Theology and Philosophy in Early Greek Thought

Reu si om Theology and Philosophy in Early Greek Thought Gregory Vlastos The Philosophical Quarterly, Vol. 2, No. 7. (Apr., 1952), pp. 97-123. Stable URL: http://links,jstor.org/sici?sici=003 1-8094%28 195204%292%3A7%3C97%3ATAPIEG%3E2.0.CO%3B2-M The Philosophical Quarterly is currently published by The Philosophical Quarterly. © Y Saye Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issuc of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www. jstor.org/journals/philquar.huml. Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. “> = N | 1N E JSTOR is an independent not-for-profit organization dedicated to creating and preserving a digital archive of scholarly journals. For more information regarding JSTOR, please contact support @jstor.org. Gad I Cu a IT http://www.jstor.org/ Mon Jul 17 10:06:22 2006

PHILOSOPHICAL QUARTERLY | Vor. 2. No.7 APRIL 1952 THEOLOGY AND PHILOSOPHY IN EARLY GREEK THOUGHT I When one reads the pre-Socratics with open mind and sensitive ear one cannot help being struck by the religious note in much of what they say. Few words occur more frequently in their fragments than the term ‘god ’.t The style itself in certain contexts is charged with religious associations ; the rhythm and sentence structure of certain utterances is unmistakably hymnodic.? In Parmenides and Empedocles the whole doctrine of Being and Nature is put forth as a religious revelation. The major themes of all the physiologoi—the creation of the world, the necessity of its order, the origin of life, the nature of the soul, and even such things as the causes of winds, rain, lightning and thunder, rivers, meteorites, eclipses, earthquakes, plagues—were matters of vivid religious import to their contemporaries. Lightning, thunder, a storm, an earthquake were ‘signs from Zeus’ (Gioomuiu) that could stop a meeting of the Law Courts or of the Assembly ;® religious feeling for an eclipse could overrule military intelligence to cause the greatest disaster ever suffered by Athenian arms.4 The philosophers who took the ‘natural’ view of these things could not be indifferent to the religious bearing of their conclusions. To think of them 1See the word-index in H. Diels and W. Kranz, Fragmente der Vorsokratiker, 5th ed., Berlin, 1934-37 (all subsequent citations of pre-Socratic fragments refer to this work), s.v. theos : eight columns of listings, against six for physis, less than six for kosmos. See Ed. Norden, Agnostos Theos, Leipzig, 1913, p. 164; K. Deichgraeber, ‘ Hymnische Elemente in der philosophischen Prosa der Vorsokratiker ’, Philologus 88 (1933) ; and now W. Jaeger, Theology of the Early Greek Philosophers, Oxford, 1947, passim. SFor the evidence see e.g. O. Gilbert, The Constitutional Antiquities of Sparta and Athens. Eng. tr., London, 1895, p. 292, notes 3 and 5. 4Thuc. 7.50.4.

as mere naturalists, bracketing off their speculations from religious belief and feeling, would be to take a very anachronistic view of their thought. Now it is just this view that was upheld quite belligerently by Burnet,® whose Early Greek Philosophy has gone through four editions since its first “ publication in 1892 to become the most influential guide to the study of the origins of Greek thought in the English-speaking world. Burnet explained away the term ‘ god’ in the pre-Socratics as a ‘non-religious use of the word’ (p. 14); and though perhaps he never thought through the meaning of this remarkable expression, the general point of his contention is clear enough. Like many a god and goddess in Hesiod, he argued, the ‘ gods ’ of the philosophers are not ‘ objects of worship ’ but ‘ mere personifications ’ of natural phenomena. Now it is true that the physiologoi maintained in all their thinking a singular independence from the public cult. If this were all that Burnet had in mind, his contention would be not only true but, as I shall argue shortly, absolutely fundamental to the proper understanding of their religious ideas. But Burnet went far beyond this when he claimed that they themselves attached no religious import to “ those ideas which they proclaimed in open or tacit defiance of the prevailing faith. It is true, of course, that their primary object is to understand nature, not to reform religion. When they discuss religious concepts, they are generally content to leave religious practices alone. But even this statement has important exceptions, and though one of them fits Burnet’s thesis, the rest go dead against it. In Empedocles it is the mystic, not the physiologos, who is exercised about the cult. His heart-wringing appeals for a religion undefiled by animal sacrifices and the eating of beans are inspired by the Orphic piety of his Purifications which admits of no rational connection with the scientific temper and doctrine of his work On Nature.® But in Xenophanes we find something quite different. When he calls nature ‘God’ he is asserting no ‘mere’ personification, but a doctrine which has urgent religious relevance, since it prompts him to attack the traditional beliefs as both irrational and irreverent.® It is impious, he says in effect, to speak about the gods : in a good cause: he was combating the error, then prevalent in some quarters, ‘of deriving science from mythology ’ (p. 14). “See below, Section V. On the relation of Empedocles’s Purifications to Orphism see e.g. W. Kranz, ‘ Vorsokratisches III’, Hermes 70 (1935) at pp. 112-115. TAristotle, Met. 986 b 24, ‘looking at the whole ouranos, the One, he said, is god ’. *B 1.13-14, ‘ Pious (<¥ppovas) men must first hymn god with reverent (eÜdbúpuous) myths and pure words’. Which myths he considered irreverentis clear from lines 21-23 of this fragment and B 11, B 12. My translation of edppovas differs from the usual rendrings (‘ joyful’ in Burnet and Kathleen Freeman), butis justified both by the context and by such usage as in Aeschylus, Choephorot 88, müs edbpov’ do, TOs Kareúfopat; edpuois = ‘auspicious, of good omen’ (Liddell & Scott, s.v.). Here, as in the following ‘ pure’ (kaPapoict), Xenophanes is reinterpreting current religious terms, transposing them into the framework of his own rational religious beliefs. The word ‘god’ in the citation is not necessarily the One God of his philosophy of nature. As Grube observes in another context (review of Jaeger’s Paideiain American Journal of Philology 78 [1947], p. 211, n. 17), ‘6 Oeds no more implies the existence of one God than 6 dvOpwros the existence of only one man. . . Both are generic’. Thisis quite

as Homer and Hesiod do, implying that his doctrine sets general standards for pious utterance. Heracleitus goes much further when he blasts away not only at what people say about the gods, but at what they do in their most sacred rites.” Such modes of worship offend not only his reason, but his religious sense ; they are not only ‘ madness ’ (B 5, B 16), but sacrilegious madness, ‘ unholy mysteries’ (B 14). There is a strong implication here that there can and ought to be a different form of worship which would qualify as ‘holy’; and this is confirmed by a passage in Iamblichus: ‘I distinguish two kinds of sacrifice : first, those of the completely purified, such as would happen rarely with a single individual, as Heracleitus says, or with a handful of men; secondly, the material . . .1° We need not take this as a wilful preference for solitary worship. It is more likely an expression of Heracleitus’s despair of the capacity of the ‘many’ to understand what he was talking about and to act accordingly. In any case, it is clear that the ‘divinity’ of his World Order™ is seriously meant as a genuine religious object which could be worshipped by the enlightened. Nor will Burnet’s appeal to Hesiod support his thesis. Certainly many divinities of the Theogony were not worshipped; but the same could be said of scores of figures in the traditional mythology which no one would term ‘ non-religious’. A Greek might know of no local cult to Sun or Moon, and might even think with Aristophanes? that none existed throughout the whole of Greece, and still be outraged by a denial of their bona fide divinity.1® Certainly, too, many of Hesiod’s figures are personifications of natural or human phenomena ; but to say they are ‘ mere ’ personifications is to confuse the issue. What is there more typical of Greek religion than the personification of winds, springs, rivers, earth, season, graces, love, victory, justice, peace, etc., whose religious vitality is attested in the cult ?14 It is not Hesiod’s verse that personifies everything from Lightning and Thunder to Sleep and Fear and Rumour, but the religious attitude of his people which feels the world as the theatre of super-natural and super-human forces. When Hesiod fills out the divine genealogies with innumerable clear here, since the fragment shifts to ‘gods’ in line 24. The relation between the ‘One God’ and the many ‘ gods’ in Xenophanes is obscure. All we can get from the fragments is that the moral, non-anthropomorphic properties of the ‘One God’ are normative for the ‘ gods ’ as well. *Attacks on purificatory rites and on image-worship, B 5: on the mystery cults, especially those of Dionysus, and magicians, B 14, B 15. In B 92 I follow H. Fraenkel in ascribing only the words ‘ Sibyl raving mouth ’ to Heracleitus : ‘raving ’ (mainesthai) is the antithesis of Heracleitean logos, sophie, and is used in his denunciation of mystic rites at B 15. Diels’s (and the usual) editing of this fragment would make the raving Sibyl the prototype of Heracleitean wisdom ! 10B 69; omitted in Burnet, as W. K. C. Guthrie observes, Orpheus and Greek Religion (London, 1935), p. 230. 11B 67, B 102, B 114, and see H. Fraenkel, ‘ Heraclitus on God’, Trans. Amer. Philol. Ass. 69 (1938), pp. 230-244. 13 Peace 406 ff. 130f, the decree of Diopeithes, Plutarch Pericles 32; and Plato Ap. 26 de. MSee L. R. Farnell, Cults of the Greek States (London, 1909), Vol 5, Chapter 11 and notes. When Aristophanes jokes, ‘ pour a libation to Stupidity ” (Knights 221) he is not even lampooning this habit of mind; he is just taking it for granted,

persons, some of whom doubtless never figured in earlier mythology, he is simply pursuing the logic of this animistic view of nature and life. They all belong to the same ‘race’ as the gods of the cult, they all have some share, great or small, of that mysterious power which exalts divine beings above the rigid limits of natural necessity. And this is precisely what, on any theory, Heracleitus (for example) did not mean, when he called his World Unity ‘god’. What then did he mean? Burnet’s theory stops him from so much as asking the question, and leaves him with a blind spot for that part of pre-Socratic thought which is its unique contribution to religion. Thus he can see no more in Xenophanes than a denial of ‘ the existence of any gods in the proper sense’; the words “One god’ mean ‘No god but the world’ (p. 128). The result has the effect of a distortion. It turns the pre-Socratics into purely ‘scientific’ thinkers, ignoring the fact that, for better or for worse, their ‘ science’ was far more (and less) than science, in our sense, has any business to be.1° Doubtless their concept of nature as a self-enclosed, self-regulative system is the intellectual foundation of science, and they who built it out of incredibly inadequate materials have every right to be considered pioneers of the scientific spirit. But neither can we forget on this account that those who „discovered this concept of nature believed that they found in it not only the principles of physical explanation, but also the key to the right ordering of human life and the answer to the problem of destiny. They began with the faith that nature itself was animated!’ by that Wisdom and Justice which the most enlightened conscience of their race had imputed to Zeus. So long as this faith lived they could transfer to nature the reverence hitherto reserved for Zeus and could therefore call nature ‘ god’ without indulging in an empty figure of speech. 18See below, n. 90; p. 11: and cf. B. Ehnmark, The Idea of God in Homer (Upsala, 1935) ‘the criterion of divine power is its supernatural power’. Jaeger (op. cit., above, n. 2) holds that Hesiod’s gods ‘ are really subject to what we should call natural law ’ because they have all been ‘ generated by the mighty power of Eros’. The premise goes a long way beyond anything in the Theogony, where Eros gets five lines altogether (120-123, 201-202). But even if we choose to read the potency of Eros into every ‘birth ’ in the Theogony, we are still a long way from ‘ natural law ’, in a realm where the natural pattern of sexual generation can be broken ad libitum without occasioning the slightest surprise, as e.g. in the birth of Athena, or the birth of Erinyes, Giants, ' Nymphs, and Aphrodite (183 ff.), to say nothing of the (preumably asexual) generation of the originals, Chaos, Earth, Eros himself, and Night. 16This is recognized by Burnet himself in a remarkable passage. In Greek Philosophy (London, 1914) pp. 11-12, he notes that Greek philosophy ‘ is dominated from beginning to end by the problem of reality (rd ôv) ’, à problem ‘ which at once takes us beyond science ’, and adds: ‘ Greek philosophy is based on the faith that reality is divine. . . It was in truth an effort to satisfy what we call the religious instinct ’. Here is impressive evidence that the vision of this great scholar was not blinkered by his theory. Had he pursued this line of thought he would have forestalled my objections. Unfortunately he did not; on p. 29 of this later book he repeats, almost verbatim, the statements, made many years before in Early Greek Philosophy, which I have criticized here. 17] use the word advisedly. See my ‘ Equality and Justice in Early Greek Cosmologies ’, Classical Philology 42 (1947), p. 177 and notes,

There is a kind of poetic justice in the fact that Professor Jaeger’s Theology of the Early Greek Philosophers should have been delivered as the Gifford Lectures (1936) at St. Andrews where Burnet had held for many years the chair in Greek. It is not polemical in tone, and there is no mention of Burnet except at incidental points where Jaeger agrees with him as often as not. But it is doubtless the strongest reply Burnet’s thesis has yet received,!® and it is all the more telling for keeping clear of the fanciful speculations that marred earlier statements of the anti-thesis.!® It works with the sound methodological principle that pre-Socratic philosophy is generally marked ‘by close-knit coherence and should therefore be studied ‘as an indivisible organism, never considering the theological components apart from the physical or ontological’ (p. 7). Coming as it does from one of the foremost living students of Greek thought, it cannot fail to command atten; tion. It will remain for years to come one of the ‘standard’ books with which every student of the pre-Socratics will have to deal. Such a work does not lend itself to summary, and does not need it. The many insights by which it illuminates and enriches our understanding of the first chapter of Western philosophy can best be appreciated by those who will read the book as a whole, with the leisureliness and care it invites and deserves. I shall therefore forgo here any thought of surveying its positive contributions and keep to a more limited and somewhat invidious task. I wish to discuss briefly those aspects of the author’s thesis which strike me as open to grave objection. The critical tone of my remarks will not be misunderstood, I hope, as any reflection on the solid value of the work.2 It is merely incidental to the expression and documentation of an alternative point of view which agrees with Jaeger against Burnet about the authenticity of the religious component in pre-Socratic thought, but prefers to interpret it along somewhat different lines. My main question springs out of the very use of the word ‘ theology’ both in the title and constantly throughout the book as a description of the religious ideas of the pre-Socratics. The word, of course, means no more than ‘account of god’ and could be so applied to any doctrine of . divinity. But the historian of ideas must scruple to use fundamental terms 18For a sample of the clash compare their views of Xenophanes. Burnet: ‘He. would have smiled if he had known that one day he was to be regarded as a theologian ’ (Early Greek Philosophy, p. 129); Jaeger: ‘only as a theologian can he be really understood’ (p. 49). 1] am thinking particularly of K. Joel’s Der Ursprung der Naturphilosophie aus dem Geiste der Mystik, Jena, 1906, whose title tells its own tale, and F. M. Cornford’s From Religion to Philosophy (London, 1913), the early work of a great scholar which over-works the hypothesis that the categories of Greek philosophy were ‘ already implicit ’ in Greek mythology, and is further misled by uncritical borrowings from the then fashionable school of French sociology. It remains for all its faults a valuable, suggestive study. 20One’s best testimony to the value of the work of a serious scholar is to take it seriously, either by way of assimilation or else by way of criticism where one is compelled. to disagree. The former I have done repeatedly in earlier published work, expressing my personal indebtedness to Jaeger for many things I have learned from him, and will do so again both in this paper and hereafter.

without regard to their exact historical fitness. He must ask himself in this connection: Would the physiologoi themselves have used the word of their own speculations about divinity ? We know that Aristotle so terms his metaphysics qua science of divinity,?4 and may assume that Plato would not be averse to have his arguments about the existence and providence of the gods described as ‘ theology.” But we also know that neither Aristotle nor Plato assumed so much for the natural philosophers who preceded them. As Jaeger observes of Aristotle, ‘in historical contexts . . . he used the term to designate certain non-philosophers such as Hesiod and Pherecydes, whom he contrasts sharply with the oldest genuine philosophers or physicists ’ ;2 and Plato’s practice is quite similar.24 The basis of the contrast in Aristotle’s mind seems to be mainly that the theologians talk myth, while the philosophers speak the rational language of natural inquiry. Jaeger «would then seem to be on safe ground in taking this over, and so juxtaposing to the mythical theology of the older poets the rational or natural theology of the philosophers. But Aristotle never talks that way. As a historical term, ‘ theology ’ for him (and for his school) has a clearly marked denotation which’ excludes the physiologoi. Why so? The answer is not to be found in Aristotle, who never stopped to analyse the problem.2> We can get at it only by looking at the actual work of the theologians themselves, beginning Met. 1026 a 19; 1064 b 3. Jaeger holds that the very word theologia was coined by Plato (and adds, here unconsciously controverting the major thesis of his own book, ‘ and he [Plato] evidently was the creator of the idea’, p. 4). But Plato is a fastidious linguist. When he makes up a word he is very self-conscious about it (e.g. at Theaet. 182 a). There is no hint of terminological innovation at Rep. 379 a; the word is introduced by Adeimantus (not Socrates) as a variant for ‘ tales about the gods’: Socrates, ‘ The founders ought to know the canons (fypous) in accordance with which the poets should tell their stories (mythologein) [sc. about the gods] . . .’; Adeimantus, ‘. . . What then are the canons for stories [or, accounts] about the gods (typoi peri theologias) ?’ The casualness with which the word is used here (and, curiously enough, never again in Plato in contexts such as Laws X where we should most expect it had it been ‘ coined ’ to indicate ‘ the importance from Plato’s point of view of the mental attitude which it tries to express ,’ Jaeger, p. 194) suggests that it was in common use at the time. And I see no good reason for doubting that those who ‘spoke about the gods’ (hoi legontes peri theon, Plato, Laws 886 c, of the theogonists ; cf. Empedocles B 131, amphi theon . . . logon, of his own theogony) would be referred to as theologoi long before this time ; similar compounds with -logos such as chresmologos, meteorologos, occur in fifth-century texts; Philolaos B 8, if genuine, would settle the point. Whether theologia was used as the title of any of the sixth-century theogonies (as Diels assumes, e.g. in the case of Pherecydes) is a different, and secondary, question, which cannot be settled from the available cvidence. He adds a little further: ‘ Eudemus [of Rhodes, the first writer of a history of theology] would never have included his master Aristotle, the creator of metaphysics or theology in the philosophical sense, among the theologians’, p. 5; he have included Anaximander or Anaxagoras. still less would #At Laws X. 886 cd he distinguishes between (a) those who ‘ speak about the gods ’ and propound ‘ theogonies ’ and (b) those ‘ recent wise men’ (véov Kat copiv) whose cosmology is materialistic, further identified later at 891 c as ‘all the men who have undertaken natural investigation’. The distinction, assumed at the beginning, is rigorously observed in the body of the ensuing argument which explicitly ignores the former (886 d) to concentrate on the refutation of the latter (886 e ff.). Neither did Plato. But his practice comes much closer to what, I shall argue, is the real difference, taking it for granted that the theologians are talking about the gods of the cult (cf. Tim. 40d-41a) while the physiologoi are either downright atheists or else deny (the primary assumption of the cult) that the gods ‘ care for human affairs ’

with Hesiod whose fully preserved text and well known impact on Greek thought gives us our best basis for judgment. When Herodotus (2.53.2) speaks of those who ‘ composed a theogony for the Greeks and gave the gods their names and divided their honours and occupations and designated their forms’, he gives pride of place to Hesiod.2° His very wording here echoés the Theogony : ‘Tell how at the first the gods and earth came to be, . . . and how they divided their wealth and shared their honours’.2” The gods Herodotus is thinking about are quite specifically the gods of Greek worship. Hesiod’s range is broader. It takes in gods whose connection with the cult is marginal, indeterminate, or wholly non-existent. But the ultimate concern of the Theogony is with the major deities of the cult, ‘the gods, givers of good things ’.® The great drama of the Theogony turns on the ‘ struggle for honours ’?? between Titans and Olympians. It tells how the former lost to the latter just that ‘ wealth ’ and ‘honour’ whose possessors qualify as cult gods by virtue of having ‘good things’ to give to those who honour them with sacrifice,. libation, ‘and prayer. There is no need to slight in this connection the cosmogonic and cosmological interests of the Theogony. An adventurous mind could hardly inquire into the origin of the gods without passing over into the kindred question of the origin of the world. This would come all the more naturally to one who thinks, as Hesiod does, of the four main articulations of nature, Earth, Sky, Sea, and Night, as themselves divine beings, whose generation accounts at one stroke for both the origin of the physical universe and the generation of the whole ‘race’ of the gods. Thus theogony broadens out easily into cosmogony, and even passes at times into pure cosmology, as in the famous description of the geography of the universe (720 ff.), with equidistant intervals from Sky to Earth and from Earth to Tartarus, and its remarkable account of the ‘ sources and limits’ in Chaos of Earth, Sky, Sea, and Tartarus (736 ff.). One may recognize all this, and still assert that the great bulk of the epic is not cosmology or cosmogony, but theogony and theology.*° Its primary purpose is two-fold: first, to sort out the motley mass of divinities, both the ones that Hesiod found ready-made in cult and those added by his own inspired fancy, into well-defined stems of descent; and secondly, to vindicate the reigning order among the gods. The latter not only establishes the supremacy of the gods of the cult over all the rest, 26An earlier testimony of the crucial role ascribed by the Greeks to Homer and Hesiod in the shaping of the popular creed comes from Xenophanes, who makes them the main butt of his attacks (B 11, B 12). 27108 ff. ; I follow Solmsen (Hesiod and Aeschylus, Ithaca, 1949, p. 8, n. 7) in accepting the genuineness of vv. 111-114. For ‘honour’, ‘ gift’ (ru, yépas, É@pov), ‘lot’ (totpa) among the gods, see also vv. 74, 204, 393-399, 413, 882, 885. Vv. 46, 111; also 633 and 644 where the ‘givers of good things’ are clearly identified with the Olympian faction. 29V. 882. In Homer (Il. 15.165 ff.) the ‘ lots’ and ‘ honours ’ of the gods are fixed by primordial moira. In Hesiod they have to be fought for and apportioned as spoils of war. 30Cf. Solmsen, op. cit., p. 58: ‘itis wrong to call the Theogony a cosmogonic poem ’,

but ends the quarrelsome anarchy ‘of the Homeric pantheon and assures law and order on Olympus under the stern authority of Zeus. Hesiod’s audience is now assured that each cult god has and keeps his proper province, so that each may be worshipped without risking offence to his peers and thus causing more trouble than he is worth. This is what makes Hesiod’s work a theology in a sense which cannot, by any stretch, take in the physiologoi. It is not merely that his forms of thought are mythical, his standards of rationality more primitive, his con-. clusions more traditional than theirs, All these things are true, but do not get at the heart of the difference, which is just this: The divinity of the physiologoi has no direct connection with the public cult, and is indeed so independent of it as to leave the very existence of the cult-gods in doubt and expose the most sacred ritualistic acts to Heracleitus’s scornful rejeetion.?! Hesiod’s teaching of divinity, on the other hand, puts the objects of the public cult at its centre. The information it conveys and the assurance it offers about the divine order makes the acts of the cult sensible propositions to a thrifty, calculating, peace-loving worshipper, such as Hesiod himself and the rural public to which he spoke.3? What the ‘ Greeks ’ got from his Theogony—not later philosophers, like Aristotle, who regarded it so patronizingly as uncouth, archaic cosmology, but the people at large who read it as a religious text—was a creditable and satisfying account of the gods they worshipped. III It is just this relation to the public cult that Jaeger ignores in his account of Hesiod’s theology* and systematically belittles when he gets to the theogonies which are contemporary with the first generations of pre-Socratic philosophy. His thesis here is that these ‘theogonic writers cannot be understood except in the light of their close reciprocal relationships with the philosophers of their own period’ (p. 57). Now we do not know what influence, if any, these theogonies had on the philosophers, nor does Jaeger profess to tell us.?* Nor is there any evidence of philosophical influence #1The average man would at least gather from the teaching of the physiologoi that it makes the worship of the traditional gods perfectly pointless. Cf. the reaction of Strepsiades (Clouds 425-26), when converted to the new philosophical divinities : ‘I absolutely will not talk to the other gods, not even if I run into them on the street ; I will not sacrifice, nor pour libations, nor offer frankincense to them ’. . #Hesiod’s work is not, of course, apologetics for the cult ; none would be needed where no doubt of its validity has yet arisen. But when he comes across a specific feature of the cult which must have struck him and his hearers as decidedly queer, he is ready to explain away the difficulty with an aetiological myth: vv. 533 ff., and Solmsen, op. cit., pp. 48-49, for the interpretation of the Prometheus story as an aetiological myth. Though not in Paideia (Vol. I, 2nd Eng. edition, New York, 1945, p. 65), where he was not preoccupied with the thesis of the present book. S4Reconstructions of such influence are not lacking in the literature. With the single exception of the influence of Orphism on Pythagoreanism, they are wholly conjectural, and amount to nothing more than the detection of certain supposed resemblances between the theological and philosophical cosmogonies. For the best statement of this point of viee see Guthrie, op. cit., Chapter VII. I can only record the impression that the ingenious ‘ parallelisms ’ which are traced here between, say, Anaximander

on any of the theogonies prior to that of Pherecydes.%5 With this one exception, all that remains of their ‘ close reciprocal relationships ’ with the philosophers is that they dealt with the common problem of the origin of the world, the nature of the gods, the destiny of man. But their attack on these problems is decidedly different. Is the difference merely that they tell myths, while the philosophers do not? If that were all, Jaeger might be justified in treating them as a half-way house between Hesiod and Ionian physiologia. But in so doing he misses—ignores or explains away?*—the remarkable fact that these theologies are one and all ascribed to men who, as historical or legendary characters, were leading figures in the new religious movements and activities of the time. Orpheus is the founder of mystic rites par excellence ;2? so is Musaeus, and also a patron of cathartic medicine.% Of the historical figures Epimenides is the famous purifier of Athens, prophet, founder of shrines ;39 Onomacritus a no less famous expert in and the Orphic cosmogony are (a) unconvincing at such points as the alleged correspondence between a supposed gonimon in Anaximander and the Orphic Egg or between ‘ Anaximander’s Moist and the Orphic Eros and (b) constitute no proof of an ‘ indelible impression ’ which the Orphic theogony is here (p. 224) said to have made on Anaximander, and cannot even be taken as evidence of Anaximander’s acquaintance with this cosmogony. If we could generalize from the fragments of Xenophanes and Heracleitus, we would have to say that Ionian philosopgy ignores the sixth-century theogonies: Xenophanes makes Homer and Hesiod the butt of his polemic against the poets (above, note 26); Heracleitus inveighs against the same pair but names also Xenophanes, Hecataeus, and Archilochus (B 40, B 42, B 56, B 57, B 105, B 129); neither of them mentions Orpheus or Musaeus or any of the rest except Epimenides, who is mentioned by Xenophanes as having lived 154 years (B 20). 85Nothing in Jaeger to this effect except for (a) the conjecture that the two Titans in Epimenides B 5 are Oceanus and Tethys followed by the further conjecture, ‘ possibly the philosophy of Thales has been influential here ’ (p. 66), and (b) the assertion that ‘as in the older [sc. Pythagorean] philosophy, Air [in Epimenides B 5] is thought as the void’ (p. 65). (a) needs no comment. As for (b), there is no evidence in the text that ‘ air is thought as the void ’, nor, if there were, would it be chronologically plausible to infer Pythagorean influence on a sixth-century theogony; but perhaps Jaeger’s statement is not intended as assertion of philosophical influence, which would then reduce the ‘close’ influence of philosophy on the theogonies prior to Pherecydes to the compound conjecture in (a). 36On the ground (p. 60) that the nominal authorship is frequently demonstrably false. But what matters here is (a) the fact (not the veracity) of the ascription and (b) the use to which these writings were put, as, e.g., by Orphic sectaries, prophets and priests (Eur. Hippol. 925-27; Plato Rep. 364 e and Meno 81 a-c). 97Elur. Rhesus 943-44, ‘ Orpheus taught her [Athens] the torch-processions of mystic rites’; Aristoph. Frogs 1032-33, ‘Orpheus taught us mystic rites and to abstain from slaughter; Musaeus the healings of diseases and oracles’; Plato Prot. 316 d {ancient ‘ sophists ’ assumed various guises, one of them being that of] ‘ those occupied with mystic rites and oracles, followers of Orpheus and Musaeus’; Ephorus ap. Diod. 5.64.4: ‘he was the first to introduce mystic rites to the Greeks’. For the interpretation see especially I. Linforth, The Arts of Orpheus, Berkeley, 1941, pp. 291 ff. Jaeger’s remark (p. 60), ‘ he [Orpheus] was not a specifically religious figure but rather a mystical singer of primeval times ’, seems to me to pose a false contrast. Orpheus was both singer and religious figure ; one could refer to him qua singer ; but is there any instance where a theological doctrine would be referred to him qua mere singer ? Jaeger himself assumes without question a little later that the account of the soul in the ‘ so-called Orphic poems ’ (Arist., de An. 410 b 22 ff.) is ‘ Orphic ’ in the specifically religious sense of the word. 38A glance at the fragments in Diels-Kranz will show how persistently he is associated with the foundation of mystic rites, especially those of Eleusis. On his association with religious (purificatory) healing Aristoph. Frogs 1033 is decisive, See the fragments in Diels-Kranz,

oracles 34 Pherecydes’s religious ventures are obscure, but we do know his reputation as a wonder-worker.“ Is it likely that theogonic writings would be imputed to such figures if there were no connection between the religious speculations contained in these works and the religious enterprises of their reputed authors ? The only way we can account for both the unprecedented proliferation of such a literature and the peculiar authority attached to it by the sectaries of the new cults‘? is to consider how urgently the sponsors of new rites would need to explain and justify their meaning over against the massive authority of the traditional ceremonials sanctioned by long established usage. Thus the new worship of Dionysus produced ‘ sacred tales ’ (kieroi logoi) which ‘ explained’ to the Greeks ‘ both the name of Dionysus and the sacrifice and the procession of the phallus ”.# A similar function was served by the tales which expounded the Orphic belief in reincarnation : the Platonic Socrates says he heard them from ‘men and women wise in _ things divine ’ whom he identifies expressly as ‘ priests and priestesses who have made it their business to give an account of the matters with which they are occupied ’.“ Even more instructive, I think, is Plato’s account of the itinerant prophets at Rep. 364 b 2 : they appeal to Hesiod and Homer for support on some doctrinal generalities ; but they produce ‘a mass of books by Musaeus and Orpheus ’ as authorities for their ritual. The bond Herodotus 7.6, ‘ a chresmologos who set in order the oracles of Musaeus’. I do not know why Diels-Kranz do not include his fragments among those of the theogonists ; see the relevant testimonies in Kern, Orphicorum Fragmenta, Berlin, 1922, pp. 55-56, and for the interpretation, M. P. Nilsson, ‘ Orphism and Kindred Movements ’, Harv. Theol. Review 28 (1935) at pp. 195-198. 4Arist. frag. 191 Rose says of Pythagoras : ‘and he did not hold back from Pherecydes’s mircale-mongering ’. Theopompus’s Mirabilia included a section on Pherecydes (Pherec. A 1 and A 6). “Eur. Hippol. 954. 43Herodotus 2.48.3 ff. The original explanation Herodotus attributes to the quasimythical Melampous who, he thinks, imported the cult from Egypt ; but he adds that it was improved by ‘ subsequent wise men (sophistat)’. How close the bond between such ‘ sacred tales’ (hieroi logoi) and the cult might be in certain cases we learn from another report in Herodotus where the explanatory ‘ sacred tale ’ appears to be actually incorporated in the mystic rite (2.51.4). (One cannot help comparing in this connection his account of Persian sacrifices [1.132.3] where the recitation of a theogony by a Magos is part of the ritual; this report is now confirmed, I think, by students of early semitic ritual, who show the ceremonial function of Babylonian and Palestinian myths of Creation: for the references see Cornford’s essay, ‘ A Ritual Basis for Hesiod’s Theogony ’ in The Unwrit en Philosophy, Cambridge, 1950, though Cornford’s interpretation of Hesiod along similar lines strikes me as forced). That readings from sacred bibloi were a part of the ceremonial in some fourth century mystic rites is clear from Demosth. 18. 259. 44Meno 81 a. The importance of this passage was called to my attention by Linforth, op. cit, p. 294, whose interpretation I am largely following here. 45BB\wv 58 dnadov mapexovras.... Kaas OuyTohoicw, literally, ‘ they produce a pile of books . . . in accordance with which they perform their ritual’. I follow the general view in taking the ‘ they ’ of this sentence, to be the ‘ mendicant priests and soothsayers ? of 364 b 5; I am not convinced by Linforth’s suggestion (op. cit., pp. 90-91) that the subject is the holders of the general view introduced at 364 a, since the people in question here are obviously priests, while the general view would be that of the public at large. Incidentally, I believe that Lindsay’s and Cornford’s rendering for agyrtai, * mendicant priests ’, is correct and clearly supported by the dictionary (Liddell & Scott, new ed., s.v.) : from ageiro, to collect, agyrtes generally ‘ beggar ’, in special contexts ‘ begging priest ’). P. Boyance, ‘ Platon et les Cathartes Orphiques ’, Revue des Hiudes Grecques

of such a literature with the cult is not only as close as Hesiod’s, but closer, since it moves into that area of religious procedure which Hesiod had so largely left alone. But I am not suggesting that we should think of the major theogonies as ad hoc fabrications to explain this or that rite. We may certainly credit their authors with vigorous, adventurous minds that would hardly be satisfied with piecemeal aetiologies, but would weave their interpretations of particular myths or rites into a far-flung pattern of creation and salvation. The few surviving fragments attested by good authorities suggest that these theogonies were no less ambitious in scope than Hesiod’s epic.4 Vast canvasses they must have been, thronged with gods and goddesses, drawing the whole universe into their design, accounting in their own fashion for the origin and nature of gods, world, and men. But somewhere along the line their story would make good the claim of the religious enterprise favoured by those who composed or recited it and undermine, by implication or open attack, the claims of their rivals. Such a hypothesis will account for the known facts. And though the dearth of evidence makes it impossible to prove it, it can at least be partially documented in fragments of the one theology whose outlines can be reconstructed with some measure of confidence, that of Pherecydes. Following Aristotle’s famous statement that Pherecydes ‘does not say everything in myths ’,* scholars are generally agreed that this comes closer 55 (1942), at p. 225 and ff, rightly insists that there is no ground for taking the word to mean ‘charlatan, quack’; Plato takes obviously a very harsh view of these people, but he is not contrasting them as bogus-Orphics with ‘real’ Orphics. On the other hand, I see no reason to assume that Orphism was anything like a coherent, homogenous "movement ; not only was Orphism itself ‘ but one of the many currents of mystic and cathartic beliefs emerging in the archaic age ’ (Nilsson, op. cit., p. 185), but there were different currents within it, so that Plato could speak with deep respect of some Orphic functionaries in the Meno and adopt its doctrine of COpa-o ja. (Crat. 400 c) in the Phaedo, yet still feel quite free to vent his scorn on the itinerant soothsayers of the present passage. 4] hope this will meet Professor Rose's objection (‘ Theology and Mythology in Aeschylus ’, Harv. Theol. Review 39 [1946] at pp. 15-16) that Linforth’s view (referred to at note 44, above) is not ‘ the whole truth ’. Rose speaks of the theologians as ‘ minds not so radically tempered as to insist on going to the very foundations of the subject, in this case their own religion . . . ., taking nothing for granted and arguing from first principles, yet sufficiently alert to ask themselves what the time-honored names, legends, and rites meant’ (p. 16). This is perfectly acceptable except that in two respects it would hardly apply to the major theogonists who, unlike such poets as Pindar and Aeschylus, (i) did go in their own way ‘ to the foundations of the subject ’ and (ii) were concerned with more than the ‘ time-honoured names, legends, and rites ’ since their data would include new rites, legends, and even (as certainly in Pherecydes) names. “Fully confirmed in the Aristophanean imitation of Orphic theogony at Birds 685 ff., which is modelled on Hesiod not only in the scope and style of the theogony but also in the immediately following (709 ff.) parody of Works and Days ; see e.g. Rogers’s notes ad loc. The Aristophanean theogony, incidentally, also confirms the general relation of theogony to cult : the Birds’ claim to world sovereignty having been asserted against Zeus on the ground that they are ‘ more ancient and older than Cronus and Titans and Earth’ (467-68) and therefore (562) deserve to be worshipped above the (traditional) gods, the theogony in due course puts their claim to cosmic priority into the framework of a full-fledged account of creation. 4 Met. 1091 b 9. But note that Aristotle does not consider him any less of a ‘ theologian’ on this account : he puts him in the same company with the Magi, whose connection with the Persian cult we know from Herodotus (above, note 43).

to philosophy than any of the surviving theogonies.* But does this justify the statement that the names of his cosmogonic deities ‘ are merely a transparent archaistic veil which by no means obscures their purely speculative character ’?°° Speculative some of them certainly are. But are they ‘ purely’ speculative? Have they no definite connection with the cult? This it seems is what Jaeger is concerned to assert, presumably on the ground of borrowings from the philosophers. He holds that Pherecydes was influenced by Anaximander in substituting the beginningless, ever-living Zas, for the Hesiodic Zeus, great-grandson of Earth, and also in turning the traditional Cronus into the time-god, Chronos.® But would these transformations, whatever their debt to philosophy, serve ‘merely [to] express the recent utterances of speculative thought’ (p. 69)? The first thing we are told about Zeus (Bl and B2) is that he gave Gé (Earth) as a wedding gift®? to Chthonié (another name for Earth, here co-original with Zeus and Chronos) and that thus Chthonié acquired the ‘name’ Gé. Now Wilamowitz has pointed out that in Mykonos, hard by Pherecydes’s native Syros, there is a cult to GE Chthonié, whose apparent redundancy would be very properly 49See e.g. von Fritz in Pauly-Wissowa Realenzyklopaedie, s.v.; Kathleen Freeman, Companion to the Pre-Socratic Philosophers (Oxford, 1946) pp. 36 ff. 50Jaeger, p. 72; asserted (in the concluding paragraph of the chapter) of the sixthcentury theogonies generally, but intended (I assume) particularly of Pherecydes, since it would apply much less to any of the rest. SlJaeger I think has a reasonably good case in deriving Pherecydes’s doctrine of the eternity of the cosmogonic deities from Anaximander (pp. 67-68), though we cannot exclude the possibility that such an idea occurred independently to Pherecydes. The case for deriving Chronos from Anaximander (p. 68 and notes) seems to me quite another matter. There is simply nothing in Anaximander’s system to correspond to Time as a substance, still less a cosmogonic one; and this, I think, should be the decisive consideration in our interpretation of the closing words of Anaximander B 1, kard rnv ToD xpövov rfi, on which Jaeger bases his argument. That these words are a verbatim citation has been questioned in an acute paper by F. Dirlmeier, ‘ Der Satz des Anaximandros’, Rhein. Mus. 87 (1938), pp. 376-82, whose arguments, though answered in part by K. Deichgraeber (‘ Anaximander von Milet ’, Hermes 75 [1940], at pp. 16-17) make it hazardous to base any argument on the assumption that the words in question are Anaximander’s own. Assuming that they are, Jaeger insists that the word taxis has the active sense, ‘ ordinance’ or ‘ decree’, rather than the merely passive sense ‘order ’ (p. 35 and notes; also earlier in Paideia, I, 2nd English ed., New York, 1945, p. 455, note 50; a similar view had been taken still earlier by H. Fraenkel, ‘ Parmenidesstudien ’, Goett. Nachrichten, 1930, p. 183). I concede that if taxis was in the original fragment it probably had the active sense; but if so, it could only have been pure metaphor (for the periodic, cyclical order of ‘ reparation ’), part of the ‘ poetic language ? to which Theophrastus so pointedly refers in making the citation. It could not be literally meant because the idea of Time as an agent, issuing ordinances and decrees, has absolutely no place in the system as we know it. I conclude, accordingly, that there is no evidence for Jaeger’s assumption that Pherecydes borrowed his Chronos from Anaximander, since no one would seriously argue that Pherecydes copied a cosmogonic | deity from an incidental metaphor in a physicist. As for the equation of Chronos- Cronus, it is not asserted in the fragment ; but it is in the testimonies (A 9), and parallels the equations Zeus = Zas and Ge = Chthonie; one may perhaps conjecture that Chronos became Cronus through Zas’s intervention, as in the transformation of Chthonie into Ge. I agree with Gomperz (‘ Zur Theogonie des Pherekydes ’, Wiener Studien 47 [1929], at p. 16, n. 6) that Wilamowitz’s sententious ‘Ich halte einen Urgott Zeit im 6. Jahrundert fiir undenkbar ’ is sheer dogmatism. 52de is thus Chthonie’s ‘ gift of honour ’ (yépas, B 1; Tovrw ce rıu®, B 2). Various cities claimed the honour of having been Zeus’s wedding-gift (Acragas, scholiast to Pind. O1.2.10 ; Thebes, Euphorion ap. scholiast to Eur. Phoen..687; both from Zeus to Persephone eis ävakakvrTipia). Pherecydes extends this idea, making the whole earth a wedding-gift to Chihonie.

explained here by Pherecydes’s theology : dark, barren Chthonië becomes the multi-colored, fertile Earth when fructified by Zas, the principle of life.5® Here the Hesiodic bestowal of ‘honours’ by Zeus to other deities®* is merged with the notion of a divine marriage (hieros gamos), the prototype both of the Greek marriage-rites and of a solemn fertility-festival, the hieros gamos, celebrated in various parts of Greece.” The deities to which the hieros gamos is now referred by Pherecydes, including the explicit reference to them of the presentation of the gifts of Unveiling (the Anakalyptêria) in the nuptial rites,>> would surely be more than ‘ purely speculative ’ entities, mere expressions of ‘the recent utterances of speculative thought’. As for Cronus, he too is quite different from the Hesiodic figure, since he now fights against the Titans, retains possession of the heavens, and is there crowned king of all.57 That Cronus is not, as in Hesiod, Zeus’s prisoner in Tartarus, but his honoured associate in the upper world, may be asserted. on good evidence as Orphic doctrine ; and a fragment of Pherecydes explicitly connects the fight of his Cronus against the Titans with ‘the mysteries about the Titans and the Giants who are said to have made war on the gods and the [sc. mysteries] in Egypt about Typhon and Horus and Osiris ’.5° Finally we have the testimonies of Pherecydes’s teaching about the soul which Jaeger completely ignores—just why, I do not know. Cicero tells us that his doctrine of the eternity of the soul is the first extant in literature; another source that he believed in transmigration; a third 5%‘ Pherekydes ’, Preuss. Akademie der Wissenschaften, Klasse, 1926, at p. 125 : Glaube der Hellenen, I, at p. 210. 54$ee above, note 27. Sitzungsber., Phil.-hist. 55See L. R. Farnell, op. cit., I, pp. 184 ff. and notes; and W.K. C. Guthrie, The Greeks and their Gods (London, 1950) pp. 53 ff. 56B 2: ‘from this arose the custom (nomos) among both gods and men’. Rose interprets the presentation of the gifts of Unveiling as ‘in itself a rite of Union, for to give part of one’s property is to give a piece of one’s self and therefore a kind of communion with the recipient ’, and adds that traces of the Unveiling-rite still survive in the marriage-rites in modern Chios (Ancient Greek Religion, pp. 33 and 146). 57B 4, The leader of the Titans is now Ophioneus, an obscure figure in Greek mythology, but mentioned also (as Ophion) in the theogony sung by Orpheus in Apollod. Rhod. Argon. 1. 496 ff. (= Orpheus B 16 in Diels-Kranz), as ‘ holding sway on snowy Olympus ’ before he was expelled by the might of Cronus and ‘fell into the waves of the Ocean ’. 58 It occurs in one of our earliest and best sources of Orphuc belief : Pindar’s Second Olympian Ode (cf. also Pyth. 4.291). It is implied at v. 70, where the souls of the just ‘pass by the highway of Zeus unto the tower of Cronus’. The life of the just here (vv. 62 ff.) ‘ not vexing the soil with the might of their hands, nor the water of the sea, to gain a meagre livelihood ’ is pictured quite obviously as that of the ‘ age of Cronus’ (Hes. Op. 111 ff.), no longer a remote antiquity, as in Hesiod, but a perpetual present in the Isles of the Blest. (Hesiod, Op. 169-169 e Rzach, where Cronus, reconciled with Zeus, reigns over the Isles of the Blest, is generally recognized as a later interpolation : see Solmsen, op. cit., p. 156, note 142). 5B 4, The ‘ mysteries about the Titans and the Giants ’ must be Dionysiac and/or Orphic (Paus. 8.37.5, ‘ Taking the name of the Titans from Homer, Onomacritus both composed rites to Dionysus and made the Titans the perpetrators of the sufferings of Dionysus’). For the connection of Dionysiac and Orphic rites see Nilsson, op. cit., p. 202 and ff. The reference to the Egyptian rites is to be understood in the light of the widespread Greek belief that various Greek divinities and rites, especially Dionysiac and Orphic, came from Egypt ; see e.g. Herodotus 2.48 (note 43, above), and Hecataeus . of Abdera ap. Didorus 1.23 and 1.96, the latter including the statement that ‘ the rites of Osiris and Dionysus are the same ’,

that he taught that duality of heavenly and earthly elements in the soul which would be, on any theory, the most obvious corollary to the belief in transmigration.” This dualistic conception of man would match and would doubtless be connected in his theology with the ‘original [Zeus-Chthonië] dualism ’ which Jaeger rightly finds ‘the distinguishing feature of Pherecydes’ theory ’ (p. 69). The same belief in transmigration and in a dualistic conception of the world recurs in Pythagoreanism, whose founder is presented by so many of our sources as the junior associate of Pherecydes.fl Here for the first time we find a chapter of pre-Socratic philosophy which can be confidently classed as a theology, since its ultimate data were the hopes, rites, and tabus that centred about the doctrine of transmigration.®? Like Epimenides, ‘Listed under A 2 and A 5 in Diels-Kranz ; fragments B 6 and B 8 also bear on his doctrine of immortality. The heavenly and earthly parts of the ‘ soul ’ (Aponius in A 5) is probably a reference to the dual nature of man’s life, whose mortal part is earth and returns to earth, while its immortal part is ‘ heavenly ’ or ‘ aethereal ’ and returns to its heavenly source. There are many fifth-century instances of this belief, some of them listed in Gomperz, op. cit., p. 24, note 18; also Pindar frag. 131 : the ‘ body of all men ‘is subject to over-mastering death ’, but the eidolon, which ‘ alone comes from the gods ’ remains alive (a clear reference to mystic faith, since the first line mentions the ‘ rite that releaseth from toil’; also in the well-known Orphic tablets from Petelia and Eleuthernai [Crete], B 17 and 17a in Diels-Kranz, ‘I am the son of Earth and starry Heaven ’. “The references in W. Rathmann, Quaestiones Pythagoreae, Orphicae, Empedocleae, diss. Halle, 1933, p. 12, note 10. “In his remarkably learned dissertation (cited in the preceding note) Rathmann has taken it upon himself to dispute the general view that transmigration was certainly taught by Pythagoras. His arguments on this score are completely unconvincing. Thus Dicaearchus’s testimony is not discredited (as he suggests, p. 3 ff.) but on the contrary greatly strengthened by his personal disbelief in immortality and tendency to assimilate Pythagoras to his own preference for the ‘ practical life’. Rathmann offers no good reason to doubt the reference of Xenophanes B7 to Pythagoras in Diog. Laert. 8.36 (see W. Kranz, ‘ Vorsokratisches II ’, Hermes 69 [1934] at pp. 226-27) which is conclusive contemporary testimony to Pythagoras’s belief in transmigration. A more serious argument which has swayed Nilsson is drawn from the conflicts in the testimonies concerning Pythagorean abstinence from flesh, ‘ some speaking of a general prohibition of certain parts of the animal or certain animals, e.g. the matrix, the heart, the brain, the sea-urchin, especially of such animals as were not sacrificed. If the general prohibition against killing animals and eating their flesh existed originally among the Pythagoreans, these special prohibitions would be meaningless’; and if the general prohibition was not original, the belief in transmigration could not have been original either (Nilsson, op. cit, p. 206). This leads Nilsson to think that the belief in transmigration came later, under Orphic influence. But surely, the conflict of our testimonies on this point can be explained in other ways. We know that there were different grades of membership in the Pythagorean order; and there are explicit reports in our sources that the rules for the lower grade (the akousmatikoi) were (as we would expect) less stringent than for the inner circle not only in other matters such as the sharing of property (see contrast of koinobioi with akousmatikoi in Iambl. Vita Pyth. 29-30; and cf. ibid. 80-81) and doctrinal proficiency (ibid. 81 and 87 ff.) but also quite specifically for the observance of the ritual and dietary tabus (ibid. 108-109, 150). Nilsson objects (Geschichte der Griechischen Religion, I, Munich, 1941, p. 666, n. 8) that this resolution of the difficulty assumes that the distinction in grades of membership had been established in Pythagoras’s own life-time. But (i) I see no good reason to doubt the testimonies of the biographical sources on this point, and (ii) even if the distinction did arise in the fifth century after his death, it could still account for conflicting dietary prescriptions in our fourth-century sources. There are other factors, too, especially in the case of Aristoxenus (the most vehement opponent of the tradition of total abstinence), whose testimony is self-contradictory (cf. frag. 27 Wehrli, where he does concede that the Pythagorean diet was vegetarian after all, but tries to explain it away by giving medical reasons for it) as well as in contradiction to some of the best authenticated original prohibitions (as e.g. that of beans, which Aristoxenus denies, frag. 25 Wehrli), and can only be explained by his obvious effort to produce a rationalized, prettified portrait of Pythagoras, purged of magical features,

Onomacritus, and Pherecydes, Pythagoras was a religious sage : what is more, he is known to have founded a coherent religious sect.®® In our best and earliest sources we hear even more of his religious activities than of his philosophy. Herodotus mentions the Pythagoreans only to refer to the peculiar rites they shared with the Orphics. Plato mentions his name only to speak of him as the founder of what was ‘ still the Pythagorean mode of life”; Isocrates, to say that he was distinguished for wisdom in ‘sacrifices and religious rites’ as much as ‘in the rest of philosophy ’. That he enjoyed quasi-divine status; that sundry miracles were ascribed to him; that his order observed tabus which savour of primitive magic— all this is vouched for by Aristotle.6* The only writings ascribed to him by an early (fifth century) authority are the verses he is said to have published under the name of Orpheus.°® Apart from the doctrine of transmigration and the belief that ‘ things are numbers ’, the only idea we can credit to him with any measure of probability is the conception of the world in terms of a duality of principles, the finite and the infinite, the first being the principle of the good, the second of evil.”” Here is something absolutely without parallel in antecedent or contemporary natural inquiry ;”! and the contrast 6A prominent feature of all the biographical traditions. *4Herodotus’s reference to him as ‘not the weakest wise man (sophisies) among the Hellenes ’ (4.95) is, of course, no exception. Burnet wrongly translates sophistes here as ‘ scientific man ’ (op. cit., p. 85). The context (Pythagoras as a teacher of Salmoxis) makes it clear that Herodotus uses sophistes here to mean ‘ religious sage’ as e.g. at 2.49 : Rathmann (p. 47) compares also Eur. Rhesus 949, which refers to Orpheus and Musaeus as sophistai ; so also Protagoras in Plato Prot. 316d. But I do not mean, of course, to minimize Pythagoras’s encyclopaedic interests in philosophy, mathematics, music, etc., which are amply attested by the general tradition and get near-contemporary witness in Heracleitus B 40 and B 109; the latter is now generally accepted as genuine (vs. Diels) and is a good testimony to his wide learning, in spite of the fact that the words ‘ practised inquiry (historien) above all men’ is a sneer (so Verdenius, ‘ Notes on the Pre-Socratics ’, Mnemosyne, S. III, Vol. 13 [1947], at pp. 283-4). 52.79, shorter version of the Florentine manuscripts, and Linforth’s commentary (op. cit, 38 ff.). 66 Rep. 600 b. Our best commentary on the sense that Plato’s words here would convey to his audience are the numerous references to the way of life of Pythagorean sectaries in middle comedy (collected in Diels-Kranz, Pythagoreische Schule, E) where the dietary tabus are most prominent. 87 Busiris 28-29. ‘Fragments 191 ff. Rose. 68Ton of Chios B 2; ef. Iambl. Vita Pyth, 146. For the interpretation see Linforth, op. cit., pp. 110 ff. ‘ Aristotle, H.N. 1106 b 28, ‘ for evil is a [form] of the unlimited, and good of the limited, as the Pythagoreans imagined’; also Met. 986 a 16 ff., 987 a 13 ff, 1093 b 11 ff. What is ‘ peculiar ’ to the Pythagoreans says Aristotle (987 a 13 ff., and cf. W. D. Ross ad loc.) is that the infinite and the One (= the finite) are the ‘ substance’ of things and ‘this is why they said that number is the substance of all things ’—a remarkable presentation of the doctrine that things are numbers as an inference from the ultimate cosmological dualism of finite-infinite. On this general topic see W. A. Heidel, ‘ Peras and Apeiron in the Pythagorean Philosophy,’, Archiv für Geschichte der Philosophie 14 (1901), F. M. Cornford ‘Mysticism and Science in the Pythagorean Tradition ’, Classical Quarterly 16 (1922) and 17 (1923), and the latest study, J. E. Raven, Pythagoreans and Eleatics (Cambridge, 1948) Chapter I which finds (rightly, I think, against Cornford) that on Aristotle’s evidence ‘ there was in earliest Pythagoreanism an eternal dualism ’, p. 18. : "The difference would hold even as over against Alcameon whose ‘ opposites’ are cited by Aristotle, Met. 986 a 27 ff., as a doctrine ‘similar’ to that of the Pythagoreans : (i) the finite-infinite duality do not appear in Alemaeon’s list, nor do the implied arith-

is all the more striking in that the only Ionian philosophers Pythagoras could have known made the Infinite itself (Anaximander) or the infinite Air (Anaximenes) the ultimate cosmogonic principle and endowed it with the attributes of divinity. Pythagoreanism not only rejects the monistic concept of nature axiomatic in the physiologoi, but implicitly condemns their highest principle as evil. We can only infer that, whatever its details, this cosmological innovation was the speculative ground for the dualistic conception of man implied by the belief in transmigration which is the one doctrine we can certainly ascribe to Pythagoras. IV No one who has ever plunged into the bottomless pit of research into Pythagoreanism will find it in his heart to blame Jaeger over-much for leaving out this whole chapter from his book.’ The omission is disconcerting, all the same. For if it is a philosophical theology that we are looking for in the pre-Socratics, it is only here that we shall properly find it. For here we do get a system of thought which, on the evidence, must have served to justify the beliefs and practices of a religious cult. And this is precisely what we do not get in any system of natural inquiry from Anaximander to Democritus. There are no moorings in the cult for any of them from which reason is not free to cut loose. They are free to condemn the cult with the savage irony of a Heracleitus, or to explain its foundations away with the relentless persistence of a Democritus.’ They are also free to ignore it metical (odd-even) and geometrical (square-oblong) dualities ; (ii Alemaeon’s opposites are not discriminated as ‘ good ’ and ‘ evil’ respectively (in spite of Aristotle’s citation of ‘good and evil’ at Met. 986 a 34), since Alemaeon’s norm of health is equipoise (isonomia, B 4), not subordination, as would be required if one term in each pair of opposites were held ‘good ’ and the other ‘evil’. Alcmaeon was doubtless acquainted with the Pythagoreans at Croton, and may have been on intimate terms with them ; but there is no evidence for regarding him as a Pythagorean : see W. A. Heidel, ‘ The Pythagoreans and Greek Mathematics’, American Journal of Philology 61 (1940) at pp. 3-6. “There are scattered references to Pythagoreanism in the book (especially at pp. 83 ff. and 151 ff.), but these are only incidental to the discussion of other doctrines. A brief sketch of his view of the teachings of Pythagoras had been offered earlier in Paideia, Vol. I (pp. 161 ff. of the second English edition). ?3Not in Empedocles’s Purifications, for (as I shall argue in Section V) this work is not a genuinely philosophical theology, but a theology (in the more traditional sense) tacked on to the philosophical system of the work On Nature. See my ‘ Ethics and Physics in Democritus’, Philosophical Review 54 (1945) at p. 581, note 24 : and‘ Religion and Medicine in the Cult of Asclepius ’, Review of Religion 13 (1949) at p. 284 and notes. When Jaeger (p. 181) says that Democritus ‘ did not deny the gods altogether ’, but ‘ recognized eternity and imperishability as properties belonging to the gods ’ and retained ‘ prayer as the most fundamental way of expressing one’s faith in the reality of the Divine ’, he misunderstands B 166. There Sextus clearly tells us that (according to Democritus) the eidola were not imperishable and that there was ‘no other god [sc. other than the eidola] having an incorruptible nature’; whence it follows that there are no gods in the recognized sense of the word. When Sextus says that Democritus ‘eÜyero that he might chance to meet lucky eidola’, ebyero can not mean ‘ prayed’ (what sense would there be in praying to Democritean eidola ?) but only ‘ hoped, wished for ’ as e.g. in Aristotle Z.N. 1118 a 32 and 1129 b 5. I must also dissent from Jaeger’s interpretation (pp. 183-84) of B 30. I cannot see that this projects the style of the Ionian philosophers and their monotheistic doctrine into primitive times. The style of the prayer is at least as old as Homer (cf. Il. 3. 177). to ‘ Zeus ’ carries no necessary monotheistic implications. The reference

altogether, and this is even more characteristic of them, quite symptomatic of their general temper. Their theme is nature, and their object to explain the how and why of its unfailing order. When they find in this a moral meaning—and they all do before the atomists—they may express the trust and reverence they feel for it by calling it ‘god’. But they may not go so far. Thus there is no good conclusive evidence that either Anaximander” or Anaxagoras’ called their cosmogonic principle ‘god’ or even ‘ divine ’. If they did not, their reticence would be significant and easy to understand from what we know of their position. Consider the case of Anaximander : When one comes to his fragments fresh from the Theogonies,”” one moves into a strange new world of thought and feeling. So many of the familiar landmarks have vanished that one can hardly guess which of the old names if any, its discoverer would have wished to conserve. Not only is it true “Jaeger argues at length (pp. 29 ff., 203 ff.) that Anaximander did call the Apeiron ‘divine’. But the crucial text kal Tovr eivaı Td Oefov may just as well be (a) Aristotle’s own interpretation of the view that the all-encompassing arche must be divine as (b) citation or even paraphrase of Anaximander. (See J. S. Morrison in Journal of Hellenic Studies 69 [1949], p. 90). I see no definite evidence whatever in favour of (b), while (a) is favoured by two considerations: (i) Td Qetov does not occur as a substantive for ‘ divinity ’ in any of the pre-Socratics or any other text prior to Aeschylus and Herodotus, while it is one of Aristotle’s favourite terms; (ii) the ancients did not understand this particular text or any other text at their disposal to say’that Anaximander himself taught that the Apeiron was TÔ Oeiov : even the chapter in Aetius (1.7) which generously supplies even Democritus with a god (= fire !), does not say that Anaximander’s Apeiron was god, but only that ‘ Anaximander declared that the infinite ouranot were gods’. Cicero (de Nat. Deorum = Anaximander A 17) speaks, like Aetius, of the worlds as nativos deos, and objects that they cannot be gods, since god must be eternal. 78As Jaeger observes (p. 161) there is no such statement in the fragments; he disregards, I think wisely, such doxographic reports as those of Aet. 1.7.5. and 14. He nevertheless holds ‘ that this must have been his doctrine’. . . . both (a) from the hymn-like form in which the predicates of Nous are expressed, and even (6) from the content of these statements . . . [e.g.] the epithets “ infinite ”, “ self-ruling ”, “ unmixed ”, and “itself by itself” ’ (p. 161). Both (a) and (6) beg the question which is precisely whether Anaxagoras may not have used just these epithets in just that style, and still held back from calling the Nous ‘ god’ or ‘ divine’. On similar grounds one might have argued even more strongly that Parmenides taught the Divinity of his Being ; but there Jaeger respects the silence of the fragments and concludes that Parmenides ‘ definitely fails to identify Being with God’ (p. 107). Though the ex silentio argument is always dangerous, I think it unlikely that Plato would not -have alluded to the divinity of Anaxagoras’s Nous, had he known of it, in the crucial passage (Phaedo, 97 b ff.: and cf. Laws 967 b) which argues that, having said so much (‘ nous is the world-ordering principle ’), Anaxagoras should have said more (that ‘ everything has been disposed for the best ’) ; since the latter would have followed even more strongly from a teaching of the divinity of the Nous, some reference to such a teaching, had Plato known of any in Anaxagoras, would almost certainly have crept into Plato’s argument. “The standard ones, Homer’s and Hesiod’s, mainly the latter, since it was by all odds the most influential. As for the sixth-century theogonies, I have suggested above (note 34) that there is no evidence of Anaximander’s acquaintance with them. In any case, they are infinitely closer to Hesiod than to Anaximander, since they continue to assume the generation of the gods. The important exception is Pherecydes, whose cosmogonic gods are explicitly said to be eternal. I have expressed above, note 51, reserved agreement with Jaeger’s view that in this respect Pherecydes was probably influenced by Anaximander. This is chronologically permissible on the general view which makes Pherecydes a somewhat younger contemporary of Anaximander. To push Pherecydes towards the end of the sixth century (so Jaeger, p. 67) in order to facilitate the possibility of philosophical influence on him seems to me unwarranted. Incidentally, there is no evidence that Pherecydes ‘assimilated the philosophical criticisms of anthropomorphism ’ (Jaeger, p. 69); the divine marriages and battles which are so prominent in his theogony are definite evidence to the contrary.

that properties and functions traditionally reserved to the gods are now transferred to an utterly different sort of entity ; what is more, the properties and functions themselves have changed. In creation the pattern of sexual generation has been replaced. by a mechanical process, which simply sorts out the physical components of an original mixture, and does so not once, to creat the single world of traditional thought, but endlessly, to produce innumerable worlds throughout an infinity of time. What is there here that Anaximander would wish to denote by the same noun as that used of Hesiod’s Zeus? We simply have no a priori way of answering the question; we can only go to the evidence, such as it is. This does not tell us that Anaximander called the Apeiron ‘god’, but that he so called ‘the infinite worlds ’.7® This, then, Anaximander may have felt, was the best he could do for the gods in his system, presumably because the gods of the Theogonies were themselves generated.” Reserving ‘ agelessness ? to the Apeiron®, he was taking away from the gods their most characteristic prerogative, but doing so by the compulsion of his logic. In his system whatever has a beginning must have an ending. If the gods have a birth, they cannot be deathless; only the beginningless Apeiron can be truly ageless and immortal. No less different from the popular gods is the Apeiron in its role of world ‘governor’. ‘Traditionally the justice of Zeus is ‘ordained unto men’ and, quite explicitly in Hesiod, no further.®! Only occasionally are the ‘beasts’ too drawn into the circle of his Justice.®? The forces of nature, uncertainly personified as earth-born deities, are subdued by Zeus in the battle with the Titans and brought more or less under his power. But there is no notion of natural laws issued and maintained by Zeus. So far from maintaining natural regularities, Zeus himself and the other gods over-ride them right and left. To do so is their prerogative and indeed provides the main medium of omens, oracles, marvels, punitive thunderbolts, storms, plagues, etc., through which they make known their will to men and enforce it upon them. What could be further from the Justice of the Apeiron whose 784 17; see above, note 75. There is a certain analogy here in Parmenides, who calls Dike- Ananke a divinity in the realm of Becoming (A 37; B 12), but not in the realm of Being. As implied, of course, by the literal meaning of the word theo-gonia. As Jaeger observes (p. 32) the term ‘ ever being’ (aièv êóvres), frequently applied to the gods in Homer and Hesiod, ‘ shows merely that the gods are thought of as immortal, living for ever’. That this usage is continued in the sixth-century theogonies is confirmed by the Aristophanean imitation in the Birds; v. 688 speaks of the gods as ‘ the immortal, the ever being ’, while at v. 691 the theme of the theogony is announced as ‘the nature of the birds [= “ gods ” here], the generation of the gods . . … cydes’s is the first known deviation from the tradition. soB 2, B 3. Phere- 81Hesiod Op. 276-78. Solmsen, op. cit., p. 104, n. 6 observes : ‘ Hesiod had restricted these “ ideas ” [such as dike] to the dispensation of Zeus and to human society, setting them far apart from the physical entities like Earth, Heaven, Land, and Sea, which took shape early and have no part in the moral world order. In this important respect, the Presocratics may be said to have destroyed the Hesiodic pattern’. See also his remarks at p. 65 and pp. 159-160. 82Archilochus, frag. 84, Hiller-Cr.

laws, fixed in the physical structure of the world, are cosmic in their scope and natural in their execution? When Jaeger tells us that Anaximander’s cosmology offers ‘the first philosophical theodicy’ (p. 36), it is essential to remember that the ‘justice’ and ‘reparation’ of fragment 1 operate simply through the self-regulative periodicities of a mechanical equilibrium. This is certainly more, as Jaeger observes, that a ‘mere explanation of nature ’ in our sense of these words, or the words ‘ justice ’ and ‘ reparation ’ would be meaningless. But it is both more and less than a ‘ theodicy * in any sense in which the Justice of the gods had been conceived or, for that matter, ever was conceived, by the clientele of the Greek cult. No theodicy could satisfy the cult which did not include a doctrine of individual providence, and no such doctrine could be squeezed out of Anaximander, or even Xenophanes, Heracleitus, and others who did call nature ‘ god 84 Philosophy could fill the order only by grafting Zeus’s all-seeing eye on a cosmic power, watchful of every single life to requite its vice and virtue in minute proportion in this life and the next. It was probably the Pythagoreans®® and after them certainly Plato® and the Stoics who performed this remarkable operation, and thus produced a properly theological theodicy. 83] offer justification of this view at pp. 168 ff. of the paper cited above, note 17. I wonder whether Jaeger is true to his own methodological principle (cited above, p. 101) in speaking of ‘ divine justice’ in Anaximander without explicit consideration of the (purely physical) mechanism through which ‘justice’ and ‘reparation’ are maintained ? “Jaeger takes it for granted that the Justice and Reparation of Anaximander B 1 occur not only in the physical cosmos but also in the world of politics and morals (p. 35: and earlier in Paideia, I, 2nd. English ed, p. 160). There is no evidence for this assumption, nor is it possible to understand how it would fit Anaximander’s physics. How would the balanced equalities of hot-cold, dry-moist, etc., enforce reparation of human wrong-doing ? Heracleitus goes far beyond Anaximander in annexing ‘ human ? to ‘ divine ’ law and justice. But even here the self-regulative execution of the ‘ common law ’ inherent in the soul and the state falls far short of a doctrine of individual providence. So does Anaxagoras’s Nous as, I think, is clear from the disappointment of the youthful Socrates with his cosmology in the Phaedo. Jaeger takes Anaxagoras B 12 to mean that the ordering of creation by Nous proceeds in accordance with ‘ preconceived world-plan ’ (p. 163); but I do not know how he gets this out of exdopnoe (‘ ordered, arranged ’) or mávra éyvw vovs (* Nous knew all things’). Anyhow, I think Jaeger would agree that, so far as we can judge from the fragments, Anaxagoras’s Nous was not meant to fulfil the function of a personal providence. Neither would the Cosmic Intelligence of Diogenes’s Air. Jaeger follows closely W. Theiler (Zur Geschichte der teleologischen Naturbetrachtung bis auf Aristoteles, Zurich, 1924) in assuming that Diogenes was the source for the providential arrangements of nature recited by Xenophon (especially at Mem. 1.4 and 4.3.), but adds that we cannot read into this source the meaning they now have in Xenophon himself, ‘ namely, that these are all ways in which the gods provide for man’s needs ’ (p. 168). Theiler’s learned and ingenious thesis has never received to my knowledge the criticism it deserves. Earlier S. O. Dickerman (De Argumentis Quibusdam apud Xenophontem etc., Halle, 1909) had rejected the hypothesis that Diogenes was Xenophon’s source (p. 48); to Dickerman’s argument I would add that if Diogenes were as much of a teleologist as Theiler takes him to be, he could hardly have been so completely ignored in Socrates’s vain search for a teleological cosmology in the Phaedo. 85Not explicit in any reliable tradition of early Pythagorean teaching, but implicit in the whole doctrine of transmigration with its teaching of divine judgment and retribution. Cf. Philolaus’s doctrine (B 15) that the gods keep a ‘ guard ’ over men. ®Notably at Laws 903 b ff, Cf. F. Solmsen, Plato's Theology (Ithaca, 1942) Chape

I hope my argument?” has made it clear that it concerns more than linguistic propriety, and would not be met had Jaeger dropped the word ‘theology ’ and spoken less provocatively of ‘the religious ideas’ of the pre-Socratics. The real issue here is not verbal usage, but historical matter of fact, i.e. the actual relation of the beliefs of the pre-Socratics to those of contemporary religion. I do not mean to suggest that Jaeger is unaware of the gulf between the two. He often alludes to it in one way or another, and remarks in one place that the gap which exists in this period ‘is never again entirely closed ? (p. 174). My point is simply that, taking the gap for granted, he never stops to measure it. The result of this omission is actually, though doubtless unintentionally, an under-estimation of the vast distance between the two types of religious belief, and consequently a failure to exhibit the full dimensions of the unique achievement of the pre-Socratics as religious thinkers. This, in a word, lies in the fact that they, and they alone, not only among the Greeks, but among all the people of the Mediterranean world, Semitic or Indo-European, dared transpose the name and function of divinity into a realm conceived as a rigorously natural order and, therefore, completely purged of miracle and magic. To moralize divinity was not their main, and certainly not their unique contribution. Pindar and Aeschylus here labour in the same cause as Xenophanes and Heracleitus : and the Hebrew prophets were doing the same a good two centuries before Xenophanes, and with a passionate intensity unequalled by any Greek philosopher or poet. But the world of Pindar and Aeschylus is thick with magic of almost every description. The prophets of Israel and Judah fought a valiant rear-guard action against wizards, necromancers, and soothsayers. But they lacked the conceptual equipment to see that magic was not only a religious impropriety but a sheer impossibility; and they never cleared their minds of the notion of miracle which is the intellectual foundation of magic. Miracle remained a permanent feature of Hebraic as of Greek®® and, later, Christian piety. To present the deity as wholly ST have made no mention of the Eleatics who occupy a unique position among the pre-Socratics, partly because I have already sketched my interpretation of Parmendides’ peculiar blend of mysticism and logic in ‘ Parmenides’ Theory of Knowledge ’, Trans. Amer. Philol. Assoc. 77 [1946] at pp. 74-77, partly too because I find Jaeger’s account of Parmenides the most satisfying chapter in the whole book. ‘ What Parmenides has done ’, he writes, ‘is to take over the religious form of expression and transpose it to the sphere of philosophy, so that in truth a whole new intellectual world takes shape ’ (p. 97)—a fine statement, I believe, of the treatment of religious ideas not only by Parmenides, but by all the physiologoi from Anaximander to Democritus. “The same may be said of their contribution to religious universalism. Jaeger holds (p. 48) that this began in the western world ‘ neither with the Christian nor with the prophets of Israel, but with the Greek Philosophers’. But the idea of one God who governs nature and all men is already in Amos (1.3-2.3 ; 5.8 ; 9.6-7). SPlato’s position here is most instructive. Transmigration itself is a miracle, since the immortal part of the soul is reserved to the reason and is disjoined from the passionate and appetitive parts which are explicitly said to be mortal (Tim. 69 c ff.); how then could the identity of the immortal soul be preserved throughout its incarnations in apimals, reptiles, etc., which are conspicuously lacking in reason? It is no use trying to explain away transmigration as a myth, since it is (i) the premise for the epistemological doctrine of Recollection in the Meno; (ii) the premise for the metaphysical doctrine of the immortality of the soul in the Phaedo; (iii) repeatedly recognized in the cosmology of the Timaeus, Plato's attitude to the irrationalities of the public

immanent in the order of nature and therefore absolutely law-abiding was the peculiar and distinctive religious contribution of the pre-Socratics, and it should be put in the forefront of any account of their religious thought. They took a word which in common speech was the hallmark of the irrational, unnatural, and unaccountable” and made it the name of a power which manifests itself in the operation, not the disturbance, of intelligible law. The transposition opened new religious possibilities. Had these been realized, Greek religion would have been freed of those evils which Lucretius in retrospect so justly imputes to it. v Had Jaeger done justice to this phase of the religious teachings of the pre-Socratics he would have altered the perspective of much of what he has to say. And the main beneficiary would have been the chapter on ‘The Origin of the Doctrine of the Soul’s Divinity’, which makes Anaximenes, of all people, the bridge between earlier Greek beliefs about the soul The argument is roughly as follows: We know and the Orphic doctrine. that the Orphics thought of the soul as (a) divine and (6) independent of the body. Jaeger holds that they also thought it (c) incorporeal (p. 84) which, he thinks, would follow from (b), and (d) air-like, on the strength of Aristotle’s reference to ‘ the account in the so-called Orphic poems, [which] says that the soul comes in from the whole when breathing takes place, being borne in upon the winds ’.* He thinks that (d) ‘already presupposes the philosophical theory [of Anaximenes] that air is the principle of life’ (p. 80); and he also finds a connection with Anaximenes at (c), assuming that his breath-soul was ‘ as incorporeal as possible ’ (p. 84), and also at (a), since Anaximenes taught that Air was divine.® This raises more questions than it answers : Briefly, cult which his state religion would conserve (Rep. 427 be, Laws 738 cd) is a more com plicated question. It may be, as E. R. Dodds argues (‘ Plato and the Irrational’, Journal of Hellenic Studies 65 [1945], pp. 22-23) that there is ‘ little reason, and certainly no necessity, to credit Plato with a serious belief in the personal gods of Greek mythology and Greek cult’. Yet Plato speaks as though he did accept the reality of good and evil daemons with power to intervene unaccountably in human life (Laws, 877 a Ep. 7. 336 b), and at Tim. 71 divination appears as a vehicle of ‘ some apprehension of reality and truth ’ which, though quite inferior to reason, may still be due to ‘ divine possession ’. 9080 e.g. in Herodotus in spite of his very considerable rationalism. Note his use of ‘ the divine ’ to denote events such as these : the Cnidian workmen suffer an unaccountably large proportion of injuries (1.174.4) ; the cats in Egypt rush headlong into the fire (2.66); Ariston is convinced that not he, but an apparition, has had intercourse with his wife (6.69); the wrath of Talthybius descends on the perfectly innocent sons of Spertheias and Bulis (7.134). When the event is so strange that it is no use even looking for a natural explanation, then it must be ‘ divine’. Per contra, if a thing can be explained by natural means, it cannot be divine (7.16.2). So the author of On the Sacred Disease (1) : ‘men think it [the “sacred disease ’’] divine ( = delov Typa same expression as in Hdt.2.66 and 6.69) because of their inexperience, and its marvellous character, and the fact that it does not resemble anything else ’. 91De An. 410 b 22 ff. (= Orpheus B 11), in the Oxford translation. 92In attributing the doctrine of the divinity of the cosmogonic power to Anaximenes, Jaeger seems to argue (p. 36) mainly from the fact that it was already taught by Anaximander and would thus be conserved by Anaximenes. If, as I have argued, there is no definite evidence for ascribing the divinity of the arche to Anaximander, independent

(i) Chronology : Where is the evidence that the Orphic theory of the soul comes after Anaximenes? But (ii) why need Jaeger make such an assumption? He argues convincingly°® that the breath-soul is an old, pre-Homeric notion, as evidenced in such expressions as ‘ breathing out one’s soul’. Why need the Orphics get from Anaximenes an idea long embedded in their mother tongue ? (iii) If the Orphics thought of the soul as air,®* would they think of it as incorporeal ? Is air incorporeal ? (iv) Anaximenes certainly did not think the soul ‘as incorporeal as possible ’, but the reverse. To say that everything, from fire to earth, is air, is to say that soul, as air, is as corporeal as anything else.°5 But none of this really gets to the heart of the question. The stark, irreconcilable contrast between Ionian naturalism and Orphic dualism would be untouched by any theory of the material composition of the soul the Orphics may have entertained, even if this were borrowed from philosophy. Anaximenes himself may have taught the divinity of the soul, though there is no such statement in either fragments or testimonies. This way of speaking and thinking is quite congenial to the temper of Ionian philosophy. grounds for crediting it to Anaximenes must be offered. These I can only find in later testimonies (Cicero, Aetius in A 10), which, however, may be suspected of confusing the doctrine of Anaximenes with that of Diogenes of Apollonia who did teach the divinity of the cosmogonic Air. It may be worth noting that our most comprehensive and reliable account of his system, that of Hippolytus (fidissimum eacerptorem, Diels found him, Doxographi Graeci, p. 132) makes no mention of the divinity of Air, but only says that ‘ gods and things divine ’ were ‘ generated ’ from Air (A 7); and that St. Augustine (Civ. Dei. 8.2) seems to agree with Hippolytus, saying that ‘ the gods arose from the air ’. %In agreement with the view of E. Bickel, Homerische Seelenglaube (Schriften der Koenigsberger Gelehrien Gesellschaft 1 [1925], Heft 1). 94] see no good reason to doubt Jaeger’s view that they did, though no such conclusion is warranted by one of the sources to which he looks for support (pp. 80 and 84), sc. Aristotle’s report (de/An. 213 b 22 ff.) of the Pythagorean view that the ouranos ‘breathed’ in ‘air’ and ‘void’ from the surrounding infinity: Aristotle’s report does not say that this inhalation is soul, nor could it, since, qua infinite, the void would come for the Pythagoreans under the heading of the ‘evil’ principle which would certainly not be that of the (divine) soul. Aristotle’s single reference to the Orphic theogony (n. 91) seems slim ground on which to base the reconstruction of their theory of the soul, but perhaps good enough where positive evidence is so hard to come by. Here, however, we should note that Aristotle’s report does not strictly say that in this view soul was air: fire, or the fiery aether, could also be ‘ borne about’ by winds and ‘ inhaled ’ (so in the philosophers, e.g. Heracl. A 16, Democr. A 106). Jaeger doubtless means to use ‘ air ’ broadly enough to include fiery aether (clearly so in another context, p. 208, n. 63, where he refers to Heracleitus’s theory of soul as ‘ air’), and with good reason since the distinction between air and aether is hazy enough in the philosophers (for Heracl. cf. B 31, B 36, B 76; for Emped. note that ‘air’ in B 17.18 = aether in B 71, etc., while at B 38 aether = fire !) and would be even hazier in popular thought. In Anaximenes soul is warm air (B 1 sub fin.). When the mystery cults contrast the aethereal soul with the earth-born body (see above, note 60) it is safe to assume that the contrast is between the warm, bright aether of the soul and the cold, dark earth of the body no less than between the volatile consistency of the ‘ breath ’-soul and the heavy, compact ‘ earth ’-body. *5Cf. Diogenes of Apollonia B 7: ‘fair is] an eternal immortal body’. Jaeger says that Anaximenes ‘ did not identify it [sc. soul = air] with any corporeal substance of the world of experience ’ (p. 84). I believe that this is precisely what he did (see B 1, which offers a primitive sort of experimental evidence for the ‘warm air’ theory of soul), his choice of Air in place of Anaximander’s Apeiron being dictated by the desire to find a cosmogonic substance in ‘the world of experience ’, taking pains to explain the particular physical circumstance under which the air about us is ‘ imperceptible ’

We find it in Heracleitus,®® Diogenes of Apollonia’, and even Democritus !?® There is no reason why Anaximenes too would not have welcomed the idea on the same terms. But what would it have meant to him on these terms? In Ionian philosophy the divine is nature itself, its basic stuff and ruling principle. To say that the soul is divine is then to naturalize it; it is to say that it is subject to the same sequence of law and effect which are manifest throughout the whole of nature. And this is the very opposite of the Orphie doetrine of the divinity of the soul, whose content is rather obscure, but whose intent is perfectly clear : that the soul is not a natural, but a super-natural, entity. The word ‘supernatural’ would not, of course, figure in their vocabulary. But it would describe their meaning better than any word at our disposal. When they said that the soul was divine they meant that it is an exile from another world: its stuff is god-stuff; its powers are not bound by the limits of the observable uniformities of nature but include the oracular and other magical powers of divine being ;®® its destiny is determined not by the natural properties of air, fire, etc., but by the mysterious will of superior gods who impose on this lesser daimon the penance of transmigration, and prescribe, through their priests and prophets, the necessary purificatory rites and tabus. Could one imagaine a sharper antithesis to any doctrine of the divinity of the soul that Anaximenes could possibly have taught ? The only answer I can think of would be by way of Empedocles. Here we do have a philosopher, and of the first order, who is also a devotee of Orphic purity. If it could be shown that what his natural philosophy tells us about the soul will square with his doctrine of transmigration, we would have strong, indeed conclusive, grounds for saying that the two doctrines are not as incompatible as we must think them on the strength of all our other evidence. And this is precisely what no one has yet succeeded in showing. Jaeger does not attempt and, if I read him aright, does not intend so much. He recognized the ‘ basic incompatibility 1% between the ‘ mysticoIf we may read so much into the equation of Ovyroi dÔdvaro: at B 62 (cf. B 77). 97In the valuable testimony of Theophrastus (de sens. 42) to which Jaeger draws attention, p. 171. 98See ‘ Ethics and Physics in Democritus ’, (cited above, n. 74), pp. 580-82. %For the oracular powers of the soul which ‘ has come from the gods’ see Pindar’s famous frag. 131 Schroeder. For the sundry magical powers (including oracles) claimed by the Orphic adept our best texts are Empedocles B 111, B 112, B 129; with Empedocles’s claims (in his capacity of Orphic god-man) to control the winds and make rain, cf. those of the magicians and purifiers in On the Sacred Disease (4): ‘they profess to know how to bring down the moon, eclipse the sun, make storm and sunshine, rain and drought, the sea impassable and the earth barren’. For other instances of Orphic magic see Eur. Cycl. 647-49; and Plato Rep. 364 be, which includes black magic (katadesmoi here = katadeseis in Laws 933 a, the tabellae defixionis, on which see M. P. Nilsson, Greek Popular Religion, New York, 1940, pp. 114-15). Such things were practised no doubt only by the lowest type of Orphic priests who, however, cannot be dismissed as merely ‘ quack’ Orphics (see above, note 45); the fact that priests who invoke the authority of books by Orpheus and Musaeus for their rites practise this kind of magic is a significant commentary on the Orphic concept of the soul. 100P, 133. For a sample of the opposite view at its boldest extreme see Cornford, op. eit., p. 241: ‘Empedocles . . . exemplifies, in a most remarkable way, the... yiew that men’s cosmological views were almost entirely dictated by, and deduced

theological’ thought-forms of the Purifications and the rationalistic logic of the poem On Nature. But he still argues for some sort of imaginative coherence between the two: ‘In the mythical space of a world pervaded by divine figures, the two attitudes so irreconcilable from our abstract point of view will be seen to fit together as two distinct, but in the last analysis basically homogeneous, spheres for the interplay of divine forces ’.1 I can only say that, after reading this chapter with care and accepting gratefully much of its detail, I do not find that the promised reconciliation comes off. Not only from our ‘ abstract’ point of view, but from Empedocles’s own as well, the two pictures of reality remain not only heterogeneous but contradictory at crucial points; they admit of no rational or, for that matter, even imaginative harmony. Take, for instance, the major fragment of the Purifications where the fatality of transmigration is somehow blamed on Strife.12 ‘In this way ’, Jaeger remarks, ‘ the philosopher tries to interpret the fundamental religious facts of the Orphic theory in accordance with the supreme principle of his natural philosophy ’ (p. 150). But the natural philosophy does not single out Strife as the power which sets in motion the wheel of becoming. Love is at least as necessary.1% What is more, On Nature speaks of Love as the creator of ‘ mortal forms ’, while on Jaeger’s interpretation, this fragment of the Purifications assigns just this function to Strife. Again, if the Golden Age of the Purifications, as Jaeger interprets it (pp. 150 ff.), stands for a world where Love holds undisputed sway, the imaginative construct would clash badly with that of the physical poem. For without Strife, the cosmologist holds, there could be no world at all, good or bad, only an indiscriminate mixture, no individual gods or daemons from, their religious convictions’. A similar view is expressed in more temperate language in Cornford’s chapter on ‘ Mystery Religions and Pre-Socratic Philosophy ? in Vol. IV of the Cambridge Ancient History, pp. 566-569. 101P, 134. I cannot be sure that I understand exactly what Jaeger means here, but suspect that he reads more into the deification of the four roots in On Nature than the evidence allows. All Empedocles means by calling them ‘ gods’ is to call attention to their (a) privileged immortality against the mortality of all other physical substances «as I have argued in ‘The Physical Theory of Anaxagoras’, Philos. Rev. 59 [1950], pp. 77-78) and (6) joint and equal share along with Love and Strife in the maintenance of the cosmic order (see the paper cited above at note 16, pp. 159-160). This sort of divinity builds no bridges to the Purifications. 102B 115.13-14, ‘. . . an exile from the gods and a wanderer, for that I put my trust in insensate Strife ’ (translation after Burnet) which Jaeger takes for ‘a mythical way of expressing its (sc. the wandering soul’s) entanglement in the cosmic machinery under the rule of Hate’ (p. 150). He had offered a different interpretation in Paideia (Vol. I, 2nd ed. of Eng. tr., p. 169, with reference to B 115.9 ff.) : ‘ In the cosmos revealed by the physicists, the soul (sc. of Orphic piety) can find no home’. But if this is (as I believe) the right interpretation, I do not see how Empedocles may be said to present ‘a synthesis which shows very instructively how these two different ways of viewing the world (sc. Orphism and physiologia) could supplement and complete each other ’ (loc. cit). 103A8 Jaeger justly observes (p. 140): ‘ Hate is as necessary as Love to maintain its (sc. the world’s) dynamic structure, even though Empedocles loves Love and hates Hate’. Hence Hate’s ‘ equality’ (B 17.27) with Love, a powerful testimony to the subordination of religious feeling to the requirements of the physical system in this poem. . WCF, propévors mavrofa Sid xpdvov eidea Hvnrav, B 115.7 with Ovyr’ épéovro and xeir’ &Ovea pupia Ovyrov, mavrotars idenurv dpypéra of B 35,14-17, under the influence of PrÀórnros Opp.

or anything else.1% But the Purifications speaks of men, beasts, and birds, as well as gods, when ‘ the flames of good fellowship glowed ’ in the Golden Age (B 128). Just how then are the Purifications ‘in the last analysis basically homogeneous ’ here with the cosmology 4106 As for the concept of the soul, Empedocles’s natural inquiry presents us with a cardinal doctrine, explicit in the surviving fragments, and enormously influential on Greek medical thought. It is the well-known theory that perception and thought are the functions of the blood, which is conceived as a mixture of the four elements in equal ratio.’ But blood can hardly exist without ‘flesh’; and in the Purifications flesh is termed in Orphic fashion an ‘ alien garment ’1® which the wandering daemon puts on when, and presumably only when, he serves his sentence in the wheel of transmigration. How then does he get along without it in his first and last estate of discarnate blessedness ? How does he think the thoughts of love when he has nothing to think with ?!% It is better to drop the question and therewith the assumption that Empedoclean physics and mysticism are ‘ basically homogeneous’. Jaeger spoke of Empedocles in Paideial® as a ‘ philosophical centaur, . . . a prodigious union of Ionian elemental physics and Orphic theology’. ‘ Prodigious ’ is the right word for the union of physics and theology, as it is for the junction of immortal god and mortal flesh. The one is as much of a miracle as the other, and Empedocles doubtless devoutly believed it to be such. He left us no explanation of either, and it would be futile to try to supply it by rationalizing the theology of the mystic or mystifying the logic of the cosmologist. To set the contribution of pre-Socratic philosophy to the concept of the soul in its just historical perspective we must see how here, as in its concept of God, it is its peculiar genius to transpose a religious idea into 105Not only men, animals, and plants, but also the ‘long-lived gods’ come into being only when the world-process is set in motion €K ToÚTwv, sc. Love and Strife (B 20.9 106Two more of the many points which cannot be dealt with here: (1) ‘It has already been correctly observed that the theory of the four elements, as it is generally called, is presupposed in the Katharmoi as well’ (p. 143). This on the strength of B 115.9 ff., which, however, on Jaeger’s earlier interpretation (above, note 102), voices the hostility of the natural universe to the Orphic soul. Is there any suggestion, here or elsewhere, that the four elements would (along with Love and Strife) pervade the whole of reality in the Purifications, as they certainly do in On Nature? (2) It is not quite clear whether Jaeger means to identify (as he seems to do, pp. 140 ff.) the Sphairos of On Nature with the ‘holy and unutterable mind’ of B 134. See K. Reinhardt’s arguments against the identification by W. Kranz, Empedocles (Zurich, 1949), p. 48, in his review of this book in Classical Philology 45 (1950), pp. 176-177. 107B 98; B 105. 108B 126. To Jaeger’s comment, ‘in this image we see corporeality regarded as a mere transient, non-essential wrapping—a conception just as strange to the Greek of Homer’s time as it is to the Ionian philosopher ’ (p. 147), I would merely add: and just as strange to Empedocles himself gua philosopher and medical man. 10] cannot imagine how this and many other difficulties could be met by Cornford’s view (op. cit., p. 239 ; now revived by H. S. Long, ‘ The Unity of Empedocles’s Thought’, American Journal of Philology, 70 [1949] at pp. 156-57) that the transmigrating soul is a compound of Love and Strife. In any case, such a view remains purely conjectural ; there is no support for it in.fragments and/or testimonia, which always. treat mental processes as functions of the ratio not of Love-Strife but of earth-water-air-fire, 110Vol. I, 2nd English ed., p. 295,

the medium of natural inquiry, transforming, but not destroying, its associated religious values. It is the function of high religion to foster man’s reverence for himself no less than his reverence for God. In this respect the record of Greek religion can hardly be called a success. The fault lay at least in part with its traditional concept of the soul as the ‘image’ or ‘shadow ’ of the real man, which was somehow the principle of life but neither of feeling nor judgment.!!! This psyche’s ghostly survival after death was a wretched affair; Homer regards with mixed feelings of gloom and contempt the flitting, squeaking, batlike!!? creatures in the ‘ joyless ’ regionst!3 of nether darkness"4, This flimsy support for the sense of reverence for life could be strengthended in one of two opposite ways. The super-natural psyche could be retained and reformed, turned into an infinitely more robust affair, human and superhuman, an incarnate god or daemon, possessing in life and conserving after death all the powers of thought, will, and passion of the fullblooded man, with an extra load of divine powers thrown in for good measure.H5 This was the way of the mystery cults. The Orphic theogonies, Pythagoras, the Empedocles of the Purifications, and finally and most triumphantly Plato, adopted and justified this faith. Plato’s influence made it a dominant doctrine of Hellenistic thought; thence it passed with appropriate modifications into Christian theology, as Jaeger observes (p. 73). Whatever our estimate of the wisdom of this conception, we have no right to call it ‘the’ Greek view of the soul. No less Greek was the radical alternative chosen by the pre-Socratics. They too merged the psyche with the feeling thymos and the thinking nous. But they did so without regard to the cult and with no concessions to magic. They assumed from the start11° that they could apply to the soul the same categories of under111] follow here Jaeger’s view (p. 74 and ff. with notes) that psyche in Homer refers not only to the ghostly residue which survives death but also to the principle of life in the living person. The latter had been ignored in Rohde’s classic on this theme. Jaeger seems to make a judicious synthesis of the best features of Rohde’s work with those of his critics. 1120d. 24.1 fi. 1180d. 11.94; an idea repeatedly echoed in tragedy, Aesch. Zum. 301, 423. Eur. Or. 1082-84. IMM. P. Nilsson, ‘ The Immortality of the Soul’, Eranos 39 (1941), pp. 1-16, at p. 3: ‘The Greek Underworld was a gloomy and terrible place, and its terror consisted just in its emptiness, its nothingness. . . . the soul a worthless, useless shadow. 115See above, n. 99. The afterlife is at most a vacuum, It is evaluated accordingly ’. “Anaximander (A 11, A 30), in assuming that man arose from purely natural substances (earth and water) by a purely natural process (heating by the sun); Anaximenes (B 1, B 2), in assuming that the properties of soul are simply those of air. This initiates a tradition in which a dualistic conception of soul-body is unthinkable ; soul is body, either a particular physical substance in the body (air in Anaximenes and Diogenes, fire in Heracleitus and the atomists) or else the mixture (krasis) of the physical components of the bcdy as a whole (Parmenides) or of the blood (Empedocles). The detachable ghost-soul of Homer or God-soul of the mystery cults is precluded. Immortality is implied or asserted (Heracleitus, Alemaeon) in quite another sense, d.e. the physical survival of the substance(s) which make it up, but which does not imply the preservation of personal identity so essential for the mystery cults. Anaxagoras’s doctrine of the ‘ unmixed ’ nous is the first break in the tradition, though it is not made the basis of any doctrine of personal immortality.

standing which formed the framework of their natural inquiry. They thought of it as a part of nature, with a natural origin and a natural ending, but as no less divine for being just that, since it shared the powers of wisdom and justice writ large throughout the universe and could therefore realize within the human microcosm some measure of the order which ruled the infinite worlds. This was its destiny, natural and divine, to be that unique thing, ‘ a self-increasing logos ’.H? Except for Heracleitus, the pre-Socratics were too pre-occupied with cosmological and physiological problems to develop the potentialities of this idea. But we can see its vitality in Democritus, where the sense of reverence for man survives the loss of faith in God. He can no longer believe in the divinity of nature, but he still believes in the divinity of the soul.™® GREGORY VLASTOS Cornell University. WHeracleitus B 115. 1188ee above, note 98. 119For a thorough criticism of an earlier draft I am indebted to my colleague, Friedrich Solmsen. The opinions in this paper are my own; but many suggestions which have improved it have come from him.