Sacred Texts  Hinduism  Satapatha Brahmana  Index  Previous  Next 
Buy this Book at

Satapatha Brahmana Part V (SBE44), Julius Eggeling tr. [1900], at

p. xiii


THE present volume completes the theoretic exposition of the sacrificial ceremonial, and thus brings us to the end of our task. The remaining six chapters of the last book of the Brâhmana form the so-called Brihad-âranyaka, or great forest-treatise, which, as one of the ten primitive Upanishads, is included in Professor F. Max Müller's translation of those old theosophic treatises, published in the present series. The portion of the work contained in this volume forms practically continuation of the first five kândas, the intervening five books being devoted to the consideration of the Agnikayana, or construction of the sacred brick-altar, which had come to be recognised as an important preliminary to the Soma-sacrifice. The circumstances which seem to have led to this somewhat peculiar distribution of the different sections of the work have been explained in the introduction to the first volume of the translation. As was there shown, the inclusion of the Agnikayana in the sacrificial system of the Vâgasaneyins, or theologians of the White Yagus, appears to have resulted in a definite settlement of the sacrificial texts of the ordinary ritual, as contained in the first eighteen adhyâyas of the Vâgasaneyi-samhitâ, as well as of the dogmatic explanation of that ritual as given in the first nine kândas of the Satapatha-brâhmana. Considerable portions of the remaining sections of both works may have been, and very likely were, already in existence at the time of that settlement, but, being excluded from the regular ceremonial, they were naturally more liable to subsequent modifications and additions than

p. xiv

those earlier sections which remained in constant use. Whilst the tenth kânda, included in the preceding volume of the translation, consisted of speculations on the sacred fire-altar, as representing Purusha-Pragâpati and the divine body of the Sacrificer--whence that book is called the Agnirahasya, or mystery of the fire-altar--the present volume contains the supplementary sections connected with the sacrificial ceremonial proper.

The eleventh and twelfth kândas are mainly taken up with additional remarks and directions on most of the sacrifices treated of in the first four kândas, especially with expiatory ceremonies and oblations in cases of mishaps or mistakes occurring during the performance, or with esoteric speculations regarding the significance and mystic effect of certain rites. In this way the eleventh book deals with the New and Full-moon sacrifices; the Seasonal offerings (XI, 5, 2), the Agnihotra (XI, 5, 3; 6, 2), the Soma-sacrifice (XI, 5, 5; 9), and the Animal-sacrifice (XI, 7, 2-8, 4); whilst the twelfth kânda treats of the 'Gavâm ayanam'--or most common sacrificial session lasting for a year, thus offering a convenient subject for dilating upon the nature of Pragâpati, as the Year, or Father Time;--of additional expiatory rites for Soma-sacrifices (XII, 6), and of the Sautrâmanî, consisting of oblations of milk and spirituous liquor, supposed to obviate or remove the unpleasant effects of any excess in the consumption of Soma-juice (XII, 7-9). Though supplementary notes and speculations on such ceremonial topics cannot but be of a somewhat desultory and heterogeneous character, they nevertheless offer welcome opportunities for the introduction of much valuable and interesting matter. It is here that we find the famous myth of Purûravas and Urvasî (XI, 5, 1); and that of Bhrigu, the son of Varuna, vividly illustrating the notions prevalent at the time regarding retribution after death (XI, 6, 1) as also the important cosmogonic legend of the golden egg from which Pragâpati is born at the beginning of the evolution of the universe (XI, 1, 6). Of considerable interest also are the chapters treating of the way in which

p. xv

the dead body of the pious performer of the Agnihotra, or daily milk-offering, is to be dealt with (XII, 5, 1-2); of the initiation and the duties of the Brâhmanical student (XI, 3, 3; 5, 4); and, last not least, of the study of the Vedas (XI, 5, 6-7) and their subsidiary texts amongst which we meet, for the first time, with the Atharvâṅgiras as a special collection of texts recommended for systematic study.

With the commencement of the thirteenth kânda, we enter once more upon a regular exposition of a series of great sacrifices like those discussed in the early books; the first and most important of them being the Asvamedha, or Horse-sacrifice. Like the Râgasûya, or inauguration of a king, the Asvamedha is not a mere sacrifice or series of offerings, but it is rather a great state function in which the religious and sacrificial element is closely and deftly interwoven with a varied programme of secular ceremonies. But whilst the Râgasûya was a state ceremonial to which any petty ruler might fairly think himself entitled, the Asvamedha, on the contrary, involved an assertion of power and a display of political authority such as only a monarch of undisputed supremacy could have ventured upon without courting humiliation 1; and its celebration must therefore have been an event of comparatively rare occurrence. Perhaps, indeed, it is owing to this exceptional character of the Asvamedha rather than to the later origin of its ritual and dogmatic treatment that this ceremony was separated from the Râgasûya which one would naturally have expected it to succeed. It is worthy of remark, in this respect, that, in Kâtyâyana's Anukramanî to the Vâgasaneyi-samhitâ, the term 'khila,' or supplement, is not applied to the Asvamedha section 2 (Adhy. XXII-XXV), while the subsequent sections are distinctly characterised as such. As a matter of fact, however, the Asvamedha has

p. xvi

received a very unequal treatment in the different rituals. Of the two recensions of the Brâhmana of the Rig-veda priests, the Aitareya-brâhmana takes no account whatever of the Horse-sacrifice, whilst its last two books (VII, VIII)--generally regarded as a later supplement, though probably already attached to the work in Pânini's time--are mainly taken up with the discussion of the Râgasûya. The Kaushîtaki-brâhmana, on the other hand, passes over both ceremonies, their explanation being only supplied by the Sâṅkhâyana-sûtra, along with that of some other sacrifices, in two of its chapters (15 and 16), composed in Brâhmana style, and said to be extracted from the Mahâ-Kaushîtaki-brâhman1. In the principal Brâhmana of the Sâman priests, the Pañkavimsa-brâhmana, the Asvamedha, as a trirâtra, or triduum, is dealt with in its proper place (XXI, 4), among the Ahînas, or several days’ performances. As regards the Black Yagus, both the Kâthaka and the Maitrâyanî Samhitâ give merely the mantras of the Asvamedha 2, to which they assign pretty much the same place in the ritual as is done in the White Yagus. In the Taittirîya-samhitâ, on the other hand, the mantras are scattered piecemeal over the last four kândas; whilst, with the exception of a short introductory vidhi-passage, likewise given in the Samhitâ (V, 3, 12), the whole of the exegetic matter connected with this ceremony is contained, in a continuous form, in the Taittirîya-brâhmana (VIII and IX). Lastly, in the Vaitâna-sûtra of the Atharva-veda--doubtless a comparatively late work, though probably older than the

p. xvii

[paragraph continues] Gopatha-brâhman1--the Asvamedha is treated immediately after the Râgasûya, and followed by the Purushamedha and Sarvamedha; these four ceremonies being characterised at the end as the Kshatriya's sacrifices 2 (medha).

With regard to the earliest phase of Vedic religion, there is no direct evidence to show that the horse-sacrifice was already at that time a recognised institution. Two hymns of the Rig-veda (I, 162; 163), it is true, relate to that sacrifice 3, but they evidently belong to the latest productions 4 of that collection, though still sufficiently far removed from the time of the oldest of the ritual works just referred to. Seeing, however, that animal sacrifices generally are not alluded to in the Riksamhitâ 5, whilst there is every reason to believe that they were commonly practised from remote antiquity, this absence of earlier positive evidence regarding the horse-sacrifice cannot be taken as proving the later origin of that institution. As will be seen further on, there are sufficient indications to show that even human sacrifices were at one time practised amongst the Aryans of India, as they were amongst their

p. xviii

[paragraph continues] European kinsmen. The fundamental idea which underlay this practice doubtless was the notion that man, as the highest attainable living being, could not but be the most acceptable gift that could be offered to the gods, and, at the same time, the most appropriate substitute for the human Sacrificer himself. For the same reason no doubt only domesticated animals were considered suitable for sacrifice; and amongst these the horse was naturally looked upon as ranking next to man (Sat. Br. VI, 2, I, 2), although considerations of practical expediency and even of social distinction might prevent its use for ordinary sacrificial purposes.

In the speculations of the Brâhmanas, a deep mystic significance is attached to the Horse-sacrifice. In the last two chapters of the 'Mystery of the Fire-altar' (Sat. Br, X, 6, 4, I; 4), the Asvamedha--i.e. the sacrificial horse itself--is coupled with the Arka, the mysterious name of the sacred fire, as the representative of Agni-Pragâpati, the Sun. The horse-sacrifice is called the bull (XIII, 1, 2, 2), and the king (XIII, 2, 2, 1), of sacrifices, just as the horse itself is the highest and most perfect of animals 1 (XIII, 3, 3, 1; Taitt. Br. III, 8, 7; 8, 9, 1); the horse selected for sacrifice, in particular, being said to be worth a thousand cows (XIII, 4, 2, 1). The connection of the sacrificial horse with 'the lord of creatures' is, of course, fully accounted for by the theory of the identity of the sacrifice generally with Purusha-Pragâpati, discussed in the introduction to part iv of this translation. The sacrificial horse accordingly belongs to Pragâpati, or rather is of Pragâpati's nature (prâgâpatya); nay, as the Taitt. Br. (III, 9, 17, 4) puts it, it is a form of Pragâpati himself (pragâpate rûpam asvah), and is, of all animals, the one most conformable (anurûpatamah) to Pragâpati. Hence also, in the cosmogonic account at the commencement of the Agnikayana section (VI, 1, 1, 11), the horse is represented as having originated, immediately after the Brahman (sacred lore) and Agni, directly from the

p. xix

egg produced by Pragâpati from the cosmic waters; whilst, according to other accounts (VII, 5, 2, 6; XIII, 3, 1, 1), the horse originated from Pragâpati's eye. But, since the offering also represents the offerer himself, or rather his divine self awaiting him in the other world (XI, 1, 8, 6; 2, 2, 6), the sacrificial horse is also identified with the Sacrificer (yagamâno vâ asvah, Taitt. Br. III, 9, 17, 4) who thereby obtains the fellowship of the Lord of creatures and a place in his world (ib. III, 9, 20, 2).

Besides Pragâpati, there is, however, another deity who lays claim to the possession of the sacred steed; for the horse is Varuna's sacrificial animal (Sat. Br. V, 3, 1, 5; VI, 2, I, 5; Taitt. Br. III, 9, 16, 1); nay, Varuna is even the lord of all one-hoofed cattle (Vâg. S. XIV, 30; Sat. Br. VIII, 4, 3, 13). This connection of the horse with Varuna seems natural enough, seeing that this god, as the king of heaven and the upholder of the law, is the divine representative of the earthly king; whence the Râgasûya, or coronation-ceremony, is called Varuna's consecration (Sat. Br. V, 4, 3, 21; cf. II, 2, 3, 1). For this reason the barley also is sacred to Varun1 (XIII, 3, 8, 5); and accordingly, during the same ceremony, the king offers a barley-mash to Varuna, in the house of his Sûta, or charioteer and herald; a horse being the sacrificial fee for this offering (V, 3, I, 5). In the Vedic hymns, this association o1 the god Varuna with the noble quadruped finds a ready, If rather commonplace, explanation in a common natural phenomenon: Varuna's horse is none other than the fiery racer who pursues his diurnal course across the all-encompassing arch of heaven, the sphere of Varun2, the all-ruler. It is in the

p. xx

form of the horse that the Sun is thus lauded in the hymn Rig-v. I, 163, recited by the Hotri on the second Soma-day of the Asvamedha, after the horse has been led up to the sacrificial stake, and to the slaughtering-place 1:--1. 'When, first born (just born), thou didst neigh, uprising from the sea, or from the vapoury region, the falcon's wings and the deer's feet--praiseworthy greatness was innate in thee, O steed!' . . . 4. 'Three fetterings 2, they say, are thine in the sky, three in the waters, three within the sea; and like unto Varuna, O steed, dost thou appear to me, where, they say, thy highest birth-place is.' And since, as in these verses, the upper regions commonly present themselves to the eye of the Vedic singer under the semblance of a heavenly sea, Varuna also comes to be looked upon as the divine representative of the waters; whilst the horse, for the same reason, is supposed to have sprung from the waters. Of any connection of the sacrificial horse with Pragâpati, on the other hand, as of the Pragâpati theory of the sacrifice generally, clearly shadowed forth in the Purusha-sûkta, and so decidedly dominant during the Brâhmana period, no trace is to be found in the earlier hymns. Indeed, if we have any right to assume that the horse-sacrifice was known and practised in the earlier times, it can scarcely be doubted that King Varuna must have been the deity to whom this victim was chiefly consecrated.

The close and natural relations between the sun and the heavens find their hallowed expression in the divine duad Mitra and Varuna. Though, judged by the number of

p. xxi

hymns addressed to them, singly or jointly, this pair of deities occupies a somewhat subordinate position in the Vedic pantheon, there is reason to believe that it formed a more prominent feature of a phase of belief lying beyond the period reflected in the hymns of the Rig-veda. Judging from the peculiar character of these deities, one might indeed be inclined to claim for the people that formed religious conceptions such as these a long period of peaceful dwelling and normal intellectual growth. if such was the case, the occupation of the land of the seven rivers and the gradual eastward drift certainly proved a turning-point in the development of this Aryan people. But, in any case, the decided change of climate 1, and the close contact with aboriginal tribes of inferior culture, could hardly fail, along with the changed conditions of life, to influence considerably the character of the people, and to modify their religious notions and intellectual tendencies. As, in their struggles against hostile tribes, the people would naturally look to leaders of deed and daring rather than to mild and just rulers, so the violent war of elements, periodically convulsing the heavens in these regions, after long and anxious seasons of heat and drought, and striking awe and terror into the minds of men, might seem to them to call for a heavenly champion of a different stamp than the even-headed and even-tempered Varuna,--it would need a divine leader of dauntless, and even ferocious, spirit to fight the worshipper's battle against his earthly and unearthly foes. Such a champion the Vedic Aryans indeed created for themselves in the person of Indra, the divine representative, as it were, of their warlike kings, and the favourite subject of their song. And side by side with him, and sharing with him the highest honours--nay, even taking precedence of him--we find the divine priest, Agni, the deified fire of sacrifice, as representing the all-pervading, all-supporting

p. xxii

light of heaven; just as we found Mitra, the sun, by the side of Varuna, the god of the all-encompassing heaven. Not as if Agni and Indra had ever entirely superseded Mitra and Varuna. On the contrary, all these gods continue to share, in a greater or less degree, the affections of the Vedic singers; and as regards Varuna and Indra in particular, their relations are well expressed by Vasishtha when he says (Rig-v. VII, 82, 2; 5), that the one (Varuna) is 'samrâg' (universal ruler, overlord); and the other (Indra) 'svarâg' (self-ruler, independent lord);--and that, ever since the time when these two, by their power, created all the beings in the world, Mitra serves Varuna in peace, whilst the mighty (Indra) goes forth with the Maruts in quest of glory. Even in the sacrificial ritual, Mitra and Varuna continue to play an important part, seeing that one of the priests--the Maitrâvaruna--is named after them, that they receive various oblations, and that at the end of every Soma-sacrifice at least one sterile cow is offered to them, apparently as an expiatory victim, for shortcomings in the sacrifice 1, thus accentuating once more the ethical character of these deities. It is thus not to be wondered at that, whilst Agni and Indra are most commonly referred to in the Brâhmanas as the divine representatives of the Brahman and Kshatra, or the spiritual and the political powers-the high priest and king-respectively, the very same is the case as regards Mitra and Varun2; and the Maruts, representing the common people, are accordingly associated with Varuna,

p. xxiii

as their king or ruler (Sat. Br. II, 5, 2, 34), just as they are with Indra (II, 5, 2, 27). One might thus expect that Indra would claim the same special connection 1 with the sacrificial horse as that which is conceded to Varuna. The reason why this is not the case probably is that, in the Brâhmana period, the notion of the horse having, like the sun, originated from the cosmic waters had become as firmly established as was the traditional connection--nay, even identity 2--of Varuna with the element of water generally.

As regards Varuna's and Pragâpati's joint connection with the sacrificial horse, the Taitt. S. (II, 3, 12, 1) records the following legend which may perhaps have some bearing on this point:--Pragâpatir Varunâyâsvam anayat, sa svâm devatâm ârkhat, sa pary adîryata, sa etam vârunam katushkapâlam apasyat, tam nir avapat, tato vai sa varunapâsâd amukyata, Varuno vâ etam grihnâti yosvam pratigrihnâti, yâvatosvân pratigrihnîyât tâvato vârunâñ katushkapâlân nir vaped, Varunam eva svena bhâgadheyenopa dhâvati, sa evainam varunapâsân muñkati:--'Pragâpati led up the horse to Varuna: he (thereby) impaired his own godhead, and became racked all over with dropsy. He beheld that four-kapâla (cake) sacred to Varuna, and offered it, and thereupon was freed from Varuna's noose; for Varuna seizes him who takes (receives) a horse,--as many horses as one takes so many four-kapâla (cakes) one ought to offer to Varuna: one (thereby) hastens up to Varuna with his (V.'s) own share, and he (V.) frees him from Varuna's noose.'

The interpretation of this legend presents, however, some difficulties. Dr. Hillebrandt, Varuna and Mitra' (p. 64), translates the first sentence by--'Pragâpati führte dem

p. xxiv

[paragraph continues] Varuna das Ross fort 1'--'Pragâpati led the horse away froth Varuna': this would undoubtedly make better sense, but, unfortunately, the construction of 'nî' with the dative in this sense would involve a decided solecism. Sâyana, on the other hand, takes it in the same sense as we have done, and he explains that it is just by giving away the horse whose deity he himself is that Pragâpati forfeits his godhead 2. He feels, however, afterwards constrained to assign to 'pratigrihnâti' the causal force of 'he causes it to be taken, he gives it away,' which is clearly impossible. But whatever the correct interpretation of the opening clause may be, it seems at all events-clear that the sacrificial horse is represented in the legend as undergoing a change of ownership from the one deity to the other.

When one compares the ceremonial of the Asvamedha, as expounded in the Brâhmana, with the ritual indications contained in the two hymns already referred to, one is struck by the very marked contrast between the two. For whilst, on the central day of the Asvamedha alone, the ritual requires the immolation of not less than 349 victims bound to twenty-one stakes (p. 311, n. 1)--not counting two sets of eleven Savanîya victims (p. 383, n. 3) subsequently added thereto--the hymns (I, 162, 2-4; 163, 12) seem only to mention two victims, viz. the horse itself, and a he-goat. This latter animal which is to precede the horse when led to the sacrificial ground (and stake), and to

p. xxv

be slaughtered first in order to carry the welcome news of the sacrifice to the gods, is in one place referred to as Pûshan's share, and in another as going forward to the dear seat of Indra and Pûshan 1. Sâṅkhâyana (Srautas. XVI, 3, 27-30), however, takes these statements of the Rishi to refer to two different he-goats, both of which he includes amongst the victims tied to the horse's limbs, viz. one, sacred to Pûshan, tied to the forehead, and another, sacred to Indra and Pûshan, fastened to the navel, of the horse 2. The corresponding 'paryaṅgya' victims recognised by the Maitrâyanî Samhitâ (III, 13) and the White Yagus (Vâg. S. XXIV, 1), on the other hand, are a black-necked he-goat for Agni, tied to the forehead, and a black or grey (syâma) one, bound to the navel, and consecrated by the one authority to Pûshan, and by the other to Soma and Pûshan. But, curiously enough, the Taittirîya school (Taitt. Br. III, 8, 23; Âp. Sr. XX, 13, 12) recognises not only Sâṅkhâyana's two victims, but also the one for Agni; whilst in regard to the other victims also it differs considerably from the other schools of the Yagur-veda. Seeing, then, that there is so little agreement on these points even amongst different branches of the same Veda, one can hardly escape the inference that, in this respect at least, there was no continuity of ritual practice since the time of those two hymns. As regards the other points therein alluded to, the he-goat and horse are referred to

p. xxvi

as being led round thrice in accordance with the sacred ordinance. Now, this ceremony is quite foreign to the later practice in animal sacrifices. Sâyana accordingly takes it to refer to the rite of 'paryagnikarana,' or carrying fire round the victims 1; but the text of the passage evidently does not admit of such an interpretation; and, besides, in Rig-v. X, 155, 5, the sacrificial cow is apparently referred to as first being led round, and then fire being carried round it. It is therefore more probable that the victims were in the first place made to circumambulate the fire, or the fire and stake combined.

Further, the allusion to the pasu-purodâsas, or cakes offered in connection with the victims, as well as to the two cloths and the piece of gold placed on the ground, as they are in the later practice, for the dead horse to lie upon, might seem to suggest that even then this sacrifice was not performed in quite so simple a manner, but somewhat more in accordance with the later ceremonial than the scanty allusions in the hymns might lead one to suppose. At all events, however, we shall probably not be far wrong in assuming that, from the very beginning, the performance of the horse-sacrifice must have had connected with it a certain amount of ceremonial of a purely secular and popular character. Even at the time of the fully developed ritual this was almost certainly the case to a larger extent than would appear from the exposition of it given in the Brâhmanas and Sûtras which, indeed, are mainly concerned with the religious side of the ceremonial. For this reason considerable interest attaches to the description of the horse-sacrifice given in the Âsvamedhika-parvan of the Mahâbhârata in which much greater stress is laid on the popular and chivalrous aspect of this religious observance. Though this epic account manifestly emanates from a much later period 2, it seems, upon the whole; to present the

p. xxvii

traditional features of this royal ceremony, embellished no doubt by all the exercise of that poetic fancy to which the occasion so readily lends itself.

On the completion of the great war between the Pândava and Kaurava princes, Yudhishthira, having re-ascended the throne of his fathers, resolves on performing the horse-sacrifice, as calculated to cleanse him of all guilt 1 incurred by the slaughter of his Kaurava kinsmen. Having been initiated on the day of the Kaitra full-moon (beginning of spring), 'the king, clad in a linen (? silk) garment and the skin of a black antelope, bearing a staff in his hand, and wearing a gold wreath, and a round gold plate 2 round his neck, shone like a second Pragâpati at the holy cult.' The chosen steed 3, of black and white colour like the black buck, is then led up, and is set free by the sage Vyâsa himself; and that model of knightly perfection, Arguna, the king's second brother, is appointed to guard the priceless victim during its year's roaming. He accordingly starts after it on his chariot yoked with white steeds, attended by a picked body-guard 4, amidst the rejoicings and fervent blessings of all Hastinâpura--men, women, and children. Thus followed by its martial escort, the noble steed roams at will over the lands

p. xxviii

over which sovereign sway is claimed by the Pândava king--to wit, the whole of India from sea to sea--first pressing eastwards towards the sea, then turning southwards along the eastern shore as far as the extreme point of the peninsula, and finally northwards again, on its homeward way, along the western coast. Time after time the determined attempts to impede its progress, or even to capture and retain it as a precious trophy and token of national independence, are successfully repelled by the dauntless son of Prithâ; but, mindful of his brother's injunctions, he spares the lives of the kings and princes who oppose him, and, having obtained their submission, he invites them to attend the sacrifice of the horse at Hastinâpura. On the other hand, not to take up the challenge implied in the progress of the horse was considered a sign of weakness or cowardice. Thus the king of Manipura is censured severely by Arguna for receiving him meekly, accompanied only by Brahmans and with presents to offer to the intruder, being told that he had lamentably fallen away from the status of a Kshatriya, and acted the part of a woman. At length tidings of the approach of the horse reach the king, and forthwith preparations are made for getting ready the sacrificial ground, and to provide accommodation, on a right royal scale, for the numerous, guests expected to witness the ceremonial. Specimens of all available species of animals are brought together to serve as victims 1 along with the sacred horse; and dialecticians,

p. xxix

eager to vanquish one another, foregather to discuss the nature and origin of things. At last Arguna arrives, and, having met with an enthusiastic welcome, he 'takes repose like a seafaring man who has reached the shore after crossing the ocean.' Then commences the performance of the sacrifice, the general outline of which, as sketched in the epic 1, fairly corresponds to the ordinary ceremonial; the chief points of difference being the form and material of the altar, which is described as three-cornered, like the heavenly bird Garuda, and as being composed of a 'trunk,' measuring eighteen cubits, and made, like the wings, of gold bricks,--the structure thus shining like the altar of Daksha Pragâpati. The sacrifice over, a great public festival ensues for which 'mountains of food and sweetmeats, rivers of spirituous and other beverages, and lakes of ghee' are provided, and the feasting goes on through day and night till every one has had his fill,--a festival, indeed, of which the poet remarks people continued to talk to his day.

From the fanciful narrative of Arguna's martial exploits whilst following his precious charge, one could not of course venture to draw any conclusion as to the kind of adventures the sacred horse might have met with, at the time of the Brâhmana, during the period of its roaming at large. As a rule, however, the closely-watched animal would probably not range very far from the place where the sacrifice was to be performed; and though its body of guardians were not permitted at any time to force it to retrace its steps, they could have had little difficulty in keeping it within a certain range of grazing. Indeed, on the occasion of King Dasaratha's Asvamedha 2, described in the first canto of the Râmâyana, no mention whatever is made of

p. xxx

anything having happened to the horse during its time of grace. The expedient mentioned in the Brâhmana (XIII, 4, 2, 5) that a hundred worn-out horses should be sent along with the horse to keep it company would doubtless, as a rule, prove a sufficient check; but seeing that neither the Taitt. Brâhmana nor Sâṅkhâyana alludes to this expedient, it is probably meant as a practical suggestion rather than as a positive injunction. That the horse intended for sacrifice was by no means always safe from violent assaults 1 is clear from the directions given in the Brâhmanas as to what should be done in the event of foes getting possession of it 2. Even more pointed, in this respect, are the stanzas quoted in our Brâhmana (XIII, 5, 4, 21. 22),--'Satânîka Sâtrâgita seized a sacrificial horse in the neighbourhood, the sacrifice of the Kâsis, even as Bharata (seized that) of the Satvats. The mighty Satânîka having seized, in the neighbourhood, Dhritarâshtra's white sacrificial horse, whilst roaming at will in its tenth month 3, the son of Satrâgita performed the Govinata (form of) sacrifice.' As a rule, however, the fortunes of the roaming horse would doubtless depend largely on personal circumstances. Whilst a strong ruler who had already made his power felt amongst his neighbours would probably run little risk of having his consecrated victim kidnapped even though it were to stray beyond its master's boundaries, a prince of greater pretensions than resources might find it very difficult to secure the safety of his horse even if it kept well within the territory over which he ruled. In any case, however, the capture of the noble beast would doubtless

p. xxxi

cause not a little bad blood, and might lead to complications and struggles not less serious than those occasioned by Vasishtha's cow, or, in Irish legend, by the brown bull of Queen Medb (Mab) of Connaught.

Whilst the epic account of the Asvamedha thus presents an instructive, though extravagant, illustration of possible occurrences during the preliminary period of the sacrifice, some items of the ceremonial on which further information might have been acceptable are altogether ignored in it. Two of these at least one might have expected to find mentioned there, seeing that they are of special interest to Kshatriyas, viz. the practice of a Brâhmana and a Kshatriya lute-player singing 1, morning and night, stanzas composed by themselves in honour of the king; and the so-called 'revolving legend' (XIII, 4, 3, 1 seqq.) related by the Hotri, in a ten days’ cycle all the year round. It is especially in regard to this latter point that the statements of the ritualistic works might with advantage have been supplemented. During the ten days’ cycle a different god, or some mythic personage, is assumed, on each successive day, to be king, having some special class of beings assigned to him as his subjects, and a certain body of texts as his Veda from which a section is then recited. But from the particulars given it even remains uncertain whether any legend connected with the respective deity was actually related; whilst regarding the form and nature of some of the specified texts--such as the sarpavidyâ (snake-science), devaganavidyâ (demonology), mâyâ (or asuravidyâ. magic art)--we really know next to nothing. Nay, even regarding the Itihâsas and Purânas, likewise figuring as distinct texts, additional knowledge would by no means be unwelcome. And though regarding some of the divinities referred to the Hotri might easily have made up some kind of short tale, others would have required some

p. xxxii

exercise of ingenuity, unless he had at his disposal materials other than those accessible to us. As a rule, however, legends of this kind would seem to have been of the simplest possible description, as may be gathered from the particulars regarding the 'Nârâsamsâni,' or recitals in praise of (pious) men, which, according to Sâṅkhâyana (XVI, 11), take the place of the 'revolving legend' in the ten days’ cycle of the Purushamedha. The Hotri's recitals on that occasion consist simply of certain verses, or hymns, of the Rig-veda, generally celebrating the liberality shown by some patron to his priest, preceded by a brief statement merely consisting, it would seem, of a prose paraphrase of the respective verses recited thereafter. This latter set of recitations and legends thus consists entirely of matter taken from, or based on, the Rig-veda, which is indeed the proper source for the Hotri priest to resort to for his utterings. The recitations required for the Asvamedha, on the other hand, consist of matter drawn not even from the three older Vedas alone, but also from the Atharvans and Aṅgiras whose names combined usually make up the old designation of the hymns and spells of the Atharva-veda, whilst they are here taken separately as if still representing two different collections of texts;--nay, the materials, as we have seen, are even drawn from other, probably still later, sources 1. This circumstance, added to the fact that the texts of the Black Yagus make no mention of this item of the ceremonial 2, might well make one suspect its comparatively late introduction into the Asvamedha ritual; though even this would not, of course, make it any the less strange that no allusion should be made, in the epic account, to this by no means the least interesting feature of the performance. One must, however, bear in mind that the poet's mind was evidently more intent on telling about the wonderful deeds

p. xxxiii

of the semi-divine bowman in foreign lands than on recording the regularly recurring rites performed, in the meantime, at home in the presence of the royal sacrificer himself. Even in cases where the horse was kept within a convenient distance from the sacrificial compound all the year round, its warders, themselves partly of royal blood, could hardly have had an opportunity of attending the performance of these rites; though the popular character of some of these rites, as well as certain expressions used in connection with the 'revolving legend,' would lead one to suppose that they were meant to be witnessed by at least representatives of the various classes of the population.

The ritual arrangements of the Purushamedha, or human sacrifice, of which the Brâhmana treats next, seem to have been developed out of those of the Asvamedha. Its first three Soma-days are essentially the same as the three days of the horse-sacrifice, except as regards the difference of victims on the second day. To these the authorities of the White Yagur-veda--and apparently also those of the Black Yagus 1--add two more days, whilst the Sâṅkhâyana-sûtra 2, on the other hand, recognises but one additional day. Like the Vaitâna-sûtra, Sâṅkhâyana also differs from the other authorities in giving an entirely different character to the central feature of this performance, inasmuch as he makes it a teal human sacrifice instead of a merely symbolic one. A peculiar interest thus attaches to this difference of theory, seeing that it involves the question as to how far down the practice of human sacrifices can be traced in India 3. That such sacrifices were practised

p. xxxiv

in early times is clearly shown by unmistakable traces of them in the ritualistic works; but in this respect India only shares a once almost universal custom. The question, then, which chiefly interests us here is whether or not this practice was still kept up at the time with which we are here concerned. Now, as regards the texts of the Yagur-veda--that is, the text-books of the sacrificial priest κατ᾽ ἐξοχήν--it seems pretty clear that they no longer recognise the sacrifice of human beings; and the same may be said of the remaining ritualistic literature with the exception of the two works above referred to with regard to this particular sacrifice. The points bearing on this question being very few in number, may be briefly reviewed here.

First as regards the story of Sunahsepa which is recited at the Râgasûya sacrifice 1, and has been several times treated before 2. King Hariskandra, being childless, prays to Varuna to grant him a son, vowing to sacrifice him to the god. A son is born to him, and is called Rohita; but, in spite of the god's repeated demands, the fulfilment of the vow is constantly deferred; till at last the youth, having been invested in armour, is told of the fate awaiting him. He, however, refuses to be sacrificed, and escapes to the forest. The king thereupon is seized with dropsy; and the son, hearing of this, hastens homeward to save his father. On the way he is met by Indra who urges him to wander, and he accordingly does so for a year. The same is repeated five different times. In the sixth year, the prince, while wandering in the forest, comes across a starving Brâhman, Agîgarta, who lives there with his wife and three sons, and who consents to sell him one of his sons for a hundred cows to serve him as a ransom to Varuna. The Brâhman wishing to keep his eldest son, whilst the mother refuses to part with the youngest, the choice falls upon

p. xxxv

the second boy, called Sunahsepa. Rohita now returns to his father who, having been told of the transaction, then proposes to Varuna to offer the Brâhman youth in lieu of his son; and the god, deeming a Brâhman better than a Kshatriya, consents to the exchange, and orders the king to perform the Râgasûya sacrifice, and to make the youth the chief victim on the Abhishekanîya, or day of consecration. Four renowned Rishis officiate as offering-priests but when the human sacrifice is to be consummated, no one will undertake to bind the victim. The boy's own father, Agîgarta, then volunteers to do so for another hundred cows; and subsequently he even undertakes to slay his son for a similar reward. But when the poor lad sees his own father coming towards him, whetting his knife, and becomes aware that he is really to be slain, 'as if he were not a man,' he bethinks himself of calling upon the gods for help; and by them he is successively referred from one to another, till by uttering three verses in praise of Ushas, the Dawn, he is released from his fetters, whilst the king is freed from his malady. Subsequently one of the four priests, the royal sage Visvâmitra, receives Sunahsepa as his son, conferring upon him the name of Devarâta (Theodotos), and refuses to give him up to Agîgarta; and when the latter calls on his son to return to him, and not to desert his ancestral race, he replies, 'What has never been found even amongst Sûdras, thou hast been seen with a knife in thy hand, and hast taken three hundred cows for me, O Aṅgiras!' And on his father avowing his guilt, and promising to make over the cattle to him, he again replies, 'He who has once done wrong will commit another sin; thou hast not abandoned the ways of a Sûdra: what thou hast done is irremediable'; and 'is irremediable,' echoes Visvâmitra, who then formally adopts him as his son.

This legend 1, so far from bearing witness to the existence

p. xxxvi

of human sacrifices as a generally recognised practice, at the time when it originated, would rather seem to mark this particular case as an exceptional one. For, if it were not so, how comes it that the king's four high-priests--who, if any, must have been looked upon as thorough masters of the sacrificial science--should have refused to assist in the immolation of the human victim ordered by the deity, leaving it to be accomplished by the sullied hands of the wretched father? But there is another feature of the story which cannot but strike one as very peculiar. Why should the childless king pray for the birth of a son only to make a sacrifice of him? He has been told to do so by the holy sage Nârada: is one then to understand that the sage's advice, as well as Varuna's consent, is given merely to try the king's faith and truthfulness? If so, the case is similar to that of Abraham's sacrifice in the land of Moriah, only that the king's faith proves less intense and exalted--perhaps more humanly faint-hearted--than that of the Jewish patriarch. But the most striking feature of the legend doubtless is the part played in it by the unnatural father; and this feature seems indeed to impart to the tale something of the character of an allegorical representation of the contrast between a barbarous (and perhaps earlier) and a more civilised phase of life and moral feeling 1. In this respect two points deserve to be noticed, viz. the coarseness of the synonymous names ('dog's tail') of the three sons of the Brâhman 2; and the fact that the latter belongs to the Aṅgiras stock, a name intimately associated with superstitious

p. xxxvii

rites 1 on the one hand, and with the ritual of the fire-altar 2 on the other.

Now, it is exactly in connection with the building of the fire-altar that the clearest, and most unmistakable trace of an old practice of human sacrifices--or rather of the slaying of men for sacrificial purposes--occurs. In laying down the bottom layer of the altar, the pan which had been used by the Sacrificer for carrying about the sacred fire for a year is built into this layer, with heads of the five recognised sacrificial animals 3--man, horse, ox, sheep, and goat--put therein, in order to impart stability to the altar (Sat. Br. VII, 5, 2, I seqq.). In a previous passage of the Brâhman4 (I, 2, 3, 6 seq.),where the relative value of non-animal offering-materials and the five sacrificial animals is discussed, it was stated that, whilst the gods were making use of one after another of these animals, the sacrificial essence gradually passed from one to the other, thus rendering the previous one useless for sacrifice, until it finally passed into the earth whence it entered the rice and barley afterwards used for sacrificial dishes. The general purport of this passage would seem to be to indicate a gradual tendency towards substituting the lower for the higher animals, and ultimately vegetable for animal offerings; though, as a matter of fact, animals continued of

p. xxxviii

course to be commonly sacrificed in later times. Now as regards the heads of the five victims, the author subsequently (VI, 2, 1, 37 seqq.) makes some further remarks which go far to show that his previous statements referred only to the traditional practice which, however, was no longer in use in his own day, and had probably not been so for generations past. He mentions various expedients adopted by some priests with a view to keeping up at least some semblance of the old custom,--viz. either by procuring real heads from some source or other, or by using heads made of gold or clay; but they are summarily dismissed as profane and fraudulent counterfeits; and the author then remarks somewhat vaguely and diplomatically that 'one may slay those five victims as far as one may be able (or inclined) to do so, for Pragâpati was the first to slaughter them, and Syâparna Sâyakâyana the last, and in the interval also people used to slaughter them; but at the present day people slaughter only (one of 1) those two, the (he-goat) for Pragâpati, and the one for Vâyu;' after which he proceeds to explain in detail the practice then in ordinary use. Later on (VII, 5, 2, 1 seqq.), the Brâhmana expounds in the usual way the formulas used in the traditional, and theoretically still available procedure, though in the actual performance perhaps only the formulas relating to the particular heads 2 used would be muttered.

While Yâgñavalkya thus, at least in theory, deals rather cautiously with this feature of the traditional custom, the theologians of the Black Yagus 3 take up a somewhat bolder position. Indeed it is evidently against this older school of ritualists that some of the censure of our Brâhmana is directed. For though they too allow, as an alternative practice, the use of a complete set of five heads, they make

p. xxxix

no mention of a man being killed for this purpose, but enjoin that a dead man's head is to be bought for twenty-one beans 1, which is then to be laid against an ant-hill with seven holes in order to again supply it with the seven 'vital airs of the head'; whereupon three stanzas relating to Yama are to be sung round about it to redeem it from the god of death. Besides the four animals, there is also to be a he-goat sacred to Pragâpati, the offering of which is to complete the animal sacrifice 2, In this school also 3, the ordinary practice, however, is to kill only a he-goat for Vâyu Niyutvat, and to use its head for putting it in the pan placed in the bottom layer of the altar. As regards the Rig-veda ritual, the Kaushîtaki-brâhmana, as Prof. Weber has pointed out, leaves a choice between a he-goat for Pragâpati and one for Vâyu; whilst the Sâṅkhâyana-sûtra, curiously enough, again adds the alternative course of using the set of five heads.

The same scholar has drawn attention to another rite in the sacrificial ceremonial which seems to him to show clear traces of human sacrifice. At the purificatory bath at the end of the Asvamedha, the Sacrificer is to be purged of any guilt he may have committed against Varuna by an oblation made to Gumbaka (Varuna) on the bald head of a man possessed of certain repulsive features, whilst standing in the water. To these particulars,--as given in the present work (XIII, 3, 6, 5), the Taitt. Brâhmana (III, 9, 15), and Kâtyâyana's Sûtra (XX, 8, 16),--Sâṅkhâyana (XVI, 18)

p. xl

again adds further particulars, viz. that the man is to be a Brâhmana of the Âtreya family, bought (or hired) for a thousand cows, and that he is to enter the river till the water flows into his mouth. Now Prof. Weber is of opinion that this ceremony would be meaningless if the man were not actually drowned. I fail, however, to see the necessity of this assumption, seeing that even a purely symbolical interpretation of the ceremony will give it all the significance of the real act. That the Yagus texts contain nothing that could make one suspect that the man was actually drowned is beyond doubt; but even Sâṅkhâyana's statement that the water is to flow into his mouth is probably only meant to suggest the nearness and semblance of death by drowning. Otherwise the oblation could hardly have been performed in anything like a decent form. Besides, Sâṅkhâyana further states that, after the completion of the oblation, 'they drive him (the man) out, thinking that the guilt of the village-outcasts is (thereby) driven out 1.' Here the verb 'nih-sidh' could hardly have been used if the man was to be driven farther into the water. What is meant is probably that the man was to be driven out from the water, and possibly also from the village, to live an anchorite's life in the forest.

If now we turn our attention to the Purushamedha, or 'human sacrifice' proper, we find that the Yagus texts, as far as they deal with this ceremony at all 2, treat it as

p. xli

a purely symbolical performance. A large number of men and women, apparently intended to represent all classes of the community, are bound to eleven sacrificial posts, and after the necessary rites, concluding with the 'paryagnikarana'--or the carrying of fire round the oblations--have been performed on them, they are one and all set free; the sacrifice then proceeding with the offering of the set of eleven animal victims. That the ceremony in this form, with its pedantically elaborate array of symbolic human victims, cannot possibly lay claim to any very great antiquity is self-evident; the only question is whether it has not come to take the place of some other form of human sacrifice. Now, after the foregoing statement of facts, it would be idle to deny that the existence, at one time, of a simple form of human sacrifice is not only quite possible, but is indeed highly probable; and it would be no more than might be expected, if such a practice should eventually have revolted the moral sense of the more refined classes of the community 1, just as it happened, little more than a hundred years ago, in the case of the scarcely less odious practice of the burning of witches in Christian lands.

The practice of human sacrifices seems, however, to receive evidence of a yet more direct and unmistakable kind than the facts hitherto mentioned, from the ceremonial of the Purushamedha, as set forth in the Sâṅkhâyana and Vaitâna Sûtras. If this evidence has been reserved here to the last, it is because there seems reason to believe that, in the form in which it is presented in those works, the sacrifice was never actually performed, and probably never meant to be performed, but that we have here to do with a mere theoretical scheme intended to complete the sacrificial system. The importance of the subject makes it, however, desirable that we should take a somewhat closer view of the procedure of the 'human sacrifice,' as laid down in those two Sûtras.

p. xlii

Saṅkh. XVI, 10, 1. Pragâpati, having offered the Asvamedha, beheld the Purushamedha: what he had not gained by the Asvamedha, all that he gained by the Purushamedha 1; and so does the sacrificer now, in performing the Purushamedha, gain thereby all that he had not gained by the Asvamedha. 2, 3. The whole of the Asvamedha ceremonial (is here performed); and an addition thereto. 4-8, First oblations to Agni Kama (desire), A. Dâtri (the giver), and A. Pathikrit (the path-maker) . 9. Having bought a Brâhmana or a Kshatriya for a thousand (cows) and a hundred horses, he sets him free for a year to do as he pleases in everything except breaches of chastity. 10. And they guard him accordingly. 11. For a year there are (daily) oblations to Anumati (approval), Pathyâ Svasti (success on the way), and Aditi. 12. Those (three daily oblations) to Savitri 2 in the reverse order. 13. By way of revolving legends (the Hotri recites) Nârasamsâni . . .--XVI, II, 1-33 enumerate the Nârasamsâni 3, together with the respective Vedic passages.--XVI, 12, 1-7. There are twenty-five stakes, each twenty-five cubits long . . .; and twenty-five Agnîshomîya victims. 8. Of the (three) Asvamedha days the first and last (are here performed). 9-11. The second (day) is a pañkavimsa-stoma one . . . 12. The Man, a Gomriga, and a hornless (polled) he-goat--these are the Prâgâpatya 4 (victims). 13. A Bos Gaurus, a Gayal, an elk (sarabha), a camel, and a Mâyu Kimpurusha (? shrieking monkey) are the anustaranâh. 14-16. And the (other) victims in groups of twenty-five for the twenty-five seasonal deities . . . 17. Having made the adorned Man smell (kiss) the chanting-ground, (he addresses him) with the eleven verses (Rig-v. X, 15, 1-11) without 'om,'--'Up shall rise (the Fathers worthy of Soma), the lower, the

p. xliii

higher, and the middle ones.' 18. The Âprî verses are 'Agnir mrityuh' . . . 20. They then spread a red cloth, woven of kusa grass, for the Man to lie upon. 21. The Udgâtri approaches the suffocated Man with (the chant of) a Sâman to Yama (the god of death).--XVI, 13, 1. The Hotri with (the recitation of) the Purusha Nârâyana (litany). 2. Then the officiating priests--Hotri, Brahman, Udgâtri, Adhvaryu--approach him each with two verses of the hymn (on Yama and the Fathers) Rig-v. X, 14, 'Revere thou with offering King Yama Vaivasvata, the gatherer of men, who hath walked over the wide distances tracing out the path for many.' 3-6. They then heal the Sacrificer (by reciting hymns X, 137; 161; 163; 186; 59; VII, 35). 7-18. Ceremonies analogous to those of the Asvamedha (cf. XIII, 5, 2, 1 seqq.), concluding with the Brahmavadya (brahmodya).--XVI, 14, 1-20. Details about chants, &c.; the fourth (and last) day of the Purushamedha to be performed like the fifth of the Prishthya-shadaha.

Vait. S. XXXVII, 10. The Purushamedha (is performed) like the Asvamedha . . . 12. There are offerings to Agni Kama, Dâtri, and Pathikrit. 13. He causes to be publicly proclaimed, 'Let all that is subject to the Sacrificer assemble together!' 14. The Sacrificer says, 'To whom shall I give a thousand (cows) and a hundred horses to be the property of his relatives? Through whom shall I gain my object?' 15. If a Brâhmana or a Kshatriya comes forward, they say, 'The transaction is completed.' 16. If no one comes forward, let him conquer his nearest enemy, and perform the sacrifice with him. 17. To that (chosen man) he shall give that (price) for his relatives. 18. Let him make it he publicly known that, if any one's wife were to speak 1, he will seize that man's whole property, and kill herself, if she be not a Brâhmana woman. 19. When, after being bathed and adorned, he (the man) is set free, he (the priest) recites the hymns A.V. XIX, 6; X, 2.-20. For a year (daily) offerings to Pathyâ Svasti, Aditi, and

p. xliv

[paragraph continues] Anumati. 21. At the end of the year an animal offering to Indra-Pûshan. 22. The third day is a Mahâvrata. 23. When (the man 1) is bound to the post, he repeats the three verses, 'Up shall rise' . . .; and when he is unloosened, the utthâpanî-verses. 24-26. When he is taken to the slaughtering-place (the priest repeats) the harinî-verses; when he is made to lie down, the two verses, 'Be thou soft for him, O Earth'; and when he has been suffocated, (he repeats) the Sahasrabâhu (or Purusha Nârâyana) litany, and hymns to Yama and Sarasvatî--XXXVIII, 1-9 treat of the subsequent ceremonies, including the recitation, by the Brahman, of hymns with the view of healing the Sacrificer.

Now, even a slight consideration of the ritual of the Purushamedha, as sketched out in these two works, must, I think, convince us that this form of human sacrifice cannot possibly be recognised--any more than the one propounded in the Satapatha and Taittirîya Brâhmanas--as having formed part of the traditional sacrificial ceremonial; and that, in fact, it is nothing more than what Sâṅkhâyana appears to claim for it, viz. an adaptation, and that a comparatively modern adaptation, of the existing Asvamedha ritual. Indeed, it seems to me by no means unlikely that the two different schemes of the Purushamedha originated at about the same time, and that they were intended to fill up a gap in the sacrificial system which seemed to require for Man, as the chief sacrificial animal, a more definite and, so to speak, a more dignified place in the ceremonial than was up to that time accorded to him. The circumstance that the account of this sacrifice, as given in the Sâṅkhâyana-sûtra, presents some of the ordinary features of Brâhmana diction, and that it is indeed actually assigned by the commentary to the Mahâ-Kaushîtaka, should not be allowed to weigh with us, since this is most likely done for the very purpose of securing for this scheme some sort of authoritative sanction of respectable

p. xlv

antiquity 1. For seeing that the older ritualistic works know nothing of it, it seems sufficiently evident that this human sacrifice could not possibly have been rite performed in Sâṅkhâyana's time, since no proper priest--no genuine Adhvaryu and Udgâtri, at all events--could have been found to perform it. And, indeed, it can scarcely be without significance that the Atharva-sûtra is the only other work which recognises the ceremony; and that nearly all the hymns and verses used in connection with the immolation of the human victim are taken from the Atharvan and the tenth mandala of the Rik. Nay, the very fact that, in both Sûtra works, this sacrifice is represented as being undertaken, not for the great object of winning immortal life, but for the healing of the Sacrificer's bodily infirmities, might seem sufficient to stamp the ceremony as one partaking more of the nature of the superstitious rites of the Atharvan priests than of that of the great sacrifices of the traditional Srauta ritual.

If thus we find it impossible to recognise the Purushamedha as a genuine member of the sacrificial system, this is still more the case as regards the Sarvamedha, or all-sacrifice, a ten days’ performance which includes amongst its component parts, not only the Purushamedha, but also the Asvamedha, the Vâgapeya, and the Visvagit with all the Stomas and Prishthas,--it thus being the very ceremonial performance that might seem calculated to fitly crown the edifice of the sacrificial theory. As regards the ritualistic treatment of this sacrifice, the number of authorities dealing with it shows a further diminution from that of the Purushamedha. For whilst the Satapatha-brâhmana agrees with the Sâṅkhâyana and Vaitâna Sûtras on the general features of its ritual--with the exception, of course, of the radical difference as to the character of the human sacrifice--the Taittirîya-brâhmana, which gave at least the list of the symbolic victims of the Purushamedha, is altogether silent on the Sarvamedha; this ceremony being,

p. xlvi

however, dealt with in some of the Sûtra works connected with the Black Yagus.

The concluding chapter of the thirteenth kânda contains a valuable and interesting account of the preparation of the burial-place or sepulchral mound, and the interment of the charred bones previously preserved, in an urn or jar, for some indefinite period since the burning of the dead body. Of especial interest, in this account, is the statement that the bones, when committed to the grave, are to be arranged in accordance with their natural position, the spaces between them being then filled up with bricks in such a way as to present, as in the case of the fire-altar, a fancied resemblance to the shape of a bird. It is difficult to see what explanation could be offered for this feature of the obsequies, except a vague belief in some form of future resurrection.

The fourteenth kânda, up to the beginning of the Brihad-âranyaka, is entirely taken up with the exposition of the Pravargya, an important, though optional and subsidiary, ceremony performed on the Upasad-days of Soma-sacrifices. Whilst the central feature of this sacrificial performance consists of a ceremony of an apparently simple and unpretending character, viz. the preparation of a hot draught of milk and ghee, the Gharma, which the Sacrificer has to take, after oblations have been made thereof to various deities, the whole rite is treated with a considerable amount of mystic solemnity calculated to impart to it an air of unusual significance. A special importance is, however, attached to the rough clay pot, used for boiling the draught, and manufactured and baked in the course of the performance itself; it is called Mahâvîra, i.e. the great man or hero, and Samrâg, or sovereign lord, and is made the object of fervid adoration as though it were a veritable deity of well-nigh paramount power.

Although the history of this ceremony is somewhat obscure, the place assigned to it in the Soma-ritual would lead one to suppose that its introduction must have taken place at a time when the main procedure of the Soma-sacrifice had already been definitely settled. This conclusion

p. xlvii

is also borne out by the position taken up towards this ceremony by the authorities of the Black Yagur-veda. For whilst the Maitrâyanîyâ Samhitâ gives at least the formulas used for it, the Kâthaka, on the other hand, takes no notice whatever of it, and the Taittirîya school only deals with it in its Âranyaka. Nevertheless, this ceremony can boast of a respectable antiquity, seeing that it is treated of at some length in the Brâhmanas of the Rik--viz. Ait. Br. 1, 18-22; Kaush. Br. VIII, 3-7; and this circumstance alone might almost seem to justify the inference that it was in that very school of ritualists that this item of the sacrificial ceremonial was first elaborated. It is very doubtful, however, whether such an inference would find any support in the dogmatic explanation of the ceremony offered by some of the theologians of the Rig-veda. At the end of the Pravargya section, in a passage which has a somewhat disconnected appearance, and seems hardly in consonance with previous dogmatic explanations, the Aitareya-brâhmana makes the secret import of the ceremony to be that of a mystic union of the gods resulting in the generation of a new, divine body for the Sacrificer. This explanation, having been previously adopted by Haug and Garbe, was recorded without question in a note to part ii (p. 104) of this translation. Further consideration of this matter has, however, convinced me that the theory. referred to fails altogether to account for the origin of the ceremony, as well as for important points in its performance which find a ready explanation in the theory applied to it by the present work, as well as by the Taittirîya-Âranyaka and the Kaushîtaki-brâhmana. For seeing that the main object of sacrificial performances generally is the reconstruction of Pragâpati, the personified universe, and (the divine body of) the Sacrificer, it is difficult to see why, for this latter purpose, a new and special ceremony should have been thought necessary; and, besides, the rejected theory, if it is at all to account for the high honour rendered to the Mahâvîra pot, would almost involve the recognition of a form of Liṅga-worship which surely would require very much stronger evidence than the isolated and

p. xlviii

[paragraph continues] (to my mind) somewhat suspicious passage on which this theory is based.

Now, as regards the rival theory underlying the exposition of the Pravargya, as given in the Satapatha-brâhmana, it makes the Mahâvîra pot a symbol of the sun, whilst the hot milk draught represents the divine flood of life and light with which the performer of the ceremony becomes imbued. These symbolic interpretations, whatever we may think of them otherwise, certainly adapt themselves admirably to the general sacrificial imagery. As the sun is the head of the universe--or, in figurative language, the head of Pragâpati, the world-man--so its earthly, and earthen, counterpart, the Mahâvîra pot, is the head of Vishnu, the sacrificial man, and the Sacrificer; and this ceremony is thus performed in order to complete the universe and sacrifice, as well as the divine body of the Sacrificer, by supplying them with their head, their crowning-piece, so to speak; and to imbue them with the divine essence of life and light. For this purpose the theory rather ingeniously avails itself of certain myths vaguely alluded to in the Rig-veda, according to which (X, 171, 2) Indra cut off the head of Makha (here identified with Vishnu, the sacrifice and the sun-god); and (I, 116, 12; 117, 22; 119, 9) Dadhyañk, the son of Atharvan, was fitted by the Asvins with a horse's head, and this hippocephalous creature then communicated to them the Madhu, or sweet thing,--that is, as would appear, the sweet doctrine of the Soma, the drink of immortality. This symbolism readily explains some points connected with the Pravargya ceremony, for which no obvious reason seems otherwise to suggest itself. For one thing, it accounts for the deep reverence shown to the Gharma vessel, which, in fact, is no other than the giver of light and life himself; whilst the optional character of the ceremony explains itself from the fact that the Soma-cup, of which the Sacrificer will subsequently partake, might of itself be expected to supply him with the blessings which he hopes to derive from the Pravargya. And, finally, it also becomes clear why the Pravargya must not form part of a man's first performance of a Soma-sacrifice.

p. xlix

[paragraph continues] For the Pravargya, as we have seen, is performed on the preliminary days of the Soma-sacrifice, before the pressing of the Soma has taken place; and it obviously is only after he has actually partaken of the Soma-drink, and has thereby potentially 'put on immortality,' that he can partake of the Gharma, and thus become imbued with the celestial light 1. The dogmatical explanation of this ceremony thus puts, as it were, the finishing touch to that strange allegory by which the Indian theologians sought to make the sacrificial ceremonial a practical illustration of that unity of the All which speculation had been striving to compass since the days when the emptiness of the Vedic pantheon had dawned upon the thinking mind, and when critically inclined bards ventured to sing of the national god 2: 'Not for a single day hast thou fought, nor hast thou any enemy, O Maghavan: illusion is what they say concerning thy battles; no foe hast thou fought either to-day or aforetime.'

As regards the optional and somewhat recondite character of the Pravargya ceremony, it is probably not without significance that the section dealing therewith is combined with the speculative Brihadâranyaka so as to make up with it the last book of the Brâhmana; the Âranyaka-kânda, or forest section. Such, at least, is the case in the Mâdhyandina text, where the Pravargya section occupies the first three adhyâyas of the last (fourteenth) book; whilst the Kânva text presents a slight difficulty in this respect. What passes generally as the seventeenth (and last) kânda of that version, consists of the Brihadâranyaka; whilst the sixteenth kânda begins with the section on funeral rites, corresponding to the last

p. l

[paragraph continues] (eighth) adhyâya of kânda XIII of the Mâdhyandina recension, and is stated 1 to include also the Pravargya section (Madhy. XIV, 1-3). Now it is a strange fact that the six adhyâyas of the Brihadâranyaka (XIV, 4-9 in the Mâdhyandina text) are counted 3-8 in the Kânva text,--a circumstance which manifestly can only be explained by the Pravargya section being taken to form the first two adhyâyas of the last book of that version. This, indeed, is probably implied in the remark added to the description of a MS. of the Kânva text in the catalogue of the MSS. of the Sanskrit College, Benares (p. 44), according to which 'Pravargyakândasya patrâni' are 'bhinnapramânâksharâni,'--that is, 'the leaves of the Pravargya section have a special pagination' (? i.e. they are numbered independently of the section on funeral rites preceding them).

And now my task is done, and I must take leave of this elaborate exposition of the sacrificial ordinances of Indian theology. For well-nigh a score of years the work has 'dragged its slow length along,' and during that time it has caused me--and, I doubt not, has caused some of my readers, too--not a few weary hours. In the early stages of the work, my old teacher, Professor Albrecht Weber, than whom no one is more deeply versed in the intricacies of the sacrificial ritual, wrote to me: 'You have undertaken a difficult, a most difficult task; and I can only hope that your courage and patience will not fail you before you are through with it.' And, indeed, I must confess that many a time I felt as if I should never be able to get through my task; and but for Professor Max Müller's timely exhortations and kindly encouragement, the work might perhaps never have been completed. 'I know,' he once wrote to me, 'you will thank me one day for having pressed you to go on with your work;' and now I do indeed thank him most sincerely and with all my heart for the kindness and patience he has shown me these many years. But, strange to say, now that the work is completed, I feel as if I could not do without working at it; and certainly, if

p. li

a second edition could ever have been required of a work of this kind, it would have found me ready once more to work my way through the bewildering maze of rites; and I know only too well that I should have to correct many a mistake, and could improve many an awkwardly expressed passage. In conclusion, a word of cordial thanks is due to the staff of the University Press, whose patience must often have been severely tried in the course of the printing of this work, and who, by the excellence of their presswork, and by their careful supervision, have materially lightened my task, and saved me much tedious and irksome labour.


EDINBURGH, December 30, 1899.


xv:1 Cf. Taitt. Br. III, 8, 9, 4,--parâ vâ esha sikyate yobalosvamedhena yagate;--'Verily, poured away (dislodged) is he who, being weak, performs the Asvamedha;' Âp. Sr. XX, 1, 1, 'a king ruling the whole land (sârvabhauma) may perform the Asvamedha;--also one not ruling the whole land.'

xv:2 Cf. Weber, History of Indian Literature, p. 107; Max Müller, History of Ancient Sanskrit Literature, p. 358.

xvi:1 Besides the two chapters referred to, nothing more than quotations are known of this work. Possibly, however, the difference between it and the Kaushîtaki-brâhmana consisted merely of such supplements which would thus be very much of the same character as the last two pañkikâs of the Aitareya-brâhmana, except that they never became so generally recognised.

xvi:2 Though this circumstance seems to favour the supposition of the more recent ritualistic treatment of the Asvamedha, it may not be out of place to notice that, in the Maitrâyanî Samhitâ, the Asvamedha section is followed by several Brâhmana sections; amongst them that of the Râgasûya which is not found in the Kâthaka at all. Sat. Br. XIII, 3, 3, 6, calls the Asvamedha an 'utsannayagña'; but it is not quite clear what is meant thereby, seeing that the same term is applied to the Kâturmâsyâni, or Seasonal offerings (II, 5, 2, 48).

xvii:1 See Professor M. Bloomfield's paper on 'The Position of the Gopatha-brâhmana in Vedic Literature,' Journ. Am. Or. Soc., vol. xix.

xvii:2 Cf. Mahâbh. XIV, 48, where these four sacrifices are specially recommended by Vyâsa to Yudhishthira as worthy of being performed by him as King.

xvii:3 Possibly also, the hymn Rig-veda I, 164 (Ath.-v. IX, 9, 10)--on which see P. Deussen, Allg. Geschichte der Philosophie, I, 1, p. 105 seq.--may have been placed after the two Asvamedha hymns to supply topics for the priests’ colloquy (brahmodya) at the Asvamedha. Cf. XIII, 2, 6, 9 seqq.; 5, 2, 11 seqq. The fact that the Asvamedha is not treated of in the Aitareya-brâhmana cannot, of course, be taken to prove the later origin of the hymns referred to, though it might, no doubt, fairly he used as an argument in favour of assuming that those parts of the Asvamedha ceremonial in which the Hotri takes a prominent part were probably not introduced till a later time.

xvii:4 Haug, Ait. Br. I, introd., p. 12 seqq., argues against the assumption of a comparatively late origin of the hymn I, 162; but his argument meets with serious lexical and other difficulties.

xvii:5 We may leave out of account here one or two vague allusions, such as X, 155, 5 'these have led around the cow (or hull) and have carried around the fire; with the gods they have gained for themselves glory: who dares to attack them?' The question also as to whether the so-called Âprî-hymns, used at the fore-offerings of the animal sacrifice, were from the very beginning composed for this purpose, cannot be discussed here.

xviii:1 'They (the Massagetae) worship the sun only of all the gods, and sacrifice horses to him; and the reason for this custom is that they think it right to offer the swiftest of all animals to the swiftest of all the gods.' Herod. I, 216.

xix:1 Dr. Hillebrandt, 'Varuna and Mitra,' p. 65, is inclined to refer this connection to Varuna's character as the god of waters and the rains, as favouring the crops and fertility generally.

xix:2 Whilst it may be a matter of opinion whether, with Professor Brugmann (Grundr. II, p. 154), we have to take the original form of this name to be 'vorvanos,' or whether the 'u' of the Sanskrit word is merely due to the dulling influence of the preceding r (cf. taruna, dhâruna, karuna), the etymological identity of 'varunas' and οὐρανός is now probably questioned by few scholars. The ethical attributes of this mythological conception seem to p. xx me to find a sufficiently intelligible explanation without resorting to outside influence to account for them. Indeed, Dr. Hillebrandt's 'Varuna and Mitra' gives a fairly complete and satisfactory account of this figure of the Indian pantheon in all its relations.

xx:1 See Sat. Br. XIII, 5, 1, 17, 18.

xx:2 That is, probably, three halting-places (? the points of rising, culminating, and setting). Perhaps also the three statements are merely meant as an emphatic repetition of one and the same locality--the sky, the sea of waters; though, possibly, three different strata of the upper region may be intended. Professor Ludwig, on the other hand, takes 'trîni bandhanâni' in the sense of 'three fetters,' and Professor Hillebrandt, l.c., in that of 'three relations (or connections, Beziehungen).'

xxi:1 Whilst the climate of Baluchistan is regulated, as in Europe, by the succession of four seasons, the climate of the districts east of the Indus, as of India generally, shows the characteristic threefold division of rainy, cool, and hot seasons (S. Pottinger, Beloochistan, p. 319 seqq.).

xxii:1 Taitt. S. VI, 6, 7, 4, explains this offering as symbolically smoothing down the sacrifice torn up by recited verses and chanted hymns, even as a field, torn up by the plough, is levelled by a roller ('matya,' taken however by Sây. in the sense of 'cow-dung'). The Sat. Br. does not allude to the expiatory character of the offering, but there can be no doubt that it is of an essentially piacular significance. It need scarcely be mentioned that the 'avabhritha,' or lustral bath, at the end of Soma- and other sacrifices, is distinctly explained (II, 5, 2, 46; IV, 4, 5, 10) as intended to clear the Sacrificer of all guilt for which he is liable to Varuna. Cf. Taitt. Br. III, 9, 15, 'At the lustral bath he offers the last oblation with "To Gumbaka hail! " for Gumbaka is Varuna: he thus finally frees himself from Varuna by offering.'

xxii:2 See, for instance, Sat. Br. IV, 1, 4, 2; V, 3, 2, 4; IX, 4, 2, 16; Maitr. S. IV, 5, 8; Taitt. Br. III, 1, 2, 7 (kshatrasya râgâ Varunodhirâgah).

xxiii:1 Since all the gods are concerned in the Asvamedha-whence the horse is called 'vaisvadeva'--Indra would of course have a general interest in it. Indra is also associated with the horse in so far as he is said to have first mounted it, Rig-veda I, 163, 2, 9. Indra's two bays (harî) of course belong to a different conception.

xxiii:2 Âpo vai Varunah, Maitr. S. IV, 8, 5.

xxiv:1 This, no doubt, might possibly be taken to mean 'Pragâpati led away the horse for Varuna,' but Dr. Hillebrandt could hardly have meant it in this sense, since his argument apparently is that the horse (like Varuna himself) represents the aqueous element, and that thus, by taking to himself the horse, Pragâpati incurs dropsy. The exact point which interests us here, viz. the relation between Pragâpati and Varuna as regards the sacrificial horse, lies outside Dr. Hillebrandt's inquiry.

xxiv:2 In Dr. Hillebrandt's interpretation, it is also not quite easy to see in what way Pragâpati, by carrying off Varuna's horse, impaired--'griff an,' attacked, assailed--his own godhead. One might possibly refer 'svâm' to the horse, but this would make the construction rather harsh. The verb 'nî' here would seem to refer to the leading up of the sacrificial horse to the offering-ground, either for being set free for a year's roaming, or for sacrifice, for both of which acts the verb 'ud-â-nî'--i.e. to lead up the horse from the water where it was washed--is used (Sat. Br. XIII, 4, 2, 1; 5, 1, 16).

xxv:1 Rig-veda S. I, 162, 2. 'When, held by the mouth (by the bridle)., they lead round the offering of the (horse) covered with rich trappings, the all-coloured he-goat goes bleating in front right eastwards to the dear seat of Indra and Pûshan. 3. This he-goat, fit for all the gods, is led in front of the swift horse as Pûshan's share; like (?) the welcome cake, Tvashtri promotes it, along with the steed, to great glory. 4. When thrice the men duly lead around the horse meet for offering along the way to the gods, then the he-goat walks first, announcing the sacrifice to the gods. . . . 16. The cloth which they spread (for the horse to lie upon) and the upper cloth and the gold, the halter, the steed, the shackle--these they bring up as acceptable to the gods.'--I, 163, 12. 'Forth came the swift steed to the slaughter, musing with reverent mind; his mate, the he-goat, is led in front; and behind go the wise singers.'

xxv:2 According to the Taittirîyas, this second he-goat is tied to the cord surrounding the horse's limbs somewhere above the neck of the horse.

xxvi:1 See , note 5.

xxvi:2 It has even been supposed to be merely a condensed version of a comparatively modern work ascribed to Gaimini, the (Asvamedha-parvan of the) Gaimini-Bhârata.

xxvii:1 Vyâsa remarks to Yudhishthira (XIV, 2071), For the Asvamedha, O king of kings, cleanses away all ill-deeds: by performing it thou wilt without doubt become free from sin.' Cf. Sat. Br. XIII, 3, 1, 1, 'Thereby the gods redeem all sin, yea, even the slaying of a Brahman they thereby redeem; and he who performs the Asvamedha redeems all sin, he redeems even the slaying of a Brahman.' As a rule, however, greater stress is laid in the Brâhmana on the efficacy of the ceremonial in ensuring supreme sway to the king, and security of life and property to his subjects.

xxvii:2 The 'rukma' is borne by the Agnikit, or builder of a fire-altar, which is required for the Asvamedha; cf. VI, 7, 1, 1.

xxvii:3 It is carefully selected by charioteers and priests, Mahâbh. XIV, 2087.

xxvii:4 Whilst, according to the Brâhmana (XIII, 4, 2, 5), the body of 'keepers' is to consist of 100 royal princes clad in armour, 100 noblemen armed with swords, 100 sons of heralds and headmen bearing quivers and arrows, and 100 sons of attendants and charioteers bearing staves; the epic gives no details, except that it states that 'a disciple of Yâgñavalkya, skilled in sacrificial rites, and well-versed in the Veda, went along with the son of Prithâ to perform the propitiatory rites,' and that 'many Brâhmanas conversant with the Veda, and many Kshatriyas followed him at the king's behest.'

xxviii:1 That is, real or symbolic, only the domesticated animals being offered, whilst the wild ones are set free after the ceremony of 'paryagnikarana.' Amongst these animals the poet curiously enough also mentions (XIV, 2542) 'vriddhastriyah,' which Pratâpa Chandra Ray translates by 'old women.' This is of course impossible; if it is not a wrong reading, it has doubtless to be taken in the sense of 'old female (kine),' probably the (25) barren cows offered at the end of the Asvamedha to Mitra-Varuna, the Visve Devâh, and Brihaspati (XIII, 5, 4, 25) being intended. In its enumeration of the victims, the Taitt. Samhitâ (V, 6, 21) indeed mentions 'vairâgî purushî,' taken by the commentator to mean 'two human females consecrated to Virâg.' If it be for this or a similar purpose that the 'vriddha-striyah' were intended, we may refer to Taitt. Br. III, 9, 8, where it is distinctly stated that 'the man' and the wild animals are to be released as soon as the 'paryagnikaranam' has been performed on them. But no 'man' being mentioned amongst the victims, Sâyana p. xxix takes the 'purusham' here to refer to the 'vairâgî purushî' mentioned in the Samhitâ. Perhaps, however, this passage has rather a wider sense, referring to human victims generally at any sacrifice.

xxix:1 Draupadî's Asva-upasamvesanam is referred to, but no further particulars are mentioned.

xxix:2 The king's object, in performing the sacrifice, was to obtain the birth of a son. Cf. Sat. Br. XIII, 1, 9, 9, 'for from of old a hero was born to him who had performed the (Asvamedha) sacrifice.'

xxx:1 Whilst cattle-lifting generally, such as formed the object of the invasion of the land of the Matsyas by the Trigartas (as related in the Virâta-parvan), was probably a practice pretty prevalent from ancient times, the stealing of the sacrificial horse would offer an additional temptation, from the political point of view, on account of the exceptional character of the animal as the symbol of its master's claim to paramountcy.

xxx:2 Sat. Br. XIII, 1, 6, 3; Taitt. Br. III, 8, 9, 4.

xxx:3 One might feel inclined to take this specification of that month as implying the existence, at the poet's time, of the practice of confining the horse in a pen or shed (made of Asvattha palings) during the last two months, mentioned Taitt. Br. III, 8, 12, 2.

xxxi:1 See XIII, I, 5, 1 seqq.; 4, 2, 8 seqq.; Taitt. Br. III, 9, 14. In connection with the revolving legend,' the conductors of bands of lute-players seem to have sung additional stanzas in which the royal Sacrificer was associated with pious kings of old; see XIII, 4, 3, 3.

xxxii:1 It is hardly likely that some of the texts mentioned (devaganavidyâ, sarpavidyâ, &c.) refer merely to portions of the Vedic texts.

xxxii:2 The singing of stanzas in honour of the king, by a Brâhmana and a Kshatriya, with the accompaniment of lutes, on the other hand, does form part of the Taittirîya ritual. Taitt. Br. III, 9, 24.

xxxiii:1 Whilst the three Samhitâs contain no section relating to the Purushamedha, the Taittirîya-brâhmana (III, 4) enumerates the (symbolic) human victims in much the same way as does the Vâgasaneyi-samhitâ (see the present vol. p. 413 seqq.); and the Âpastamba-sûtra makes the performance similar to what it is in the White Yagus texts. The Vaitâna-sûtra of the Atharva-veda also makes it a five days’ performance.

xxxiii:2 Like the chapter on the Asvamedha, that on the Purushamedha is stated to be taken from the Mahâ-Kaushîtaki-brâhmana.

xxxiii:3 On this question see especially A. Weber, Zeitsch. d. D. M. G. 18, p. 262 ff., repr. in Indische Streifen, II, p. 54 ff.

xxxiv:1 See part iii, p. 95.

xxxiv:2 Cf. Max Müller, History of Ancient Sanskrit Literature, p. 108 ff.; M. Hang, Aitareya-brâhmana, II, p. 460 ff.; R. Roth, Weber's Ind. Stud. I, 475 ff.; II, 112 ff.

xxxv:1 The earliest reference to the myth or story of Sunahsepa is in Rig-veda I, 24, 11-13; V, 2, 7, where he is apparently alluded to as having been actually p. xxxvi rescued from the stake, or from (three) stakes to which he was bound either for sacrifice, or, as Roth prefers, for torture.

xxxvi:1 In the Sabhâparvan of the Mahâbhârata (II, 6275 seqq.),, as was first pointed out by Lassen, Krishna accuses Garâsandha, king of Magadha and Kedi, residing at Mathurâ, of having carried off numerous vanquished kings and princes to his city, and keeping them confined in his mountain stronghold with a view to afterwards sacrificing them (at his Râgasûya) to the lord of Umâ (Rudra); adding subsequently (v. 864) that 'the immolation of men was never seen at any time.'

xxxvi:2 His own mime 'Agîgarta,' on the other hand, is taken by the St. Petersburg Dictionary to mean 'one who has nothing to swallow,' and would thus be merely descriptive of his condition of life.

xxxvii:1 Viz. in their connection with the Atharva-veda. In Mahâbh. V, 548-51 Aṅgiras praises Indra by means of 'Atharvavedamantraih.' Cf. Weber, Ind. Stud. I, p. 237.

xxxvii:2 Both in making the fire-pan (ukhâ) and in laying down the bricks of the fire-altar, the expression 'aṅgirasvat' (as in the case of Aṅgiras) frequently occurs in the formulas; cf. VI, 1, 2, 28; 3, 1, 38 ff.; 4, 1, 1 ff.

xxxvii:3 All that is said in the Brâhmana regarding the headless bodies of the five victims is (VI, 2, 1, 7 seqq.) that Pragâpati, having cut off the heads, and put them on (the altar, i.e. on himself), plunged four of the trunks into the water, and brought the sacrifice to a completion by (offering) the he-goat (not a he-goat, as translated), and that he subsequently gathered up the water and mud (clay) in which those corpses had lain, and used them for making bricks for the altar. The view that the other four bodies should likewise he offered is rejected by the author, who rather seems to suggest that they should be allowed to float away on the water.

xxxvii:4 A very similar passage occurs in Ait. Br. VI, 8, on which cp. Max Müller, History of Ancient Sanskrit Literature, p. 420.

xxxviii:1 This doubtless is what is meant (cf. Kâty. XVI, 1, 38); and 'atha' at the beginning of VI, 2, 2, 6 ought accordingly to have been taken in the rather unusual sense of 'or' (? 'or rather'), instead of then.' Cf. VI, 2, 2, 15.

xxxviii:2 According to Âp. Sr. XVI, 17, 19-20, however, even if there is only one head (that of Vâyu's he-goat) all the formulas are to be pronounced over it.

xxxviii:3 The Maitr. Samhitâ, however, does not seem to refer to this particular point in its Brâhmana sections.

xxxix:1 Or, according to Âpastamba, for seven beans; the head to be that of a Kshatriya or a Vaisya killed either by an arrow-shot or by lightning, and apparently to be severed from the body at the time of purchase (which, as Professor Weber rightly remarks, is a merely symbolic one). As, however, the particulars given by Âpastamba are not mentioned in the older works, they may not unlikely have been introduced by him to meet some of the objections raised by the Vâgasaneyins to whose views he generally pays some attention. Otherwise the transaction might seem rather suspicious.

xxxix:2 Taitt. S. V, 1, 83, indeed, seems to speak of the other four animals being set free after fire has been carried round, so that their sacrificial use would he merely symbolical. Whether in that case only the head of the one animal would be used, or the man's head along with it, seems doubtful.

xxxix:3 Cf. Taitt. S. V, 5.

xl:1 The compound 'nihshiddhapâpmânah (apagrâmâh) 'may possibly be meant in the sense that the evil deeds of the outcasts are driven out (prevented from troubling the peace of the village); Katy. XX, 8, 17-18, however, states that when the Sacrificer has stepped out (of the water), evil-doers enter (to bathe in the water) without having performed any (other) rites, and that they are then said to be 'purified by the Asvamedha.'

xl:2 Besides the description of the ceremony in the present work (XIII, 6, 1-2, 20), only the Taittirîya-brâhmana (III, 4) seems to refer to it, enumerating merely the would-be victims who, according to Âpastamba, as quoted by Sâyana, are eventually set free. Professor Weber's suggestion that they may possibly at one time have been intended to be all of them slaughtered can hardly have been meant seriously. One might as well suppose that, at the Asvamedha, all the 'evil-doers' who, according to Kâtyâyana, are to bathe in the river, were meant to be drowned.

xli:1 When the practice became generally recognised that the Sacrificer (and priests) should eat a portion of the offered victim, this alone would, as Professor Weber suggests, have tended to make human sacrifices impracticable.

xlii:1 The Asvamedha section of the same work begins:--Pragâpati desired, 'May I gain all my desires, may I attain all attainments.' He beheld this three days’ sacrificial performance, the Asvamedha, and took it, and offered with it; and by offering with it he gained all his desires, and attained all attainments.

xlii:2 See XIII, 4, 2, 6-17.

xlii:3 See p. xxxii.

xlii:4 See XIII, 2, 2, 2, seqq.

xliii:1 That is, as it would seem, with a view to dissuading her husband from offering himself as a victim.

xliv:1 Dr. Garbe, in his translation, makes this and the subsequent rules refer (erroneously I think) to the animal victims of rule 21.

xlv:1 On this and other passages referred to the Mahâ-Kaushîtaka, cp. Professor Aufrecht's judicious remarks, Ait. Br., p. v.

xlix:1 The Kaushîtaki-brâhmana (VIII, 3), on the other hand, seems to justify the prohibition on the ground that, prior to the first complete Soma-sacrifice, the body of the Sacrifice (and Sacrificer) is incomplete, and therefore not ready to receive its head, in the shape of the Pravargya. Hence also the same work allows the Pravargya to be performed at the first Soma-sacrifice of one who is thoroughly versed in the scriptures, since such a one is himself the body, or self, of the sacrifice.

xlix:2 See XI, 1, 6, 10.

l:1 Cf. A. Weber, Satapatha-brâhmana, p. xi.

Next: XI, 1, 1. First Adhyâya. First Brâhmana