Professional Documents
Culture Documents
Accent in Sanskrit - Abhyankar
Accent in Sanskrit - Abhyankar
ACCENT IN SANSKRIT Author(s): K. V. Abhyankar Source: Annals of the Bhandarkar Oriental Research Institute, Vol. 50, No. 1/4 (1969), pp. 4155 Published by: Bhandarkar Oriental Research Institute Stable URL: http://www.jstor.org/stable/41694274 . Accessed: 15/06/2013 16:32
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp
.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].
Bhandarkar Oriental Research Institute is collaborating with JSTOR to digitize, preserve and extend access to Annals of the Bhandarkar Oriental Research Institute.
http://www.jstor.org
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
ACCENT
IN SANSKRIT* By
K. V. Abhyankar [ 1 Field of accent not sufficiently explored. 2 Recital first : 3 has accent as its essential reflection afterwards. Language factor. 4 Stress accent in the original Rgveda hymns. 5 Utterance of vowels and consonants. 6 Five types of accents. 7 The syllabic accent. 8 The Sentential accent. 9 The Metrical accent. 10 The modulatory accent. 11 The Musical accent. 12 Accents in compound words. 13 Referencesto the syllabict the sentential and the metrical accents in the Astdhyyl. 14 References to the modulatory accent. 15 Referencesto the Musical Accent. 16 The original verse text of the Rgveda. 17 Peculiar features of the . original text. 18 frfffojtn for thepreservation of the original text s 19 Scrutiny of the and the 20 Father E steiler1 . theory. 21 Critical observations on Father Esteller's theory 22 Comparison of the Father Etellers restoration, and the 24 Tramsir re-reading. 23 The Re-reading of the stanza position of words or lines impossible. 25 Importance of accents and pauses. 26 The five important factors for the re-reading of the Samhit text.]
3FigTqi rt sfiil:
n swngiTTCt il ||
im 3TT^*tM FITT3FT
1 Field of accent not sufficientlyexplored : It has to be stated first that accent in Sanskrit is a field which is not yet explored or rather not sufficiently explored by * R. G. Bhandarkar Address deliveredat the Bhand&rkar O,R. Anniversary on 28th 1968. Iostitute, Ripacam, August, B. 0. R. I. J [ Annals,
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
42
Sanskrit scholars of Veda and Vykarana. There are many scholars who are of opinion that accent is a factor relegated to Vedic Literature only, and there is no accent in Sanskrit Language at all. Many authors of Alamkra works have maintained this dictum fearingthat the abdalesa figure of speech would disappear in case accents are given. Foreign scholars have done a good deal of work in trying to interpretthe sktas of Veda, but they do not appear attention to accents. They have dealt sufficiently to have paid sufficient with Vedic metre. They have scanned the lines of many sktas and tried to findout the iambic rhythmin them, but they have not taken into consideration the role which the acute accent or udtta has played in the moulding of the original Rk verses. The reason possibly forthis is they have not heard the traditional Vedic recital going on at differentplaces in the country on differentoccasions, nor they have committed to memory a sufficientnumber of Vedic stanzas in the different metres with accented syllables. Recital first , reflection afterwards : Recital first and reflection afterwards was the secret of the traditional method of studies and many scholars of Vykarana till recent times had a number of Veic hymns committed to memory many Vedic memory. I was required to commit to mantras and many Vykarana Stras when I received my first lessons orally in those subjects from my maternal and spiritual grandfathers. Later on, when I studied Veda and began to teach in some portions of it, the long standing riddle as to what ijhft in and TT STTcTTT iftf |pn 3J4N4I*pTT TT^TT^W could mean was solved easily by looking to (RV. 1.167.4) It became clear that the unnecessary prolongation of the Padaptha. in cTRTT an(i *Tm ^TT^TT^TT, the svarita vowels resulting into an awkand mmWM, in the middle of the words ward pause at m and of of the line. the sense in a presented difficulty the easy understanding Immediately a thought arose in my mind that in the ancient Vedic manner with the vowel times the recital must have been in a different udtta uttered with a kind of stress, and pauses given at the end of words as strict metrical considerations required them. I have been thinkingover the subject for several years and now I am taking this opportunityof placing my thoughts beforeyou in the matter. I frankly say that I am open to reconsideration and correction. Now, let me turn to my proper subject of this evening - accent in Sajisknt, 2
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
Accent in Sanskrit 3
43
Language has accent as its essential factor : It is a proposition universally admitted that every spoken language of the voice which is called tone is charcterized by a sort of inflection or accent. Sanskrit is not now a spoken language, and, if it is found in use at present, it is only in the debates, discussions and discourses of the Sanskrit Shastris and Pandits. The language used by them is, however, based more or less upon their study of the existing vast classical epic and Shastric Literature wherein no accents are found. Stress accent in the original Rgveda hymns : In the very ancient days of the seers of the Yedic hymns as Sanskrit was a spoken language at the time, the inspired hymns of the Rgveda must have been recited with a stress or emphasis on almost all words in a sentence excepting the unimportant ones and the expletives f and the like. In the language in is those as spoken days generally the case with spoken languages, used sentences made up mostly of simple words of not more small people than three or four syllables in which only one syllable was accented, the . rest being left unaccented. This accent of the nature of stress was named udtta , and was looked upon as a propertyof vowels ( ). Consonants which were looked upon as mereappendages to vowels, could not be accented by their very nature and they were not accented. In popular language, however, the accented vowels were called udtta , and the unaccented ones, in contrast were called an-udtta i. e. not-udtta. The term udtta is translated into English by the word acute. The term ( udtta ) is definedas by Panini as also by the authors of the Prtikhya works. The word does not mean loudly pronounced. The author of the Mahbhsya has given in explanation the following line showing clearly therebythat on the pressure sound-producing organs, hardness,and contractionofthe are the causes of the acute accent which, in short, is the tone gullet attended with stress. In contrast, the term anudtta is explained by the words which show that looseness of the sound-producing organs, softness,and widening of the gullet are the causes of the anudtta or the grave tone. It is worth-noticingthat the term svarita is not explained in a similar way by showing any peculiar characteristics of that tone, but it is defined as merely a combinationof the udtta and the anudtta tones. Utterance of vovjels and consonants : ' to The word r? which is derived from the root ^ produce sound,' literally means sound in general, which is of the nature of air-vibrations falling on the tympanum of the ear. Air, 5 4
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
44
breathed out from the lungs, on its way to the mouth, gets a produced on it by the vibrating laryngeal cord at the end of wavy effect the wind-pipe. When this vibrated air, beforeescaping out of the mouth merelypasses beside the five places of articulation without touching them, it produces the vowel-sounds. If, however,on its way it touches the places of articulation, an additional sound-element becomes attached to it which is very helpfulin the productionofarticulate sound i. e. speech (srr^O, which is a speciality of human beings by means of which they easily communicatetheir thoughts to others. This additional element is the consonantal sound. Rightly the author of the Mahbhsya has described the complete dependence of the consonants on vowels in the wordsi crtoti *Rrrit i ix m r<r qt ^rfrfir I MBh. VI. 1.2. The name svara is given to primary sounds or YFr? air-vibrations which are not affected by places ofarticulation. The effect on air vibrated at the the produced places ofarticulation is called a consonant or Vyajana. Panini in his grammar,has used the short terms and ^forvowels and consonants respectively. He has used the term no doubt, but it is used in the sense of accents thrice, and once in the sense of sound in general. Earlier writers viz. the authors of the Prtikhya works as well as later writers viz. Patajali and others have used the terms ^ and 3-37^in the senses of vowels and consonants which now are current. The term 3131^ is used in the sense of a vowel with or without a consonant i. e. a syllable, while the term ^ar is used fora letter or phoneme which is produced with a single effort. 6 Five types of accents : Accent of the type of emphasis is a peculiar feature of a spoken language especially in conversations and addresses. In a sentence, with a view to conveying the main idea, the accenti which is suitable to carry home the intended sense, is naturally used by the speaker and many a time an accent of the type of intonation is very helpful in conveying the right sense. This intonation has a variety of types, of which the syallabic, the sentential, the metrical, the modulatory and the musical accents are the chiefones. The syllabic accent : The syllabic or the stress accent is more or less governed by the sense element of a word. In a sentence, words which are important in sense, are emphasized, and in each of those words also, that syllable is stressed which possesses a comparatively more important sense factor. For instance, in the $<^<3 and the words the vowel of the affixis udtta as in sotto: , and the like ; but, if the sense of the base predominates,the base retains its 7
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
Accent in Sanskrit
45
> accent, as beforethe affixesmarked with mute 5^ or ^ as in 1 nouns with the case affixes, the base retains its accent, while the case-affix is unaccented, as the case-relation, which is easily supplied by the vicinity of two related words is, comparativelyless important. It is worth noticing in this connection that in compound words the case-affixis dropped. 8 The Sentential accent : The sentential accent is not much differentfrom the syllabic accent. Instead of the importance of a syllable in a word, it is the importance of a word in a sentence that is at the root of the sentential accent. As a result, sometimesthe whole word and others remains unaccented as for example, the expletives or verbs placed after nouns or the short forms of the pronouns or nouns in the vocative case, if not placed at the beginning and of a sentence. A verb preserves its accent at the beginning of a sentence or a verse-foot as also in the conditional and the relative clauses, of course because, it has got there a sort of importance. 9 The Metrical accent : The metrical or poetic accent occurs at the end of a word when there is a pause in the course of the recital of a verseline. The vowel at the pause is a long vowel. If it is not a long one, it is lengthened in the recital. Many times, verbs and indeclinables, ending in short vowels have got their final vowel lengthened. Generally, the metrical accent sets aside the natural stress accent in a word if they come in conflict. The case is otherwise in the by original Rgveda where the stress accent or udtta is never affected the metrical accent. The pauses change their place, but the udtta accent remains where it is. In the later Sanskrit poetryhowever,wherein the pauses and the sequence of short and long vowels in a verse were rigidly fixedby means of the eight ganas of three syllables in each gana or by means of mtrs, four mtrs generally making a unit, the stress accent came in conflict with the metrical accent and the result was that the former was totally ignored in poetry. For example see or the line sr^rg which the stanza ^ in as of four also units mtrs the stanza have got syllabic succession, and the line ffer ^3% whereinthere is the iambic rhythmmade up by short and long syllables in succession. The modulatory accent : The modulatory accent or accent accompanied by a modulation of the voice, is found in conversations where the different emotions 10
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
46
are conveyed by means of peculiar changes in the voice. It is not possible to find it in the recital of the various Veda texts as they are the records of the glorious exploits and high thoughts of the ancient seers. These modulatory accents were, however, used in the Sanskrit language currentat the time, and they are found echoed in the Jater story and dramatical literature. The modulatory accent is given the general name which is definedas The stanza forer Wit *T JRJrffT f.SlTOTW wf ^3 f q<rn f 1 II
T^?rr^qf;
in the Vensamhra and the stanza ^ m the Uttararmacarita furnish good examples of this modulatory accent. A verb in the interrogatorysense without any interrogativepronoun is uttered also in this modulatory tone when the verb generally stands at the beginningof a sentence. The famous lines, " Breathes there the man with soul so dead Who never to himself hath said " This is my own, my native land. furnish also an example ; the lines could be rendered in Sanskrit as h srfr ^ns^r srrw i *nf mfljiRr 11 Musical accent : The musical accent has got its origin in the Samhit texts of all the Vedas wherein the various verses of the Rgveda are recited like prose passages without any serious attention paid to are their metrical form. As the various rules of coalescence or tffr, and in Samhit these sometimes observed texts, corresponding strictly vowels, added beforethe semivowels forthe sake of metre,are removed, *TT ) they are rightlygiven the name Samhit ( fsT:pjt:STCTT For the preservation of the text intact with every syllable distinctly uttered, the Samhitkras uttered the vowel preceding the udtta vowel as anudtta with a lower pitch, while the vowel succeeding the udtta was uttered as svarita with a higher pitch, and the rest were left practically unaccented ( ekaruti ). These three accents with a and hence, they can be differencein pitch, produced a musical effect, musical accents. called 12 Accents in compound words : All these five kinds of accents appear to have been referred to in the Sutras of Panini. There are in the Astdhyy of
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
Accent in Sanskrit
47
Panini 375 stras on accents of which more than half deal with accents of compound words. Panini has laid down a general rule that the last syllable of a compound word is accented udtta ; while Patarijali has laid down that the first member in the Avyaybhva, the second in the Tatpurusa, none in the Bahuvrlhi and both in the Dvandva compound occupy the principal position syntactically. Naturally, the principal member is expected to retain its accents leaving the subordinate member unaccented i. e. with all the vowels uttered as anudtta. But, such is not the case. It is not an invariable rule that the principal memberof a compound has got more importance attached to it. Importance depends upon the will of the speaker which is governed by circumstances, and many times an adjective or a subordinate member has got more importance, and it is emphasized. Panini has made a thorough study of accentuation in the Sanskrit language which was actually spoken in his days, in all the different provinces, and has given a very long list of deviations fromthe general rules which state that the Tatpurusa Compound has the acute accent on the last syllable, while the Bahuvrlhi and Avyaybhva retain the accents of the first member. The total number of rules discussing the accents in compound words, is two hundred out of three hundred & seventy five. 13 Referencesto the syllabic, the sentential and themetrical accents in the Astdhyy : Out of the remaining 175 rules, dealing with accents of single words, 72 rules deal with the syllabic accent udtta , which is given on the base or the affix as Remanded by the sense-factor. There are 55 stras treating the sentential accent which is of two kindsthe anudtta accent for the whole word known as ( 1 ) on the word having comparatively small importance in the sentence, and ( 2 ) st^TtT accent on the firstsyllable of a word when the word has got a special importance in the sentence. A separate section treating the sentential accent is given under the heading ciT^T^ by some authors of independent treatises on accents, the prominent among which are and the like. The metrical accent is not directly referredto in any rule of Panini. There are four stras and ^ (VI. 3, 133-136 ) which prescribe the lengtheningof short vowels at the end of some indeclinables and verbal formsformetrical purposes. This lengthened vowel, when standing at the end of a verse-footor markinga pause, has got a special additional tone for the sake of rhythm, which is called the metrical accept.
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
48 14
References to the modulatory accent : The twentyrules prescribingthe prolation or protractionof vowels, known as gf^r, referto the modulatory accents. They are given by Panini in the second quarter or pda of the eighth adhyya afterthe treatment of word-formation is over. This fact has a kind of significanceviz. that these accents are in a way supplementary. The udtta accent and the svarita accent in a few cases, are prescribed by Panini for all prolated vowels showing therebythat prolation does not come in conflict with the acute or the svarita accent. There are no instances of prolation o? jpluti in the Rgveda. In the spoken language, however, in those days, the final vowels of verbs as also those of nouns in the vocative case were many times prolated and given the acute accent also. Instances of pluta vowels are foundin treatises dealing with sacrifices as it is in the actual rituals that the pluta vowel is heard. There are some rules prescribingthe prolation of a vowel along with the svarita or udtta accent on occasions of question, anger, rebuff, censure, wonder and other emotions. This udtta or svarita is accompanied by an additional modulation of the voice. Instances of this modulated acute or svarita accent are found only in the spoken language and Panini's treatmentof this accent can safely be taken as a powerful testimonyto show that Sanskrit was a spoken language at his time. References to the Musical Accent : The rules dealing with musical accents are, in the firstplace, the same as those which prescribe the syllabic accents, as it is, in fact,these very accents which, when the text is recited in a differentmanner, as noticed in the present traditional recital of the Vedic Samhits, become musical accents. In addition to the sutras prescribingthe three accents the four stras giving the definitionsofthe three accents ( P.I. 2. 29-32 ) and the eleven stras stated as applying to the Vedic Samhits only, referto the musical accents. There are some rules dealing with ekasruti ( P. I. 2. 33 and the following) which is looked upon as a fourth accent by some scholars of Veda. These rules can be said to refer to musical accents. The word f^rpnn, used by Panini in the rules prescribingthese musical accents in contrast to the word 5^% in the the rules prescribingthe lengtheningof the short vowel forpurposes of metre,furnishesa good testimonyforbelieving that there was a strictly metrical original text of the Rgveda of which the present Rksamhit is a re-reading. 15 The original verse text of the Rgveda : The Rk Samhit represents the oldest record of articulate sounds or speech in verse-form. It is a well-known fact that a verse or a song, on the jnijid of the listener; when heard, produces very easily a deep effect 16
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
Accent in Sanskrit
49
and hence it is, that verses or songs ar remembered veryeasily, especially when they are recited with pauses and caesura at regular intervals. The poets ofYedic hymnsgave out theirinspired utterancesin verse,with accents of the type of stress on a word, or a part of a word, the sense element of which they wanted to emphasize. The hymns were metrical consisting of three or fourfeet which were made up of a definitenumber ofsyllables with short pauses at the end of one, two or three words. The feetconsisted of 8, 11 or 12 syllables, the last fourof which made generally a couple of iambuses or a couple of one trochee and one iambus, especially in a foot of 11 syllables. The inspired verse of the Yedic poets, in short,contained ( 1 ) a definitenumber of syllables in each of the three or fourfeetof the stanza, ( 2 ) short stops at the end of a word or two, and (3 ) a regular cadence at the end of each foot. As words ( simple as well as compound ) used by the Vedic poets, contained many times more than three or four syllables, and as a word could not have more than one acute-accented syllable, it was not possible to have a rhythm of the trochaic or iambic type based on accent, as in the English poetry of Chaucer, Milton, Scott and others ; but the rhythmwas made up of a regular succession of short and long syllables which was sometimes foundin the whole footas in There are many instances 9rrar:* of this rhythmin the later classical poetry as for example in or . 17 Peculiar features of the original text: In their inspired utterances of the verse-feet of the hymns, the rules of co-alescence were many times ignored by the Yedic poets in the case of two vowels in juxta-position, while sometimes a corresponding vowel was added before a semivowel for the sake of metre. This addition has been noticed in the Prtikhya works under the name Svarabhakti, while the non-coalescence of vowels was termed or On rare occasions a diphthong vowel was split up into its two componentvowels or a long vowel was split up into two vowels, but it is remarkable that this splitting was done in such cases only where the long vowel or the diphthong had resulted from a combination of two vowels and not promiscuously anywhere. 18 for thepreservation of the original text: In course of time after the lapse of some centuries, the original Vedic hymns, which were recited in a strict metrical form with the stress accent called udtta on the important syllables, lost their original fervour. They assumed the form of historical records of the sublime thoughts of the ancient revered sage-poets, which posterity wanted to preserve as a sacred valuable trust. Yskcrya has recorded this preservationof the sacred lore in the oft-quoted stanza in the Vedi Tristubh metre : B. O. R, I. J 7 [ Annals,
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
50
i fsn s IfmimmsRm tim m H * TTT ^TT ^R<T The necessary steps fortheir preservation were taken : The hymns were of arranged in different groups as became suitable forthe performance the various sacrificesand their text was made similar to the running text of the later explanatory prose works such as the Taittirlya , the Maitryam and the Kthaka samhits and the various Brhmana works by combining together the vowels in juxta-position and removingthe vowels f, 3Tand sr which had been added forthe sake of metrebeforethe and The process of this combinationof two vowels semi-vowels in juxta-position and the removal of theadded vowel was called and the resultant running text was called the Samhit text with which names of sages like aunaka, Kyapa and others are found associated. A prominentfeatureof the Samhit formof the original metrical hymns was the changed manner of their recital in which every syllable was recited with the same force, and as a result, the element of stress was practically ignored. As the several syllables had to be distinctly uttered, the vowels beforeand afterthe udtta were uttered with the same force, but with difference in pitch. : The anudtta vowel was uttered with a low pitch, the udtta with a higher one and the svarita with a still higher one. The Samhit was thus characterizedby three accents which could be called musical accents. These accents later on developed into the seven and the udtta musical notes, the anudtta being the same as and the svarita the same as q^rsr, and the same as and This is shown by the following stanza :n ^ ) ( nfatfNreisfrT
The samhit text which has come down to us is, as shown above, only a manner;it cannot presentationof the original metrical text in a different be called a revision or redaction of the old text. : and the 19 Scrutiny of the ' ' If the well-known hymnknown as the Purusa-skta in the Rgveda and the hymnknown as the Pavamnaskta in the Yajurveda , as theyare recited in the Samhitptha to-day, are examined in the light of the remarks made above, it will be noticed that every foot is made up of eight syllables, of which the fifthand the seventh syllables are generally short and the sixth and the eighth are generally long, making a couple of iambuses. In a few cases the Samhitptha has seven syllables and very rarely nine in a foot,and there is no iambic rhythmin the last four syllables. It is possible, however, to read eight syllables in these feet a technia by reading a correspondingvowel before semi-vowel, process
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
Accent in Sanskrit
SI
by ancient scholars of Veda. In a few cases a cally known as coalesced vowel can be read as two vowels-a process which is technically These cases are, of course, very few. There are only seven called svarabhakti and six cases of vislesana , in the whole Purusacases of skta consisting of 64 feet.Similarly, there are nine cases of svarabhakti and only one case of vislesana in the Pavamnaskta of 65 lines. As far as the iambic rhythmis concerned,it is not noticed at several places and it caunot be broughtabout by the svarabhakti or vislesana processes. Other types of rhythm are, however, noticed provided the recital has the udtta syllable uttered with a stress accent and regular pauses are given where necessary, but always at the end of a word and never in the middle of a word. The need of an additional syllable for such Samhit feet as have seven syllables, could be fulfilled by following the process of svarabhakti and reading as as Hlf^TT: *nTT: as as as as as and even ftpTTO By followingthe process of could be read as as ^ and and as as ^ snSr*, wren?* fibroma, sraraiarar as swnPrand as Sometimes, although rarely in order to STFR shorter: reduce the number of syllables fromnine to eight, two syllables have to be read as one, as for example, into In all these processes a clear influenceof the acute accent is noticeable especially in the matter of the reduction of a syllable. When two short syllables precede or follow a syllable with the udtta accent or stress,the short vowel of one for ^ and syllable is lost as could be seen fromwords like for and and 20 Father Esteller's theory: Father Esteller, a distinguished Vedic scholar, has worked on the problems of the original Rgveda in strict verse, fora number of years and has written several papers and articles on the subject. He names the Rgveda Saihit as a ' palimpsest ' Recently he has given the result of his researches, the restored text in strict verse form of two sktas of Rgveda ( II. 23 and X. 122 ) as a specimen. His observations are given by him in an article named " Problems in the text-critical reconstructionof the Rgveda-Palimpsest ( samhit ) " which is printed as the first article in the Golden Jubilee Volume of theB. O.R.L. He remarks that ' the traditional Samhit text of the Rgveda is called palimpsest ( a written-over text ) because it is redactorially tampered with by the transmitting agency which is here called by the name Samhitkra. In transcribing the Samhit text all the samdhis are dissolved by him, the diphthongs restored to their original formsand the resolved long vowels are given as two short ones, while
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
52
other resolutions are also possible, if demanded by the metre' He has selected two hymns and shown the original verse-text according to him under the heading ' complete correction of text concerningX. 23 and I. 122 on pages 9 to 14, and remarked hat ca careful analysis of the metre proves that the latter in combination with archaism is a secure basis for the text-critical restoration of the Rgveda original text and that both together are paramount as against the Samhit 4 palimpsest ' He has stated that the Samhitkras have superimposed and an artificial orthoepy Samdhi, tampered with the word-order, and even suppressed text words ' modifiedgrammatical forms s theory: 21 Critical observations on Father E steiler1 A simple glance at the restored text given by Father Esteller on pages 8 to 14 is enough to show that in his enthusiasm to see that the iambic rhythmis observed at the end of every line, he has split up almost into all diphthongs into their constituent vowels, as for example on and on. so into into occasions Similarly many ftfg-, 35717, he has split up long vowels into two short ones as forexample, and so on. In this way while into grsTr, into into of a syllable or two in some feet, in the he has filledup the deficiency case of a large number of lines the number of the necessary syllables has been increased unnecessarily by one or two syllables. This has resulted into a serious fault of metre especially so, as he himself has endorsed the standard view of Vedic scholars that the number of syllables must be 11 and 12 respectively in the Tristubh and JagatI metreswhich he himselfappears to have violated here. The retention of the "visarga" everywhere, which he has done, cannot fail to on the ears of a listener,if the latter is to hear produce a jarring effect the recital of the restored text. It is simply beyond imagination to think especially when Sanskrit was a spoken language that the seers of the original Vedic poems recited the verses,as Father Esteller maintains, with diphthongs split up and visarga kept intact. It can only be said that the addition of a corresponding vowel beforea semi-vowel called in justified cases and the splitting up of a vowel into two were resorted to by the Vedic poets, coalesced vowels called as them. Omission of a syllable or more took demanded metre when only of the oral recital but that was in course the occasions on rare place has pointed out that the old text of only by haplology. Syancrya in his was prefissi* fsrs Iwtcft firer? comment on the Father E steller's restoration and 22 Comparison of the sfg^n, : the re-reading As a specimen to illustrate almost all observations made above, the Rksamhit text of stanza I. 122. 6 could be shown below
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
Accent in Sanskrit
53
s A, and its restoration according to Father Esteller could be cited as B, side by side with the re-read text in conformitywith the observations made above given as C. A ( Samhit text ) a if foawjii agtf *fa(.l stg5: rg?if: *[: H $!T B ( Restored text ) T? rawir ST3 81%:3rT: i 33 sto*. 51^5 5: wig^rfir: f tg: arfe: 11 F^rfr il C ( Re-read text ) Si fmrc<uu
metre of 11 syllables with the rhythm The stanza is in the of the at the end made up long short,long long vowels. The third line is defectivein rhythmas also in the numberof syllables, while the last line has foursyllables wanting. The splitting of diphthongs made by Father Esteller in all the seven places in the stanza is unjustified as also unrequired,as thereby there are unnecessarily added 2, 3, 1 and 1 syllables respectively to the four lines a matter which goes against the metre. He has unnecessarily transposed the second and the third of four syllables in the fourth line by lines and supplied the deficiency words the q*; taking sftg occurring in the third line ( his second line ) as understood in the fourthline. Father Esteller has not given any accents in his restored text nor discussed their utility. Does he mean that the original Rgveda poems were without accent ? The Re-reading of the stanza ^ fto: The stanza can be re-read as shown under the heading C stressing the udtta syllables and uttering all the rest as unaccented. The stanza is in Tristubh metre with 11 syllables in each line. It is and separately in the firstline possible to read the words and to look upon that line as a foot in the JagatI metre. The deficiencyof a syllable in the third line can be filledup by means of in srtg. The root is changed into ^ svarabhakti by reading as srft of Guna vowel the substitution the arwith attached to it. As the by vowel sirin the word gsrig: is udtta a with stress, the preceding vowel g-can be recited as a short vowel although followed by the conjunct consonant 3^ and the tristubhrhythm can be secured. The deficiencyof foursyllables can be filled up by reading the word gshj: twice. It is quite natural that, in the Samhit recital the word gsftg:at the beginning of the fourth line was lost by haplology. The deficiency of a syllable still, even after the lost word gsiVg:has been supplied, can be as ggNrsu,as an is the case-affix and hence, filledup by reading can be read as detached fromthe base 23
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
54 24
Transposition of words or lines impossible : Regarding transposition of words by the Samhitkras as alleged by Father Esteller,.it can be said that no transposition of words, not even that of syllables, was possible in a text which was handed down by oral recital from generation to generation for several centuries. The oral recital was so perfect in ancient times that even a single error, committedby a pupil in his recital of the Veda, was severely dealt with by the preceptor as could be seen by Patajal''s remark qin tlie ire of in of the context the Yedic recital. very Mahbhsya perfectness The riciters also, who preserved the Veda were men of first rate intelligence and brilliant scholars of Veda and Vykarana like Balashastri Ranade of the 18th century whose strong memory and sharp intelligence are attested well by the stanza often quoted by Pandits of Vrnasl: ^ iwsnfrwnflr T sn f^TT fcnfi fa gia'umtfril srrcifasTRre Importance of accents and pauses : Accent and pause played a very important part in the ancient Vedic verse in controlling the rhythm and producing a musical effect. If GyatrI was the first metre as is supposed by many Vedic scholars, there was a pause generally at the end of four syllables, and the last four syllables generally formed two iambic feet or a trochaic foot first and the next an iambic one. Sometimes there was a pause at the end of the third or the fifth syllable. The. pause was marked always by a long vowel. If the vowel at the pause happened to be a short one, it was lengthened. The foot of the Gyatrx metregave rise to the Tristubh and JagatI metres by the addition of three and four syllables respectively. There was rarely a pause in the middle of the last three syllables in the Tristubh and the last fourin th Jagat. The accent udtta which was of the type of a stress, helped the understanding of the sense which could be grasped by the listeners, while the solemn accented recital' with regular pauses at the end of three to five syllables, was going on. The udtta accent many times resulted into the loss of a vowel, especially when the accented syllable in a word had beforeit or afterit two syllables possessed of short vowels. The loss was unavoidable as the stress accent, which was based on the importance of the sense, could not be shifted. and others with the accented vowel a* at the The words as cases where the omission of the vowel be mentioned could beginning 25 i
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions
Accent in Sanskrit
55
and others is noticed. Similarly, the ending vowel 9Tof the roots krt affix^ is lost when that affixis followed by of the non-accented 3T and preceded by a root which has got the the non-accented affixti; acute accent of the affix^ shifted to it. See for instance the words and others. In the same way, causal roots have the vowel in lost when they are followed by the affix^ as could be seen sr and the like. from the forms The five important factors for the re-reading of the text: A mere mechanical scrutiny is not of much use in producing research-work,especially in fixingan original text from a derived one. A firmgrasp of the spirit and mental attitude of the original author is an essential factor for interpreting and explaining that author. For example, in fixing the extremely obscure text of Bhartrhari's Mahbhsyadpik Acharya V. P. Limaye and myself had now and then to guess what Bhartrhari had in his mind when he wrote the lines. The Yedic poets had certainly the sense element of what they composed prominentlyin their mind which they wanted to convey to their listeners. Poetic expressions convey not only some sense, but a charmingsense preponderated by emotion, which makes the poets emphasize the important sense-factorsby putting a stress accent on them. This stress accent ofthe Vedic poets was the udtta which was given on only one syllable in a word,the other syllables remainingunaccented or anudtta . The pr imary Vedic poetry is to be re-read from the present Samhit and not to be reconstructed ; and in doing so, the important factors such as the accent and the pause cannot be ignored. Of course, side by side with these, the superb and majestic tone of the recital also, which is being heard at present in different places on certain occasions only, has to be carefully considered. If the traditional text or palimpsest is re-read with as much consideration given to the stress-accent and pauses as to metre,rhythmand archaism, many of the difficultiesin re-reading the text are sure to disappear. An attempt in this direction is certainly worth-making by modern Yedic scholars of erudition and a team foraccomplishing this deep study, preferably through forming task, 1 i i afr 55 *tr i s up-hill r m 26
This content downloaded from 202.41.10.3 on Sat, 15 Jun 2013 16:32:55 PM All use subject to JSTOR Terms and Conditions