________________
3 28
THE INDIAN ANTIQUARY.
[NOVEMBER, 1882.
A.D. 1662-1737; and (13) Pan-chhen Lo-ssa Paldan-Yee, A.D. 1737-1779, who died on a visit to Pekin, of which an account is given. 6, Life and legend of Lo-safi-tagpa, the great Buddhist reformer of Tibet A.D. 1378-1441, and 7, Rise and progress of Buddhism in Mongolia (8 pp. of Tibetan text) with translation,---in all 141 pages). The paper is accompanied by plates representing the lamas, &c., but the 2nd called Rigdan Tagpa' is not referred to in the text, and it can hardly be meant for Masijusri Kirtti, whose Tibetan name is not given. The only other paper in No. 1 of 1882 is a short memoir of Maulana Minhâju'd-din Abu 'Umari-'Usman, the author of the Tabakit-i Nisiri.
Part Il of the Journal contains several Geogra. phical, Meteorological, and numerous Natural History papers; but the Index and Contents of this part is very late in being published. No part of this division of the volume for 1882 has yet reached us.
The American Oriental Suciety has issued the 1st part of the XIth volume of its Journal containing five papers read between October 25th 1877 and October 28th 1880, which are of the usual high character that distinguishes the published papers of this Society. The let by Mr. A. Hjalmar Edgren is “On the Verbal Roots of the Sanskrit Language and of the Sanskrit Grammarians," in which the author separates the authenticated from the unauthenticated radicals of Sanskrit, and classifies the former ;--meaning by "authenticated" root forms such as have been actually found in any form in Hindu literature as well as in Panini's Dhatupathd and explained by native commentators.
The object of Dr. Edgren's paper is to distinguish the authenticated roots and root-forms in Sanskrit from the unauthenticated, to make a general classification of the former, and to attempt
determination of the character and value of the latter. The author refers first to the familiar fact that a majority of the roots given by the Hindu Grammarians have never been met with in use, and to the suggestions made in explanation of it. The importance of the matter to Indo-European etymology makes desirable a more systematic inquiry. Of the more than two thousand roots catalogued by the grammarians, 974 have been authenticated by being found in use in the litera. ture: and there are besides over 30 Vedic roots which the catalogues do not contain. A considerable number of the former, however, are only dupli. cates, of slightly different form: if these are subtracted, the number is reduced to 879. Taking from this number, again, evident denominatives, there are left 832; and by further deduction of
essentially duplicate and derivative forms, we arrive at the number of 788 radicals, which are either entirely distinct roots, or secondary formations by accretion, or vowel-change and transposition, outside the ordinary grammatical processes and even this number may be further considerably reduced, if we are strict in detecting and casting out such secondary formations.
Of the 832' which remain after taking away graphical variations and denominatives only 549 occur in both the Rig Veda and the later literature; 62 are found in the Rig Veda alone (11 having later derivatives); of the remaining 221, about 30 have derivatives in that Veda, and a considerable part of the rest occur in the other Vedas or in the Brahmanas-not a few only there. Of course, the absence of any root in a single work is no proof of its absence from the language of the period. Yet there are sufficient reasons for believ. ing that a considerable part of the roots here in question are of later origin.
An important characteristic of the authenticated roots is their productiveness, by combination with prepositional prefixes and by formation of derivatives : very few of them remain barren and isolated in the dictionary.
Of the other great class of radical forms, the unauthenticated, there are 1119. Allowing, as before, for slight variations of form in roots of identical meaning, the number will be reduced to rather less than 1000. It is to be noted, however, that meanings wholly diverse and incompatible are freely attributed to these roots, just as to the authenticated roots similar unauthenticated senses are assigned. Of these meanings, as virtually in. creasing the number of roots, no account is here made. The character of the class is discussed under the following heads: 1. The disproportion between the two classes. While Westergaard and other early scholars might hope that the unauthenticated roots would yet be found in parts of the literature then unexplorad, all hope of such a result is now long past. 2. The different relation which the classes sustain to the material of the vocabulary : only a small proportion of the unauthenticated (less than 150) even seem to have any connection with derivative nominal bases 3. The different relation between authenticated radicals of kindred form and meaning on the one hand, and unauthenticated ones of the same kind on the other; and the artificial aspect of the latter. Nearly four-fifths of the second class can be arranged in groups, numbering from two to twenty and more, of identical meaning and of analogous but obviously not historically related form. For example: kev, khev, geo, glev, pev, pleo, mev, mlev, 6sv; meb, peb; mop, lep, are all defined
