Brugmann's Law and the Sanskrit Vrddhi 9781463221911

Citation preview

Brugmann's Law and the Sanskrit Vrddhi

A n a l e c t a Gorgiana

353 Series Editor George Anton Kiraz

Analecta Gorgiana is a collection of long essays and

short

monographs which are consistently cited by modern scholars but previously difficult to find because of their original appearance in obscure publications. Carefully selected by a team of scholars based on their relevance to modern scholarship, these essays can now be fully utili2ed by scholars and proudly owned by libraries.

Brugmann's Law and the Sanskrit Vrddhi

Carl Darling Buck

l gorgias press 2009

Gorgias Press LLC, 180 Centennial Ave., Piscataway, NJ, 08854, USA www.gorgiaspress.com Copyright © 2009 by Gorgias Press LLC Originally published in All rights reserved under International and Pan-American Copyright Conventions. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise without the prior written permission of Gorgias Press LLC. 2009

1

ISBN 978-1-60724-607-7

ISSN 1935-6854

Extract from The ^American Journal of Philology 17 (1896)

Printed in the LTnited States of America

III.—BRUGMANN'S L A W AND T H E S A N S K R I T VRDDHI. It is a curious irony of fate that ' Brugmann's Law ' refers not to one of this eminent master's numerous discoveries which have been welcomed without dispute and have become the common property of science, but to a dogma which was vigorously attacked at the outset, and which at times has had but few adherents, even having been officially pronounced dead by its opponents. The famous equation of European o with Sanskrit ä in open syllables, when first advanced by Brugmann in 1876 (Zur Geschichte der stammabstufenden Declinationen, Curt. Stud. I X 363 ff.), was intended as an important piece of evidence in support of the primitiveness of the European vowel variation a, e, o, and this fact gave a special animus to the earlier discussions. But the primitiveness of a, e, 0 is no longer a matter of dispute, and Brugmann may well be content with that share of the credit in bringing this about which is universally accorded to his article on the nasalis sonans, removing, as it did, the worst stumbling blocks in the way of such a recognition. The question, then, should be and is, at present, discussed entirely on its own merits. For a time the earlier discussion participated in by Collitz (Bz. B. 2, 291 ff.; cf. also 10, 1 ff.) and J . Schmidt ( K . Z. 25, 1 ff.) in opposition, and by Osthoff (M. U. I 208, note), de Saussure (Système primitif des voyelles) and Hübschmann (Der indogerm. Vocalsystem) in support of the doctrine, gave place to a simple registration of opinion, the doctrine being rejected by various scholars, including Gustav Meyer, Griech. Gram.2, p. 7, note ; Wackernagel, Das Dehnungsgesetz der griech. Composita, 1 1 6 ; Bartholomae, Bz. B. 17, 95. Within the last few years the discussion has been renewed, being opened by Bechtel, Hauptprobleme, 46 ff., with results adverse to the law, while soon after an energetic convert to the same appeared in the person of Streitberg (I. F . I I I 364 ff.). Fay, Am. Journ. Phil. X I I I 478 f., offers a general explanation of the Sanskrit vrddhi as an extension from the a-series. Zubaty, Bz. B. 18, 254, promises a defense of Brugmann's equation, but it has not yet appeared.

446

AMERICAN

JOURNAL

OF

PHILOLOGY.

B a r t h o l o m a e , G r u n d r i s s d. iran. Phil. I 27, renews his opposition a n d H i i b s c h m a n n , I d g . A n z . 6, 35, his allegiance. Finally W a c k ernagel, w h o h a d formerly e x p r e s s e d the opinion that the doctrine was d o n e a w a y with, has a d o p t e d it in part in his A l t i n d i s c h e G r a m m a t i k . A n d still m o r e recently Meillet, M 6 m . S o c . L i n g . I X 142 if., has a r g u e d in opposition to the law. 1 B r u g m a n n ' s o w n attitude, as e x p r e s s e d in his G r u n d r i s s (cf. I, p. 70 f.; II, p. 1205, note), is that, while h e d o e s not maintain the equation as p r o v e n , the probabilities are in his opinion in favor o f its acceptance, and that, in a n y event, it is still an open question, t o be settled b y a continued examination of the facts, not b y d e c l a r a t i o n s a n d c o u n t i n g of hands. A n d s u r e l y it is idle for a n y one to d e n y that the question is a living one, especially since the recent utterances of S t r e i t b e r g and W a c k e r n a g e l . It is, m o r e o v e r , one of the most important questions of I n d o - E u r o p e a n p h o n o l o g y , one u p o n w h i c h an a g r e e m e n t w o u l d be most beneficial. F o r it is impossible to t a k e u p word-formation o f a n y sort in S a n s k r i t without b e i n g b r o u g h t face to face with this problem. It is not m y intention to follow t h e discussion in detail, as B e c h t e l has d o n e . M a n y statements m a d e on both sides w o u l d no l o n g e r b e a d v a n c e d and m a y b e eliminated. T h e f o l l o w i n g propositions are, in m y opinion, b e y o n d dispute and m a y s e r v e t o b r i n g out what is now the crucial point. 1. W h i l e there is n o t h i n g impossible in the supposition of a middle-time v o w e l which a p p e a r s in o n e l a n g u a g e under certain conditions as l o n g , in a n o t h e r as short, y e t the b u r d e n of p r o o f rests distinctly o n t h o s e w h o a s s u m e this. T h i s s e e m s to be a c k n o w l e d g e d b y B r u g m a n n , M . U . I l l 119. 2. O f individual forms there are m o r e w h i c h s p e a k against t h e equation of 0 with S k t . a than for it. N o t that w e can establish a definite numerical ratio after the manner o f J. S c h m i d t . The e x a m p l e s v a r y w i d e l y in respect to the w e i g h t w h i c h can b e attached to them. In s o m e the 0 is not in ablaut with e a n d s o d o e s not c o m e within the r a n g e o f B r u g m a n n ' s h y p o t h e s i s a c c o r d i n g to his later formulation. In s o m e the probability in favor of the o v o w e l is no greater than for e. B u t there is a l a r g e n u m b e r of instances of S k t . a in o p e n syllables w h e r e the 1 T h i s article, a reprint of which I o w e to the kindness of the author, reached m e after my own article w a s practically completed. I h a v e b e e n able to add from it a f e w examples to the lists already c o l l e c t e d , b u t in g e n e r a l our s t a n d p o i n t s are w i d e apart, especially in regard to the causative formation.

BRUGMANN'S

LAW

AND

THE

VBDDHI.

447

evidence of other languages is plainly in favor of I. E . 0. O n e m a y indeed explain these a w a y , as Brugmann has done, b y supposing analogical influence (as in katara, explained once as due to influence of ka-, more recently as due to a katra-~) or by supposing that the Sanskrit represents the ¿-form of the root as against the o of other languages. But then we can also explain away janu and daru, and I believe that B r u g m a n n will admit that the same reasoning which sees I. E . 0 in these forms would apply equally to those other and more numerous instances in which Sanskrit shows a. 3. T h e r e are certain form categories in Sanskrit which receive their simplest explanation through Brugmann's law. S o , in the case of the causatives and perfects, Brugmann maintains that no explanation of these formations which is in any w a y satisfactory has yet been advanced by the opponents of his hypothesis ( G r d . II, pp. 1146, 1205-6). A n d it is through the consideration of the ablaut of the n-, r- and i-stems that Streitberg has been converted. H e says (I. F . I l l 365) that though this or that individual form may yet need explanation, this cannot weigh against the fact that a whole inflectional system is made intelligible through Brugmann's law. W a c k e r n a g e l too, Altind. Gram., p. 13 f., assumes the equation for the categories in question, at the same time admitting that elsewhere I. E. o, even that which is in ablaut with epa>, etc.), and if, in addition to this, the first dual bhdrävas stands for I. E . -dues ( G o t h , bairös, B e z z e n b e r g e r , B z . B . 5, 3 1 9 ; J. S c h m i d t , K . Z. 26, 1 1 ; S t r e i t b e r g , B e i t r ä g e z. i d g . S p r a c h g e s c h i c h t e , 108), the a p p e a r a n c e of ä in the plural is not o n l y natural, but a l m o s t inevitable. 2. T h e inflection o f consonant stems ( B r u g m a n n , padarn = n68a, u§asam — ¡¡¿a, datäram — §¿ropa, dfmänam = aKpova). A s a l r e a d y r e m a r k e d , it is u p o n this point that S t r e i t b e r g bases his s u p p o r t o f B r u g m a n n ' s l a w . A n d y e t this is b y n o means a necessary c o n s e q u e n c e o f his general t h e o r y . O n e m a y well believe with him and with B r u g m a n n (against J. S c h m i d t and Collitz) that the l o n g v o w e l b e l o n g s p r o p e r l y o n l y to the nominative singular. B u t the t y p e s represented b y G r k . aymv, ay&vor, ayäva, L a t . sermö, -önis, -önetn, G r k . Trevßrjv, -rjvos, L a t . hen, lienis, G r k . fi^armp, -rapa, L a t . dator, datöris, G r k . ¿¡orijp, Sori}pos, G o t h , tuggö, tuggöns, O . B . grazdane, s h o w a n actual a d v a n c e o f the l o n g v o w e l into the other cases, w h i c h B r u g m a n n , G r d . II, p. 324, admits m a y h a v e t a k e n p l a c e in the I n d o - E u r o p e a n p e r i o d , — i n part at least, a d d i n g " D o c h k a n n diese S t a m m f o r m , wie umbr. tribris-in-e, na-tin-e, air. er-mi-tin z e i g e n , d a m a l s nicht durch alle C a s u s d u r c h g e d r u n g e n g e w e s e n sein." E x a c t l y so. In G r e e k and L a t i n the 1

T h o u g h the question is strictly one of A r y a n p h o n o l o g y , the Iranian forms

agree so nearly with the Sanskrit that it will not be necessary to refer to them except in special cases.

BRUGMANN'S

LAW

AND

THE

449

VRDDHI.

long vowel appears in all the oblique cases, if at all outside of the nominative, but the earliest extension, at the time (whether proethnic or ur-Greek and ur-Italic) when the weak ablaut forms had not yielded to the strong, would naturally be to the other strong cases only. 1 A n d m a y this not be precisely the state of things represented b y S k t . data, dataram ? It is certainly arbitrary to d e n y for the Sanskrit the possibility of what one admits in the case of G r e e k and L a t i n — n a m e l y , the extension of the l o n g vowel from the nominative. Cf. also Bechtel, Hauptprobleme, 59. Nor can any one deny this possibility, and if Streitberg, nevertheless, prefers the explanation afforded b y B r u g mann's law, it must be on account of certain coincidences which indicate to him that the Sanskrit variation of long and short vowels in the strong cases corresponds to an European variation of e- and 0-vowels. A n d now W a c k e r n a g e l , Altind. Gram., p. 13, while laying stress on certain individual correspondences of a with European 0, expressly states that in the other instances, such as dataram., rajanam, vacant, the a, may be connected with t h e a o f t h e n o m i n a t i v e , a s in G r k . 80r^pa, L a t . datdrem,

sermonem.

It can only be to the advantage of the adherents of the law to eliminate in this w a y what is clearly only a fictitious support, and to recognize that the only genuine argument is to be derived from those forms in which a parallelism is to be observed between the Sanskrit quantitative and the European qualitative variation. A n d there is no d e n y i n g that such parallels as naripa : *eopa = pitdram

: svasaram

and

evpcvca : ^oa =

sumdnasam

: u$asam

are

of striking effectiveness. N o n e o f the previous attacks have attempted to meet this point, with the exception of that of Meillet, M6m. Soc. L i n g . I X 147, and even this is only partially satisfactory. H e explains w h y , in the case of the -on-, -¿?^-stems, an analogical extension of the 0 (a") should affect the accusative in -a°n-am and not the locative in -a'ni; but the inverse argument in relation to the -en-, -