf. Burnet, , p. 172Google Scholar, with Rohde, , p. 442Google Scholar). Pherekydes, who was certainly older than Pythagoras (see Rohde, , p. 461,Google Scholar n. 1; Zeller, , pp. 71, 327;Google Scholarcontra Gomperz, i, 542Google Scholar), is said to have held the doctrine by Cicero, , Tusc, i, 38.Google Scholar Empedokles held the doctrine in full form : he knew Pythagoras' view, but there is no reason to suppose he borrowed it; his treatment varies from that of Pythagoras (Rohde, , pp. 473Google Scholar seq.). For the Thracians transmigration appears in the tale of Zamolxis or Salmoxis (Hdt., iv, 94; 95Google Scholar). The demons of Hesiod, , Op., 250Google Scholar seq., form a preliminary stage (cf. Rohde, , pp. 89Google Scholar seq.; Meyer, E., Gesch. des Alterthums, ii, §§ 453 seq.Google Scholar). Cf. also Murray, in App. to Harrison's Prolegomena and JHS., iii, 114 seq.Google Scholar, on gold plates of Petelia with Orphic verses, dating from the fourth or third century B.C. These verses are of great importance inasmuch as they conclusively disprove what was the main difficulty in Zeller's time of dealing with Orphism, the theory that the Orphic fragments were all late (cf. Zeller, , i, 98Google Scholar seq.; Purser, , Dict, of Antiq., ii, 302Google Scholar). These verses establish the existence of the cosmological and psychological doctrines of the Orphic school as they are revealed in the Orphic rhapsodist theology, and Diels' investigations (Archiv für Gesch. der Phil., ii, 91Google Scholar) lead him to the conclusion that the original form of the Orphic theogony belongs to the sixth century, and that the Orphic eschatological mysticism is a good deal older still (see Gomperz, , op. cit., i, 539Google Scholar). The world-egg is alluded to in Aristophanes, Aves, 695;Google Scholar Phanes occurs on a plate from Thurioi; but to go further into the question of the evidence for Orphism would be out of place here. Philolaos quotes as his authority (Clem, ., Strom., iii, 433Google Scholar) for the doctrine of the bondage of the soul, (11) Gomperz' own argument rests (i) on Xenophanes: if the belief in metempsychosis had existed, he says (i, 126), Xenophanes would not have ridiculed Pythagoras especially on this account. This criticism is quite unsound. In the first place, the doctrine of transmigration was never a universal belief in Greece: a satirist like Xenophanes could always make effective fun of it. Secondly, the point criticized by Xenophanes is a very remarkable one: Pythagoras goes beyond all early Indian transmigration ideas by claiming to recognize in a dog's howls the voice of a friend. Again, (ii) Gomperz. says that this episode is based on kindness to animals, and the Greeks, were not especially friendly to animals; there were, with a few isolated exceptions (a statement which, cf. Harrison, , Trans. Third Inter. Congress of Relig., ii, 154,Google Scholar is somewhat exaggerated), no sacred animals as these were in Egypt and India. But this is very doubtful: (a) the. doctrine is sufficiently accounted for by the Pythagorean (Porph, ., Vit. Pyth., 19; Rohde, p. 465Google Scholar); cf. also Westermarck, , Origin and Development of the Moral Ideas, ii, 500Google Scholar seq.;, (b) the indifference of the Greeks to animals does not apply in any case, even assuming its general truth, to the dog, cf. Odysseus' dog, and Geddes, , Problem of the Homeric Poems, pp. 221Google Scholar seq.; (c) thereis no necessary connection between the existence of sacred animals and kindness to animals generally. In India the existence of sacred animals did not prevent the contemporary existence of a brutal and cruel ritual for the slaying of animals in sacrifice; see Aitareya Brāhmaṇa, ii, 6;Google ScholarHopkins, , p. 198, n. 5.Google Scholar Nor is India in any special degree remarkable for kindness in theory or fact to animals generally even now. (iii) Vegetarianism is common to India and Pythagoras: the statement is true of neither, as we have seen above, (iv) The formulae which summarize the whole creed of the “circle and wheel” of birth are likewise the same in both. This statement (for which cf. Oldenberg, , Ancient India, p. 96Google Scholar) is applicable to the Orphic conception and is a mere case of natural coincidence, (v) Pythagoras, he thinks, learned of the doctrine viâ Persia, and he points out that the Asiatic Greeks while Pythagoras dwelt in Ionia were united with a part of the Indian nation under the sway of Cyrus. This merely means that Cyrus conquered Asia Minor and a small part of the north-west of India. The Indians do not occur in Greek literature before Aischylos (cf. Maspero, , The Passing of the Empires, p. 694;Google ScholarBusolt, , Griech. Gesch., ii, 515Google Scholar), nor the Greeks in Indian literature before Pāṇini's Yavanānī (see my Aitareya Āraṇyaka, pp. 22 seq.Google Scholar), which points to the fifth century B.C. at soonest). (12) It may be argued that if we admit foreign influences it is absurd to exclude Indian. But there is; a great difference. India was remote from Greece, and, unlike Egypt, not in any close touch with Greek travellers. A good instance of this close touch is seen in the Orphic cosmology itself, where the worldegg, the twofold nature of Phanes, etc., correspond very closely with Egyptian ideas (see Gomperz, i, 92 seq.Google Scholar). I see no reason to question here Egyptian influence (if the idea is originally Babylonian, still it no doubt came viâ Egypt) on Orphism, as the world-egg idea is not found elsewhere in early Greek thought. Cf. also Jastrow, , Trans. Third Inter. Congress of Relig., i, 237;Google Scholar and see now Petrie, Flinders, Trans. Third Inter. Congress of Relig., i, 196Google Scholar seq. Harrison (ibid. ii, 164) finds the origin of the egg in a primitive bird-worship, but I doubt this. This fact may also be cited as supporting the attribution to Egypt of the germs of the Greek doctrine of transmigration, and Maspero (Bibl. égypt., i, 349Google Scholar) thought it existed there when the country came into contact with Greece, but probably the Greek was only secondarily influenced in this regard by Egypt. It may also be pointed out that any theory can hardly be satisfactory which attributes to any individual influence the growth of the belief in transmigration. The facts are that the belief appears widespread over a considerable part of Greece; it is not universal by any means, and therefore cannot be regarded as quite a normal development of Greek religious feeling. On the other hand, it is much too widespread to be the creation of a single mind, and the theory of Thracian influence, which Gomperz rejects, receives most important support from the fact that the doctrine is unquestionably closely connected with Orphism, which beyond doubt came in from Thrace, and by the fact that the Phrygian religion is marked by its orgiastic character. The Thracians were an uncultured people who held their religious beliefs in a much deeper way than natural in an enlightened Hellas; contrast with Euripides' general attitude the Bakchai written in the north. (13) The fact must of course be emphasized that we may at any time be confronted with new evidence proving Oriental influence, though I think such evidence will be more likely to point to Egypt as the source of the doctrine than to India (ef. e.g. Foucart, Recherches, on the Mysteries of Eleusis, which, however, differ essentially, as Rohde, and Farnell, , Cults of the, Greek States, iii, 146Google Scholar seq., have shown, from those of Orphic societies). What I have endeavoured to show is that the arguments of von Schroeder are not convincing, and that any new investigation must rest on fresh arguments. At the same time I gladly recognize that von Schroeder's arguments were for the time of the appearing of his book practically decisive. It may be added that I have not attempted to deal with the quite different question of whether Indian influences may not have been at work to produce the Thracian doctrines: I do not see any evidence for that, but following von Schroeder I am merely concerned with the theory of Indian influence on Pythagoras himself. (14) Garbe finds other early evidence of Greek borrowing from India, especially (Phil. of Ancient India, pp. 54, 55Google Scholar) in the derivation from the Vāc doctrine of the Logos idea by Herakleitos. That this is impossible I think certain; see for the real sense of Logos in Herakleitos, Burnet, op. cit., pp. 146, n. 3, 153, n. 1. In fact, Hopkins, , Rel. of India, p. 558Google Scholar rejects even Weber's view (Ind. Stud., ix, 473Google Scholar) of Indian influence on neo-Platonism in this regard. See also the doctrine as it appears in Egypt (fifth century B.C. onwards), and is described by Petrie, Flinders, Trans. Third Inter. Congress of Relig., i, 196 seq.Google Scholar (15) Oldenberg, , Ancient India, p. 96,Google Scholar points out the similarity between the Buddhist and Orphic conceptions of the wanderings of the soul in the next world, but the Egyptian Book of the Dead is a much more obvious source of the Greek version if foreign influence be demanded. (16) With reference to the question of remembrance of former births (p. 577), it should be noted that Windisch, Buddha's Geburt, p. 62, n. 2,Google Scholar accepts the interpretation of Aitareya Āraṇyaka, ii, 5,Google Scholar which finds in it an assertion of Vāmadeva having remembrance of his former births. But I have tried in my edition (p. 233) to show that this is at least very improbable. Nor does Bṛhadāraṇyaka, i, 4, 22,Google Scholar bear out the theory: that passage merely asserts in Ṛgveda, iv, 26, 1,Google Scholar a recognition of the unity of the universe, and does not illustrate recollection of previous births, which is in no sense the subject of the passage in the Upaniṣad.
Āraṇyaka, p. 30).
praised by 
Much has wrongly been made of this word, as of his polymathy. It is quite a mistake to read into this the modern conception of historical research and comparative study of religion.
The game is believed to have been borrowed from Egypt (
is quoted not only by von Schroeder (p. 39), but also by Garbe (
, the middle the seven planets and the sun and moon, the third the sublunar and terrestrial region. Nor is there any force in the argument that Pythagoras ascribed to spirits the middle region: there is no evidence for any such formal ascription. (7) The arguments of Colebrooke, derived from a distinction between øρήν and θνμός similar to that between jīvātman and manas, and from a distinction between the coverings of the soul analogous to those between the sūkṣma and sthūla śarīra, are not borne out by any Pythagorean writings, and probably refer to Neo-Pythagoreanism. For the real position of θμός in early Greek thought, see 
(11) Gomperz' own argument rests (i) on Xenophanes: if the belief in metempsychosis had existed, he says (i, 126), Xenophanes would not have ridiculed Pythagoras especially on this account. This criticism is quite unsound. In the first place, the doctrine of transmigration was never a universal belief in Greece: a satirist like Xenophanes could always make effective fun of it. Secondly, the point criticized by Xenophanes is a very remarkable one: Pythagoras goes beyond all early Indian transmigration ideas by claiming to recognize in a dog's howls the voice of a friend. Again, (ii) Gomperz. says that this episode is based on kindness to animals, and the Greeks, were not especially friendly to animals; there were, with a few isolated exceptions (a statement which, cf. 
(
