On Instability in the Use of Moods in Earliest Sanskrit 9781463222352

Citation preview

On Instability in the Use of Moods in Earliest Sanskrit

A n a l e c t a Gorgiana

398 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.

On Instability in the Use of Moods in Earliest Sanskrit

Maurice Bloomfield

gorgia* 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 978-1-60724-663-3

ISSN 1935-6854

Extract from The ^American Journal of Philology 33 (1912)

Printed in the LTnited States of America

AMERICAN

JOURNAL VOL. X X X I I I ,

i.

OF

PHILOLOGY WHOLE NO. 129.

I . — O N I N S T A B I L I T Y IN T H E U S E O F M O O D S EARLIEST SANSKRIT.

IN

T h e f o l l o w i n g p a g e s are in partial fulfilment o f p r o m i s e s m a d e in an article, entitled, ' O n certain w o r k in continuance of the V e d i c C o n c o r d a n c e ' , J A O S . X X I X , p. 286-298. O n the last five p a g e s o f that a n n o u n c e m e n t I outlined the k i n d and e x t e n t of information w h i c h the V e d i c variants m i g h t b e m a d e to contribute to the s u b j e c t o f use o f m o o d s . P h i l o l o g i s t s , not least A m e r i c a n philologists, h a v e f r o m time immemorial b e e n s o diligent a n d p r o d u c t i v e a l o n g this line o f s y n t a x , as to warrant the h o p e that the following statements, t h o l a r g e l y statements o f austere facts, will arrest their attention in an u n c o m m o n d e g r e e . T h e gist of the following p a g e s is, that, as far as earliest H i n d u s p e e c h is c o n c e r n e d , ideas w h i c h are e x p r e s s e d in a g i v e n m o o d m a y be, and are, on a l a r g e and s u r p r i s i n g scale, e x p r e s s e d e q u a l l y well in another m o o d , the c i r c u m s t a n c e s under which the t w o statements are m a d e b e i n g p r e c i s e l y the same. T h e so-called V e d i c mantras, b y far the most i m p o r t a n t part o f the V e d a , constitute the earliest form o f H i n d u s p e e c h o f w h i c h w e h a v e r e c o r d . T h e y consist either o f metrical verses (such as those o f the entire R i g - V e d a , and the b u l k of the A t h a r v a - V e d a ) , or p r o s e liturgic p r a y e r s a n d formulas, which p e r v a d e the entire, v e r y e x t e n s i v e ritualistic texts o f the V e d a . T o g e t h e r t h e y m a k e up a vast s t o c k o f sentences, h a n d e d d o w n in varied forms current in a l a r g e n u m b e r o f s c h o o l s or branches (fakhas) o f the S a c r e d L e a r n i n g . It is important to r e m e m b e r that these variants i n v o l v e no c h a n g e of situation w h a t s o e v e r . A g i v e n sentence is e v e r y time the s a m e sentence, e m p l o y e d in the same material 1

2

AMERICAN

JOURNAL

OF

PHILOLOGY.

situation. If there is any psychological shift of attitude in a changed mood, that shift is at the most and solely due to an arbitrary change in the appraisal of the original mood. There is no conceivable motive for the change, except in the subjective feeling of the repeater, or reciter of the second mouth. It is impossible to avoid the conclusion that just such changes of the repeater's mood are, to some extent, and in varying degrees, at the bottom of these changes in grammatical mood ; and yet such assumption is, in the circumstances, the purest kind of argument in a circle. In any case the frequency of these changes testify eloquently to the instability of mood. And because they concern all moods, the following pages are a kind of negative syntax of the ancient Sanskrit moods. I need scarcely remark that conditions such as are sketched in the paragraph above happen to be unparalleled elsewhere in the history of recorded literature and speech. The interest of these interchanges is greatly increased by the uncommonly large formal apparatus for expressing mood which is at the disposal of the Vedic language. This apparatus is considerably larger than that of Classical Sanskrit, and, I believe, also considerably larger than that of any other Indo-European language. The following forms carry with them modal value, under circumstances to be dealt with in detail later on: in various ways and various degrees they interchange one with the other : 1. Present Indicative. 2. Various Preterite Indicatives, notably the Aorist Indicative. 3. Imperative (including Imperatives in tàf). 4. Imperatival forms in si. 5. Subjunctive. 6. Mixed Imperative-Subjunctive forms. 7. Injunctive. 8. Optative. 9. Precative. 10. Future. 11. Desiderative. 12. Infinitive. The rôle of the last two of these twelve classes is unimportant and, as regards strict modal quality, somewhat dubious. This still leaves ten modal categories whose inter-relations are the theme of the following pages.

INSTABILITY

OF MOODS IN EARLIEST

SANSKRIT.

3

Before entering upon a detailed discussion of the modal interchanges in two different texts there are two preliminary matters which lend atmosphere to the subject as a whole. First, one and the same text sometimes varies its mood in what may be regarded as a repetition of the same passage. Secondly, the interchanges in different texts sometimes, tho rarely, bring in three, or even more different moods. Interchange of moods in passages repeated in one and the same text. Any single Vedic text is liable to iteration on an extensive scale. 1 In these repeated passages mood may be changed either, apparently, for no other than subjective reason, or because of some external circumstance provoked by the general technique of the text in question. Thus the R V . : adityair no aditih farma yahsat 1. 107. 2; 4. 54. 6 ddityair no aditih farma yachatu 4. 42. 6. ' Aditi, together with the Adityas, shall offer us protection !' In the first form yansat is aorist subjunctive; in the second form yachatu is present imperative. 2 The motive of the variation is the metre: the first line is tri§tubh (11 syllables); the second jagatl (12 syllables). Any other reason for the change of mood would be sheer hair-splitting: one sentence says exactly the same thing as the other. Similarly it is possible to assign metric convenience for the following case of modal difference: suveda no vasu krdhi 7. 32. 25 ' Make wealth easy for us to g e t ! ' suveda no vasu karat 6. 48. 15 ' Let him make wealth easy for us to g e t ! ' krdhi is 2. sg. aor. impv.; karat is 3. sg. aor. subj.: there is no reason why the poet of 7. 32. 25 should not have used karah to match karat, but we may observe that it would be against the metre to use the aor. impv. *kartu, or the pres. impv. kri^otu or karotu in 6. 48. 15. 1 2

See the author, J A O S . X X X I , 49 ff. Cf. the same relation between R V . 10. 128. 8 a and A V . 5. 3. S.a

4

AMERICAN

JOURNAL

OF

PHILOLOGY.

Similarly metrical considerations may account for the modal change in: svadhvara krnuhi jatavedah

3. 6. 6 ; 7. 17. 3

' Prepare effective sacrifices, O Jatavedas (Agni) !' svadhvara karati jatavedah 6. 10. 1 ; 7. 17. 4 ' Let Jatavedas (Agni) prepare effective sacrifices!' Here krnuhi is 2. sg. pres. impv.; karati, again, 3. sg. aor. subj. T h e last two examples show, what I am certain is true on a larger scale, that 3. person subjunctives are just as good imperatives as the true imperative forms. 1 Remembering that the so-called imperatives of the first person are all thematic subjunctives we may say confidently that there is no real difference between subj. and impv. in the mantras, at least as far as principal sentences are concerned. 2 This is borne out by the very large number of interchanges of the two moods, cited below, p. 18. Not even metrical convenience accounts for the interchange between opt. and subj. in pairs like jyok pagyema suryarn uccarantam 10. 59. 6 jyokpagyat suryarn uccarantam 4. 25. 4 ' A long time may we behold (and, let him behold) the rising sun!', i. e., ' l o n g may we (and, he) live!' It would be just as well to have subj. pagyama in the first passage, or opt. pagyet in the second. Or, in A V . a slight and apparently unnecessary change of tad to idam causes exchange between impv. and inj.: indro marutvan sa dadatu tan me (11. 1. 27, dadad idam me), ' Indra with the Maruts, may he give that to me !' Or, again, there is no apparent reason for the variation between impv. and prec. in one and the same formula in K S . , durmitras tasmdi santu (38. 5, bhuyasur) yo 'sman dve§ti, 3. 8; 38. 5, ' may they be hostile to him that hates u s ! ' 1 Consider especially the third person mixed subjunctive-imperatives, treated below, p. 24, and see also p. 20. 8 T h e chief distinction b e t w e e n the two moods is, in all probability the exclusion of the 2. sing, imperative from dependent sentences: R V . 8. 103 14, madayasva svarnare, ' d e l i g h t thyself in the house of S v a r n a r a ' (impv.), but, 8. 65. 2,yad. . . madaya.se svarnare ' w h e n thou delightest thyself in the house of Svarnara' (subj.).

INSTABILITY

OF MOODS IN EARLIEST

SANSKRIT.

5

On the other hand there is again possible metrical reason for the interchange between opt. and subj. in : jlvema faradah fatam R V . 7. 66. 16 jiväti faradah, fatam R V . 10. 85. 39 ' May we (and, he) live a hundred autumns!' T h e opt. jivet, is barred from the second case, because it is a syllable short. It is of interest to observe that A p M B . 1. 5. 2 changes jiväti of R V . 10. 85. 39 to jivätu, hybrid subj. impv., showing the tense sympathy between subj. and impv. in the third person, alluded to above; it also shows that opt., subj., and impv. are much the same in such connection. Again, the impv. in tät, the so-called future impv. (Whitney, Skt. Gr. § 5 7 1 ; Delbrück, Altindische Syntax, § 2 0 7 ; Speyer, Vedische und Sanskrit-Syntax, § 188, note 1), fails to differentiate itself modally from the ordinary pres. impv. in two R V . parallels, to wit: pra noyachatäd avrkaih prthu ckardih, 1. 48. 15 präsmäiyachatam avrkaih prthu chardih 8. 9. 1 ' D o thou furnish us (and, do y e two furnish htm) broad protection against enemies!' The difference, at most, is chronological: yachatäd is a more archaic form, and, at the same time, occurs here in an older part of the R V . ; see Oldenberg, Prolegomena, p. 262. It is interesting to observe that in one instance in the R V . there is no real difference between a thematic pres. subj. and an indie, aor.: jvan mä somäso mamadan yad ukthä ubhe bhayete rajasl apäre 4. 42. 6, ' W h e n the soma libations and the songs of praise incite me, both boundless hemispheres (heaven and earth) are afright'. purü sahasrä ni fifämi däfu$e yan mä somäsa ukthino amandisuTi 10. 48. 4. W e must of course in verbal translation reproduce the aor. of the second passage: ' Many thousands do I secure to my worshiper when the soma libations accompanied by songs of praise have incited m e ' . In truth, the aor. is here just as modal as the subj.: it is the so-called ' prophetic aorist' which states a wish as an accomplished fact; see below, p. 15.

6

AMERICAN Instances

JOURNAL

OF

PHILOLOGY.

of more than two modal

varieties.

In quite a number of cases more than two moods, usually three, but occasionally even more than three, interchange in different versions of one and the same passage. These are of especial interest, because they show, in a superior degree, that the modal distinctions are the reverse of sharp. The cases here listed are not repeated below under the heads of the much more numerous interchanges between two moods ; they may be added to these without any great inconvenience In the iormulaic prayer which says, ' may I ("thou, we, or he) live a hundred autumns ! t h e r e is a confluence of almost all the modal varieties: subj. jivati, ' may he live ' ; impv. jlva, ' live t h o u ' ; mixed impv.-subj. jivatu, ' may he l i v e ' ; opt. jtvema ' may we live,; subj. first persons jlvava, ' may we two live', and jivani, ' may I live'; and, finally pres. ind. jivami, ' I live'. All these are in final effect wish moods, and nothing more. Their citations may be found in my Vedic Concordance; see, e. g., jivati

faradah

fatam.

W e have, next, almost any three moods in passages that are much more adventitious: O p t a t i v e , P r e c a t i v e , a n d F u t u r e : carum

adya

devebhyo

vacam

udyasam A p f . , ' may I this day speak speech pleasing to the g o d s ! ' : madhumatirh udyasam vacam

(QQ- madhumatlm adya) devebhyo vacam vadi^yami) QQ. T S . T A . ' m a y I (or,

I shall) this day speak honeyed speech to the g o d s ! ' : madhumatim vacam udeyam AV., ' m a y I speak honeyed speech !' Injunctive, Subjunctive and O p t a t i v e : prati vamjihva ghrtam uc caranyat

( A V . caranyat;

T S . caranyef)

AV. TS. MS. KS.

KB. A£. ' may your tongue move up to meet the ghee !' Present Indicative, Imperative, and Precative: surya bhraji§tha bhrajisthas ( T S . QQ. A p £ . bhrajasvin bhrajasvl; VSK. bhrajasvan bhrajasvans; M Q . bhrdjaskara bhrajasvans) tvam deve$v asi ( T S . devequ bhuyah-, M S . devesv edhi) V S . V S K . T S . M S .

' O brilliant Sun, be thou (or, thou art) brilliant among the g o d s ! o r the like. Present Indicative, Imperative, and Subjunctive : k§eme tiqthdti ( P G . tistkatu;

H G . ti^hati)

ghrtam

uk§amana

AV. PG.

HG.,

' may (the dwelling) stand in security ! o r the like. 1

Hereafter the abbreviations for the several texts are those of my Vedic Concordance. To save space citations of places are, as a rule, omitted ; they may be readily supplied from the same work.

INSTABILITY

OF MOODS

IN EARLIEST

SANSKRIT.

7

Subjunctive, Imperative, and Imperative in tat: etarh janatha (KS.jdnzta; T B . janitdf) parame vyoman V S . K S . T B . ; jdnita smdinam ( T S . M Q . j a n i l d d enam) parame vyoman A V . T S . M £ . , ' acknowledge him in highest heaven ! ' Present Indicative, Prophetic Aorist, and Imperative : suga vo devdh sadana akarma ( M S . krnomi; K £ . K a u g . sadanani santu) A V . V S . M S . K £ . Kaup., ' w e have made (or, I make), O y e gods, your seats easy of access for y o u or, ' easy of access, O y e g o d s , shall y o u r seats be for y o u ! ' Imperative, Injunctive, and Imperfect: havya te svadantdm ( M S . svadan; K S . asvadari) V S . T S . M S . K S . , ' delightful to taste be (or, were) thy oblations ! ' 1 Systematic classification of modal interchanges. W e may turn now to a systematic account of the modal variations in one and the same passage when repeated in the texts of different schools. A s far as I can observe these schools show no constitutional preference for some one mood at the expense of another. W h a t they do seem to show is indifference to distinction between them. It is possible, of course, when we find one text, e. g., substituting imperative for subjunctive, or precative for optative, that it approaches the passage from a slightly altered direction, or with a different quality of emotion. But it is e v e r y time the same passage, in the same connection, uttered in the midst of the same real properties; hence any attempt to establish such differences lies outside of the limits of grammatical investigation. A g o o d deal depends upon the constancy of these interchanges; the more frequent these are, the greater the chance that they imply indifference to modal distinction, and nothing more. I would remind the reader that there are in this same sphere about 200 cases of interchange between active and middle, also entirely bare of distinction. 2 W e shall deal with the moods in the order stated above. T h e lists of passages in each class m a y be reinforced here and there from the groups involving more than two modal varieties, above, p. 6. I . Present Indicative in interchange w i t h other moods. T h e most frequent interchange is between the present indicative and other moods. These, again, amount to about 200 cases 1 See in addition some of the cases under the heading ' Mixed Imperative8 See J A O S . X X I X . 293. Subjunctive Forms *, below, p. 27.

8

AMERICAN

JOURNAL

OF

PHILOLOGY.

distributed somewhat unevenly between imperative, subjunctive, and optative. But this interchange is, in the main, temperamental, rather than logical; philological, rather than grammatical. The Vedic mantras deal almost entirely with the praise of fictitious gods; with efforts to coax them into good humor and generosity; and with all sorts of magic or hocus-pocus that is supposed to fulfill wishes. In such atmosphere the indicative which is by nature a sort of modus rectus is in truth a modus obliquus (subjunctivus); almost everything that is stated categorically is meant modally. The indicative states things as certain; as a matter of fact these things are merely wished for, hoped for, requested, or importunately insisted upon. So, e. g., to illustrate by one of the keenest desires of every stratum of the Veda, the desire for dak^ind (baksheesh). A poet priest states, apparently with serene confidence, therefore in the pres. ind., that a certain god is clever ( p r a j a n a n ) in making even the stingy man give gifts to the priests ; aditsantam dapayati prajanan VS. KS., ' he cleverly makes the stingy man give'. In truth this poet is whistling in the woods. What is really meant is, that he wishes, hopes, or requests that the god may, should, or shall do so. Accordingly three other texts change the ind. dapayati, 'makes give', to the impv. dapayatu, 'shall make give to wit: aditsantam (AV. utdditsantam)

dapayatu prajanan AV. TS. MS.

If this were question of logic or grammar, and not of temperament and manner of speaking, we might enrich the vocabulary of grammatical terminology by yet one more item, namely hortative indicative. This kind of indicative is very common, as may be supposed after this description. Its real interest for grammar is to be found in the fact that it varies impartially with pretty much all of the oblique modes, showing that there is in this particular kind of wish statements no real difference between any of these moods. It does not seem to me necessary to cite all cases of such interchange. But I shall illustrate them by a sufficient number of examples for each one of the modes that takes the place of the indicative.

INSTABILITY

OF MOODS

Present Indicative

IN EARLIEST and

SANSKRIT.

Q

Imperative.

T h i s is b y far the most frequent of these interchanges. We m a y be reminded of the indicative in interchange with two other moods (above, p. 6): once, ind. with impv. and prec. ; another time, ind. with impv. and subj. T h e first persons of the impv. are in reality subjunctives. T h e y involve peculiar conditions as far as the older language is concerned ; we m a y reserve them for the next subdivision. A s for the other two persons their extreme readiness to take the place of this ind. calls to mind the fact that throughout Sanskrit literature the impv. is in a marked degree a mode of wish as well as of command, as when, in contrast with Lat. vivai crescat floreat, or vivat rex, Sanskrit uses the i m p v . , j a y a t u räjä, ' may the k i n g be victorious ! ' 1 T h e sequel will show that the impv. encroaches upon the more timid wish-moods to a larger extent than is to be expected in a mood of command, 5 W e m a y illustrate the play between ind. and impv. b y a half a dozen examples each for the second and third persons; the ind. form precedes the impv. in each instance : k§atränäm k§atrapatir asi ( V S . edhi) V S . T S . , ' thou art (or, be thou) sovereign lord of sovereignties ! ' Prayer in behalf of a k i n g on his coronation. syonä casi su$adä cäsi V S . , ' pleasant art thou, and a fair seat to rest on ! ' : syonä ca me su§adä cäidhi T B . , ' be thou pleasant and a fair seat for me ! ' P r a y e r addressed to Earth. pr$thena dyäväprtkivi ( M S . adds äprna) antarikqam ca vi bädkase ( M S . bädhasva) V S . M S . K S . , ' w i t h thy back thou sunderest (or, sunder thou) heaven and earth and the firmam e n t ! ' , or the like. Prayer addressed to the brick altar. ä rohatho ( V S . T S . rohatarh) varuna mitra gartam R V . V S . T S . M S . K S . N., ' y e sit (and, sit ye), O Mitra and Varuna, upon y o u r throne ! ' jätavedasam adhvaränäm janayathah, ( K S . M S . janayatam) purogäm K S . T B . A p Q . M £ . ' y e beget (or, beget ye) Jätavedas ( A g n i ) as the leading-steer of sacrifices ! ' A d d r e s s e d to the two fire-drills. 1 T h e passive impv. is a favorite means of expressing polite request; see Speyer, 1. c., § 192. * Cf. "Whitney, Skt. Gr., § § 572, 575; Delbrück, Altindische Syntax, p. 361 ; Speyer, Vedische und Sanskrit-Syntax § § 188, 192.

IO

AMERICAN JOURNAL OF PHILOLOGY.

somo viram karmanyam dadati (TB. dadatu) . . . yo dadagad asmai RV. VS. MS. TB., 'Soma giveth (or, shall give) a pious son to him that revereth him !' vigvam hi (KS. ha) ripram pravahanti (MS. pravahantu) devih, RV. AV. VS. MS. KS., ' for the goddesses (the Waters) carry off (or, shall carry off) all defilement!' dhanuh gatror apakamam krnoti (MS. krnotu) RV. VS. TS. MS. KSA., ' o u r bow brings (or shall bring) sorrow to the enemy!' a devo yati (so MS. MG.; all the rest yatu) saviid suratnah RV. MS. KS. QB. TB. MG., ' God Savitar comes hither (or, shall come hither) rich in treasures!' edhante asyd jndtayah, RV. AV. ApMB., ' her (the bride's) relations thrive': edhantam jnatayo mama SMB. PG. HG. MG., ' may my (the bridegroom's) relations thrive !' In a weddingcharm ; the point is the same in both forms. In a single instance the impv. takes the place, secondarily, of the pres. ind. in a relative clause : ye . . . svadanti (MS. TB. svadantu) deva ubhaydni havya RV. VS. MS. KS. TB. N., ' the gods who enjoy (or, shall enjoy) both sorts of our oblations '. Present Indicative and

Subjunctive.

I have given mere samples of the interchange between ind. and impv. On the other hand ind. and subj. vary with one another so rarely as to call for all instances. It is quite impossible to feel any distinction between the two classes of interchange; we may conclude that, in this sphere of expression, impv. and subj. perform the same r&le, even if we suspect, as I do, that the impv. is the milder mood of the two. The following are the instances in point of second and third persons: agnir no vanate (VSK. vanute; SV. TS. KS. vansate) rayim RV. SV. VS. VSK. TS. MS. KS., Agni obtains (or, shall obtain) for us wealth!' Here vanate is ambiguous, either pres. ind., or aor. subj.; vanute is pres. ind.; vansate aor. subj. indro jayati (AV. TS. jayati) na para jayate (AV. TS. jaydtai) ' Indra is (or, shall be) victorious, is not (or, shall not be) vanquished!' uta praham atidiva jayati (RV. atidlvya jayati) RV. AV., 'moreover the superior gamester gains (or, shall gain) the stake!'

INSTABILITY

OF MOODS IN EARLIEST

SANSKRIT.

IX

trnam vasana sumana asi ( A V . asas) tvam A V . H G . , ' O house, clothed in grass (i. e. thatched), thou art (or, be thou) kindly disposed towards us ! ' vafi vafarh nayasa ( A V . nayasa) ekaja tvam R V . A V . , 1 thou (Manyu, ' W r a t h ' ) , controling, bringest (or, shalt bring, sc. our foes) under control!' More frequent are the interchanges between the first persons of pres. ind. and subj. The latter forms figure in later Sanskrit as the first persons of the impv. (see above, p. 9); at no time is there any real basis for distinguishing the two varieties of modality in that person. tenaham. asya brahmana ni vartayami ( T B . vartayani) jlvase T B . ApQ. M £ . , ' through that holy rite of his do I return (or, let me return) to life ! ' tayanantam kamam (QQ lokam) aham jay ami (thus A (J.; the rest jayani) AQ. QQ. ApQ., ' through this (oblation) I conquer (or, let me conquer) endless bliss (or heaven) !' brahmaham antaram krnve ( K Q karave — karavat) A V . K £ . , ' I make (or, let me make) the brahma my inner (defence) ! ' T h e preceding examples of 1. person still partake of that temperamental distinction between indicative and the ' oblique m o o d s ' , described above. There, are, however, also plain cases of such interchange as when we say in English either, 'let me eat', or, ' I am going to eat'. F o r the present is so little of a tense as to be at times a future; again, future and subjunctive, especially in Sanskrit, are close allies. Hence the following cases: un nayami A £ . : un nayani K S . K £ . A p £ . M £ . , ' I ladle out', or, ' let me ladle o u t ! ' niharam ni hardmi ( V S . harani) te V S . V S K . T S . K S . , ' to thee I give (or, let me give) my w a r e s ! ' dadamity (AQ. dadaniiy) agnir vadati T B . A (J. ' I give (or, let me give), saith A g n i ' . saiyasya dharmana vi sakhyani srjamahe V a i t . , ' in accordance with the law of truth we dissolve our union'.: sakhyasya dharmai}d vi sakhya visrjavahdi M Q . , ' in accordance with the law of truth let us two dissolve our union!' manve (so M S . ; the rest, manai) babhrUnam ahafu gataih dhamani sapta ca R V . V S . MS. K S . N., ' of these (herbs) whose hue is brown I declare (or, let me declare) the hundred powers and s e v e n ! '

12

AMERICAN JOURNAL OF

PHILOLOGY.

There is yet one more kind of interchange between indicatives and subjunctives of all persons, which is a genuine syntactical interchange. Namely, they vary with one another in relative clauses; the indicative states the fact; the subjunctive assumes it potentially (for the subj. see Delbrück, 1. c., p. 317): yo nah .. . abhidäsati bhrätrvyah . . . idam aham tam adharam pädayämi T S . 3. 2. 10. 2: yo nah . . . abhidäsät sapatnah . . . id am aham, tam adharam pädayämi MS. 4. 5. 8: 76. 13, 'the rival who contends (or, may contend) against us, him do I here lay l o w ! ' sam ajäi^am imä aham sapatnir . . . yathäham asya virasya viräjäni janasya ca ( A p M B . viräjämi dhanasya ca) R V . A p M B . , ' I have conquered these rival women, so that I control (or, may control) this man and his folk (or, and his wealth)'. jägratsvapnah . . . yarn dvismas tam sa rchaiu R V . 10. 164. 5 : ' may wakefulness worry him whom we hate !' päpmä . . . yam dveqäma tam rchatu A V . 6. 26. 3, ' may Evil worry him whom we hate (or, happen to hate)!' tam (sc. devam) äbhara . . . rayim yena vanämahäi (SV. vanämahe) R V . S V . 'bring that (god) hither . . . so that through him we obtain (or, shall obtain) wealth!' yunajmi ta uttarävantam indram yena jayanti (TB. jayäsi) na pari jayante (TB. jayäsäi) A V . TB., ' I join to thee Indra who is eminently superior, by whom men conquer, are not conquered (or by whom thou mayest conquer, not be conquered) yena bhüyag caraty ay am jyok ca pagyati süryah, tena te vapämy äyuqe MG., ' with (the razor) through which he shall live on, and be beheld by the sun, with that do I shear him unto life!': yena bhüyag caräty ay am jyok ca pagyäti süryam tenäsyäyu$e vapa ApMB., 'with (the razor) through which he shall live on and long behold the sun, with that shear him unto life!' kva tyäni näu sakhyä babhüvuh sacävahe (MS. sacävahäi) yad avrkam purä cit R V . MS., ' What has become of the friendship of us two, when formerly we trained together, free from harm ? ' 1 In one case, very similarly, an interrogative clause, which has the effect of a conditional clause, introduces interchange between indie, and subj.: apärn napäd äguhemä kuvit sa supegasas karati ( K S . karoti) jo^ad dhi R V . MS. K S . A p £ . , ' will the Son of the Waters . . . adorn (my songs), so as to enjoy them ?' That is 1

Cf. Delbrück, 1. c., 278.

INSTABILITY

OF MOODS

IN EARLIEST

SANSKRIT.

to say: ' if he will adorn them, he will enjoy them'. karoti is secondary.

13

The present

Present Indicative and Injunctive. The injunctive, or ' improper subjunctive', or ' augmentless subjunctive', is identical in form with augmentless preterits of all classes. T h e distinction between modal value and preterite value is at all times difficult to make. Especially in the Veda which abounds in the quasi-modal use of preterite indicatives, as well as present indicatives (see above, p. 8, and below, p. 14), the distinction between inj. and pret. is not to be made with certainty. Moreover, a large number of inj. forms figure at all times as impvs. tending to efface the individuality of the inj. in this direction also. This last uncertainty is, however quite unimportant, as far as our purposes are concerned. The present class of interchange is rare in any case. In so far as it does occur it puts the inj. completely in line with the subj. The following two cases involve, with reasonable certainty, injunctives in exchange with indicatives, in principal clauses: pra te divo na stanayanti fusmah R.V., ' your lightning fires (O Agni) thunder from heaven': pra te divo na stanayanla fusmdih MS., 'may your (fires) thunder from heaven with lightning!' yad amayati nis krtha (TS. MS. K S . krla),' whatever causes disease ye cure (or, cure y e ) ! ' krta may, of course, be regarded as impv. as well as inj. In two cases again, as in the case of the subjunctive (p. 12), relative clauses are the seat of variation between indicative and injunctive, to wit: hiranyayi arani yam nirmanthato (£B. yabhydfn nirmanthatani) afvina, tam te garbham havamahe da fame mast siltave R V . £ B . , ' the foetus which the two Afvins drill out (or, may drill out) with (their) golden drill (or, two drills), that do we call, that thou mayest beget in the tenth month !' yd rajana saratham yatha (MS. yata) ugrd ta no muncatam dgasah MS. KS., ' y e two kings (Mitra and Varuna) who go (or, may go) in company, do ye free us from sin!' Present Indicative and Optative. The optative, a rather indifferent wish mood, stands, perhaps, more in the centre of the entire sphere of modality than any

14

AMERICAN JOURNAL OF

PHILOLOGY.

other mood. This accounts, to some extent, for its survival in the later language as the heir of the majority of the modi subjunctivi. It is, as we shall see, on terms of lively reciprocity with subj., being, however, rarer than the subj. in the metrical parts of the Veda. With the pres. ind. it alternates in only a moderate number of cases, especially as compared with the impv.; cf. our remarks on the latter mood, above, p. 9. The present makes, as usual, the impression of greater certainty or insistence: The interchanges are, for the most part, in the first person: asyed indro madesv a grabham grbhnati ( R V . grbhnita) sanasim R V . SV., ' when exhilarated by this very soma Indra makes (or, may make) a rich haul!' so 'ham vajam sanamy (thus K S . ; the rest, saneyam) agne V S . T S . MS. KS., ' I here gain (or, may I here gain) substance!' vifvair vifvahgdih saha sam bhavami ( A V . bhavema) A V . M£., ' I come into being (or, may we come into being) with all sound-limbed persons!' marulvantam havdmahe (SV. havemahi) R V . S V . ' (Indra) with the Maruts do we call (or, we wish to call) '. indravanto vanamahe (PB. vanemahi) T S . PB., 'with Indra do we obtain (or, may we obtain), sc. offspring, etc II. Preterits in interchange with other moods. I have used the term ' temperamental' more than once in the preceding pages, in order to describe uses of the categorical indicative that really carry within them modal values of all kinds and degrees. The Vedic poets show even greater tenseness of feeling in that they employ preterite indicatives when they really experience moods. More particularly the aorist, that perfective aorist which is the equivalent of the Greek perfect, looms in this sense. Were there any question as to this use all doubt vanishes when we find that an aorist in the text of one school is replaced in another school by a modal form. So, e. g., some texts say : ' I have speedily attained unto truth ! anjasa satyarn upagdm. MS. K S . But the majority, nine in number, say, using an optative aorist: ' M a y I speedily attain unto truth !' anjasa satyam upa gesam V S . T S . GB. £ B . AQ. g g . Vait. L £ . £ G . Or one text says: ' T h e moon and the constellations have helped thee a l o n g ! c a n d r a m a nak§atrair anu ivavit K S .

INSTABILITY

OF MOODS

IN EARLIEST

SANSKRIT.

15

Another text says, using the imperative : ' May the moon and the constellations help thee along!' candrama nak§atrair anu tvavatu TB. This kind of aorist, especially common in the literature of magic and conjuration, has with some propriety been called 'prophetic aorist'. W e cannot quite look into the heart of those poets and poetasters, those inspired Rishis and voodoo-men, so as to put a precise estimate either upon their fiery prayers, or their bathetic babble. To some extent the prophetic aorist holds serene faith ; more often it has a touch of vulgar slyness, perplexed cocksureness, and even bluster. Underneath it all lurks, in any case, the modal element of desire and doubt which the hot-headed statement does not disguise. The following lists contain the prophetic aorists in interchange with impv., subj.,or opt., in addition to the cases just mentioned. Aorist and Imperative. sa na agan (AV. aitu) varcasa samvidana AV. TS., ' she (the lovely goddess) hath come (or, shall come) to us, endowed with lustre!' suryasya caksur aruham (VS. aroha) VS. TS. MS. K S . , ' the eye of Surya (the Sun) I have ascended (or, ascend thou) !' sam dpo adbhir agmata VS. TS. QQ., 'waters have commingled with waters!': sam apo osadhibhih gachantam M S . , ' let waters commingle with plants !' syonam asadam susaddm asadam (thus LQ.; the rest asida, both times for asadam) VS. TS. MS. KS. L£., 1 1 have seated myself (or, sit thou) upon a (throne) that is fair, upon one that offers a pleasant seat!' viqnus tvakransia (VS. tvd kramatam) VS. Ap£., ' Visnu hath mounted thee (or shall mount thee) (namely, a cart) !' fucim te (SV. ca) varnam adhi gosu dldharam (SV. dharaya) RV. SV., ' t h y bright color, (O Soma), I have infused into the milk', or, ' and bright color do thou infuse into the milk!' ud asau suryo agdt RV. AV. Ap. MB., 'yonder sun hath risen': ud asdv etu suryah TB., yonder sun shall rise !' agnir janavin makyam jay am imam adat Kaug., ' Agni, who owns people, hath given me this wife': agnir janita sa me 'mum jayarn dadatu £G., ' may Agni, the begettor, give me yonder woman for a wife !' a tvaharsam antar abhuh, (RV. edht) RV. AV. VS. TS. MS. KS., ' I have brought thee; thou hast entered (or, enter) within!'

16

AMERICAN

JOURNAL

Aorist and

OF

PHILOLOGY.

Subjunctive.

anyavak^ld (VS. any a vaksad) vasu varyani VS. TB., ' the other (goddess) hath brought (or, shall bring) boons and treasures!' Aorist and Optative. gravavadid (Ap£. grava vaded) abhi somasyahfum (Ap£., °funa) KS. Ap£., ' the press-stone hath spoken (or, shall speak) over the soma-shoot!', or the like. Other preterite forms in interchange with modalforms. The aorist is the most definite of the preterites, stating that a thing has actually taken place. It holds the extreme distance from the true modal types. In between lie the other preterites. They are merely narrative, and, so to speak, assume no real responsibility for the actuality of the event narrated. Yet they also, whether imperfect or perfect, interchange with modal forms. W e may assume that in such cases these preterites state the event not so certainly as the aorist, but with more certainty, and, that too, assumed certainty, than the moods. Both imperfect and perfect are less frequent in these relations than the aorist. Imperfect and

Imperative.

pr§the§v airayad (RV. eraya) rayim RV. SV., ' upon his back (Soma) carried (or, shall carry) hither wealth !' mahyam apo madhumad airayanta (AV. erayantam) AV. KS., ' to me the waters sent (or, shall send) what is sweetness !' praty auhatdm(MS. uhatam) afvina mrtyum asmdt(AV. asmat) AV. VS. TS. MS. KS., ' the Ajvins swept (or shall sweep) away death from him (or, us) !' tvarh gavo 'vrnata rajyaya TS. MS., ' thee the cows chose for kingship': tvarh vifo vrnattim rajyaya, ' thee the clans shall choose for kingship ! ' Imperfect and Subjunctive, a vo rohito afrnod abhidyavah (AV. (rnavat sudanavah) AV. TB., ' Rohita listened to you, ye heavenly (Maruts) ', or ' Rohita shall listen to you, ye liberal (Maruts)!' tad agnir agnaye 'dadat (KS. M£. dadat) KS. Ap£. M£., ' Agni gave (or shall give) this to A g n i ! ' asann a patram janayanta (KS. janayantu) devah RV. VS. TS. MS. KS., ' t h e gods begot (or shall beget) him (Agni) as a

INSTABILITY

OF MOODS

IN EARLIEST

SANSKRIT.

17

vessel at their m o u t h s ! ' . Offerings come to the g o d s through the mediation of A g n i . In this example janayanta may be, less probably, injunctive. Perfect

and modal forms

of various

sorts.

sa vigvä prati cäklpe A V . , ' he (Väigvänara A g n i ) shaped himself into all t h i n g s ' : savigvam prati cäklpat Aak§ ' grow'. Mahldhara refers the word to vah ' carry', something like ' therein carry comfort and ward off (misfortune)!' In either alternative we have a modal form in si, in exchange with the imperative in sva, in Yajus-texts. V. Subjunctive in interchange with other moods. The subjunctive alternates with indicatives (above, pp. 10, 16); with imperative (above, p. 18); and furtherwith injunctive,optative, precative, future, and desiderative. The first persons, as I have remarked before, are indifferently subj. or impv.; they are treated

INSTABILITY

OF MOODS IN EARLIEST

SANSKRIT.

25

here, rather than under impv. In principal clauses there is no perceptible difference between subj. and its modal correspondents. In dependent (relative) clauses 2. pers. impv. forms do not alternate with subjs.; but the indicative (p. 12), the third person imperative (p. 10), injunctive (p. 13, and below), and optative (below) are also fairly common and natural. Subjunctive and

Injunctive.

Alternation between the two moods is common in principal clauses, and occasional in relative clauses : tasmai deva, adhi bravan (MS. K S . A p £ . bruvan) V S . T S . MS. K S . T B . Ap£., ' may the gods comfort him !' tasmai somo adhi bravat ( K S . bruvat) R V . A V . K S . , ' may Soma comfort him!' isarn urjam anya vak§at (TB. 2. 6.10. 3C, vak$it) V S . MS. K S . V . T B . 2. 6. 10. 3 C ; 3. 6. 13. 1, 'let the one bring refreshment and f o o d ! ' praclnam sidat (MS. sidat) pradiga prthivy&h V S . MS. K S . TB., ' may (Indra) sit eastward by the direction of the earth !' pra gmagru ( S V . gmagrubhir) dodhuvad urdhvatha bhut (SV. urdhvadha bhuvaf) R V . SV., 'may (Indra) tossing his beard stand e r e c t ! ' sa tvaitebkyah pari dadat ( T A . dadat) piirbhyah R V . A V . T A . V . , ' he shall hand thee over to these Manes!' sa ( A V . sa) nali garma trivarutham viyansat ( A V . ni yachaf) R V . A V . MS. T B . A p £ . , ' may he (or, she) grant us thricedefending protection !' In relative clauses which, as stated before, (p. 18) greatly favor the subjunctive, the injunctive also appears occasionally : yad adya hotrvarye hotrvurye) jihmam cakquli parapatat parapatat) agni§ tatpunar abharat (£B. abkriyat) QB. QQ. ApQ., ' that which, at the choice of the Hotar-priest, may escape the crooked (i. e. faulty) eye, that may Agni bring back h e r e ! ' It is interesting to observe that there is in the principal clause, as well as in the dependent, interchange of mood, namely subj. and prec., so that there is indifferent inter-play between three moods. tisroyad agne garadas ivam ic chucim ghrtena gucayah saparyan ( T B . saparyan) namani cid dadhire yajhiyani, ' when the three pure autumns honored just thee, O Agni, pure one, they obtained sacred names '.

26

AMERICAN

JOURNAL

Subjunctive

OF

and

PHILOLOGY.

Optative.

Interchange between subj. and opt. in principal clauses is frequent in the first persons where subj. and impv. blend. Only the first of the following examples brings in the third person. In two cases, moreover, first persons alternate with optatives in relative clauses: vidad (S V. vided)

urjaih gatakratur

vidad (SV.

vided)

i$am

R V . S V . , ' may (Indra) who hath hundred-fold wisdom obtain food, obtain refreshment!' vibhum kamarn ( V S . vibhun kaman) vy agnavai

( M S . aglya)

V S . MS. K S . TB., ' may I obtain my vast desire (or, desires) ! ' ima nu kam bhuvana

si^adhama

(SV. T A .

Ap