1

a certain importance only as auxiliaries to the chief argument derived from the imperfect development of the method or formal treatment of the law. But considering all that has been said in the preceding discussion, it is, I think, not too much to say that there is no obstacle against, and some reason for, our accepting as true the assertion, which is made in the Manu-smriti itself and supported by the tradition preserved in the Skanda-purâza, that Bhrigu's Samhitâ is the first and most ancient recast of a Dharmasâstra attributed to Manu, which latter, owing to the facts pointed out in the first part of this Introduction, must be identified with the Mânava Dharma-sûtra. Though this recast must be considered the work of one hand, the possibility that single verses may have been added later or altered, is of course not excluded. A perfectly intact preservation of an Indian work which has been much studied, is a priori improbable, and the divergence of the commentators with respect to certain verses shows that some of those contained in our text were suspected by the one or the other of them. But the number of Slokas with regard to which real doubts can be entertained is comparatively small, and hardly amounts to more than a dozen 2.

The above discussion has also to a certain extent defined the relative position of our Manu-smriti in Brâhmanical literature, and has thus opened the way for the consideration of the last remaining problem, the question when the conversion of the Mânava Dharma-sûtra into a metrical law-book

1 A clear and definite explanation of the question why the Hindu tradition ascribes the promulgation of Manu's laws to Bhrigu has hitherto not been traced. Bhrigu's only connexion with Manu is that mentioned in the text, according to which he is one of the mind-born sons or creatures of the father of mankind. This version of the legend of his origin is, however, by no means common. In the Mahâbhârata XII, 182-192, we find a condensed Dharmasâstra,' which is said to have been revealed by Bhrigu to Bharadvâga. It includes an account of the creation, but makes no mention of Manu. As Bhrigu appears also elsewhere as the author of a Dharmasâstra, it is just possible that the legend may be based on Bhrigu's fame as a legislator and as the offspring of Manu.

2 Many more verses are left out partly in Medhâtithi's Bhashya and partly in Nandana's commentary. But see below, pp. cxxvi and cxxxv, where it has been shown that omissions in the accessible MSS. of these two works alone do not mean much.

may have taken place. The terminus a quo which has been gained for the composition of Bhrigu's Samhitâ is the age of the Mahâbhârata, and the terminus ad quem the dates of the metrical Smritis of Yâgйavalkya and Nârada. Though we are at present not in a position to assert anything positive regarding the period when the Mahâbhârata and especially its twelfth and thirteenth Parvans were written, and though the date of Yâgñavalkya's Dharmasâstra is very doubtful, yet some facts known regarding the Nârada-smriti are not without importance for framing our answer to the difficult question now proposed. Both Professor Jolly and myself1 have lately discussed the significance of the mention of golden dînâras or denarii in the longer and more authentic version of Nârada and of the circumstance that Asahâya, a predecessor of Manu's earliest commentator, Medhâtithi, explained it and have arrived at a very similar conclusion, viz. that the Nârada-smriti dates either shortly before or shortly after the middle of the first thousand years of our era. If that is so, Bhrigu's Samhitâ must, in consideration of the arguments just stated, be placed not only earlier, but considerably earlier, and the assertion that it must have existed at least in the second century of our era is not unwarranted. This latter inference is also made inevitable by the discovery that we have to admit the former existence of very ancient commentaries, and of at least one ancient Vârttika or Kârikâ which referred to the text of Manu, known to us. With respect to the commentaries, Medhâtithi, the author of the Manubhâshya, is a most valuable and clear witness. This author, who probably wrote in the ninth century A. D.2, very frequently quotes opinions and various readings, expressed or men

Jolly, Tagore Lectures, p. 56; West and Bühler, Digest, p. 48. To the arguments adduced there I would add that Bâza, the friend of SriharshaHarshavardhana (606-7–648 A.D.), makes a pretty clear allusion to the Nâradîya Dharmasâstra in the Kâdambarî, p. 91, 1. 13 (Peterson's edition), where he calls a royal palace nâradîyam ivâvarṇyamânarâgadharmam, similar to the Nâradîya (Dharmasâstra), because there the duties of kings were taught (by the conduct of the ruler) just as they are taught (in the law-book).'

"For the details, see below, pp. cxxi-cxxiii.

tioned by his predecessors, and shows by the number of the conflicting explanations which he sometimes adduces for a passage of the text, that in his time a very large number of commentaries on the Manu-smriti existed. Among the persons thus quoted, he designates some by the terms Pûrva and Kiramtana. Pûrva, which means both 'former' and 'ancient,' is an ambiguous word. It can be applied to all persons who wrote before the author, though it frequently is used in speaking of those who lived centuries ago. Kiramtana, 'long previous or ancient,' is much stronger, and, according to the usage of Indian authors, denotes a predecessor belonging to a remote antiquity. As Medhâtithi, writing in the ninth century, knew of commentaries. to which he was compelled to assign a remote antiquity, it is only a moderate estimate if we assume that the earliest among them were in his time from three to four hundred years old. But if in the sixth or even in the fifth century A. D. glosses on our text existed, its composition must go back to much earlier times. For the widely divergent and frequently very questionable explanations of the more difficult passages, which Medhâtithi adduces from his predecessors, indicate that even the earliest among them were separated by a considerable interval from the compilator of the Manu-samhitâ, an interval so great that the real meaning of the text had been forgotten.

The merit of the discovery that one of the lost metrical Dharmasâstras, the Brihaspati-smriti, was a Vârttika on our text of Manu, belongs to Professor Jolly, whose careful investigation of the fragments of the lost law-books, contained in the modern Digests, has contributed very materially to the elucidation of a difficult chapter in the history of Indian legal literature. He shows that Brihaspati not only allots to Manu's Smriti the first place among all lawbooks, but that he explains, amplifies, and occasionally corrects its rules on various portions of the Vyavahâra. The particulars from Manu which Brihaspati mentions are such as to leave no doubt that the text which he knew in

1

1 Tagore Lectures, pp. 60-62; see also above, p. xvi.

no way differed from that known to us. He explains, as Professor Jolly points out, the curious terms, used Manu VIII, 49, for the various modes by which a creditor may recover a debt, as well as the expression asvâmin, which occurs in the title of law, called Asvâmivikraya. He further mentions that Manu IX, 57-68 first teaches and afterwards forbids the practice of Niyoga, and gives, as it seems to me1, the correct explanation of this contradiction. He also notes that Manu IX, 221-228 forbids gambling, which other writers on law permit under due supervision, and he corrects Manu's rules regarding the indivisibility of clothes and other objects enumerated IX, 219. An apparent contradiction in Brihaspati's rules with respect to subsidiary sons 2 proves that he knew and accepted Manu's teaching on this subject. He declares that the substitutes for a legitimate son of the body are forbidden in the Kaliyuga, and yet admits the rights of a Putrikâ or appointed daughter, who mostly is reckoned among the substitutes. This difficulty is easily solved, if it is borne in mind that Manu, differing from the other ancient law-books, does not reckon the Putrikâ among the subsidiary sons. He separates her, IX, 127-140, from the Gauna Putras, IX, 158-181, and strongly insists on her rights, while he restricts those of the others very much. The list of instances where Brihaspati alludes to, annotates, or amplifies rules of Manu might, I think, be enlarged still further, and it seems to me that a comparison of those verses of his, which Colebrooke's Digest contains, with Manu gives one the impression that Brihaspati's work is throughout a revised and enlarged edition of the Bhrigusamhitâ, or, to use the Indian expression, a Manuvârttika or Manukârikâ. Professor Jolly, finally, has pointed out that this evidence concerning the relation between Manu and Brihaspati agrees with and gives some weight to the tradition preserved in the Skanda-purâṇa, according to which Brihaspati composed the third of the four versions of Manu's Dharmasâstra. The age of the Brihaspati-smriti

1 See also above, p. xciv.

2 Jolly, Tagore Lectures, p. 158.

is circumscribed by its definition of the value of golden dînâras, and by the quotations from it which occur already in the oldest commentaries and Nibandhas from the ninth century A. D. downwards. Since the latter period it has been considered as a work of divine origin, revealed by the teacher of the gods. Hence Professor Jolly's supposition, that it must have existed some two or three hundred years earlier, places it not too early, but, in my opinion, rather too late. But even if the Brihaspati-smriti dates only about 600 A. D., its statements regarding the high authority of Manu's teaching show that our version of the latter must have preceded it by many centuries.

The three points just discussed are, in my opinion, the only ones that are really useful for fixing the lower date of our Manu-smriti. All the other facts known to me which bear on the question are made valueless by flaws of one kind or the other. Thus if we find that another metrical Dharmasastra, the Kâtyâyana-smriti, which probably belongs to the same period as the Brihaspati-smriti, repeatedly quotes doctrines of Manu or Bhrigu found in our text, it is nevertheless not permissible to assume confidently with Kullûka on Manu VIII, 3501, that its author knew and explained our text. For, as Professor Jolly has shown 2, there are other cases in which the teaching attributed by Kâtyâyana to Bhrigu or Manu differs from the opinion advanced in our Smriti. It is, of course, possible that the author, who assumes the name of Kâtyâyana, may have made a slip, or may have known several Manu-smritis or Bhrigu-smritis, and have referred in different places to different works. But, making every allowance for such possibilities, it cannot be said that his references furnish a really conclusive argument. Again, it has been pointed out that the author of the Bhavishya-purâna has largely drawn on the first three chapters of our Manu, whom he also names, and nobody who carefully compares the two

1

' कात्यायनश्च भृगुशब्दोल्लेखनेन मनूक्तं श्लोकं व्यक्तं व्याख्यातवान् ॥

2 Tagore Lectures, p. 62, 11. 22 and 24, Brihaspati has been printed twice by mistake for Kâtyâyana.

3 Professor Aufrecht's Catal. Sansk. MSS. Bodl. Libr. p. 30.