Sacred Texts  Hinduism  Index  Previous  Next 

INTRODUCTION.

   THE Vishnu-smriti or Vaishnava Dharmasâstra or Vishnu-sûtra is in the main a collection of ancient aphorisms on the sacred laws of India, and as such it ranks with the other ancient works of this class which have come down to our time^[1]. It may be styled a Dharma-sûtra, though this ancient title of the Sûtra works on law has been preserved in the MSS. of those Smritis only, which have been handed down, like the Dharma-sûtras of Âpastamba, Baudhâyana, and Hiranyakesin, as parts of the respective Kalpa-sûtras, to which they belong. The size of the Vishnu-sûtra, and the great variety of the subjects treated in it, would suffice to entitle it to a conspicuous place among the five or six existing Dharma-sûtras; but it possesses a peculiar claim to interest, which is founded on its close connection with one of the oldest Vedic schools, the Kathas, on the one hand, and with the famous code of Manu and some other ancient law-codes, on the other hand. To discuss these two principal points, and some minor points connected with them, as fully as the limits of an introduction admit of, will be the more necessary, because such a discussion can afford the only safe basis for a conjecture not altogether unsupported regarding the time and place of the original composition of this work, and may even tend to throw some new light on the vexed question as to the origin of the code of Manu. Further on I shall have to speak of the numerous interpolations traceable in the Vishnu-sûtra, and a few remarks regarding the materials

p. x

used for this translation, and the principles of interpretation that have been followed in it, may be fitly reserved for the last.

   There is no surer way for ascertaining the particular Vedic school by which an ancient Sanskrit law-book of unknown or uncertain origin was composed, than by examining the quotations from, and analogies with, Vedic works which it contains. Thug the Gautama Dharmasâstra might have originated in any one among the divers Gautama Karanas with which Indian tradition acquaints us. But the comparatively numerous passages which its author has borrowed from the Samhitâ and from one Brâhmana of the Sâma-veda prove that it must belong to one of those Gautama Karanas who studied the Sâma-veda^[1]. Regarding the code of Yâgñavalkya we learn from tradition that a Vedic teacher of that name was the reputed author of the White Yagur-veda. But this coincidence might be looked upon as casual, if the Yâgñavalkya-smriti did not contain a number of Mantras from that Vedic Samhitâ, and a number of very striking analogies, in the section on funeral ceremonies particularly, with the Grihya-sûtra of the Vâgasaneyins, the Kâtiya Grihya-sûtra of Pâraskara^[2]. In the case of the Vishnu-sûtra an enquiry of this kind is specially called for, because tradition leaves us entirely in the dark as to its real author. The fiction that the laws promulgated in Chapters II-XCVII were communicated by the god Vishnu to the goddess of the earth, is of course utterly worthless for historical purposes; and all that it can be made to show is that those parts of this work in which it is started or kept up cannot rival the laws themselves in antiquity.

   Now as regards, first, the Vedic Mantras and Pratîkas (beginnings of Mantras) quoted in this work, it is necessary to leave aside, as being of no moment for the present purpose, 1. very well-known Mantras, or, speaking more

p. xi

precisely, all such Mantras as are frequently quoted in Vedic works of divers Sâkhâs; 2. the purificatory texts enumerated under the title of Sarva-veda-pavitrâni in LVI. The latter can afford us no help in determining the particular Sâkhâ to which this work belongs, because they are actually taken, as they profess to be, from all the Vedas indiscriminately, and because nearly the whole of Chapter LVI is found in the Vâsishtha-smriti as well (see further on), which probably does not belong to the same Veda as this work. Among the former class of Mantras may be included, particularly, the Gâyatrî, the Purushasûkta, the Aghamarshana, the Kûshmândîs, the Vyâhritis, the Gyeshtha Sâmans, the Rudras, the Trinâkiketa, the Trisuparna, the Vaishnava, Sâkra, and Bârhaspatya Mantras mentioned in XC, 3, and the Mantra quoted in XXVIII, 51 (= Gautama's 'Retasya'). Among the twenty-two Mantras quoted in Chapters XLVIII, LXIV, LXV (including repetitions, but excluding the Purushasûkta, Gâyatrî, Aghamarshana) there are also some which may be referred to this class, and the great majority of them occur in more than one Veda at the same time. But it is worthy of note that no less than twelve, besides occurring in at least one other Sâkhâ, are either actually found in the Samhitâ of the Kârâyanîya-kathas, the Kâthaka^[1] (or Karaka-sâkhâ?), or stated to belong to it in the Commentary, while one is found in the Kâthaka alone, a second in the Atharva-veda alone, a third in the Taittirîya Brâhmana alone, and a fourth does not occur in any Vedic work hitherto known^[2]. A far greater number of Mantras occurs in Chapters XXI, LXVII, LXXIII, LXXIV, LXXXVI, which treat of daily oblations, Srâddhas, and the ceremony of setting a bull at liberty. Of all these Mantras, which,--including the Purushasûkta and other such well-known Mantras as well as the short invocations addressed to Soma, Agni, and other deities, but excluding the invocations addressed to Vishnu in the spurious Sûtra, LXVII, 2,--are more than a hundred in number, no more than forty or so are found in Vedic

p. xii

works hitherto printed, and in the law-books of Manu, Yâgñavalkya, and others; but nearly all are quoted, exactly in the same order as in this work, in the Kârâyanîya-kâthaka Grihya-sûtra, while some of them have been traced in the Kâthaka as well. And what is even more important, the Kâthaka Grihya does not contain those Mantras alone, but nearly all the Sûtras in which they occur; and it may be stated therefore, secondly, that the Vishnu-sûtra has four long sections, viz. Chapter LXXIII, and Chapters XXI, LXVII, LXXXVI, excepting the final parts, in common with that work, while the substance of Chapter LXXIV may also be traced in it. The agreement between both works is very close, and where they differ it is generally due to false readings or to enlargements on the part of the Vishnu-sûtra. However, there are a few cases, in which the version of the latter work is evidently more genuine than that of the former, and it follows, therefore, that the author of the Vishnu-sûtra cannot have borrowed his rules for the performance of Srâddhas &c. from the Kâthaka Grihya-sûtra, but that both must have drawn from a common source, i. e. no doubt from the traditions current in the Katha school, to which this work is indebted for so many of its Mantras as well.

   For these reasons^[1] I fully concur in the view advanced by Dr. Bühler, that the bulk of the so-called Vishnu-smriti is really the ancient Dharma-sûtra of the Kârâyanîya-kâthaka Sâkhâ of the Black Yagur-veda. It ranks, like other Dharma-sûtras, with the Grihya and Srauta-sûtras of its school; the latter of which, though apparently lost now, is distinctly referred to in the Grihya-sûtra in several places, and must have been in existence at the time when the Commentaries on Kâtyâyana's Srauta-sûtras were composed, in which it is frequently quoted by the name

p. xiii

of Katha-sûtra on divers questions concerning Srauta offerings, and at the time, when the Kasmîrian Devapâla wrote his Commentary on the Kâthaka Grihya-sûtra, which was, according to the Kasmîrian tradition, as explored by Dr. Bühler, before the conquest of Kasmîr by the Mahommedans. Devapâla, in the Introduction to his work, refers to 'thirty-nine Adhyâyas treating of the Vaitânika (= Srauta) ceremonies,' by which the Grihya-sûtra was preceded, from which statement it may be inferred that the Kâthaka Srauta-sûtras must have been a very voluminous work indeed, as the Grihya-sûtra, which is at least equal if not superior in extent to other works of the same class, forms but one Adhyâya, the fortieth, of the whole Kalpa-sûtra, which, according to Devapâla, was composed by one author. It does not seem likely that the Vishnu-sûtra was composed by the same man, or that it ever formed part of the Kâthaka Kalpa-sûtra, as the Dharma-sûtras of Baudhâyana, Âpastamba, and Hiranyakesin form part of the Kalpa-sûtras of the respective schools to which they belong. If that were the case, it would agree with the Grihya-sûtra on all those points which are treated in both works, such as e. g. the terms for the performance of the Samskâras or sacraments, the rules for a student and for a Snâtaka, the enumeration and definition of the Krikkhras or 'hard penances,' the forms of marriage, &c. Now though the two works have on those subjects a number of such rules in common as occur in other works also, they disagree for the most part in the choice of expressions, and on a few points lay down exactly opposite rules, such as the Vishnu-sûtra (XXVIII, 28) giving permission to a student to ascend his spiritual teacher's carriage after him, whereas the other work prescribes, that he shall do so on no account. Moreover, if both works had been destined from the first to supplement one another, they would, instead of having several entire sections in common, exhibit such cross-references as are found e. g. between the Âpastamba Grihya and Dharma-sûtras[1]; though the absence of such

p. xiv

references might be explained, in the case of the Vishnu-sûtra, by the activity of those who brought it into its present shape, and who seem to have carefully removed all such references to other works as the original Dharmasûtra may have contained. Whatever the precise nature of the relations between this work and the other Sûtra works of the Kârâyanîya-kâthaka school may have been, there is no reason for assigning to it a later date than to the Kâthaka Srauta and Grihya-sûtras, with the latter of which it has so much in common, and it may therefore claim a considerable antiquity, especially if it is assumed, with Dr. Bühler, that the beginning of the Sûtra period differed for each Veda. The Veda of the Kathas, the Kâthaka, is not separated from the Sûtra literature of this school by an intermediate: Brâhmana stage; yet its high antiquity is testified by several of the most eminent grammarians of India from Yâska down to Kaiyata^[1]. Thus the Kâthaka is the only existing work of its kind, which is quoted by the former grammarian (Nirukta X, 5; another clear quotation from the Kâthaka, XXVII, 9, though not by name, may be found, Nirukta III, 4), and the latter places the Kathas at the head of all Vedic schools, while Patatañgali, the author of the Mahâbhâshya, assigns to the ancient sage Katha, the reputed founder of the Katha or Kâthaka school of the Black Yagur-veda, the dignified position of an immediate pupil of Vaisampâyana, the fountain-head of all schools of the older or Black Yagur-veda, and mentions, in accordance with a similar statement preserved in the Râmâyana (II, 32, 18, 19 ed. Schlegel), that in his own time the 'Kâlâpaka and the Kâthaka' were 'proclaimed in every village^[2].' The priority of the Kathas before all other existing schools of the Yagur-veda may be deduced from the statements of the Karanavyûha^[3], which work assigns to them one of the first places among the divers branches of

p. xv

the Karakas, whom it places at the head of all schools of the Yagur-veda. Another argument in favour of the high antiquity of the Kathas may be derived from their geographical position^[1]. Though the statements of the Mahâbhâshya and Râmâyana regarding the wide-spread and influential position of the Kathas in ancient times are borne out by the fact that the Karanavûtha mentions three subdivisions of the Kathas, viz. the Kathas proper, the Prâkya Kathas, and the Kapishthala Kathas, to which the Kârâyanîyas may be added as a fourth, and by the seeming identity of their name with the name of the {Greek Kaðaîoi} in the Pañgâb on the one hand, and with the first part of the name of the peninsula of Kattivar on the other hand, it seems very likely nevertheless that the original home of the Kathas was situated in the north-west, i. e. in those regions where the earliest parts of the Vedas were composed. Not only the {Greek Kaðaîoi}, but the {Greek Kambísðoloi} as well, who have been identified with the Kapishthala Kathas^[2], are mentioned by Greek writers as a nation living in the Pañgâb; and while the Prâkya Kathas are shown by their name ('Eastern Kathas') to have lived to the east of the two other branches of the Kathas, it is a significant fact that adherents of the Kârâyanîya-kâthaka school survive nowhere but in Kasmîr, where all Brâhmanas perform their domestic rites according to the rules laid down in the Grihya-sûtra of this school^[3]. Kasmîr is moreover the country where nearly all the yet existing works of the Kâthaka school have turned up, including the Berlin MS. of the Kâthaka, which was probably written by a Kasmîrian^[4]. It is true that some of the geographical and historical data contained in that work, especially the way in which it mentions the Pañkâlas, whose ancient name, as shown by the Satapatha Brâhmana (XIII, 5, 4, 7) and

p. xvi

Rig-veda (VIII, 20, 24; VIII, 22, 12), was Krivi, take us far off from the north-west, the earliest seat of Aryan civilization, into the country of the Kuru-Pañkâlas in Hindostân proper. But it must be borne in mind that the Kâthaka, if it may be identified with the 'Karaka-sâkhâ,' must have been the Veda of all the Karakas except perhaps the Maitrâyanîyas and Kapishthalas, and may have been altered and enlarged, after the Kathas and Karakas had spread themselves across Hindostân. The Sûtras of a Sâkhâ which appears to have sprung up near the primitive home of Âryan civilization in India, which was probably the original home of the Kathas at the same time, may be far older than those of mere Sûtra schools of the Black Yagur-veda, which have sprung up, like the Âpastamba school, in South India, i. e. far older than the fourth or fifth century B. C.^[1]

   But sufficient space has been assigned to these attempts at fixing the age of the Kâthaka-sûtras which, besides remaining only too uncertain in themselves, can apply with their full force to those parts of the Vishnu-sûtra only, which have been traced in the Kâthaka Grihya-sûtra. It will be seen afterwards that even these sections, however closely connected with the sacred literature of the Kathas, have been tampered with in several places, and it might be argued, therefore, that the whole remainder of the Vishnu-sûtra, to which the Kâthaka literature offers no parallel, may be a subsequent addition. But the antiquity of the great majority of its laws can be proved by independent arguments, which are furnished by a comparison of the Vishnu-sûtra with other works of the same class, whose antiquity is not doubted.

   In the foot-notes to my translation I have endeavoured to give as complete references as possible to the analogous passages in the Smritis of Manu, Yâgñavalkya, Âpastamba, and Gautama, and in the four Grihya-sûtras hitherto printed. A large number of analogous passages might have been traced in the Dharma-sûtras of Vâsishtha^[2]

p. xvii

and Baudhâyana as well, not to mention Hiranyakesin's Dharma-sûtra, which, according to Dr. Bühler, is nearly identical with the Dharma-sûtra of Âpastamba. Two facts may be established at once by glancing at these analogies, viz. the close agreement of this work with the other Sûtra works in point of form, and with all the above-mentioned works in point of contents. As regards the first point, the Sûtras or prose rules of which the bulk of the Vishnu-sûtra is composed, show throughout that characteristic laconism of the Sûtra style, which renders it impossible in many cases to make out the real meaning of a Sûtra without the help of a Commentary; and in the choice of terms they agree as closely as possible with the other ancient law-books, and in some cases with the Grihya-sûtras as well. Numerous verses, generally in the Sloka metre, and occasionally designed as 'Gâthâs,' are added at the end of most chapters, and interspersed between the Sûtras in some; but in this particular also the Vishnu-sûtra agrees with at least one other Dharma-sûtra, the Vâsishtha-smriti, and it contains in its law part, like the latter work, a number of verses in the ancient Trishtubh metre^[1]. Four of these Trishtubhs are found in the Vâsishtha-smriti, and three in Yâska's Nirukta as well, and the majority of the Slokas has been traced in the former work and the other above-mentioned law-books, and in other Smritis. In point of contents the great majority both of the metrical and prose rules of the Vishnu-sûtra agrees with one, or some, or all of the works named above. The Grihya-sûtras, excepting the Kâthaka Grihya-sûtra, naturally offer a far smaller number of analogies with it than the Smritis, still they exhibit several rules, in the Snâtaka-dharmas and otherwise, that have not been traced in any other Smriti except the work here translated. Among the Smritis again, each single one maybe seen

p. xviii

from the references to contain a number of such rules, as are only met with in this work, which is a very important fact because, if the laws of the Vishnu-sûtra were found either in all other Smritis, or in one of them only, its author might be suspected of having borrowed them from one of those works. As it is, meeting with analogous passages now in one work, and then in another, one cannot but suppose that the author of this work has everywhere drawn from the same source as the other Sûtrakâras, viz. from ancient traditions that were common to all Vedic schools.

   There are, moreover, a number of cases in which this work, instead of having borrowed from other works of the same class, can be shown to have been, directly or indirectly, the source from which they drew, and this fact constitutes a third reason in favour of the high antiquity of its laws. The clearest case of this kind is furnished by the Vâsishtha-smriti, with which this work has two entire chapters in common, which are not found elsewhere. I subjoin in a note the text of Vâsishtha XXVIII, 10-15, with an asterisk to those words which contain palpable mistakes (not including blunders in point of metre), for comparison with Chapter LVI of this work in the Calcutta edition, which is exceptionally correct in this chapter and in Chapter LXXXVII, which latter corresponds to Vâsishtha XXVIII, 18-22^[1]. In both

p. xix

chapters Vishnu has mainly prose Sûtras and throughout a perfectly correct text, whereas Vâsishtha has bad Slokas which, supported as they are by the Commentary or by the metre or by both, can only be accounted for by carelessness or clerical mistakes in some cases, and by a clumsy versification of the original prose version preserved in this work in others. Another chapter of the Vishnu-sûtra, the forty-eighth, nowhere meets with a parallel except in the third Prasna of the Dharma-sûtra of Baudhâyana, where it recurs almost word for word. An examination of the various readings in both works shows that in some of the Slokas Baudhâyana has better readings, while in one or two others the readings of Vishnu seem preferable, though the unsatisfactory condition of the MS. consulted renders it unsafe to pronounce a definitive judgment on the character of Baudhâyana's readings. At all events he has a few Vedic Mantras more than Vishnu, which however seem to be very well-known Mantras and are quoted by their Pratîkas only. But he omits the two important Sûtras 9 and 10 of Vishnu, the latter of which contains a Mantra quoted at full, which, although corrupted (see Vâgas. Samh. IV, 12) and hardly intelligible, is truly Vedic in point of language; and he adds on his part a clause at the end of the whole chapter[1], which inculcates the worship of Ganesa or Siva or both, and would be quite sufficient in itself to cast a doubt on the genuineness and originality of his version. It is far from improbable that both Vâsishtha and Baudhâyana may have borrowed

p. xx

the sections referred to directly from an old recension of this work, as Baudhâyana has borrowed another chapter of his work from Gautama, while Vâsishtha in his turn has borrowed the same chapter from Baudhâyana^[1]. It may be added in confirmation of this view, that as far as Vâsishtha is concerned, his work is the only Smriti, as far as I know, which contains a quotation from the 'Kâthaka'(in XXIX, 18). The Dharma-sûtras of Âpastamba and Gautama have nowhere a large number of consecutive Sûtras in common with the Vishnu-sûtra, but it is curious to note that the rule, which the latter (X, 45) quotes as the opinion of 'some' (eke), that a non-Brahmanical finder of a treasure, who announces his find to the king, shall obtain one-sixth of the value, is found in no other law-book except in this, which states (III, 61) that a Sûdra shall 'divide a treasure-trove into twelve parts, two of which he may keep for himself. Of the metrical law-books, one, the Yâgñavalkya-smriti, has been shown by Professor Max Müller^[2] to have borrowed the whole anatomical section (III, 84-104 including the simile of the soul which dwells in the heart like a lamp (III, 109, III, 201), from this work (XCVI, 43-96; XCVII, 9); and it has been pointed out by the same scholar, that the verse in which the author of the former work speaks of the Âranyaka and of the Yoga-sâstra as of his own works (III, 110) does not occur in the Vishnu-sûtra, and must have been added by the versificator, who brought the Yâgñavalkya-smriti into its present metrical form. Several other Slokas in Yâgñavalkya's description of the human body (111, 99, 105-108), and nearly the whole section on Yoga (Y. III, 111-203, excepting those Slokas, the substance of which is found in this work and in the code of Manu, viz. 131-140, 177-182, 190, 198-201) may be traced to the same source, as may be also the omission of Vishnu's enumeration of the 'six limbs' (XCVI, 90) in the Yâgñavalkya-smriti, and probably all the minor points on which it differs from this work. Generally speaking, those

p. xxi

passages which have been justly noticed as marking the comparatively late period in which that law-book must have been composed^[1]: such as the allusions to the astrology and astronomy of the Greeks (Y. I, 80, 295), which render it necessary to refer the metrical redaction of the Yâgñavalkya-smriti to a later time than the second century A. D.; the whole passage on the worship of Ganesa and of the planets (I, 270-307), in which, moreover, a heterodox sect is mentioned, that has been identified with the Buddhists; the philosophical doctrines propounded in I, 349, 350; the injunctions regarding the foundation and endowment of monasteries (II, 185 seq.)--all these passages have no parallel in this work, while it is not overstating the case to say that nearly all the other subjects mentioned in the Yâgñavalkya-smriti are treated in a similar way, and very often in the same terms, in the Vishnu-sûtra as well. Some of those rules, in which the posteriority of the Yâgñavalkya-smriti to other law-books exhibits itself, do occur in the Vishnu-sûtra, but without the same marks of modern age. Thus the former has two Slokas concerning the punishment of forgery (II, 240, 241), in which coined money is referred to by the term nânaka; the Vishnu-sûtra has the identical rule (V, 122, 123; cf. V, 9). but the word nânaka does not occur in it. Yâgñavalkya, in speaking of the number of wives which a member of the three higher castes may marry (I, 57), advocates the Puritan view, that no Sûdra wife must be among these; this work has analogous rules (XXIV, 1-4), in which, however, such marriages are expressly allowed. The comparative priority of all those Sûtras of Vishnu, to which similar Slokas of Yâgñavalkya correspond, appears probable on general grounds, which are furnished by the course of development in this as in other branches of Indian literature; and to this it may be added,

p. xxii

as far as the civil and criminal laws are concerned, that the former enumerates them quite promiscuously, just like the other Dharma-sûtras, with which he agrees besides in separating the law of inheritance from the body of the laws, whereas Yâgñavalkya enumerates all the laws in the order of the eighteen 'titles of law' of Manu and the more recent law-books, though he does not mention the titles of law by name.

   However much the Vishnu-sûtra may have in common with the Yâgñavalkya-smriti, there is no other law-book with which it agrees so closely as with the code of Manu. This fact may be established by a mere glance at the references in the foot-notes to this translation, in which Manu makes his appearance far more frequently and constantly than any other author, and the case becomes the stronger, the more the nature of these analogies is inquired into. Of Slokas alone Vishnu has upwards of 160 in common with Manu, and in a far greater number of cases still his Sûtras agree nearly word for word with the corresponding rules of Manu. The latter also, though he concurs in a very great number of points with the other law authors as well, agrees with none of them so thoroughly as with Vishnu. All the Smritis of Âpastamba, Baudhâyana, Vâsishtha, Yâgñavalkya, and Nârada contain, according to an approximate calculation, no more than about 130 Slokas, that are found in the code of Manu as well. The latter author and Vishnu differ of course on a great many minor points, and an exhaustive discussion of this subject would fill a treatise; I must therefore confine myself to notice some of those differences, which are particularly important for deciding the relative priority of the one work before the other. In a number of Slokas Manu's readings are decidedly older and better than Vishnu's. Thus the latter (XXX, 7) compares the three 'Atigurus' to the 'three gods,' i.e. to the post-Vedic Trimûrti of 'Brahman, Vishnu, and Siva,' as the commentator expressly states, whereas Manu in an analogous Sloka (II, 230) refers to the 'three orders' instead. At the end of the section on inheritance (XVIII, 44) Vishnu mentions among other

p. xxiii

indivisible objects 'a book,' pustakam; Manu (IX, 219) has the same Sloka, but for pustakam he reads prakakshate. Now pustaka is a modern word^[1], and Varâhamihira, who lived in the sixth century A. D., appears to be the first author, with a known date, by whom it is used. It occurs again, Vishnu-sûtra XXIII, 56 (prokshanena ka pustakam), and here also Manu (V, 122) has a different reading (punahpâkena mrinmayam). The only difference between Vishnu-sûtra XXII, 93 and Manu V, 110 consists in the use of singular forms (te, srinu) in the former work, and of plural forms (vah, srinuta) in the latter. Now there are a great many other Smritis besides the Manu-smriti, such as e. g. the Yâgñavalkya and Parâsara Smritis, in which the fiction is kept up, that the laws contained in them are promulgated to an assembly of Rishis; but there are very few Smritis of the least notoriety or importance besides the Vishnu-sûtra, in which they are proclaimed to a single person. Other instances in which Manu's readings appear preferable to Vishnu's may be found, LI, 60 (pretya keha ka nishkritim) = Manu V, 38 (pretya ganmani ganmani); LI, 64 (iti kathañkana) = M. V, 41 (ity abravînmanuh); LI, 76 (tasya) = M.V, 53 (tayoh); LIV, 27 (brâhmanyât) = M. XI, 193 (brahmanâ); LVII, 11 (purastâd anukoditâm) = M. IV, 248; Vâsishtha XIV, 16; Âpastamba I, 6, 19, 14 (purastâd aprakoditâm); LXVII, 45 (sâyamprâtas tvatithaye) = M. III, 99 (samprâptâya tvatithaye), &c. But these instances do not prove much, as all the passages in question may have been tampered with by the Vishnuitic editor, and as in sonic other cases the version of Vishnu seems preferable. Thus 'practised by the virtuous' (sâdhubhiska nishevitam, LXXI, 90) is a very common epithet of 'âkâra,' and reads better than Manu's nibaddham sveshu karmasu (IV, 155); and krikkhrâtikrikkhram (LIV, 30) seems preferable to Baudhâyana's and Manu's krikkhrâtikrikkhrau (XI, 209). What is more important, the Vishnu-sûtra does not only contain a number of verses in the ancient Trishtubh metre, whereas Manu has none, but it shows those identical three Trishtubhs of Vâsishtha and Yâska, which Dr. Bühler

p. xxiv

has proved to have been converted into Anushtubh Slokas by Manu (II, 114, 115, 144)^[1]; and Manu seems to have taken the substance of his three Slokas from this work more immediately, because both he (II, 144) and Vishnu, (XXX, 47) have the reading âvrinoti for âtrinatti, which truly Vedic form is employed both by Vâsishtha and Yâska. The relative antiquity of Vishnu's prose rules, as compared to the numerous corresponding Slokas of Manu, may be proved by arguments precisely similar to those which I have adduced above in speaking of the Yâgñavalkya-smriti. As regards those points in the code of Manu, which are usually considered as marks of the comparatively late date of its composition, it will suffice to mention, that the Vishnu-sûtra nowhere refers to South Indian nations such as the Dravidas and Andhras, or to the Yavanas; that it shows no distinct traces of an acquaintance with the tenets of any other school of philosophy except the Yoga and Sânkhya systems; that it does not mention female ascetics disparagingly, and in particular does not contain Manu's rule (VIII, 363) regarding the comparatively light punishment to be inflicted for violation of (Buddhist and other) female ascetics; and that it does not inveigh (see XV, 3), like Manu (IX, 64-68), against the custom of Niyoga or appointment of a widow to raise offspring to her deceased husband. It is true, on the other hand, that in many cases Vishnu's rules have a less archaic character than the corresponding precepts of Manu, not only in the Slokas, but in the Sûtra part as well. Thus written documents and ordeals are barely mentioned in the code (if Manu (VIII, 114, 115, 168; IX, 232); Vishnu on the other hand, besides referring in divers places to royal grants and edicts, to written receipts and other private documents, and to books, devotes to writings (lekhya) an entire chapter, in which he makes mention of the caste of Kâyasthas, 'scribes,' and he lays down elaborate rules for the performance of five species of ordeals, to which recourse should be had, according to him, in all suits of some importance. But in nearly all such cases the antiquity of Vishnu's

p. xxv

rules is warranted to a certain extent by corresponding rules occurring in the Smritis of Yâgñavalkya and Nârada; and the evidence for the modifications and entire transformations, which the code of Manu must have undergone in a number of successive periods, is so abundant, that the archaic character of many of its rules cannot be considered to constitute a sufficient proof of the priority of the whole code before other codes which contain some rules of a comparatively modern character. To this it must be added that the Nârada-smriti, though taken as a whole it is decidedly posterior to the code of Manu^[1], is designated by tradition as an epitome from another and more bulky recension of the code of Manu than the one which we now possess; and if this statement may be credited, which is indeed rather doubtful, the very particular resemblance between both works in the law of evidence and in the rules regarding property (see LVIII) can only tend to corroborate the assumption that the Vishnu-sûtra and the Manu-smriti must have been closely connected from the first.

   This view is capable of further confirmation still by a different set of arguments. The so-called code of Manu is universally assumed now to be an improved metrical edition of the ancient Dharma-sûtra of the (Maitrâyaniya-) Mânavas, a school studying the Black Yagur-veda; and it has been shown above that the ancient stock of the Vishnu-sûtra, in which all the parts hitherto discussed may be included, represents in the main the Dharma-sûtra, of the Kârâyanîya-kathas, another school studying the Black Yagur-veda. Now these two schools do not only belong both to that Veda, but to the same branch of it, as may be seen from the Kârânavyûha, which work classes both the Kathas and Kârâyanîyas on the one hand, and the Mânavas

p. xxvi

together with the six or five other sections of the Maitrâyanîyas on the other hand, as subdivisions of the Karaka Sakhâ of the Black Yagur-veda. What is more, there exists a thorough-going parallelism between the literature of those two schools, as far as it is known. To begin with their respective Samhitâs, it has been shown by L. Schröder^[1] that the Maitrâyanî Samhitâ has more in common with the Kâthaka, the Samhitâ of the Kathas, than with any other Veda. As the Kathas are constantly named, in the Mahâbhâshya and other old works, by the side of the Kâlâpas, whereas the name of the Maitrâyanîyas does not occur in any Sanskrit work of uncontested antiquity, it has been suggested by the same scholar that the Maitrâyanîyas may be the Kâlâpas of old, and may not have assumed the former name till Buddhism began to prevail in India. However this may be, the principal Sûtra works of both schools stand in a similar relation to one another as their Samhitâs. Some of those Mantras, which have been stated above to be common to the Vishnu-sûtra and Kâthaka Grihya only, and to occur in no other Vedic work hitherto printed, have been traced in the Mânava Srauta-sûtra, in the chapter on Pinda-pitriyagña (I, 2 of the section on Prâksoma)^[2], and the conclusion is, that if the Srauta-sûtra of the Kâthaka school were still in existence, it would be found to exhibit a far greater number of analogies with the Srauta-sûtra of the Mânavas. The Grihya-sûtra of this school^[3] agrees with the Kâthaka Grihya-sûtra even more closely than the latter agrees with the Vishnu-sûtra, as both works have not only several entire chapters in common (the chapter on the Vaisvadeva sacrifice among others, which is found in the Vishnu-sûtra also), but concur everywhere in the arrangement of the subject-matter and in the choice of expressions and Mantras. The Brâhmana stage of Vedic literature is not represented by a separate work in either of the two schools, but a further argument in

p. xxvii

favour of their alleged historical connection may be derived from their respective geographical position. If it has been rightly conjectured above, that the original seats of the Kathas were in the north-west, whence they spread themselves over Hindostân, the Maitrâyanîyas, though now surviving nowhere except in some villages 'near the Sâtpuda mountain, which is included in the Vindhyas^[1].' must have been anciently their neighbours, as the territory occupied by them extended 'from the Mayûra mountain into Gugarât,' and reached 'as far as the north-western country' (vâyavyadesa)^[2]. Considering all this evidence regarding the original connection between the Kathas and Mânavas, it may be said without exaggeration, that it would be far more surprising to find no traces of resemblance between their respective Dharma-sûtras, such as we possess them, than to find, as is actually the case, the contrary; and it may be argued, vice versâ, that the supposed connection of the two works with the Vedic schools of the Kathas and Mânavas^[3], respectively, is confirmed by the kinship existing between these two schools.

   In turning now from the ancient parts of the Vishnu-sûtra to its more recent ingredients, I may again begin by quoting Professor Max Müller's remarks on this work, which contain the statement, that it is 'enlarged by modern additions written in Slokas^[4].' After him, Dr. Bühler pointed out^[5] that the whole work appears to have been recast by an adherent of Vishnu, and that the final and introductory chapters in particular are shown by their very style to have been composed by another author than the body of the

p. xxviii

work. If the latter remark were in need of further confirmation, it might be urged that the description of Vishnu as 'the boar of the sacrifice' (yagñavarâha) in the first chapter is bodily taken from the Harivamsa (2226-2237), while most of the epithets given to Vishnu in I, 49-61 and XCVIII, 7-100 may be found in another section of the Mahâbhârata, the so-called Vishnu-sahasranâma. Along with the introductory and final chapters, all those passages generally are distinctly traceable to the activity of the Vishnuitic editor, in which Vishnu (Purusha, Bhagavat, Vâsudeva, &c.) is mentioned, or his dialogue with the goddess of the earth carried on, viz. I; V, 193; XIX, 24; XX, 16-21; XXII, 93; XXIII, 46; XXIV, 35; XLVII, 10; XLIX; LXIV, 28, 29; LXV; LXVI; LXVII, 2; XC, 3-5,17-23; XCVI, 97,98; XCVII, 7-21; XCVIII-C. The short invocation addressed to Vishnu in LXVII, 12 is proved to be ancient by its recurrence in the corresponding chapter of the Kâthaka Grihya-sûtra, and Chapter LXV contains genuine Kâthaka Mantras transferred to a Vishnuitic ceremony. Chapter LXVI, on the other hand, though it does not refer to Vishnu by name, seems to be connected with the same Vishnuitic rite, and becomes further suspected by the recurrence of several of its rules in the genuine Chapter LXXIX. The contents of Chapter XCVII, in which it is attempted to reconcile some of the main tenets of the Sânkhya system, as propounded in the Sânkhya-kârikâ, Sânkhya-pravakanabhâshya, and other works, with the Vaishnava creed and with the Yoga; the fact that the two Slokas in XCVI (97, 98) and part of the Slokas in XCVII (15-21) have their parallel in similar Slokas of the Bhagavad-gîtâ and of the Bhâgavata-purâna; the terms Mahatpati, Kapila, and Sânkhyâkârya, used as epithets of Vishnu (XCVIII, 26, 85, 86); and some other passages in the Vishnuitic chapters seem to favour the supposition that the editor may have been one of those members of the Vishnuitic sect of the Bhâgavatas, who were conspicuous for their leaning towards the Sânkhya and Yoga systems of philosophy. The arrangement of the Vishnu-sûtra in a hundred chapters is no doubt due to the same person, as the Commentary points out that the number

p. xxix

of the epithets given to Vishnu in XCVIII is precisely equal to the number of chapters into which the laws promulgated by him are divided (II-XCVII); though the number ninety-six is received only by including the introductory and final invocations (XCVIII, 6, 101) among the epithets of Vishnu. It seems quite possible, that some chapters were inserted mainly in order to bring up the whole figure to the round number of a hundred chapters, and it is for this reason chiefly that the majority of the following additions, which show no Vishnuitic tendencies, may also be attributed to the Vishnuitic editor.

   1. Most or all of the Slokas added at the end of Chapters XX (22-53) and XLIII (32-45) cannot be genuine; the former on account of their great extent and partial recurrence in the Bhagavad-gîtâ^[1], Mahâbhârata, and other works of general note, and because they refer to the self-immolation of widows and to Kâla, whom the commentator is probably right in identifying with Vishnu; the latter on account of their rather extravagant character and decidedly Purânic style, though the Gâruda-purâna, in its very long description of the hells, offers no strict parallel to the details given here. The verses in which the Brâhmanas and cows are celebrated (XIX, 22, 23; XXIII, 57-61) are also rather extravagant; however, some of them are Trishtubhs, and the verses in XIX are closely connected with the preceding Sûtras. The two final Slokas in LXXXVI (19, 20) may also be suspected as to their genuineness, because they are wanting in the corresponding chapter of the Kâthaka Grihya-sûtra; and a number of other verses in divers places, because they have no parallel in the Smriti literature, or because they have been traced in comparatively modern works, such as the Bhagavad-gîtâ, the Pañkatantra, &c. 2. The week of the later Romans and Greeks, and of modern Europe (LXXVIII, 1-7), the self-immolation of widows (XXV, 14; cf. XX, 39), and the Buddhists and Pâsupatas (LXIII, 36) are not mentioned in any ancient Sanskrit work. Besides, the passages in question may be easily removed, especially the Sûtras referring to the seven days of the week, which

p. xxx

form clearly a subsequent addition to the enumeration of the Nakshatras and Tithis immediately following (LXXVIII, 8-50), and the rule concerning the burning of widows (XXV, 14), which is in direct opposition to the law concerning the widow's right to inherit (XVII, 4) and to other precepts regarding widows. That the three terms kâshâyin, pravragita, malina in LXIII, 36 refer to members of religious orders seems clear, but it maybe doubted whether malina denotes the Pâsupatas, and even whether kâshâyin (cf. pravragita XXXVI, 7) denotes the Buddhists, as dresses dyed with Kashâya are worn by Brahmanical sects also, and prescribed for students, and for ascetics likewise, by some of the Grihya- and Dharma-sûtras. Still the antiquity of the Sûtra in question can hardly be defended, because the acquaintance of the Vishnuitic editor with the Buddhistic system of faith is proved by two other Sûtras (XCVIII, 40, 41), and because the whole subject of good and evil omens is not treated in any other ancient Smriti. On the other hand, such terms as vedanindâ and nâstikatâ (XXXVII, 4, 31, &c.) recur in most Smritis, and can hardly be referred to the Buddhists in particular. 3. The Tîrthas enumerated in LXXXV, some of which are sacred to Vishnu and Siva, belong to all parts of India, and many of them are situated in the Dekhan, which was certainly not included within the limits of the 'Âryâvarta' of the ancient Dharma-sûtra (LXXXIV, 4). As no other Smriti contains a list of this kind, the whole chapter may be viewed as a later addition. 4. The ceremonies described in XC are not mentioned in other Smritis, while some of them are decidedly Vishnuitic, or traceable in modern works; and as all the Sûtras in XC hang closely together, this entire chapter seems also to be spurious. 5. The repetitions in the list of articles forbidden to sell (LIV, 18-22); the addition of the two categories of atipâtakâni, 'crimes in the highest degree,' and prakîrntakam, 'miscellaneous crimes' (XXXIII, 3,5; XXXIV; XLII), to Manu's list of crimes; the frequent references to the Ganges river; and other such passages, which show a modem character, without being traceable in the Smritis of Yâgñavalkya and Nârada, may have been added by the Vishnuitic

p. xxxi

editor from modern Smritis, either for the sake of completeness, or in order to make up the required number of chapters. 6. All the passages hitherto mentioned are such as have no parallel in other ancient Smritis. But the Vishnuitic editor did evidently not confine himself to the introduction of new matter into the ancient Dharma-sûtra. That he did not refrain, occasionally, from altering the original text, has been conjectured above with regard to his readings of some of those Slokas, which are found in the code of Manu as well; and it can be proved quite clearly by comparing his version of the Vrishotsarga ceremony (LXXXVI) with the analogous chapter of the Kâthaka Grihya-sûtra. In one case (LI, 64; cf. XXIII, 50 = M. V, 131) he has replaced the words, which refer the authorship of the Sloka in question to Manu, by an unmeaning term. The superior antiquity of Manu's reading (V, 41) is vouched for by the recurrence of the same passage in the Grihya-sûtra of Sânkhâyana (II, 16, 1) and in the Vâsishtha-smriti (IV, 6), and the reference to Manu has no doubt been removed by the Vishnuitic editor, because it would have been out of place in a speech of Vishnu. References to sayings of Manu and other teachers and direct quotations from Vedic works are more or less common in all Dharma-sûtras, and their entire absence in this work is apparently due to their systematical removal by the editor. On the other hand, the lists of Vedic and other works to be studied or recited may have been enlarged in one or two cases by him or by another interpolator, namely, XXX, 37 (cf. V, 191), where the Atharva-veda is mentioned after the other Vedas by the name of 'Âtharvana' (not Atharvângirasas, as in the code of Manu and most other ancient works), and LXXXIII, 7, where Vyâkarana, 'Grammar,' i. e. according to the Commentary the grammars of Pânini and others, is mentioned as distinct from the Vedângas. The antiquity of the former passage might indeed be defended by the example of Âpastamba, who, though referring like this work to the 'three Vedas' both separately and collectively, mentions in another place the 'Âtharvana-veda^[1].' Besides the above works,

p. xxii

and those referred to in LVI, the laws of Vishnu name no other work except the Purânas, Itihâsas, and Dharmasâstras. 7. As the Vishnuitic editor did not scruple to alter the import of a certain number of passages, the modernisation of the language of the whole work, which was probably as rich in archaic forms and curious old terms as the Kâthaka Grihya-sûtra and as the Dharma-sûtra of Âpastamba, may be likewise attributed to him. As it is, the Vishnu-sûtra agrees in style and expressions more closely with the Smritis of Manu and Yâgñavalkya than with any other work, and it is at least not inferior to the former work in the preservation of archaic forms. Thus the code of Manu has seven aorist forms^[1], while the Vishnu-sûtra contains six, not including those occurring in Vedic Mantras which are quoted by their Pratîkas only. Of new words and meanings of words the Vishnu-sûtra contains also a certain number; they have lately been communicated by me to Dr. von Böhtlingk for, insertion in his new Dictionary.

   All the points noticed render it necessary to assign a comparatively recent date to the Vishnuitic editor; and if the introduction of the week of the Greeks into the ancient Dharma-sûtra has been justly attributed to him, he cannot be placed earlier than the third or fourth century A. D.^[2] The lower limit must be put before the eleventh century, in which the Vishnu-sûtra is quoted in the Mitâksharâ of Vigñânesvara, From that time downwards it is quoted in nearly every law digest, and a particularly large number of quotations occurs in Aparârka's Commentary, on Yâgñavalkya, which was composed in the twelfth century^[3]. Nearly all those quotations, as far as they have been examined, are actually found in the Vishnu-sûtra; but the whole text is vouched for only by Nandapandita's Commentary, called Vaigayantî, which was composed in the

p. xxiii

first quarter of the seventeenth century. The subscriptions in the London MSS. of the Vaigayantî contain the statement, which is borne out by the Introduction, that it was composed by Nandapandita, the son of Râmapandita, Dharmâdhikârin, an inhabitant of Benares, at the instigation of the Mahârâga Kesavanâyaka, also called Tammasânâyaka, the son of Kodapanâyaka; and a passage added at the end of the work states, more accurately, that 'Nandasarman (Nandapandita) wrote it at Kâsî (Benares) in the year 1679 of the era of Vikrâmabhâsvara (= A. D. 1622), by Command of Kesavanâyaka, his own king. These statements regarding the time and place of the composition of the Vaigayantî are corroborated by the fact that it refers in several cases to the opinions of Haradatta, who appears to have lived in the sixteenth century^[1], while Nandapandita is not among the numerous authors quoted in the Vîramitrodaya of Mitramisra, who lived in the beginning of the seventeenth century^[2], and who was consequently a contemporary of Nandapandita, if the above statement is correct; and that he attacks in a number of cases the views of the 'Eastern Commentators' (Prâkyas), and quotes a term from the dialect of Madhyadesa.

   The subjoined translation is based upon the text handed down by Nandapandita nearly everywhere except in some of the Mantras, which have been rendered according to the better readings preserved in the Kâthaka Grihya-sûtra. The two Calcutta editions of the Vishnu-sûtra, the second of which is a mere reprint of the first, will be found to agree in the main with the text here translated. They are doubtless based upon the Vaigayantî, as they contain several passages in which portions of Nandapandita's Commentary have crept into the text of the Sûtras. But the MS. used for the first Calcutta edition must have been a very faulty one, as both Calcutta editions, besides differing from the best MSS. of the Vaigayantî on a very great number of minor points, entirely omit the greater part of Chapter LXXXI

p. xxxiv

(3-22), the genuineness of which is proved by analogous passages in the other Smritis^[1]. An excellent copy of the Vaigayantî in possession of Dr. Bühler has, together with three London MSS. of that work and one London MS. containing the text only, enabled me to establish quite positively nearly in every case the readings sanctioned by Nandapandita. I had hoped to publish a new edition of the text prepared from those MSS., and long ready for the press, before publishing my English version. This expectation has not been fulfilled, but it is hoped that in the mean time this attempt at a translation will be welcome to the students of Indian antiquity, and will facilitate the understanding of the text printed in Givânanda Vidyâsâgara's cheap edition, which is probably in the hands of most Sanskrit scholars. The precise nature of the relation in which the text of my forthcoming edition stands to the Calcutta editions may be gathered from the large specimens of the text as given in the best MSS., that have been edited by Dr. Bühler in the Bombay Digest, and by myself in two papers published in the Transactions of the Royal Bavarian Academy of Science.

   Nandapandita has composed, besides the Vaigayantî, a treatise on the law of adoption, called Dattaka-mîmâmsâ[2], a commentary on the code of Parâsara, a work called Vidvanmanoharâ-smritisindhu, one called Srâddhakalpa-latâ, and commentaries on the Mitâksharâ and on Adityâkârya's Âsaukanirnaya. All these works belong to the province of Hindu law, and both his fertility as a writer in that branch of Indian science, and the reputation enjoyed by some of his works even nowadays, must raise a strong presumption in favour of his knowledge of the subject. The

p. xxxv

general trustworthiness of his Commentary on the Vishnu-sûtra is further confirmed by the frequent references which it contains to the opinions of earlier commentators of that work; and the wide extent of his reading, though he often makes an unnecessary display of it, has been eminently serviceable to him in tracing the connection of certain chapters and Mantras with the Kâthaka literature[1]. On the other hand, his very learning, combined with a strict adherence to the well-known theory of Hindu commentators regarding the absolute identity between the teaching of all Smritis, has frequently misled him into a too extensive method of interpretation. Even in commenting the Slokas he assigns in many cases an important hidden meaning to such particles as ka, vâ, tathâ, and others, and to unpretending epithets and the like, which have clearly been added for metrical reasons only^[2]. This practice, besides being contrary to common sense, is nowhere countenanced by the authority of Kullûka, in his remarks on the numerous identical Slokas found in. the code of Manu. With the Sûtras generally speaking the case is different: many of them would be nearly or quite unintelligible without the explanatory remarks added, in brackets from Nandapandita's Commentary^[3], and in a number of those cases even, where his method jars upon a European mind, the clauses supplied by him are probably correct[4]. The same may be said of his interpretations of the epithets of Vishnu, excepting those which are based on utterly fanciful etymologies[5],

p. xxxvi

as the style of the introductory and final chapters is as artificial, though in another way, as the Sûtra style. Though, however, in works composed in the latter style, every ka, vâ, or iti, &c., which is not absolutely required by the sense, was probably intended by their authors to convey a special meaning^[1], it is a question of evidence in every single case, whether those meanings which Nandapandita assigns to these and other such particles and expletive words are the correct ones. In several cases of this or of a similar kind he is palpably wrong^[2], and in many others the interpretations proposed by him are at least improbable, because the authoritative passages he quotes in support of them are taken from modern works, which cannot have been known to the author of the Vishnu-sûtra. Interpretations of this class have, therefore, been given in the notes only; and they have been omitted altogether in a number of cases where they appeared quite frivolous, or became too numerous, or could not be deciphered completely, owing to clerical mistakes in the MSS. But though it is impossible to agree with some of his general principles of interpretation, or with his application of them, Nandapandita's interpretations of difficult terms and Sûtras are invaluable, and I have never deviated from them in my translation without strong reasons to the contrary, which have in most cases been stated in the notes^[3]. Besides the extracts given in the notes, a few other passages from the Commentary and several other additions will be given in p. 312; and I must apologize to my readers for having to note along with the Addenda a number of Corrigenda, which will be found in the same page. In compiling the Index of Sanskrit words occurring in this work, which it has been thought necessary to add to the General Index, I have not aimed at completeness except as regards

p. xxxvii

the names of deities and of penances. My forthcoming edition of the Sanskrit text will be accompanied by a full Index of words.

   In conclusion I have to express my thanks in the most cordial manner to Dr. Bühler, who has constantly assisted me with his advice in the preparing of this translation, and has kindly lent me his excellent copy of the Vaigayantî; and to Dr. von Böhtlingk and Professor Max Müller, who have favoured me with valuable hints on divers points connected with this work. My acknowledgments are due, in the second place, to K. M, Chatfield, Esq., Director of Public Instruction, Bombay, to Dr. von Halm, Chief Librarian of the Royal Library, Munich, to Professor R. Lepsius, Chief Librarian of the Royal Library, of Berlin, and to Dr. R. Rost, Chief Librarian of the India office Library, London, for the valuable aid received from these gentlemen and the great liberality, with which they have placed Sanskrit MSS. under their care at my disposal.

p. 1