Oldenberg

INTRODUCTION

TO THE

GRIHYA-SŪTRAS.

We begin our introductory remarks on the literature of the Gr̥hya-sūtras with the attempt to collect the more important data which throw light on the development of the Gr̥hya ritual during the oldest period of Hindu antiquity.

There are, as it seems, no direct traces of the Gr̥hya ceremonies in the most ancient portion of Vedic literature. It is certain indeed that a number of the most important of those ceremonies are contemporaneous with or even earlier than the most ancient hymns of the Rig-veda, as far as their fundamental elements and character are concerned, whatever their precise arrangement may have been. However, in the literature of the oldest period they play no part. It was another portion of the ritual that attracted the attention of the poets to whom we owe the hymns to Agni, Indra, and the other deities of the Vedic Olympus, viz. the offerings of the Srauta-Ritual with their far superior pomp, or, to state the matter more precisely, among the offerings of the Srauta-Ritual the Soma offering. In the Soma offering centred the thought, the poetry, and we may almost say the life of the Vasishṭḥas, of the Viśvāmitras, &c., in whose families the poetry of the Rig-veda had its home. We may assume that the acts of the Gr̥hya worship, being more limited in extent and simpler in their ritual construction than the great Soma offerings, were not yet at that time, so far as they existed at all, decked out with the reciting of the poetic texts, which we find later on connected with them, and which in the case of the Soma offering came early to be used. Probably they were celebrated in simple unadorned fashion;

what the person making the offering had to say was doubtless limited to short, possibly prose formulas, so that these ceremonies remained free from the poetry of the above-mentioned families of priests ¹. We think that the character of the verses given in the Gr̥hya-sūtras, which had to be repeated at the performance of the different ceremonies, justifies us in making these conjectures. Some of these verses indeed are old Vedic verses, but we have no proof that they were composed for the purposes of the Gr̥hya ceremonies, and the connection in which we find them in the Rig-veda proves rather the contrary. Another portion of these verses and songs proves to have been composed indeed for the very Gr̥hya ceremonies for which they are prescribed in the texts of the ritual: but these verses are more recent than the old parts of the Rig-veda. Part of them are found in the Rig-veda in a position which speaks for their more recent origin, others are not contained in the Rig-veda at all. Many of these verses are found in the more recent Vedic Saṁhitās, especially in the Atharva-veda Saṁhitā which may be regarded in the main as a treasure of Gr̥hya verses; others finally have not as yet been traced to any Vedic Saṁhitā, and we know them from the Gr̥hya-sūtras only. We may infer that, during the latter part of the Rig-veda period, ceremonies such as marriage and burial began to be decked out with poetry as had long been the case with the Soma offering. The principal collection of marriage sentences ² and the sentences for the

burial of the dead ³ are found in the tenth Maṇḍala of the Rig-veda, which, for the most part, is known to be of later origin than the preceding portions of the collections ⁴. If we look into the character of the verses, which these long Gr̥hya songs are composed of, we shall find additional grounds for assuming their early origin. A few remarks about their metrical character will make this clear ⁵. There is no other metre in which the contrast between the early and later periods of Vedic literature manifests itself so clearly as in the Anushṭubh-metre ⁶. The Anushṭubh hemistich consists of sixteen syllables, which are divided by the caesura into two halves of eight syllables each. The second of these halves has as a rule the iambic ending ( ), though this rule was not so strictly carried out in the early as in the later period ⁷. The iambic ending is also the rule in the older parts of the Veda for the close of the first half, i.e. for the four syllables before the caesura ⁸. We know that the later prosody, as we see it in certain late parts of Vedic literature, in the Pāli Piṭakas of the Buddhists, and later in the great epic poems, not only departs from the usage of the older period, but adopts a directly contrary course, i.e. the iambic ending of the first pāda, which was formerly the rule, is not allowed at all later, and instead of it the prevailing ending is the antispast ( ) It goes without saying that such a change in metrical usage, as the one just described, cannot have

taken place at one jump. And accordingly a consideration of the Vedic texts reveals a transition period or rather a series of several transition periods between the old and the new standpoints. The first change is that every other ending of the first pāda is allowed by the side of the iambic ending. The two forms of the ending, the one prevailing in the earliest, and the one prevailing in the later period of the prosody, the iambic ( ) and the antispastic ( ), are those that occur most frequently in the intermediate period, but besides them all other possible forms are allowed ⁹.

This is precisely the stage of metrical development which the great Gr̥hya songs of the tenth Maṇḍala of the Rig-veda have reached. Let us consider, for instance, the marriage songs and the marriage sayings, X, 85, and see what kind of ending there is at the end of the first pāda. Of the first seventeen verses of this Sūkta sixteen are in Anushṭubh metre (verse 14 is Trishṭubh); we have therefore thirty-two cases in which the metrical form of these syllables must be investigated. The quantity of the syllable immediately preceding the caesura being a matter of indifference, we have not sixteen but only eight a priori possible combinations for the form of the last four syllables of the pāda; I give each of these forms below, adding each time in how many of the thirty-two cases it is used:

8

5

5

4

3

3

3

1

32

We see that all the possible combinations are actually represented in these thirty-two cases, and accordingly the metrical build of this Sūkta shows that it belongs to a period to which only the latest songs of the Rig-veda collection can be referred, but the peculiarities of which may be often noticed in the Atharva-veda and in the verses scattered throughout the Brāhmaṇa literature ¹⁰.

A hasty glance suffices to show that those verses of the Gr̥hya ritual which do not appear in the Saṁhitās, but which are quoted at full length in the Gr̥hya-sūtras, are also in the same stage. For instance, the seven Anushṭubh verses which are quoted Sāṅkhāyana-Gr̥hya I, 19, 5. 6, give us the following relations, if we investigate them as we did those in Rig-veda X, 85:

4

3

2

2

1

1

1

14

Thus even the small number of fourteen hemistichs is enough to give us seven of the eight existing combinations, and no single one occurs at all often enough to allow us to call it predominant.

Or we may take the saying that accompanies the performance of the medhājanana on the new-born child. In the version of Āśvalāyana ¹¹ we have:

In the version adopted in the school of Gobhila ¹² the

context of the first line is different, but the metre is the same:

Or the saying with which the pupil (brahmacārin) has to lay a log of wood on the fire of the teacher ¹³:

There would be no object in multiplying the number of examples; those here given are sufficient to prove our proposition, that the development of the Gr̥hya rites in the form in which they are described to us in the Sūtras, that especially their being accompanied with verses, which were to be recited by their performance, is later than the time of the oldest Vedic poetry, and coincides rather with the transition period in the development of the Anushṭubh metre, a period which lies between the old Vedic and the later Buddhistic and epic form.

Besides the formulae intended to be recited during the performance of the various sacred acts, the Gr̥hya-sūtras contain a second kind of verses, which differ essentially from the first kind in regard to metre; viz. verses of ritualistic character, which are inserted here and there between the prose Sūtras, and of which the subject-matter is similar to that of the surrounding prose. We shall have to consider these yajñagāthās, as they are occasionally called, later; at present let us go on looking for traces of the Gr̥hya ritual and for the origin of Gr̥hya literature in the literature which precedes the Sūtras.

The Brāhmaṇa texts, which, as a whole, have for their subject-matter the Vaitānika ceremonies celebrated with the three holy fires, furnish evidence that the Gr̥hya fire, together with the holy acts accomplished in connection with it, were also already known. The Aitareya-Brāhmaṇa ¹⁴ gives this

fire the most usual name, the same name which is used for it in the Sūtras, gr̥hya agni, and describes a ceremony to be performed over this fire with expressions which agree exactly with the style of the Gr̥hya-sūtras ¹⁵. We often find in the Brāhmaṇa texts also mention of the terminus technicus, which the Gr̥hya-sūtras use many times as a comprehensive term for the offerings connected with Gr̥hya ritual, the word pākayajña ¹⁶. For instance, the Satapatha Brāhmaṇa ¹⁷, in order to designate the whole body of offerings, uses the expression: all offerings, those that are Pākayajñas and the others. It is especially common to find the Pākayajñas mentioned in the Brāhmaṇa texts in connection with the myth of Manu. The Taittirīya Saṁhitā ¹⁸ opposes the whole body of sacrifices to the Pākayajñas. The former belonged to the gods, who through it attained to the heavenly world; the latter concerned Manu: thus the goddess Iḍā turned to him. Similar remarks, bringing Manu or the goddess Iḍā into relation with the Pākayajñas, are to be found Taittirīya Saṁhitā VI, 2, 5, 4; Aitareya-Brāhmaṇa III, 40, 2. However, in this case as in many others, the Satapatha Brāhmaṇa contains the most detailed data, from which we see how the idea of Manu as the performer of Pākayajñas is connected with the history of the great deluge, out of which Manu alone was left. We read in the Satapatha Brāhmaṇa ¹⁹:

‘Now the flood had carried away all these creatures, and thus Manu was left there alone. Then Manu went about singing praises and toiling, wishing for offspring. And he sacrificed there also with a Pāka-sacrifice. He poured clarified butter, thickened milk, whey, and curds in the water as a libation.’ It is then told how the goddess Iḍā arose out of this offering. I presume that the story of the Pākayajña as the first offering made by Manu after the great flood, stands in a certain correlation to the idea of the introduction of the three sacrificial fires through Purūravas ²⁰. Purūravas is the son of Iḍā; the original man Manu, who brings forth Iḍā through his offering, cannot have made use of a form of offering which presupposes the existence of Iḍā, and which moreover is based on the triad of the sacred fires introduced by Purūravas; hence Manu’s offering must have been a Pākayajña; we read in one of the Gr̥hya-sūtras ²¹: ‘All Pākayajñas are performed without Iḍā.’

There are still other passages in the Brāhmaṇa texts showing that the Gr̥hya offerings were already known; I will mention a saying of Yājñavalkya’s reported in the Satapatha Brāhmaṇa ²²: he would not allow that the daily morning and evening offering was a common offering, but said that, in a certain measure, it was a Pākayajña. Finally I would call attention to the offering prescribed in the last book of the Satapatha Brāhmaṇa ²³ for the man ‘who wishes that a learned son should be born to him;’ it is there stated that the preparation of the Ājya (clarified butter) should be performed ‘according to the rule of the Sthālīpāka (pot-boiling),’ and the way in which the offering is to

be performed is described by means of an expression, upaghātam ²⁴, which often occurs in the Gr̥hya literature in a technical sense.

We thus see that the Brāhmaṇa books are acquainted with the Gr̥hya fire, and know about the Gr̥hya offerings and their permanent technical peculiarities; and it is not merely the later portions of the Brāhmaṇa works such as the fourteenth book of the Satapatha Brāhmaṇa, in which we meet with evidence of this kind; we find it also in portions against the antiquity of which no objections can be raised.

While therefore on the one hand the Brāhmaṇa texts prove the existence of the Gr̥hya ceremonial, we see on the other hand, and first of all by means of the Brāhmaṇa texts themselves, that a literary treatment of this ritualistic subject-matter, as we find it in the Brāhmaṇas themselves with regard to the Srauta offerings, cannot then have existed. If there had existed texts, similar to the Brāhmaṇa texts preserved to us, which treated of the Gr̥hya ritual, then, even supposing the texts themselves had disappeared, we should still necessarily find traces of them in the Brāhmaṇas and Sūtras. He who will take the trouble to collect in the Brāhmaṇa texts the scattered references to the then existing literature, will be astonished at the great mass of notices of this kind that are preserved: but nowhere do we find traces of Gr̥hya Brāhmaṇas. And besides, if such works had ever existed, we should be at a loss to understand the difference which the Hindus make between the Srauta-sūtras based on Sruti (revelation), and the Gr̥hya-sūtras resting on Smr̥ti (tradition) alone ²⁵. The sacred Gr̥hya acts are regarded as ‘smārta,’ and when the question is raised with what right they can be considered as a duty resting on the sacrificer alongside of the Srauta acts, the answer is given that they too are based on a Sākhā of the Veda, but that this Sākhā is

hidden, so that its existence can only be demonstrated by reasoning ²⁶.

But the Brāhmaṇa texts furnish us still in another way the most decisive arguments to prove that there have been no expositions of the Gr̥hya ritual in Brāhmaṇa form: they contain exceptionally and scattered through their mass sections, in which they treat of subjects which according to later custom would have been treated in the Gr̥hya-sūtras. Precisely this sporadic appearance of Gr̥hya chapters in the midst of expositions of a totally different contents leads us to draw the conclusion that literary compositions did not then exist, in which these chapters would have occupied their proper place as integral parts of a whole. Discussions of questions of Gr̥hya ritual are found in the Brāhmaṇa literature, naturally enough in those appendices of various kinds which generally follow the exposition of the principal subject of the Srauta ritual. Accordingly we find in the eleventh book of the Satapatha Brāhmaṇa ²⁷, among the manifold additions to subjects previously treated, which make up the principal contents of this book ²⁸, an exposition of the Upanayana, i.e. the solemn reception of the pupil by the teacher, who is to teach him the Veda. The way in which the chapter on the Upanayana is joined to the preceding one, is eminently characteristic; it shows that it is the merest accident which has brought about in that place the discussion of a subject connected with the Gr̥hya ritual, and that a ceremony such as the Upanayana is properly not in its proper place in the midst of the literature of Brāhmaṇa texts. A dialogue (brahmodya) between Uddālaka and Sauceya precedes; the two talk of the Agnihotra and of various expiations (prāyaścitta) connected with that sacrifice. At the end Sauceya, filled with astonishment at the wisdom of Uddālaka, declares that he wishes to come to him as a pupil (upāyāni bhagavantam), and Uddālaka

accepts him as his pupil. It is the telling of this story and the decisive words upāyāni and upaninye which furnish the occasion for introducing the following section on the Upanayana ²⁹. The subject is there treated in the peculiar style of the Brāhmaṇa texts, a style which we need not characterize here. I shall only mention one point, viz. that into the description and explanation of the Upanayana ceremony has been inserted one of those Slokas, such as we often find in the Gr̥hya-sūtras also, as a sort of ornamental amplification of the prose exposition ³⁰. ‘Here a Sloka is also sung,’ says the Brāhmaṇa ³¹:

ācāryo garbhī bhavati hastam ādhāya dakshiṇam

tr̥tīyasyāṁ sa jāyate sāvitryā saha brāhmaṇaḥ ³².

From this passage we see, on the one hand, that the composition of such isolated ³³ Slokas explaining certain points of the Gr̥hya ritual goes back to quite an early period; on the other hand, we are compelled to assume that the Slokas of this kind which are quoted in the Gr̥hya-sūtras differ nevertheless from the analogous Slokas of the early period, or at any rate that the old Slokas must have undergone a change which modernized their structure, so as to be received into the Gr̥hya-sūtras; for the metre of the Sloka just quoted, which has the antispast before the caesura in neither of its two halves, and which has even a double iambus before the caesura in one half, is decidedly of an older type than the one peculiar to the Slokas quoted in the Gr̥hya-sūtras ³⁴.

Another Gr̥hya section in the Satapatha Brāhmaṇa seems to have found its place there through a similar accidental kind of joining on to a preceding chapter as the above-mentioned passage. In XI, 5, 5 a story of the battle of the gods and Asuras is told: the gods beat the Asuras back by means of constantly larger Sattra celebrations and conquer for themselves the world of heaven. It seems to me that the description of the great Sattras celebrated by the gods is the occasion of the joining on of a section beginning with the words ³⁵: ‘There are five great sacrifices (mahāyajñas); they are great Sattras: the offering to Beings, the offering to men, the offering to the Fathers (i.e. the Manes), the offering to the Gods, the offering to the Brahman.’ After this introduction follows an account of one of the five great offerings, namely of the Brahmayajña, i.e. of the daily Veda recitation (svādhyāya). The third Adhyāya of Āśvalāyana’s Gr̥hya-sūtra begins in exactly the same way with the sentence: ‘Now (follow) the five sacrifices: the sacrifice to the Gods, the sacrifice to the Beings, the sacrifice to the Fathers, the sacrifice to the Brahman, the sacrifice to men,’ and then follows here also a discussion of the Brahmayajña, which is entirely analogous to that given in the Satapatha Brāhmaṇa. Āśvalāyana here does not content himself with describing the actual course of ceremonies as is the rule in the Sūtra texts; he undertakes, quite in the way of the Brāhmaṇa texts, to explain their meaning: ‘In that he recites the Ricas, he thereby satiates the gods with oblations of milk, in that (he recites) the Yajus, with oblations of ghee,’ &c. It is plain that the mode of exposition adopted by Āśvalāyana in this passage, which is different from the usual Sūtra style, finds its explanation in the supposition that exceptionally in this case the author of the Gr̥hya-sūtra had before him a Brāhmaṇa text, which he could take as his model, whether that text was the Satapatha itself or another similar text.

Among the extremely various prescriptions which we find

in the last sections of the Satapatha Brāhmaṇa, there is a rather long section, which also really belongs to the Gr̥hya domain. To quote from this section ³⁶: ‘If a man wishes that a learned son should be born to him, famous, a public man, a popular speaker, that he should know all the Vedas, and that he should live to his full age, then, after having prepared boiled rice with meat and butter, they should both eat, being fit to have offspring,’ &c. Then follows a description of an Ājya offering, after which the marital cohabitation is to be performed with certain formulas. This, however, is not the last of the acts through which the father assures himself of the possession of such a distinguished son; certain rites follow, which are to be performed at birth and after birth, the Āyushya ceremony and the Medhājanana. These rites are here prescribed for the special case where the father has the above-mentioned wishes for the prosperity of his child; but the description agrees essentially with the description of the corresponding acts in the Gr̥hya-sūtras ³⁷, which are inculcated for all cases, without reference to a determined wish of the father. It is a justifiable conjecture that, although this certainly does not apply to the whole of ceremonies described in the Gr̥hya-sūtras, many portions of these ceremonies and verses that were used in connection with them, &c., were first developed, not as a universal rite or duty, but as the special possession of individuals, who hoped to attain special goods and advantages by performing the ceremony in this way.

It was only later, as I think, that such prescriptions

assumed the character of universality, with which we find them propounded in the Gr̥hya-sūtras.

It is scarcely necessary to go through the sections of the texts of other Vedic schools referring to the Gr̥hya ritual in the same way in which we have done it in the case of the Satapatha Brāhmaṇa. The data which we have produced from the great Brāhmaṇa of the white Yajur-veda, will be sufficient for our purpose, which is to give an idea of the stage in which the literary treatment of the Gr̥hya ritual stood during the Brāhmaṇa period. As we see, there were then properly no Gr̥hya texts; but many of the elements which we find later in the Gr̥hya texts were either already formed or were in the process of formation. Most of the verses which are used for the Gr̥hya acts - in so far as they are not verses composed in the oldest period for the Soma offering and transferred to the Gr̥hya ceremonies - bear the formal imprint of the Brāhmaṇa period; the domestic sacrificial fire and the ritual peculiarities of the Pākayajñas which were to be performed at it, were known; descriptions of some such Pākayajñas were given in prose; there were also already Slokas which gave in metrical form explanations about certain points of the Gr̥hya ritual, just as we find in the Brāhmaṇa texts analogous Slokas referring to subjects connected with the Srauta ritual.

Thus was the next step which the literary development took in the Sūtra period prepared and rendered easy. The more systematic character which the exposition of the ritualistic discipline assumed in this period, necessarily led to the taking of this step: the domain of the Gr̥hya sacrifices was recognised and expounded as a second great principal part of the ritual of sacrifices alongside of the Srauta domain which was alone attended to in the earlier period. The Gr̥hya-sūtras arose which treat, according to the expression of Āśvalāyana in his first sentence, of the gr̥hyāṇi ³⁸ as distinguished from the vaitānikāni, or, as Sāṅkhāyana says, of the pākayajñās, or, as Pāraskara says, of the gr̥hyasthālīpākānāṁ karma. The

Gr̥hya-sūtras treat their subject of course in exactly the same style in which the sacrifices of the Srauta ritual had been treated by the Srauta-sūtras, which they constantly assume to be known and which are the works of teachers of the same Vedic schools, and oftentimes even perhaps the works of the same authors. Only certain differences in the character of the two groups of texts are naturally conditioned on the one hand by the greater complexity of the Srauta sacrifices and the comparative simplicity of the Gr̥hya sacrifices, on the other hand by the fact that the Srauta-sūtras are entirely based on Brāhmaṇa texts, in which the same subjects were treated, while the Gr̥hya-sūtras, as we have seen, possessed such a foundation only for a very small portion of their contents.

It goes without saying that the above-mentioned statement that the subjects treated of in the Gr̥hya-sūtras are Pākayajñas ³⁹ or Gr̥hyasthālīpākas should not be pressed with the utmost strictness, as though nothing were treated in the Gr̥hya-sūtras which does not come under these heads. First of all the term Sthālīpāka is too narrow, since it does not include the offerings of sacrificial butter which constituted a great number of ceremonies. But besides many ceremonies and observances are taught in the Gr̥hya-sūtras, which cannot in any way be characterised as sacrifices at all, only possessing some inner resemblance to the group of sacrifices there treated of, or standing in more or less close connection with them ⁴⁰.

The Sūtra texts divide the Pākayajñas in various ways; either four or seven principal forms are taken up. The

commonest division is that into the four classes of the hutas, ahutas, prahutas, prāśitas ⁴¹. The division into seven classes is doubtless occasioned by the division of the Haviryajñas and of the Somayajñas, which also each include seven classes ⁴²; for the nature of the sacrifices in question would hardly of itself have led to such a division. The seven classes taken up are either those given by Gautama VIII, 15 ⁴³: ‘The seven kinds of Pākayajñas, viz. the Ashṭakā, the Pārvaṇa (Sthālīpāka, offered on the new and full moon days), the funeral oblations, the Srāvaṇī, the Āgrahāyaṇī, the Kaitrī, and the Āśvayujī.’ Or else the seven classes are established as follows, the fourfold division being utilised to some extent ⁴⁴: ‘Huta, Prahuta, Āhuta (sic, not Ahuta), the spit-ox sacrifice, the Bali offering, the re-descent (on the Āgrahāyaṇa day), the Ashṭakā sacrifice.’ According to the account of Prof. Bühler ⁴⁵, the exposition of Baudhāyana, who gives this division, keeps closely to the course which it prescribes. For the rest, however, the Gr̥hya texts with which I am acquainted do not follow any of these divisions, and this is easily accounted for, if we consider the artificial character of these classifications, which are undertaken merely for the sake of having a complete scheme of the sacrifices. On the contrary, as a whole the texts give an arrangement which is based on the nature of the ceremonies they describe. In many instances we find considerable variations between the texts of the different schools; often enough, in a given text, the place

which is assigned to a given chapter is not to be explained without assuming a certain arbitrariness on tlfe part of the author. But, as a whole, we cannot fail to recognise in the arrangement of the different texts a certain agreement, which we will here merely try to explain in its main traits; the points of detail, which would complete what we here say, will occur of themselves to any one who looks at the texts themselves.

The domestic life of the Hindus represents, so to speak, a circle, in which it is in a certain measure indifferent what point is selected as the starting-point. Two especially important epochs in this life are: on the one hand, the period of studentship of the young Brahmacārin devoted to the study of the Veda; at the beginning of this period comes the ceremony of the Upanayana, at the end that of the Samāvartana; on the other hand, marriage (vivāha), which besides has a special importance for the Gr̥hya ritual, from the circumstance, that as a rule the cultus of the domestic sacrificial fire begins with marriage. One can just as well imagine an exposition of the Gr̥hya ritual, which proceeds from the description of the studentship to that of the marriage, as one which proceeds from the description of the marriage to that of the studentship. The Samāvartana, which designates the end of the period of studentship, gives the Hindu the right and the duty to found a household ⁴⁶. On the other hand, if the exposition begins with the marriage, there follows naturally the series of ceremonies which are to be performed up to the birth of a child, and then the ceremonies for the young child, which finally lead up to the Upanayana and a description of the period of studentship. The Hiraṇyakeśi-sūtra alone, of the Sūtras treated of in these translations, follows the first of the two orders mentioned ⁴⁷; the other texts follow the other order,

which has been already described by Prof. Max Müller almost thirty years ago, and we cannot do better than to give his description ⁴⁸: ‘Then (i.e. after the marriage) follow the Saṁskāras, the rites to be performed at the conception of a child, at various periods before his birth, at the time of his birth, the ceremony of naming the child, of carrying him out to see the sun, of feeding him, of cutting his hair, and lastly of investing him as a student, and handing him to a Guru, under whose care he is to study the sacred writings, that is to say, to learn them by heart, and ’to perform all the offices of a Brahmacārin, or religious student.’

In this way we find, as a rule, in the foreground in the first part of the Gr̥hya-sūtras this great group of acts which accompany the domestic life from marriage to the studentship and the Samāvartana of the child sprung from wedlock. We find, however, inserted into the description of these ceremonies, in various ways in the different Sūtras, the exposition of a few ritualistic matters which we have not yet mentioned. In the first place a description of the setting up of the sacred domestic fire, i.e. of the ceremony which in the domain of the Gr̥hya ritual corresponds to the agnyādheya of the Srauta ritual. The setting up of the fire forms the necessary preliminary to all sacred acts; the regular time for it is the wedding ⁴⁹, so that the fire used for the wedding acts accompanies the young couple to their home, and there forms the centre of their household worship. Accordingly in the Gr̥hya-sūtras the description of the setting up of the fire stands, as a rule, at the beginning of the whole, not far from the description of the wedding.

Next the introductory sections of the Gr̥hya-sūtras have to describe the type of the Gr̥hya sacrifice, which is universally available and recurs at all household ceremonies. This can be done in such a way that this type is described for itself, without direct reference to a particular sacrifice. This is the case in Pāraskara, who in the first chapter of his

Sūtra describes the rites recurring at each sacrifice, and then remarks: ‘This ritual holds good, whenever a sacrifice is offered ⁵⁰.’ Similarly Āśvalāyana, in one of the first chapters of his work, enumerates the rites which are to be performed ‘whenever he intends to sacrifice ⁵¹.’ Other texts give a general description of the Gr̥hya sacrifice by exemplifying it by one special sacrifice. Sāṅkhāyana ⁵² chooses for this the sacrifice which the bridegroom has to offer, when a favourable answer has been granted to his wooing; Gobhila ⁵³ gives at least the greater part of the rules in question à propos of the full moon and of the new moon sacrifice; Hiraṇyakeśin ⁵⁴, who opens his account at the period of the studentship of the young Brāhmaṇa, describes the sacrificial type à propos of the Upanayana rite.

The sacrifices which are to be offered daily at morning and at evening, those which are celebrated monthly on the days of the new moon and of the full moon - the Gr̥hya copies of the Agnihotra and of the Darśapūrṇamāsa sacrifices - and, thirdly, the daily distribution of the Bali offerings: these ceremonies are commonly described along with what we have called the first great group of the Gr̥hya acts, immediately preceding or following the Vivāha.

We find, as a second group of sacred acts, a series of celebrations, which, if the man has founded his household, are to be performed regularly at certain times of the year at the household fire. So the Sravāṇa sacrifice, which is offered to the snakes at the time when, on account of the danger from snakes, a raised couch is necessary at night. At the end of this period the festival of the re-descent is celebrated: the exchanging of the high couch for the low couch on the ground. Between these two festivals comes the Pr̥shātaka offering on the full-moon day of the month Āśvayuja; it receives in the Gr̥hya texts the place corresponding to that which actually belongs to

it in the series of the festivals. As a rule ⁵⁵ the acts we have just mentioned are followed, in accordance with the natural series, by the Ashṭakā festivals, which are celebrated during the last months of the year.

Alongside of these acts which are connected with fixed points of the year we find in the various Gr̥hya texts an account of a series of other ceremonies, which, in accordance with their nature, have no such fixed position in the system of the ritual. Thus, for instance, the rites which refer to the choice of a piece of ground to build a house or to the building itself; further, the rites connected with agriculture and cattle raising. In many texts we find together with this group of acts also an account of the ceremonies, related to fixed points in the year, which stand in connection with the annual course of Vedic study: the description of the opening festival and of the closing festival of the school term, as well as a point which generally follows these descriptions, the rules as to the anadhyāya, i.e. as to the occasions which necessitate an intermission in the study of the Veda for a longer or for a shorter period. As a rule, the Gr̥hya-sūtras bring the account of these things into the group of acts which refer to the household life of the Gr̥hastha; for the Adhyāpana, i.e. the teaching of the Veda, held the first place among the rights and duties of the Brāhmaṇa who had completed his time at school. On the other hand these ceremonies can naturally also be considered as connected with the school life of the young Hindu, and accordingly they are placed in that division by Gobhila ⁵⁶, between the description of the Upanayana and that of the Samāvartana.

The sacred acts connected with the burial and the worship of the dead (the various kinds of Srāddha rites) may be designated as a third group of the ceremonies which are described to us in the Gr̥hya.-sūtras. Finally, a fourth group comprises the acts which are connected with the attainment of particular desires (kāmyāni). Among the

texts here translated we find a somewhat detailed account of these ceremonies in the Gobhila-sūtra and in the Khādira-Gr̥hya only ⁵⁷.

These remarks cannot claim to give a complete outline of the contents and arrangement of the Gr̥hya texts; they only aim at giving an idea of the fundamental traits, which in each particular text are modified by manifold variations, but which nevertheless are to these variations as the rule is to the exceptions.

We must now speak of the relations of the Gr̥hya-sūtras to the two other kinds of Sūtra texts, with which they have so many points of contact in the Srauta-sūtras and the Dharma-sūtras.

Prof. Bühler, in several places of the excellent introductions which he has prefixed to his translations of the Dharma-sūtras, has called attention to the fact that the relation in which the Sūtra texts of the same school stand to each other is very different in different schools. Many schools possess a great corpus of Sūtras, the parts of which are the Srauta-sūtra, the Gr̥hya-sūtra, &c. This is, for instance, the case with the Āpastambīya school ⁵⁸; its Sūtra is divided into thirty Praśnas, the contents of which are divided as follows:

I-XXIV: Srauta-sūtra.

XXV: Paribhāshās, &c.

XXVI: Mantras for the Gr̥hya-sūtra.

XXVII: Gr̥hya-sūtra.

XXVIII-XXIX: Dharma-sūtra.

XXX: Sulva-sūtra.

In other cases the single Sūtra texts stand more independently side by side; they are not considered as parts of one and the same great work, but as different works. Of course it is the Dharma-sūtras above all which could be freed from the connection with the other Sūtra texts to such an extent, that even their belonging to a distinct Vedic school may be doubtful. The contents

of this class of Sūtras indeed have hardly any connection with the subdivisions and differences of the Vedic texts handed down in the various schools; there was no reason why Brahmans, who studied various Sākhās of the Veda, should not learn the ordinances concerning law and morals given in these Sūtras as they were formulated in the same texts. The Gr̥hya-sūtras are not so independent of the differences of the Vedic schools. The close analogy between the sacrificial ritual of the Gr̥hya acts and that of the Srauta acts, and the consequent necessity of taking into account the Srauta ritual in the exposition of the Gr̥hya ritual, necessarily brought the Gr̥hya-sūtras into closer connection with and into greater dependence on the Srauta-sūtras than in the case of the Dharma-sūtras ⁵⁹. But above all, the Gr̥hya ceremonies demanded the knowledge of numerous Mantras, and accordingly as these Mantras were borrowed from the one or the ether Mantra Sākhā ⁶⁰, there followed in the case of the Gr̥hya text in question an intimate connection with the corresponding Mantra school ⁶¹. We find accordingly as a general rule, that each Gr̥hya-sūtra presupposes a Vedic Saṁhitā, whose Mantras it quotes only in their Pratīkas ⁶², and that besides each Gr̥hya-sūtra presupposes a previous

knowledge of the ritual which is acquired through the study of the proper Srauta-sūtra ⁶³. It is not necessary to quote the numerous places where the Gr̥hya-sūtras either expressly refer to the Srauta-sūtras, or point to them by repeating the same phrases or often even whole Sūtras. It will be sufficient to quote one out of many places, the opening words of the Āśvalāyana-Gr̥hya, which in a way characterise this work as a second part of the Srauta-sūtra: ‘The rites based on the spreading (of the three sacred fires) have been declared; we shall declare the Gr̥hya rites ⁶⁴.’

Thus it is not difficult to perceive the dependence of the Gr̥hya-sūtras on the Srauta-sūtras; but there remains the much more difficult question whether in each particular case both texts are to be regarded as by the same author, or whether the Gr̥hya-sūtra is an appendix to the Srauta-sūtra composed by another author. Tradition accepts the one alternative for some Sūtras; for other Sūtras it accepts the other; thus in the domain of the Rig-Veda literature Āśvalāyana and Sāṅkhāyana are credited with the authorship of a Srauta-sūtra as well as of a Gr̥hya-sūtra; the same is true of Āpastamba, Hiraṇyakeśin, and other authors. On the other hand, the authorship of the Gr̥hya-sūtras which follow the Srauta-sūtras of Kātyāyana, Lāṭyāyana, Drāhyāyaṇa, is not ascribed to Kātyāyana, Lāṭyāyana, Drāhyāyaṇa, but to Pāraskara, Gobhila, and Khādirācārya.

It seems to me that we should consider the testimony of tradition as entirely trustworthy in the second class of cases. Tradition is very much inclined to ascribe to celebrated masters and heads of schools the origin of works which are acknowledged authorities in their schools, even though they are not the authors. But it is not likely that tradition should have made a mistake in the opposite

direction, that e.g. it should designate Pāraskara as author when Kātyāyana himself was the author.

We shall not be able to trust so implicitly to tradition where it puts down the same author for the Gr̥hya-sūtra as for the corresponding Srauta-sūtra; the possibility that such data are false is so large that we have to treat them as doubtful so long as we have not discovered certain proofs of their correctness. At present, so far as I can see, we are just as little justified in considering that such a proof has been made as we are able to prove the opposite state of things. It is easy to find the many agreements in contents and expression which exist, for instance, between the Srauta-sūtra and Gr̥hya-sūtra of Sāṅkhāyana, or between the Srauta-sūtra and the Gr̥hya-sūtra of Āśvalāyana ⁶⁵. But these agreements cannot be considered as sufficient proof that in each case the Gr̥hya-sūtra and the Srauta-sūtra are by the same author. Even if the author of the Gr̥hya-sūtra was not Āśvalāyana or Sāṅkhāyana in person, still he must have been at all events perfectly familiar with the works of those teachers, and must have intended to fit his work to theirs as closely as possible, so that agreements of this kind can in no way astonish us ⁶⁶. On the other hand, if the Srauta-sūtras and Gr̥hya-sūtras are read together, it is easy to discover small irregularities in the exposition, repetitions and such like, which might seem to indicate different authors. But the irregularities of this kind which have been detected up to the present are scarcely of such

a character as not to be easily ascribable to mistakes and carelessness such as even a careful author may be guilty of in the course of a large work ⁶⁷. It seems to me then that until the discovery of further circumstances throwing light on the question of the identity of the authors of the Srautas and of the Gr̥hyas, it would be premature if we were to venture on a decision of this question in one direction or the other.

Prof. Bühler’s investigations have made perfectly clear the relation in which the Gr̥hya-sūtras and the Dharma-sūtras stand to each other in those cases, where we have texts of both kinds by the same school. In the case of the Gr̥hya-sūtra and the Dharma-sūtra of the Āpastambīyas he has proved ⁶⁸ that both texts were the work of the same author according to a common plan, so that the Gr̥hya-sūtra is as short and terse as possible, because Āpastamba had reserved for the Dharma-sūtra a portion of the subject-matter generally treated of in the Gr̥hya-sūtras. Besides there are references in each of the two texts to the other which strengthen the proof of their being written by the same author. In the Sūtra collection of Hiraṇyakeśin the state of things is different. Here, as Prof. Bühler has also shown ⁶⁹, we find numerous discrepancies between the Gr̥hya and the Dharma-sūtra, which are owing to the fact, that while this teacher took as Dharma-sūtra that of Āpastamba with some unessential changes, he composed a Gr̥hya-sūtra of his own. Of the two Sūtras of Baudhāyana, the same distinguished scholar, to whom we owe the remarks we have just mentioned, has treated in the Sacred Books of the East, vol. xiv, p. xxxi.

I believe that every reader who compares the two kinds of texts will notice that the frame within which the exposition of the Dharma-sūtras is inclosed, is an essentially

broader one than in the case of the Gr̥hya-sūtras. We have here, I think, the same phenomenon that may also be observed, for instance, in the domain of the Buddhist Vinaya literature, where the exposition of the life of the community was at first given only in connection with the explanation of the list of sins (Pātimokkha) which was promulgated every half month at the meetings of the spiritual brethren. It was not till later that a more comprehensive exposition, touching all the sides of the life of the community was attempted ⁷⁰, an exposition which, on the one hand, no longer limited itself to the points discussed in the Pātimokkha, and which, on the other hand, necessarily had much in common with what was laid down in the Pātimokkha. The relation of the Gr̥hya-sūtras and Dharma-sūtras seems to me to be of a similar nature. The Gr̥hya-sūtras begin to treat of the events of the daily life of the household, but they do not yet undertake to exhaust the great mass of this subject-matter; on the contrary they confine themselves principally to the ritual or sacrificial side of household life, as is natural owing to their connection with the older ritualistic literature. Then the Dharma-sūtras take an important step further; their purpose is to describe the whole of the rights and customs which prevail in private, civic, and public life. They naturally among other things touch upon the ceremonies treated in the Gr̥hya-sūtras, but they generally merely mention them and discuss the questions of law and custom which are connected with them, without undertaking to go into the technical ordinances as to the way in which these ceremonies are to be performed ⁷¹.

Only in a few cases do portions treated of in the domain of the Dharma-sūtras happen to coincide with portions treated of in the Gr̥hya-sūtras. Thus especially, apart from a few objects of less importance, the detailed rules for the behaviour of the Snātaka and the rules for the interruptions

of the Veda study (anadhyāya) are generally treated in an exactly similar way in the texts of the one and those of the other category.

---

We have spoken above of the metrical peculiarities of the Mantras quoted in the Gr̥hya-sūtras, the metre of which clearly proves what is indubitable from other reasons, that most, if not all, of these verses were composed at a perceptibly older period than the descriptions of the sacred acts in the midst of which they are inserted ⁷². A second kind of verses which are quoted in the Gr̥hya-sūtras must be carefully distinguished from these. It is doubtful whether there are any to be found among them which the authors of the Sūtras have themselves composed; but they were composed at a period decidedly more recent than those Mantras ⁷³, and they therefore exhibit metrical peculiarities which are essentially different. The verses I mean are Slokas of ritual contents, which are quoted to confirm or to complete what is stated in the prose, and which are introduced by such expressions as tad apy āhuḥ ‘here they say also,’ or tad api ślokāḥ ‘here there are also Slokas,’ and other similar phrases ⁷⁴.

We called attention above (p. xix) to the fact that a verse of this kind occurs in one of the Gr̥hya chapters of the Satapatha Brāhmaṇa, in a metre corresponding to the peculiarities of the older literary style. On the other hand, the verses appearing in the Gr̥hya-sūtras differ only in a few cases from the standard of the later Sloka prosody, as we have it, e. g. in the Mahābhārata and in the laws of Manu. In the Zeitschrift der Deutschen Morgenländ. Gesellschaft, vol. xxxvii, p. 67, I have given tables for the verses in question out of the Sāṅkhāyana-Gr̥hya, and these tables show that the characteristic ending of the first

Sloka Pāda for the later period , which, for instance, in the Nalopākhyāna of the Mahābhārata covers precisely five-sixths of all the cases, occurs in Sāṅkhāyana in thirty cases out of thirty-nine, that is in about three quarters of the cases ⁷⁵; Sāṅkhāyana has still twice the ending which is the rule in the Rig-veda, but which is forbidden by the later prosody: prahutaḥ pitr̥karmaṇā, uktvā mantraṁ spr̥śed apaḥ ⁷⁶. It may be observed that a similar treatment of the Sloka metre appears also in the Rig-veda Prātiśākhya of Saunaka. Here too the modern form of the ending of the first pāda dominates, although sometimes the old iambic form is preserved, e. g. II, 5 antaḥpadaṁvivr̥ttayaḥ, III, 6 anudāttodaye punaḥ.

It seems evident that we have in this Sloka form of the Sūtra period, the last preparatory stage which the development of this metre had to traverse, before it arrived at the shape which it assumes in epic poetry; and it is to be hoped that more exhaustive observations on this point (account being especially taken of the numerous verses quoted in the Dharma-sūtras) will throw an important light on the chronology of the literature of this period lying between the Vedas and the post-Vedic age.

We add to these remarks on the Slokas quoted in the Gr̥hya-sūtras, that we come upon a number of passages in the midst of the prose of the Sūtras, which without being in any way externally designated as verses, have an unmistakable metrical character, being evidently verses which the authors of the Sūtras found ready made, and which they used for their own aphorisms, either without changing them at all, or with such slight changes that the original form remained clearly recognisable. Thus we read in Āśvalāyana (Gr̥hya I, 6, 8), as a definition of the Rākshasa marriage: hatvā bhittvā ca śīrshāṇi rudatīṁ rudadbhyo

haret: the approximation of these words to the Sloka metre cannot escape attention, and it is only necessary to make rudadbhyaḥ and rudatīṁ change places in order to obtain a regular Sloka hemistich. In Gobhila the Sūtras I, 2, 21-27 represent three hemistichs, which with one exception (na ca sopānatkaḥ kvacit) exactly conform to the laws of the Sloka metre. II, 4, 2 gives also a hemistich by slightly changing the order:

Mahāvr̥kshān śmaśānaṁ ca nadīś ca vishamāṇi ca ⁷⁷.

Somewhat more remote from the original verses is the wording of the Sūtras I, 6, 8. 9 na pravasann upavased ity āhuḥ, patnyā vrataṁ bhavatīti; we have the metrical order in one of the Slokas quoted by Sāṅkhāyana (Gr̥hya II, 17): nopavāsaḥ pravāse syāt patnī dhārayate vratam.

The verses which are thus either expressly quoted, or at any rate made use of by the authors of the Gr̥hya-sūtras, do not seem to be taken from connected metrical works any more than the yajñagāthās quoted in the Brāhmaṇas; on the contrary in a later period of literature, when texts similar to Manu’s Code were composed, they evidently furnished these texts with some of their materials ⁷⁸.

---

Leaving out of consideration the Khādira-Gr̥hya, which is evidently a recast of the Gobhilīya-Gr̥hya, and the Sūtra of Hiraṇyakeśin, which is, at least in part, based on that of Āpastamba ⁷⁹, we are not in regard to the other Gr̥hya texts in a condition to prove that one of them borrowed from the other. It often happens that single Sūtras or whole rows of Sūtras agree so exactly in different texts that this agreement cannot be ascribed to chance; but this does not - so far at least - enable us to tell which text is to be looked upon as the source of the

other, or whether they have a common source which has been lost.

I will content myself with mentioning two such cases of agreement, in the one of which we can at least prove that a certain Sūtra cannot originally spring from one of the texts in which we find it, while in the other case we are able by means of a possibly not too uncertain conjecture to reconstruct the opening Sūtras of a lost Gr̥hya-sūtra.

The description of the vr̥shotsarga (i.e. of the setting a bull at liberty) agrees almost word for word in the Sūtras of Sāṅkhāyana (III, 11), Pāraskara (III, 9), and in the Kāṭḥaka-Gr̥hya. In Sāṅkhāyana we read:

§ 15: nabhyasthenumantrayate mayobhūr ity anuvākaśesheṇa.

(‘When the bull is in the midst of the cows, he recites over them the texts “mayobhūḥ, &c.,” down to the end of the Anuvāka.’)

On the other hand in Pāraskara we have:

§ 7: nabhyastham abhimantrayate mayobhūr ity anuvākaśesheṇa.

(‘When the bull is in the midst of the cows, he recites over it the texts “mayobhūḥ, &c.,” down to the end of the Anuvāka.’)

The quotation mayobhūḥ is clear, if we refer it to the Rig-veda. Hymn X, 169, which stands about in the middle of an Anuvāka, begins with this word ⁸⁰. On the other hand in the Vājasaneyi Saṁhitā there is no Mantra beginning with Mayobhūḥ; we find this word in the middle of the Mantra XVIII, 45, and there follow verses whose use at the vr̥shotsarga would seem in part extremely strange. There can thus be no doubt that Pāraskara here borrowed from a Sūtra text belonging to the Rig-veda, a Pratīka, which, when referred to the Vājasaneyi Saṁhitā, results in nonsense.

The other passage which I wish to discuss here is Pāraskara

I, 4,1-5. Pāraskara, being just on the point of describing the marriage ritual, prefixes a few sentences, the position of which here it is not very easy to understand. A general division of all Pākayajñas - general remarks on the nature of the place for sacrificing: this looks very strange between a discussion of the Arghya and marriage ceremonies. Now these same sentences are found almost word for word and with the same passing on to the marriage ritual in Sāṅkhāyana also (Gr̥hya I, 5, 1-5). Here, as in other cases, we have the borrowing word for word of such portions of text from an older text, and, closely related to this phenomenon, the fact that the sentences in question are awkwardly woven into the context of the Gr̥hya where we read them, and are poorly connected with the surrounding parts. Unless we are much deceived, we have here a fragment from an older source inserted without connection and without change. It would seem that this fragment was the beginning of the original work; for the style and contents of these Sūtras are peculiarly appropriate for the beginning. Thus, if this conjecture is right, that old lost Gr̥hya began with the main division of all the Pākayajñas into four classes, and then proceeded at once to the marriage ritual. Later, when the texts which we have, came into existence, the feeling evidently arose, that in this way an important part of the matter had been overlooked. The supplementary matter was then inserted before the old beginning, which then naturally, as is to be seen in our texts, joins on rather strangely and abruptly to these newly-added portions.