But what confirms me even more in my view that such strict uniformity must not be looked for in the ancient hymns of the Rishis, is the fact that in many cases it would he so very easy to replace the irregular by a regular dipodia. Supposing that the original poets had restricted themselves to the dijambus, who could have put in the place of that regular dijambus an irregular dipodia? Certainly not the authors of the Prâtisâkhya, for their ears had clearly discovered the general rhythm of the ancient metres; nor their predecessors, for they had in many instances preserved the tradition of syllables lengthened in accordance with the requirements of the metre. I do not mean to insist too strongly on this argument, or to represent those who handed down the tradition of the Veda as endowed with anything like apaurusheyatva. Strange accidents have happened in the text of the Veda, but they have generally happened when the sense of the hymns had ceased to be understood; and if anything helped to preserve the Veda from greater accidents, it was due, I believe, to the very fact that the metre continued to be understood, and that oral tradition, however much it might fail in other respects, had at all events to satisfy the ears of the hearers. I should
have been much less surprised if all irregularities in the metre had been smoothed down by the flux and reflux of oral tradition, a fact which is so apparent in the text of Homer, where the gaps occasioned by the loss of the digamma, were made good by the insertion of unmeaning particles; but I find it difficult to imagine by what class of men, who must have lived between the original poets and the age of the Prâtisâkhyas, the simple rhythm of the Vedic metres should have been disregarded, and the sense of rhythm, which ancient people possess in a far higher degree than we ourselves, been violated through crude and purposeless alterations. I shall give a few specimens only. What but a regard for real antiquity could have induced people in VIII, 2, 8, to preserve the defective foot of a Gâyatrî verse, sămâ̄nē ădhĭ bhâ̄rmān? Any one acquainted with Sanskrit would naturally read sămâ̄nē ădhĭ bhâ̄rmănī. But who would have changed bhârmani, if that had been there originally, to bhârman? I believe we must scan sămâ̄nē ădhĭ bh̆â̄rmān, or săm̄ănē ădhĭ bhâ̄rmān, the pæon tertius being a perfectly legitimate foot at the end of a Gâyatrî verse. In X, 158, 1, we can understand how an accident happened. The original poet may have said: S̆û̄ryō nō dĭvās pâ̄tŭ pâ̄tŭ vâ̄tō āntărī̆kshâ̄t, āgnīr nāh pâ̄rthĭvēbhy̆āh. Here one of the two pâtu was lost. But if in the same hymn we find in the second verse two feet of nine instead of eight syllables each, I should not venture to alter this except in pronunciation, because no reason can be imagined why any one should have put these irregular lines in the place of regular ones.
In V, 41, 10, grĭnî̄tē āgnĭr ētărî̄ nă sû̄shāih, sōkīshkēsō nĭ rĭnâ̄tĭ vănâ̄, every modern Pandit would naturally read vanâni instead of vanâ, in order to get the regular Trishtubh metre. But this being the case, how can we imagine that even the most ignorant member of an ancient Parishad should wilfully have altered vanâni into vanâ? What surprises one is, that vanâ should have been spared, in spite of every temptation to change it into vanâni: for I cannot doubt for one moment that vanâ is the right reading, only
that the ancient poets pronounced it văn̄â̆. Wherever we alter the text of the Rig-veda by conjecture, we ought to be able, if possible, to give some explanation how the mistake which we wish to remove came to be committed. If a passage is obscure, difficult to construe, if it contains words which occur in no other place, then we can understand how, during a long process of oral tradition, accidents may have happened. But when everything is smooth and easy, when the intention of the poet is not to be mistaken, when the same phrase has occurred many times before, then to suppose that a simple and perspicuous sentence was changed into a complicated and obscure string of words, is more difficult to understand. I know there are passages where we cannot as yet account for the manner in which an evidently faulty reading found its way into both the Pada and Samhitâ texts, but in those very passages we cannot be too circumspect. If we read VIII, 40, 9, pû̄rvî̄sh tă īndrōpămâ̄tăyāh pû̄rvî̄r ŭtă prăsāstăyāh, nothing seems more tempting than to omit indra, and to read pû̄rvî̄sh tă ŭpămâ̄tăyāh. Nor would it be difficult to account for the insertion of indra; for though one would hardly venture to call it a marginal gloss that crept into the text—a case which, as far as I can see, has never happened in the hymns of the Rig-veda—it might be taken for an explanation given by an Âkârya to his pupils, in order to inform them that the ninth verse, different from the eighth, was addressed to Indra. But however plausible this may sound, the question remains whether the traditional reading could not be maintained, by admitting synizesis of opa, and reading pû̄rvî̄sh tă īndrō͡pămâ̄tăyāh. For a similar synizesis of ̄ ̆, see III, 6, 10. prâ̄kî̄ ādhvărē͡vă tāsthătūh, unless we read prâ̄ky ādhvărēvă.
Another and more difficult case of synizesis occurs in
VII, 86, 4. ăvă tvâ̄nēnâ̄ nămăsâ̄ tŭ͡ră(h) ĭyâ̄m.
can be proved to be legitimate, and if that is the case, then even the synizesis of tŭ͡ră, though hard, ought not to be regarded as impossible.
In II, 18, 5,
Even the epitritus tertius ( ̄ ̄ ̆ ̄) might easily have been avoided by dropping the augment of apâm in X, 119, 1-13, kuvit somasyâ̄pâ̄m ĭtī. It is, in fact, a variety of less frequent occurrence than the rest, and might possibly be eliminated with some chance of success.
Lastly, the choriambus ( ̄ ̆ ̆ ̄) could have been removed in III, 24, 5, sĭsî̄hĭ nāh sû̄nŭmătāh, by reading sû̄nŭmătāh sĭsî̄hĭ nāh, and in VIII, 2, 31, sănâ̄d ămri̅ktō dăyătē, by reading ămri̅ktō dăyătē sănâ̄t.
But I am afraid the idea that regularity is better than irregularity, and that in the Veda, where there is a possibility, the regular metre is to be restored by means of conjectural emendations, has been so ably advocated by some of the most eminent scholars, that a merely general argument would now be of no avail. I must therefore give as much evidence as I can bring together in support of the contrary opinion; and though the process is a tedious one, the importance of the consequences with regard to Vedic criticism leaves me no alternative. With regard, then, to the final dipodia of Gâyatrî verses, I still hold and maintain, that, although the dijambus is by far the most general metre, the following seven varieties have to be recognised in the poetry of the Veda a:
1. ̆ ̄ ̆ ̄ , 2. ̆ ̆ ̆ ̄, 3. ̄ ̆ ̄ ̄, 4. ̆ ̆ ̄ ̄, 5. ̄ ̄ ̄ ̄, 6. ̆ ̄ ̄ ̄, 7. ̄ ̄ ̆ ̄, 8. ̄ ̆ ̆ ̄.
I do not pretend to give every passage in which these varieties occur, out I hope I shall give a sufficient number in support of every one of them. I have confined myself almost entirely to the final dipodia of Gâyatrî verses, as the Ânushtubha verses would have swelled the lists too much.
§ 2. ̆ ̆ ̆ ̄.
I, 12, 9. tasmai pâvakă mrĭlăyā. (Instead of mrilaya, it has been proposed to read mardaya.)
I, 18, 9. divo na sadmămăkhăsām.
I, 42, 4; 46, 2; 97, 1-8; III, 11, 3; 27, 10; IV, 15, 7;
32, 4; 52, 1; V, 5, 9; 7, 4; 7, 5; 7, 7; 9, 4; 53, 12; 61, 3; 61, 11; 64, 5; 65, 4; 82, 9; VI, 16, 17; 16, 18; 16, 45; 45, 17; 61, 4; VII, 15, 14; 66, 2; VIII, 6, 35; 6, 42; 32, 10; 44, 28; 45, 31; 72, 6; 72, 13; 80, 1; 83, 3; 93, 27; IX, 61, 5; 64, 1; X, 118, 6.
§ 3. ̄ ̆ ̄ ̄
I, 22, 11. akkhinnapatrâ̄h săkāntâ̄m.
I, 30, 13. kshumanto yâbhīr mădēmā.
I, 41, 8; 90, 1; 90, 4; 120, 1; V, 19, 1; 70, 3; VI, 61, 10; VIII, 2, 2; 2, 4; 2, 5; 2, 11; 2, 12; 2, 13; 2, 14; 2, 15; 2, 16; 2, 17; 2, 29; 2, 30; 2, 32; 2, 33; 2, 36; 2, 37; 7, 30; 7, 33; 11, 2; 11, 3; 11, 4; 16, 3; 16, 4; 16, 5; 16, 7; 46, 2; 71, 2; 81, 1; 81, 3; 81, 4; 81, 7; 81, 9; 94, 2; IX, 62, 5; X, 20, 4; 20, 7.
§4. ̆ ̆ ̄ ̄.
I, 3, 8. usrâ iva svăsărâ̄nī.
I, 27, 4. agne deveshŭ pră vōkāh.
I, 30, 10; 30, 15; 38, 7; 38, 8; 41, 7; 43, 7; II, 6, 2; III, 27, 3; V, 82, 7; VI, 16, 25; 16, 26; 61, 12; VIII, 2, 1; 2, 3; 2, 8; 2, 18; 2, 19; 2, 21; 2, 22; 2, 23; 2, 26; 2, 35; 16, 2; 16, 6; 16, 8; 71, 9; 79, 3; IX, 21, 5; 62, 6; 66, 21; X, 20, 5; 185, 1; 185, 2; 185, 3.
§5. ̄ ̄ ̄ ̄.
I, 2, 7. dhiyam ghritâkî̄m sâ̄dhāntâ̄.
I, 3, 4. anvîbhis tanâ̄ pû̄tâ̄sāh.
I, 27, 3; 90, 2; II, 6, 4; III, 41, 8; V, 68, 3; 68, 4; VIII, 2, 10; 2, 24; 16, 1; 16, 12; 79, 2; IX, 66, 17; X,
20, 6; 20, 8.
§6. ̆ ̄ ̄ ̄.
I, 15, 6. ritunâ yagñăm â̄sâ̄thē.
I, 38, 2. kva vo gâvo nă rānyāntī (see note to I, 38, 2).
I, 38, 9; 86, 9; III, 27, 2; 41, 3; IV, 32, 23; V, 68, 5; 70, 2; VI, 61, 11; VIII, 2, 20; 2, 25; 7, 32; 26, 19; 79, 4; 79, 5; 81, 6; X, 158, 4.
§7. ̄ ̄ ̆ ̄.
I, 10, 8. sām gâ̄ āsmābhyām dhû̄nŭhī.
I, 12, 5. āgnē tv̆ām rākshāsvĭnāh.
I, 37, 15; 43, 8; 46, 6; III, 62, 7; IV, 30, 21; V, 86, 5; VIII, 5, 32; 5, 35; X, 119, 1-13; 144, 4.
§ 8. ̄ ̆ ̆ ̄.
I, 2, 9. daksham dadhâtē ăpăsām (or § 2).
I, 6, 10. indram maho vâ̄ răgăsāh.
I, 27, 6; 30, 21; 41, 9; 90, 5; III, 24, 5; V, 19, 2; 70, 1; 70, 4; 82, 8; VIII, 2, 27; 2, 31; 16, 9; 55, 4; 67, 19; 81, 5; 81, 8; IX, 47, 2.
But although with regard to the Gâyatra, and I may add, the Ânushtubha pâdas, the evidence as to the variety of their vrittas is such that it can hardly be resisted, a much more determined stand has been made in defence of the vritta of the Traishtubha and Gâgata pâdas. Here Professor Kuhn and those who follow him maintain that the rule is absolute, that the former must end in ̆ ̄ ̆, the latter in ̆ ̄ ̆ ̄, and that the eighth syllable, immediately preceding these syllables, ought, if possible, to be long. Nor can I deny that Professor Kuhn has brought forward powerful arguments in support of his theory, and that his emendations of the Vedic text recommend themselves by their great ingenuity and simplicity. If his theory could be carried out, I should readily admit that we should gain something. We should have throughout the Veda a perfectly uniform metre, and wherever we found any violation of it, we should be justified in resorting to conjectural criticism.
The only question is at what price this strict uniformity can be obtained. If, for instance, in order to have the regular vrittas at the end of Traishtubha and Gâgata lines, we were obliged to repeal all rules of prosody, to allow almost every short vowel to be used as long, and every long vowel to be used as short, whether long by nature or by position, we should have gained very little, we
should have robbed Peter to pay Paul, we should have removed no difficulty, but only ignored the causes which created it. Now, if we examine the process by which Professor Kuhn establishes the regularity of the vrittas or final syllables of Traishtubha and Gâgata pâdas, we find, in addition to the rules laid down before, and in which he is supported, as we saw, to a great extent by the Prâtisâkhya and Pânini, viz. the anceps nature of e and o, and of a long final vowel before a vowel, the following exceptions or metrical licences, without which that metrical uniformity at which he aims, could not be obtained:
1. The vowel o in the body of a word is to be treated as optionally short:
II, 39, 3. prătĭ vāstō̆r ūsrâ̄ (see Trisht. § 5).
Here the o of vastoh is supposed to be short, although it is the Guna of u, and therefore very different from the final e of sarve or âste, or the final o of sarvo for sarvas or mano for manas a. It should be remarked that in Greek, too, the final diphthongs corresponding to the e of sarve and âste are treated as short, as far as the accent is concerned. Hence ἄποικοι, τύπτεται, and even γνῶμαι, nom. plur. In Latin, too, the old terminations of the nom. sing. o and u, instead of the later us, are short. (Neue, Formenlehre, § 23 seq.)
VI, 51, 15. gō̆pâ̄ ămâ̄.
Here the o of gopâ is treated as short, in order to get ̆ ̄ ̆ ̄ instead of ̄ ̄ ̆ ̄, which is perfectly legitimate at the end of an Ushnih.
2. The long î and û are treated as short, not only before vowels, which is legitimate, but also before consonants.
VII, 62, 4. dyâ̄vâ̄bhû̄mî̄ ădĭtē trâ̄s̍î̄̆thâ̄m nāh (see Trisht. § 5).
The forms î̄sî̄yā and râ̄sî̄yā in VII, 32, 18, occur at the end of octosyllabic or Gâyatra pâdas, and are therefore
perfectly legitimate, yet Professor Kuhn would change them too, into î̄sî̆yā and râ̄sî̆yā. In VII, 28, 4, even mâyî is treated as mâ̄yî̆ (see Trisht. § 5); and in VII, 68, 1, vî̄tām as vî̆tām. If, in explanation of this shortening of vîtam, vîhi is quoted, which is identified with vĭhi, this can hardly be considered as an argument, for vĭhĭ occurs where no short syllable is required, IV, 48, 1; II, 26, 2; and where, therefore, the shortening of the vowel cannot be attributed to metrical reasons.
3. Final m followed by an initial consonant is allowed to make no position, and even in the middle of a word a nasal followed by a liquid is supposed to make positio debilis. Several of the instances, however, given in support, are from Gâyatra pâdas, where Professor Kuhn, in some of his later articles, has himself allowed greater latitude; others admit of different scanning, as for instance,
I, 117, 8. măhāh kshōnāsyă āsvĭnâ̄ kānvâ̄yā.
Here, even if we considered the dispondeus as illegitimate, we might scan kă͡nv̆â̄yă, for this scanning occurs in other places, while to treat the first a as short before nv seems tantamount to surrendering all rules of prosody.
4. Final n before semivowels, mutes, and double n before vowels make no position a. Ex. III, 49, I. yāsmī̆n vīsvâ̄ (Trisht. § 5);. I, 174, 5. yāsmĭ̄ñ kâ̄kān; I, 186, 4. sāsmĭn(n) û̄dhān b.
5. Final Visarga before sibilants makes no position c. Ex. IV, 21, 10. sātyă̄h sāmrâ̄t (Trisht. § 5). Even in I, 63, 4.
7. The group ăvă is to be pronounced aūă. Ex. ăvăsē becomes aūăsē; săvĭtâ̄ becomes saūĭtâ̄; năvă becomes naūă.
8. The group ăyă is to be changed into aīă or ēă. Ex. năyăsĭ becomes naīăsĭ.
9. The group vă is to be changed into ua, and this ua to be treated as a kind of diphthong and therefore long. Ex. kānvătămāh becomes kā̆nu͡a̅tămāh; vărŭnāh becomes u͡a̅rŭnāh.
10. The short vowel in the reduplicated syllable of perfects is to be lengthened. Ex. ta̅̆tănāh, da̅̆dhĭrē.
11. Short vowels before all aspirates may be lengthened. Ex. răthâh becomes rāthâh; săkhâ becomes sākhâ.
12. Short vowels before h and all sibilants may be lengthened. Ex. măhĭnĭ becomes māhĭnĭ; ŭsĭgâm becomes ūsĭgâm; rĭshătē becomes rīshătē; dăsāt becomes dāsāt.
13. The short vowel before t may be lengthened. Ex. vâgavătah becomes vâgavātah; ătithih becomes ātithih.
14. The short vowel before d may be lengthened. Ex. ŭdaram becomes ūdaram; ŭd ava becomes ūd ava.
15. The short vowel before p may be lengthened. Ex. ăpâm becomes āpâm; tăpushim becomes tāpushim ; grihăpatim becomes grihāpatim.
16. The short vowel before g and g may be lengthened. Ex. sânushăg asat becomes sânushāg asat; yunăgan becomes yunāgan.
civ:a See some important remarks on these varieties in Mr. J. Boxwell's article 'On the Trishtubh Metre,' Journal Asiat. Soc. Beng., 1885, p. 79.
cvii:a A very strong divergence of opinion is expressed on this point by Professor Bollensen. He says: 'O und E erst später in die Schrifttafel aufgenommen, bewahren ihre Länge durch das ganze indische Schriftenthum bis ins Apabhramsa hinab. Selbstverständlich kann kurz o und e im Veda erst recht nicht zugelassen werden.' Zeitschrift der D. M. G., vol. xxii, p. 574.
cviii:a Professor Kuhn has afterwards (Beiträge, vol. iv, p. 207) modified this view, and instead of allowing a final nasal vowel followed by a mute to make positio debilis, he thinks that the nasal should in most cases be omitted altogether.
cviii:b Here a distinction should be made, I think, between an n before a consonant, and a final n following a short vowel, which, according to the rules of Sandhi, is doubled, if a vowel follows. In the latter case, the vowel before the n remains, no doubt, short in many cases, or, more correctly, the doubling of the n does not take place, e. g. I, 63, 4; 186, 4. In other places, the doubling seems preferable, eg. I, 33, 11, though Professor Kuhn would remove it altogether. Kuhn, Beiträge, vol. iii, p. 125.
cviii:c Here, too, according to later researches, Professor Kuhn would rather omit the final sibilant altogether, loc. cit. vol. iv, p. 207.
