kuvalayamala1

8/13/2019 kuvalayamala1

1/7

Gleanings from the Kuvalayaml Kah I: Three Fragments and Specimens of the EighteenDesabhssAuthor(s): Alfred MasterSource: Bulletin of the School of Oriental and African Studies, University of London, Vol. 13,No. 2 (1950), pp. 410-415Published by: Cambridge University Press on behalf of School of Oriental and African StudiesStable URL: http://www.jstor.org/stable/609283 .Accessed: 13/08/2011 08:25

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 formsof scholarship. For more information about JSTOR, please contact [email protected].

Cambridge University Press and School of Oriental and African Studies are collaborating with JSTOR todigitize, preserve and extend access to Bulletin of the School of Oriental and African Studies, University of London.

http://www.jstor.org

8/13/2019 kuvalayamala1

2/7

Gleanings from the Kuvalayam_ala Kahi IThree fragments and specimens of the Eighteen Desabhasas

By ALFRED MASTER

?1

SHORT account of the Kuvalayamala was given in "An UnpublishedFragment of Paisaci ", BSOAS., xii, p. 659 foil. It is dated A.D. 778 and

is known from two MSS.-J dated A.D. 1083 and P undated, of which P 2 isa copy. The bulk of the work is written in Pdyaya-bhdsd, .e. Jaina Maharastri.

Other styles are used, one of which is Avabbha.msa, which Uddyotana, theauthor, describes as Sakkaya-Pdy'ubhaya-suddh'dsuddha-paya-samataramga-ramgamta vaggiram " A spoken language rippling with even waves of Sanskritand Prakrit words correct and incorrect ". It is therefore not the classical

Apabhramsa of Hemacandra and others, which does not contain Sanskritwords apart from the tatsamas or words common to Sanskrit and Prakrit.

?2The first and longest fragment is printed by L. B. Gandhi in GOS. 37 (G.),

Intro.,p.

109. Thespelling

is much lessregular

than in the Prakritprecedingand following it, and cannot be ascribed for the most part to clerical errors.

The two MSS. do not differ widely. For reasons into which it is unnecessary toenter here it is clear that J and P were copied from the same MS. and neitheris dependent upon the other. P 2, although an inferior MS., often usefullysupplements J and in the present extract such aid is particularly welcome, as

owing to the habit of readers putting a finger on the text where difficultiesoccur, much of the J palm-leaf has been smudged and would be almostunreadable without the help of P 2. The rotograph copy is therefore unsuitablefor

reproduction.As I have not

always acceptedthe

readingsof

G.,the

passageis reproduced here with some more lines by way of introduction.

Bhaniyarm Mdydiccena "Aho, gdma-mahattard mahdpdvam mae kayamnmittadojjha.m dma, td aham jaliam kudsanam pavissdmi. Deha majjha pasiahakatthdim alanam va " tti. Tau bhaniyarn ekkena gdma-mahattarena 1) Eduehau, dummanas-sahu, savvam je thuj'aaridu, tujjhanau vamka valitaum

paraddhaum. Etu praim sugai, bhratu-vara, bhramti sampratu. Tau annenabhaniyam (2) "Thuja viraidu dhana-lavaya suha-lampade, etu praimdutth'attha-mana-moha-luddhaum. Tumr samprati brolitaum. Etu etu

praraddhubhallaum. Tau

a.nnenabhaniam

cira-jara-junna-dehena 3)Ettha

sujjhati kira suvannam. E vaisanara-muha-gataum, kaum pau mittassavamcana ? Kamalia-vrata-dharane etu pau tujjhe ppanahiya ". Tau sayala-Dramga-samind bhaniam jettha-mahi-mayaharena, (4) "Dhavala-vahana-dhavala-dehassa siri bhrameti; ja vimala-jala-dhaval'ujjala, sa bhadari-yati-Gamga pravesi tuhum, mitradrojjhu to nama sujjhati."

8/13/2019 kuvalayamala1

3/7

GLEANINGS FROM THE KUVALAYAMALA KAHA I

" Mayaditya said ' 0 village-elders I have committed a great sin, that isto say, betrayal of a friend, therefore I shall enter an evil place aflame. Save

my body from pains and burning.' Then a village elder said (1) 'This is so,dejected sir. All indeed is thy action. Thy destiny has turned awry. Therefore

thy blameless life, excellent brother, is now confusion.' Then another said

(2) 'What thou hast done is for a little wealth from lust of happiness;because of this thou wert seduced by the mental lure of wretched riches.Thou hast now transgressed. This, this is thy destiny, my friend.'Then another with a body withered by old age said, (3) 'Surely goldis purified here 0 thou who art the mouth of the Supreme Being, howis the deceiving of a friend a sin ? By taking the vows of a Kapalika, let

this sin of thine be destroyed.' Then spoke the Lord of all the Drangas, thesenior High Elector, (4) ' The sheen of the white body of a draught bullock is

dazzling. Do thou enter that venerable Ganges of ascetics, which gleams withthe whiteness of pure water, then indeed thy betrayal of a friend is cleansed

away.' "

(1) ehau: cf. Hem. 4. 362 eha, eho, ehu and O.Guj. ehavau. dummanas-sdhu:so G., but dumma- is not clear. P 2 has houmanussdharm. The true reading is

possibly bhoumanussdu human life of enjoyment; du, life, is neuter like savwam.

je is a padapuran a : Hem. 2. 217. thuj': G.'s reading, but uncertain; if correct

it derives from Pkt. tuhya (Pischel transliterates tuyha), a form of Gen.-Abl.based on Skt. mahyarm Pali has tuyhamn at.-Gen.) which passed into *tuhuya,*tuhhuja, uja, cf. Marwari, Mewari thum from tuhum and Guj. tuj from *tuhuja.tujjhdnau: Gen. with -nau = O.Guj. -nau (G.). sampratu: cf. Pkt. sampayam;P 2 sa.mprai.

(2) thuja: G. has thuje, but the karna of j is very faint; P 2jamji. lavdya:G. reads lavdsd ta and glosses lavasayd tathd; J lavdsd va (or ta) with sd can-

celled, P 2 lavasae, which gives good sense (with the desire for), but is too easy;dsdya is a regular Pali idiom with the Dat. to express longing (W. Geiger, Pali

Literature and Language, p. 117) and lavadsya is perhaps correct. samprati:the Skt. form. brolitaum: Pkt. bolai sink, pass away, bolae transgress(Ratnachandraji); G. glosses abravth, but in that case we should expectbrollitaunm, and the sense would be poor. The anusvaras may be due toattraction to turz, but the use of anusvaras is everywhere erratic.

(3) e: G. reads re as P 2, but e is clear and has the same meaning. kdmdlia:J has komalia tender lady, with the karna over the matra of ko faint; possiblykdpdlia should be read, as P 2 has kdvdliya. ppandhiya (nda > ndh) is notrecorded, but can be paralleled (Pisch., PG., ? 264): G. glosses pranaiksyati,but the form seems to be a 2 sg. impv.; P 2 is wildly astray with eueueu sujjhe

jjandhim. Dra in Dramga is indistinct; P 2 just ga. mayaharena frommatadhara-, grdma-pradhdna Hargovind Das T. Sheth, cf. Guj. matadar coun-cillor, voter, matdddr Persianized from matadhara) a village elector to the postof patel in certain Gujarat villages; the word should mean "property-holder"(Persian matd'ddr), but the qualification of matadars is not from property.

411

8/13/2019 kuvalayamala1

4/7

ALFRED MASTER-

bhrameti: J bhramiti corrected by later hand, from bhramita; P 2 bhramati,which may be right. bha.ddr : an Apa. word. prdvesi : a denominative.

Theonly

clue to thelocality

of this scene is the mention of theDramga-svami. If Dramga is the correct reading, it may refer to the Dangs, the hilly

forest region between the Surat and Nasik districts,l and Gandhi's opinion,based on the use of -nau and bhallau, that the language is that of the Gujjaras,will not be disturbed thereby. The word dramga occurs in the Kharosthidocuments and is discussed by T. Burrow, Language of Khar. Doc., p. 98.He interprets it as " office ", but cannot reconcile this meaning with dramgaof the Rajatarangini and udranga in Kuttanimatam and suggests that the

original meaning may be " fortified place" or "elevated structure ". The

latter meaning would not be too remote from that of Dang, especially as raigais rather " arena " or " area " than " structure ". Udranga is also used in theValabhi copper-plates of A.D. 605-10 (e.g. JJBRAS., N.S. i, 39) in the phrasesodraigamr oparikaram, where it seems to mean a tax, and drdngika in the

phrase -viniyuktaka-drdngika-mahattara-cdtabhata- eaning a particular official.The language contains several Apa. features, which are common with Old

Gujarati. Peculiar are the retention of Skt. ai as ai in vaisanara and the

irregular alternation of -t-, -d-, and zero. P 2 has also the y-sruti and the

irregularity operates for different words, so dyarium, valiyaum, gayaum, vraya;

yadi with d and sugati with t. It is only the dentals which show irregularity, andwe seem here to make contact with the practice of dramatic Sauraseni.Whether we take this extract as an example of Apabhra.msa prose or of

written Gujjara desabhasa, it is unique.

?3The second fragment (G. Fol. 37, Intro. p. 110) is a Doha of the regular

type-6 + 4 + 3/6 + 4 + 1; the first half line ends v L, the second - u, therime being -ei. It is introduced as follows.

Java ya imam cimtei Can.daso mohienam, tdva ima.m giyayam gyyamgdmanadze,

Jo jasu manusu vallahaum, tam jai annu ramei,jai so janai jivai va, to tahu prana laei.

J jdsu, P 2 jasu: these are alternative forms, but the metre requires thelatter. va to: G. reads patto, P 2 vi so; J is clumsily corrected with doubtfulintention.

"And while Candaso was thinking of her with infatuation, a village dancinggirl sang this song, 'He who has a beloved one, if another woos that one, if hekuows and is alive, he will take the other's life '."

" Is alive " " is a true man ".The fourteenth century Ratnaprabhasuri, who wrote an abstract of the

Kuvalayamala, called Kuvalayamala-katha-samksepa, has

1 Bombay Gazetteer, Vols. Surat, Nasik, and Khandesh.

412

8/13/2019 kuvalayamala1

5/7

GLEANINGS FROM THE KUVALAYAMALA KAHA I

Tavad idam natya gitam " Istam yanmanusam yasya tadanyena rametacet/sa janannevamirsyaluradatte tasya jivitam.

This has the same meaning, except that jivai va is replaced by Trsalurjealous. Possibly Ratnaprabha read Tsa-vasu or jwvai va so(to) which breaksthe metre. Isa-vasu gives better sense than jivai, but the construction isawkward.

The language is Apabhramsa of the classical type, but being verse isremarkable only for its date, which is the earliest recorded, the sporadicexamples in the Paumacariya, Bharata's Natyasastra and the Vikramorvasiyabeing all under suspicion as interpolations or through uncertainty of readingor date. It is impossible to say whether the dancing girl was using her own

language, but it was certainly one she understood and one which was ordinarilyused for songs in small towns.

?4The third extract (G. Fol. 47, p. 110) is also a Doha. The caesura is marked

by a matra in J. It is introduced as follows:-

Rade puna kahimpi devaule padiuna pasutto. Rad-pacchima-jame kena viGujjara-pahiena imam Dhavala-duvahayam gTyam-avi ya

Jo navi, vihure, visajjanau dhavalau kaddhai bharu,so gotthamgana-mandanau sesau vva jamsaru.

duvahayam in the Prakrit portion is a hapax, Skt. *dvddhaka dvidhaka.It is not the Dodhaka of the Piingala, which is a 16 matra quatrain, but a Dohaor Daua. visajjanau: G. has vibhajjannau against the MSS. and the metre.

"Now at night he lay down in a temple and went to sleep. In the last watchof the night a Gujjara traveller sang the Bullock-couplet as follows:-' Thefreed bullock, 0 widow, which draws no load, is an ornament of the farm-yard,resembling, in short, the leavings of a temple-offering'."

Ratnaprabha glosses, "Pascatyayame kenapi Gurjara-pathikena gitamDhavala iva yo'tra, vidhure, svajano no bhara-karsane pravanah, sa cagostfagana-bhuitala-vibhiusanam kevalam bhavati ". This shows that he readvi sajjanau, which does not give the best sense. He misses the allusion to thesesa or stale remnants of temple-offerings given by the priest to worshippers assacred relics. Ratnaprabha has therefore turned the comparison of the sacredbull with a sacred but useless relic to one of a widow's useless relative with asacred bull.

The language is once more conventional Apabhramsa, associated with a

Gujjara speaker. Again there is no evidence that it was a specific Desabhasa.

?5The Eighteen Desabhasas or Desibhasas are frequently mentioned in books

of the Jaina Canon (e.g. Nayadhammakahasutta) as necessary for the educationof Princess Princesses, and Hetairai. Jinadasa in Nisithacurn. (G. Intro., p. 87)

413

8/13/2019 kuvalayamala1

6/7

ALFRED MASTER-

writes in A.D. 667 ahavd atthdra-desTbhdsd-niyatam ddhamdgadham, "alter-natively Ardha-Magadhi is the standard form of the Eighteen Desibhasas ",

indicatingthe reason

why grammariansand others

kept Ardha-Magadhiapart from the usual Tetrad (or Hexad) of Classical Bhasas. Uddyotana, writing100 years later, is one of the few authors who mention both the Tetrad and theEighteen Desibhasas. But he does more than mention them, and his account ofthe latter (G. Fol. 131-2, p. 91) includes specimens in the shape of a word or twofor each of sixteen Desabhasas, two being missing. They are introduced asfollows:-

" So after a short time he saw a road with a market resounding with thechatter of men actively engaged in the business of buying and selling with

bundles of various goods outspread. And crossing over to that place he sawsome country merchants, distinguished by different Desabhasas."Then follow Arya couplets containing a description of each class of speakers,

their name and word specimens. The names are Gollae (Abhira herdsmen),Majjhadese, Mdgadhe, Amtavete ? Amtavee or -vette), K;rae (MSS. Kire, but themetre requires another matra: Kashmiris), Takke (Panjabis), Sendhave Sindhis),Mdrue (Marwaris), Gujjare, Lade, Mdlave, Kannd.dae, Tdie (Tajiks or Persian

Arabs), Kosalae, Marahatthe, Amdhre (an Apa. spelling: Andhras). The two

missing Desabhasas are possibly Odra and Drdvid7, as in the Natyasastra.

Some of the words are quite corrupt. The Gollae say U, da, de according toJ, arade P 2 and arare G. But tere, mere, au 1 of the Majjhadese may be themodern Hindi words. Kitto and kimmo may have once been Braj words, astheir assignment to the Amrtavee ndicates, but kittd is now Hindi, not Braj(with Marathi kiti for the indeclinable form), while kimmo is only representedby the indeclinable form kimi how much ? in Hindi, while Gujarati kevo, kemare respectively declinable and indeclinable forms of Apa. kemva, kema, whichalso appear as kima, kimva. Eham, tiham (Takke) are close to Panjabi eha,teha, this, that. The Sendhave vaudayame does not scan, but vahud.i sandyami

my bride, scans and has a meaning, which P 2 vamsedaino has not. Appd,tuppd (? we, you) assigned to the Mdrue recall Marwari adpdm we (incl.) and

Manjhi-Panjabi tupa you (LSI). Re, bhallau (Gujjare) are not typical, and theidentification of nau with the Genitive suffix -nau in Old Gujarati is veryuncertain. The Lade dhamha, tumham (emended to fit the metre) resembleforms found in Old Gujarati and dmhd is found in the Arda Viraf (1415 A.D.)written in the Surat area. The Mdlave words bhdua, bhai.n, tumhe are not dis-tinctive and occur in Ardha-Magadhi. Adri, pondi, mare (P 2 adi, pdn.di,ramare) are not Kanarese, but reading adi (metri causa) the first two words are

good Telugu for "that, go ". Isi, kisi, misi (Tdie) suggest Perso-Arabickismis currants, but the three words seem to form the same phrase as in asi-masi-

kasi-vdnijja-sippa-samgayd n the Samaraiccakaha of Haribhadra (Bib. Ind.

169, p. 2, 17) " concerned with asi-masi-kasi, trade and handicrafts", which is1 Most of the specimens appear as homogeneous pairs with, if needed, a third word as make-

weight (pada-pirana).

414

8/13/2019 kuvalayamala1

7/7

GLEANINGS FROM THE KUVALAYAMALA KAHA I

unsatisfactorily translated by the commentators as " sword, ink, and cultiva-tion ". The Kosalae jala, tala are suggestive of the Chattisgarhi jeld, teld (acc.

of the relative and correlative pronouns), used in an area formerly known asMaha-Kosala. The Marahatthe words dinnalle, gahiyalle (the reading of P 2is required by the metre and gahille, the reading of J, means " possessed (of adevil)" and does not correspond to dinnalle) are not found in Pkt. or Apa.,although possible forms, and only d7nhald occurs in the oldest Marathi, whiledidhalem is frequent in the Jnanesvari. There is a close connection betweenall these forms. The Andhre words adi, puti, ratirm may be intended for Teluguadi, pondi, randi that, go, come, and Hem. Des., vi, 51, has punde go; P 2 hasatthinmadhi.

In prose at the end of the couplets are added the names of three non-Aryanpeoples: the Khasas, a mountain tribe mentioned in the Panhavagaran.aim,the Markandeya Purana, and the epics; the Parasas, inhabitants of Pars orFars; and the Babbaras, who are mentioned in the Jaina Canon and elsewhere,but are unidentified.

To sum up, the words are few, isolated, and often of uncertain meaning.But some important points emerge :-(i) There is no sign of recognition of thesimplification of geminate consonants and compensatory vowel lengthening;(ii) only a few of the words are Prakrit; (iii) there is a reasonably close corre-

spondence of identifiable words with those now used. And in spite of changes inthe number and constitution of the Desibhasas, there emerges a continuity,which appears, for example, in the -e termination (probably dir. pl. masc.) of theMadhyadesa and the Marahatthas as contrasted with the -o and -au(dir. sg. masc.) of the Antavedans and Gujjaras. The Desibhasas were dialectsrather than languages if, as it appears, they depended for their literarymedium on an Apabhramsa which, in its early stages, allowed a great varietyof inflexions and vocabulary. The absence of Apabhramsa prose shows that itwas not used as a medium of oral communication, as were Maharastri and other

Middle Indian languages. It is hoped to deal with this question in a subsequentpaper.

415