* See Professor Weber’s pertinent remarks in Kuhn’s Beitrige, vol. iii, p. 394. T do not think that in the verses adduced by Professor Kuhn, in which final o is considered by him as an iambus or trochee, this scanning is inevitable. Thus we may scan the Samhita text:
I, 88, 2. rukmo na Aitrah svadhitivan.
I, 141, 8. ratho na ydtak sikvabhik krito.
I, 174, 3. sho na dame apAmsi vastoh.
VI, 24, 3. aksho na éakryoA sGra brehan
X, 3, 1. ino rdgann aratié samiddho.
This leaves but one of Professor Kuhn’s examples (Beitrige, vol. iv, p. 192) unexplained: I, 191, 1. kahkato na kafkato, where iva for na would remove the difficulty.
> This subject, the shortness of e and o in the Veda, has been admirably treated by Mr. Maurice Bloomfield, ‘ Final as before Sonants,’ Baltimore, 1882. Reprinted from the American Journal of Philology, vol. iii, No. 1.
PREFACE TO THE FIRST EDITION. Ixxxv
Pazini says, VI, 1, 127, that i, u, 7¢ (see RV. Bh. IV, 1, 12) at the end of a pada (but not in a compound®*) may remain unchanged, if a different vowel follows, and that, if long, they may be shortened. He ascribes this rule, or, more correctly, the first portion of it only, to Sdakalya, Pratisikhya 155 seq.> Thus #akri atra may become éakri atra or akry atra. Madhd atra may become madhii atra or madhv atra. In VI, 1, 128, Pasini adds that a, i, u, 7% may remain unchanged before ri, and, if long, may be shortened, and this again according to the teaching of SAkalya, i.e. Pratiskhya 136°. Hence brahma rishi becomes brahma vishi# or brahmarshi#; kumari risyah becomes kuméri risya# or kumary risyak. This rule enables us to explain a number of passages in which the Samhita text either changes the final long vowel into a semivowel, or leaves it unchanged, when the vowel is a pragrzhya vowel. To the first class belong such passages as I, 163, 12; IV, 38, 10, vagi arva, Samh. vagyarva ; VI, 7, 3, vagi agne, Sash. vagyagne; VI, 20, 13, pakth? arkafh, Samh. pakthyarkath ; IV, 22, 4, sushmi 4 gé%, Sah. sushmyd gék. In these passages i is the termination of a nom. masc. of a stem ending in in. Secondly, IV, 24, 8, patni a4k&ha, Samh. patnydkkha ; IV, 34, 1, devi dhnam, Samh. devydhnam; V, 75, 4, vaniki &-hita, Samh. vanifyéhita; VI, 61, 4, avitri avatu, Sah. avitryadvatu. In these passages thei is the termination of feminines. In X, 15, 4, dti arvak, Samh. dtyarvdk, the final i of the instrumental dt? ought not to have been changed into a semivowel, for, though not followed by iti, it is to be treated as pragrihya; (Pratis. 163, 5.) It is, however,
avitryadvatu. In these passages thei is the termination of feminines. In X, 15, 4, dti arvak, Samh. dtyarvdk, the final i of the instrumental dt? ought not to have been changed into a semivowel, for, though not followed by iti, it is to be treated as pragrihya; (Pratis. 163, 5.) It is, however,
® There are certain compounds in which, according to Professor Kuhn, two vowels have been contracted into one short vowel. This is certainly the opinion of Hindu grammarians, also of the compiler of the Pada text. But most of them would admit of another explanation. Thus dhanvarsasah, which is divided into dhAnva-arszasah, may be dhanu-armasah (RV. V, 45, 2). Dhanarfam, divided into dhana-arkam, may have been dhana-rv7#am (RV. X, 46,5). Satarkasam (RV. VII, 100, 3) may be taken as satd-ri4asam instead of sata-aréasam.
> In the Pratisakhya the rule which allows vowel before vowel to remain unchanged, is restricted to special passages, and in some of them the two vowels are savarna; cf. Sitra 163.
mentioned as an exception in Sitra 174, 9. The same applies to II, 3, 4, védi {ti asydm, Sash. védyasyfm. The pragvzhya i ought not to have been changed into a semivowel, but the fact that it had been changed irregularly, was again duly registered in Sftra 174, 5. These two pragrzhya i therefore, which have really to be pronounced short, were irregularly changed in the Samhita into the semivowel ; and as this semivowel, like all semivowels, may take vyavaya, the same object was attained as if it had been written by a short vowel. With regard to pragrzhya Q, no such indication is given by the Samhita text; but in such passages as I, 46, 13, sambha {ti sam-bhQ 4 gatam, Samh. sambhi 4 gatam; V, 43, 4, baht iti ddrim, Sash. bahd adrim, the pragrzhya 0 of the dual can be used as short, like the 0 of madhd atra, given as an example by the commentators of Pazini.
To Professor Kuhn, I believe, belongs the merit of having extended this rule to final 4. That the 4 of the dual may become short, was mentioned in the Pratisdkhya, Sitra 309, though in none of the passages there mentioned is there any metrical necessity for this shortening (see p. lii). This being the case, it is impossible to deny that where this 4 is followed by a vowel, and where Sandhi between the two vowels is impossible, the final 4 may be treated as short. Whether it must be so treated, depends on the view which we take of the Vedic metres, and will have to be discussed hereafter. I agree with Professor Kuhn when he scans:
VI, 63, 1. kva tya valgd puru-hatd adya, (Samh. puruhdtAdya); and not kva tyd valgd puruhdtAdya, although we might quote other verses as ending with an epitritus primus.
IV, 3, 13. m4 vesasya pra-minatak mA Aped, (Samh. mapez,) although the dispondeus is possible.
I, 77, 1. kathd ddsema agnaye kA asmai, (Samh. kAsmai.)
VI, 24, 5. aryak vasasya pari-etA asti. °
IV, 3, 13. m4 vesasya pra-minatak mA Aped, (Samh. mapez,) although the dispondeus is possible.
I, 77, 1. kathd ddsema agnaye kA asmai, (Samh. kAsmai.)
VI, 24, 5. aryak vasasya pari-etA asti. °
Even in a compound like tv4-(ta, I should shorten the first vowel, e.g.
- vrovrv = although the passage is not mentioned by the Pratisakhya
PREFACE TO THE FIRST EDITION. 1xxxvii
among those where a short final vowel in the eighth place is not lengthened when a short syllable follows *.
But when we come to the second pada of a Gayatri, and find there a long 4, and that long 4 not followed by a vowel, I cannot agree with Professor Kuhn, that the long 4, even under such circumstances, ought to be shortened. We may scan:
V, 5, 7. vatasya patman ita daivya hotarA manushad.
The same choriambic ending occurs even in the last pada of a Gayatri, and is perfectly free from objection at the end of the other padas.
So, again, we may admit the shortening of au to o in sAno avye and sano avyaye, as quoted in the Pratisdkhya, 174 and 177, but this would not justify the shortening of au to av in Anushtubh verses, such as
V, 86, 5. martdya devau adabha, amsi-iva devAu arvate, while, with regard to the Trishfubh and Gagati verses, our views on these metres must naturally depend on the difficulties we meet with in carrying them out systematically.
There is no reason for shortening 4 in
V, 5, 10. devandm guhya namani.
It is the second pada of a Gayatri here; and we shall see that, even in the third pada, four long syllables occur again and again.
For the same reason I cannot follow Dr. Kuhn in a number of other passages where, for the sake of the metre, he proposes to change a long 4 into a short one. Such passages are in the Pada text:
VI, 46, 11. didyavak tigma-mdrdhanad, not mardhdnas.
I, 15, 6. ritund yag#am Asathe, not Asdthe.
V, 66, 2. samyak asuryam AsAte, not dsdte.
V, 67, 1. varshish¢Zam kshatram 4sAthe, not 4sdthe. See Beitrage, vol. iii, p. 122.
I, 46, 6. tim asme rasathdm isham, not rdsdtham isham.
* I see that Professor Kuhn, vol. iv, p. 186, has anticipated this observation in eshfau, to be read A-ishyan.
Ixxxviii VEDIC HYMNS.
IV, 32, 23. babhrO yAmeshu sobhete, not sobhete.
IV, 45, 3 uta priyam madhune yuagathim ratham, not yuagdtham ratham.
V. 74, 3. kam akéha yuagathe ratham, not yuégathe ratham.
IV, 55, 1. dyAvabhimi ({ti) adite trasitham naA, not trasithim na.
V, 41, I. ritasya va sadasi trasithdm na, not trasithim nah.
I must enter the same protest against shortening other long vowels in the following verses which Professor Kuhn proposes to make metrically correct by this remedy:
I, 42, 6. hiranyavasimat-tama, not vasimat-tama.
V, 41, I. ritasya va sadasi trasithdm na, not trasithim nah.
I must enter the same protest against shortening other long vowels in the following verses which Professor Kuhn proposes to make metrically correct by this remedy:
I, 42, 6. hiranyavasimat-tama, not vasimat-tama.
Here the short syllable of gazasri-bhiz in V, 60, 8, cannot be quoted as a precedent, for the i in gavasri, walking in companies, was never long, and could therefore not be shortened. Still less can we quote nfari-bhya# as an instance of a long i being shortened, for nari-bhya’ is derived from narif4, not from nari, and occurs with a short i even when the metre requires a long syllable; I, 43, 6. nri-bhyak ndri-bhyad gave. The fact is, that in the Rigveda the forms narishu and n4ri-bhya# never occur, but always nérishu, nari-bhya#; while from vasi we never find any forms with short i, but always vasishu, vasi-bhiz.
Nor is there any justification for change in I, 25, 16. gavak na gavy(ti£ anu, the second pAda of a Gayatri. Nor in V, 56, 3. rikshah na vak marutak simi-vin amak. In most of the passages mentioned by Professor Kuhn on p. 122, this peculiarity may be observed, that the eighth syllable is short, or, at all events, may be short, when the ninth is long:
VI, 44, 21. vrishne te induk vrishabha. pipaya.
I, 73, 1. syona-sih atithié na’ prindnah.
VII, 13, 1. bhare havi# na barhishi privdnas.
II, 28, 7. enak krinvantam Asura bhrivanti.
Before, however, we can settle the question whether in
PREFACE TO THE FIRST EDITION. Ixxxix
these and other places certain vowels should be pronounced as either long or short, we must settle the more general
Metre and question, what authority we have for requir-
Grammar. ing a long or a short syllable in certain places _ of the Vedic metres.
If we declare ourselves free from all authority, either grammatical or metrical, we may either sacrifice all grammar to metre, or all metre to grammar. We may introduce the strictest rules of metre, determining the length or shortness of every syllable, and then ignore all rules of grammar and quantity, treat short syllables as long, or long ones as short, and thus secure the triumph of metre. Or, we may allow great latitude in Vedic metres, particularly in certain pAdas, and thus retain all the rules of grammar which determine the quantity of syllables. It may be said even that the result would really be the same in either case, and that the policy of ‘thorough’ might perhaps prove most useful in the end. It may be so hereafter, but in the present state of Vedic scholarship it seems more expedient to be guided by native tradition, and to study the compromise which the ancient students of the Veda have tried to effect between grammar on one side and metre on the other.
It may be so hereafter, but in the present state of Vedic scholarship it seems more expedient to be guided by native tradition, and to study the compromise which the ancient students of the Veda have tried to effect between grammar on one side and metre on the other.
Now it has generally been supposed that the Pratiskhya teaches that there must be a long syllable in the eighth or tenth place of Traishtubha and GAgata, and in the sixth place of Anushzubha padas. This is not the case. The Pratisakhya, no doubt, says, that a short final vowel, but not any short syllable, occupying the eighth or tenth place in a Traish¢ubha and Gagata pada, or the sixth place ina Gayatra pada, is lengthened, but it never says that it must be lengthened ; on the contrary, it gives a number of cases where it is not so lengthened. But, what is even more important, the Pratisikhya distinctly adds a proviso which shows that the ancient critics of the Veda did not consider the trochee as the only possible foot for the sixth and seventh syllables of Gayatra, or for the eighth and ninth, or tenth and eleventh syllables of Traishtubha and Gagata padas. They distinctly admit that the seventh and the xc VEDIC HYMNS.
ninth and the eleventh syllables in such pAdas may be long, and that in that case the preceding short vowel is not lengthened. We thus get the iambus in the very place which is generally occupied by the trochee. According to the Pratisdkhya, the general scheme for the Gayatra would | be, not only 6 teea] eos, but also 67 ++4+4+[ 44-4; and for the Traishtubha and GAgata, not only
p. 8
+++ 4 [+ t+-lud + (+), but also B 9
+++ ¢]+ + + 4[-+ + (4). And again, for the same p4das, not only oil t++ttl[ +++ 4+[4+-(4), but also IO
+++t¢] + ++ 4+] 4+ 4-(4).
Before appealing, however, to the Pratisakhya for the establishment of such a rule as that the sixth syllable of Anushtubha and the eighth or tenth syllable of Traishtubha and Gagata padas must be lengthened, provided a short syllable follows, it is indispensable that we should have a clear appreciation of the real character of the Pratisakhya. If we carefully follow the thread which runs through these books, we shall soon perceive that, even with the proviso that a short syllable follows, the Pratisakhya never teaches that certain final vowels must be lengthened. The object of the Pratisakhya, as I pointed out on a former occasion, is to register all the facts which possess a phonetic interest. In doing this, all kinds of plans are adopted in order to bring as large a number of cases as possible under general categories. These categories are purely technical and external, and they never assume, with the authors of the Pratisékhya, the character of general rules. Let us now, after these preliminary remarks, return to the Sdtras 523 to 535, which we discussed before. The Pratisdkhya simply says that certain syllables which are short in the
PREFACE TO THE FIRST EDITION. xci
Let us now, after these preliminary remarks, return to the Sdtras 523 to 535, which we discussed before. The Pratisdkhya simply says that certain syllables which are short in the
PREFACE TO THE FIRST EDITION. xci
Pada, if occupying a certain place in a verse, are lengthened in the Samhita, provided a short syllable follows. This looks, no doubt, like a general rule which should be carried out under all circumstances. But this idea never entered the minds of the authors of the Pratisikhya. They only give this rule as the most convenient way of registering the lengthening of certain syllables which have actually been lengthened in the text of the Sashita, while they remain short in the Pada; and after having done this, they proceed to give a number of verses where the same rule might be supposed to apply, but where in the text of the Samhita the short syllable has not been lengthened. After having given a long string of words which are short in the Pada and long in the Samhita, and where no intelligible reason ° of their lengthening can be given, at least not by the authors of the Pratisakhya, the Pratisakhya adds in Sdtra 523, ‘The final vowel of the eighth syllable is lengthened in padas of eleven and twelve syllables, provided a syllable follows which is short in the Samhita.’ As instances the commentator gives (Samhita text):
I, 32, 4. tdditnd sdtrum nd Kila vivitse.
- - - - t vo7 bbe v
I, 94,1. A4gne sakhyé m4 rishdma vaydm tava.
Then follows another rule (Sdtra 525) that ‘The final vowel of the tenth syllable in padas of eleven and twelve syllables is lengthened, provided a syllable follows which is short in the Samhita.’ As instances the commentator gives:
- i] A
III, 54, 22. hd visvd sumand didihi nad.
II, 34, 9. ava rudrd asiso hantana vidhas.
Lastly, a rule is given (Sdtra 526) that ‘The final vowel of the sixth syllable is lengthened in a pada of eight syllables, provided a syllable follows which is short :’ -
I, 5,10. tsdno yavaya vadhdm,
If the seventh syllable is long no change takes place: IX, 67, 30. 4 pavasva deva soma.
While we ourselves should look upon these rules as xcii VEDIC HYMNS.
tsdno yavaya vadhdm,
If the seventh syllable is long no change takes place: IX, 67, 30. 4 pavasva deva soma.
While we ourselves should look upon these rules as xcii VEDIC HYMNS.
founded in the very nature of the metre, which, no doubt, to a certain extent they are, the authors of the Pratisikhya use them simply as convenient nets for catching as many cases as possible of lengthened syllables actually occurring in the text of the Samhita. For this purpose, and in order to avoid giving a number of special rules, they add in this place an observation, very important to us as throwing light on the real pronunciation of the Vedic hymns at the time when our Samhita text was finally settled, but with them again a mere expedient for enlarging the preceding rules, and thus catching more cases of lengthening at one haul. They say in Stra 527, that in order to get the right number of syllables in such verses, we must pronounce sometimes one syllable as two. Thus only can the lengthened syllable be got into one of the places required by the preceding Sftra, viz. the sixth, the eighth, or the tenth place, and thus only can a large number of lengthened syllables be comprehended under the same general rule of the Pratisakhya. In all this we ourselves can easily recognise a principle which guided the compilers of the Samhita text, or the very authors of the hymns, in lengthening syllables which in the Pada text are short, and which were liable to be lengthened because they occupied certain places on which the stress of the metre would naturally fall. We also see quite clearly that these compilers, or those whose pronunciation they tried to perpetuate, must have pronounced certain syllables as two syllables, and we naturally consider that we have a right to try the same expedient in other cases where to us, though not to them, the metre seems deficient, and where it could be rendered perfect by pronouncing one syllable as two. Such thoughts, however, never entered the minds of the authors of the Pratisikhyas, who are satisfied with explaining what is, according to the authority of the Samhita, and who never attempt to say what ought to be, even against the authority of the Samhita. While in some cases they have ears to hear and to appreciate the natural flow of the poetical language of the Rishis, they seem at other times as deaf as the adder to the voice of the charmer.
PREFACE TO THE FIRST EDITION, Xcili
While in some cases they have ears to hear and to appreciate the natural flow of the poetical language of the Rishis, they seem at other times as deaf as the adder to the voice of the charmer.
PREFACE TO THE FIRST EDITION, Xcili
A general rule, therefore, in our sense of the word, that the eighth syllable in hendecasyllabics and dodecasyllabics, the tenth syllable in hendecasyllabics and dodecasyllabics, and the sixth syllable in octosyllabics should be lengthened, rests in no sense on the authority of ancient grammarians. Even as a mere observation, they restrict it by the condition that the next syllable must be short, in order to provoke the lengthening of the preceding syllable, thereby sanctioning, of course, many exceptions; and they then proceed to quote a number of cases where, in spite of all, the short syllable remains short*. In some of these quotations they are no doubt wrong, but in most of them their statement cannot be disputed.
As to the eighth syllable being short in hendecasyllabics and dodecasyllabics, they quote such verses as,
VI, 66, 4. antar ({ti) santak avadyAni pundndd.
Thus we see that in VI, 44, 9, varshiyak vayak krinuhi sakibhif, hi remains short; while in VI, 25, 3, gahi vrishnyAni krinuht pardkaf, it 1s lengthened in the Samhita, the only difference being that in the second passage the accent is on hi.
As to the tenth syllable being short in a dodecasyllabic, they quote
II, 27, 14. adite mitra varuma uta mrda.
As to the tenth syllable being short in a hendecasyllabic, they quote
II, 20, 1. vayam te vayak indra viddhi su nad.
As to the sixth syllable being short in an octosyllabic, they quote
VIII, 23, 26. mahak visvan abhi satak.
A large number of similar exceptions are collected from
* ‘Wo die achtsilbigen Reihen mit herbeigezogen sind, ist es in der Regel bei solchen Liedern geschehen, die im Ganzen von der regelmissigen Form weniger abweichen, und fiir solche Fille, wo auch das Prdtisdkhya die Langung der sechsten Silbe in achtsilbigen Reihen vorschreibt, namlich wo die siebente von Natur kurz ist. Die achtsilbigen Reihen bediirfen einer erneuten Durchforschung, da es mehrfach schwer fallt, den Samhitatext mit der Vorschrift der Pratisikhya in Ubereinstimmung zu bringen.’ Kuhn, Beitrage, vol. iii, p. 450; and still more strongly, p. 458.
XCciv VEDIC HYMNS.
528, 3 to 534, 94, and this does not include any cases where the ninth, the eleventh, or the seventh syllable is long, instead of being short, while it does include cases where the eighth syllable is long, though the ninth is not short, or, at least, is not short according to the views of the collectors of these passages. See Sitra 522, 6.
See Sitra 522, 6.
Besides the cases mentioned by the PratisAkhya itself, where a short syllable, though occupying a place which would seem to require lengthening, remains short, there are many others which the Pratisakhya does not mention, because, from its point of view, there was no necessity for doing so. The Pratisdkhya has been blamed® for omitting such cases as I, 93, 6, urum yag#aya akrathir u lokam; or I, 96, 1, deva agnim dharayan dravizodam. But though occupying the eighth place, and though followed by a short syllable, these syllables could never fall under the general observation of the Pratisakhya, because that general observation refers to final vowels only, but not to short syllables in general. Similar cases arg I, 107, 1%; 122, 93 130, 10; 152,63; 154,13; 158, 5%; 163, 2; 167, 10"; 171, 4; 173,6; 179, 1%; 182, 8; 186, 6, &c.
If, therefore, we say that, happen what may, these metrical rules must be observed, and the text of the Veda altered in order to satisfy the requirements of these rules, we ought to know at all events that we do this on our own responsibility, and that we cannot shield ourselves behind the authority of Saunaka or Katyayana. Now it is well known that Professor Kuhn? has laid down the rule that the Traish¢ubha p4das must end in a bacchius or amphibrachys v—-™, and the G4gata pddas in a dijambus or pzon secundus u—w¥. With regard to Anushéubha pddas, he requires the dijambus or pzon secundus v—v™ at the end of a whole verse only, allowing greater freedom in the formation of the preceding pddas. In a later article,
* ‘Dazu kommt, dass der uns vorliegende Samhitatext viclfaltig gar nicht mit Saunaka’s allgemeiner Regel iibereinstimmt, indem die Verlaingerung kurzer Sil ben nicht unter den Bedingungen eingetreten ist, die er vorschreibt.’ Kuhn, Beitrage, vol. iii, p. 459.
> Beitrage zur. Vergleichenden Sprachforschung, vol. iii, p. 118.
PREFACE TO THE FIRST EDITION. XCV
however, the final pada, too, in Anush/ubha metre is allowed greater freedom, and the rule, as above given, is strictly maintained with regard to the Traishfubha and Gagata pAdas only.
This subject is so important, and affects so large a number of passages in the Veda, that it requires the The four prine Most careful examination. The Vedic metres, cipal Padas’ though at first sight very perplexing, are very simple, if reduced to their primary elements. The authors of the Pratisikhyas have elaborated a most complicated system. Counting the syllables in the most mechanical manner, they have assigned nearly a hundred names to every variety which they discovered in the hymns of the Rig-veda*. But they also observed that the constituent elements of all these metres were really but four, (Sitras 988, 989): °
1. The Gdyatra pada, of eight syllables, ending in U-.
2. The Vairaga pAda, of ten syllables, ending in ——-.
3. The Traishtubha p4da, of eleven syllables, ending in --.
4. The GAgata pada, of twelve syllables, ending in v-.
The Gdyatra pada, of eight syllables, ending in U-.
2. The Vairaga pAda, of ten syllables, ending in ——-.
3. The Traishtubha p4da, of eleven syllables, ending in --.
4. The GAgata pada, of twelve syllables, ending in v-.
Then follows an important rule, Sdtra 990: ‘The penultimate syllable,’ he says, ‘in a Gayatra and Gagata pada is light (laghu), in a Vairaga and Traishtubha pada heavy (guru).’ This is called their vritta.
This word vrztta, which is generally translated by metre, had evidently originally a more special meaning. It meant the final rhythm, or if we take it literally, the turn of a line, for it is derived from vvit, to turn. Hence vritta is the same word as the Latin versus, verse; but I do not wish to decide whether the connection between the two words is historical, or simply etymological. In Latin, versus is always supposed to have meant originally a furrow, then a line, then a verse. In Sanskrit the metaphor that led to the formation of vvétta, in the sense of final rhythm, has nothing to do with ploughing. If, as I have tried to prove (Chips from a German Workshop,
Varitta = versus.
* See Appendix to my edition of the Pratisikhya, p. ccclvi.
xcvi VEDIC HYMNS.
vol. i, p. 84), the names assigned to metres and metrical language were derived from words originally referring to choregic movements, vvztta must have meant the turn, i.e. the last step of any given movement; and this turn, as determining the general character of the whole movement, would naturally be regulated by more severe rules, while greater freedom would be allowed for the rest.
the last step of any given movement; and this turn, as determining the general character of the whole movement, would naturally be regulated by more severe rules, while greater freedom would be allowed for the rest.
Having touched on this subject, I may add another fact in support of my view. The words Trishéubh and Anushtubh, names for the most common metres, are generally derived from a root stubh, to praise. I believe they should be derived from a root stubh, which is preserved in Greek, not only in orugeAds, hard, crudeAl(w, to strike hard, but in the root creud, from which ordudvaop, stamped or pressed olives or grapes, and doreygijs, untrodden (grapes), then unshaken; and in oréu8w, to shake, to ram, oroBéw, to scold, &c. In Sanskrit this root is mentioned in the Dhatup4tka X, 34, shtubhu stambhe, and it exists in a parallel form as stambh, lit. to stamp down, then to fix, to make firm, with which Bopp has compared the German stam pfen, to stamp ; (Glossarium, s. v.stambh.) I therefore look upon Trish¢ubh as meaning originally tripudium, (supposing this word to be derived from tri and pes, according to the expression in Horace, pepulisse ter pede terram, Hor. Od. iii. 18,) and I explain its name ‘ Three-step,’ by the fact that the three last syllables U—u, which form the characteristic feature of that metre, and may be called its real vritta or turn, were audibly stamped at the end of each turn or strophe. I explain Anush¢ubh, which consists of four equal padas, each of eight syllables, as the ‘ After-step,’ because each line was stamped regularly after the other, possibly by two choruses, each side taking its turn. There is one passage in the Veda where Anushtubh seems to have preserved this meaning :
Poets by their wisdom discovered Indra dancing to an Anush/¢ubh.
In V, 52, 12, &handah-stibhak kubhanydvah utsam 4
PREFACE TO THE FIRST EDITION. XCVii
kirfzak nrituh, in measured steps (i.e. stepping the metre) and wildly shouting the gleemen have danced toward the spring.
Other names of metres which point to a similar origin, ie. to their original connection with dances, are Padapankti, ‘Step-row;’ Nyanku-sérini, ‘Roe-step;’ AbhisArini, ‘Contre-danse,’ &c.
If now we return to the statement of the PratisAkhya in reference to the vvsttas, we should observe how careful its author is in his language. He does not say that the penultimate is long or short, but he simply states, that, from a metrical point of view, it must be considered as light or heavy, which need not mean more than that it must be pronounced with or without stress. The fact that the author of the Prdtisakhya uses these terms, laghu and guru, instead of hrasva, short, and dirgha, long, shows in fact that he was aware that the penultimate in these pAdas is not invariably long or short, though, from a metrical point of view, it is always heavy or light.
The fact that the author of the Prdtisakhya uses these terms, laghu and guru, instead of hrasva, short, and dirgha, long, shows in fact that he was aware that the penultimate in these pAdas is not invariably long or short, though, from a metrical point of view, it is always heavy or light.
It is perfectly true that if we keep to these four p4das, (to which one more pAda, viz. the half Vairdga, consisting of five syllables, might be added,) we can reduce nearly all the hymns of the Rig-veda to their simple elements which the ancient poets combined together, in general in a very simple way, but occasionally with greater freedom. The most important strophes, formed out of these padas, are,
1. Three Gayatra padas=the Gayatri, (24 syllables.)
2. Four Gayatra padas=the Anush¢ubh, (32 syllables.)
3. Four Vairaga padas=the Virdg, (40 syllables.)
4. Four Traish¢ubha padas=the Trish¢ubh, (44 syllables.)
5. Four Gagata padas=the Gagati, (48 syllables.)
Between the Gayatri and Anush¢ubh strophes, another strophe may be formed, by mixture of Gayatra and Gagata padas, consisting of 28 syllables, and commonly called Ushzih ; likewise between the Anushfubh and the Virag, a strophe may be formed, consisting of 36 syllables, and commonly called Brzhati.
In a collection of hymns, however, like that of the Rigveda, where poems of different ages, different places, and
(32) g
XCVill VEDIC HYMNS.
different families have been put together, we must be prepared for exceptions to many rules. Thus, although the final turn of the hendecasyllabic Traish¢ubha is, as a rule, the bacchius, .—-, yet if we take, for instance, the 77th hymn of the tenth Mandala, we clearly perceive another hendecasyllabic p4da of a totally different structure, and worked up into one of the most beautiful strophes by an ancient poet. Each line is divided into two halves, the first consisting of seven syllables, being an exact counterpart of the first member of a Saturnian verse (fato Romae Metelli); the second a dijambus, answering boldly to the broken rhythm of the first member*. We have, in fact, a Trish¢ubh where the turn or the three-step, v—-, instead of being at the end, stands in the middle of the line.
X, 77, 1-5, in the Pada text: 1. abhra-prusha/ na v4#d prusha vasu,
havishmantad na yagwds vi-ginushak | &c.
Another strophe, the nature of which has been totally misapprehended by native metricians, occurs in IV, 10. It is there called Padapankti and Mahdpadapankti; nay, attempts have been made to treat it even as an Ushzih, or as a kind of Gayatri. The real character of that strophe is so palpable that it is difficult to understand how it could have been mistaken. It consists of two lines, the first embracing three or four feet of five syllables each, having the ictus on the first and the fourth syllables, and resembling the last line of a Sapphic verse. The second line is simply
It consists of two lines, the first embracing three or four feet of five syllables each, having the ictus on the first and the fourth syllables, and resembling the last line of a Sapphic verse. The second line is simply
® Professor Kuhn (vol. iii, p. 450) is inclined to admit the same metre as varying in certain hymns with ordinary Traishfubha pAdas, but the evidence he brings forward is hardly sufficient. Even if we object to the endings v-—v- and --.u-, V, 33,4, may be a Gagata, with vyfha of dasa, the remark quoted from the Pratisikhya being of no consequence on such points; and the same remedy would apply to V, 41, 5, with vydha of eshe. In VI, 47, 31, vyfha of asvaparzaik; in I, 33, 9, vydha of indra and rodasf ; in II, 24, 5, vydha of m&dbhiA would produce the same effect ; while in I, 121, 8, we must either admit the Traishfubha vttta -— J — or scan dhukshan. In III, 58, 6, I should admit vydha for nard; in IV, 26, 6, for mandram; in I, 100, 8, for gyotiA, always supposing that we consider the ending — - v — incompatible with a Trishfubh verse.
1. Agne tam adyalasvam na stomaikikratum na bhadram, hridisprisam ridhy4md ta ohaid. 2, Adhd hy agneikrator bhadrasyaidakshasya sddhoé, rathir ritasya brzhato babhitha, &c.
I give the Samhita text:
1. Agne tam adyalasvam na stomaikikratum na bhadram, hridisprisam ridhy4md ta ohaid. 2, Adhd hy agneikrator bhadrasyaidakshasya sddhoé, rathir ritasya brzhato babhitha, &c.
Now it is perfectly true that, as a general rule, the syllables composing the vritta or turn of the different metres, and described by the Pratisakhya as heavy or light, are in reality long or short. The question, however, is this, have we a right, or are we obliged, in cases where that syllable is not either long or short, as it ought to be, so to alter the text, or so to change the rules of pronunciation, that the penultimate may again be what we wish it to be?
If we begin with the Gayatra pada, we have not to read g2 c VEDIC HYMNS.
long before we find that it would be hopeless to try to crush the Gayatri verses of the Vedic Rzshis on this Procrustean bed. Even Professor Kuhn very soon perceived that this was impossible. He had to admit that in the Gayatri the two first pAdas, at all events, were free from this rule, and though he tried to retain it for the third or final pada, he was obliged after a time to give it up even there. Again, it is perfectly true, that in the third pada of the Gayatri, and in the second and fourth pAadas of the Anush¢ubh strophe, greater care is taken by the poets to secure a short syllable for the penultimate, but here, too, exceptions cannot be entirely removed. We have only to take such a single hymn as I, 27, and we shall see that it would be impossible to reduce it to the uniform standard of Gayatri padas, all ending in a dijambus.
Gfyatra Padas.
But what confirms me even more in my view that such strict uniformity must not be looked for in the ancient
Conjectural hymns of the Rishis, is the fact that in many emendations. cases it would be 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 Pratisakhya, 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
PREFACE TO THE FIRST EDITION. cl
I should
PREFACE TO THE FIRST EDITION. cl
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 Pratisakhyas, 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 GAyatri verse, samAne adhi bharman? Any one acquainted with Sanskrit would naturally read samAne adhi bharmazi. But who would have changed bharmazi, if that had been there originally, to bharman? I believe we must scan sam4ne adhi bharman, or samane adhi bharman, the pon tertius being a perfectly legitimate foot at the end of a Gayatri verse. In X, 158, 1, we can understand how an accident happened. The original poet may have said: Saryo no divas p4tu p4tu v4to antarikshdt, agnir nah parthivebhyaz. Here one of the two p4tu 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, grinite agnir etari na sdshaik, sokishkeso ni rindti van, every modern Pandit would naturally read vandni instead of vana, in order to get the regular Trish¢ubh metre. But this being the case, how can we imagine that even the most ignorant member of an ancient Parishad should wilfully have altered vanani into vand? What surprises one is, that van4 should have been spared, in spite of every temptation to change it into vanani» for I cannot doubt for one moment that van4 is the right reading, only
Cli VEDIC HYMNS.
What surprises one is, that van4 should have been spared, in spite of every temptation to change it into vanani» for I cannot doubt for one moment that van4 is the right reading, only
Cli VEDIC HYMNS.
that the ancient poets pronounced it vana. 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. Ifa 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 Samhit4 texts, but in those very passages we cannot be too circumspect. If we read VIII, 40, 9, parvish fa indropamatayas parvir uta prasastayah, nothing seems more tempting than to omit indra, and to read parvish fa upamatayak. 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 Afarya 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 parvish fa indropamatayah. F or a similar synizesis of — vu, see III, 6, 10, prafi adhvareva tasthatu/, unless we read pray adhvareva. Another and more difficult case of synizesis occurs in VII, 86, 4. ava tvdnend namasd tura(A) iydm. It would be easy to conjecture tvarey4am instead of tura iyam, but tvarey4m, in the sense of ‘let me hasten,’ is not Vedic. The choriambic ending, however, of a Trishfubh
PREFACE TO THE FIRST EDITION. clii
can be proved to be legitimate, and if that is the case, then even the synizesis of tura, though hard, ought not to be regarded as impossible.
In II, 18, 5, 4 vimsatyd trimsatd yahy arvn,
A Aatvarimsatd haribhir yugdnah,
4 pa#hdsata surathebhir indra,
4 shashtya saptatya somapeyam, Professor Kuhn proposes to omit the 4 at the beginning of the second line, in order to have eleven instead of twelve syllables. By doing so he loses the uniformity of the four padas, which all begin with 4, while by admitting synizesis of haribhi# all necessity for conjectural emendation disappears.
By doing so he loses the uniformity of the four padas, which all begin with 4, while by admitting synizesis of haribhi# all necessity for conjectural emendation disappears.
If the poets of the Veda had objected to a pzon quartus (vvv-) at the end of a Gayatri, what could have been easier than to change IV, 52, 1, divo adarsi duhitd, into adarsi duhita diva? or X, 118, 6, addbhyam grzhapatim, into grihapatim adabhyam?
If an epitritus secundus (— v——) had been objectionable in the same place, why not say VI, 61, 10, stomya bhat sarasvati, instead of sarasvati stomya bhdt? Why not VIII, 2, 11, revantam hi srinomi tvd, instead of revantam hi tva srinomi?
If an ionicus a minore (» v——) had been excluded from that place, why not say I, 30, 10, gatitribhyah sakhe vaso, instead of sakhe vaso garitribhyah? or I, 41, 7, varunasya mahi psaraA, instead of mahi psaro varunasya ?
It a dispondeus (- —— —) was to be avoided, then V, 68, 3, mahi vam kshatram deveshu, might easily have been replaced by deveshu vam kshatram mahi, and VIII, 2, 10, sukra Asiram ydkante, by sukra yakanta Asiram.
If no epitritus primus (UV — — —) was allowed, why not say VI, 61, 13, nidas pdtu sarasvati, instead of sarasvati nidas patu, or VIII, 79, 4, dvesho yAvir aghasya Ait, instead of yavir aghasya kid dveshah?
civ VEDIC HYMNS.
Even the epitritus tertius (-— v—) might easily have been avoided by dropping the augment of apam in X, 119, I-13, kuvit somasydpdm iti. 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 (— v v-—) could have been removed in III, 24, 5, sisthi nad stinumatad, by reading sdnumatak sistht nah, and in VIII, 2, 31, sandd amrikto dayate, by reading amrikto dayate sant.
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
Seven the final dipodia of Gayatri verses, I still Gayatra Vrittas. 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*:
I. ur-un-, % VuU-y 3. -Um my, 4 UUm nm, 5. HO,
6. vrnrnm-, J --Vn-,~ 8. -—UuU-,
I do not pretend to give every passage in which these varieties occur, but 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 Gayatri verses, as the Anushtubha verses would have swelled the lists too much.
I have confined myself almost entirely to the final dipodia of Gayatri verses, as the Anushtubha verses would have swelled the lists too much.
§ 2vvyvu-n-. I, 12, 9. tasmai pavaka mri/aya. (Instead of mrilaya, it has been proposed to read mardaya.)
I, 18, 9. divo na sadmamakhasam. I, 42, 43 46,2; 97, 1-8; III, 11, 3; 27, 10; IV, 15, 73
* See some important remarks on these varieties in Mr. J. Boxwell’s article ‘On the Trish/ubh Metre,’ Journal Asiat. Soc. Beng., 1885, p. 79.
But although with regard to the Gayatra, and I may add, the Anushtubha pAdas, the evidence as to the variety Traishfubha and Of their vvittas is such that it can hardly be Gagata Pédas. resisted, a much more determined stand has been made in defence of the vritta of the Traishtubha and Gagata pddas. Here Professor Kuhn and those who follow him maintain that the rule is absolute, that the former must end in u-v, the latter in U—vU-, 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 GAgata lines, we were obliged to repeal all rules of prosody, to allow almost every short vowel to be used as long, and eyery long vowel to be used as short, whether long by nature or by position, we should have gained very little, we
PREFACE TO THE FIRST EDITION. evil
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 vrttas or final syllables of Traishtubha and Gagata p4das, 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 Pratisakhya and P4zini, 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 : Prosodial 1. The vowel o in the body of a word is to Licences. be treated as optionally short :
II, 39, 3. prati vastor usrd (see Trishz. § 5).
Here the o of vasto/ is supposed to be short, although it is the Guaa of u, and therefore very different from the final e of sarve or Aste, or the final o of sarvo for sarvas or mano for manas*. It should be remarked that in Greek, too, the final diphthongs corresponding to the e of sarve and Aste are treated as short, as far as the accent is concerned. Hence &moixot, rénrerat, and even yvapuat, 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, 53, 15. gopa ama.
Hence &moixot, rénrerat, and even yvapuat, 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, 53, 15. gopa ama.
Here the o of gopa is treated as short, in order to get v-—v-— instead of -—v-, which is perfectly legitimate at the end of an Ushzih.
2. The long f and Q are treated as short, not only before vowels, which is legitimate, but also before consonants.
The forms isiya and rasiya in VII, 32, 18, occur at the end of octosyllabic or Gayatra padas, and are therefore
* A very strong divergence of opinion is expressed on this point by Professor Bollensen. He says: ‘O und E erst spater in die Schrifttafel aufgenommen, bewahren ihre Linge durch das ganze indische Schriftenthum bis ins Apabbramsa hinab. Selbstverstandlich kann kurz o und e im Veda erst recht nicht zugelassen werden.’ Zeitschrift der D. M. G., vol. xxii, p. 574.
CVill VEDIC HYMNS.
perfectly legitimate, yet Professor Kuhn would change them too, into istya and rasiya. In VII, 28, 4, even mayt is treated as mAyi (see Trish¢. § 5); and in VII, 68, 1, vitam as vitam. If, in explanation of this shortening of vitam, vihi is quoted, which is identified with vihi, this can hardly be considered as an argument, for vihi 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 Gayatra padas, 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. mahak kshomasya asvind kanvaya.
Here, even if we considered the dispondeus as illegitimate, we might scan kanvaya, for this scanning occurs in other places, while to treat the first a as short before xv seems tantamount to surrendering all rules of prosody.
4. Final n before semivowels, mutes, and double n before vowels make no position*. Ex. III, 49, 1. yasmin visva (Trish¢. § 5); 1,174, 5. yasmi# #4kan ; I, 186, 4. sasmin(n) dhan >.
5. Final Visarga before sibilants makes no position*® Ex. IV, 21, 10. satya samr4¢ (Trish¢. § 5). Even in I, 63, 4.
* Professor Kuhn has afterwards (Beitrage, 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.
IV, 21, 10. satya samr4¢ (Trish¢. § 5). Even in I, 63, 4.
* Professor Kuhn has afterwards (Beitrage, 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.
> 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, e.g. I, 33, 11, though Professor Kuhn would remove it altogether. Kuhn, Beitrige, vol. iii, p. 125.
kodik sakhd (probably a Gagata), and V, 82, 4. sdvié saubhagam (a Gay. § 7), the long i is treated as short, and the short a of sakha is lengthened, because an aspirate follows.
6. S before mutes makes no position. Ex. VI, 66, 11.
ugr4 aspridhran (Trish?. § 3). 7. S before makes no position. Ex. visvaskandrik, &c.
8. Mutes before s make no position. Ex. rakshas, according to Professor Kuhn, in the seventh Mandala only, but see I, 12, 5; kutsa, &c.
vw Y¥ - g. Mutes before r or v make no position. Ex. susipra, dirghasrut.
10. Sibilants before y make no position. Ex. dasyin.
11. R followed by mutes or sibilants makes no position. Ex. Ayur givase, khardif, varshishtham.
12. Words like smaddishfin, &c. retain their vowel short before two following consonants.
We now proceed to consider a number of prosodial rules which Professor Kuhn proposes to repeal in order to have a long syllable where the MSS. supply a short :
1. The vowel 7 is to be pronounced as long, or rather as ar. Ex. I, 12,9. tasmai pavaka mrilaya is to be read mardaya ; V, 33, 10. _Samvaranasya risheh is to be read arshek. But why not samvaranasya rishek (i.e. siarsheh) ?
2. The a privativum may be lengthened. Ex. agarak, amritah.
3. Short vowels before liquids may be long. Ex. narah, tarutA, tarati, marut4m, harivah, arushi, dadhur iha, suvita (p- 471).
4 Short vowels before nasals may be lengthened. Ex. ganAn, sanitar, tands, upa nah.
5. Short vowels before the ma of the superlative may be lengthened. Ex. nritama.
6. The short a in the roots sam and yam, and in am (the termination of the accusative) may be lengthened.
cx VEDIC HYMNS.
vvue
7. The group ava is to be pronounced aua. Ex. avase becomes auase; savitd becomes sauita; nava becomes naua.
8. The group aya is to be changed into aia or ea. Ex.
wv ve nayasi becomes naiasi. g. The group va is to be changed into ua, and this ua to be treated as a kind of diphthong and therefore long.
10. The short vowel in the reduplicated syllable of perfects is to be lengthened. Ex. tatanah, dadhire.
11. Short vowels before all aspirates may be lengthened. Ex. rathas becomes rathaz; sakhA becomes sakha.
12, Short vowels before h and all sibilants may be lengthened. Ex. mahini becomes mahini ; usigam becomes usigdm ; rishate becomes rishate; dasat becomes dasat.
13. The short vowel before t may be lengthened. Ex. vagavatahk becomes vagavatah; atithi becomes atithiz.
14. The short vowel before d may be lengthened, Ex. udaram becomes udaram; ud ava becomes ud ava.
15. The short vowel before p may be lengthened. Ex, ap4m becomes ap4m; tapushim becomes tapushim ; g7‘hapatim becomes grthapatim.
16. The short vowel before g and g may be lengthened, Ex. sAnushag asat becomes sdnushag asat; yunagan becomes yunagan.
Let us now turn back for one moment to look at the slaughter which has been committed! Is there one single rule of prosody that has been spared? Is there one single short syllable that must always remain short, or a long syllable that must always remain long? If all restrictions of prosody are thus removed, our metres, no doubt, become perfectly regular. But it should be remembered that these metrical rules, for which all this carnage has been committed, are not founded upon any a priori principles, but deduced by ancient or modern metricians from those very hymns which seem so constantly to violate