the Winter solstice as sun-spells-we have another argument for the connection of the Vișuvant with the Winter sun and the Mahāvrata with the Summer solstice.

The argument clearly rests on too many hypotheses to be convincing. In the first place, it should be noted that the Āśvalāyana Śrauta Sutra1 sets only the Mahādivākīrtya Sāman in connection with the Mahāvrata. Professor Hillebrandt meets this objection by regarding the version of the ritual in Āśvalāyana as later than that in Sankhāyana, but Dr. Friedländer 2 and I have adduced a good deal of evidence to prove that the Śānkhāyana ritual generally is of a more elaborate and artificial type than that of Asvalayana, and that the relation in time of the two Sutras is the reverse of that accepted by Professor Hillebrandt. It is therefore very difficult to eliminate from the Mahāvrata the Mahādivākīrtya Sāman, which by both Sankhāyana and Āśvalāyana is assigned to the Mahāvrata. Again, there is no evidence whatever for the connection of the Rathantara and the Vișuvant beyond the improbable and unsupported conjecture that it originally occupied the place of the Mahadivākirtya. Moreover, the Bṛhat and Mahādivākīrtya cannot be assigned solely to Indra and Surya respectively, without the arbitrary elimination of portions of the received forms of these Samans as used in this ritual.*

Secondly, even if we accepted as true all these hypotheses, and assumed that the Mahavrata was connected with Indra alone and the Vișuvant with Surya alone, nevertheless we would not be bound to accept the theory that the former must be placed at the Summer solstice. There is no obvious reason why Surya should not be celebrated at the Summer solstice as at the Winter solstice, and the wheel rite of Schleswig at the Winter solstice quoted by Professor Hillebrandt may be balanced

1 viii, 6. The Kauṣītaki Brāhmaṇa, xxv, 4, on the other hand, mentions that both the Brhat and Mahādivākirtya Samans were by some assigned to the Visuvant. The Sankhayana and Aitareya Aranyakas recognise both Brhat and Rathantara for the, Mahāvrata.

2 In his edition of the Sankhāyana Āraṇyaka, Mahāvrata section, pp. 9 seq.

Aitareya Aranyaka, pp. 30 seq.

Cf. also Sankhayana Srauta Sutra, xi, 14, 8; Eggeling, S.B.E., xli, p. xv,

n.

by the similar use of a wheel at the Summer solstice at the present day both in France and Germany, a custom which, according to a medieval writer cited by Frazer,1 was one of the three great features of the Midsummer ritual. In the case of Indra we can now quote Professor Hillebrandt against himself, for he has in his Vedische Mythologie 2 abandoned the idea that Indra's foe, Vṛtra, is a drought demon, and now finds in him the Winter, without, however, giving up his theory of the Mahāvrata. It is not necessary here to discuss in detail how far the view, which converts Indra into a sun-god, is an accurate representation of the facts of the Rgveda as they stand. It is sufficient for our purpose to accept the view of Weber that the conflict of the sun and Winter is Indo-European, or at least Indo-Iranian, and that this conflict is inseparably confused and combined with the later and more specially Indian conception— naturally adopted under the climatic conditions of Hindustanof a conflict between the drought and the thunderer.

The further arguments adduced by Professor Hillebrandt may be dismissed more briefly. The third ground brought forward by him is the fact that the Vișuvant forms the middle of a period of twenty-one days, and this period may be compared with a period of like duration, apparently dating from the end of November or the beginning of December, of which faint traces are found in German mythology. But no stress can be laid on this argument, for no special significance attaches to the period of twenty-one days in the Vedic ritual—it is merely one of various similar groupings-while the Germanic evidence is not merely very scanty and doubtful, but, if it shows anything, shows that the period lay just before the Winter solstice, whereas the Vișuvant is preceded by and followed by ten days on either side.

The next argument rests on the fact that according to one theory mentioned in the Kausītaki Brāhmaṇa, xix, 2, the Mahavrata would have fallen in the month Taișa. This month derives its name from the asterism Tiṣya, which is equated with the

2 iii, 162 seq.

1 Frazer, Golden Bough, ii, 260 seq.

3 See Bloomfield, Vedic Religion, pp. 179 seq.; J.R.A.S., 1908, p. 883.

1

Avestan Tistrya and that again with Sirius. Sirius represents the heat of Summer, and hence it is deduced that the Mahāvrata must fall in the Summer. But even if we accept the equation of Tistrya and Sirius, which is by no means without phonetic difficulties, there remains the fact that there is no evidence that Tiṣya was ever to the Vedic Indians a Summer month. The asterism Tisya in the Taittirīya Samhita2 holds the same position to the others as Pusya in the Atharvan 3 list, and the commentators on the passages where Taiṣa as a name of a month is found concur in equating it with Pausa, while Professor Hillebrandt himself admits that this was probably already the view of the Kauṣitaki Brāhmaṇa.

3

4

It is not necessary to discuss the minor arguments adduced by Professor Hillebrandt, as he naturally lays no stress on the allegorical plays on the Samans and on their connection with the length of the day, etc., which he adduces and explains on his theory just as little convincingly as on any other. It remains, however, to be seen whether the actual rites throw any light on the season at which they were held.

"

The chief characteristic of the rite is the bird shape 5 ascribed to the litany, the Mahad Uktha, as also to the altar and to the sacred fire. Now the bird is undoubtedly the sun-bird, for it is addressed in the ritual by the word garutman, winged,' which in the Rgveda itself denotes the sun-bird." Both the sun (Aditya) and the fire (Agni) receive formal worship, and there can be no doubt of the sun-character of the swing which is set up and pushed from east to west by the priests. Already in the Ṛgveda the sun is described as the golden swing in

8

1 Cf. Weber, Altiran. Sternnamen, p. 15; Zimmer, Altind. Leben, p. 355. Professor Mills kindly informs me that he thinks the identification probable. Cf. also Geigen, Ostiran Cult., p. 708; Roth, Z.D.M.G., xxxiv, 713. 3 xix, 7, with Lanman's note.

2 iv, 4, 10, 1.

4 One, from the use of ekāṣṭaka as the mother of Indra, he has withdrawn; see Vedische Mythologie, iii, 198, n. 2.

5 So in the certainly older version of the Aitareya; the Śānkhāyana presupposes the human form of fire, altar, and hymn; cf. Friedländer, op. cit., p. 10; above, p. 1, n. 2.

6 Śānkhāyana Āraṇyaka, i, 8; Aitareya Āraṇyaka, v, 1, 5, with Sāyana's note.

Cf. Macdonell, Vedic Mythology, p. 39.

8 vii, 87, 5.

•

heaven, and the direction of the motion is extremely significant. Moreover, in one of the formulæ accompanying the bringing of the swing into contact with the ground, occur the words, the great hath united with the great,' this being explained as Agni-here clearly in his celestial form-with the earth.1 Still more significant is a struggle between an Arya, normally a Vaisya, and a Śūdra for a round white skin, which is won by the Arya and used by him to strike down his defeated rival. The old tradition, already known to the Kāṭhaka, equates the skin and the sun,2 and, like all the other details mentioned, this equation suits admirably the conception of the rite as an attempt to stimulate the sun at the Winter solstice both by worship and by magic. The movement of the swing stimulates its motion; the Arya rescues it from the hostile powers which threaten to extinguish its light. Neither act is quite so appropriate at the Summer solstice, when the sun's heat is strong and needs no recruiting.

In this connection can also be explained the use of a drum by the priest and of various musical instruments—a long list of names of these instruments is given made up of rare popular words-by women, whose presence and activity are characteristic of the popular character of the ritual. These noises may have been designed, like the gong at Dodona, to drive away evil demons, and to protect at once the sun and the performers of the rite from their onslaught, and the sounds of the musical instruments were reinforced by the shout of the priests. Professor Hillebrandt prefers to regard the use of the drum as an imitation of the thunder, designed to evoke real thunder, but the simpler explanation is here quite adequate.

3

Nor can any support of Professor Hillebrandt's theory be

1 Śańkhāyana Aranyaka, i, 5.

2 See Oldenberg, Religion des Veda, p. 88, n. 4, and Kathaka Samhitā, xxxiv, 5, cited by Weber, Ind. Stud., iii, 477, which Oldenberg appears to have overlooked. This passage shows clearly that the analysis of sudrāryau as sūdra and arya is incorrect and strengthens Geldner's view that arya never means the Aryan (Ved. Stud., iii, 94–7).

Cf. Frazer, Adonis, Attis, Osiris, pp. 164 seq.; Warde Fowler, Roman Festivals, pp. 40 seq.; Crooke, Popular Religion, pp. 60, 108; Cook, J. H.S., xxii, 4, 20-8; Hillebrandt, Vedische Mythologie, ii, 32.

derived from other parts of the rite. The performance was accompanied by a running commentary of praise and criticism by two persons selected for that purpose. This is probably a priestly refinement, for this feature of the rite is one of the least well authenticated. On the other hand, all the versions agree in mentioning the brahmacāripumścalyoḥ sampravāda, a contest in ritual aioxpoλoyía between a hetaera and a Brahmin scholar vowed to chastity. Various theories have been advanced to explain such instances of aioxpoλoyía; the simplest 2 perhaps is that it is merely another method of demon-scaring, but the evidence for this view is hardly convincing, and it seems best to regard the ritual here, as in the Thesmophoria,3 as undoubtedly calculated to promote the fertility of beings and the earth. This view is probably rendered certain by a further custom merely referred to in the Aitareya Āraṇyaka1 in the terse words bhūtānāṁ ca maithunam, and by a singular example of priestly or general moral progress repudiated as puranam utsannam in the Sankhāyana Śrauta Sutra," but referred to in the Taittirīya Samhitā and described fully in the other Sūtra accounts. This rite must be compared with the iepòs yáμos7 of the Greek and Asianic ritual and brought into connection with the symbolic union of earth and sun, the touching of the earth by the swing. Probably originally the rite was one celebrated by two Aryans, the pumścali, like the Māgadha of Latyāyana, representing a degradation of the rite, and was a solemn ceremony, at once a counterpart of the union of sun and earth, whence sprang fertility for the crops, and a powerful spell to promote human fertility. All this is perfectly

1 It is noticed in neither of the Rgvedic Aranyakas, but it occurs in Taittirīya Samhitā, vii, 5, 9, 3.

2 Cf. Campbell, Ind. Ant., xxiv, 263 seq.

3 Farnell, Cults of the Greek States, iii, 104.

4

v, 1, 5.

5

xvii, 6, 2. The sense of utsanna is here made clear by the continuation na kāryam.

6 The plural in the Aitareya suggests that the ritual involved the union of more pairs than one, representing the different sides of animal life. The Taittiriya, vii, 5, 9, 4, has only one pair.

7 I assume that the iepòs yάuos is a remnant of a real marriage, not a mere symbolism, but intended to promote fertility as a magic spell. Cf. Farnell, op. cit., i, 184 seq.