BUDDHIST HYBRID SANSKRIT READER (An Old And Rare Book)

Buddhist Hybrid Sanskrit Reader is a collection of selection from the Mahavastu, Mahaparinirvana Sutra, Udanavarga and Lalitavistara, which have been edited according to the principles to be adopted for Buddhist Hybrid Sanskrit. The purpose behind this work is to facilitate the practical use of the author's Buddhist Hybrid Sanskrit Grammar and Dictionary (2 Vols.) by scholars and students as well as teachers interested in the language.

About the Author:

FRANKLIN EDGERTON was Salisbury Professor of Sanskrit and Comparative Philology in Yale University. He has many books and articles to his credit in the fields of Indian literature, Philosophy and religion, Sanskrit and general comparative philology.

It is hoped that this Reader will facilitate the practical use of my Grammar and Dictionary by scholars and students who may wish to acquaint themselves with the language, and by teachers who may wish to conduct courses in it. The most important texts are largely out of print and hard to find, except in large libraries; and even there, as a rule, only a single copy of each text will be found. Furthermore, it would hardly be an exaggeration to say that not one of the texts has been, in my opinion, satisfactorily edited. The selections here printed have been edited according to the principles which I think should be adopted for Buddhist Hybrid Sanskrit (BUS), so far as this is made possible by the variant readings furnished in the critical notes to the printed editions. The editors of the Mahavastu, Mahãparinirvanasutra, Udanavarga, and Lalitavistara, especially the first three, seem to have been careful and conscientious in reporting the exact readings of the mss, they used. Those of the SaddharmapundarTka (SP) were far less so; it has been proved (see my §1.74) that they were very careless; their critical notes often report readings of their mss. Wrongly, and far oftener fail to report at all differences of reading which are found in some or even most of the mss, they used. They also obviously attempted to change the samdhi of the prose of SP to standard Sanskrit Samdhi, while only rarely reporting the samdhi of the mss. For these reasons the SP selections printed here cannot claim to be very close to a real critical edition, and in particular look far more like standard Sanskrit than such an edition would look.

It is unnecessary to repeat here what has been said in the first chapter of the Grammar (see especially §1.33-56; 1.69.77) on the BHS tradition and the way to deal with it. Luders’ principle (1.4O) should be universally applied: any non Sanskritic form presented in the mss. must, in general, be regarded as closer to the original form of the text than a ‘correct’ Sanskrit variant. Most editors, even down to the present, have proceeded on the opposite principle. Indeed, many have gone farther, and ‘corrected’ into Sanskrit non-Sanskrit readings found in all their mss. The plain fact is that BHS is not Sanskrit. Copyists and late redactors did much to Sanskritize it, but never fully succeeded, and modern editors are wrong in carrying the process further. Every Middle Indic or semi-Middle Indic form found in any stream of tradition of any BHS work should, as a rule, he welcomed and adopted in the text, even if Sanskritized substitutes are recorded in the same sentence. All BHS texts, even the Mahavastu, have been subjected to a good deal of Sanskritization, some of it very likely going back to the original composition of the work, but much of it, in the case of most if not all BHS works, introduced by copyists and redactors in the course of the tradition. The Middle-Indicisms, or hybrid forms, which escaped this process should be put into the texts, as a general principle; they constitute precious evidence of an earlier time when the texts were (as most of them certainly were) much less Sanskritized than they seem in our mss. (Such relic forms, by the way, are considerably more numerous, in the prose of such texts as SP, Lalita Vistara (LV), and Divyavadana, than is often supposed.) Instead, many editors try to suppress them, reporting them in notes if they are conscientious, but too often (like the SP editors) failing to do even that. The principles here set forth, like most sound general principles, are not to be applied mechanically; the context, as well as variant ms. readings, will vary from case to case, and each must be separately studied.

The verses present special problems of their own. Here the very brief statements in my Introduction (especially § 1.38) must be supplemented by my article ‘Meter, Phonology and Orthography in Buddhist Hybrid Sanskrit’, JAOS 66.197-206. In this place I can only mention briefly a few general principles of fundamental importance. Most BI-IS verses belong to types known in Sanskrit (but LV, at least, also contains some verses in Apabhrama meters). Their alternations of long and short syllables are as rigidly applied as in Sanskrit, except that in many meters two shorts may be substituted for one long, and one long for two shorts. An initial consonant cluster never ‘makes position’; that is, a short vowel at the end of a preceding word constitutes a short syllable. In the seam of compounds, this rule is optional; that is, juncture may be close or open between the parts of compound, which may be treated as one word or two in this respect. The reason for this peculiarity obviously is that what is written as an initial consonant cluster was originally pronounced in BUS as a single consonant, in Middle Indic fashion.

Still more strange, from the Sanskrit standpoint, is the free and seemingly arbitrary lengthening or shortening of a syllable metri causa. This is accomplished most commonly by lengthening or shortening a vowel, but also by nasalization or decasualization, and by doubling a consonant after a short vowel, or conversely simplifying a double consonant (or orthographic cluster). All these alterations metri causa are commonest at the end of a word, or of a part of a compound; but they also occur internally. In general, the last syllable of a path counts as long (that is, an automatic pause is implied); but occasionally, in some meters, lengthening m.c. seems to occur there, and even at the end of a line.

The recognition of these principles brings with it the corollary that once the meter of a verse is recognized it is sometimes necessary to emend the mss, in accord therewith. This is justified by the fact that the mss. themselves so regularly present such ‘arbitrary’ lengthening and shortenings, when meter requires them, that we must assume copyists’ errors when they fail to do so. In the verses of most texts, such failures are relatively rare. In the Mahavastu they are commoner; but many of the verses of that text show in other respects that the copyists did not understand the meters; the mss. are often full of gross and obvious corruptions. This will be clear from the Mv verses found in this Reader. It is, in fact, sometimes hard to determine the meters of Mv verses; and sometimes the editor failed to see that they were verses at all. To establish the text of them a good deal of bold emendation is at times required. I cannot claim certainty for all of my attempts.