Skip to main content

Full text of "Analecta"

See other formats



on behalf of the Committee 

P. K. CODE, M. A., 




Published by 





&, analecta 



V. S. SUKTHANKAR, M. A. (Cantab.), Pb.D. (Berlin) 

Generai Editor, Critical Edition of the Mahdbharata, 
Editor4n-Chiei , Journal of the Bombay Bianch 
of the Royal Asiatic Society, Honorary 
Member^ American Oriental Society, etc. 

^Ist January 1945 

POONA 4 (India) 

This copy is 'Numbered 

Prated ^nd PubKdied for the V. S. Sukthankar Memorial Editioa Committee 
by M. N. Kulkarni, at the Kamatak Printing Press, Kamatak Hpuse^ 
Q)iT^ Bazar, Bombay 2, 


The First Volume of the Sukthankar Memorial Edition, cx)ntaining 
Dr. Sukthankar’s Critical Studies in the Makdbhdrata, was published 
by me on behalf of the Memorial Edition Committee on 21st January, 
1944, the First Anniversary of Dr. Sukthankar’s demise. On the 
occasion of this Anniversary the Committee was fortunate enough to 
have as President Dr. Baba Sahib (M. R.) Jayakar, m.a., ll.d. and 
Shri K. M. Munshi, b.a., ll.b. as lecturer. These two great friends of 
the departed savant paid glowing tributes to the sacred memory of Dr. 
Sukthankar and his epoch-making work on the Critiopl Edition of 
the Mahdbhdrata. A full account of the Anniversary function has been 
published in the New Indian Antiquary, Vol. VI (pp. 225-234) for the 
information of Dr. Sukthankar’s friends all over the world. I have 
to convey the best thanks of the Committee to Dr. Jayakar and Shri 
Munshi for making this function a grand success. 

In his Presidential remarks Dr. Jayakar expressed his apprecia- 
tion of the work of the organizers of the Memorial Edition and 
observed that there should be no hiatus between the publication of 
the First Volume of the Edition and that of the Second Volume 
promised by the Memorial Edition Committee. In accordance with 
this observation of an eminent friend of the departed scholar coupled 
with an additional personal donation of Rs. 200/- for the Second 
Volume announced by Dr, Baba Sahib Jayakar I lost no time in 
commencing my work of collecting funds for this volume. The pub- 
lished Volume of the Edition, copies of which were distributed to 
donors and subscribers immediately after the Anniversary, proved my 
great friend and ally in niy arduous work. This Volume was hailed 
with delight by scholars in India and outside and before any reviews 
of the Volume appeared in Oriental journals it put me in touch with 
an eminent friend of Dr. Sukthankar, I mean Sir C. R. Reddy, kt., 
D.LITT., the Vice-Chancellor of the Andhra University, who informed 
me that he was a contemporary of Dr. Sukthankar at Cambridge 
as early as 1906 and that he desired to have a copy of the Sukthankar 
Memorial Edition as a souvenir of his life-long friendship with the 
eminent Orientalist On getting the First Volume of the Edition Sir 
C. R. Reddy wrote to me as follows on 11th March, 1944 : 

“ I have gone through the First Volume which you sent and I 
am wonder-struck at the deep scholarship, penetrating judgment and 
elegant style of Sukthankar. When I looked at the photos of Suk- 



THANKAR included as illustrations, I missed his dear old Cambridge 
face with its fine wealth of curly hair, which he subsequently seems 
to have mislaid ! I have a photograph* of his, taken in 1906 or a 
while before, with his autograph. If required I can send it to you for 
making a block." 

The sentiments of deep devotion to his old friend Dr. Sukthan- 
KAR evinced by Sir C. R. Reddy in his letter referred to above embold- 
ened me in my appeal to him to use his good offices in collecting some 
funds for the Second Volume of the Sukthankar Memorial Edition. 
My confidence in this genuine old friend of Dr. Sukthankar was more 
than justified as I found to my agreeable surprise that with Sir Reddy 
words meant acts. On 12th May, 1944 Sir Reddy forwarded to me a 
copy of the appeal sent by him to his personal friends for funds to 
complete the work of the Memorial Edition. The eminent friends 
of Sir Reddy were prompt and generous in their response to his appeal 
as will be seen from the following donations received and kindly for- 
warded to me by Sir Reddy between 3rd June and 30th August 1944 : — 

Rs. 500— Raja Saheb of Bobbili, K.C.I.E., d.litt. 

Rs. 500— Hon’ble the Maharaja of Parlakimidi, Prime Minister, 
Cuttack (Orissa). 

Rs. 500— Raja Saheb of Munagala, Saifabad, (Hyderabad) . 

Rs. 250 — ^Hon’ble Rajah Sir Annamalai Chettiar, k.c.le., ll.d., 
Raja of Chettinad, Madras. 

Rs. 250 — Sir C.. P. Ramaswami Aiyer, K.C.S.I., K.C.I.E., Dewan 
of Travancore, Trivandrum. 

Rs. 100 — ^Hon’ble Sir Manohar Lai, Kt., Finance Minister, Govt, 
of Punjab, Lahore. 

Rs. 25— Sir AUadi Krishna Swami Iyer, kt., Madras. 

Rs. 2,125 

I cannot adequately express the sense of gratitude both of the Memo- 
rial Committee and myself to these distinguished donors for tlieir 
generosity and unstinted response so promptly given to Sir Reddy’s 
personal appeal to them. How true are the words of the Dhamma- 
pada ? — 

The scent of flowers, incense and jasmine cannot travel against 

^ ^ this iSTe Cambridge 

^ been reproduced in the present volume. On 
of ae \femanal Edition Committee I have to convey to Sir Reddy their 
^ photograph to their notice as also for permitting its 
reproduction which has greatly enriched the Edition ^ 



the wind, but the fragrance of good deeds travels in all directions. 
Sweeter than the scent of incense and jasmine is the fragrance of good 

I am personally indebted to Sir Reddy for his continuous active 
interest in this work to such an extent that I must ever remain grate- 
ful to his obligations at a time when his help came to me almost by a 
Providential arrangement. It was the darion-call to duty from Maha- 
rsi V3^sa with which Sukthankar closed his Introduction to the 
Aranyakaparvan of the Great Epic and to which he made a prophetic 
and pointed reference in the following parting words : 

“Across the reverberating corridors of Time we his (Vy&sa’s) des- 
cendants can still hear dimly his darion-call to Duty.” 

That this “luminous message of Maharsi Vyasa” as Sukthankar 
put it, was heard by Sir Reddy himself will be clear from the following 
extracts from his personal appeal to his friends issued on 9th May, 
1944 : 

“ Dr. V. S. Sukthankar, whose early death iwas the most serious loss 
to SansJoitic and Oriental Learning generily that India has sustained since 
the death of Dr. R. G. Bhandarkar, was a contemporary of mine at Cam- 
bridge Even in those early years he had quite a reputation for original 
thinking. By far the greatest undertaking of modem India is the Critical 

Edition of the Mahdbhdrata The Editorship of such an undertaking 

required colossal scholarship, vast patieice and a critical acumen of the 
highest type. Dr. Sukthankar who had studied Sanskrit both in Cambridge 
and in Berlin was appointed Editor. How well he has done the work is 
proved by the remarkable reception given to it by Sanskritists of all the 
Universities of the world. It may be remarked that under him the American 
Professor of Sanskrit in the University of Yale was editing one of the Par- 
vans. Dr. SuK'dHANKAR died before the Critical Edition, as it is ^ed, 
of all the Eighteen Parvans could be published. But by the publication of 
5 or 6 Parvans he had set the standard and the model for all future work. 
It will be recalled that this Critical Edition has heai under preparationi for 
over 25 years. Imagination staggers at the volume of labour, comparative 
study and critical work involved. 

Dr. Sukthankar embodied some of the result^ of his studies in Sans- 
krit Mahabhafata in a series of astoundingjy profound and brilliant Prefaces, 
Essays and Lectures. These are now under publication and the First Volume 
has been published. / have read this First Volumd and my advice to every 
Hindu is that he must regard the study of it as an indispensable part of his 

As a colleague of Dr. Sukthankar for seventeen years prior to 
his lamented demise I fully endorse the foregoing estimate of Dr. 
Sukthankar’s sdiolarly work in the field of Indology» the begimiiiigs 
of which were noticed early by his Cambridge contemporaries like Sir 
C. R. Reddy, Dr. M. R, Tayakar and others but which found a con- 



genial soil and a bracing atmosphere at the Bhandarkar Oriental 
Research Institute and put forth its richest blossom and fruit in the 
shape of his Cntical Studies in the Mahabharata and the publidied 
volumes of the Critical Edition of the Great Epic. 

The tabula gratuVztoria appearing m this Second Volume of the 
Me mo rial Edition includes the names of all donors and subscribers 
whose generosity has been responsible for the publication of this 
volume. I convey to these friends the best thanks of the Memorial 
Edition Committee. In particular I have to convey the special thanks 
of the Committee to the following contributors whose timely help as 
indicated below has enabled the Committee to complete the Memorial 
Edition in spite of all difficulties : — 

Rs. 300— The Government of Bombay purchased 20 copies of 
the First Volume of the Edition. 

Rs. 200 — ^Right Hon’ble Dr. M. R. Jayakar, Bombay. (This is 
an additional donation lor the present volume in 
addition to Rs. 100/ received for the First Volume.) 

Rs. 200 — ^The Umvorsity of Bombay (in addition to Rs. 150/- 
received for the First Volume). 

Rs. 100 — Shri Han Narayan Purohit, b.a., Vidyabhushan, 
Jaipur (in addition to his subscription for the Edi- 

Rs. 50— Raja Saheb of Aundh (in addition to Rs. 100/- received 
for the First Volume) . 

I hope I shall not be exceeding the bounds of official decorum if I record 
here the best thanks of the Memonal Edition Committee to Mr. S. N. 
Moos, C.I.E., M.A., I E.S., the Director of Public Instruction, and Prof . 
R. P. Patwardhan, M.A., I.E.S., the Deputy Director of Public In- 
struction, who recommended to Government the purchase of 20 copies 
of the First Volume of the Edition. Similarly I must not fail to convey 
my personal thanks to our Vice-Chairman, Diwan Bahadur K. M. 
jHAVERi, and other friends at the University of Bombay, 
whose good offices have been responsible in securing from the 
University an additional donation of Rs. 200/- for the pr^ent volume. 
To my wtogenerian friend Shri Hari Narayan Purohitji of Jaipur, 
whose dose intact with me durmg the last ten years has enlivened 
rny mtoest in the history of Jaipur and Rajputana, I am deeply 
obliged for ks vduntary gift of Rs. 100/- on his receiving the First 
Volume. Wlule sending me his blessings for the completion of liie 
EditKm P^ditji wrote : " The present edition of Vol. I is so valuable 
a ^eduction that even crores of rupees would not be equal to its 
value . In my prrface to the First Volume I observed that “the 


valuable and sdiolarly contents of the present volume speak for them- 
selves and will continue to speak with greater resonance as years pass 
by”. Judging by the correspondence from scholars received by me 
smce the publication of the First Volume I have reason to bdieve that 
my observation has been fully vindicated. 

I am personally indebted to the Raja Saheb of Aundh, the Chair- 
man of the Memorial Edition Committee and Dewan Bahadur K. M. 
JHAVERI, the Vice-Chairman for their continued help and guidance m 
my work on the present volume. They were kind enough to attwid the 
function arranged by me on 21st January 1944 and encourage me in 
my efforts to pu^ on the work projected by the Memonal Edition 
Committee by sending me tokai donations for the Second Volume in 
addition to the donations sent by them for the First Volume. The 
blessings of these two grand old friends of Dr. Sukthankar, I mean 
the Raja Saheb now running his 77th year, and the Dewan Bahadur 
now in his 76th year, have been responsible for bringing the work of 
the Memorial Edition to a successful conclusion and while conveying 
to them my humble thanks for these blessings I wish them happy long 
lives and mcreasing prosperity to continue their disinterested services 
to the sacred cause of Indology in the widest sense of the term with 
which Dr. Sukthankar had completely identified himself to the last 
moment of his conscious life. 

As regards the writings of Dr. Sukthankar included in the 
present volume I tender my most grateful thanks — 

(1) To Mr. B. T. Anklesahia, m a , the Hon. Secretary of the 
K. R. Cama Institute for securing the pomission of his Institute to 
indude Dr. Sukthankar’s two papers in the Memorial Edition, one 
of whidx viz. “ Arjunami§ra ” has been published in the First Volume 
while the othor on “An Excursion on the Periphery of Indological 
Research ” has been induded in the present Volume. I diall not for- 
get the kindness of this sincere friaid of Dr. Sukthankar in sending 
me free copies of these papers. It is unfortunate that this learned 
friend of ours should pass away* before Dr. Sukthankar’s second 
paper is reprinted in this Volume ! May his soul rest in peace ! 

(2) To Rao Bahadur K. N. Dikshit, m.a., Director-General ol 
Archseology m India for permission to indude in the Memorial Edi- 
tion Dr. Sukthankar’s papers in the Epigraphia Indica. 

(3) To the authonties of the Bombay Branch of the Royal 
Asiatic Society for permission to indude in the Memorial Edition 
Dr. Sukthankar’s papers originally published in the Society’s 

* Mr. Anklesaria passed away m November 1944, 



(4) To Dr. R. N. SAEDESAI, L.C.P.S., Proprietor, Oriaital Book 
Agency, Poona, for permission to reproduce Dr. Sukthankar’s Eng- 
lish Translation of the VdsavadattS in the Memorial Edition. 

(5) To the Editors of the Oriental Literary Digest, Poona, for 
permission to include in the Memorial Edition some reviews of books 
by Dr, Sukthankar, 

(6) To the Editors of the Annals (B. O. R. Institute) , Dr. R. 
N. Dandekar and Prof. K. V. Abhyankar for permission to repro- 
duce in the Memonal Edition Dr. Sukthankar’s papers originally 
publi^ed in this journal. 

(7) To the Editors of the Journal of the American Oriental 
Society for permission to reproduce Dr. Sukthankar’s Studies in 
Bhasa in the present Volume. 

(8) To the Editor of the Journal of the Mythic Society, Banga- 
lore, for permission to include Dr. Sukthankar’s papers originally 
publidied in their Journal. 

(9) To Dr. N. P. Chakravarti, Deputy Director-General, of 
Ardueology in India for sending me a complete list of Dr. Sukthan- 
kar’s contributions to Eptgraphia Indtca and other publications of 
the Archaeological Department along with extracts from Dr. Sukthan- 
kar’s apiplication at the time of his joining the Archaeological De- 
partment, In forwiarding the extraicts Dr Chakravarti wrote to me 
on 12-6-1943 : “ Professor LxJders always thought very highly of his 
pupil Sukthankar. Even when I met him as late as 1922 he told me 
that Dr. Sukthankar was still the best pupil he had from India.” 

It was not possible for me owing to the present war to sedc permis- 
sion of the publishers of Dr. Sukthankar’s thesis on “Die Gram- 
matik Sakatdyarufs ” and the Editors of the Z. D. M. G. who published 
Dr. Sukthankar’s paper entitled “Miscellaneous Notes on Mam- 
mata’s Kdvyaprahaia.” I offer to these publishers the apologies of 
the Memorial Edition Committee for mduding these writings of Dr. 
Sukthankar without their formal pmnission owing to circumstances 
beyond the control of the Committee. I b^ also to be excused for 
any infringement of the nghts of any publishers that I may have over- 
looked in bringing out the Memori^ Edition in haste solely with 
the object of commemorating Dr. Sukthankar’s services to Indology 
and thus redeeming at least partially the debt I owe to his inspiring 
scholarly contact of seventeen years at the Bhandarkar Oriental Re- 
search Institute^ Poona. 

As in the case of the First Volume of the Memorial Eldition the 
eitire editing of the present volume has been carried out by my most 



esteemed friends Dr. S. M. Katre and Rof. D. D. Kosambi. They 
have tried their best in editing this volume as neatly and accurately 
as possible in spite of the diversity of material which reqmred lynx- 
eyed proof-correcting, coupled with an expert knowledge of printing 
and typography, not to say a close knowledge of German in which Dr. 
Sukthankar's thesis appears in the present volume. The task of 
editing this thesis has been considerably lightened by the willing and 
dismterested co-operation of our fnend Dr. V. V. Gokhale, who as a 
friend and admirer of Dr. Sukthankar’s work joined the Memorial 
Edition Committee m the very first wedc of our enterprise and offered 
his ungrudging co-operation m the execution of the Committee’s pro- 
ject. I have, therefore, to thank most cordially all these three friends 
for their harmonious co-operation which has crowned the Committee’s 
efforts with success. Dr. Sukthankar and Dr. Katre were Matiie- 
maticians in their early careers though later they took to Indology. 
Prof. Kosambi, though at present a renowned Mathematician, is lean- 
ing towards Indology and let me hope that Indology is benefited before 
long by Eis rigid mathematical training and scientific outlook on life 
and literature. 

In the preliminary appeal issued by me on behalf of the Memorial 
Edition Committee reference was made to the intention of the Com- 
mittee to include in the present Edition a hterary biography of Dr. 
Sukthankar on the strength of materials gathered by me from the 
numerous friends and admirers of the great Savant. A few of these 
fnends* have forwarded to me some letters of Dr. Sukthankar receiv- 
ed by them but they are hardly sufficient for a comprehenave literary 
biography of Dr. Sukthankar contemplated by the Committee. Under 
these drcumstances it was thought advisable to indude in the Edition 
Dr. I^tre’s daborate monograph on “ Vishnu Sitaram Sukthankar 
and his Contribution to Indology ” which was published last year in the 
Sukthankar Memorial Volume of the Bulletin of the Deccan College 
Research Institute, Poona. This monograph prepared as it is with 
meticulous care by my learned friend Dr. Katre now takes the pl^irp 
of the contemplated literary biography and has accordingly been indud- 

■* Among friends who were kmd enough to send me some correspondence from 
Dr Sukthankar received by them I may mention Rev. H. Heras of St Xavier’s 
College, Bombay, Dr. S K. Db of Dacca, Dr. Ruben of Ankara (Tmiey), Mr. 
Y. R. GUPTE of Poona and Dr A. N. Upadhyb of Kolhapur. Some other f^da 
had promised to send some letters of Dr Sukthankar but they havfe not stUl 
been recrived. I have^, however, to amveiy the best thanks of the Cbmmittee to 
the above mentioned fnendsi for the material sent by them. If arffii+{n.npi | maB-P- 
nal 18 received by the Committee it may still be’ possible to use it Iot some mpmm'r 
Dr. Sukthankar as a man and scholar with a view to supipJement Dr. Katre’s 
monogri^ pnibhshed m the present volume. 


ed in the present Volume with the kind permission of the authorities 
of the above institute. I have to convey the best thanks of the Com- 
mittee to these authorities for this permission. I have also to thank 
Dr. Katre for his devoted labour of love in the preparation of this 
monograph which is based on the published writings of Dr. Sukthan- 
KAR and as such contains a literary biography of this great Orientalist 
as revealed by his own writings. Dr. Katre’s dose personal contact 
with Dr. Sukthankar during a decade preceding the latter’s demise 
and his thorough tmderstanding of Dr. Sukthankar’s cntical philo- 
logical method have enabled him successfully to trace the growth and 
expansion of his monumental scholarship which gave a stately stature 
to Indian cntical scholarship by his masterly editing of the Maha- 

In conduding this preface to the Second Volume of the Memorial 
Edition I cannot adequately express my sense of gratitude to cui 
liiend Mr. M. N. Kulkarni, who has done yeomen service to Indo- 
logy by shouldering the heavy responsibilities of publishing many 
works on Indology on bdialf of his Kamatak Publishing House in the 
best possible form and character. The name of Mr. Kulkarni and 
his Kamatak Publishing House and Kamatak Printing Press have 
now became proverbial as a guarantee for good printing and pubhsh- 
ing of every work imdertaken by them and the Sukthankar Memonal 
Edition has enjoyed the fullest benefit of this guarantee. In spite of 
every conceivable difficulty consequent upon war conditions such as 
scardty of paper and lalx)ur, mconvemences and delays created by 
the recent paper control order and similar handicaps, Mr. Kulkarni 
has stood by me and fulfilled his guarantee to the letter in completang 
the work of this Edition most promptly, efficiently and zealously like 
my esteemed friends Dr. Katre and Prof. Kosambi. But for the loyal 
co-op«ration of these sincere friends it would have been impossible 
for me to undertake the work of the Manorial Edition an d complete 
it within two years. 

In presenting this Second Volume to' the public on the Second 
Adversary of Dr. Sukthankar’s demise the Memorial Edition Com- 
mittee has cmnpleted one of its projected tasks. The second task of 
the Committee which remams to be completed is the investment of 
the proceeds of the Edition for instituting a special medal, fellowdiip 
or lecture^p in connection with Epic Studies. The execution of this 
ta^^ depends on the quick realization of the sale-proceeds of the entire 
edition. It is hoped, therefore, that friends and admirers of Dr. Suic- 
thankar all over the world will readily come forward to purchase the 
completed Memorial Edition and thus help the Ccanmittee to com- 
memorate Dr. SuKTH^VNKAR's signal services to the Great Epic of 


India, the MahahMraUt, the richest heritage of the Aryan race and 
the national saga of India. 

Finally I convey my most grateful thanks to all my colleagues on 
the Memorial Edition Committee with whose initial blessmgs, good 
wishes and sincere co-operation I started my work on the Memorial 
Edition and with which alone I have been able to carry it to a success- 
ful conclusion without a hatus. I fully endorse the hope expressed by 
one of my colleagues, Dr. N, P. Chakravarti in the following memor- 
able words : 

“ So long as the Bhandarkar Research Institute will be in exist- 
ence and his colleagues and pupils will be there, the same spirit with 
which the MahSbhIarata work was started, I am sure, will prevail.” 

Though Dr. Sukthankar has done his part of the MaMbhIarata 
work nobly the responsibility of completing it wholly lies not only on 
the shoulders of his colleagues and pupils at the Bhandarkar Institute 
but on those of all his countrymen and the Memonal Edition com- 
pleted to-day stands as a permanent rermnder to his countrymen to 
revere “this deathless traditional book of divine inspiration un- 
approachable and far ronoved from possibilities of human consti- 


Poona 4, 

21st Jamtary, 1945 

P. K. Code, 

Hon Secretary and Manaiini Editor, 

Dr, V. S. Sukthankar Memorial Edttton Committee. 



Preface v 

Die Grammatik Sakatayana's 

Einleitung 1 

L Tea . 9 

II- Tell . . , 43 

arUDIES IN Bhasa 

I Introduction 82 

II On the Versification of the Metncal Portions of the Dramas 92 

III On the Relationship between the Carudatta and the 

Mrcdiakatika 113 

IV. A Concordance of the Dramas 125 

V A Bibhographical Note 141 

VI. On the Prakrit of the Dramas , 159 

VII. The Bh^sa Riddle A Proposed Solution . , 170 

Epigrafhic Studies * 

I The Porumarailla Tank Inscription of Bhaskara Bhava- 

dura . iSaka 1921 , 184 

II. Bhandak Plates of Kri^maiaja I iSaka 694 . 199 

III. A New Inscription -of Siri-Fulumavi 213 

IV. Three Kshatrapa Inscnptions 217 

V. Two Kadamba Grants from Sirsi 229 

VI. A Viakataka Inscnption from Ganj 240 

VII Two New Grants of Dhruvasena (I) from Paiitana 243 

VIIL On the Home of the so-called Andhra Kings 251 

IX. Besnagar Inscnption of Heliodoros 266 

X. Palaeographic Notes ... . 273 

XI. Progress Report of the Archaological Survey of India, 

Western Circle, 1916-17 . . . 283 

XII. Progress Report of the Archaeological Survey of India, 





Miscellanea • 

An Excursion on the Periphery of Indologkal Research 311 

Miscellaneous Notes on Manunata’s Kavyaprafc5isa— 

(i) The two Authors of the Kavyaprakasa 321 

(ii) A Note on Mammata's Saniucca 3 /a 334 

(hi) Another Case of the Practice of Quoting Names 

merdy Honons Causa 342 

The Satavahanas . 345 

An Assyrian Tablet found in Bombay 346 

“ Charudatta ” — ^A Fragment . 347 

Cunosities of Hmdu Epigraphy 352 

Book Notices and Reviews . . 360 

Oldest Hindu Drama . . . . 368 

The Art of India . . 369 

In Memonam . 380 

The Position of Linguistic Studies in India . . 386 

VasavadattA . . . . 401 

Vishnu Sitaram Sukthankar and His 'Contribution to 

Indology . . , . 463 

Tabula Gratulatoria . . , 504 




Die vorliegende Arbeit gibt ein Specimen der grammatischen Sutras 
gakatayana's nebst dem Kommentar Cmtamai;^. Den ersteti ausfuhrhchen 
Beiicht^ uber diesen Grammatiker gab Georg Buhler, Onent wid Occident, 

2 (1864), 691 ff. Er hielt den unter dem Namen des iSSkatayana iiberUe- 
ferten Traktat fur das Werk des alten von PBujini erwahnten ^Sakatayana und 
glaubte, dass die Grammatik P§mm’s nur “ erne verbesserte, vervoMandigte 
und teilweis umgearbeitete Aullage der Grammatik ” Sgkatayana’s sei (a a. 
0. S. 703), wobei er sich hauptsachlich darauf stutzte, dass sich zwei von 
Banmi seinem iSakatayana zugesdiriebene Regeln auch in den ihm vorlie- 
genden Blattem des igabd'anu^sana fanden Diese Ansicht spricht er winder 
in einer Wemen Mitteilung aus, die kurz darauf m derselben Zeitschnft eis- 
chien, Or. md Occ. 3 (1864), 181 ff. Sie hat sich jedochnichtbestabgt. In dem 
Aufsatz “On the Grammar of Sakatayana”, Indian Antiquary, 16 (1887), 
24 ff. wies Franz Kielhokn darauf hm, dasa unser Grammatiker nidit nur 
im voUen Besitz alles dessen sei, was I^ini, KStyayana und Patafijah gdehrt 
haben,® sondrni dass er auch die Ldirai, die zum erstenmal bei Candra auf- 
korainen, berucksichtigt babe Diesdbe Abhandlung enthalt femer eine 
iJbersicht des Inhalts der Grammatik und dn Verzeidinis einer Anzahl vcm 
grammatischen Werken, wie Kommentare, prakriyas usw., die sich an das 
Werk eng anschliessen * Eimge Jahre spater (1893) gab Gustav Ofpert die 
grammatischen Sutras iSakatSyana’s nut dem Prakiiyasamgraha von Abhaya- 
candra-{6}-siddh5ntasun* unter dem Titel Sakatayana’s Grammai, Vol. 1, 

* Nodi frtihere Notizen bei Wilson, Mackenzie Collection, 1, 160 und 
Golostuckbr, Pdnini kts ^iUtcA tn Sanskrit literature (1861), S. 163. 

~ Burnell hatte ubngens auch sdion erkannt, dass die Sakatiayana-Grani* 
mahk junger als die Giaminatik Panim’s stem musse, meinte aber, da^ das uns 
vodwgende Werk due Neuredakbon der Grammatik des alten Sakatayana sd. Cf. 
On the Atndra School of Semkrtt GrammarianSi (1875), 8. 97ff, 

8 Diesen sind noth folgende hinzuzufi^en : Momprakasika (Kom. zum Cmta- 
maiji) von Ajitasena ; Amoghavrtti (ein ausfiihrlidier Kom. etwa wie die KS&ka) , 
Nyfisa (Kom zu der Amo^vrtti) von PraHiacandia , endhch nodi erne TikS von 
Bhavasenatnvidyadeva. Eiese Liste haba ich aua der Bombayer Ausgabe entnom.' 
men Mir waren die Werke unzuganglidi 

‘ Zum zwdtenmal abgedruckt von den Jama-Gdehrten PauKjit-Jyesthara- 
mamukundajISarmB und Pannalala unter dem Titel irtmadabhayacandrasur^a^ 
taprakriyasamgrahasahitam Sakafayanmn vyakaranam (Bombay, 1907), Trotz der 
zahlreidiQi Memen Drudcfebler emp&ehlt sich diese Ausgabe durch die gdegent- 
lichen Auszaige aus dem Cintamani imd die Erlauterungen, die in den Fussnoten 
enthalten smd. Die letzteren smd mir erne wesenthche Hilfe zum Veistandnis der 
Grammatik gewesen. 


DIE grammatik Sakatayana's 

heraus. Der zweite Band soUte die Amoghavrtti, einen ausfuhrlichen Kom- 
mentar zu alien Regeln .Sakatayana’s enthalten, ist aber mdit erschienen. In 
der kuizen Einleitung zu dem erschienenen Band vertntt Oppert die irrtum- 
liche, von Kielhorn endgultig zurvickgewicsene Ansicht uber das Alter 
unsers Graminatikers und ubergeht stillschweigaid den oben erwahnten im 
Indian Antiquary erschienenai Aufsatz Kelhorns. Dies veranlasste Kiel- 
horn nochmals auf die Sakatayana-Grammatik zuriickzukomnien. In pmorn 
Aufsatze in den Nachrichten von dek kontgl Gesellsch der Wss in Gottm 
gm (phil-hist K1 1895) vergleicht Kielhorn erne fortlaufende Reihe von 
Regeln der iSakaiiayana-Graminatik (2. 4. 128-289) nut den Regain PBjiini’s und 
den Ldiren seiner Nachfolger und gibt das Resultat dieser Vergldchung. 
Diese UntersMchung stellte in bezug auf unsere Gr ammat ik folgendes fest. 
Sie enthalt R^eln : 1. die dassdbe lehren wie die Regeln PSomi's ; 2. „ fiir 
die bei PBinini Aequivalcnte nur dann sidi finden, wenn wir seme Regeln so 
erweitem, besdiranken oder anderiweitig andern, wie dies m den einzdnen 
Fallen von den Verfassem der VSrttikas oder des MahSbhishya vorgeschne- 
ben wird “ (a a O. S. 10) und solche, die erst von K!aty3yana und Patan- 
jali vorgetragen iwotden smd ; 3. fiir die man enteprechendes nur m den 
Ganas zu F&mni’s Regeln oder in der Ka4ik& nachweisen kann ; 4. die eret 
bei Candra aufkommen und 5 endlich auch solche, die weder bei PSinini und 
semen Erklarem noch bei Candra najdiweisbar smd Wichtig war der Nach- 
weis Kielhorns, dass die Verfasser der Ka^ika auf erne Regel 'Sakalayana’s 
kemen Bezug nehmen, in der ^k&tayana „allem das nchtige gelehrt hat, 
und wo sie [d i, die Verfasser,] . . eineni sprachlichen Faktum gegenUber- 
stehen, das dutch kerne Regel P%imi’s order dessen Interpreten seme Er- 
Wdrung findet " (a a. O. S. 13). Die ganzhche Abhangigkeit Hetnacandra’s 
von Sakatayana hat Kielhorn schon in dem Aufsatz im Indutn Antiqumy 
(Bd. 16) behauptet und nachgewiesen. Weitere Belege dafur wird man m 
meinen Erlauteningen finden Damit ist die relative Chronologie der Gram- 
matiker vai FSoini bis Hemacandra emwandfrei festgelegt 


Ich wende midi jetzt zu den Ergebnissen memer Untersuchung des 1. 
pdda des 1. adhy&ya. Die emleitenden Strophen in dem {7} Kommentar 
geben wichtige Aufschlusse sowohl fiber l^katayana als uber den 
Kommentar und dessen Verfasser, die z. T. schon von verschiedenen Gelehr- 
ter, mitgeteilt worden sind®. Ich fiige eine wortlicbe Ubersetzung bed da 
sie em dem Kommentar bezugliches wichtiges Faktum zu Tage bringen wird, 
das von anderen ubersehen zu sein scheint. Die Ubersetzung lautet ; 

undOo^eSf A ^ eistenmal JBuhler in Or. 

^ Bombayer Ausgabe haben einige 

TS^er abgedrudct. Erne Auswahl gibt Whber m Hand- 

^ Btbl m Berlm (1886), S. 205. Vgl. water Bubneu, 

op. At. (passim) ; Ind. Ant. 16, 24 ff, 



1. Es mSge das allwissende Licht der Erkenntms, (namlich) der 
das Weltall erleuchtende, alle Wunsche gewahrende Cintamaini Euch un- 
verganghches Gludc bnngen. 

2 Verehrung der Sonne, (namlich der Offenbarung des) Brahman 
als Wort, die die Welt fordert, (indem sie) die Erleuchtung der von der 
Macht der Finsterms iiberwaltigten Erde bewirkt. 

3 Heil iS&katayana, das Oberhaupt der grossen Gemeinde der 
Mdnche, der die Kaiserwurde (un Reiche) aller Erkenntnis erlangt hat, 

4. der allein den Ozean der Worte mit dem Mandara(-berg iaeines) 
Geistes quirlte und den ganzen Nektar der Granunatik nebst der Sti des 
Ruhmes herauszog, 

5 van dem eine Grammatik erfunden*’ ist, die ^nngen Umfang 
hat, leicht zu erlemen, vollstandig, alien Nutzen bringend und die beste 
ist (und daher) der Lehre der Aihats gleicht (die dieselben Vorzuge 

6. in dessen Grammatik ausserhalb der Regeln (swtra) kein Desid- 
eratum {isti) aufzustellen ist, kern Nachtrag {vaktavya) zu machen, 
kein Zusatz {upasamkhyma) hinzuzufugen ist, — 

7 indem Yabsavarman dessen (d i iSakat&yana’s) sehr umfangrei- 
chen Kommentar zusammengezogen hat, wird er diesen kiirzeren, (den- 
noch) in alien Bestandteilai vollstandigai Kommentar^ vortragen. 

8 Dieser Versuch, (das Lehrbuch) zusammenzufassen dient dazu, 
denjenigen, die sich vor umfangreichen Texten scheuen (und) deren 
Verstandeskraft noch unentwickelt ist, Tugenden wie Gehorsam gegen 
kanonische Werke und andere beizubnngen. 

9. Die Zahl der Slokas des Cintamani, des Konunentars des Sabda- 
Husdsana, der desQ Sinn treu wiedergibt, ist als 6000 festgestdlt. 

{8} 10. Die von den Grammatikem Indra, Candra usw gdehrten 
grammatischen Regeln stdien alle hier Was nicht hier steht, stelit 

M Man wisse, dass die ga^a^ und die Wurzeln in den gana- und 
dhatupdtha, alles was das Genus betrifft in dem lingdnusdsana, die mit 
un und anderen (Suffixen) gebildeten (Nominalstamme) m den unddi 
{-sutras), das iibrige alles m diesem Kommentar zu finden sind, 

12. Sicherlich werden infolge des Stadiums dieses Konunentars so- 
gar Kinder und Frauen innerhalb ernes Jahres die ganze Sprache 

® Fur die Bedeutung von upakrama, neutr. am Ende ernes Tatpunx^a vgl. 
P 2 4. 21, upafi&pakramain tadadyacikhydsayam, 

7 D h. ein Kommentar, der die anuvrttt, uddharma^ pratyuddhara^, die 
apavadas usw. und etwa die in Betracht kommenden pairibhdsas angibt. 



Aus den Worten Yaksavarman’s geht deutlidi hervor, dass der Verfas- 
ser des umfongretchen Komnuhttars, dessen gurze Fassung der Cintamam 
darstellt, Sdkatdyana seUrst ist. Denn das tasya in Vers 9 muss das Korrelat 
der in den vorangehenden Strophai befindlichen Relativa seui. Sonst wiirde 
iiberhaupt den relativen Satzen em entsprechender unabhangiger Satz fehlen. 

Diese Tatsachei, wdche Buhler und Kielhorn® ubersehen zu haben 
scliemen, macht den Cintamam um so iwertvoller Der Umstand, dass i§ak- 
atayana seme eigenen Sutras kommentiert hat, bietet nichts aussergewohu' 
liches Hat doch HemacandrazweiKommentare zu semer Grammatik verfasst. 
Es ist auch sehr wahrscheinlich, dass Candra ebenfalls seine Sutras kommen- 
tiert hat* Also hat ^Skatayana emen Kommentar zu seiner Grammatik ver- 
fasst Daraus erklart sich dber die Kurze der Sutras Sie ist entstanden nicht 
sowohl aus emem „ krankhaften Streben die Sutras moglichst kurz aus- 
zudrucken, als vielmehr daraus^ dass zwischen der Zeit Pamni’s undlSakata- 
yana’s das Schwergewicht in dieser Literaturgattung verschoben war Nun 
bildet der Kommaitar emen mte^nerenden Teil des Werkcs. Die Sutraperiodc 
war langist zu Ende Der Sutrastil geht dann in den. Bha§yastil uber.^i Und 
unsere grammatschen Sutras sind schwache Reflexe einer cigentlichen Sutra- 
literatur, smd nur Stichworte, blosse Hilfsmittel zum Manoneren, die bis zum 
heutigen Tage dne grosse Rolle un Unterrichtsverfahren der Inder spielen. 
Denn nicht nur smd die spateren Sutras an und fiir sich volkcwnmen unver- 
Biandlich ; sie smd sogar in sidi nicht voIlstSudig.^® Es ist wohl bekannt, dass 
m der Cdidra-Grammatik edmge von den unentbehrlichsten Panbhagas 
fehlen und dass sie in der Regel kerne Definitionen der Ter mini gibt. Die pari- 
bhSsd : yathosarfikhyam aitudeSah samdndm (P. 1 3 10) ist in der iSakata- 
yana-Grammatik nur im Kommentar erwahnt. Das Sutra Saka'layana’s 
Sfdanidrd (S. 1. 1. 49) das dem Sutra PSnim’s anekdUit sarvasya (1 1 55) 
entspncht, heisst an und fur sidi gar nichts. Die Beispiele kann man nach 
Belidien vermehren. 

Dennoch hat iSakailfiyana die technische Seite seiner Grammatik kemes- 
we|^ vemachlassigt. Er hat aufs gewissenhafteste versucht sdin Werk von 
Fefalem des anukUi und durukta frei zu machen Man vergleiche den^Gebraucb 
des Wortes bhavya in 1. 1. 4, airaya in 50, das Sutra 65, die Formuherung 
des Sutra 51, usiw., wie dies eben m d^ einzelnen Fallen in den Vartdkas 

* Kielhorn, Ind, Ant. Bd 16 In the mtroductory verses the author 
states that he has compiled bis work from a more extensive commentary (S. 25) . 

* Cf. CiEBiCH, CandrorVyakarana, Abh. f. d. Kunde des Morgenlandes htsg. 
von d D M. G. Bd. 11, No. 4, Vorwort S. VIII. 

1# So Kielhorn, Nachnchten von der Kgl. Gesellsok der Wiss in Gottingen 
(phiL-hist KI. 1895), S 10. 

“ Cf. Hermann Jacom, uber die Echtheit des Kautiliya, Sitzmgsb. d. 
komg}. preuss Akad. d. Wks fur 1912, S 842. 

Nur insofem tnlden die Sutras eme Einheit, als Stidiworte auch von anderen 
benutzt und zum Geeenstand der Kommentienimr crminrlit tnvrdpn 



bezw. dem Mahabhasya vorgeschrieben wird. Vor allem zeigt sich dies in 
dem Sutra sludgenat (I. 1 . 52). Dies ist eine von den sdir schwierigen pan- 
bliS^Ss P. 1. 1. 56 ff., die zur Erklarung, Berichtigung und Erweiterung uber 
75 vartttkas hervorgerufei haben. Der Einsdiluss von enad ist notwendig zur 
Bildung der Fonn enad acc. neutr., das dem Sutra iSakai&yana’s zugrunde 
liegende varttika wird aber von Kiatyayana nicht unter semen Bemerkungien 
zu den Sutras P. 1. 1. 56-59, wo es logisch hingdiort, vorgetragen, sondem 
an einer ganz entlegenen Stelle. 

Da i&aka|tayana offenbar ein Jaina war, versteht es sich von selbst, dass 
er den vedischen Dialekt gamiclit hat berdcksichtigen woUen. Demzufolge 
hat er nicht nur alle Akz^tregeln bei Fanmi weggelassen, sondem audi z. 
B. R^ieln uber die Bildung zahlreidier yedischer Inhnitiva, Absolutiva usw. 
usw. Doch ist er nicht ganz konsequent verfahren und auf Sdiritt und Tntt 
begegnen uns Regdn uber Worte und Bildungsdemente, die nur fiir die 
vedische Literatur gielten Oder jedenfalls im klaasischen Sanskrit me zur 
Anwendung kommar. Man vergleiche z. B. acchavad (1 1. 30), tipaje, 
arimje^^ (33), vt^adryac, adadryac, atmtmyac, amudryac^* (1. 2 . 45 und 
2. 2. 65) usw. Lebrreidi ist die Regd 1 . 1 . 104, die die Substitution von 
S fiir u lehit, weil diese Substitution ihre Stelle eigentlich nur im Padapatha 
hat^*’ Die Aufnahme dieser Regd ba 'SakafSyana ist daraus zu erklaren, 
dass PSinini diese Substitution fur die “ nicht-veidische ” Sprache (,andr§e) 
lehrt TJnd was “ nicht-vedisdi ” ist, konnte mit gutem Gewissen nicht ausge- 
lassen werden Man darf also behauptm, dass Sakafayana ausser den Akzent- 
regetn nur dte Regeln Pd^ints ausldsst, die ausdnicMich mit dem Vermerk 
chandasi vsto, gelehrt werden. 

Wie scdion oben erwahnt, begniigt Sakatayana sich nicht damit, 
die Lehren semer Vorgang^ zusammenzufassen und sie systematiscb 
anzuordnen, sondern er geht fiber sie hinaus und tragt — freilich nur s^r selten 
— auch neue Lehren vor. Idi konatatiere folgmde Neuerungen im 1. pSda des 
1'. adhyaya : Der Auslaut der Partikdn ca usw (ausser a) darf nicht in dei 
Pause nasaliert werden (Sutra 6 B), zulassig sind im Kompositum die Formen 
sukharta, pfapja usw. (89) ; gavak^a kann nur “ Fenster " bedeutm, sonst 
mussi man goaksa Oder g<fk^a sagen (95-98) ; die auf eindn anusvara bezw. 
marjanSya folgenden Tenues kotmen verdoppelt werden^’^ (115) ; nach emem 
/liuto-Vokal am Ende eines pada kann ch verdoppelt werden (125) ; vor sea 
dfirfen 4 und n bezw- » die Gleitlaute t bezw. 7 nicht angefugt werden (146, 
147) ; saskartr als eine Nebenform yon sofftskartr (152) ; fiber den sandfti in 
Fallen iwie yafus + pitakam, sar^ + kalakam, usw. (172). 

In der Literatur sind sie nicht belegt. 

Belegt ist nur vkvadryac itn Rgveda. 

Ahnlidi ist die Aussefaliessung von iti in i. 1. 99 zu beurtalen. 

Me Verdoppelung ist in sudindisdien Handsduiften sehr verbmtet. Nadi 
WackermaKxl {Aitind, Oram. 1. 42 § 96 a) auch insfcbriftlidi hinter rmusvara. 



Hiennit ist die Wichti^t der Grammatik Sakal&yana’s fur, die Gesch- 
ichte der Entwiddung der mdischen Giaimnatik seit Patanjali an die Hand 
gegdben. Auf die Rolle die sue fiir das Verstandms der Grammatik Hema- 
candia’s spielt, hat schon Kielhorn hingewissen Ich eiwahne nur, dass all 
die obengenannten Neuerungen ausser der Zulassigkeit der Formen sukharta, 
prajna usw., sich bei Hemacandra wiederfindei. Der Ansatz des Wortes 
saskartr von Hemacandra, das auf Missverstandnis einer tsH Patanjah’s 
seitens iSakati.yana beruht, zeigt, dass Hemacandra seinem Vorganger auch 
m Fehlem Mgt Doch zeigen die Sutras Hemacandra’s 1. 1. 25, 26 , 2. 3. 14 
uud andere, die Lehren des MahSbhai?ya emeuern, auf wddie Sakalayana 
nicht Bezug nimmt, dass der grosse Jama^KompiIator gelegentlich auch die 
alteren Quellen benutzt hat 


Zur Herausagabe dieses Specimens der €akatayana-Grammatik nebst dem 
Kommentar Crntamami habe ich die folgenden Manuskripte benutzt.i^ 

B = London, India Office Biihler MSS 138, 141, 142, 143. Aufrecht, Cat. 
Cat. 1, 638 , Buhler, Two lists of Sanskrit Manuscripts, ZDMG, 42, 
544. Ein ganz junges Papiermanusknpt in EtevanagaiS Schrift, namlich 
die von BiiHiER veranlasste “ Umsdirift eines alten Hala-Kamata-Mscpt 
der Madras-E. T H. [llj library im alphabetischen Catalogs mit nro. 
1083 bezeichnet”^® — schon und deutlich geschneben und im grossen 
ganzen fdilerfrei. Leider ist es unvollstandig, da es m der Mitte des 
42, SQtra des 3 PSda des 1. AdhySya abbncht” Im 2. Pada hat der 
Schreiber mehiere Lhcken — haufig von betrachtlicher Ausddmung — 

P = London, India QEBce, Mackenzie Coll. XII. 8. Wilson’s Catcd. Vol. 1, 
S 160 No. XXXIV. — Vorzughches Manusknpt, sorgfaltig geschrieben 

IT Ausserdem habe idi noch die von dem Heraustgidier der Bombayer Auagabe 
des Ptakriyasarngraha in den Anmerfcungen gdegentlich zitierten Ausziige aus dem 
Cmiamam benutzt und veiglicfaen. Die Gottmger HSS. der SakainyanarGiaimnatik 
enthaltsi nur den Teict der Sutras^ dn alphabetisches Verzdchnis der Sutras und 
eine modeme Abscbnft des PTaknySsarfigrako Die and bier weiter nicht beiiuck- 
Bichtigt worden 

w BdHtER, Uber die Grammatik des Cakatayana, Or. Occ 2, 691. 

19 Dass das Berlmer Manudmpt des CmtSmani (Ms. or. fd. 872, Weber, 
Verzeichnias Bd 2 [1886], S. 205) eme Abechnft des MS. B ist, ergibt sich aus 
folgendax Erwagungen Es reicht gjeidifalls Ins zu 1. 3 42. Die Mehrzahl soinpy 
Fdiler sind in B schon vorhanden, andere lassen sich durch die typograiAischen 
EigentOmlichkeiten von B eiklaren ; urn nur zwd von den letzteren bprommigrpifam : 
B zeigt Formen von fa und tra, die na re^ pra sehr ahnlich sind. Ifeufig gibt 
das BerHner MS. tatsachhdi jene Budistaben mit diesen wiedor, so z. B. gleich das 
Wort des Kom. napra fur tatra. In einem; Fdle flndet sich dassdbe Zdchen 
^ fur jho m beiden MSS. 



vind fast fdderfrei. Es ist erne kurzere Rezension des Kom, indem die 
udaharanas und die pratyudahorai^as und was sonst einen vdlstkndigen 
Kom.®® zugehort, ausgelassen wird. enthall mit einer Ausnahme nur 
eine Paraphrase dei Sutras mit Angabe der Worter, die durch amvjtti 
fortgelten. Est ist gut erhalten, abgeseheii davon, dass der obere Rand 
vtm einigen 50 Blattem am Anfang beschadigt ist, wodurch aber in der 
Regd nur der Anfang der ersten bezw. der letzten Zeile gelitten hat. 

H = London, Indian Office, Bubnexl Sanskrit Manuscnpt No. 405, die von 
Burnell veranlasste Umschnft in Tdugu eines MS. Hala Earuata- 
Schrift.®^ Es ist vollstandig, aber voUer Fehler. Die ersten Seiten sind von 
anderer Hand korrigierit worden. In dem Verzachnis der variae lectiones 
(p 46 ff.) sind die Verbesserungen vorausgesetzt. 

Die did Handschriften, disdion sie der Hauptsache nach von einander 
wenig abwdch^, sind unabhangig von einander. Bd der Feststeilung des 
Textes bin ich hauptsachlich B gefolgt. Idi habe ea ftir unzweckmSssig 
gehalten, jeden Fehler von H zu verzeichn^i, da die Anzahl von solchen un- 
gemem gross ist; die abweichenden Lesarten von P aber sind vollst^dig 
ang^eben H^ufig verwechsdt H postkonsonantisches o mit a, dh mit d, 
V mit d Hingegen schreibt B haufig t fat k, v fat p und steits lu fur das 
sonantisdie I Betrefifs der Verdoppelung von Konsonanten verhalten sich die 
MSS ausserst inkonsequent Das ava£raha-Zachm wird m B regelmkssig 
ausgelassen, in P aber in der Regei eingesetzt. B gibt {12} das Zeidien (3) 
fill die Plutierung durdi nu wieder, was dch aus grossen Ahnlichkeit der 
Zeichen in Hala Karnataka erklart 

Man wird in meinem Specimen sehr oft die sandht-Reg^ verletzt ffiidai. 
Bd der Entscheidung bin idi in der Regd dem Manuskript B gefolgt, doch 
habe ich mich stets dutch die Deutlichkeit Idten lassen. Nach dem Voi^ang 
Kielhorns, und zwar aus dem MBbSs. Vol. 1, Einleitung S. 9 f. angegebenen 
Grunde, habe ich die Verdoppdung vrai n, n und n zwisdien Vokalen durch- 
gangig unterlassen. 


der von mir benutzten Textausgaboi. 

Pinims Grammatik, herauagegeben, iibersetzt, erlfiutert usw. von Otto BbmLiKGK, 
Leipzig 1887. 

VyQkar(ina'Mt(hSbhdshy/i of Patahjali edited by F. Kielhckin, Vol. 1. 2. 3, Bombay 

MaMbhdshya by Patanjali Mum with M. M KaiyaVipadhyaya’s Pradipa and 
M M. NagojlUiatta’s uddyota [Babarana] edited by Pandit Devi Datta 
Pasajuli. Ctiawldi^ba Sanskrit Sales, Benares 1908. 

®° Ich verwdse auf die dnleitenden Strophen 7, 10 und 11 

®'' Nadi der handscriftlidioi Angabe Bubnelis auf dem Titelbatt des MS. 



KMika, edited by Pandit BEla SastrI Second Edition, Benareal 1898.. 

PanbJidskendusekhma of Nagojibhatta edited and explained by Kielhotn. Part. 
1, The Sandnit Text and various readings, Bombay 1868. Part. 2, Transla- 
tion and Notes, Bombay 1874 

Stddh^akmmudi with the Tattvabodhini Commentary of Jnanendra Saraavati 
and the Subodh^ Commentary of Jayakpshna edited by V^dev 
IjilcahtT^an Siastri PAijiglKAB. Fourth EtStion. Nimaya^gar Press, Bonhay 

Laghukamudi ed. by James R. Ballantyne. Fourth edition, Benares 1891. 

CmdrorVyakarima herausg^eben von Bruno Liebicr [ = Abhandlungen fiir die 
Kunde des MorgeDlandes, XI. Bd No. 4], Leipzig 1902. 

Sttkatdyatta-VyakaTaaa mat dem PraknyasamiTaha von Abhayacandra Siiii, Bom- 
bay 1907. [Es wird hier nach dieser und racht nach der Oppert'sdien 
Ausgabe zitiert.] 

Siddhahemlsic’l-SabdanuSSna by Kalikfila SarvajnaHSri-Hemacandrfichfiryavarya 
.. BenaresI 1905. 

Nur die folgenAn AWcurzungien bedurfoi besondwer Erwhhnung ; 

P. = Paijim , CJ= Candra ; S,'= ffikatSyana ; H.= Hemacandra ; 
Va. = VSrttika ; BL = MahSbhisya. 

£ 13 ) 

1. TEIL. 

Text der Sutras nebst dem Kommaxtar. 

II tfiMataiaya namh || 

4riyain kriyad vah sarvajnajnanajyotir ana4vanm ] 
vi4vaip praka4ayam4 cintama]ai4 cintarthasadhanah 1| 1, H 
namas tatoahprabhavabhibbiitabhudyotahetave | 
lokopakSnitie 4abdabrahinaii}e dv&da4atmane || 2 || 
svasti 4tfsakalaj!66nasaim|ajyapadani optavan | 
tnahai4rama^sam^dhipatir yah ^katayanah || 3 || 
ekah eabdambudhiip buddhimandareina pramathya yah | 
saya4a§4ri samuddadhre vx4vam vyakara|nSinrtam 1| 4 |j 
svalpagranthaip sukhopayam saippuitpam yadupakramam [ 
4abdanu£sanam sarvahii arhaccMsanavat param || 5 || 
i§tir ne9ta na vaktavyaira vaktavyam sutratab pithak | 
samkhyataip nopasaipkhyonam yasya 4abdBnu4asane (| 6 |1 
taay&timahaGqi yrttim samhTtyeyaip lagbiyasi | 
saippiuipalaksaioa vjttir vak§yate yaksavarmaiija |1 7 jj 
granthavistarabUiiniipiam sukumaradhiygm ayam j 
4u4FQ^igu]]an kartuip ^stre saipharaiciodyainab 1| 8 || 
4abdBnu4a8ana8ySnvarMy&§ dntamaiijer idam ( 
vxtter granthapramanam tu satsahasram nirQpitam 1| 9 1| 
indiacandiSdibhih 4abdair yad uktaip iabdala^apam | 
tad ihgfiti samastaip ca yan neMsti na tat kvacit || 10 t| 
gapacMtupSthayor gaipadhatiH 
liing^u4gsane Imgagatam | 
auibadikSn unjSdau 4ei$axa 
niSSeeam atra vrttau vidyB,t || 11 || 

(14) baliabaEjano 'py asyla vitter abhySsavjttitah | 
samastaiji yaiunayam vetti var§ejjaikena ni4cay§t |1 12 j] 

tatra sutrasyadSv ayaip mangala41okah | 
nama^ Snvardhatnandya prabuddhSSeiovastave | 
yma Sabddrthasaritbandhah sarvetfo Sunifu^tSh {| 

4a b dia r t h a 8 a t pbandha vScakavacyayogyatfib | athava. SgamaprayojanopByo- 
peyabh&vBh te yena sarvasattvahitena tattvatah prajnBpitalb §rimate 

mahaviiaya aekeatfcrtasakaladravyfiya namalj | namaskaronflty adhyaMrab | 
td vighnapraiamaniarthanii arbaddevatatiamasharatp paramamaibgalain fira’ 



bhya Wiagavan acaryah ■SkatSyanah SabdanuSasanaip Sastram idaip prSra- 
bhate || 

dharmBrthakamaniokgesu tattvarthavagatir yata]j | 
iSabdarthajiianapurveti vedyam vyfikaraijaip budhaih |1 
aiun\rk\eon\ai auc | ha ya va ra laB ] 
na ma na nam I ja ba ga da daS | jka bha gha dha 
dhas I kha pha cha tha that |i <ra ta tav \ ka pay | 

Sa sa s am ah X ka x | bal |) 13 || 
iti vaiJjasamlainiiayah ] kramanubaiidhopadlanah pratySharayan fiSstrasya 
laghavarthah \ sSmanySiSrayaiiad dlrghaplutanunasikagrahapaip | 
hrasvadirghaplutahalo by ekadvitryardhanfitrikaih 1 
iSsilram anuyata^ ca vanqab sylad anunasika b 1 1 
uccair udatto nScaili syad anudSttah e^raras tatha | 
vyami4rah svanto jfieyah pratyekam vibudhair iha ] | 

r ity anena pvarnasySpi grahaiijam bhavati 1 diuiad SmaBtryasya gurur vaiko 
lanrt [2. 3. 27,] iti Igrahaajat ] tatha ca | rty akajj [I. 1. 75] iti Jkaie 'pi dd- 
dhato I hak§rasya dvir upadeSo a 5 adau vaMdau^' ca grahaparthab ] hakgiS. 
di 5 v akaradaya uccaraparthah [j 

sStmetet || 1 1| 

sarpjfSasutram etat | ita sahoaaryamSiDO vaipah samudayo va atmangh 
prabhrty a tasmad ito vyavasthitanBip sajpjfia bhavaty atmaiiB sahaj ap ] 
ak I ac I hal I sup | sut | tin | ptasu®' jj 

saipiflaniyamarasedhadhikaranityapavadavidhiparibhasah | 

atidcSavikalpav iti gatayahi dabdinuSasane wntrair^'m || 

tl5} uta svah ||2|| 

uktrojeta sahopadiybnano vaipah svasya vaigasya sajpjfia bhavaty 

atmana saha I ku I cu I tu I tu 1 pu 1 1 

teyan ||3|| 

_ takar^eta sahop§diyaniano vaipa iySn | yavanmatra upfittas tavanmatra 
evBsau veditavyah | at | it ] ut I| 


vedi^Tl’ 1 ,^?^ PratyajaviBiagamaiapai agaldralm vawa ijto eva 
I bh^suh i 38 ,=. I asBbba, I wta II ag iti Bm II amm 1 

aprayofit || 5 
ihopadi^yamano vaipab saWudSyo vS 
diiSyatB sa itsatnjfio bhavati | edhi I edhate 1 
dukja I kjtriman || 


yo laukike 4abdaprayoge na 
aiup I an | tuveprii | v^athub | 


* 7 . 1 . 39 und Cintam. dazu. 

» J. 2. 41. 

1. TEIL H 

&vah sthanSsyatkye 11 6 || 

sthanaip kaiQt^di 1 3syaim mukham 1 oQthat prabhfti piSk kfikalak&t | 
tatra bhavam sprejtatadi prayatnapaficakam ^syara 1 kanta^tbamurdhaji- 
hvadantorastSlunasika vamianaith sthaiilaQy asyaim spp^te^at^^ir^taviyrtasai)!- 
vrtesadvivrtam 1 tayor abhede varujo vanjasya svo nama veditavyab 
akuhavisarjaifiyajihvarauEyaJb kaKlhylah 1 kur jihvamflle 1 havisarjaniyav 
urasyau J jihvSmiySyo jihvyah 1 sarvalmukhasthanam avamam ity dee ] 
leaicuya^s tSlavySb 1 eai kaijiliiiatalavyQv eke$am 1 uoaupupadhm&nlyS 
o§thy@h 1 oau kanitiiajthyav, djesaip 1 vo dantosthya]^ 1 srk vasthanam eke§am | 
tturasamurdhanySli | rgriio dantamula ekfigam 1 Jtulasa dantyajj 1 nasikyo 
’nusvarah 1 kanthanlsikya eke$iam 1| asyam H sp|i$tam karajijatp spatiSan^ii | 
IjatepTstam antassthanam 1 vivitam usimnam svaianam ca | eo vivitatarau 1 
tabhyam aiau | tSbhylam avan^ { l^adviviiiam u^uiapSm | saipvrtam 
akarasya 1 d a k ity akaralj udiatto ’nudSttalii svantaS eSnunSsiko 'nanuna.- 
siltag ceti I evaip (firg^plutiv iti dv&da'§^vai|aabhec@h parasparasya sve 
bhavanti 1 evam ivaiinadSnaitp tv [16} astadaSa bhedSJj 1 Ivamasyanukaranad 
anyatra dlrgho nasidti dvada^a bhediab | eoam hrasvabhUviad dvadaila bbe(^ | 
yavaMnSm anumsiko ’nanunasikaS ceti dvau bhedau | vargyaib panca panca | 
rephoi$maii;jBip eve na santi 1 1 

asamtah 1| 7 H 

MsaimanSsaimaprasange sthanagusTaprainapadibhir yathasvam asanna 
eva vidhir upStto veditavyaHj 1| tatra sthanena |1 dirghaiji ,[J. 1. 77] 1 IdSg- 
ram 1 mui^drab H gimena H kte 'nitcajah kur gghiti [4 1. 171]' | pSk^ 1 
tyagalh 1 cakgrasyagho^asyalpapigioasya 1adr§a eva kakaro bbavati | jakar- 
asya gha$avato ’IpapiSnasya tSdeSa eva gakSro bhavati H pran£|nena H do 
mo ’sytdaso mSd gu§ cSi^my asan [2. 2. 44] J amui^mai 1 amubbyim 1 
matribasya matnkdb 1 dvimfitrasya dvimatraji H arthena H ministry ekar- 
thayoh stryanyata [2 2. 41] 1 vStanjdyayuvatiib 1 d&radavrnd&TiiS | 
vata|ni$4abdasya apatyarthasya tadartho viataodyabhavaib | daracdiabdasya 

dSradab ll 

satfibandkinatn sembandhe 1 1 8 1 1 

eaipbandhidabdatigiii yat karyatn acyate tat sapibandhe saty eva bhavati 
nanyatra 1 evaiui&d yab [2 4. 94] Iva'Suryab | saipjfiayaip dvaiiiuad in'* 
eva I dvBguriih || 

gha4dal:i samkhya 1|9|1 

ghatudatipratyaygntaip*’ saipkhyavad bhavati | dmriifca eaipkhyS 
tatkaiyapd pratipadyata ity arthah 1 y&vatkam 1 ytvaddha 1 yavatkjtvajb. 1 
yatidha | yatikjivah® H 

* 2. 4. 21. 

» 3. 3. 68-71. 

® 3. 4. 27, 32. 


bahugco^atfi bhede 1 1 10 1 1 

bahugapa ity etau Sabdau bhede vartanfinau saiiikhyavad bliavatah | 
bhedo nadatvam dcatvapratiyogi | bahukah | bahudha ) bahukrtvaJj* j 
gainakaib | gaiaadha | ganak^tvalh” || bbeda iti kim || vaipulyc saqighe ca 
mS. bhut II 

kasamas^ 'dhyardkab^ || 11 1 | 

adhyardhaiiabdaibi kapratyaye vicMtavye samase ca samkhyavad bhavati | 
adhyardhakam 1 adhyardhaiSfirpaip krite || pratyayasya dvigoh^ §Iuk || 

|[17} ardhapurmpado dot 1| 12 || 

ardhapurvapado idatpratyaySntah ^bdah kasamgsayoh saipkhyavad 
bhavati | dad iti satpkhyapurape dat [3. 3, 76] ity arabhya S. dvitres® 
tTyatas takarepa pratyaharah | ardhapaficamakam | ardhapancamaSurpam || 

pautrddi vfddham 1 1 13 1 1 

paramaprakrter apatyavatah yat pautrSdy apatyaip tad vrddhasaipjiiatn 
bhavati | gargasyapatyajp pautrSdi gprgyah | vatsyah* | anantai&patyaip 
gargih I vatsir^*' ity eva bhavati || 

prapautrddy astn varnhajydyobkrdiroh sati yuvd || 14|1 

prapautraJj pautiapatyaip paramaprakrtes caturthaij | vapiSe bhavo 
vaitp^ab pitidir atinanah k&ranam j jySy&n bhiata vayo’dhika ekapittka 
ekaai&tjrko vB | paramaprakiteh prapautcady apatyarp stiivaqilaip vaiplye 
sati jfvati puttiadi jySyasi ca bhiBtari kanlySn bhtata yuvasairpjiio bhavati | 
gargyiayaiijah | vatsyayanab^^ 1 1 paramapiakrtir gargah | tasySnantarapatyaip 
gargb'’® I tadanantatBpatyam vrddbo gBrgyab* trtSyalj syat | caturdio 
^gyayanjo yuva 1 1 prapautiBdib kim 1 1 pautro glargyaJh | j astriti kim 1 1 stri 

soi sapiifde ’dhhayassthane v3 || 15 ]| 

yayofe pQrvab saptamah puru^a dias tSv anyonyasya sapipdau | vayo 
yauvauadi | sthSnaip pita putra ity&di ] paramaprakrteb prapautrady 
apat^ strtvarjitam vayassthanSbhyaip dvabhygm apy adhike sapipde 
jivati sati saj jlvad eva yuvasaipjfiaiiii vB bhavati | pitrvya pitfvyasya pitan 
pitamahe putre va vayo’dhike jlvati gBrgyasyapatyajp jivad gBrgyah gSrgyB- 
SltirS* vBtsyayanoii vB || sad ityadi kim || anyatia gBrgyah® H 

yuvavrddhcun kutsSrce || 16|| 
yuvB ca vrddhajn cBpatyam yathBkramaip kutsByBm 
yuvasamjfiam va bhavati | gargasyBpatyaap yuvB kutsito 

aicaySip ca visaye 
^gyah I gBrgySi- 

» 2. 4. 18, 38 
« 1. 3. 15. 

» 3. 3. 64, 
« 2, 4. 21. 

» 3.3.86. 
« 2. 4. 33. 

1. TEIL 


yaino^® va H jalmalj guruman bhutva svatantra ucyate ] anyatra gargy6yana 
eva 1 1 gargasyiapatyaiji vrddham ardtaip gfirgySyanah [ gargya vS. ] anyatra 
gargya eva |1 

[18]} nSnta dul^ 1 1 17 1 1 

yan namadhyeyam satpvyavahaiaya hathan niyujyate devadattadi tad 
dusamjnarp va bhavati | devadatfiyihi* ] daivadatfgh || 

tyadadih |] 18 ]| 

tyadadayab §abda. nityam dusamjfiS. bhavanti | tyadlyam^^ | tadlyam | 
kimiyam | ladayanih^s | yadayamb || tyadiidilj saiv&dyantargaijah || 

yasyak$v ddir adaic || 19 {| 

yasya ^abdasyaaam madhyc adir ac Skara aij va sa dusaipjfio bhavati | 
amraguptayamhis | lamba^thyab^® | sauviryah^ | aitMyailyalhw ] aupagavi- 
yahi‘ll . 

deia emtn chadau 1| 20 || 

deia eva vartamanasya yasya SabdasyBcam adir en bhavati sa chadan 
pratyaye vidiratavye dusaTjijno bhavati | saipurikl | saipurik& | skaxmaganki | 
skaunagankfi ] sepuram skonagaratp ca b&hikagtmnau || evak^ niyamar- 
ttiab I tena defe 'nyarthatra ca vartamanasya na bhavati 1| chadav iti kim || 
phifiladau na bhavati ]| 

pragdeSe |j 21 |1 

ptggde§e vartam^asya yasya 4abdaisy&cam adir eh sa cMdau pratyaire 
dusatpjho bhavati | §ai3vata nama nad! | tadapek^ piSgudagvyavastha [ 
enipacaimyah^* | gonardiyab ] ekacakrakah 1 niyamanivirttyartham vaca- 
nam || 

kriy ortho dhStuh || 22 ]] 

kriyapravrttih piirvSparibhuta sadhyamanarupa [ sa artho ’bhidheyaip 
yasya sa 6abdo dh&tusaipjfio bhavati | bhil | tdiavati | edhi | edhate g(^ya | 
gopayati I papacya | pgpacyate | patrakainya^^ | putrakamyati |1 Sistapra- 
yog^usSiitvSl laksaipasya alnapayatyadinivTttih || 

dadha ghv ab || 23 || 

dSdhafupopalaksito yo dhStuh so ’bakaranubandho glwsaipifio bhavati ] 
darQpa'S catv&rah | dhanupau dvau || dan? ] praajidaiS | den | prani- 
dayate | idudaiB. [ praipidadati | do | prapidyati | dhet j praajidhayati ] dadlian | 
praipidadhSti^^ 1 1 ab iti tam 1 1 dab [ d^taip baifailh I daib | avadataip mu- 
kham )| 

IS 2. 4 . 38, 33, 20. 3. 1. 26. 3. 1. 28. 

Cf. P. 4. 1. 149, 171. ” 4. 1. 17. w J 2. 77. 



prMir mpraiyaye || 24 || 

piadiih svaradyaiitargairialj j sa na dhatuh | dhStoi avayavo na bhavati ] 
lam vyudasya tatah para eva dhatusamjfio veditavyalh 1 apratyaye \ na cet 
t ptah paraJi pratyayo bhavati | abhyamanayatai® | abhimimai©yi?ate I abhi- 
manayya®® | pi&sadlyat ] prasisadlyjsati J prasadiyya 1| apratyaya iti kim || 
autsukayata‘‘i | utsukayi?ate | utaukayitvSl®* |i aaamgramayatasura ity atra 
saiiigrama ity etavian yuddhSitho dhatur natra sam piadiai || 

tasySgatarthddhiparyarcasvatyattkramdty upasargah pf&k ca \\ 25 \\ 

tasya dhlatdh sambandhi tadarthadyoti pradir upasargasamjno bhavati 
prak ca tato dhator bhavati | yau gatarthSv adhipan iti yau cSrcaAdsayau 
suati iti yaS catikramavisayah ati iti ten varjayitva | pialambhah®-* | 
parinaniati** | abhi?i!5cati*“ |1 upaaargatvena ir&di || tasyeti kim || vrksam 
vrkisam abhi sicyate || agatetyadi kim || adhySgacchati j agacchaty adhi j 
parydgacchati ] Sgacchati pan | adhyagamanikah | paryanitam | upari 
bhavasya sarvatobhavasya ca prakairanSdeh pratipattau gatarthatvam | apra- 
lipattau adhyagacdiati paryagacchatlti pr^ktvam eva || arcasvati 1] su sik- 
tam bhavatfi | ati stutam bhavata | atra dhStvarthah praiSasyate | anyatra 
susiktatn bhavateti kutayate || atikramati || ati siktam eva bhavalS [ ati 
stutvS I yadartham kriyia tasmin kSiye nDSpanne 'pi knySpravrttir atikra- 
mah I an 3 ?atra ati^ayya 1 1 piSk cety adhikHrah prSg avyayasamjnayah 1 1 

daccvyuryadyanukaraijaiti ca 1 1 26 1 1 

dajantam cvyantam uiityevamddy anukaramam upasargasamjfiaip ca 
dhatoih sambandhi tisatpjfiam bhavati || dac || patapaltikrtya^® | sapattra- 
kftya II cvi II 6uklikirtya®‘* | ghaiBkrtya || luryadi || luilkrtya | uraiikTtya'® || 
anukaranam || khatkrtya | phatkrtya®* || upasargah || prakrtya parihrtyaso | 
cviidacs6dhannyad uryadinSm krbhvastibhir eva yoge tisaitpjfia. || uryadayo 
ganapgithe drai^tavyalh || 

{^ 20 } karikMamado’nialfsadasat sthityadibhmonupadeidpan- 
grahddarak?epe || 27 || 

sthity&dau UiusS. anupadeSa-apangtaha-ddara-ksepa ity et€i$u caithe$u 
yathesalmkhyam karika-alam-adas-antar-sat-asat ity ete Sabda dhlatos tisamjna 
bhavanti || sthitir maiyiada vrttir va. | ddii§abcldd yatnadi grhyate | tatra 
kankakjtya®® |l bhusd mapdanam | tatra alamfcrtya** || svayam paidmarso 
’nupadeSab | tatra adabkrtya“ || paiigrahah svikarah I tadabhave antar- 
hatya*« || fidarah piitya sambhriunab | tatra satbrtya»« || ksepah paribha- 
vab I tatra asatkjiya®® || sthityadBv iti kim |1 karikStp brtvS | kartifm ity 
arthab I alam krtvS | ma karity arthah | adab krtvB gatab 1 ayaip parasyo- 

2 ® 2 2 . 171 . 21 3. 3 113 . 

4. 2 . 207 . 24 4 . 2 23 ». 

*« 2 . 2 . 171 . 

4 2 . 131 
*2 4 . 4 . 144 . 
“ 4. 2 . 219 . 


padeSah | antar hatvS mQ§ikaip ^yeio gatal;i | parigrhyety arthalj 1 sat krtva j 
vidyamlanam ity arthalj | asat kjtva | avidyamanam ity arthah jj 

kanemanah sraddhocchede || 28 || 

kaiQie-inanas ity etau §abdau Sraddhaya abhilasasyocchede dhatos lisatp- 
jfiau bhavatah || kanehatya^^ | manohatya j Sraddl:rain ucchidyety arthah | 
anyatra tanjidulasya kafje hatv& mano hatva. gatah 1 1 

ostampuro 'vyayam 1 1 29 1 1 

astam-puras ity etav avyayau dhatos tisamjiiiau bhavatah ] astaipgatya 1 
puraslqtya''*’' | astam iti ria§e vartate || anavyaye | astam krtva fc?iptatn ily 
arthaih | purah fcftva nagarir ity arthah jj 

gatyarthavado ’cchah |1 30 H 

accha ity etad avyayam abhi^abdirthe dndharthe ca vartate | tad gat- 
yarthasya vade4 ca dhatdh satnbandhi tisamjiRaiii bhavati j acchagatya®’ j 
acchavrajya ] acchodya || avyayam iti Idm || accham udakam gatva || 

tiro ’ntardhau || 31 || 

tiras ity etad antardhau vyavadhSne vartamSnaip dhatoli sambandhi 
tisaipjfSam bhavati | tirobhuya®" [ tirodhaya 1 anyatra tiro bhiitva sthitah 1 
tiryag bhutvety arthah 1 1 

kriio va || 32 |j 

tiras ity etad antardhau vartamanam krno dhStoh sambandhi tisaqi- 
jnam va bhavati j tiraskrtya*® | tirahkrtya ] tirah krtva ( 1 anyatra tirah krtva 1 1 
» plj monasywasyupa)tfnva}emadhyepademvacane || 33 1| 

manasi-urasi-upaje-anvaje-madhye-pade-mvacane ity etany avyayam 
krfio dhatoh sambandhini tisaipjnani v& bhavanti | urasi manasi anatya- 
dhanavi$aye | atyadhanam upai§lesa ticaryam ca | manasikrtya^” | manasi 
krtva I urasikrtya | urasi krtva | upajekrtya | upaje krtva | anvajdcrtya | anvaje- 
krtvS i' madhyekrtya | madhye k?rtva | paddcrtya | pade fcrtvS 1 nivacane* 
krtya j nivacane krtva j] 

svamye ’dhih || 34 || 

adhir ity ayaitl updsargah svamibhavavi$aye kudo dhatoh satpbandh! 
tisaipjho bhavati vfi | devadattam giame 'dhikrtya*® | adhi krtvS | sviiranaip 
krtvety arthah | anyatra acity adhikrtya®* j) piadir®* upasarga*^ iti. vartate | 
tenopaaargasaijijinapi vikalpyata iti krtvadMti priaktvasyaniyamah |[ 

saksadddy acvi || 35 |]. 

sahsBdityadi gabdarupam acvi cvyartbam acvyantam krfiah sambandhi 

1 . 1. 168. 1 . 1. 167. s® 2. 2- 171. 

i. 1. 24. i. 1. 26. 



Usatpjfiaqi v§ bhavati | sgki^tkrtjra^’' | siak$at krtva | mithySkrtya | mithy& 
krtva j[ acvSti kim |{ lavaiQikirtya | UQilkttya^^ || 

nityom hastepanau svlkrtau || 36 || 

haste-panav ity etav avyayau attnlylkaraoe krnab sagibha ndhi nan 
tisa^jiSaa rntyaip bhavatah | hastdqrtya | p3|Qauk]tya^o | anyatra haste kltva 
kai;$3pa!nam gatah i | 

jivtkopani?ttd ive |1 37 || 

jIvika-upaniGad ity etau ^hdau ivSrthb gamyamane krnah saipbandhi- 
nau tisaapjfiau lutyam bhavatah j jlvikam iva kitva jivikakrtya*® 1 jlvifca- 
karoti** I upajoi^atlcFtya | upani$atkaroti [ anyatra jivikam Iqtva | upani^adaip 
kitv6 II 

pradhvam bandhe |l 38 |] 

pridhvam ity etad makatiantam avyayam anukulye vartate | tadanu- 
kulye bandhahetuke vartanianaip IqSiah sarpbandhi tisaipjfiarp bhavati | pr5- 
dhvamkrtya*® i anyatra pragatam adhvanatp piSdhvam fcitva Sakatam 
gatah II 

{[22} tasvannamadhaniasyStriktvaijitutritisunptasvdbhasvaTMlny 
avyayam || 39 || 

tas-vat-n&m-ityetadantSni dhainvarjitatasyantlani am-kjrtva-am-tum ityeta- 
danteni tisajpjinSni suhptasupratirupBini svaiadini ca labdarupany avyaya- 
sainji^i bhavanb || tas || eka±^talh pSlum^ato®® vidyotate || vat || munivad 
vittam II ham || uccaistaraam |] adhaititasi || ramatah^^ | iSvafpatah | sarvatah | 
saiVatra®® | bahu^ | ta^ti vya4raye las [3. 4 . 4] ity larabhya 6aser®® ikSrena 
pratyaharah H adhan iti kim j] pathidvaidhani | 8am§ayatraidh5ni®^ || &m || 
daySipcakre | vidSipkarotu®® || ktva || krtvi. ] hrtvt || am || piirvambhojarp | 
kanjfidanSaip®® varayati || turn || kartum | hartum H ti |j adahkrtya j 
avyaySn na sift || suihaWiaih |j ihtrau | velaylam | asti | syat || ptasv&bhah || 
yadiS. I tatha | katham | kutah | ptasu iti ptaspratyayBd*® arabhya B katham 
itthamor [3. 4. «;] ukBreipa pratyBhBrah (j svaiadih [ svas tiethati | dntas 
ti$thati I upa kafoti || 

sadr^aqpL tri§u lmge§u sarvasu ca vibhaktiju \ 
vacane&u ca sarve^u yan na vyeti tad avyayam ||' 

, tas&digta h aip ai p kim 1 1 ekah | dvau | bahavah | Bpah | vati^ah 1 1 svaradayo 
gaioapaithe dra^tavy^ || 

ghy asakhyadvandvapatldut |)401| 

ika^ntam ufcarantam ca &bdarQpam ghisamjaam bhavati sakhifejbdaip 
dvandvBnavayavaip ca patifiabdaip varjayitva | munina | sadhuM | muni- 

*» 2 . 2 . 1 . 
« 3. 4. 17. 
»« J. 4. 83, 

3. 1. 179, 180. 

3. 4. 64. 

4. 4. 150, 

3. 4. 4. 
3. 4. 31. 
*0 3. 4. 12. 

1. TEIL 


sutau I sadhuguptau*^ ll asakhyadvandvapatiti kim || sakhya | sakhye ] 
patya | patye || advandvagrahainam kim |1 patisutau ] pjatisakhayau |] 
patisakhi^bdayor ayam pratisedhah | na samudayasya | teneha bhavaty 
eva I atisakher ggacchab | bahupateib svam |{ 

pratyayah kjto ’sasthy^ 1 1 41 1 ] 

iha yah krto vihitaJi sa pratyayasamjno veditavyah | a^asthyah [ §asthy- 
anlarthah $a$thl | na cet sa $ai$lhyantarthasya vihito bhavati | aganio 
vikaro vety arthali | ni | [ su au jas*® | vrksah vrksau vrtosSh [ | 

^[23} ijttdy aydt taddhitah 1 1 42 ] | 

i] yuddhe [2 1 135] ity arabhya gupaudbubvicchipanpaner ayah [4. 1. 
1] ity Byapratyayat prag yat pratyayasaipjnajn tat taddhitasamjnam 
bhavati | ke§ake§i*'' aupagavah^^ |1 

ghyMy 'atin kit 1 1 43 [ | 

ghyanfidi*® pratyayaSamjfiam tmvarjitam krtsamjfiatp bhavati | shana 
gtiatyah godByo vrajati || atin iti kim || pranimste j] 

parah 1 1 44 1 1 

yah pratyayah sa prakrtdi para eva bhavati | vrhgah | vrksau ] vrkwh 1 1 
mid aco ’ntydt H 45 |1 

maklaianubandhako yasya vidhlyate tasyacSm antyBt paro bhavati | 
vandate 1 vanani^®’ [[ 

spardhe 1 1 46 ( ] 

dvayor vidhyor anyatra savaka§ayos tulyabalayor ekatra vinipatah 
spardhas tatra yah sutrapajthe parah sa vidhir bhavati ) utvam" | ko 
basati I ko dhavati | luk*® | esa karoti | sa saratiti ubhayapraptau paratval 
luk I e$a hasati | sa dhavati || 

parajp syat purvaparayor nityam syal paranityayoh | 
nityat tathantarangam syBt tato ’py anavakaiSakam || 

softhyah sthme ’nte ’lah || 47 || 

^(hyantarthasya vidhlyaixiBno vidhis tasya yo ’ntyo ’1 tasya stiiane 
prasango bhavatiti veditavyam ] napo ’co hrsvah [!■ 2. 1] giBmaini kulam | 
senlam kulam || 

tasmad adeh l|48|| 

tasmad ^agthySh paBcamiviSistasya easthyantarthasya parasya vidhi- 

« 2 . 1 . 119 . 
« 2 . 3 . 82 
« 1 . 1 . 157 . 

*2 J . 3 . 97 . 
« 4 3 . 60 . 
*8 1 . 1 . 158 . 

« 2 . 1 135 . 

88 1 . 2 . 11 . 


yamlano vidhis tasySder sis]} sthSne bhavati ] dvyaDtarupasarg&d id apo 
’nat [2. 2. 138] [ dvipam ] antaiipam H 

[24} Manidid ||49|| 

&c caAidal cade§^ ga^thyani^irthasya tasyaiva atMne bhavati nSdet 
antasya vaiah || 6it 1| jaiSasaJi m [I. 2. 1&] vanani | dhanlani || afiidal \\ 
sam amali [i. 2 176] ] sarvesam | viivesSm |1 nidalparyudBsah kim || jaraya 
nas« 1 jaraaa ] jarase | jho 'ntab [7- 4. 88] iti yajj sa nirdi'SyamanasyadeSah || 

sthamvanalasraye ||50|| 

yasya ya yidhjyate sa sthanS j itara adegab | stMnIva bhavaty 

arfftSqb I sthamkaryam pratipadyata arthah | anaJggraye | na cet tat 
karyatn sti^yaHugrayarp bhavati ] yuva | mja*® | suval lope 'pi ny ak 
[ 1 , 2 134 ] iti djrghaih padatvadi ca | kasmai | kasmat | kiipvat sarvaditvat 
sm&y&di^ II anaMiaya iti kim |1 sah | panthah®^ | atm athamVattvabhavat 
halaji paratvalafcsapa sor lug na bhavati®® || &§rayagrahainam kim |1 
pradivya | prasivya !■ valader id na bhavati®* || 

pare’caf} pnraco ’kvidirghayadvyesadasklugvidhou j| 51 || 

ajadeSab paranimittakas tatab purvavidhau kartavye sthamvad bhavati | 
kvlvidhup dlighas}^ vidhnp yakarasya vidhiip dvitvasya vidhim & etasmad 
arabhya domo ’sySdaso mgd yu$ o§i$iny asan [i. 2, 44] ity asadadhikarad 
yo vidhih sajmyogasyadiskor lug [i. 2. 91] iti lugvaijitas taip ca varjayitvia | 
kathayaP | avadhit || atiallugupantyavidhau kartavye ethSnivad bhavati || 
padikab®® 1| atra padbhave kvividhyadipratisedhah kim || devayater dyub I 
lavam aca$te lavayater lauh | atia Qilugallopau kvmdhav uci na sthSnivat |i 
^amajtp Simam | aigmi | 4aip£imaip daip^mam a^ajp^Smi | atra ipyantat 
niyahpyantat ca khamunihau |rulugallucau dir^bavidhau || sauii balaka | 
brahmajnakainiddtih | atra allug yavidhau 1 1 daddhy atra | maddhv apanaya j 
atia yafl dhakarasya dvitvavidhau || nayanam | lavanam®® | vaiyakaranah j 
sauvadvab®’’ 1 ySni santi | flam sanP®* | abhi^anti [ vii?anti | apayanti | 
viyanti®® | atraidaiiyalluco ’ySdav Ssadvidhau || sklugvidhiprdtisedhah®** 1 
kim II sukusmayateh suktih | ki^thaip tak^ayatiti kS^thatak | atra sarpyoga- 
syadiskor luk [1. 2. 91,] iti lua sthanivadbhavat padasya [J. 2. 92] iti luk || 
ka^thataid ity apsrante || asklug iti piSyiko 'yam nii^dhas tena madhu4cutam 
acaksSjpo [ 25 } madhug ity atra ipilopasyastbanivattvSt dnas tat so ’acab 
[2. 1. 146] iti paiyu^s aa a m a r thyiat ^akSrasySpi satpyogasyadiskor lug iti 
luk I ^adika i^ a«aid ity ato ]a4 || 

« 1. 2. 37. 
®® J. 2. 213, 
«S 3. 2, 3p. 
«* J. 1. 73. 

®« 1. 2 96, 134, 120. 
®* J. 2. 120. 

«« J. 1. 71. 

»» J. 1. 77. 

«* 2. 2 217, 166. 

®* 4. 2. 134. 

®r 2. 3. 87 , 1. 1. 71. 
*» L 2. 91, 


1. TEIL 

iludgenat || 52 || 

parasya pratyayasya §Iuci samjatayiaip ^lugbhutapararunuttakaiji pfirva- 
karyam ik-enad ity etfivad eva bhavati ] veveddhi ] &i6a\dti | jarigrhiti | 
enat paSya | enacchritakaljBi || sthaifivfinalaSraye [1. 1. 50] iti siddhe niya- 
piarf haip vacanam 1 teraanyani na bhavairti | tat ) gargSh 1| 

tid ddil). 11 53 II 

tid yasya vidhiyate sa tasySdih pratharnSvayavo bhavati j dnas tat f-o 
scab [2 1 146] | gu<?alitt aaye || valader it“* | vadita | vaditum || 

Hd antah 1 1 54 1 1 

kid yasya vidhiyate sa tasyanto ’vaan&vayavo bhavati || nah 6i jak 
[J 1 147] bhavafic churah || hrasvasya tak®* | agnicit || yogavibhaga 
uttaiarthah 1| 

visesanam || 55 || 

viSesanjam vi^yasya samudayasyanto 'ntavayavo bhavati || napo ’co 
hrasvati [2 2 1] kflalaparji | giamaxu kulam || yvf ity ac®* [ jayab | 
stavaih | tarah || 

prdk pancaml 1 1 56 1 1 

pahcamyantarthavisesainam asamanladhikarainam vi40$yltt pi3k puivarn 
bhavati || padad vakyasya vasnasau yugvibhaktelj [2 2 191] dharmo vo 
vardhatam | dhanno no vardhatam || iha na bhavati || yiDstnakaip dharnw 
vardhatam 1| 

na saptamy aghyadku l| 57 || 

iha saptaminirdi^tam viSeeanam vile^yat purvam na bhavati ghyaltedi- 
vidher®’ anyatra |1 eco ’ey ayavSySv [2. 1 71] munaye | sadhavoghah J 
anena parasyayadi na bhavati || aghjSdiigv iti kim |1 smarasi vatsySmah 
kahihgesu®® | ma bhfit®^ || 

tasyadii^ ]| 58 || 

tasya saptamyantasya vi4e?anain tasyadir avayavo veditavyah |1 jaiay§ 
nas indrasySci [2. 2. 37] jarasah | jarasam || iha na bhavati |[ jarasu tapyate 
’nena || saptamity asya stiStvam na paiSmrSyate || 

[26} pifatyaycmyak^yat prakriyadeh || 59 H 

§yad iti gurupottamasyanaise ’patye ’niSah syaift [2. 3. 2] ity arabhya a 
yunas tit [2. 3. 76] iti titas takSrapa pratyaharah | pratyayo nyah upasar- 
janarp ca igyat viie§anani prakrty&deh samudayasyeti veditavyam nonSdhi- 
kasya || pratyayah || matrbho^iijah 1 kh&rapayalnah || sun padam [2. 1. 62] 
iti padasapijfS iinasya na bhavati | tena ahhinne [2. 2. 56] iti inajj || iSjhah 

2. 1. 33 4. 2, 134. ®® 4. 1 42. 4. 4. 14. 

«» 4, 4. I S. «e 4, 3. 209, 4. 4. 137, 


puru§ah tiajapuntsah | eastliy ayatnat 12, 1. 43] iti samasah | adhikasya 
samud&yasya na bhavati | vrddhasya iSjfiaJj puru^ah |1 putram icchati 
putratemyati | supaih kartuh kamyah [4 1 17] [ adhikan na bhavati ] 
mabSntatn putram icchati [ j nyaksyat 1 1 atikarlsagandhyabandhuh | atikau- 
mudagandhyabandhuh H bandhau ?yasyei [2. 2. 115] nabhavati“8 | anupasar- 
jinas tu §yad adhikasySpi bhavati || paramakarisagandhibandhuh | parama 
kaumudagandhibandhuh [1 

krt satik^rakasyapt |1 60 H 

krtpratyayah satisamjfiikasy&pi salsarakasyapi satikaiakasyapi piakrt- 
yadeh samudayasya viie§anam bhavati \ apisabdat kevalasyapi | udakevi^Ir- 
nam | avataptenakulasthitam [ devadattanakhanirbhinnam | bhasmanihutain ] 
samkfltinam | vyavakro® || 

tina vakyani || 61 || 

iha safcsat partimparyeina va tinantasya vi^esanam prayujyamanam 
aprayujyamanarfi va tena bhantena prayujyamianenapiayujyamanena va 
saha vakyasamjflam bhavati ! dharmo vo rak^atu®” | dharmo no rak§atu | 
sadhu VO raksatu | sadhu no lalcjatu | SaKnam ta odanaip dadati ] halinam 
ma odanam dadab | katam kuru3 gramam ca gaccha | yavan Iunihi3 sak- 
tQinS ca piba 1 devadattena vo datavyam | devadattena no dStavyam 1 1 s5kan- 
fcsatve ’pi tiinantabhede vakyabhedartham vacanam | odanaip paca tava 
bhavijyab [ mama bhavi§yab | paca tava bhavi§yab | mama bhavi^yati 
odanam tava bhavii?yab 1 mama bhavi^yab ] aithat prakarajiad vavagatav 
aprayogah || 

sun padam 1 1 62 1 1 

sun^o iti prathamaikavacariad arabhya a mahino” tiakar«(ija piatyah- 
araih | sunantaqi Sabdarupam padasaipjfiam bhavati | dharmah 1 karma ] 
pacatah | apacan ] briimahe j vafc l| 

£273 kye II 63 || 

nak&imtam Sabdanupam kye pratyaye paratalj padasaipjfiaiP bhavati 1 
kya ib kyac-pkyan-kyaij-kyanam^® ■ vi46$akaran anubandhan utsrjya aamM- 
yena grahaioam 1 iSjryati''* | i&jayate | carmayati ] usnayate’* 1 1 kya iti 
kim f| sSmanyaih |1 suin ity eva [ manya H 

stdvaly adhatoh. i| 64 || 

sib valadau ca pratyaye pare purvam padasarajfiam bhavab ) adhatoJj | 
na cet sa poratyayo dhator vihito bhavati |{ siti |j bhavadiyaih | urijayuh ij 
vail 11 payobhySm j iiayahsu | iSjata ] va^am 1| adMtor iti kim |1 yajvS | 
vacmi || 

«8 1, 3 2. 88 1, 2. 191 TO 2. 3, 97, 100, 1-27, 135, 152, 165, 171. 

4. 2. 12 4. 1. 18, 22, 27, 35. ts 4, i, 36. 


1 . IfilL 

m vrttyontah || $5 || 

padarthabhidMnaip vrttih | tadvad §abdasamudlayaib aamasadih | tas- 
yantajb ^abdab padasamjno na bhavati j paraitiagirau | patamadivau | 
gyalihau I goduhau | bahudapdinau || antagrahanam kmi || iSjavSk || 

storn matvmthe || 66 || 

sakaiSntam takatantatp ca 4 abdarupatp matvarthlye pratyaye pare 
padasamjnam na bhavati ( yaiSasvl | yagasySn [ vidusman | tnamtvtn |1 

mammabho’n^ro vati || 67 || 

manua-nabhas-ajngiras ity etani vati pratya 3 re pare padasaipjfSani na 
bhavanti | manur iva manu^vat | nabhasvat | angirasvat [j 

vvrSme ’^diotancady m vammasikah ||i 6S || 

vuratir vwamah 1 vii§ine vartamSnasySino gidanSncadivarjitasya tadS' 
aanno''* 'nunasika adeio va bhavati \ gaina 1 sama j khatvS 1 khastva j 
^s,ri}^ i I dadhi | madhti 1 niadhu H yuSma iti kim 1 | dadhi karoti || agida- 
nancaditi kim jf mun! | sadhii^’t | kmi u || anafigraha|naip kim |1 pSLali* 
putifid 2 1 pataliputrad a |1 

car jaialf. || 69 || 

yitamie vartamanaaya jaiSajh sthgne tadgsannaiS'''^ canadeSo vS. bhavati ( 
tristup I tristubT® I vak j vag 1 eat 1 sad 1 tat | tad H viraina iti kim |i vag 
atra’« ||> 

C28l«all 70 11 

ita iiirdhy ayp yad vaksyate tad vitame vartamSnasya na lAiavatity adhi- 
kftajp veditavyami a padapansamapteh 1 te Shuh 1 bhavan lutfiti H 

eco ’ey ayavayav || 71 H 

ecah sthane aci pare ay-avnay-^v liy ete kramedade^ bhavanti | naya> 
nam 1 lavanam | iSyau 1 i&vau |j 

yan evad ikab 1| 72 H' 

ecalh sftbane avarnat paro ya ig idistas^^ tasya sthSne aci pare yafia* 
deSa eva bhavati i&ayaSo. \ agamaSh agnibhutSSy^'' atiagaocha | agamaSh 
pataSv^’ atiSgaccha |1 evakSro dirghah&dhanarthalh^® | anyatha hi pQrve 
'pavadS. anantaiSn vidhin bSdhante nottarSn itt hrasvasyaiva^* bddha syat |j 

asve 11 73 II 

ikah sthSne asve aci paratas tadasanno yaB&deSo bhavati | dirgha* 
pavadaih^® 1 dadhy®“ a^Sna | inadhv®“ apanaya 1 pitrarthah ] lakfbh 

** 1 1. 7. 78 i. 2. 22 , 1. 4. 76 j, 2. 75. 77 2. 3. 30. 

78 1. 1, 77. 7» j. 1. 74. 60 1, 2. 7, 



dadhy Itakaya diyatfim ity aha || asva iti kim || dadhidam | madhustrali || 
iko yafibhir vyavadt^am ity eke 1 teeam ikah paficamisi 1| dadhiy atra [ 
madhuv atra | tinyan J bhiiviadayah || 

hrasvo vapade || 74 || 

iira^i stKane asve aci pare hrasvadeSo bhavati va yanapavladah®® | na 
cel tav igacav ekatra pade®* bhavatah | nadi®* e$a | nady®^ e^a | dadlu®* 
atra | dadhy®* atra | madhu®® atra | madhv atra | ati eti | aty eti j anu 
eti I anv eti || hrasvasyiapi hrasvab | parjanyaval lakeaiijiapraviittih || apada 
ib kim || nadyau®®) | vadhvau | nadyudakara | vadhvananam || 

jty aka^i || 75 || 

akah sthSne rti fklare Jkare®® cSci pare hrasvadeio via bhavati j maha- 
isih I mahaisli®® | dhulirtaji | dhiulyrtaib®® | vadhumam | vadhmam®* | 
kartjriigyah [ kartRyah®* || Iti |1 tava Ik&raih | tavalkSrah®® || takarah kim || 
kanyg fk^ah j kanyarkarah®® || 

(29'} fi col^ sacah |1 76 || 

uh rvainjasya Jvanjasya c&kah sthane Tti rkgre Ikare caci pare parenaca 
sahitasya fn ity acsamudayo 'jvyafijanasamudayp vanjBntaraim va RatspiRta- 
karanam adeSo bhavati vS | pitrreabhalb | pitmsabhaji 1 pitpsbhali |! Jti j| 
pitrlk&rah 1 pitrlkarah | pitrkaralh || Ivatnasya H pitJrtaka ity aha | pdtlrtaka 
ity aha ] pitrtaka ity i^a || Iti Ivaijjasya H pitjltaka ity aha | pitlltaka ity 
aha I pitrtaka ity &ha 1 rvai|ijalvaittiayor dkatvapratijSanad Jity adefe jphida- 
didarSaeSl latvam || cakaro vety®® asySnukaisa|i:igrtham | tenottaratra nanu> 
vartate || s&ca ity aj^fcarah padante 'ty enah [I. 1. 94] iti yavat || 

dirgm 11 77 H 

akah sthSne aci pare parepaca sahitasya tadasanno mtyarp dirgha adeso 
bhavati 1 daodSgram 1 s&gata 1 mutdhdraih 1 nadiyam | madhudakam j 
vadhudaram 1 pitRabhah H Ivaiipasyanukaraip&d anyatra dirgho nastiti ]var- 
tiasya ^mra eva dirgbah ll 

Sasy ak || 78 H 

akah 4asy aci pareipSicia sahitasya yathasaipkhyam ag ^rgho bhavati | 
tah®» 1 daiajji 1 buddih 1 dhenuh 1 nadilj 1 vadhGh matih paiya H 

nantak purm^ II 79 II 

akah p u xpli i fl g asan ibaiidhipi 4asy aci pare parephcS. sahitasya yathasaip* 
khyaitp djirgho nakSiSnta Bdaio bhavati | jinSn»« ( munin 1 sadhiun 1 pitfn H 

81 J. 1. 4a, 8* 2. 1. 73. 8s J. 1, 62 f. 8* 1 . 2 120. 

88 CX. Cinffim. zu den PtatyShfirasutras. se 82. 

8* 1 . 1. 77. 88 J. 1. 74. 88 J 2. 216. •• 1 . 2. 92, 95, 49. 

1. TEIL 23 

dhralucy || 80 || 

i^hakSrasya rephasya ca luci®^ purvasySiio dllrgho bhavati | li'tjhani ] 
gu(Jham 1 puna ratrau | agra rathena®* \ patu raja |1 ijhraluks^acaryad iha 
na bhavati || esa karoti | sa dadati*® H 

sahtvaho ’symh || 81 || 

sahivahor avaih^sya idhralucy oMro bhavati | sodha | vodhS | sodhum | 
vcxihum II asyety adhikaras tathanmedhy eny en [i 1 93] iti yavat || 

' II I) 

asya aaca iti vartate | avamasya sthane iki pare paresnaca sahitasya 
yathasamkhyam en-ar ity eta adesa bhavanti | devendrah ] tifileyam | gan- 
dhodakam | tnalodha | paraman^ih mahatigih | tavalkarah | salkaia®^ ityadi jj 

ejucy etc II 83 H 

avamasya sthine eci ujade^ ca pare sacas tadasanna ai] ade§o bhavab | 
tavai^ I khatvaii^a tavaindd tavaudanah | tavaupagavah 1 1 <ucx 1 1 dhautah | 
dhautavSn || 

prasyodho4hyiihai$aiiye || 84 || 

praSabdasya yad avamam tasya sthSne udha-udhi-uha-esa-eisya ity eteisu 
mca asanna aijadeso bhavati | praudhah | prauidhih | prauhah | prai$ah j 
praisyah ll' 

svtdrasv(dryaksauhii}yam ]j 85 || 

svaira-svairm-afciauhiipi ity etesv avaipasya ^ca ejSdeio bhavati | svasrya 
Irah I svairaih | svayam Intum SQam asyeti | svEiiii | ak#riiiii uho ’syain 
astSti I ak^auhiiQi sen& || 

omant para}}, || 86 || 

avamasya sthane omiabde Snadede ca sacah paro ’jadeSo bhavab | 
tavoipkEfah | kom ity avocat j| Siu || a pSyat | ai4yat®® | adya an§- 
yat I adyanSyiat | khatvaii§yat 1 a ihi | ehi*® | upa‘ ehi | upehi 1 paiehi j a 
udha I ojdha®® | adyodha | khatvodlfi || 

eve ’niyoge || 87 || 

avarpasya evailabde sScah paro ’jiade^ bhavati | na cet sa eva^bda 
niyogavieaye ’vadhSrane vartate | niyogah | idam eva kartavyam iti | ihcva 
dplyate ] adyeva tisthati ] tattvSnvakhySnam etat 1| aniyoga iti kim || 
atraiva®* tvaip ti§theti niyujyate |l 

I. 1. 131. ®® J. 2. 72 ; 1. 131. ®® J 1. 158, 46. 

®* Cf Cintam. zu I. 1. 76. ®® I. 1. 82. *® I. 1. S3, 


vausthautau samase 1 1 88 1 1 

avamasya ostha&ibde otu&ibde ca pare! sacaj? paro ’jadelso bhavati va j 
tau cen nmuttanimittiniav ekatra samase bhavataJj | bimbosthi ] bimbau 
1' sthulotuh I sthulautub 1 1 saroasa iti kim 1 1 rajaputrau§tham paSya 1 
devadattautuvijfmbhitam pa'gya l| 

£31} or tftiyaya rte \ \ 89 1| 

trtiyantasambandhino 'vamasjra sthane rta§abde pare saca &tade§o bha- 
vati samase | sukbartah | dulikhlartah || rty ar upasargasya [i. 1. 91] iti 
punar argrahanjad hrasvo®^ na badhyate | dubkhajtalb 1 aukharlah®^ 1 1 trti- 
yfiyS iti kim 1 1 paramartaih®® | j samasa iti kim 1 1 duljkhenartah»* 1 1 

pradasartfovasanakambalavatsatarasyame || 9D || 

pra-daisa-iTna-vasana-kambala-vatsatara ity ete§5m avaiinasya fiOSsabde 
pare sSca Sx bhavati samase | pragatam mam prai|nam | da^^ maip 
daiamam ( da^ mam yasylam aa daSama nadi daSamo janapadaib | mapa- 
nayanaya mam rnSiinam | masya mam m5mam | vasanam eva j.mam vasanar- 
ipam 1 1 evam kambalamam ] vatsataramaw | j hrasvo na b&dhyata iti praf- 
nam®’’ ity&di bhavati || 

rty or upasargasya || 91 || 

upasargasya yad avamaip tasya sthane pklaiSdau cMtau pare sSca ar 
bhavati | sarvapavadah | prardhnoti | piarcchati || punar argrahaipaip hras- 
vabSdhanSxtham*r || 

supi va II 92 |[ 

upasargasya yad avamam tasya sthlane supi subantS-vayave rkaiadau 
dhatau pare saca Sr bhavati va 1 pak§e yathg pifiptam ) upai^abhlyati | 
upaieabhiyaU®® ] upar§abhiyati»r || upalkariyati»» | upalkariyati | upajkari- 

tathaninedhy «Ay an 1 1 913 1 1 

upasargasya yad avamam tasya sthBne iln gatau edhi vrddhau ity eta 
bhyam anyasminn enadau dh&tau pare slaca efiadefe bhavati | tqtha subdliS- 
tau tu via I prdayati j pire$ayati | upcddiati | prokhati || subdhStau |( upe- 
Iakl3^ti I upailaklyati^®® | upodanlyati | upaudanlyati || aniinedhy iti kim || 
upaiti“® I praidhate [| 

padante 'ty enah |1 94 || 

padinte ya eft tasya akBre pare saca eift bhavati | te ’tra | paita ’tra j] 
padBnta iti kun 1 1 nayanam^ | lavanam 1 1 takarah kim 1 1 paitav Sssva’^ j | 

« 2.1.75. 
““ 2. 1. 83. 

»« 2. J. 82. 
1 2. 1. 71. 

»» Cf. obeti S. 14, 24 f. 


1. TEn, 

{32} gor od II 95 II 

aSca® iti nivirttam | gty§abdas 3 ra yah padanta eh tasya akSre pare okaxo 
va bhavati | or odvacanam prakrtibhavSrtham | goagram | gavagram® \ go- 
’gram ^ 1 1 he citrago ’gram ity atra citragoSabdasya® lHk§aimkatvSn. tia bha- 
vati I lak^anapratipadoktayoh pratipadoktasyaiva grahainam | na tu ]gk$atoi- 
kasya 1| 

avo ’ey anak$e || 96 || 

go4abdasya padante vartamanasya e&ah ad pare ava ity adeSo va bha- 
vati I na cet so ’j akga^abdasthah I gavagram | go’gram* | goagram* | gaves- 
varah | gavffSvarali’’ || padanta iti kim || gavi’’ || anak!§a iti kim || goakgam* ( 
go’bsam* || 

indre || 97 || 

godabdasya padante vartairgnasya enah mdra^abdasthe ’d pare ava ity 
adeSo nityam bhavati | gavendrah || 

vStayane 'kse || 98 || 

gosabdasya padante vartarnSnasya enah ak^a^abdasthe ’ci pare vfitayane 
vacye avety 5de^ bhavati | gavaksajj | vatayanam ity arthab |[ anyatra || 
go’ksam®' | goaksam® || 

na plutasySnitau || 99 || 

plutasySnitav aci pare yat pifiiMioti tan na bhavati | devadattaS® atra 
nv asi | jinadattaS idam lanaya || anitav iti kim | sudlokSS id | su^oketi® || 

gitaf^ II 100 II 

ganubandhakasyad pare tannuiuttam yat piapnoti tan na bhavati | 
muni®<* etau j sadhu etau ) pacete atra ] pacavahe*® avam || 

coder aco ’nanah |{ 101 || 

eSdir asattvavac* Sjivarjito yo ’c tasy&d pare tannimittaiip na bhavati | 
a apehi | i indratp paiSya | u uttistha 1 S evaiji ml manyase | a cvaip kila 
tat 1 1 anana iti kim 1 1 a uisnam | oi;nam° | lead ispiam | {33} a ihi | ehi | a 
udakiantat odakantat priyam anuvrajet | a aryebhyalh | aryebhyo 'sya ya4o 
gatam || 

feadarthe kriyayoge marySdabhividhau ca yah | 
etam Btaap hitam vidyad vakyasmaranjayoir aiflt 1 1 

otttk II 102 1 1 

cader okarantasySd pare yat prapnoti tan na bhavati | aho idam | 
utaho evam | atho asmai | no indnyam || 

® 1. 1. 76. » 1. 1. 96, * 1. 1. 94. • 2. 1- 123. 

« I 1 95. ''1.1. 71. ® 2. 3. 27, • 1. 1. 82. 

1® J. 2.. 22. 11 1. 4. 93. 


DIE grammatik SakaitAyana’s 
sou vetau \ \ 103 1 1 

sunmutto ya okaras tasya itisabde pare yat prapnoti tan na bhavati va | 
pato iti 1 patav iti“) 1| sav iti kim |1 gav ity aha H 

comfi. II 104 II 

ufi ity etasya itau pare Q ity ayam dlrghiinunasika adeso bhavati va J 
caSabdIad yad anyat prapnoti tac ca va bhavati | evaip trairupyaip bhavati | 
g iti I u iti« I V itii*. 1 1 

mayo ’ci vo 'son || 105 jj 

may iti pratyahSrah | maya uttarasya ufiah sthSne aci pare vakaro 
bhavati va | sa casan abhutavat | krufm^® v aste | krufiii u &8tei® | kim 

V ii-yigini I kim u uipam’^®’ | tad v asya matam I tad u asya matam | kim 

V Iti ! kim u iti“ | kim u iti“ | knji v iti^’’ || asve \1. 1. 73] iti yafl || 
asattvad“ dvitvam^*' anu8varanunlsik&bh&va§ ca || 

hdo ’nunasikd 'mmasikai, svah 1 1 106 1 1 

padinte vartamanasya halah sthane anunSsike pare sthdninah svo 
'nunELsika !ade§o bhavati vd | v&h madhuiS | vlag^o madhuiS | gam nayth | 
sad nay^ | tan nayanam | tad®® nayanam ] kakummaimdalam | kakub- 

maDdaJam®* | halmatram ) halmatram | tvafi g®®) iti | tvag g id |1' asan“ 
ity eva | tvang iti { hrasvan idamab '[!• 1- 123] iti dvitvaip na bhavati || 

pratyaye |1 107 H 

padante vartamSnasya halah stl^e anunasikadau pratyaye pare sthani- 
nah svo ’nunasika 9de§o bhavati mtyam | vanmayam | $aininam || padanta 
iti kim 11 yajfiah | svapnah II 

C34J myok'W 108 || 

padSnte vartairiane rephe sakare ca ikaranubandhe pare yah purvas 
tasya sthane svo ’nungsika ade^ bhavati | n|h pihi®<>^ l^skan®®} | bbav g 
diadayati*®,' || 

mnam }ayy apadante || 1109 || 

mak&ranajkai^m apadante vartamlanaaim jayi pare nimittasvo 
’nunadko bhavati*|| masya || ganta | gantum || nas 3 ra || Sankita | lahkitum || 
bahuvacanarp nasya ipatvabadbanartham || visrambhah | abhi^ti || apa- 
danta iti kim || bhavSn paiamah |j 

« 1. 1 71. “11 JOl. 

« 1. 1. 12^ M 1 1. 104. 

M 1. 1. 105. w Cf. J. 1. 123 

« 1. 1. 14a » 1. 1. 149. 

1. 1 73. 

IT 1. 1. 73, 111. 
*® J. 2. re. 
a» I. 1. 150. 


1. 1£IL 

saly onusvarafi 1| 110 || 

makaranakarainam apadante vartamananatp sthane feili pare ’nusvaro 
bhavati || masya H pumsi | gaipsyate \\ nasya \\ daipgah, |' ya'^si 1| 

mafjimo halt tau || 111 || 

padanta iti vartate matpgrahanat | mamagamasya padante vartamanasya 
ca makarasya sth&ne hali pare tau ninuttasvo ’nunasiko ’nusvaraS ca patya- 
yejja bhavatali [ cabkramyate \ cajnkraraayate“ 1 abhraUiho vayub jabhraiji- 
lihoi v&yuh I masya || tvan karoai | tvaip karoji | sayyanta j sarpyanta || 
padanta iti kim || gamyate [ ramyate || 

hi Ivyatmt || 112 || 

padante vartam&nasya makarasya stbSne lavayamana ity etadvanoapare 
hakSre pare te^Sup svo ’nunasiko ’nusvaraS ca parySyeifja bhavatafe | kil 
hladayati | kim hladayati 1 kiv hvayate j kup hvayate | kiy byah | kup 
hyab I kim hmalayati | kup lunalayati | kin hnu$es [ kup hnu^ || 

samtat || 113 || 

sam ity etasya rfijatau kvibante pare anusvarabhavo mpatyate j 
samratsu paiicamaili i^ntdi || 

kkay khayalf. Sari va || 114 || 

padanta iti nivrttam | khayab ^axi pare khayiadeiio bhavati vd | taccb 
^te I tac 4ete j vathsah | vatsah | aphsaiah | apsaidh || 

£35 J Sara ’nu dve |! 115 H 

4aral)i parasya khayah sth&ne anu yad anyat piapnoti tasmm krte pa4c&d 
dve r,upe bhavato va ] kad cchBdayati*® 1 kaiS chadayati ] tvaap kkhanasi-' | 
tvajp khanasi | stthaii | sth&li || anv ity uktatvad asan®®^ iti nivrttEim iti 
dvitve cartv&di®® bhavati {| punaih khayaiji parasya larati sthane dve rupe 
bhavato v& | tac d^te | tac 4ete | vatssah | vatsah 1 1 

yaiio mayah |j 116 || 

yafialb parasya mayah sthane dve nupe bhavato va [ vrkjav kkaroti 1 
vrksav karoti 1 valmmSkaJj 1 valirfikah |1 anv*'' ity eva || prowunava j 
urjjijayisati || punar mayab parasya yanah stHSne dve mpe bhavato va j| 
dadhyy atra \ dadhy atra 1 madhw atra ] madhv atra | trapw atra 1 trapv 
atra || 

aco hro ’hracaJj^ || 117 || 

acah paro yo hak&ro repha§ ca tabhyatp parasya ahracah hakii&d repbad 
acad cSnyasya vamasya sthane dve rape bhavato vS | brahmma | brahma | 

** 4. 1. 88. 

** 1 . 1 . 11 , 

J. 1. 135. 

2® 1. 1. 105, 


sarvva}; | sarvah ] | dirghah 1 1 ahraca iti kun ] | barhah | dahrah j 

ahatn || 

adirghat || 118 |1 

adlrghad acah paiasyShracah sthane dve tiupe bhavato vh | daddhy^** 
atra | dadhy atra | patthy®®., adanam | pathy adanam | tvakk | tvak 1 
tvagg 1 tvag j goSttratah j goStrataib H anv*^ ity adhjkSrlat kutvadau*'*' 
kite dvitvam || adir^d ekahaJIty anuktva na samyogla .[J 1. 119,] tv aci 
[2 1. 121,] iti yogadvaygrambMd vimme ’py ayam ade§a!h jj ahraca iti kim | 
sahyam | varyah j titauh H adarghad iti kim H sutram j patram | vak j) 

na samyoge || 119 1| 

halo ’nantarajh samyogah | satpyoge pare ahracah sthSne dve rape tia 
bhavatajb [ indrah | krtsoani [j 

putrasyadiputaradwy dkro§e j| 120 || 

putxa§abdasya adin^abde pare puti&dinlabde ca pare akroSavi^ye dve 
rape na bhavatah | putiidui tvam asi pape | putraputiadmi bhava 1 1 anya^ 
tra puttigdira || 

£36} aci II 121 \\ 

adirgjhat parasya ahracalji stbSne aci pare dve tjlpe na bhavata^ | dadhi | 
madhu [[ 

Sarab ll 122 || 

Saro ’d pare dve ifipe na bhavatah | dar^anam | van^ah | tarsam || 
hrasvan Aamak padante || 123 |l 

hrasv&t parasya padante vartaingnasya nanvah sthane ad pare dve rOpe 
bhavatah | kiuM gste | soganin iha | kn^ann iha || asiddhatp bahirai'igana 
antarahge iti |iio na bbavati || 

dtrgkac cko va || 124 || 

padante vartamgnlad dirghat parasya chak&rasya dve rape bhavatd vS ] 
kanya ochatram®® | kanyg chatram || 

pMat 11 125 II 

padante, vartamSnad (highasthSnihat plutat parasya chakarasjra dve rape 
bhavato v<i | agaccha bho mdrabhuteS cdiatram®® anaya | lagaodia bho 
indiaUiuteS chatram inaya || dirghad iti kim || agaccha 1::^ devadattSS 
chatram anaya || 

as J. I. 136. 

as 2. 2. 88. 

s» 2. 1. 135. 

1. lEIL 


ajanniMah || 126 || 

acaJj &no mafial cavyayad uttarasya chakarasya dve nupe nityaip 
bhavatah | icchati®® | mlecchati’o ] aochmatti | ma cchidat || anv*^ itj‘ 
eva II praSnah | pra)?ta II 

dajbkaja ’to lug itau || 127 || 

dacbhBjah*® anekaco ’vyaktlanukaramsya yah atSabdas tasya itiSabde 
pare lug lopo bhavati | chamat ili | diam iti | patat iti | pat iti | asiddhatp 
bahiraxigam antarange iti luci jaStvatps^ na prlLpnoti || 

cakad ib taditapi krtam ] 

iti dafcarantajin dra?tavyam || idajbha] iti kirn || chat iti |i chad iti ) jagat 
ih I jagad iti || 

p7j na dvyukteh jj 128 || 

dve ukU yasya tasya dajbhajo®* yo ’tSabdas tasya itau pare lug na 
Kuvati I patatpatad iti | gha'tadghalad iti | vipsayam dvyuktit®* | patatpatad 
ill samudayanukaraiiami || 

tah II 129 II 

dvyukter jd&jbhajcH^ ’to yas takSras tasya itau pare lug bhavati \ patat- 
pateti'** karoti | ghatadghateti karoti || 

d&cy adau || 130 || 

dvyukter adau piirvasy6m uktau ato yas takaras tasya <3B,ci parato tiit- 
yaip lug bhavati | paiapaijgkaroti ] dhaniadham^i'oti 1 1 

4hro dhri H 131 || 

dhakarasya rephasya ca yatUasamkbyam dhakare r^he ca pare lug 
bhavati | lidhams’' | gfldham j agni rathena j puna rauti |1 

hdo yond yamo va \ | 132 1 1 

halah parasya yamo yathasaipkhyam yami pare lug bhavati vS. | adityah | 
Milyyajh'^ 1 1 ke^aip cid yaiifim iti palhah | tes^ vacanabhedad yathasatp- 
khyaip nastily udShaiamam idam || babhyate | babhryate ||i 

j(ffi jarah sve va\\ 133 || 

halaJh parasya jarah. sve jari pare lug bhavati | bhintah I bhinttah | 
bhintam | bhinttam || sva iti kim || taptvS || 

itdal} sthdstambhah || 134 || 

udah parasya sthSastambhor dhatvor avayavasya jaro jari pare nityaip 
lug bhavati | utthat& | utthatum | uttanibhita | uttarabhituih || uttthata | 

=»■ I. 1. 115. ^2 5 T 4 54. « 1. 1. 135 a* 1. 1. 136. 

®* 2- 3. 8. 88 1 . 82. sr I, 1, 80. sg j. 1 . 116 . 


uttambhita iti trisamyogab adirghat [2. 1 118] iti dvitvena bhavati |1 
skander utkandalco roga iti pr^odailadi^u drastavyab || 

car II 135 H 

jarah sthane jari pare caradeso bhavati | vedacchatram | gudalit tarati | 
natsyate | lapsyate 1| ja^i jaiS [2. 1. 186,] vacanat khari cartvam || 

[38} yast yoJ II 136 1 1 

jarah sthlane jagi pare ja^ Wiavati [ caro ’pavadah | labdha | labdhum | 
dogdha |. boddha | §adWiyah | vidyud bhadra || 

icau icw stvoh || 137 || 

sahlraaya lakSre cavarge copaili$tasya sthane 4akara Sdelo bhavati | 
lalha tavargasya cavargah | apta'g Sobhate | tapa4 carati | yai§a§ chatram | 
^yotati I bhtijati*® || tavargasya j| tac §ete | bhav^ Setd I lac carati | tac 
chadayati | taj jayati | taj jhg^yati | bhavaii jakarena | rajna | yajtlah || 
satnavacane yathasamkhyam | 4ailffyam adaryasya | na ^t [2 1. 139,] toh 
padanta iti ni^dhSt pare tdh [2. 1. 141] iti nu^edh&t pQrve ca icutva^- 
tntve‘® 1 1 

^lau stu II 138 II 

sak^asya aakSre favarge copa!§li$tasya aakaro bhavati | tath& tavargasya 
ravargah | kag $a|Q|cl£ | ka$ tikate | ka$ thakareina 1 1 tavargasya 1 1 pe$lta | tait 
tikate I tat tbakaraija 1 1 

na Sat |[ 139 || 

^ak&i&t parasya 4cutvatp na bhavati | a^ti | kliSnati || 

toh padanta ’natjmagafinavateh || 140 || 

padgnte vaitanamt itavarggd uttarasya stutvam na bhavati | naipna- 
garlnavati^bdan varjayitvg | madhulit sadati | §adnayam*^ | ^an nayah || 
anaxpnagaiuiavater iti kim | §an|niam‘* | ganmagarl | i^aonavatib [ | padanta iti 
kim 1 1 Stite 1 1 padgnta ity adhikara a pgdaparisamapteih 1 1 

toh II 141 II 

tavargasya padante vartamgnasya §akgre pare ^tutvarp na bhavati | 
agnicit aadikah | mahBn eandaih |l 

It lah II 142 II 

padante vartaman^ya tavargasya sthane lakare pare lakaradeSo bha- 
vati I tal lun&ti I bbaval likhati |[ 

89 2. 1. 136. 

*0 i. 1. 138, 

" 2. 2, 152, 

« 2. 2. 152, 34. 

1. tElL 


£39} jaSo ho jhas va |] 143 || 

padiante vartansmlij jaSah parasya hakarasya stMne yathasaipkhyam 
jhas va bhavati | ajjhalau | aj-halau f tristubbhutam | tri§!tub-hutam ] vag 
ghasati 1 vag hasati \ san? 4halani \ sad halani | taddhitam | tad-hitaxn || 

Sai cho 'mi || 144 || 

padiante vartam^naj jaia uttarasya ^akgrasya ami pare chakaro bhavati 
va I tac chobhate | tac dobhate | tristup chruyate [ tristup Sruyate 1 1 

nw> gagdak Sari || 145 || 

padiante vartamanayor nahSrainakarayoh 4ari pare yathasamkhyam gik- 
dak ity et^v agamau va bhavatab | kmnk dete | krun dete | sugaioit dete < 
sugap ^te 1 1 

dnas ta( so 'Scah || 146 1| 

padiante vartamSned d^kaian nakarac ca parasya sakarasya tadagamo 
va bhavati | adcah | dcasamyogasyavayavad cet sak&ro na bhavati J itiadhu- 
litt sidati 1 bhavSnt sarpsadi |l adca iti kim || sat dcyotanti j bhav&n. dcyotati ‘il 

nah Si jak 1 1 147 1 1 

nakhiasya padante vartamSnasya dafcare pare jagpgamo*® vS bhavati 
adcah I dcasaipyogasya tu dakare na bhavati | bhavaSc deta" ( bhavafi^* dete | 
k]:sajic dete | ktsan ^ I bhavMk: durah | bhavtn durah 1 1 

nfnah pi rlirak || 148 H 

nfn ity etasya nakorasya paddntasthasya pak^e pare ri^” iti ikaianu- 
bandha adedo rak*’ vagamah pary&yepa bhavato vS \ n^h'*' p5bi j lOPPh,*' 
pahi I nfn pShi || 

katpskan sisak || 149! || 

kSn ity etasya dasantasya dvirvacane krte purva^a si** iti ikaranu- 
bandha Sdeldah sak** vagamajj parySyena nipatyate \ liSskan“ I kaipskan** | 
lisyor ikaro risyoih [1 1. 108j] iP vide$apSrthah || 

£40} chavy amy apraSSnah |! 150 |i 

pradan varjitasya yo nakSras tasya padantasya ampare chavi paratah 
8ir*» Sdedah sak*® vagamah paryayepa bhavatah 1 bhavSi®* diadayati j bha- 
vatpid** diSdayati | bhaN^*® I bhavajps®* taraP || diavip kim || bha- 
van ii^aP || amiti bm || bhav3n tsanikah || apradana iti kim || pradaii 
caraP || 

« 1 . ). 54. 

« i. 1. 108 , 2 67. 

« 1 . 1. 108 

»* 1 . 1. 110, 137. 

• ** 1 . 1. 137. 

« 1 . 1. 110 ; 2. 67. 
*0 1 . 1. 54. 

«» J. 1. 110. 

« 1 . 1. 108. 

** 1 . 1 . 110 . 

« 1. 1. 408, 137. 


pmtffh khayi || 151 || 

pum ity etasya yad antyam tasya padiantasya ampare khayi paratah 
^sakau hhavatah | puskokilah*® 1 puipskokilah ] puskhStam 1 pumskhatani j 
pu&ali 1 pumScalT || 

samah ^Arst gluk ca || 152 || 

dcfsi sasatkasya ktfio ’vayave sakiare pare sam ity etasya aisakau gluk ca 
bhavanti | sasakarta*® i sasskartum 1 samsskarta | samsskartum | saskarta ] 
saskartum || gitvam uttarartham || 

vyo '$y aghobhobhagoj} \ \ 153 1 1 

avaipad agho-bho-bhago ity etebhya§ ca parasya padantasya vafcarasya 
yakSrasya cast pare glug bhavati | vrksa hasati | vrkgavrscam lacaksiino vtksav | 
deva®* yanti | agho hasati** [ bho dadfiti | bhago dehi |j padlnta iti kim |j 
gavyam 1 jayyam | bho vyoma 1 1 

acy aspast(^ ca || 154 I| 

avaiinad aghobhobhagobhya4 ca parayoh padantayor vyor aci pare glug 
aspa^tah avyakta4ruti4 casaono bhavati | pata u | paltav u** | ta u | lay u | 
agho u I aghoy u ] agho®* atra [ aghoy atra | bho®* atra 1 bhoy atra | bhago 
atra | bhagoy atra || glua gitah [I 1 100] ib sandhipratieedharthah 1| 

vanmy at || 155 || 

avaiio&t parasya padmtasya vyah unvarjite aci pare glug aspa§taS ciade 
Sau via bhavatah j pafese tadavaathyam ] pata iha [ patav iha [ patav iha | 
deva asate ] devay Ssate l.devBy Ssate H pad&ita iti kim || nayanam®^ | 
lavanam || 

£41 J nr yah 1| 156 H 

avai|Q&d aghbbhobhagobhyai§ ca parasya rer ikaiSnubandhasya sthane 
a^i pare bhavati | devay®* Ssate | deva®* hasanti | aghoy** atra j 
agho hasati | bhoy Sste | bho iSja \ bhagoy Sssva [ bhago dayase | [ rer iti 
kim |j antar dayate |[ 

ato 'ddhasy uh || 157 || 

akaiat parasya rdj sthane akSre ha§i ca pare ukara SdeSo bhavati | 
yalvapavSdab | ^ramapo** ’smi | dhanno®* jayati || takaralj kim || deva** 
atra | devS yanti | susrotSS atra nv asi ] susrota2 dehi | sarvajfia a3.ste 1 1 
ter iti kim l| antar asim || 

hcdy ammsamase luk ta}/. sat || 158 || 

takaraathanikat* sakSrSt parasya rer hah pare lug bhavati | na cet sa 
sakSio nafisamase bhavati 1 e§a«* karoti | sa dadati | paramai?a karoti ]■ 

’ “ I 

»* 2. 2. 72 , 1. 156. «» J. 1 155. se i. 155 , 

4 . 3. 86. «* J. 2 72. 59 j. 2 . 72 , 1. 153. 

•® 2. 2. 72 ; 1. 155. ** 2.‘2 72 ; 1. 94. ** 2. 2. 72 ; 1, 156, 156. 

•» 2, 2. 15. «* 2, 2. 15, 72. 

1. tElL 


paramasa dadati || haliti kmi \\ eso 'smi*® | so ’aim \\ anafisamasa iti kim |i 
ane$o gacchati | aso gacdiati || 

tadalt padapurane j ] 159 1 1 

tadadeSt sakSrad uttarasya rer aa pare anafisanfiae lug bhavati ( pada- 
pfirajnavisaye | lope cet pfidah puryate | 

sai?a daiarathi ramah sai?a rlja yudhisthirah | 
saisa kamo mahaty&gl saisa partho dhanurdharai 1| 
padapuraipa iti kim || 

sa eea bharato raja yo nyayye path! vartate || 

TO 'hno ’sy asubrupardtirirathmtare || 160 || 

ahan®* ity etasya rer asi pare repMadeSo bhavati na subrflpaifitriratban- 
taiesu 1 ahar®« eti ] ahar dadSti || asubrupaittrirathantara iti kim l| aho- 
bhyam®^ | ahobhih | dirgMihiayam®® | ahorupam | gatam aho®® rStrir agata | 
aho lathantaram || 

{42} visarjartiyasya 1| 161 |1 

visarjaniyasyisi^® pare repho bhavati | nuinir^^ asti 1 ^dhiir asmi | 
sajur''* jayati | pitur dayase || visarjaniyasyety adhikSira la p&daparisainap- 
tdb II 

vdharpatyadtsu || 162 || 

aharpatity evam^isu sabde^u visaijaniyasya stbane lepho bhavati va | 
ahaipatib | ahahpatih^® | g^rpatih | giiihpatilh | dhuipatih dhujhpatih | praceta'''* 
rajan ] praceto t^an 1 1 vS rephad atra rer utvab&dhayl pakse visatjaniyab 1 1 

sai chauy aSari H 163 || 

visarjaniyasya sthine alarpare diavi paratalj. sakaio bhavati | ka§ cha 
dayati ( kas^® tarati [ antas thudati j rnata; taklrepa || a&iiiti kim || aselb 
tsaruh || 

sari va || 164 || 

visarjaniya&ya aiiarpare Sari pane sakaro va bhavati | ka;S Sobhatc’® | 
kah Sobhate | iiiata$ samdhe | matalj Banjdhe ] antas siktah | antah siktah {j 

luk khayi pare || 165 || 

visarjaniyasya khayi pare Sari parato lug va bhavat | anta alrlialati j 
antas skhalati | antah skhalati | cafcsu spandate i cakgus spandate | caksulj 
spandate || 

«® i. 2 72 ; 1. 157, 94. b® 1. 2. 72. ar i. 2. 72 ; 1. 157, 

«® 1 2. 96, 134. «» 1. 1. 157. ro i g. 67. 

” J. 2. 72, 67. , rt j. 2. 72 « J. 2. 72 ; 1. 160 ; 2. 67. 

'•* 1. 1. 130, 80, 1. 2. 72, 67. 



kupau X ka X Pam || 166 || 

visariamyasya’kavargiyepavargiyecaSarparekhayipareXka>;pa ity etau 
jihvamuEyopadhnifimyau yathasamkhyam ade^u va bhavatab | ka X karoti | 

kah karoti [ kax khanati I kah khanati fanta- pacati )• antaii pacati I aiita;:^ 
phalati | antah phalati {| a§aipara iti kim || vasafti k^aumam | abdhib psa- 
tam 1 1 khaylti kim 1 1 antar gaochati | antar bba^ate | { 

ttrasas teh sth || 167 || 

tieamjnakasya''® tirasaJj iSabdasya saapbandhmo vi&arjaniyasya sthane 
kavarglye pavarglye cSiiarpare khayi pare sih ikatSnubandha bhavat< 

va I tiraskrtya | tirabkrtya | tiraskaroti | tirabkaroti |1 ter iti kim || tirah 
krtva II 

£433 namasputraSaJii || 168 1| 

namaspuras” ity etayos ti&amjfiakayoh sambandhinor visarjaniyasya 
kupav a&upare khayi pare mtyam sir bhavati | namaskitya | namaskaroti | 
puiaskrtya | puraskaroti || ter iti kim || namah krtv^ | purah k]rtv3 |l yoga- 
vibhago mtyarthah | j ^ 

catmmrdurbahiravispradusam || 169 1| 

catur-ms-dus-bahis-€vis-piadu& ity ete§aim visarjaityasya kupSv a§arpare 
khayi pare sir bhavati | catu^am^^ | catuspatram | niiskaroti | nispacati ) 
du$karoti | du^pacab | bahii$karoti [ bahi^pacati | avi^karoti |$pacati j 
prSduskaroti | piSduispibati | nii$kulah | du^purugaih || 

. suco vd 11 170 II 

sucpratyayantasya visarjaniyasya kupav asarpare khayi pare sir va 
bliavati I dvi^'^® karoti | dvih’^® karoti 1 tri? khanati 1 trih khanati 1 catus 
pacati I catuh®® pacati 1| 

iSMSo 'peksdyam |1 171 || 

isuspratyayantasya saipbandhino visarjaniyasya kup&v a^rpare khayi 
pare sir va bhavati | sthanimimttapade cet parasparasy&peksay^ bhavatab | 
larpis karoti^® | sarpih karoti | sarpis pibati 1 sarpib pibati | dbanue khap’ 
^yati I dhanuh khandayati | dhanu$ phalati | dhanuh phalati paramasarpi^ 
raroti | paramasarpih karoti | paramadhanu$ phalati | paramadhanub 
Dhalati II isa s^acary&d uso ’tmo ’tra grahanad iha na bhavati || cakiub 
ralaham | bhindyulh p&p^ || apeks^ygm iti kim || ti$thatu sarpib P^i^a 
vam udakam || 

ndkriymkarthe H 172 || 

isuspratyayantasya visarj’aniyasya stMne kupav a^ipare khayi kriya- 

1 1. SI. 2. 1. 29 la 1. 2. 66, 

"f® J. 2. 72, 67, 65. «» 1 2. 67. 

1. TEIL 


padavarjitasamanadhikarajjapadasthe pare stharammittayor apeJssayam ai 
na bhavati | saipih®^ kSlakam | yajuh pitakam || knySpratisedhalj kim j) 
saipis knyate®® | sarpih kriyate j [ ekiartha iti kim 1 1 sarpi? kumbhe 1 sarpiJj 
kumbhe jj 

{;44J samase ’samastasya || 173 || 

isuspratyayantasya pUrvemasamastasya sambandhino visarjaniyasya 
kupav ^arpare khayi pare sir bhavati | te cet sthaninimittapade ekasamaee 
bhavatah | sarpiakundam®® | sarpiapanam | dhanuskhandam j dhaiui?pha 
lam II saraasa rti kim || tiathatu sarpih®* piba tvam udakam || asamastasjTiti 
1dm II pararaasarpilikundam®* | mdradhanuhkhaitjidam || 

pade ’dhaihrasah || 174 || 

pQrveinSkrtasamasayoil'i adhas-4iraa ity etayor visarjaniyasya padaSabde 
pare samiase sir bhavati | adhaspadam®* | giraspadam || sam&sa iti kim || 
adhah®* padam | 4iraih padam || asamastasyeti kim || paiamagirahpadam ji 

krkandkoffisakuSaktfi'mkurnbhapatre ’to ’novyayasya || 175 || 

anavyayasya purvenasamastasya sambandhmo ’kaiSt parasya visarja 
niyasya sthane dukrh karane kamuh kantau kamsa-ku§a-kai3^-kumbha-pdtra 
ity eteiju paratah samiase sir bhavati | ayaskrt | ayaskarah | paya^mah | 
ya^asbamah | ayaskarpsah | ayasku^i' | aya^atpi | payaskumbhah | payas* 
kumbhi I payaspatram | payasp&tif || piatipadikagrahaoe hngaviSistasyapi 
grahamaip bhavati [| anavyayasyeti kim || svahkamah || samSsa iti kira 1[ 
ayah karoti 1| asamastasyeti kim || pararaaya§ahkamaih |1 4ilikBtnIti®® ijavi- 
dhau kamigrahaniad atraijyantagra^nam | tena striySm payadiamiti bha* 
vati 11 

pratyc^e 1| 176 || 

anavyayasya sambandhino visarjaniyasya kupay afeupate khayi pratya- 
yasthe pare sir bhavati | kBroya-kalpa-ka-paSih pratyayah | t3n kavayah 
prayojayanti | yaSa^myati | ^§kalpam®« | ya^askam | yuspBlai*® | suyfl*- 
pa§3 [I 

na rahrtah kamyd || 177 || 

rephSntasyShnafi ca visarjaniyasya kamyapratyaye sir na bhavati | 
dhuhkamyati*’’ j ahahkamsrati®* || 

{451 supas ti |! 178 || 

hrsvat parasya visarjaniyasya subantad vihite takaiadau pratyaye sir 
bhavati | tara-tama-tas-taya-tvartal-tySh pratyaySh tan kavayah prayojayanti | 

81 I. 2. 72, 67. 

«* I 2. 72. 67. 

81 J, 2. 67. 

sa 1 . 1 171 , 2. 65. 
88 4 . 3. p.10. 

8« I. 2. 63. 
88 1 , 2. 65, 


sarpistaramss | sarpistamam | sarpi§ita!h ! catu^fayam 1 catu^-tvam | catueta | 


mso 'nasevayarfi, tape || 179 || 

nisah saipbandhino visarjaniyasya tafcaradau tapatau para tah sir bha- 
vali I ni^tapatiss svai^iam || anSsevayam iti kim || nistapati svamam 
svaiipakSraih || titi kim || mratapat || i§apoirde§ad iha na bhavati || nista- 
tapti II 

tipa ^panubandhena nirdisrfajn yad gan<>Tia ca j 
yac caikfijgrahanam kim at pancaitani na yafi^luci || 

kashadtsu || 180 || 

kaska iti evampiafcaresu Sabdasu visarjaniyasya sthSne kavargjye pavar- 
giye ca^aipare khayi pare sir adeSo bhavati | kaskah | kautaskutah 1 1 sai- 
pisku|nidikiadigaijapS.thah samastarthah | tena paramasarpi§kundiketyadiS9 sid- 
dham || bahuvacanad akrtigaino ’yam | tena bhiaskara itySdi || 

iti SriiirutakevalideSy&caiyaSakatayanakrte 
&ibdanuSasane cmtamainau vrttau 
prathamah p&dalh 1 1 

1. 2 , 65 . 

*» a. 1. 1 . 173 . 


H auf dem Titelblatt : 
sSkataymavyakaranaprarairi - 
bhah \ I Srijimndraya namaJp \ | 
P fangt mit Verdming aus- 
druckenden Worten an, wo- 
von ich nur ’’mtnak | ’/JO- 
'rmk zu lesen vermag. 

13 3 prakdSayasctrfitd" P, prakdsa- 
yaccmtd° °yairisatd (corn 
°yacctmtd°) H. [St 1,] 

7 °patiyyah BP. [St 3] 

8 Die Strophen von 4 an fehlen 
bd P. 

8 sarv<d5strSmbudhtriibudh%° 

(corr wie im Text) H. 

9 sayasaMri B. fSt 4] 

10 °grandham B. [St 5] 

10 sampuiifa H. 

11 ® marhatsyasana° H. 

14 tasyatV^ H. [St 7] 

17 §d$trosafjihah{a}o° H. [St 8] 

21 °dihmhi H. [St IQ] 

22 “dhdtu B. [St 11] 

25 vTttauttau H. 

J4 4 B kiirzt den Vers mmah sri° 
usw mit n° Srtvardhamand- 
yetyadi ab 

6 yogyata athavd B. 

8 sdksatsakala° H. 

8 HB om. namak. 

14 ff Die Lesarten von H. fui 
die pratydhdra-^^a& sind 
nicht angegeben, weil der Text 
zu sdir verderbt ist. 

24 14 “varnlan B. 

18 °pddhanah H. 

22 BH om. den Veis uccaii 
uddtto usw. 

24 H om. r ity anena . . r-ya 

24 f B om. i m Ivarnasydpi. 

25 lugrahandd B 

26 lukdre° B, lukdram° H. 

29 va ya atmanak B. [!]■' 

31 at (st ok) B. 

31 H add. kia^ hinter ptasu 

15 2 vamasya B [2] 

3 aPmand sdha bhavati H. 

6 H add. ri hinter ut [3] 

7 rvyon H. [4] 

8 videyak P. 

9 agakarana H 

9 asmd H 

13 bhatfat B. [5i] 

15 Mss. sva. |[6] 

17 katdakat H. 

19 sprsie^atsprstatti mvrta" P 

20 sthanama H. 

22 tnaaicu° B. 

23 ostkyah R 

23 ka^thostham \ ekesam H. 

24 ’’ete^dm BP. 

24 jtkurapd B. 

25 bitulaha B, lutula° P, itvla° ] 
{47} 25 26 kmthy<md° P. 

29 Mss. aaa. 

16 1 luvama° B. 

8 ktenitjak H. [7] 

9 B om. alpapranasya 

10 ghasacato B. 

11 H om bhavati. 

12 amusyai H. 

13 stryanyatonuk B. 

* [Hereafter the referenceai in square brackets are to Sutias ] 


1>IE grAmmatik ^Ikatayana’s 


s. z. 

14 yuvati B- 

18 H fiigt tu binter samjndyani 

19 ivMtm B. 

22 vatkofryarfi H. [9] 

22 yavaddhd B 

29 °mdsedhyartthah B, °mdse- 
dhyarddah H [11,] 

30 vidhdtavye nach ca PH 
32 p^tuk B 

J712 bhrdtd dayodhikah H. [14] 

13 ekapitrkamdtjkah | patamo° 

14 fivasati H. jivati salt P 
14 putrddih B 

14 P hat ca nach t>hratan 

15 f. paramaprakjtir gargah sydt | 
gSrgts tadanantarah | vrddho 
gargyas triiyah H, 

17 gdrgydyane yuvd B. 

17 H om yuvd. 

24 pitrpye B. ,[15] 

24 pitam Bi 

25 iivSdgdrgyah H. 

25 B om. gdrgyah 

29 °patyam kutsau yuvd B. [161 
18 2 hardni H. [17] 

3 devadattd^ B. 

8 yasydkamvddt° B 

9 iabdasydcdmddtva° B. 

9 B om. sa hinter vd. 

12 “ ecatnchddau B, deSavyeHchd- 
dau H. [20] 

13 B om. yasya 

13 °r«M bbhavatt sa® P 

14 PH om. pratyaye 
18 14 P om bhavatt. 

14 viddtavye P 

15 H om. skaunagariki, 

15 bahtka° H 

17 H add gonarddiydh. ziwischen 
hhavati tuid chddd° 

17 phibhddau no bhavatak B. 

S. Z. 

20 °peks5m H ,[21,] . 

21 atkttcakrakah BH. 

26 H om. pdpacya [22] 

27 mvrti}), BH 

28 dada° P, ddddghvabh H. [23] 

29 ddrupo dhdrupasca B, dddd 
, Tupopalaksitaro H 

29 sak abakdra” PH 

30 dydrupau B, dhdturupd H 
30 pronild H 

19 1 dheth B. 

2 °daddti H 

3 avattdddtam B 
5 ghma B. [24] 

7 P om apratyaye. 

7 pratyayah paro B, P om. 

7 abhyamandyatah H 

8 ’Obhimamomandi^ati H. 

8 prdsddiyata H. 

8 prdsisddiyapati H. 

9 utsukailvd H. 

15 B om. itt nach °pan [25] 

16 °tikrdma'* P 

18 vrksavr° B 

20 f. °bhdvasya cakdpraka° H. 

22 pTdktameva B. 

23 parasyate H 
25 yadarthah H 

28 ddjanta B. [26,] 

30 sukavlkrtya B und H s. m. 

30 patekrtya H. 

30 uryadt B 
32 °ddcsdrthttrtnydf B. 

20 3 °upadeia}f. P, [27] 

3 parigraha H. 

5 bhavati B. 

8 pitya B. 

10 kdrikd H. 

[48]} 20 ID karttrom B 

13 visyamma° B, vtd^mema° H. 

19 °Sttm}no B. [29] 

1. TEIL 


S. Z. 

29 va nach bhavati H [31,] 

33 f. B om anyatra usw 

21 4 amtyadanamupa^esaii H, [33,] 
8 samye H [34] 

10 f H. om adhi kjtva 
12 H om tena 
12 °kdpate H, 

12 “tvadhiti B 

15, 16 In P ist die Stelle abge- 
brochen. [35] 

17 avanlkrtya B. 

18 f H voller Fdiler ' [36] 

20 PH om mtyam 

21 gatva B. krtah B. 

22 1 tasvondSmadhantasyairiktan- 

tvmtisuptasva° B, ’‘madai}- 
tasyani° P, “madanUtsamktva” 
H. [39] 

4 sunaptasu° B 

6 H om. nam u£C(ns°. 

6f adhonast B 

7 vamddttafj^ \ ravomditalj B. 

8 vy5§ayesitya° B 

9 pttdh,idvai° B, padt’^ H. 

10 dhrtva B. 

10 °mojam B 

11 B om hartum 

12 f. ptasvabham H. 

16 trsu B 

17 yatna veti B 

18 Spa B, Spa H. 

21 idudarntam P. [40] 

22 iabdadv<mdva° B. 

28 pratyaya krto sasthya H [41] 

30 pastkydnta° P, pasthantS° H. 

23 5 ghadya^ H [43] 

7 pro^imsti H 

14 sparttke B [46] 

21 Mss. sasthya [47] 

21 sthanmtdlak B. 

22 yontyorla tasya B. 

23 24 sertSfd B. 

S. Z 

28 dvipam B, dvipah H [48] 

24 1 °mdav B [49] 

3 jascassi H. 

3 B. 

3 H om dhandni. 

5 H om. }ka 'ntah itt yah. 

10 f H om. suval lope 'pi. ,[50] 
13 so H. 

13 H om na 
21 tatTallu° B. [51] 

27 B add ycduci hinter yamdhau 

32 taksatt kastaiat B. 

33 kastataditi H 

34 prayinayanni° H 

25 3 sadik ity ato jai H. 

7 macchitakah B |[52] 

10 didadilf, H. [53,] 

11 tastadsorSca H 

12 ivapitsSye H. 

12 veditS B 

15 bhavdrdchwrdh, B, bhavanchu- 
mh. H. [54] 

18 vUe^ya H. [55] 

18 samudayoyasySntontc'^ P . 

18 H om. 'ntavayavo, 

19 yvii B, yvra H 

20 jayastava B. 

26 °myaghd2 H. [57] 

28 ghanddi" H. 

30 smarani H. 

^2 saptamyarthasyaB.. [58;] 

26 1 °krtySdih H [59] 

2f. P am den ersten Satz bis 
2 patyam B 

2 ran ityd B. 

3 tid iti atitaa H. yurmitaditya- 
tas B. 

4 P om. vUesanam. 

4 samudayasya BH. 

11 '‘gandhyabmdhu^ (bddesmal 
fur ^gmdhya°) BH. 



12 syasyais B. 

16 sakma-asya B [60] 

{[49} 26 18 °nakulam° B 

19 samkothtnam B, samkodinam 


19 vySkroM B, vydkrom H 

24 H om sddhu no mksatu ,[61] 

25 kurupi gramam gacchka B, 
kamfu gramatfi ca H 

26 Mss om. yavdn und Icsen 

26 iaktutftSca B. 

26 f H om. devadattma no data- 

28 if. odanam pamca na bhavis- 
yatt 1 mama bhavisyati ] 
paca I tava bhavtsyati j oda- 
nam I tava bhavtsyati \ 
arthat usw. H 

30 prakaranadvagabd^ B, °dvd- 
gatdta!vapraka° H 

31 mahvhau B, °A(n H. [62,] 

27 2 nantam ke p<ne padamT. [63] 

3 kyaccham H 

5 B om kya iti kirn || sdmdn- 

yah II 

5 H add sund vimanyah zwi- 
scben sammyah und suit. 

6 vidvdlya° B [64] 

9 urnndlayuh H. 

10 yarpcim H. 

11 Mss vrtya''. [65] 

12 pararttha'^ B. 

13 tasyantasabdah H. 

17 saifitam tamtam P. [66] 

17 P om. pratyaye pare. 

18 padatfl P. 

21 P om. pare. [67] 

24 varttamanasyano B [68] 

25 f sama &ama BH. Das amma- 
sika-2xa.dkieD. fdhlt ebenfalls in 
dai iibrigen Beispiden bei B. 

S. Z 

28 patahputrddya B, °trddtt H 
(nur einmal ')• 

28 3 P om. a padapm'*. [70] 

10 P om. nanyah. [72] 

10 f. B gibt das Zeichen fur die 
Plutiemng stets mit nu 
wieder; bei H fehlt legliches 

28 16 rtakaya BH [73] 

17 iko yantabhir'^ B. 

20 PH om. vd. [74] 

27 svo vd fur vd PH. [75] 

28 f. Die Lesartesn H’s smd vollei 
Schreibfehler und daher un- 
berucksiditigt gdassen B best 
durchweg lu fur I 

29 43. pitnsabhab | pitrsabhab | 

luti I pitrlkdrab j pttrlukdTah j 
pitrkdrab I luvartfasya j pit- 
lurtaka° | pitlurtaka° | pUr- 
taktt° I luti luvarnasya pitkil- 
taka’‘ I pttlu lutaka° | pttr- 
tanaka° B, rti | pitrrsabhah | 
pttrrsabhah | prtrsabhah j tit 
pitrlkdrah | pitflakarak | lakd- 
vtmmya piearta ikd ity'‘ pita 
rtaka° | lati j lavarnasya pitl 
Uaka° I pitl lataka° {z7f&- 
mall) pitrtaka° H [76] 

8 rP»dddt‘‘ PL 

12 f. nityona dirggho adeio B 


16 iasyat B. [78] 

29 °vahosyoh B ,[81i] 

30 1 "nara B [82] 

5 saltaka H. 

6 eimcyaic B. [83,] 

11 yadavarjjasya B [84] 

14 svaiTasad° B. [85] 

18 omddi^ B. [86] 

21 adyarSydt (st. ad'ya or'*) B 

1. TEE, 


31 11 jnavayovah -ti}am rndrnabh 

B : H verderbt ! [90,] 

32 2 B om. sdca iti nivTttam. [95] 
5f. laksaifapratipa° usw nur in 


29 B om. a u^nam osnam [101] 

33 9 f. tacca vd na bhavati P. 


11 u B. [104] 

[50} 33 12, 14. u B. 

19 kttmi B. [105] 

21 ca bhavmtt H. 

24 °nmkasva^ B [106] 

26 hda mdtrarn B 
26 f. Das Zeidien fur das anu- 
ndsika mrd uberall weg- 

34 3 °ndsiko bhavati B. [108] 

3 Das anun^t^a-Zeichen fehlt 
bea B 

6 nimittasve B, “ttasyaso P, 
°mitrasvo H. ,[109] 

14 B om ^ha°. [Ill] 

23 f. In B fallt das Zeichen fiir 
das anundsika durchweg aus 

[ 112 ] 

24 kima hmalayati B ; beidesmal 
hrse B. 

28 Samrdt scnmdjau H ,[113] 

35 3 B om. k(^ ccha° und ka& 

chadayati H hat nur kas 
coda'’. [115] 

6 tac iete, tach sete B, tac ^ete, 
tacete H. 

11 proktundva B [116] 

12 H om. madhvv atra und 

13 trasvvatra und tra^vatra B. 

36 6 tarsah B. [122|] 

11 cd (st. vd) B. [124] 

16 ff. nu fur das Zeidien dei 
, Fhitiemng B [125] 

S. Z. 

19 °mdmlj. B. [126,] 

37 3 tnpsdyadvyaktih B. [12S] 

10 damadamd ka” H. [130|] 

16 yamddtti B. [132] 

21 taptd BH. [133] 

25 H om tnsamyogah trisam- 
yoga iti B. [134] 

■ 28 corrd B. [135] 

310 rmtsyate B. 

38 4 icauh B. [137] 

8 jha^ayati B. 

10 rmaidt B 

10 f. tospiti (st toh si tit) B. 

38 11 '‘sutve B 

12 f« B. [138] 

18 tospadd° B [140] 

26 ^mdhah B. [141] 

39 14 d'etat B. [146] 

19;atB. [147,] 

21 &cah samyogasya B. 

22 Durchweg °«cA° m bhavddc 
Sete usw. B 

26 f f^dhpdhi und nrmdhpa B. 

40 3 bhavdnschddayati B. [1510] 

5 sarukah B 

6 praSdn carati B. 

11 samaskfsi BP. [152] 

16 avanfdntdd P. [153] 

18 dcaksano B 

18 vrksac B. 

22 fif. Die undeutlich auszuspre- 
chenden y und v werden in 
B durchw^ mit eineim can- 
drabindu bezeicfanet. [154] 

28 f Das V im ersten pataviha 
und das y ha ersten devdyd- 
sate sind in B mit candTo- 
bindu bezeichnet. [155] 

41 3 deva^ dsate B. [156] 

4 bhagaydsva B. 

SSravoffoB. [157i] 



s. z. 

9 deva yatra H. 

13 cetsakaro B. ,[158] 

23 rosno” B. [160] 

26 di^ghdhdyayam B, dirgha- 
hdyayam H. 

42 3 sojSrdayate H. i[1611 

3 pitardayase B. 

7f B best ^pati (si. gth^) 
und om. [162] 
dhuhpalth und praceto rajon 

7f. ahdpatih, gilpati und dhul- 
patt H 

10 saScavy° B. [163] 

13 aseda B. 

22 kupoumkoTpam B. .[166] 

23 khayi pare nkarpa und so 
{51} durchiweg « fur X und 
r fat bei B. 

42 26 B ona antah pacaii. 

27 dadbh^ psatam BH. 

S. Z. 

29 tvtasaJ}. steh sih B, tirasaste 
sih P. [167] 

43 3 B otn. nityam [168] 

5 ter iti . purah krtvd nur ui 


26 pare nach kkayi P. [172] 

44 19 ayaskumbhafi, B. [175] 

21 ayakah karoti BH. 

44 26 f. tan kavayah prayojayanU 

nur m H ; m P ubngens fallt 
das Ganze nut kdmya begin- 
nettd [176] 

45 3f. tan kavayah° nur in H 


9 H add ntstaptd aratayah jj 
anyatra \ nach nistapaU 
svamam. [179] 

13 °kddgrakanairi B. 

13 yad sluci B. 

20ff. P om §rl B om. Sri ... 
krtou, vrtau PH. [180] 

{52} 2. TeU. 

** M «• 

Ubersetzung der Sutras und Erlauterungen. 

Wegen der Ubersetzung der einleitenden Strophen s Einleitung S. 7 f. 
Das iabddrthasambmdha (S. 14, 5) ist sidieilich dean ersten VSrttika des 
Katy&yana sUdhe sabdarthasambandhe entlehnt und ist daher im Smne Patan- 
jali’s zu fassai, der das Komp als em dragliednges Dvandva auffasst Das 
yogyata des Kom bedeutet wohl nidit mehr als sambandha oder vielmehr 
das Geeignetsem den Zusammenhang (zwischen dem Auszudruckenden und 
dem Ausdruckenden) herzustellen ^ — dhaTmmlhakama° usw • ’’Weil der 
Erkerintnis des wahren Wesens von dharma, artha, kama und moksa die 
Kenntnis der Woite und (deien) Bedeutungen vorausgeht, soil der Weise die 
Grammatik kennen lernen 

Die Astadhyayl hat 14 ^wafycfeara-Sutras Sakatayana hat die Zahl ura 
eins vermindert. Emem, Varttika Katyayana’s zufolge hat er den musvdra, 
visarjaniya, jthvdmuliya aufgenommen - — Gestutzt auf das Varttika (P 7. 1. 
9 Via. 5) rkaralkdrayok savamavidhth lasst er den Vokal I nut iwenigen Aus- 
nahmen durchweg unberucksichtigt. Die Homogenitat von r und I wrd zwar 
nicht ausdruckhch erwahnt, ergibt sidi aber nach diem Kom durch ein 
jnapaha. !§. 2 . 3 27 schieibt die Plutierung der Vokale nut Ausnahme des r. 
aber einschliesdidi des I, imter gewissen Bedingungiein vor Es ware mcht 
notig gewesen, heisst es, die Einschliessung von I dort ausdrukhch zu erwahnen 
wenn mcht r und i homogen warai und demzufolge die Ausschliessung von ir 
auch die von I nut sich gebracht hatte. Das zweideutige n des JFSaim’s^hcn 
pratyahara Ian (vgl MBha§. Vol. I, S. 34 /.) wird hier durch ein neutrales 
n ersetzt ; das t des iSiva-Sutra hayavarat wird als zweddos aufgegeben. Bei 
dieser letzten Anderung ist er Candra gefolgt {53} Femer weicht '§a3cat. 
von seinen Vorgangem darin ab, dass er die nicht aspirierten Mediae und die 
aspirierten Mediae und Taiues in dm glachen, phonetisch emheithchen Sutras 
(jabagadadaS, jhabha^hadha4ha§, khaphachathathat) angibt , dagegenerschei- 
nen sie bei Flanim und Candra in schanbar wiUkiirlidi zusanunengjesetzten 
Abschnitten {jhabhan, ghadhadHas, jabagadadai, khaphachatkathacafatav ), — 
Alle drei MSS lesen einstinimig pratyahdrayan (S 14, 18) ; ich weiss es aber 
nicht recht zu deuten 

3. Apts gibt yogyatS wieder ‘ the absence of absurdity in the mutual connec- 
tion of the thuigs signified by the words', und verweist auf die Definition un SSkitya- 
darpana (ed. Kane, Bombay 1910, & 8 oben)': yogyata paraspara satpbandhe 

* Nach Kielhobn, Ind. Ant. Bd. 16, S. 26'>. 


DIE grammatik SXecatayana’s 

1 (Ein Laut Oder ein Aggregat von Lauten, das) mit einem stummen 
Buchstaben (versehen ist, bezeichnet alles Dazwischenliegende) bis zu deni 
stnnunen Buchstaben mit Emschluss seiner sdbst 

P. I. 1. 71 , C I. 1. 1. 

Itet ist in tt& a it aufzulosen Der sandhi erfolgt nach S 1. 1. 82, 86. 
Den erstai drei Sutras iSakat's entsprechen die namlichen m dem Candra- 

2. (Em Koosooant) mit (dem stummen Laut) u (bezeichnet) den 
homogenen ( [siia] einsdiliesslich semer selbst). 

P. I. 1 69; C 7 1. 2. 

Nach 6. J 1.6 hasst sva ein homogener Laut und ist mfolgsedessen gleich- 
bedeutend mit dem PBinmi'schen savarna Dem Wortlaute nadi also gilt 
die otnge Regel Mr alle Laute, die emen homogaien Laut aufwasen In der 
Tat aber wird der stumme Laut u nur an k, c, t, t und p angefugt , daher ira 
Kom. svasya vargasya 

3 (Em Vokal) mit (folgendem) t (bezeichnet) nur emen so langcn 
(d. h. den Vbkal semer Quantitat) 

P 7. 1. 70.C 7. 1. 3 

Diese und die folgende Rieigel schliessen em, dassi em mit t unvci- 
bondener Vokal, ausser Suffixen, Augmenten und Substituten, zuglach die 
entsprechenden langen, plutierten und nasalierten Formen desselben bezeichnet 

4. Em oparativer (Vokal d h. an Vokal, der em SuflSx oda Augmait 
ist, bezachnet nur den Vokal semer Quantitat), ausser wenn er mit (dem 
stummen Laut) g versehen ist 
Cf. P 7. 1 69. 

iSlakalt. hat das Planmi’sche Sutra 7. 1 69 mit Rucksicht auf die Pari- 
bhasla (19), bhavyantanena saiiarndndm grahanam na, verandert EM wird 
dem Wort pratyaya sane etymologische Bedeutung angiewiesen Vgl. Kaiya(a 
dazu : pratlyate vidhiyate iti yaugikasydtra pratyayfasya grahcuyam iti bhavah. 

- -Zu amum, amu des Kom. : Der stumme Laut g hat zwei Funktionen ; die 
eine wird hia erwahnt, die andere erst in 7 1 . 100 Wird g einem operativcn 
Vokal wie z. B. anem Suffix angefugt, so hat nach dieser R^el das wirkliclie 
Substitut ebensoviel matras wie der sthdnin , z B nadi iS 7. 2. 44 wird 
fur das d von ados m substituiert, {54} woba ig)ki fiir den unmittdbar 
darauf folgenden Laut antritt Folgt also •au des nom oder des acc. du. auf 
das ®iga <ida, so muss das Substitut ebenfalls zwei matras haben, und infolge- 
dessen tntt amA (und nidit amu) fur adau (aus ada + au) ein. Folgt 
dagegen an kurzes a auf d, wie m adam (aus ada + am), so muss em kurzes 
u Mr den folgenden Laut substituiert waden ; infolgedessen heisst der acc. 
sing. mas. amum. 

5. (Buchstaben oda Aggregate von Buchstaben, die in der Grammatik 

2. TEIL 45 

angeftihrt, aber m der gewohnlichen Sprache) nicht gebraudit (werden, 
heissen) stunune Budistabea (tt). 

Cl P. i. 3, 2-9 ; C. 2. 1. 5 ; H. I. 1. 37. 

In tuvfprn (■= Wz. vep) smd die Buchstaben t, u, f und n it. Dieses 
Sutra vertritt die Regdn Plajjini’s 1 3. 2-9. 

6. Werai die Artikulationsstelle und die Tatigkeit (des Mundes) gleich 
smd, (heissen die Laute) homogen {sva). 

P. 2. 1. 9; H. 2. 1. 17. 

MS P gibt fur dieses Sutra ausnahmsweise den ganzen, wie in Tect 
gedrueiten Kommentar.® — Die MSS lassen einstinimig den visarjaniya hinter 
sva aus. Der Ausfall ist nach iS. 2. 1. 165 (= P. S 3 36 Va 1) freigesteUt. 
Die Lesart ohne den vtsarjasniya mag wohl die ursprunghche s^, weal dadurch 
das Sutra um emen Buchstalien kurzer ist. Der Koosequenz halber habe kh 
die vollere Schrabung bedbehalten. — ^Die Unterschadung der sechs kurzen von 
den zwolf langen und plutierten a heruht darauf, dass das asya der kurzen 
samvrta, dagegen das dex langoi vtvrta ist. Vgl P. S. 4 68. — Nach dem 
Herausgeber der Bombayer Ausgabe des Prakriyas (S. 4, Anm. 1) heisst 
%jk : osthayor agrabhagah, Es ist mir nicht bekannt, dass man sjkvan (oder 
srkkan „ Mundiwinkd “) ala das asya von v angegeben hat, was man sonst 
annehtnen nniisste. — Was PBn. asya nennt, heisst bei iSakat sthdna, dem pra- 
yatna PagQL.’s entspricht aber bei iSakat asya. Vgl. CintSmaipi und Siddhantak. 
S. 4 (unter tulydsyaprayatnam savanfom): talvSdisthanam Sbhyantara- 
prayatnaS ca,' — Diese Regd fehlt ganzhch bei Candra ' — Za amikara^ . 
Ausser in den Ableitungen von klp und m den Worten, die den Laut I bezeich- 
nen, komnit I nach den indischen Grammatikem auch bei der Nachahmung 
einer Person vor, die aus Unfdhigkeit r auszusprechen an dessen Stelle I 
sagt. So sagt z. B. erne alte Brahmanenfrau pitl, Itaka und pitltaka anstatt 
pitr beziw. rtaka und pitftaka (MBhSj?. Bd 1, S. 19, Z. 16 f.) . 

7. (Von den versdiiedenen Substituten wird dasjenige vorgezogen,) das 
(d^ urspriinglidien Laut) am nachsten (steht). 

P. 2. 1. 50; H. 7. 4. 120. 

Nach H<>m Komi, kann die Verwandtschaft auf der ArtikulaticttissteUe* 
Qualitat, Quantitdt und Bedeutung heruhen. In muni + indta £55} must fur 
die bedden t nach 2. 1. 77 ein langer Vokal substituiert werden ; es wird in 
diesem Falle nadb unserer iRegel das lange * gewahlt, weil die zwei Vokale — 
dafi kurze und das lange i — ’das g^eiche sthana haben, und kemer vmi den 
librig^ dem sthanin naher steht als dieser — Zu den Beispiden zu pvatndnena, 
vgl. Anm. zu 2. 1. 4. — vStai}dyayuvati = ^ junger wdWicher Nachkomme 
von V : daradavrndaHka-d'i& schonste unter den D.-Frauen. Das erste femi- 
nine Glied des Kompositums nimmt diegenige maskuline Form an. welche 

8 Vgl. F-iTiirit img S. 11 unter der Bescbreibung des MS. 

40 DIE grammatik Iakatayana's 

dem uTspninglichen Feminmum. was die Bedeutung anbelangt, am nachsten 
steht also wird fiir vateof^ nicht txftoftdo substituiert, sotvdem vaUa^ya ( - 
em lii&imlicher Nachkamme dea V.), das jenem noch naher steht als das 
vatan/da da beide den Abkonunling bezachnen. 

8. (Eine Operatwn, die sich auf) an Verwandtschaft (ausdrudcaides) 
Wort (bezieht, tritt nur dann em), wenn die Verwandtschaft (m da Tat) 
vorhanden ist. 

H 7 4. 121 

svahtrya = Sobn des Schwiegervaters, Schwager , aber itffls«n = Sohn 
Pinpfi Mannes, der einem Schwiegavater ahnelt* Dxese R^eJ vermag ich 
weda bei FS|n noch bei seinen Kom. nachauwasen. Vgl. jedoch Ujjvala- 
datta’s Kom. za Uin 1 45 : sambandhiSaMandm tatsadr^at pT’Otisedhah. 
H, 7. 4. 21 gibt das Sutra nut dem Kom. fast wortlich wieda— -Das Sutra 
wahrachemlich nur dazu, den abgdeiteten Nominalstamm svc&un m 
dem obenerwahnten Sinne zu rechtfertigen, dessen Bildung sonst gewisse 
Schwiaigkeiten baeitet.® 

9. Die (Wortformen) auf ghat (■='<*?) und 4ati (l='flfi) (sind wie) 
Zahlwdrta {[savikhyd] zu behanddn). 

P. 1. 1. 23, H 1. 1. 39, 

Zum Suffix ghatu . Dem Suffix ghatu bei 'Sakat steht bei Pan. vatu 
gegatuba (P. I. 1 23) Ehes smd beides Suffixe (Konsoinanti4- at), die 
gawiasen Pronotmna angefiigt waden, wie z B. in yavat, tdvat, kiyat, iyat 
usw. Da bei FSlU'. das nonnale Suffix vat(,u) (wie m den Bildungen ySvat, 
tavat) heisst, so wird hmta Mm und idam fSr das v des Suffixes daa Subshtut 
gk (= ty)* £56} vorgeschrieben, um die Bildungen kiyat, iyat zu ermoglichcn 
(P. 5. 2. 40, 41} . Umgdcehrt heisst bed €3kat. das nonnale Suffix in diesem 
Falle ghatiu) = iyat (wie in kiy<a, iyat, 6 3. 3. 68) ; infolgedessen schreibt 
er 5. 3. 69, 70 die Substitution von v fiir gh hinter etad usw ietdvat) vac. 
Das Ergebnis ist genau dasselbe. Warum iSSkat. die Anderung voigenoramen 
hat, IMsat sich vorlSufig nidit ermittdn. — ^Bei Candra wird da Terminus 
saffikhyi nidit weita dedniert. 

* VgL Paribha^du^. (Test) S. 15, Z 9 : ivahirasadrSasySpatyam ity arthake 
SvaSutih usw. Nach Ujjvaladatta (Un. 1. 45) ist ivaiura hia em nom. pr. iva&uro 
nama katcit ta&yapatyam. 

® IHe JRegel P. 4, 1. l57 rajaivaHtrSd yad wird dutch die PanbbSsS (15) . 
iautfaimkhyayor muhhye kSryasittupn&yayah modifudert und infolgedeSsai udrddas 
Suffis ya dem ivtdura na dann angefugt, wenn das Wort seine primSre Bedeutung 
Uat ; diese wiederum b'edaif da in der l^iiUiai^ selbet mcht auegedihckten Ergkn- 
zung : ktrit oSyatii nySyo na pratipadikctkarye him. tUpSttam vi^ySrthopasthSpakam 
vtfiftarSpam yatra tadjiapadakarya eva (I^ribhasaidu^. Text S. 15, Z. 5 — 6 ; 
Tiansl. S. 88 und Anm. 1, S. 89 und Anm. 1) . 

* Nach P. 7. 1. 2 , chensosicheiiidiauchba^kat., obwohl ich die £5^ Regel 
bei jhm nidit nadiweisai kann, vgl, jedodi PraknySs. S, 183 (unter Siutra 913) 
ghasya iy. 

2. TEIL 


10. bahu und geata (warden) in (dei Bedeutung) Verschiedenheit 
(Mannigfaltigkeit) (iwie Zahlworter bdianddt). 

P 1 . 1. 23 + Ka§. ; H. I 1 1 . 40. 

Der Zusatz bhede bemht auf der Angabe der TCaAiIra (zu P. J. 1 23) : 
bahnganayor vaipulye satfighe ca vartammayor tha grakaijani nastt | saifi- 
khyavacinor eva. Vgl, Cintam. 

11. (Die Wortform) adhyardha (wird) vor (dem Sufiix) ka und in 
Kompositum (wie Zahlwoiter behandelt). 

P. 1. 1. 23 Va 5, H. i 1 41. 

12. (Eine Wortform, die ein Kompositum ist, dessen) erstes GUed 
ardha (ist, und auf) dot (d h ein Ordnungszahlwort bildendes Suffik aus- 
lautet, wird wie Zahlwdrter behanddt) 

PI 1. 23 Via 7 ; £r. I. 1 42 

13. (Die Nachkoramenschaft) von Enkd an (heisst) vrddha 
P. 4 1. 162 ; H 1 2 

In der Grammatik des Candra sind die Termini vrddha und yuvan nicht 
definiert , vrddha entspricht Plaaj.’s goira 

14. Wenn ein Glied in der aufsteigenden Lime Oder em alterer Bruder 
noch am Leben ist, (heisst die) nicht weibliche (Nacfakommenschaft). von 
Urenkd an yuvan. 

P. 4. 1. 163 + Va. 7 und Bh. ; H. 6 1 3. 

varjiiya ist jeder Vorfahr, von dem Vater aufsteigeind, der die Ursache 
eigenen Dasdns isL*— Das entsprechende Sutra PBlnim’s (4 1. 163) fivatt iu 
varjiSye yttvd ist aweideutig, weil da der sa^]Hm unspezifiziert bleibt. Dei 
Uienbd ist es erst, der liberhaupt yuvan genannt werden kann. Die richUge 
Konstruktion ist aus dem foJg«iden Auszug aus dan BhSsya zu, dem Sutra 
ersichtlich : evcaji tarhy apotyam evabhisantbadhyate na tu pautraprabhrti- 
&amSnadkikara>}am apatyam | mavant vijMyate pautraprabhffti yad apotyam 
iti 1 kathom tdMi | poutrt^nr&hTter yad apatyam iti (MBhas. Vol. 2, S 265, 
Z 19 f.). Dieser ErMarungsversuch ist offenkundig dn Kunstgnff Pat.’s 
und bezeugt noch einmal dessen Bemiihung die Unfehlbarkeit PSirn 's zu doku- 
mentieren. Unser Grammatiker lehnt ihn ab und vereinfadit die Sache, indem 
er p7j prapautfadi ausdtiiddich hinzufttgt— Der Singular salt ist auffallend, 
aber fur den SQtrastil bezeichnend Er soil doch ausdrucken : wenn ««er von 
den zwei genannten am Leben ist ; soloh wurde bedeuten wenn der vatfisya 
und der bhrdtr bMe am Leben sind. Zu sati ist also anyatarastnm zu erganzen. 
Hema. ebenfalls fivati (d. h, satt. H. 6. 1. 3) .— astri des Sutra beruht 
auf Va. 7 zum oben owahnten Pajnini’schen Sutra ' — vjddha statt Painini’s 
gotra, wohl gewissetmassen als Gegensatz zu yuvan. 

15 (Die im Sutra 14 gdehrte Bezeichnung ist) friegestellt, warn ein 


an Alter und Wurde hoher stehender sapU},4o des lebenden (Nachkommen 
vom Urenkel an noch am Leben ist) 

P. 4. 1. 165 + Bh. ; H. 6. 1. 4. 

Die Erklaiiung in dem Kom von sthana ist wenig klar. Der Ausdruck 
vayassthana stanunt aua dem MBhSi? her, wo Pat bei Gelegenheit dei Erklar- 
ung von stkaviratara (P. 4. 1 165) sich folgendermassen aussert : atha- 
sthavirataragrahanam kimartham \ ubhayato visi?te yatha syat | sthanaio 
vayastaS ca. Dies passt auch gut zu der tiblichen Bedeutung von sthavira . 
bejahrt und wurdig. Man vergleiche den Gebrauch des Wortes (pa tkera) 
bei den Buddhisten. Wenn Pan. nur den Allersunterschied hdtte ausdmcken 
wollen, so hatte er emfadi jyayasi oder ahnliches sagen konnen Sidierbch 
ist der Ausdruck hier auch so zu verstehen, wie bd den alteren Grammatikem 
Warum der Kom. sich so ausdriickt, sei dahingestellt. 

16, (Die Bezeichnungen) yuvm und vrddha (sind freigestellt), weiui 
Tadd (resp ) Verdirung (auszjidrucken ist). 

P 4. 1. 166, 167 = P. 4. 1. 162 Va. 2, 163 Va. 3 ; H. 1. 5 

ymavjddham ist nom sing, und kuisarce ist lok sing eines Neut.- 
Dvandva. iSakat gebraucht das Dvandva fast durchweg als sing, neut." 

17, Em Eigenname (wird behebig) du (genannt). 

P I. 1 73 Va 5 : C 3, 2. 26 : H. 6. 1.6. 

du entspricht dem vrddha bei PSij-. das 5akat. fiir PSti ’s goira verwendet 
(s. SQtra 13, 14 Anm.). 

18, tyad usw. (beasseti du)~ 

P. 1, 1. 74; C. 3. 2 28 ; H. 6, 1. 7 

Zu beachten ist, dass der Ubergang von vibhasd zu mtya im Tert des 
Sutra mcht ausdrucklidi erwahnt wird. Es ist mir nicht klar geworden, 
woraus dieselbe iiberhaupt zu «:schlies8ai ware. 

19 (Eine Wortform) von dessen Vokalen der erste a, at oder au (ist, 
beisst du), 

P. 1. 1. 73 ; cf. C. 3, 2. 24; H. 6. 1. 8. 

20. (Eme Wortform, die) nur als Ortsname (gebraucht'wird pS} und 
von deren Vokalen das erste) e oder o (ist, heisst du) vor (den Suifixen) 
cha ( = tya) usw 

P. 1. 1. 75.+ Bh.; H, 6. 1. 9 

Vgl. Anmerkung zum folgenden Sutra. 

21. (Eine Wortform, die) einai Ort in Osten (bezeichnet und von deren 
Vokalen der este e oder o ist, beisst du vor doi Suffixen cha [ = iya^] usw.). 

P. 1. 1. 75 + Kg.4., C 3. 2. 25; H. 6. 1. 10. 

Die Sutras 20, 21 besagen folgendes : Ein ostlicher Ortsname wird in 

T Cf. 1. 1. 11, 26-28, 33, 36 f., 59, 71, 81, usw. usw. 

2. TEIL 


gewissen Fallen als ein du genanntes Wort behanddt (21), aber auch solche 
anderen Ortsnamen, die nur als Bezeichnungen von Orten auftreten (20). In 
dieser W-eise wird die Regel weder von Panini, noch von seinen Kommenta- 
toren, noch endlich von anderen Grammatikem ausser Hema, fonnuliert Die 
Regel bei Panim {1 1 75) lautet en pracdm deie, was dear Verfasser der 
Ki^ka so erklart : en yasydcdm ddts tat prdgdeidbhtdhdne vjddhasarrtjnatn 
bhav^att Dass Candra derselben Meinnng ist, geht deutlich aus dem Wort- 
laut seines Sutra (5 2 25) * enddyacah prdgdesdt hervor. pracdm kann aber 
auch erne andere Bedeutung haben, namlich, „ nach der Meanung der ostlichen 
Gramrnatiker “ ; so z B. P 5. 1 90 j 4 18 , 4. 1. 17, 43, 160 usw., wo es 
von den Konunentatoren einstixnrriig in diesem Sinne erklart wird. Leider 
gibt uns das MBha§. keinen direkten Aufschluss daniber, wie Patanjali sich 
zu dieser Sache verhalt Sein kurzes Bhasya uber das Sutra lautet : en 
prdcdm dese saipkesv iti vaktavyam ‘ saipuriki saipurikd \ skaimagariH ] 
skaunaganketi (MBhas Vol 1, S. 190, Z. 20 f ). Nach der Ansicht 
Kaiyaita's aber soil Pat die zuletztgenannte Auffassung billigen. Er sagt : 
ktmnd prdggrahmam dcaryantrdeSdrthani vydkhydtam , . | any^a"^ tu 

prdggrahanam desavtiesanatji vydkhydtam . . | bhdsyakdras tu kwytdaxianam 
asiinyat Und Kaiyata hat offenbar Recht Derm die beispidweise gegebenen 
Ortsnamen Sepura und Skonagara miissen solche Namen vertreten, die vrddha 
heissen und denen zugleich die sai^'fta-SuflBxe ikt, ikd angefiigt werden konnea 
Nun aber kdnnen diese Sufifixe iktj ikd, techmsh nith genannt, an Ortsnamen 
wie Sepura und Skonagara nur dann tretoi, wenn diese im Gama KdSi enthalten 
(P 4 2. 116) Oder Dorfnamen der BBhika sind (117) Das erste ist rucht 
del Fall, also mussen sie Dorfnamen der B§hika sem Da aber das Land der 
Bahika gar nicht im Osten liegt, sondem den Teil Indiens bildet, der heutzu- 
tage das Punjab heisst (s Nagojlbhatita's Pradipoddyota zum oben erwahnten 
Sutra Paa^Lini’s), so kdnnen die zwei Namen bd Pataxljaii, welche als Bei- 
spide der Ortschaften, auf die sich das Sutra bezieht, dienen mussen, keiue 
Namen von Orten im Osten sem Folghch kann nach der Ansicht Patafijali’s 
prdcdm mcht nut de§e verbunden sem. Die Sutras £59} iS&kait/s wollen nim 
den beiden entgegengesetzten Ansichten — der Ansicht Patahjah's und der der 
Candra-KS§ikakara — gerecht werden. Der letzten gibt er durch das 21. 
Sutra Ausdruck, der ersten durch das vorangdiende Sutra Die durch das 
Wort eva im 20. Sutra ausgedrucfcte Beschrankung kann auf der Bemerkung 
der Ka^iKa beruhen : de§a iti kim | gomatydrn bhavd matsyd gaumatdh. Das 
Sutra 20 hat meaner Ansicht nach lediglich den Zwecfc, die vier im MBha$. 
vorkommenden Bildungen saiptmfn usw zu rechtfertigen 

Die VSrttikas 6 7 und 8 zu P. 1. 1. 73^ modifiziert durch das Bha^ya 
gotrantdd vdsamastavad tty eva jydyah, bilden ein Sutra im Abschnrtt uber 
die TaddhitasufSxe (S. 2. 4. 2 = H. 6, 1. 12). 

® Darunter ist wohl der Verfasster der Ka^ilta zu verstehen. 



Die ghammAtik §akatayana’s 

Das MS H fugt als Beispiel gonardiya zwischen bhuvati und chddau im 
Kom zu Sutra 20 ein, was vollbonunen unzulassig ist Derm das Beispiel als 
pratyuddharana musste einen Ortsnamen aulfuhren 1) der nicht ausschlies- 
slich als Bezeidinung eines Landes verwendet wird, 2) dem das Suflbc cha 
(—iya) eben mcht angefugt werden kann, 3) der kein Ortsname der 
ostlichen Vblker ist Es ist offenbar aus dem Kom zu 21 emgedningen 

22. (Erne Woitfoim,) die eine Tatigkeit bezeidmet, (heisst) Wurzel 

Cf P. 1 3 1 , H. 5 3 3. 

Obwohl nach der obigen Definition jede Tatigkeit ausdnicklende Wort- 
form dhdtu heissen kdnnte, werden die halbprakntischen Verba, wie 
dnapayati, vaddhati usw aus dem Begriff dhdlu ausgeschlossen, genau wie im 
Papini’schen System (vgL VSrtt 12 zu P 1. 3. 1 und Bhasya daselbst). 

23 (Die vereduedenen Wurzelformen) dd und dhd, mit Ausnahmc der 
mit (dem stummen Laut) b versdienen, (heissen) ghu 

P. I 1. 20, C. I 1. 4; H 3.' 3 5. 

dSi} 1 Kl. {yacckati) „geben“; den 1. K1 (dayate) „schutzen“, 
dudM. 3. Kl (daddti) „geben“, do 4 Kl. (dyati) „ schneiden “ ; dket 
1. Kl (dhayatt) „8augen", dudhdn 3. Kl. (dadhdtt) ”setzen“ 

24 pta usw (gehdren) nicht (der Wurzel an), ausser wenn ein Sufik 
(darauf) folgt. 

H. 3 3. 4. 

utsuka wird nach der Ansicht der indischen Grammatiker von ut mit 
dem Suffix suka gebddet (S 5 3 113) . 

25. ipra usw.,) die sich auf eine (Wurzel) bezidien, imt Ausschluss von 
adhi und pari, wenn (sie) biedeutungslos (smd), su und ati, wenn Lob (aus 
gedrudct wird), und ati wenn Uberschreitung des Masses (ausgedruckt wild, 
heissen) Praposibon (.upesarga) und (stehen) vor (dem Verbum). 

P. 1 4 . 58, 59, 80, 93-95 ; H 3 1 1 

pralambha • Nur nach upasargas wird vor dem Suffix a m £60J tabh 
u eingeschoben , so pndambha, dagegen isallabha, Idbha usw. (cf. P. 7. 1 
67). — Wenn adhi, pm keine itpascergas smd, konnen sie nach, Verbum 
stehen, wie in dgacchaty adhi oder agacchatt pari , hier ergibt sich der Sinn 
Tnnauf, nn^^um “ auch sonst, z. B. aus dem Zusanunenhang . — su stktom 
usw : susUita druckt einen Tadel aus, wie es aus unserem Kommeotar und 
der Bemerkung der Siddhantak. (p. 136) . susiktam ktm tavdtra<> \ ksepo 
’yam hervorgeht. Es ist aber schwieriger zu sagen, was su siktam piyn tlirh 
bedeutet Es kdnnte wohl eanfach ” gut begossen ‘‘ heissen ; allein die Kom- 
mentatoren fugoi immer atra dhdtvarthah stuyate hinzu, was nur heissen 

« Ironisdi gemeint; vgl. im Deotschen- "das hast du gut eetan” obwnhl 
die Pragqiartikd (fctm) befremdend wirkL 

2. TEIL 


kanii, hier wird die Bedeutung der Wurzel (nicht etwa die Ansfuhning der 
Handlung, wie man eiwarten wurde,) „ gepriesen Leider geben die alteren 
Kommentatoren keinen Aufschluss daruber, mwiefam der dhatvartha gepriesen 
wird In der Siddhantak (p 135 Anm 2) jedoch findet sich die 
Notiz . su siktam tti \ sekagatapUjyatvadyotakah suh, W‘iyapujyatv<akrte 
tattatknydkartuh pujyatve gamya evatsd samjM „su deutet die der Vollzie- 
hung des Begiessens zukommende Ehrwurdigkeit (oder geradezu das 
Verdienst) an (wie z. B. des Begitessens ernes heiligen Baumes wie der 
Twiast). Nut m dem FaHe, dass wegen der Ehrwurdigkeit gewisser Hand- 
liingen die Ehrwurdigkeit des VoUzidiiers der betreffmden Handlungen aus 
zudrucken ist, hat der Terminus (namlich katmapTamcmtya) Gulti^eit“ 
Dieser Ansicht nach kotmte su siktam bhavatd geradezu heissm ; es ist ver- 
dienstvoll, dass du begossen hast 

Die vier Worter adht, pari, su und ati in der im Sutra gegeibenen Bedeu- 
tung heissen bei Plainmi karmapravacamya. Dieser Terminus hat dm Zweck 
dieselbm von dm Kategonm upasarga und gati auszuschliessm ; cf Kadika 
zu P. 1. 4. 93 ; gatyupasargasarfijndbadhamrtha karmapravacamyosamjna 
vidhSyate iSakat kmnt die Kategorie karmapravacamya nicht und scheidet 
bloas die obmgennanten Worter aus dem Begriff der upasargas und ti aus. — 
Bei Candra werdm die upasargas nicht defiiuert. — Der adhtkdra" prak ca 
gilt fort bis 1. 1, 38 inklusiv. 

26, (Wortformen, die auf) dac (= -a), cvi (|= -i) (auslauten, die 
Wbrter) Mil usw., Qnomatopoetika (und die Prapositionm in Verbindung 
mit Verbm^®) heissm tt 

P J 44 60-62 ; C 2. 2 25, 26 ; H. 3, 1. 2. 

Zu cviddcsddharmyat vgl. MBhas. zu P. 1. 4. 61. — ti ist erne Verstdm- 
mlung von Panim's gatt und mtspncht diesem ganz genau, wie schon BOhler 
iichtig erkannt hat. Or. u Occ. 2, 701 Anm 1. 

{^613 27. (ti heissm m Verbindung mit Verbm.) kanka, wenn Regel 
usw. ausgedriicbt wird, cdam wenn Sdunuckung ausgednickt wird, ados, wenn 
nicht Mitteilungii, und emtar, wenn nicht Ergreifm (Inbesitznahme) ausge- 
driickt wird, sat und asat, wenn Hoch- respektiv Germgachtung ausgedriickt 

P I 4 . 63-65, 70, 610 Va. 1 ; C 2 . 2 . 25, 27, 28, 32 ; H. 3. 1 3-5 

Zu karika . Amarak. gibt kdrtkd nut ydUnta und vrtH wieder vjtti ist 

“ Ortent und Occident, 2, 697 f. hat in seiner " Notiz ub» die Gram- 

matik des Qaka^yana dieses Sutra ubersetzt. Els heisst doit * ” . Upasargas 

heissm Ti, wenn sie mit Nomina verbundcn smd “ Er ^aubt {^61]} also in den ii- 
upasar£as wno Vrarsitufe der keawatvaivacatuyas bei 1*111011 wiedeizuerkennm. Das 
1 st jedoch ptTi Irrtuni, wie aus der Erlaiiterung zu dem vorangehenden Sutra her- 

Oder positiv gesagt, wenn Oberlegung {pardmarSa) ausgednickt wird. 



” Erldarung “ , also kankS „ erklarende Regd “ Darauf deutet auch sthtft, 
marydda feste R^el“, „Regel“ im Kom. hin.i* kartkakr stammt wahr- 
scheinlich aus der Untemchtssprache der Grammatiker und heisst „ zur 
Kdnka majchen also „ erne Regd oder eine Erlauterung m der Form emer 
karikd fassen “ , denn wie soli es „ durdi erne kdrtkd erMaren “ heissen, wie 
Bohtlingk m semem Worterbudi angibt ? Unter adi versteht der Kom yatna 
(Anstrengung). SoUte da mcht ydtand („Qual, Pern", also kdrtkd = 
Gelangenschaft “ entsprechend dem kdrya im Kautiliya), zu lesen sein’ 
Es ist indessen kaum anzundimen, dass kdnkdkr je im Sume von etwa 
" Gefangenschaft machen“ gebraucht wird. Im Sutra stelit ddt und der 
Kom musste eben erne zweite Bedeutung angdjen. — ^Wegen kartri vgl 
Vopadeva, Grammatik 8 21 

28 kane und manas (heissen ti), wenn das Stillen des Verlangena 
ausgedruckt wird. 

P 2 4 66, C 2. 2. 29, H. 5 1. 6. 

29 Die Indekhnabilen astam und puras (heisst ti) . 

R 2. 4. 67, 68 , C 2, 2. 30 , H 2.1. 7. 

30 accha (in Verbmdung mit Verben) der Bewegung und vad (heisst 


P. 2 4 69 ;C 2, 2 31, H 2. 1 8. 

accAtfgfi/ya =.herangekommen seiend:; acchavmjya = d. s. , acchodya 
- emgdaden habend Hier ist zu erlwahnen, dass accha vad im Sinne von 
didha neu ist accha vad heisst soost : herannifen, begrussen, bewiUkommen, 

22 Was (m Verbindung mit Verben heisst ti), wenn ein Dazwischen- 
treten ausgedruckt -wird. 

P. 2 4. 71, C. 2. 2 33; H. 2. 1. 9. 

22. (Wenn tiros) mit kr (verbunden ist, ist die Bezeichnung ti) frei- 

E. 2. 4 72 ; C. 2. 2 34 ; H. 2 1. 10. 

{62} 22. (Wenn die Indddmabilen) mcmast, urasi, updje, anvajd, 
madhye, pade und nivacane (mit kr verbunden sind, ist die Bezetdhnung ti 
freigestellt) . 

P 2. 4 . 73, 75„ 76 ; C. 2. 2 . 35, 37 ; H. 2. 1, 11, 12. 

Zu bemerken ist, dass selbst das widitige Wort anatyddhdne in dem 
Sutra. w^elassen ist; es wird dem Kom. uberlasen es zu erghnzen. Bei 
He ma . ('2. 1 11) lautet die Regel richtiger. 

1* BtiEtL® gibt zweifelnd sthttt nut „Ge8chaft'‘ wieder (Or, u Oct\ 2, 698). 
^ Zweifd ist bereditigt Nach CorajBROOKii Grammar 124, soU es in diesem 

^ W«ter das MBl^s. nocb 

die ^aSiK^ ewatt 4iei Ausdrii^ke 

2. TEIL 


34 (Wenn der upasmga) adhi zum Ausdnick der Herrschaft (mit kr 
verbunden ist, ist die Bezeichnving ti mcht notwendig) 

P. 1. 4. 97, 98,, H. 3. 1. 13. 

Die mandukapluti-datisi anuvrtU von upasmga wird durch den Umstand 
notwendig gemacht, dass adhi in diesem Sinne eigenthdi weder gatt noth 
upasarga, sondem ein karmapravacanlya (P. 1 4. 97) ist, welche letzte 

Kategorie sich bei 'Sakat nidit findeL 

35 (Die Bezeichnung U ist freigestellt fiir) sdksdt usw, (wenn sie 
die Bedeutung von den auf cvi [= *] auslautenden Wortfoimen haben), ohne 
(aber das Suffix) cm ( [ = i,] selbst zu haben) 

P. 1. 4 74 + Va. 1 , C 2 2. 36 ; H. 3 1 14. 

Wenn die Worte auf cvi ausgehen, so heissen sie gati (i§ 1 . 1 . 26) und 
mussen als soldie ein mtyasamasa mit kr usw bilden 

36. haste und pdnau (heissen m Verbindung nut fer) stets (ft), wenn 
das Sichaneignen gemeint ist 

■P. 1 4. 77 ; C 2 2. 38 ; H. 3. 1 15 

37. (Die Bezeichung ft ist freigestdlt fiir) jivika und upant§ad, wenn 

Ahnlichkeit gemeint ist. , 

P I. 4 79 , C. 2 2. 40 , H 3.1 17 

38 (In Verbindung nut Verben hdsst) pfSdhvam (ti), wenn Fessdung 
ausgedruckt wird 

P. 1. 4. 78 ; C. 2 2. 39 , H. 3. 1 16. 

39 avyaya (heissen die auf die SufiBxe) tas, vat und ndtti (- am) 
(ausgehenden Wortfoimen, auch diejemgen, welche auf die Reihe von 
Suffixen) tasi nut Ausnahme von dhan (i=dha) (ausgehen), dm (m peri- 
phrastisdtien Perf.), (die Absolutiva auf) ktvd (= tvd) und am, (die Infi- 
nitive auf) turn, (die) ft (genannten Worter, femeir diejemgen, wdche) den 
mit Flexicfflsendungen (versehenen Oder den auf die SufiBx-Reihe)' ptasu 
(ausgehenden Wdrtem) ahneln (und die Wortgnippe) svar usw 

P 1. 1 37-40 , H. J. 1. 30-36 

Der Umfang des Terminus avyaya bei PSijim und iSakat ist ganz genau 
derselbe Nur ist die Aufzahlung und Emteilung bei diesem etwas anders 
als bed seinem Vorganger. So heissen bei Pan. avyaya die folgenden Kate- 
gonen von Worten : 1. svar usw. und die (Partikeln (ntp^) (P. J. 1. 37) ; 
2 die auf em taddhtta (63^ Suflfe ausgehenden Worte, wenn sie mcht mit 
alien Kasusendungen versehen werden kbnnen (P. 1 1 38) ; 3. die mit einem 
fef-SufiSbc gdiildeten Worte, iwenn sie auf m oder einem DiiAithong ausgehen 
(P. 1. 1 39) ; 4 die Absolutiva auf Jttvd (= tvd) und' die Infimtive auf 
tosum (= fos) und kastin (— as) (P. 1. 1. 40), und endlich 5. die avyayi- 
bhttva (P. 1 1. 41). Zunachst fallen fdr unsere Grammatik naturgemass 
die vediscben Infinitive (P. 3. 4. 16, 17) auf tos und as (Kategorie 2) und 


DIE grammatik Sakatayana’s 

die ebenfalls vedischen Infinitiv-Dative auf Diphthonge wie fiva&e, pibadhym 
usw. (Kategorie 3) ,lweg. Von den libngen entgirechen der 1 Kategone bei 
Sakat. : svar ti und sunabha ; der 2. Kategpne . adhatftast, ptasvabka, 
tas (in pHiimulatah) , vat und am (in uccaiitaimm) ; der 3 ICategone ; am 
(in dayatiicakre, s. MBhas. Vol. 1, S. 96, Z. 19 f.), am (in purvambhojam) 
uiid turn ; der 4 Kategorie tva Dem SQtra P. 1 1. 41 schemen die Sutras 
5 1. 4, 6 zu entspiechen. Die in P. i. 1 35 erwahnten taddhitorSa&sR 

{iaddhitaS cdsarvavibhakuh) ausser vat und tas hat iSakat. an einer Stelle 

5 4 4-64) zusammengestellt, so dass er den pratyahara ptasu bilden 
kormte. Der Vers sadrsam usw stammt aus dem MBhas. zu P. i 1. 38. — 
Bei Candra wird avyaya lucht definiert 

40 gh% (heissen die Wortformen auf) t und u, mit ausnahme von sakkt 
und von patt, wenn dieses nidit (Ghed ernes) Dvandva ist 

P, 2. 4. 7 , a 5 2 50, 51. 

In emem Dvandva geht em gAt-Nonunalstamm voran (S. 2. 1 119) . 
daher patisutau und patisakhayau, (weil patt gki ist, dagegen suta bezw sakht 
nidit Der viprattsedha hat nut der Flexion des patt am Ende eines 
Kompositums oder, anders gesagt, m bezug auf die Frag^ ob eine 
Regel, die fur den Nominalstanun paii gilt, ebenfalls fiir einen auf patt 
auslautenden Nominalstamm gdten sollte, gar nidits zu tun. Daruber 
gibt die Panbhasa (31) Aufschluss * gtahanavatd prdtipadtkena tadantavi 
dhtr nastt, wdche nach der Kielhom’schen Ubersetzung (S. 160) heasst : That 
wliich cannot possibly be anythmg but a Prttipadika does (ccuitrary to P. 1. 
1 72) not denote that whidi ends with it, (but it denotes only iteelf), wobei 
allerdings die Gultigkeit der Panbhlaai. nidit nur fur dn Sutra, m dem em 
SufSx gdehrt wird, (wie dort ata evayatji pratyayavidhivisaya eva 'Phnbli 
Text S. 29, Z 16) sondem auch fur saut/M5-B'utras vorausgesetzt werden 

41 Ein operatives (Element heisst) Suffix {pratyaya), (wenn es sich) 
lUcht auf etwas (bezieht, was) on Genetiv (gelehrt wird) 

P-3 1.1, L 1. 49 ; H. J. 1 38 

Das im Genetiv gdehrte bezeichnet dasjenige, an dessen Stelle 
Etwas treten soli, also den sthanm, und das operative Element heisst in 
F^e pratyaya. Wenn em pratyaya emem Element angefugt werden soU, so 
wird das Element gewohnlich im Ablativ, aber memals im Genetiv, 
Mgen^ Unser Sutra £64} besagt also im Grande genotmmen dasselbe wie 
das FSffumsche Sutm : sa^lu sthaneyogS (P* i. i 49), 

f; J?" M 3 60) (bis aim Entte des Werks) 

nut Auaschluss der verbalcn Flesdonaendungen (heissen) krt 
P. 3. 1. 93; H d. 1. 1. ^ 



44 (Em Suffix \pratyaya] wird) nachgesetzt, 
fP. 5. 1 2 ; H 7 4 118. 

45. Was m zma stununen, Laut hat, folgt auf doi letzten Vokal (des- 
sen, dem das operative Element angefiigt wird). 

Pi 1. 47 ; C I. 1 14 

Die Kommentatoren eiklaren acah nut acam, da bei Namen von Klassen 
Oder Arten der Singular fur den Plural emtreten kann (jatav ekavacanam) . 
Cf Pralcriyias S 52 Anm 1 : mid aco 'ntydd tti sutre cco) ity ekavacanam 
fatav eva | ntrdharane sasthlyam. 

46 (Wain zwei Regeln, die glache Kraft haben und von daien jede 
auch noch anderswo eintreten konnte,) in Konflikt (geraten dadurch, dass 
sie sich m einem Falle beide darbielten, so tritt die in der Reihenfolge der 
Sutras spater gelehrte an^®) 

P I. 4 2; CJ. 1. 16; H 7 4.119. 

Zu den Baspielen im Kom. Nadi S i 1. 157 muss « an die Sidle des 
n (i= r) treten, das fur anlautendes s substituiert ist, wenn a, die Media 
Oder die Nasale folgen ; nach 158 aber musa luk (d i. Elision) fui das auf 
den Pronominalstamm sa folgende ri emtreten, werm em Konsonant folgt. 
Fiir e^ak hasati und ^ dhavati werdoi sidi bade R^dn darbieten ; daher 
der vipratifedha — spardha (m) als Substantiv ist (nach dem P. W.) sonst 
gar nidit bdegt 

47. (Wenn one Substitution) an SteUe ernes im (jenitiv (ausgedru<±- 
ten Elements gdehrt wird, so hasst dies, dass das Substitut an die SteUe) 
des letzten Lautes (desselben zu treten bat) . 

P. i 1 52, C i 1. 10; H. 7. 4 106 

Das Beispiel napo ’co hrasvahi^* ’’die Kiirze (tritt ein fiir. den Auslaut) 
eines auf einen Vokal (audautenden) Neutrums", illustnert die 2 wei Funk- 
tionen des Genetivs, die ausanander gdialten werden mussen Der erste 
(jenetiv napafy faUt unter unsere Regel ; dagegen ist do" zweite Genetiv acah 
ein Attnbut {vUe^cmtt) von nc^rah und fallt daher unter 1. Ij 55. Danach 
hidsst napo 'cah {65} nicht etwa „an die SteUe eines neutralen Substantivs, 
welches ein Vokal i&t“, sondern „an die SteUe eines auf einen Vokal auslauten- 
den neutralen Substantivs" 

48 (Worn one Substitution fur dwas gdehrt wird,) was auf em 
Anderes, welches im Ablativ steht, (folgt, so heisst dies, dass das Substitut 
an die SteUe) des ersten (Lautes des folgenden Elementa tritt). 

P. i. 1. 54;C.I. 1 9;H. 7. 4. 104. 

18 Fur Mtio klare der vipratisedhaparibhasa vgl KiMtHORN. Pari- 

bhaisenduiddiara (Trandation), S. 194 f. 

M g. i. 2. 1 = P. 1. 2. 47. 

56 DIE grammatik Sakatayana’s 

Weiin gdehrt wird, dass i an die Stelle von op tritt, wenn dieses auf dvt, 
antar usw. im Kompositum folgt (S 2 138), so tritt nach unserer' Regel 
das I racht an die Stelle des letzten Lautes (#>) der Gruppe (wie nach 1 1 
47 zu ei'warten), sondem an die Stelle des eisten Lautes (d). 

49 , (Em Substitut,) das a zum stummen Laut hat, (tmd ems, das aus 
mehr als emem) Laut (hesteht und) bein w zum stummen Laut hat, tntt an 
die Stelle des Ganzen 

P I. 1 53, 55 , C I 1 11, 12 

Dass der Sinn des Sutra dies sein muss, wird niemand bezweifeln Wie 
er aber aus den vier Silben des Sutra hdrauskniiUMfin soil, ist nur dunkel 
geblieben. dl (als Plur gedacht) steht schembar fur omkdl ' Man hatte 
femer erwartet, dass eine Andeutung der Tatsache, dass das Substitut an die 
Stelle eben des Ganzen und mcht ernes Teiles tntt, m dem Wortlaut des Sutra 
kaum entbehrt werden konnte Der Kom versagt vollstandig 

it ( = ») ist ein ekal und hat i zum stummen Laut , es tntt infolgedes- 
sen fur das Ganze, (})as und (i)as em , (,s)Sm ist an^l und hat femer 
kein n zum stummen Laut , es tntt daher ebenfalls fur das ganze dm ein 
(A) as 1st anekal, hat aber n zum stummen Laut ; es tntt infolgiedessen nur 
fur den letzten Laut in jard ein ; onto ist ebenfalls anekal; hat aber kein n 
zum stummen Laut, es wird daher fiir das ganze jha substituiert. 

50 (Das Substitut ist zu behandeln) wie das Ursprungliche, wenn (die 
betreffende Operation) mcht von den lautlichai (Bestandteilen des Urspriui- 
glichen) abhangig (ist)“ 

P. 1. 1 56 + Va. 4, H 7 4 109. 

Der Begnff airaya in diesem Zusammenhang stammt aius einem Varttika 
Katy@,yana’s (P. 1 1 56 Va 4) her saty diraye vtdhk istak , vgl. audi 
das Bhasya (Vol. 1, S. 133, Z 13 f) : tdam tarhi prayojanam uttarapada- 
lopo yathd vtjAdyeta | alam airayate 'Idiray/ah \ ddSdayOi vidhvr alvidhir itt, 
und dazu Kaiyata alairayo yasyeti j dairayo vidhtr alvtdhih [ oddsrayatvad 
vidhir evdl ity ucyata iti gatarthatvad aptayoga eva uttaratHfdasya lopah. 
Er gilt dann fur die Folgezeit als em unentbdirhcher Be- {663 -standteil rtips^r 
PanbhB^. So die KaSika (zu P i 1. 56) : sthamvad edeSo bhavatt sthdn- 
yairay^ kdryesu andc^ldyesu \ sthdnyal^ayani hdryani varjayttva ; Hem 
hat es m die R^[el mcht aufgenommen, aber im Kom verwendet : ddesah 
ddesiva syat | na cet stHSnivamdirayam kdryam (H. 7. 4. 109) 

51. (Das durdh das) Folgende (bedingte Substitut) fur Vokal 
(verhalt sich in bezug auf eine Regel, die einen ihm)( vorangehenden (Laut 
beUllIl, wie der ursprlmgh^^s Laut) ausser (in einer Regei, wa es sich handelt 
1. um die Substitution von Ehsion durch) km, 2. (um die Substitution der) 

In der CSndra-Giaininatik fdilen diiese und die folgenden PariMia^s 



Lange, 3. (um den Ausfall von) y, 4 (tun erne) Verdoppdung, und 5 (end- 
hch ausser in alien Regdn von hier an) bis (zum adhikara) asat (inklusivL 
ausgenommen (nur die Regel ijber die Substitution von) Elision fur s und k 
(namlidi 1. 2. 91). 

P 1. 1. 57, 58 + Via. ; H 7. 4. 110, 111. 

Das Wort asat bietet gewisse Sclrwienglaeiten 3sat heisst an und fQr 
sich ‘ von (dem adhikara) asat (inklusiv bez. eaddusiv) an belz. bis zu Es 
muss hier offenbar „bia zu asat" heissen Es fragt sich daim ab^ . von 
wo an? Das etasmat des Kom kann sich wohl nur auf eben diese sdbe 
Regel beziehen G^en diese Anffassung sprechen freilich die Beispide daddhy 
atia, maddhv atra des Kom.. Denn angenommen, dass die unter asat gedach- 
tcn Regeln von hier an bis zu I 2. 101 sich erslrecken und dass die von Kom. 
unter dvitva angegebenen Beispide die emzigen sind, vro Jinsere Regel gdten 
kann, so wnrde sich das im Sutra dem asat unmittelbar vorangehende Wort 
dvi auf erne Regel beziehen, die schon im Komplex asat eingeschlossen sein 
vinirde. Eis ist aber nicht ausgeschlossen, dass der Verfasser der Sutras audi 
andere Regeln ausser 1 1. 115-126 unter dem Ausdruck dvi nuteinschliessen 
v/iU, naralidi wo uberhaupt von Verdoppelung die Rede ist. Es sei aber 
erwahnt, dass die Beispiele des Kom unteir asat Regdn illustrieren, die sich 
mit 1. 1 51 — 1. 2. 101 decken, also im Einklang mit meiner Auffassung stehen 
—Die Verweise in den Fussnoten zum Text konnten nur soweit angegeben 
warden, als die zu Gebote stehenden unvoUstandigen Matenali^ es etmbg- 
hchen. Wegen kathayaU, avadhit vergleiche man MBhas. Bd 1, S. 146, 
Z. 1 — 3, wo Kielhorn die Verweise auf Pajjini’s Sutras angibt,— 2ii padikah : 
Nach !§ 5. 2 39 tntt an pSda das Suffix than ( = ika) Der Taddhita than 
bewirkt den Abfall des vorangdienden Vokals (vgl. P €. 4. 148). Vor 
einem vokalisch anlautenden Suffix sollte! fur pad pad substituiert werden 
(vgl. P. 6. 4. 130). Da aber die Substitution der Elision fur den Vokal a 
von pdda durch etwas folgendes bedingt ist, verhalt ach die Eliaon wie der 
uisprungliche Vokal a in bezug auf die Regel, die die Substitubon von pad 
fiir das dem sthanin vorangehende pad lehrt, — also ist die Wartfomi *» bezug 
auf die Reg^ gar nicht als pad, sondem als pSda zu betraditen — und 

verhindert die Substitution Dies geht aus atra padbhave des Kom 
hervor. — ^Zu la/oam dcadfe lauh ‘ An lava tnt (»)i, {k)v{i)\ und s(»). Das 
samprasarana kann nur vor konsonantisch anlautenden Suffixen eontreten. 
Der lopa des Vokals zwischen lava und kvi verhalt ach nicht wie der sthanin. 
~Zu sukusmayateh usw • sukuh i:St ein Denommativ von der Wurzel kusma 
mit dem Praverbium su (v^. Siddhiantak S 402) — In kdsthatat steckt ein 
Verbalnomen vcai dem Kausativum der Wz, in kdsthatak dag^en vom Sim- 
plex — pfayiko ’yam nisedkah — da das Verbot zwar die masten, aber nicht 
alle Falle betriflft — madhuk ist wahrschemlich eine Taddhita-Adedtung von 
madhidcut, wrfjei das ut abfallen muss. (Vgl. MBbSs zxx 8, 3. 17 und Kaiyata 
ad loc.) — ^adika usw. ist mir unklar geWieben. 



52. Wenn (em durch) i/«c (schwindet, so tritt die Opesration an 
dem vorangehenden Element, die dureh das geschwvmdene SufBx bedmgt ist, 
nur in Bezug auf) t, « respectiv r (fdr y, v respectiv r und m) enad (acc. 
sing neut.) (fur etad ein, aonst ajier nicht) 

P i 1 . 63, 2 4 34 Va. 1 , H 7.4. 112. 

Dass das ik-karya dassdbe wie das samprasardna ist, geht aus dem fol- 
jpntlpn hervor ihtcigenad iti myamdt parasya sluci satydm yana tk etadah 
enad iti dvayatn eva bhavatiti usw (Prakriyas. S 253, Anm. 2). Demnacb 
lUustrieren die Beispide veveddht, SosainU, jangrhtti die Substitution von, t, 
u respektiv r (in den Stammen des Frequentativums vevtdh, Sokt- respektiv 
jaiigTk) fur y, v und r (m vyadk, ivt und grab), trotzdem dass fur den 
Intensivdiarakter ya slue dngetreten ist— Die Ausnahme fur enad beruht auf 
Varttika (P. 2. 4. 34 Va. 1) enad iti napumsokavacane Zunachst 
ergibt sich im acc. smg neut etadam Nach dem Abfall des am durch slue 
(■§ 1 2. 5 = P 7. 1 23 svamor nopumsakdt, zu ergknzen luk)\ konnte das 
fur etad und tdam gdehrte ena (,S 1. 2 203 = P 2. 4 34) gar nicht 
emtreten, well nut dem Schwund duidi slue auch das durdi das Suffix in Bezug 
auf den Stamm bedingte aufgehoben wird, gerade wie bed tad das im nom 
sing, zu substituierende sw (P 7 2 106) nur in mask und fern , aber nidit 
im neut zur Erscheinung kommti® 

53. Was t zam stummen Laut hat, (wird) an den Anfang (angeffigt) 
P. 1 1. 46;C 1. 1. 13. 

54. Was k mm stummen Laut hat, (wird) an das Ende (angefugt). 

P. 1. 1. 46 ; C. I 1 13. 

{[68} Hema hat die stummen Laute t und k aufgegeben. Z B. dem 
SakalSyanal’schfin Sutra nah M jak (iS. 1. 1. 147 = P 8. 3 31) entspncht 
bei Hema (J. 3. 19|) nah H tie (Comm : padantastheesya nasya he p(ffe He 
va sySt I bhavShc surah usfw ), und dnah hah tso ’kah bei diesem (H. 1 3. 
18) ent^ncht dnos tat so ’ Scah (S I. 1 146) — Zur Trennung det Sutras 
53 imd 54 s Kom. 

55 (Was als) nShere Bestimmung (dnes zu spezifizierenden dient, 
bezeichnet den Auslaut des Aggregats) . 

P. 1. 1. 72 ; H 7. 4. 13. 

Das unserem Sutra entsprechende SQtra bei Knini {1 1. 72) ist ohne 
die Viarttikas 3 und 4 KSt^yana’a geradezu' unmid^ich Die Varttikas ver- 
langen den Ausschlus der Falle, wo es sich um Komposita und Suffix handelt 
mit Ausnahme derjenigen Falle, wo dn iigif-SuffiK erwdhnt wird Oder wo bei 
der Formulierung des Sutra nur Budistaben gebraudit werden, Indem 'SSkat 

^ddhSntak, (S 99) • atwSdehe napumsake (sic) enad vaktavayah , Tatt- 
Vabodhirfi dazu (djcnda) : amy evedarn tndhiyate na tv authasSdi^u phedabhavat | 
svamor Ttapurpsal^ IP 7. 1, 23] ity amo luka luptatve ’pi pratyayaiakiaefam ika 
pravartate vacanasamarthyod ity 3kulf< 

2 TEIL 59 

den Wortlaut des Sutra in der Wase andert, dass er das den Auslaut bildende 
Element tind das, dessen Auslaut dieses bildet, als rni Verhaltnia von viscjapa 
zu vtsesya stehend fasst, glaubt er die obenei*wahntai Varttikas K&tyiayana’s 
entbehren zu koimen. Wie wat es ihm dies gelungen ist, ist freibch sehr 
fraglich Denu m einem Sutra wie hitadibhib (S 2. 1. 33 = P 2 1. 24) 
[Komm. : dvttlyantcafi subantarfi &ntddibhih subant<dh samasyate'] mochte 
rnan fragen, warum hier §xita agentlich kem. viie^ana von (dem vise^yasa- 
mudaya) sup (zu erganzen aus dem ad&iftara-Sfitra) ist und warum es als 
solches nicht zuglach den Ausgang ernes Komplexes wie paramasnta bezeich- 
nen kann Dasselbe gilt von dem im MBhia$ (Vol. 1, S. 183, Z. 20) erwahn- 
ten Falle nadadibhyah phak (P. 4. 1. 99), dem bei iSakat. (2 4 32) vada- 
dtbhyah. phm entspncht. Man beachte, dass diese Frage nidit durcb S. 1. 
1. 59 beruhrt wird Denn dort handelt es sich lediglich darum, was die m 
einem Sutra angegdienen Suflfee (wie z. B sw« padam iS. 1 1. 62 und saslhy 
ayatndt & 2. 1. 34) und mcht prMipadtka oder Teile von denselben 


56 (Ehe im) Ablativ (angegebene Bestimmung muss unmittelfaar) vor 
(dem Spezifizierten stdien) . 

P 2. 1 67;C. 2. 1 8 ; H. 7 4. 104. 

57. Nicht (so darf die im) Ldcahv (angegebene Bestimmung vor dem 
Spezifizierten stehen, sondem sie steht hinter demselben,) au^senommen in 
(den R^dn) ghya (d h. ghyai} [4. 3. 60]) usw. 

' P. 2 1. 66; C. 2. 1. 7; H 7 4 105. 

agkyadisu Die Beschrankung bezieht sich auf die m 4 3. 60 ff. gelehrten 
Suffixte. So z. B. m yamah samnivyupe (4. 4 10) heisst scofinivyupe nicht 
V 01 den Prapositionai sam usw , sondem im Gegenteil nach denselben. Also 
heisst das Sutra an yam hinter sam, ni, vt und upa (treten gewisse SufSxe) ; 
vgl Prakriyds {69} S 392, Nr 1965. — Zu smarast usw. Dies ist das 
bekannte Beispiel fur die R^el • In Verbindung mit einem Verbum in der 
Bedeutung „sich erirmem" steht das erste Futurum (It) um die hinter dem 
Heute gelegene Vergangenheit zu bezdchnen; s. K&'Ska zu P. 3. 2. 112. 
Beide Handschriften B und H (P fallt naturhch aus) lesen vartsyamah, was 
ohne weiteres zu verbessem ist. 

58. (Die ndhere Bestimmung des im Lokativ gelehrten Spezifizierten 
bezeichnet das) Anfangs( -element) dessen (d. h des Spezifizierten) . 

P. 2. 1. 72 VE 29; H. 7. 4. 114. 

59. (Als Attribute bezeidinen) em Sufiix und syat (d. i. die Feminina 
bildendai Sufike, wenn die auf sie ausgehenden Wortformien dem ganzcn 
Komipositum) untergeordnet (sind, denjenigen Lautkomplex, wdcher) mit 
der prakrti anfkngt (d. i. damit anfangt, woran das Suffix angefiigt wird), 

P. 2. 4. 13 Va. 7 4- 6 1. 13 Bh. ; H.' 7. 4. 115, 116. 



Wegen matrbhoginak vgl Va 1 zu P S 4. 11 und wegen nyaksyat vgl. 
die Paribh. f26) stylprc^yaye cmupasarjane na MDAyUBA^Yol. 3, S 20, 
Z 15 — ^23. — Die Verwandlung von « in « nach r und s findet dann start 
wenn die betreffenden Laute in einem und demselben Pada stehen (P 5. 4 
1). Nun heisst ein Pada das, vras auf die Kasusendungen oder die Personal- 
endungen ausgeht (S. 1. 1. 62 = P. I. 4. 14). Nach unserer Paribh. 
heisst m stm padam (is 2 1. 62) Pada das, dem das SuiBx angefugt wird 
nebst dm Suffix In matrbhoglmh wird s (nom sing ) {mdtr + bhoga) 

ina" angefugt, also heisst der game Komplex [ {matr + bhoga) + + s 



60 (Em m emem Sutra aufgefiihrtes) fef (-Suffix bezeichnet sowohl die 
Wortform, der das betreffende Suffix angefugt wird, als) auch diese nebsl 
den U (genannten Worten) und den m emer K^usverbmdung stehenden 

P I 4. 13 va. 9 , H 7 4 117 

Dieses Sutra ist identisch nut der tl^ribh. (28) krdgrahane gatikara- 
kapurvasyapi graha^am — *bhasm\cmhuta (adj) = in die Asche geopfert, s. 
V. a. ein unnutzes Werk vollbracht (Bbht. Worterb ). 

61, (Eine nahere Bestimmung des verbum finitum eei es, dass sie 
ausgedruckt oder nur implicite verstanden ist) nebst (dem ausgedmckten 
Oder implicite veretandenen) verbum fimtum (heisst)' Satz {vSkya). 

P.2.1 1 Va 9, 10. H. 1. 1. 26. 

Die Bezffldinung vdkya wird am Gebrauch der enkUtischen £70|} Prooo- 
mina (was, nos, td und me § 1.2. 191, 193) und der Plutienmg des Auslautes 
(S. 2 3 17, 27) illusitriert — Obwohl die Lesart von H sehr mangelhaf t 
ist, lasst das ca hinter gramam die riditige Lesart nut Sicherheit erkennen. 
Die Bei^ide ffir Plutienmg beruhen auf Konjektur, doch macht ein Ver^eich 
nut der Kasika zu P S. 2. 104 die Lesung ganz sicher Indem Hema (Z 
1. 26) das vakya als saotsesanam dkhydtam („ ein verbum finitnm ndist den 
dieses naher bestimmenden Worten") defimert, schhesst er sich naher sowoW 
an den Smn als den Wcutlaut des BhS§ya an (Bd. 1, S 367, Z. 15) : Qpara 
Ska akhyatarti saviSe^ai^m tty emt | sarvani hy etam [sal., avyaya, hdraka 
und karaka und karakaviSesai^a] kriydvt§e$anant. 

62. (Was auf) erne Kasus- oder Persomalendung (ausgeht, heisst) Wort 
{pada) . 

P. 1. 4. 14; H. Z 1. 2(0. 

Nach der Paribhas& (23) . pratyayagrahane yasmat sa vtkttas tadddes 
tadantasya grahatfatn kann suptm nicht die Suffixe sup und tin, sondem muss 
eine Wortforna, die auf dte Suffixe sup und tin ausgeht, bezeichnen. Auf 

w ina ist ein samasantaSxs&x. Keser Fall ist von magavapi^ (von mSsavS- 
put) zu unterschaden, wa das Suffix .niw an krt ist Dies fellt unter die Panbh. 
175) ‘ gfitikarahepapadSnarp krdbkth sake santasavacanatfi prik subutpatteh. 

2. TEIL 


onsere Regd findet aber die PanbMa (27 = P. I. 4. 14 Va 1), Anvren- 
dung • sarnindvidhau pratyayagrahane tadantagrahnam nasU, wdche nach der 
Kidhom’achen Ubersetzung heisst An affix, 'whei employed in a rule, which 
teaches the meaning of a technical term does not denote a word-form ending 
with the affix, i^kat. hat die Paribhasa tatsachlich aufgenonunen (siehe das 
Buhler’sche MSS unter den PanbhBsas) Ei noacht sich hier also emer 
Ungenauigkdt schuldig. Das Sutra Hema’s (i. 1. 20) tadantam padam 
wird der Panbhla§a gerecht 

63. (Erne auf) « (auslautende Wortfomi® hdsst) vor (den verschie- 
denen Denominativsuffixen) kya ( = ya) Wort {pada) . 

P. i. 4 15 ; H J. 1 22 

In samanya fallt das n nicht aus, well das darauflolgende Sufiix nicht 
kya, sondem sy^in (cf. P 5 1 124) heisst. — Zu manyd : Nach der Regd 
samajani^adanipad manvK^ghamtnah (iS 4 4. 70 = P. 3. 3. 90) tritt 
das Suffix kyap an mm Doch fallt das n des Stammaudautes davor nicht 
ab, well der Stamm nicht auf sun ausgeht 

64. (Vor emem Suffix), wdches ein stummes s hat und vor einem mit 
bdid>igem Konsonanten ausser y (anlautenden Suffix heisst das Voraus- 
gehende padia, jedoch) nicht (wenn es erne) Wurzd (ist). 

P 1. 4. 16. 18 ; H. i. 1. 21. 

In bhavaddya hasst das SuflSx chas, wobei cha = iya^’',. in urnayu yus 
(6. 3. 1. 24; 3. 148) — Zu yajvd, vacmi-. Am Ende fTlJ eines pada 
wird fur den Endkonsonanten von yaj s subsutuiert (P 3 2. 36) , in vacna 
wurde, wenn vac pada ware, natiirhch « fdr c emtreben. Dem, adhdtok iSakat's 
entspncht ndma bei Hema. : ndma stdavyoHjane (H. 1. 1. 21). ^ 

65 Dem Ausgang (d i. dem letzten Glied) eines Aggregats von Wort- 
foimen, das zur Bezeichnung ernes G^enstandes verwendet wird, (kommt 
die Bezdduiung pada) nicht (zu). 

P. I. 1. 63 Vffl 6; H 1. 1. 25. 

In der Wiedergabe des vrtti des Sutra bin ich dem Kora, gefoigt Viel 
emfadier ist aber die Erklarung des Wortes vfttyanta bea Hema. (i. 1. 25), 
wo unser Sutra mit emer kleinen Abweichung, die spater zur Sprache kommen 
wird, wifider ersdieint. Dort heisst es : paddrthdbhidhayi samasddir vrttts 
tasyq antoJf usw. Danach heisst vjttt etwas, was emeu Gegenstand bezekhnet, 
wie z B ein Kompositum®*., Das Sutra ist erforderlich, um fur die Wort- 
formen gir, dtv usw. in paramaprau, paramadivm usw. die Bezeichnung 

IS Es handelt sich hier aber nur um Nominalstaimne; da die Suffixe kya nur 
soldien angefugt werden 

w Cf. P. 7. 1. 2. 

so Die Bedeutung ist in dem Wortwbuch Apte’s angegeben (ohne Bdeg), 
fehlt aber in P, W, 


pada aufzuhebien, die ihnen nacb S. 1. 1. 50, 52? (—P. 1. 62, 63) 

eukommt. da sonst m den angefuhrten Beispielen der Reihe nach visaiianlya 
fur r, fakultative Elision des t; (der sogenannte sakalyaprat^edha) . d fur h, 
g fur h und e ndlirh Elision des « eintreten warden AUe diese Lautverande- 
rungen werden fur das Ende des poda gelehrt Es bleibt zu erlddren, wie gtr 
usw in paramagiTOU usw die Bezeichnung pada zukomint Em Kompoeitum 
ist in seiner aufgelosten Form nut ICasusendungen versehen Bei der Eompo- 
fition aber wird fur diese luk substituiert Eine solche Wortfoim wird dann 
zu einem prattpadika und es konnen ihr neue Endungen angefugt werden. 
Eine flektierte Form von rajapurusa kommt auf folgende Weise zustande : 
rajnah puTusa + (Endung) = Irdjan + (luk) + purusa + (luk)} + (neue 
Endung) = rajapurusa + (Endung) Die Endung tritt stets an das ganze 
prStipadika rajapurusa und nicht an purusa allem an Wenn z. B . Kaiyata 
den Fall bespricht, ob in paramavaca, da ein vokalisch anlautendes Suffix 
folgt, das vorausgdbende vac em bha (also kein pada), heissen konnte, sagt 
er ausdniddich, dass die Endung d an den Komplex paramavac und nicht 
an den Teil vac tntt : paramavacett | samasartha ya vibhaktth kjta tarn 
Isupo dhatv (P. 2. 4. 71) itt hiptam, Nagogibhatta,] pratyayalaksanmdirUya 
padatvambandhandni kutvadim prdpmtvantt | bhasamjM tu yastnad yajadu 
vidhvT tti samuddyasyatva na tv amyavasya (Bbdsyapradlpa zu P. 1. 1. 
63). Demnach steht in unserem Bei^id die Sadie so . parama + (luk) + 
gtr + (luk) + au. Die Substitution des luk hmter gir wiirde die Bezdchnung 
pada fur gir ehensowenig aufheben wie fur rdjm in rdjapurufcp^ denn 
[72} nut der Substitution des luk werden nur solche Operationen aufgehoben, 
die berm Vorhandaisein des Suffixes in Bezug auf das aiiga vorgenommen 
werden miissten — Dieses Sutra beruht auf einem VSrttika Katyayana’s 
(P. i 1 63 Va 6) Idi muss ^tehen, dass mir das BhBgya dazu nicht in 
seinem ganzen Umfang klar geworden isL Das Ergebnis sdieint mir abcr 
vom Verfasser der Siddhantak (S 99) folgendermassen kurz zusammengefast 
zu sein : antarvartinlm vibhakttm dSritya piir&apadasyevoitarakhandasydpi 
padasamjnay&tfl prdptaydm ] uttarapadatve cdpadadividhau prathadhah 
IVa. 6 zu F i. 1. 63] | th pratyayalaksmam na \ apadddwidhau Um | 
dadhisecou | iha satvanisedke kartavye padatvam asty eva | kutvd tu na?^ 
Dem Zusatz iha satvanisedke usw , wddier aus dem MBhSs ( Vol 1, S. 166, 
Z.‘ 12 — 14) stammt, wird das Sutra Hema’s (1. 1. 25) vrityanto ’saje 

[Komm. . asase sasya satve padam eva] gerecht, was zugleich zeigt, dass 


21 Als pada wirft rajan das audautende n ab. 

22 In dem Kojnmentar dazu heiaart: es pufvapadasyevett \ anyathd tajapuru^o 
vas^r ftyadau nalapakutvadfkam yathasainbhavaffi na sydd iti bhdvah || uttarC’^ 
padatve ceti || uitcerapadena uttarapadam ucyate | uttarapadasya padaive pada- 
vyapadeie kartavye pratyayalak^anaTfi na bhavatity artkah j etena sudhiyau sudhtya 
tty otrantarvartisupd padatvat pak^e ^akcdaprasanga rty ^4nk3 pardstd \ uktanlyd 

2. TEIL 63 

Hema gd^iaitlich uber die Aagaben bei iSakat hinausgegangen ist, imd dass 
er sdbstandig aus alteren Qudlen gesdidpft hat. 

66 Emer auf t oder s (auslautenden Wortfonn kommt vor pmpn i Suffix) 
m der Bedeutung von mat (die Bezeichnung poda nicht zu) 

P. 1. 4. 19 , H 1. 1. 23. 

67 mams, nabhas und angvras (kommt) vor (dem Snffixe) uat (die 
Bezeichnung pada nicht zu) . 

VI 4 18 Via 3 ; H. J 1 24. 

iSakat hat hierimt die Paiuni’sche Kategone bha (P. J. 4. 18-20) ab- 
geschafft, worm ihm Hema. gefolgt ist (H 1. 1. 23, 24). 

68 a i und P, konnen in der Pause nasaliert warden, (jedoch) nirht 

die {a, S). welche nut den stummen Buchstaben g versdien sind, sowie 

(der Auslaut von) ca usw., ausser d 

P S 4. 57, C. 6. 4. 150, H. 1. 2. 41 

Die Beschrankung anancddt ist weder bei Pan.,, noch im Bh§sya, noch 
bei Candra zu finden, wohl aber bei Hema. 

69 (In der Pause kann) die tonlose mdit-a^inerte Muta fur die 
tonende (substitmert werden) 

P. S. 4. 56 , C 6 4 149 , H. 1. 3. 51. 

70. (Was bis zum Schluss des pdda gdehrt werden wird, ist) nicht (in 
der Pause vorzunehmen). 

P. 5 1 72, H. 1. 3. 52 

. t73j Zwischen te und dkuh bezw. bhavan und lumtt ist eme Pause hinzu- 

71. Fur e, o, en und au (werden) vor anem Vokal (der Reihe nach) 
ay, av, ay und, an (substitmert). 

P. 6. 1. 78 ; C. 5. 1. 75 ; H. 1. 2 . 23, 24. 

72. Fiir die auf a folgenden i und u (konnen) nur y respektiv v (sub- 
stituiert werden, wenn die ersteren fur e respiektiv o dngetreten and). 

P. S. 2. 1D8 . C. 6. 3.133 ; H. 7. 4 103. 

piirve ’pouddd anantaron usw =Panbh. 59.— Ohne dieses eva konnte 
Verlangerung vor homogenen VokaJen emtreten, wed der apavada daijenigm 
niyama aufhdit, der unmittelbar auf jenen folgt. 

73 (Fur die Vokale I, P. | (f) werden) vor einem ihnen nicht homo- 
genen (Vokal der Kedhe nach y, v, r und I substitmert) . 

P. 6 1. 77; C. 5. 1 74; H. 7. 2. 21. 

Der Komm. erwahnt, dass das Sutra auch auf eme andere Weise erldart 
wird, wonach sich die Formen dadhiy atra, rmdhuv atra usw ergehen-s^ 


as Zur Verwendung vou iy, uv statt y, v, vgl. Wackernagel, Altmd. Gramm. 

1 . § lai. 



74 (Pur dw Vokale !, & und f (2) kann (vor emen nidilj homogenen 
Vokal) die Kurze (substituiert werden, weim die aufeinander folgenden 
Vokale) nicht in emem (und detnselben) pada stehen 

P 6. 1. 127 + Va 1 +Bh ; C. 5. 1. 132; H. 1. 2. 22. 

Diese R^el iSakat’s, ebensowie die entsprechendej Regd Hema.'8 (2 2. 
22;, scheinen zu wait zu sein, denn nach dem MBhiai$ gilt der sakalyapratt- 
sidha nur ficr den Fall, dass der zweite Vokal den Anlaut eines mit den 
stummen Buchstaben s versehenen SulBxes bildet, also eines stf-SulSxes, vor 
welchem das Vorangehende pada heisst (P 2 1 16 , iSa 2. 1. 64). — parjan- 
usw. = Paribh 111. 

75. Vor r (2) kann fur jeden einfachen Vokal (die Kurze substituiert 
werden) . 

P. 6 1. 128, C. 5 1. 133, H 2. 2 2. 

76. Fur r Q) ndist dem (folgenden) Vokal (wird, wenn r (1) folgt, 
fr (allem substituiert, wo fr folgendes bezeachnet : ein Aggregat von VoKalen, 
wie z. B rr, /2 usw, oder em Aggregat von Vokal und Konsonant, wie z. 
B. jT, ri usw , Oder aber em ganz anderer Laut, wobei die Organe sich nur 
leicht beruhren) 

Vgl. P 5 1. 101 Va 1, 2 und H. I. 2. 3, 4. 

Die obige Ubersetzung ist nur provisonsch gegeben. Aus dem Wortlaut 
des Sutra sdbst ist nidits zu entnehmen Man ist mfolgedessen voUstandig 
auf den Kom. angewiesen Aus dem samuccaya m dem Kom. wird man 
mcht king Wenn der Kom. ^74} nur sagen will, dass fiir r + y rr bez. rr 
subsbtuiert werden, ist der Ausdruck samuccaya redit unglucklich gewahlt 
Zwei Vokale machen doch kdnen samuccaya acau bezw. ajhalau batte geniigt 
und ware klarer gewesen Die Beispiele smd hier wemg von Nutzen, well 
m solchen FaUen die richtige Lesart erst aus der Aussage des Sutra bez, des 
Kom. erschloesai werden kann. Die Deutung der Varttifcas zu P. 1 101 
savamadirghatva rti fvavacanam und Iti Ivdvacanam ist nur mcht gelungen. 
Die entsprechendKi Sutras Hema’s 2 2 3, 4 smd ebenfalls wemg klar. 
Wegen vomdniara usw. veiweise idi auf die Siddlfintak (p 21 unter dem 
Sutra akah savame dirghah) rti rva Iti Ivety ubhayatrapi vidheyam vattyi- 
dvayam dvmdtrOm \ ddyasya madhye dvau rephau taym eka mStrd [ abhito 
jbhakter apard j dvttiyasya tu madhye dvau lakarau | sesotfi pragvat Der 
Laut besteht danach aus zwei r, deren Lange erne matra betrdgt, und aus 
zwei die t umgebenden Vokalen, deren Gesamflange auch eme moDra betragt 
Der Laut ist also doch zwei matrcs lang 

77. (Wenn auf emen einfadien Vokal dn zweiter Vokal folgt, wird 
fur bade die entsprediende) Lange (aUem substituiat), 

P. 6. 1. 1101 ; C. 5. 1. 106 , H. 2. 2. 1. 

2. tElL 


In der Tat aber wild die Lange nur dann Q,jhe+ih.,»^ 

j , , suDstituiert, weim ein homo- 

gener Vokal folgt ; derm die apavadas I. 1 73 sp oq . 

, _ ^ ^ Sind imt diesem sutra 

zusammenzulesen. — Wegen anukara^a vgl. Anm zu Sutra 6 

, ^ w- ‘ >"» « a® piu ) 

kl!h (wird fur beide die dem voiansuhenda, Vokaf eulspiedimdu Ura® 
allem substituiert) . ® 

P. 6. 1 102 , C. 5 1. 109 

79. Im Maskulmum (wird fur emfachen Vokal vor dem ai des acc plu 
aber erne Ldnge) nut darauf folgendem n (substituiert) 

P 5. 1 102 , C 5. 1 . 109. 

munin ist aus ‘munlns entstanden , das s ist nach $ 1. 2 . 92 abgefallen 
welches den Abfall des letzten Konsonanten emer am Ende ein^ pada sb>hen- 
den Koinsonantengruppe lehrt. n am Ende dnes pada sollte nach 1. 2. 95 
abfallen, fallt aber doch mcht ab wegen 1 2 49 . Vgl. Prakriyias S 25 
Anm 2 und S. 59 unter Sutra 227. 

80. Wenn 4h und r ausfallen, wird fur (em vorangdiendes) j und « 
(eine Lange substituiert)®*. 

P. 6. 3. Ill ; C 5. 2. 137 ; H 7.3. 41, 42. 

C753 81 In sah und vah wird (beim Ausfall des dk oder r), 0 foiC a 

- P 6.Z. 112 ; C. 5. 2. 138 ; H J. 3 43. 

82 Wenn ein einfacher V(*al (auf a oder d) folgt, (wird fur bade) 
e respekt. 0 , ar (allem substituiert). 

P. 6. 1. 87, J. 1. 51 ; C. 5. 1. 82 , H. 1. 2. 6. 

83 Wenn an Diphthong oder (das fUr d subsbtuierte) u (auf a oder 
d folgt, wird flir bade) at respekt au (allem substituiert) 

P. 6. 1 . 88. 89 , C. 5. 1 814. 86; H. 1 . 2 . 12, 13 

Das uc entspncht Pan.'s Hth. Der Auslaut hat in beiden Fallen sonst 
kane Bedeutung — dhouta aus dkav + ik)ta. 

84 Fhx (das a von) pro und (den Anlaut von)i iidha, MU, aha, esa 
und e$ya (wird ai respeM. au alldn substituiert). 

P. 6. 1. 89 Va. 4; C. 5. 1. 89; H. J. 2 . 14, 

varm wird hier und im Kom. zu 91 — 95 als Neutrum gdwauchL In 

**< Etes Smra PSnini’s lautet: dhralcpe Purvasya dirgho ’nah. Das Wort 
p&Tvasya ist achdnbar iiberflussig ; denn nadi P. i. i gg Veranderung dien 

nur in bezug auf den vorangehenden Vokal vargenommen werden , es wird aber 
dadurch notwendig gemacht, dass das Wort uttarapade vksi P 6 3 1 noch in diesem 
Sutra fortgilt Infdgedessen wurde. ohne in u^em Sutra, die 

VerlangMting nur dann erfolgen, ein d* oder r m eineta uttarapada elidiert'-vmd 
(vgl MSiSg. zu P. 6. 3. 111). Diese Eiwagungen kommen fur da^ Sutra Sakata- 
yana's ebenso wie fur das entsjHechaide Sutra Candia’s (C. 5. 2 137) nir-jif m 
Betradit. ’ 



DlE grammAtik §akatayana‘s 

seinem LinganuSasana®® aetzt ^kait. vanja zuerst neut (Vers 8) und dann 
mask, und neut. (Vers 53) an. 

85. In sveara, svatrm und aksauhim (tntt ot respekt ou fur a + * 
respekt. u ein). 

P 6. 1. 89 3, 5 . C 5. 1 87, 88 , H. i. 2 15. 

68 Fur (a oder a + o von) (m und (ur das Substitut fiir die Prapo- 
sition) a (rmt dem darauf folgenden Vokal wird) der zweite Vokal (allcm 
substituiert) . 

P 6 1. 95 ; C 5. 1 99 ; H. i 2 18 

Das Beispiel adyarSydt findet sich auch im MBh&s. zu P 6. 1 . 95, wo 
itbrigens adyarsyat, mit langem a, ausdnicldich zuriickgewiesen wird 

87 Fur a + e von eva (wird der zweite Vokal allein substituiert, wenn 
das Wort) mcht zum Ausdruck der Beschrankung auf etwaa Bestunmtes 
(gebraudit wird). 

P. ^ 1 94 Va. 3 , C. 5. 1. 96; H. 1. 2. 16. 

88. Im Kompositum kaim (fur w oder 5 + o von) cstha oder otu (der 
zweite Vokal allein substituiert werden) 

F.6 1 94 Va. 5 , C. d. 1. 97 ; H. I. 2. 17. 

89. (Fur das auslautaide a ernes) auf das Instrumental (-sufiSia auslaut- 
enden Wortes+r) von rta (wird im Kompositum) dr (allem aibstitiiiert). 

P. 6 1. 89 Va. C. 5. 1. 90, H 1. 2. 8. 

{76} Das MBha§ kennt den vom Kom erlaubten Hiatus m sukhajUi, 
pTOtifa (I. 1. 90), uparfobhiyati (1. 1. 92) usw., (iwelche veruuttelst ernes 
jfittpaka zustande kommen) freilidi nidit ; weil es andereraeits kaum anzu- 
nehmen ist, dass der Kom ihn erfunden hat, mussen wir ihn wohl auf den 
Verfasser der SQtras zuruckfuhren. 

90 Pixr (das auslautende a in) pra, daSa, rna, vasana, kambala und 
vaisatara + (r von) rna (wurd im Kompositum dr allein substituiert). 

P. 6 1. 89 Va. 7, 8 ; C. 5. 1. 91 , H i. 2. 7. 

Die Hss. der C&idragrammatik lesen dasarm in dem entsprechenden 
Sutra (C. 5. 1, 91). Hema hat da5dma (H. 1. 2. 7). — Wegen praj'tia 
usw. siehe Anmerfciing zum vorangdienden Sutra— Candm hat vatsara fur 
vatsatara; Hema. gibt naturlidi beides an. — Die SiddhantaL (S 19) lost 
ptdrtfa so auf : ynasydpanayanaya yad anyad rnorfi knyate tad wdrijran ( eine 
zum Abtragen einer anderen Schuld gemadite Sdiuld). 

91. (Fiir das a) einer Prdposation (nebst dem folgenden Vokal wird), 

' wenn r (ernes Verbums darauf folgt,) dr (allein substituiert). 

P. 6 1 91; C 5. 1. 93, H. 1. 2. 9 

Hisgu. vrai B. Otto Franke in den „Indisdien Genuslehren nut dem Teirt 
usw." Kid 1890, 

2. TEIL 67 

92 (Die Substitution von ar fixr a einer Prapoation + r ist) mclit 
notv?endig, wain (das mit r anlautende Verbum) eui Denommativ (ist). 

P 6. 1 92, C. 5 1 94 ;H 1.2. 10, 11 

Wegen uparsabhlyati s. Anmerkung zu 89. 

93 (FUr das a cmer Praposition+l) e oder o (eixies Verbums) aussett 
t („gehai“) und edh („wachsen“) wird e respekt o (substituiert , doch ist 
bei Denonunativen die Substitution nicht notwendig). 

Rtf 1. 94 + Kai ; C. 5 1. 95 , H. J 2. 19, 20. 

Die Freistellung bei den Denonunativen finden wir zueist in' der KaS 
(zu P tf. 1. 94) erwahnt (vgl. kecid va supy apiialer usw.), wenn nicht der 
Verf. unter k^nt Candra im Auge hat. Die Stellung des va un uttmtiMbar 
vorangdienden SOtra (C. 5 1. 94) spridit dafur ; doch lasst sich das bei der 
Abwesenheit ernes Kommentars nicht mit Bestiinmtheit behaupten. Hema. 
(1 2. 20) ebenso wie Vopadeva (2 4) steUt die pararupa-Andeniag frei — 
Von den Varttikas hat Sakat. aufgenommea Va. 3 (iS = 1. 1. 87), Va. 5 = 
(5. 2. 1 88). VB. 6 fallt aus, da es sich auf vedische Formen bezieht. Va. 

2 hebt Va. 1 auf. Via. 4 lasst ert fort. In diesem Punkte weicht Sakat. von 
Candra ab, der das Va in seme Sutras aufgeinomiroen hat (C 5. 1. 98). 
Sakat, ebenso wie Hema , der ihm in der Ablehnung folgt, hat offaibar diese 
Worte als selbstandige Worte betrachtet, die kemer weiteren Zerlegung 


94. Fur das am Ende ernes pada stehende e und o mat dem darauf 
folgenden a (wird e respekt. o aUein substituiert) 

P. tf. 1. 109. ; C. tf. 1. 115 ; H. 1. 2. 27. 

{77} 95. Pur (das am Ende des Wortes stehende o von) go kann o 
(substituiert werden, wenn darauf a folgt) 

P. tf. 1. 122 ; C tf. 1 120, H. i. 2 31. 

laksanapratipadokta^ = Paribh. 105 

96. (Fiir das o von go kann) vor emem Vokal ava (substituiert werden, 
jedoch) nicht, wenn aksa folgt 

P tf 1 123; C 5 1 121; H. J 2 29 

Nach den ErklSrem Pinini’s — und iibrigena auch nach Candra — sand 
goaksa und go’k§a unstatthaft. Vgl. das Sutra Candra's (5. 1. 122) aksendre, 
wonach die Substitution <wa fiir o in go notwendig ist, wenn ok^a folgt, ebenso 
wie wain indra folgt KBd (zu P tf 1. 123) ': vyavastUtavibhSseyani tern 
gavdksa ity atra mtyam avan bhavati Vgl wdter SiddhSntak 22 ; Mugdha- 
bodha 2. 15 ; Eatantra 19. 43. 

ffy, ^ Vor indr a (muss ava fur. das o von go substituiert waden) 
p' tf. 1 124 ;C 5 1 122;H. 1 2 30. 

9S. Vor aksa (muss ava fur das o von go substituiert werden, wenn das 
Komposituiu} „Fenster“ bedeyitet, 

H. I 2. 38. 



So ist gavaksa nitya nur in der Bedeutung Fenster. Hema ist der einzige 
Grammatifcer der diese Lehre in ihrem ganzen Umfang aufgaiommen hat 
Vgl H I 2. 28-31 

99 Em jdutierter Vokal (ist vor emem folgend«i Vokal), ausser (i von) 
lit, (fcanen Veranderungen unterworfen, die sonst atattfinden warden). 

P 6. 1. 125, 129, C. 5. 1. 123, H. 1. 2. 32. 

Hier auch gibt B das Zeichen far die Plutierung nut iwieder, wie m 
i 1 , ?2 . iS^t lasst den cakravarma^-prattsedha weg (P 6 1. 130 = C. 5. 

1 . 124 = H. 1 . 2. 33) . — In dm Beispid ist suSlokeh nar die „ Padapatha “ 
— ^Form.*® S. Arun za 104. 

100 Die nut dem stanunen Buchstaben g versehenen (Laate sind vor 
einem folgenden Vokal keinen Veranderungoi unterworfen, die sonst stattfinden 

P. 1. 1. 11, 12, C. 5. 1, 125, 126; H. J. 2. 34, 35. 

Z B. gu and gi in i 2 44, 46 ^ 

101. (Die) aus (einem einzigai) Vokal bestehenden (Partikdn) ca usw., 
mit Ausnahme des nut dem stummen Buchstaben n versehenen d, (wenn sie 
feein Wesen bezeichnen, sind vor einem Vokal kemen Veranderungen unter- 
worfen, die sonst stattfinden warden). 

P. 1 1. 14; C. 5 1. 127; H. 2. 2. 36. 

{78} Der Vers wird in demsdben Zusammenhang im MBa^ya (Bd. 1, 
S 7) und in der Ka'§. (zu P. 2. 1 14) angefiihrt, 

102 Eine auf o (audautendfi Partikel ist vor emem Vokal keinen Ver- 
anderungen unterworfen, die sonst stattfinden warden) . 

P. 2. 1. 15 ; C. 5. 1. 128 ; H. 2 . 2 37. 

103 Em aus (der Kasusendung) s (entstdiendes o) kann vor if* (un- 
verandert bleiben). 

fP. 2 1. 16 ;C. 5. 1. 129, H. 2. 2. 38. 

104. Und (fur die Partikd) u (kann vor iti aiuih) (tretai). 

P. 2. 1. 17, 18 + Va. 2 : C. 5. 1. 130, 131 ; H 2.2. 39. 

Hier beruht das ca auf der von Kiaty. vorgeschlagenen und von Pat 
^estiitzten Zerl^jung des Pgipini’schen Sutra iyogavibhaga) Vgl. Va. 1 und 
2 zu P. 2. 1. 17, 18 und Kaiy ebenda — Diese Substituticml hat eigenthch 
ihre Stelle nur im Phdap&tha Hire Aufnahme dundi gakait. wird datauf 

28 Das mtspreohende Sutra FSnini’s lautet . aplutavad vpasthite {6, 1. 129) , 
Nach dem MBhS^ya heisst upasthita hier oitorfa itikarandk „ das nidit vt» den 
Bsis herruhmide (also nicfat dem eigentlithen vediachen Texte angdiorende) Wort 
iti Die fiigt hinzu : samudayad twaccktdya padcnit yena svatupe ’vastha- 

pyate „wodurdi dn Wort von dem Aggregat getrennt und ip seiner eieenen Gestdt 
hingestellt tvird". 

2. TEIL 


beruhen dass bei Pan die Substitution als anarse gelehrt wird*^ 

105 Nach (einem auslautenden Konsonanten von dem pratyahSra) 
may (d. h von Muten und Nasalen ausaer n, kann fur die Partifed «,) 

wenn ein Vokal folgt, v (substituiert werden. Diese Substitution ist jedoch 

als) nidit eiAgetreten (zu betraditen) 

P. 5 3 33 , C 6 4 16; H. I 2. 40. 

Im kim u iti kann der musvara fur m nur dann substituiert werden, wenn 
das V nach der Regel asve [1. 1. 7,31 fhr « substituiert wird. 

106. Em (auslautender) Konsonant (am Endc ernes pada kann) vor 
einem Nasal m den entsprechenden Klassennasal (iibergehen) . 

'P. S. 4 45; C. 5 4. 140 ; H. I 3. 1. 

• Hal und mcht ym (wie in P 5. 4 45) ist w^en des Folgenden ge- 

braucht Diese Andemng konnte geraacht werden, da h ja doch keinen ent 
sprechenden Nasal hat. 

107 (Ein auslautender Konsonant am Ende ernes pada muss) vor 
einem (mit Nasal anlautenden) Suffix (m den entsprechenden Klassennasal 
iibergehen) . 

'P. S 4 45 Via ; H. 1. 3. 2. 

108. (Pur den Vokal, der) dem mit (dem stummen Buchstaben) 
versehaien s Oder r vorangeht, (wird em nasaler substituiert). 

Cf. P. S 3. 2, 4; C. 6. 4. 6, 7; H 1.3.8. 

109. Fiir em nicht am Ende ernes pada stdhendes m tmd « wird vor 
emem folgeiden Konsonanten mit Ausnahme do: Nasale, Halbvokale, Sibi- 
lanten und h (der diesem entsprechende Nasal substituiert). 

P. 8. 3 . 24, 4 58 , C. 6. 4. 9, 151 ; H i, 3. 39. 

{79]} Das Dvandva mn- verlangt den Dual oder eigentlich deii Smgular, 
well die Dvandvas m den Sutras iSakajt.’s soost stets als neutr smg. erschienen 
Der Plural soli andeuten, dass die nach der ^ora-Regd [i. 1.- 46] dntretende 
Vetiwandlung von « in « unterbleiben soil. Dies wird bei Pan automatiscli 
durch die Anordnung der Sutras erzielt. Vgl. die Kas zu P 8. 4. 58 *. iha 
kurvantt vrscmtt tty atra ifotwt- [P. S. 4. 2] sydsiddhatvat purvatjt nakarasya- 
nusvarajj, [8. 3. 24] kriyate | tasydpi parasavarnena nakdra etfa bhavati | 
tasyapy asiddhaivat punar ftatvani na bhavati. Man bemerke die Willkiir 
der iSakat’schen Bezeichnung. Diese hat unzweifdhaft ihr Vorbild in den 
Erklaiungsversuchen Patafijali’s durch jhdpakas. 

110 (Pur inlautendes m und « wird) vor Sibilanten und h der anusvdra 
(substitmert) . 

P. S. 3. 24, C. 6. 4 9, H. J. 3 40. 

ar Einldtung S. 9. 



111. Fur das Nasalinfix m und fur das am Ende eines pada stehende 
}ti kann bcidcs (d. i dcr onusv&io odsr d6r ufMiMsiko substituidrt wsrdcrit) 
wenn ein Konsonant folgt 

P 5 . 4. 59 + Va2, 3 zu7. 4 85 ; C. « 4 152 , H. J 3.14 

Nach dera Varttika 2 zu P 7 4 85 soil das dem Vokal der RedupUka- 
tionssilbe des Frequentativum angefugte Augment nicht n, sondern edn anu- 
svara sem. Weil femer nach dem folgMiden Varttika padSntavac ca dieser 
antisvara als am Ende eines pada stehend angesdien wird, kann dafur der 
dem folgenden Konsonantai ent^chende Klassennasal emtreten (P S 4 
58). Weil iSakat. die Vertretung des Klassennasals dutch anusvafa hier 
direkt vorschreibt, kann er die Forderung Katyayana’s padSntavac ca 

112. Vor h mit folgendem I, v, y, m oder « (werden der Reahe nach 
der anusvara und der dem auf h folgenden Konsonanten entsprechende 
anunasika fur ein aim Ende eines pada stehendes m substituiert) 

P. S. 3 26 + Va. 1. 27. C 6.4 11; H. 1.3. 15. 

113. (In) samrat (bleibt das m unverandert). 

P.8.3 25 ;C 6. 4. 10 ; H. 1. 3. 16. 

114 Vor den Sibilanten konnen fur die Tenues die Tenues (substi- 
tuiert werden). 

P 8. 4. 48 Va 3 , C 6.4. 158 . H. 1 3. 59. 

Weil es em zweckloses Verfahren sem wurde, fur die k, p usw dieselben 
Buchstaben wiederum einzusetzen, ist das Sutra wohl so zu verstehen, dass 
vor den Sibilanten die aspinerten Tenues fur die homogenen rachtaspinerten 
emtreten konnen (und umgekehrt) Das diesem Sutra zugrundeliegende 
Varttika lautet unzweideutig cayo dvitiyah San pauskarasadeh (P 8 4. 48 
Va 3), „Vor den Sibilanten konnen nach der ^^ehnimg Pauekarasadi’s die 
zweiten (d. h aspinerten Tenues) fur die nichtaspirierten emtreten “ 

{^80} 115. Die auf Sibilanten, anusvara, visarjemiya, phtramuliya und 
upadhmaniya folgenden (Tenues oder die auf Tenues folgenden Sibilanten 
usw.) konnen verdc^pelt werden, (aber erst,) nachdem (was sonst zu voU- 
ziehen ist, vollzogen worden ist). 

P. 8 4. 47 Va. 2 ; C. 6. 4. 143 ; H. 1. 3. 35, 36. 

Nach der Verdoppelung wird fur den ersten Konsonanten ein nichtas- 
pinerter nach der bekannten R^el substituiert — Die fur dieses und das 
nachste Sutra zugrunddi^enden Varttikas (P. 8. 4. 47 Va 1, 2) werden von 
Patafijah ebenfalls dpppeldnmg aufgjefasst Der anusvdm ist frehch in den 
dort aufgezahlten Konsonanten mciht mit ednbegriffen^B. Dieses ist als 
Neuerung iSakat’s zu betrachten. Bd Hema. (H. 1. 3. 35) lautet die 
Regel dbenso. 

as Pur (lie Verdoppelung hinter dem anusvara d. Wackeknagel, AIttnd. 
Gramm. 1. § 98 

2. TEIL 


116 Ein auf einen Halbvokal folgender Konsonaitf, ausser h, H and 
den Rihilanti^, (oder umgekehrt, dn auf die Konsonanten ausser h usw. 
folgender Halbvokal, kai m verdoppdt werden, aber erst, nachdem was sonst 
zu voUzidien ist, vollzogen worden ist). 

P. 8 4. 47 Va. 1 ; C. 4 143 ; H. 1. 3. 33. 

vjksav ist ein Denominativ von v^ksa „Baum“ und heisst v[ksavxsc<tn 
acaksdno, nach Hema In prorrofunava (perf akt. 1. sing von urtm nut 
pra) wrd die Reduplikation zuerst vorgenommen und erst dann die Verdop- 
pelung. Denn nacli P. 5 1. 2 3, werdiein, wenn die zu reduplizierende Silbe 
vokalisch anlautet, fur den zweit«i Komples von Lauten, der nur emen Volsal 
enthalt {ekdc), zwei gesetzt d i. es wird didser zweite Komplex zweimal ge- 
setzt. n, d und r werden aber als An f ang slaute einer Konsonantengruppe nicht 
wiederholt. Die Perfectform von iimnu lautet ier + [t?M+ «u] +• a, was nach 
einigen anderen Anderungien zu urmnava und dann endlich nach unseier 
Regel auch zu urmumva wird. Wird aber die Verdoppelung zuerst vorgenom- 
men, so musste das abhyastd?° ebaifalls ein doppdtes « enthalten («r+ [»y« 
+««m] +a). 

117. Nach r und h kann, wain ihnen ein Vokal vorangeht, (ein Laut) 
nut Ausnahme van h, r oder einan Vokal (verdoppelt werdai) 

P «. 4. 46;C. 6. 4. 141; H. J. 3. 31. 

118. Nach emem nichtlangen Vokal '(kann ein Laut, nut Ausnahme von 
k, r Older emem Vokal, verdoMidt werden) . 

P. 5, 4. 52 ; C. 6. 4. 147 ; H 1 3. 32. 

Die notwendige Erganzung zu diesem Sutra folgt. m 119, 121. Dem 
Same nach umfasst dies Sutra auch das von Pat als unnbtig zunickgpwiesene 
Varttika avasane ca (P. 8. 4. 47 Va. 3. ) Sakat um-£81J-geht den von E^t 
gerugten Mangid anders als Pat Die drei folgenden Sutras iSakat’s hatten 
in ems verschmolzen werden konnen. Eben diese Zerlegung — meint dei 
Kom., und ohne Zweifd nut Redit — deutet darauf hin, dass die Verdoppel- 
ung auch in der Pause stattfindet. Doch unterschiedet sich die Regel Sakata- 
yana’s von der Auffassung des PBjjihi’sdien Sutra, die Pat vortragt, dadurch, 
dass Sakat die Verdoppelung nach einemlangea Vokal fiirunstatthaft erklSrt. 
Pat gestattet also vakk und vak, wahrend Sakat ledighch das letztaie zulasst 
Hema. veiinetet ebenfalls die Verdoppdung nadi emem langen Vokal ; vgl 
adirgfiad viranudkavyanjane (H. 1 3. 32). 

119 (Die 115 f. gdehrte Verdoppdung ftndet) nicht (statt), wenn (auf 
die betreffenden Laute) erne Konsonantengruppe folgt. 

P. 8. 4. 50. H. 1. 3. 32 

Sidie Anm zu 153 

Warum die zweite Sdbe des dbhyasta n und nicht enthalt, eridilrt die 
SiddhSntak, (S. 375 unten). 



Diese Regel habe ich in der Candra-Grammatik nicht gefunden und bm 
geneigt zu glauben, dass sie dort fehlt 

120. (Das t) von putra wird vor ad%n und putrddm (rucht verdoppelt), 
wenn mit dem Wort geschmaht wird®^ 

P S. 4. 48 + Va 1 , C 6 4. 145 , H. I 3 38. 

121. (Die Verdoppelung unterbleibt), wenn (auf die betieffenden Laute) 
ein Vokal folgt 

P S. 4. 47, C 5. 4 142, H. 1. 3. 32. 

122. Die Sibilanten (werden nicht verdoppelt, wenn ein Vokal folgt). 

P. 8. 4 49 , C 5. 4 146 , H. i. 3. 37, 

123 A, •« und « am Ende des pada nach kuizem Vokal (werden ver- 
doppelt, wenn ein Vdtal folgt). 

P. 8. 3. 32 ; C. 5 4. 126 ; H. i. 3. 27. 

Zu krsarm iha : Die Substitution des fur « ist bahirafiga und zum 
Zwecke der Substitution des fur « als mcht voUzogen zu betrachten ; infolge- 
dessen bleibt das « im Auslaut und kaim mcht m n verwanddt werden. 

124 Nach emem langen (Vokal am Ende einea pada) kann ch (ver- 
doppelt werden). 

P 6. 1. 75, 76 , C. 5. 1. 73 , H 1. 3. 28 

125. Nach emem plutierten Vokal (am Ende ernes pada, wenn ihm ein 
langer Vokal zugrunde li^, kann ch verdoppdt werden). 

H i. 3. 29. 

Diese vermag ich bei kemem anderen Grammatiker ausser Hema. 
(HI 3. 29) nachzu'weisoi 

{82J 126 Nach einem Vokal, (sowie nach den Partikeln) a und tm 
(muss ch verdoppelt werdai). 

P 6. 1. 73-75; C 5. 1, 73 ; H. 1. 3. 28. 30 

In prach + na (i= pt^asna)' findet keme Verdoppelung des ch statt, weil 
die Substitution des i fur ch nach 1 . 1 . 115 schon vorher eingetreten ist, 

127, Diejenij^ (auf at ausgehenden mehrsilbigen Wortformen), denen 
das Suffix dac [ = a] angefiigt wird, erleiden Verlust des at vor iti. 

P. 6. 1. 98 + Va. 1; C. 5. 1. Ii02. 

Nach &. 3. 4. 54 wird ddc nur an tnehrsilbige, sdiallnachahmende Wort- 
formen angefugt Das andkdcah stammt aus don Vlarttika : itav anekacgraha- 
mfn Sradartham (P. d. 1. 98 Va ) 

128 Wird (die Wortform, der das Suffix ddc [■= a] angefUgt wird) 
verdoppelt, (so wird das at vor tti) mcht (elidiert). 

P. d. 1. 99; C. 5. 1. 103. 

Die Verdoppelung hdngt wohl mit Stz des Ictus zusanunen ; andeis 
Wackernacei. (Altind, Granun. 1 § 98 a Antn.), der dieei in Zusammenhang mit 
dem Spirechtempa bringt. 

2. TEIL 


Zu vipsayam usw Das distnbutive Verhaltnis wird durch Wiederholung 
dcs Wortes ausgedrudkt ; patatpatat ist aber die Nachahmung des mehriadien 
Schallea (und drhckt kein distributives Verhaltnis aus>. 

129- (Wird die Wortform, der das Suffix dac [ = angefhgt wird,) 
verdoppelt, (so wird das) t (vou at vor iti didiert). 

P 1. 99, C. 5. 1. 104 

130 Folgt (auf die verdoppdte Wortform das Suffix) Me ([=41, so 
wird das auslautende t) in dem zuerst (ausgesprodienen Teil didiert). 

P 6. 1. 100 = P. 6. 1. 99Va 1 ; C. 5. 1 105 

Dieses Sutra beruht auf einan VSrttika K&t’s (P 6. 1. 99 V3. 1) 
Doch bemerkt er (Vartt 2), dass die Regd entbehrlich sei, da man auch von 
palapata (emetn auf a auslautenden Onomatopoetikon) ausgehen kann. 

131 dh und t (fallen) vor dh respektiv r (ab) 

P 8. 3. 13, 14 : C. 6. 4. 18, 19 , H J. 3. 41, 42 

132 Em auf emen Konsonanten. folgender Halbvofcal oder Nasal kann, 
wenn eben dersdbe Laut folgt, (ausfallen). 

P. 8 4. 64 , H 1.3 47 

Wenn yamdm gdesmi wird, wie es bei Paio steht, so wdrde mfolge der 
Versdiiedenhdt des Numerus von yatnam und yami das yathdsatjtkhya (s. 
P I. 3 10) nicht ohne wieiteres verstandlich sem. Trotzdem setzen die 
ErMarer Baja’s voraus, dass der AusfaU der Konsonanten nur dann stattfindet, 
wenn der gl&cke Laut folgt In der SiddhSntak. (p 17) heisst es ausdriick- 
lich : yamatfi yamti yathasartikhyavipiman neha \ nu&hatmyatn. Die Gleich- 
heit der Laute wird von Hema (J. 3. 47) ausdriiddich gdehit — Zu adityya 
vgl KaS. zu P. 8. 4 64. 

£83} 133 Eme (auf ttaea Konsonanten folgende) Muta oder ein 
Sibilant kann vor emem homogenen Laut (ausfalleti) 

P. 8 4. 65, C. 8 4 155; H. J. 3 48 

Der Kom gibt kemen Aufschluss dariiber, warum das Wort va hicr 
wiederholt wird, wenn es durch die anuvritt aus dem vorangehenden SG.tra 
ohne weit^es erganzt werden kikinte 

134. (Die Muta Oder der Sibilant) von den auf (die Praposition) ud 
folgenden sthd und stambh (werdai vor einer Muta odm: mnem Sibilanten 

P 8. 4, 61 ; C 8. 4 154 ; H, I. 3. 44. 

Hier wird die Regel etwas anders formuliert als bei Pginini (8. 4. 61). 
■Sakat. 1^ das aniautende s der Wurzeln ausfallen — so ist die Regel sicher 
zu verstehen , vgl Prakriyas. Sutra 51 : dkStvoh aakdrasya lug bhavati joti 
pare — und die Verdoppdung findet nach adirghat statt (S. I. 1. 118). 
Anders bei FSn. — Das von Patafijali erforderte utkandaka (aus ut+skand) 
als Bezeichnung einer bestinomten Krankheit wird in emen Ga];ia aufgoioinr- 


men — Das VSrttika zu P. S. 4. 61 fallt naturlidi weg, well es sich auf eine 
vedische Form -bezidit. 

135. (Eine Muta oder ein Sibilant geht vor emer Muta Oder einem 
Sibilanten in) erne tonlose mcht-a^inerte Muta (iiber). 

P 8. 4. 55; C. 6. 4. 148; H. I. 3. 48. 

Das nachste Sutra beschrbnkt das (Sehiet dieses Sutra’s. 

136 (Fur eine Muta Oder emen Sitnlanten wud) eme tdnende nicht- 
aspinerte Muta (substituiert), wenn eme tbnende Muta folgt. 

P. 8. 4. 53; C 6. 2 . 115; H. 2. 3. 49. 

137.. Wenn s oder ein Dental mit s Oder einem Palatal (zusaratnenstdsst, 
werden fur s) f und (fur den Dental) em Palatal (substituiert). 

P 8 4. 136; H. 2. 3. 60, 61. 

138. (Worn s Oder ein Dental) mit 5 oder einem Zerebral zusammen- 
stdsst, (werden fur s) s und (fur den Dental) em Zerebral (substituiert) 

P. 8 4 . 41 ; C. 6. 4. 136; H. 2 . 3. 60, 61. 

133. Nadi 5 (findet die in 137 gdehrte Substitution des S und der 
Palatale) nkht (statt) . 

P 8 . 4 44; C 6 4. 139, H. 2. 3 62 

140. Nach einen am Ende ernes pada (stdienden) Zerebralen (gehen s 
imd die Dentale nicht m s und die Zerebrale iiber), mit Ausnahme (der 
Kasusendung) vdm und nagarl und navati. 

P. 8 . 4. 42 + Bh. ; C. 6 4. 137; H. 2.3 63. 

141 Vor s (gdit ein am Ende ernes pada stdiender) Dental (nicht m 
s Oder 2!erei>ral itber') 

P. 8 . 4. 43 ; C 6. 4. 138 ; H. 2 3. 64 

|^84J ^adika erklart Pat zu P, 1. 4. 18 Va. 1 folgendermassoi : sa} 
aiigulayo yasya sa $adangidth \ amkcmpitab sadangulif^ sa4ikah ! 

142. (Era am Ende eines pada stehender Doital geht) vor I (in) I 

P. 8 . 4. 60 ; C. 6. 4. 153 , H. 2. 3 65. 

143. Fur h kann nach einer (am Ende eines pada stehenden) 

nicht aspirierten Muta eine tonende aspirieite Muta (substituiert werden). 

'P. 8 . 4. 62 ; C. 6. 4. 156 ; H. 2. 3 . 3 

PBip. braucht jhay statt jkaS wegen des Folgenden, wdches die tonlosen 
l^utae verlangt ; jhaS (die tQnenden Mutae) hatte aber genugt, da m Wirk- 
lichkdt die tdnenden Mutae zuerst fur die tonlosen substituiert werden 

144 ^ Fur i (kann nach emer am Ende eines pada stehenden tdnenden 
nicht-aspirierten Muta) ch (substituiert weidai,) wenn ein Vokal, Halbvokal, 
Nasal Oder h folgt 

2. TEIL 


P. S. 4. 63 + W , C. 6. 4. 157 ; H. 1. 3. 4. 

In dieser und dai folgenden Regeln ist zu beachten, dass nach J 2. 75 
iur alle Muten am Ende ernes pada ane nidit-aspmerte Media (jo^) subsb 
tuiert wird. 

145. An ein (am Ende ernes pada stehendes) n und (kann) g res- 
pektiv 4 vor dnem Sibilanten (angefugt werdoi) . 

P. S. 3. 28; C 6 4. 12; H. JT 3. 17. 

Der Vokal a vertritt bei iSakat fast, wenn nicht ganz ausnahmslos, den, 
um die Aussprache zu ermoglichen, eingeschobenen normalen Vokal, wie z. 
B. hier gak, dak. Bei PSn dagegen ubemimint der Vokal » sehr haufig 
diese Rolle z B. kuk, tuk. Ich brauche nur an das SUtra ukalo ’) jhrasvadir- 
ghaplutah (I. 2 27) zu ermnem, wo man eigentlich mcht begreift, warum 
der Vokal u vorgezogen wird, wenn a oder t ebensogut den Zweck hatte 
crfuUen konndi. 

146 (An ein am Ende eines pada stehendes) d oder » (kann) t vor s 
(antreten, ]edodi) mdit (wenn der Sibilant den ersten Bestandteil) vc«i ic 

P. 8 3. 29, 30 ; C 6. 4. 13, 14 , H. I. 3. 18. 

Den Zusatz ascak vermag ich bei kemem anderen Granunatiker ausser 
Hema (H. 1. 3. 18) nachzuiweisen. Er fiihrt als Beispiel sat kyotati an. 

J47. Vor S (kann dn am Endd eines pada stehendes) n das Augment 
f (erhalten, jedoch mcht, wenn der Sibilant den ersten Bestandtdl von ic 

P 8. 3. 31 ; 4. 63 ; C. 6. 4 15 ; H. 1 . 3. 19 

Wegen aScak fuhrt Hema. ausdrucklich das Beispid bkavoH icyotati an 
(H. 1. 3. 19). S. Amn. zum vorangdienden Sutra. 

148. Pur (das auslautende n von) .nfn kann [ = r ] (substituiert 
{85} werden) oder (es kann demsdben) am Ende em r (angefugt werdcn), 
wenn p folgt (oder nfn kann unverandert bldben). 

•P. 8. 3. 10 , C. 6. 4. 5 ; H. J 3. 10. 

149. (FUr das erste «) in kankm kann « [ = s] (substituiert werden, 
oder es kann demsdben) am Ende ein s (angefugt werden). 

P 8 3. 12;C. (5. 4. 4; H 1 3. 11. 

Ein solches s geht nicht in r uber, well man sonst dben r ala Substitut 
hatte lehren konnen. V^. PrakiiyBs S. 16, Anm 1 : alra sjsafeor vidhanat 
Hr na syat j yady atra paddntavarttnok ^sakoh sajurahassa}}, .[1. 2. 27J ity 
ddinS nh syat tarhi riTokav eva vidhlyeyatam, 

150. (FOr ein auslautaides n) mit Ausnahme des n von prasdn, (wird 
si [ = s] substituiert) vor ch, th, th, c, t oder t, mit darauf folgiendein Vokal, 
Halbvokal, Nasal oder h (oder es kann demsdben s angefugt werdm). 

P. 8. 3. 7; C. 8. 4. 3; H. 1. 3. 8. 


W^en s vgl Anm. zu dem vorangehenden Sutra - 

151 (Fur das auslautende m) von pum (kann si l=s] substitmert 
werden) vor einer tonlosen Muta (nut darauffolgendem Vokal, Halbvokal, 
Nasal Oder h, oder es kann demselben s angefugt werden). 

P. S 3. 6; C. 5. 4, 2; H. 1. 3. 8. 

Wegen s vgl Anm. zu 1. 1. 149. 

152 Vor Hpm s, (das) kr (angefugt wird, kann fiir das m) von saw, 
(si [=si] substituiert oder dem w em s angefugt werden) und (es kann 
dafur) eine Niete {[ghtk] substituiert werden) 

P. S, 3. 5 + Bh. ; Of C 6. 4, 1 ; H. i. 3 12, 13 

Die Substitution von gluk fiir das m von sam beruht auf Missverstandnis 
der tsti Patanjali’s : samo va lopam eka tcchanti (MDhS$. Bd. 3, S. 425, 
Z. 8), seitens iSakait Pat lehrt die Substitution von lopa hint^ sam fur den 
unmittelbar da/tmf folgmden Lout. Fasst man aber samah als Genetiv auf, 
wie |!^lra| t- getan zu haben sdieint, so muss man den lopa fur daa m von 
sam eintreten lassen. Also erhalten wir die Bildungen saskarid usw. Hema, 
der Sakat. auch in diesem Punkte nadifolgt, lehrt ebenfalls den Abfall von m 
m sam vor skr usiw. Diese Verwechslung mag man auf den ersten Blick kaum 
fur mbglich haltea Es sei aber daran eriimert, dass zwei von Kielhorn’s 
besten Handschriften G und A — von der emen (G) sagt er : „m settling 
my text, I have been chiefly guided by the MS.' G, which is the best of all 
the MSS of the Mahflbhflshya that have come under my notice “®®, und von 
der anderen (A) „ a carefully maid and complete copy of the Mahfl- 
bhashya‘'8»— 4atsachhch als Baspid sasharta f86} anfUhren Es ist also 
gar nicht au^esdilossen, dass die R^d, wie sie von ISakat. formuliert wordeti 
ist, fUr gewisse Grammatiker und Grammatikerschulen als feste Norm gegol- 
ten hat 

153. Fiir (die am Ende eines pada stehenden) v und y (wird) nach 
agho, bhago und bko oder (wenn den v und y em) a, (kurz oder lang, voran- 
geht,) vor Vokalen und vor tdpenden Konsonanten (eine Niete [glttk] sub- 

P. 8. Z. 22+ 17 Bh ; C. 4 26 , H. I. 3. 23. 

vfksav ist ein mit dem ^f-SulSx; vtc gebildeter Nonunal-stamm zu dem 
Denominativ Vfk?avayati. jSo die KSiSika 2 !u F. 8. 3. 17 : vjksam vfscatift 
vrk^avft I tom acaste yah sa orksavayati | vjksavayater apratyayah \ vrksav 
karoti .^. — Pur das hah PSpdni's m dem entsprechenden Sutra (P. 8. 3, 22) 
hat iSakat aji. Diese Veranderung ist duich das Bha§ya Pataifljali’s zu P. 8. 
3. 17 veranlasst • uttarariharm tarhy t^grahanam kartavyatp, hali sarvesdm 
[8. 3. 221 holy aHti yathd sydt, Damadi tntt die Substitution des lopa 

MahaWiBsya, Bd. 1, Emleitung S 8. 

MahSbhSi^ Bd. 2, S 7. 

Vgl. MBhi$ zu 8, 3. 17 und Kaiyata dazu. 




nicht vor alien Kansonanten ein, sondem nur vor denjenigen des praiyakara 
a§, wobei zu bemerken ist, dass aS bei FSsrini detn as iSakat’s ganz genau 
enlspridit — Zu bho vyoma . das v von vyoma fallt nicht ab. 

154. Vor einem Vokal ist es freigestellt (die in 153 ^lezifizierten Kon- 
sonantai y und v) undeutlich (auszusprechen). 

P 8. 3. 18, 19, 20 ; C. d. 4. 27 ; H. 1. 3. 24, 25, 

Das aspasta iSakat’s heisst laghuprayatnatara bei PSjnini (S 3. 18) . 

155 Wenn (den in 153 spezifizierten Konsonanten y und v) « vorangeht, 
wird (vor einem Vokal) ausser der Partikd u (sowohl der Ausfall als die 
undeutlidie Ausepradie als audi der Nichtausfall) freigestellt. 

P. 8 3. 18 , C. 5 4 27, H. J. 3. 25. 

156. Fur das n (= t)' (wird nach agho, bhago, bho und wenn ihm ein 
a Oder o vorangdit vor Vokalen und tbnoiden Konsonanten) y (substituiert). 

P. 8. 3. 17; C. 6 . 4 24-26; H. 1. 3. 26. 

Die hier vorgeschndaene Substitution fur das r, dem ein a oder a voran- 
geht, ist der Beschrankung durdi das folgende Sutra unterworfen. 

157 (Piir das n 1= r] wird) nach emem a “ (substituiert), wenn ein 
a Oder ein tbnender Konsonant foigt 

P. 6. 1. il3, 114, C. 5. 1. 119; H. 1. 3. 20, 21. 

158. Nach dem fdr t (in tad und etad) substituierten s fallt vor dnem 
Konsonanten (das ri [ = r] ) ab, (wenn die Worte) nicht mit der Negation 
komponiert smd. 

P. €. 1. 132 ; C. 5. 1. 134 ; H. 1. 3. 46, 

1^873 Warum das akoh des enteprechenden PBlnim’schen Sutra (6. 1 
132) ausgelassen wird, ist nur nicht klar gewordoa Hema schbesst sich an 
Paj). an (H i. 3. 46) ! 

159 (Das n [=i'] fallt nach dem) fiir t in tad (substituierten s ab), 
wenn (durch diesen Abfall) der Stcdlen metrisch vollstandig wird. 

P. 6. 1. 134; H. 1 3. 45. 

Der Halbvers scdsa ddSarathih usw. wird m demselben Zusammenhang 
m der KM. zitiert (KS^ zu 6. 1 134)**. Zu act des PSimni’schen Sutra 
bemerkt der VittMra (a. a O.) act vispastartham, „aci der Deutlichkeit 
wegen “. Den der Ausfall des s vor emem Konsonanten kann die VoUstdn- 
digkeit des Metrums mcht beeinilussen, wed die Sdbenzahl dadurch nicht 
verandert wird. Es liegt nahe zu vermuten, dass diese Bemerkung des Ver- 
fassers der Kai. Sakat. veranlasst hat, das aci fortzulassen. 

ISO. (Piir das ri [ = r] ) von ahan (wird) ein (einfaches) r (substi- ^ 
tuiert), wenn ein Vtdtal oder. tbnender Konsonant foigt, jedoch nicht vor 

** Die P&das a und b des ersten Verses saijo usw. und der Halbvers so esa 
bharato usw. finden ach audi in P, ein Beweis dafur, dass seme Vorlage mch 
PeisfMte entbidt, 


einer Kasuseadung und rUpa, rdtii und rathantara 
P. 8 . 2 . 69 + 68 Va ; C. 6. 3. 100. 

I6J, Fur den vtscerjamya (wird r sulsatituiert, wenn ein Vokal odei 
tbnender Konsonant folgt). 

P. S 2 66 ; C. 6. 3 98 

162. In aharpati us\f kann (fur den visarjamya r substituieit werden). 
P. 8. 2 70 Va und Bh. , C. 6. 3. 102 ; H. 2 3 58. 

B. liest ei?pati Die richtige Lesart ist sicherlich gihpatth, wie es in 
der Kielhom’schen Ausgabe des MBha^ya (Vol. Ill, S. 412, Z. 15) und 
audi bei Hema. steht. Die Ka6 liest gtspatth, wie es ubngens auch in emer 
MBhasya-Handschrift stdit. — Das vd rephdd atra usw des Kean, ist nur 
nicht Idar gevrordoi. 

163. (Fur den visarjamya wird) vor ch, th, th, c, t und t s (aubeti- 
tuiert), iwenn (auf jene Konsonanten) kein Sibilant folgt 

P. 8. 3. 34, 35 , C 6 4. 28 , H i 3. 7. 

164 Vor einem Sibilanten, (dem kein Sibilant folgt, ist die Substitution 
von s fiir den visorjamya) freigestdlt 

P. 8. 3. 36; C 6. 4 29; H. 2. 3 6. 

aSarpare im Kom. wird wegen der anmrtti fortgefdhrt. 

165. (Folgt dem visarjamya em Sibilant), dem eine tonlose Muta folgt, 
(so kann der visarjamya) aiiRfallpn 

P. 8. 3. 36 VS, 1 , C. 6. 4 30 ; H. 2. 3. 56. 

Kat. Idirt in dem VSrttika zu. P 8. 3. 36 den bdiebigen Abfall des 

visarjamya vor §ar, dem khar folgt. 

166. Vor emem toolosen Guttural und Labial, (auf die kein Sibi- 
lant folgt, kann der Reihe nach fur den visarjardya) X reqiefctiv x (substi- 

tuiert werden). 

P. 8. 3, 37; C. 6. 4. 31 . H. 2. 3, 5. 

Wegen der Lesung adbhijj psdtam gegen die Hss. sieihe MBha^ und 
KSfikS zu P. 8. 3. 37. 

167. par (den visarjamya) der gati tiros (kann vor einem toolosen 
Guttural oder Labial) « (= s) (substituieit werden). 

P. 8 3. 42 ; H. 2. 3. 2. 

Auf Grund dieser Substitution kann ein solches s mcht wieder in visar- 
jamya ubergehen; nach 2. 2 65 aber kann s(t) zu $ werdlen 

168 Pdr (den visarjamya in den gatis) natnas und punas (wird vor 
einem toolosen Guttural oder Labial si [ = s] substituiert) 

P. 8. 3. 40;C. 6. 4. 35 ; H 2. 3. 1, 

169. (Pdr den visarjamya) von catur, nis, dus, bahts, avis und prddus 
(wird vor einan tonlosen Guttural oder Labial s* [= a] substituiert), 
p. 8. 3. 41 ; C. 6. 4. 35 ; H. 2. 3. 9, 

2. TEIL 79 

170. (Fiir dea vtsarjamya) des (Suffixes) sue (= s) kflnn (vor pini^m 
tonlosen Guttural oder Labial si [=s] substituiert warden). 

P. S. 3. 43 , C. 6. 4. 36 , H. 2. 3. 10. 

Indetn iSakat. in der Formulierung der Regdn uber die Verwandlung m 
? von s der Prapositionen ms, dus usw und der Zahladverbien dm usw. 
(Sutras 169, 1710) von Pan), abweicht, schliesst er sidi an Candra an. Die 
Sutras Candias lauten : mrdurbahiravUcatuspraduspurasdm (5. 4. 35) und 
swo vd (36) Abgesdien davon, dass in dem Sutra Candra’s puras zusammen 
nut den andearen Adverbien, die auf is bez. iis auslauten, aufgefuhrt iwird, 
stimmai die Sutras der zwet Grammatiker ganz genau uberein Die Bemer- 
kung Putanjali’s, dass der Ausdruc^ dvistriicatur in dem Sutra dvistriscatwr 
lit kftvo'rthe (P. 8 3 43) entbehrlich sei (MBha? Bd 3, S 435, Z. 3 f ), 
durfte wohl die veranderte Formulierung von Candra veranlasst haben. 

171. (Fair den visar^amya) ernes auf is oder us (auslautenden Wortea 
kann vor emem tonlosen Guttural oder Labial si [= s] substituiert wearden), 
wenn (die zusammenstossenden Worte mitemander) in KoffreleAion (stehen) . 

P. 8. 3.44, C. 6. 4. 37 ;H. 2. 3, 11. 

Bd P&nini lautet das entsprediende Sutia isusoh samarthye (8 3 44,) . 
Wegen e^eksd v^ die KasikS . sarmrthyam iha vyapek$d | na punar ekdrtM- 
bhavah | ubhayerm va. Candra hat soriibandhe (6 4. 37) . 

172 (Die in 171 gddute Substitution findet) mcht (statt, wenn die 
mit Guttural oder Labial anlautenden Worte) nut Ausnahme von Zdtwdrtem 
(mit denem auf is und us auslautoiden) im Kongruenzverhaltnts (stehau). 

£89} H 2. 3. 12. 

ekariha ist hier nach dem Kom im Sinne von samdnadhikaraiia ge- 
braucht ; ebenso bd Hema (2. 3. 12) Ein Verbum imd an Nooien konnen 
schednbar samanddhikaranapada sem. Merke aber, dass hier das Vabum 
(kriyate) passivisch gebraucht ist. Dag^en m den Bdspielen zu 171 ist ea 
akttvisch (karoti) gebraucht. 

173. In der Komposition (ist die in 171 gdehrte Substitution notwen- 
dig), vorausgesetzt (dass das Wort auf ts oder its) kern (zwdtes Glied) 
eines Kompodtums (bildet). 

P 8 3 . 45; C. 5. 4. 39 ;H. 2. 3. 13. 

Nach den Regdn 1 1. 171, 172, 173 ist die Substitution von s Mr den 

1 fakuttativ : 

a) Wenn die Worte in Korrelation stdien, e.g sarpis karoti, sarpik 
karoti (171), sdbst wenn das erste Wort das zwate Glied eines Kompositums 
bildet, z. B. paramasafpis karoti, paratnasarpik karoti (171). 

j8) Wenn die zwd Worte miteinand^ nkht im Kongruenzverhaitnis 
stehen, z. B. sarpi? kutnbhe, stffpik kumbhe (172) . 



7 ) Wenn das zwdte Wort ein yerbum ist, mit dem das erste Wott 
im Kongruenzverhaltms steht, 2 . B. sarpis kriyate, sarptk knpate (172) . 

2. notwmdtg : 

Im Kompositum, vorausgesetrt dass das erste Wort nicht das zweite 
Ghed eines Kompositums bildet, z B. sarpiskundam (173) . 

3. unstattkaft 

a) Im Kompositum, wenn das erste Wort das rweite Ghed eines 
Kompositums bildet, z B paramasarpikkundam (173) . 

( 8 ) Wenn die zwei Worte mitemander nickt in Korrelation steiben, 
2 6 tisthatu sarpih piba tvam udakam (171). 

7 ) Wenn die zwd Worte sowohl m Korrelation als im Kongxuoiz- 
verhaltms stehen, z. B sarpth kahkam (172) 

Die Regdn I 7 und 87 sind wohl Neuerungen von iSSkat Ich babe bd 
den alteren Grammatiken nidits entsprechendes bnden konnen Die Frd- 
stdlung pmamc&cffpi^ kaxoti, paramasarpih karoh beruht auf der Lehre des 
MBhi©. (Bd 3, S 436, Z. 6 , 7) . 

174 (Im Kompositum wird fur den vtsarjamya) von adhas und iiras 
(si [ = s]) vor pada (substituiert, wenn diese Worte nicht das zweite Ghed 
eines Kompositums bilden). 

P. 8 3 47, C. 6 4 41 , H 2.3 4 

175 (Fur den) auf a (folgenden visarjamya eines pada), mit Ausnahme 
eines Indddinabile, (wenn ea nidit das zweite Ghed ernes Kompositums bildet, 
wird im Kompositum si [=s] substituiert), wenn (dne Bildung vonO *r 
und katn Oder (die Worter) kar/isa. kuSa, karvi, kumbha, pdtra folgen. 

t;90} P. 8 . 3. 46 , C 6 . 4 40;H 2 3 5 

pratipadikagrahane usw. = Panbh. 71. Cf. auch MBh^. Vol II, S. 193, 

Z. 6f. 

176. Vor emem (mit einem tonlosen Guttural oder Labial anlautenden) 
Suffix (tntt si [ = s] an die Stelle des visarjantya ; jedoch mcht nach dnem 

P. 6 . 3. 38 + Va 1 , C 6 . 4.32, H 2 3 6 . 

J77. (Fbr den vtsariamya von) auf r (auslautenden Worten) und von 
ahan (tntt) vor kamya (s»[=s']) tiicht (em). 

P .8 3. 3SVa 2jC. 6. 4. 33,H 2.3 7 . 

178 Vor (dnem nut) t (anlautenden), oinem Nomen angefugten (Suffix 
wird fiir den) auf kurzen Vbkal folgenden {visorjaraya si [= s] substituiert) 

P. 8. 3. 101 + Va ; C 6 4 . 87 ; H. 2. 3. 34 

Vgl. PrakriySs. S. 22 , Anni 2 . 

17'9. (FUr dm visarjaniya) von nts vor tap (wird si [= s] substituiert), 
wenn nicht von Wiederholung (d. i. von wiederholtm Gluhen die Rede ist) 

P. 8 . 3. 102; C. 6 . 4. 88 ; H. 2 . 3 . 35 

2. TEIL 


lipa §apa usw. = Panbh. 120, 3 (a. Utpa v. 1. fur Itpa, c. yatrai^ fuj 
yaccd°, caiva fur kim cit, d. °hikt fur °&luci), vgl. Prakriyas. S. 253 ; nicht 
im MBha?. Der Verfasser des PanbhasenduS Idmt sie auch ab ; vgl Kiel- 
horn, PiinbMeendu^ Transl S 519 (unten). 

180. In kaska usw (wird fui den matjantya si I = s] substituiert) . 

P. S. 3. 48 ; C. 6. 4. 85 , H. 2. 3. 14. 

Das paramasarpt'skundtka des Kom ist bemerkenswert. Pat. erkennt 
keine solche Form an (s. Anmerkung zu 173). Nach der Anacht emiger 
Grammatiker enthalt der Gama kaskadt erne Anzahl Komposita wie sarpk- 
kundikS, dhemu^kapSlam, yajxispdtram usw , deren s sich scaist nach der allge- 
meinen Regel mtyam santase ’nutiarapadasthasya (P. 8. 3. 45 = 6. J. 1. 
173) ergibt. Die Aufnahme dieser Komposita in dem Gama soli nun an- 
deuten, dass in densdben der visarjmtya in s, respectv $, auch dann ubergeht, 
wain die mit dem tiisarjamya auslautenden Worte das zweite Glied eines 
Kompositum bilden. Dies ist die Ansicht der Pcraycnjtkas. Vgl. die Kasika 
zu P. 5. 3. 48 sarpi$kundika j dhanuskapalam | barhtfputam | yajuspaP'am 
ity e^am pStha uttarapadasthasyapi satmrn yathd sydd tti ] pai'amasarptk- 
phalam (so zu lesen !) ity epamddipratyuddharandd (d. i in P. S. 3. 45) lit 
pardyatitikd ahuh. — Demzufolge hat Hema. das Baspiel panmayajufpatram 
(H. 2. 3. 14). Zu bhratusputram bei Hema. im selben Sutra, vgl. P. 8. 
3. 41 Va. 4 : bhrdtusputnagrahanam indpakam ekadesantmittat satvaprati- 
sedhasya und das Bhasya dazu ; yadayam kaskadisu bhrSiu^putrasabdant 
paihati usw. 

j[91}. Lebenslauf. 

Ich, Vishnu Sitaram Sukthankar, brahmanischer Konfession, wurde am 
4. Mai 1887 zu Bombay (Indien) geboreri als Sohn des Ingenieurs Sitaram 
Vishnu Sukthankar und seiner Frau Dhaklibai, bezog nach dem Elementaruii- 
toricht in deti Hochschulen zu Bombay die Universitat zu Cambridge 
(England), wo idi mxch vor allem dem Studium der Mathematik widmete 
Als meine Reifepnifung gilt das ffakkalaureatsexamen an ddr dortigen Um- 
versitdt, das ich im Juni 1906 bestand. Sommer 1911 kam ich nach Berlin 
und gab midi von da an hauptsadilich dem Studium der indischen Philo- 
logte hin. Idi besuchte die Vorlesungen der Herren Beckh, Erdmann, Immel- 
mann, Ed. Lehmann, Loeschke, Liiders, Marquart, Mittwoch, Riehl, E. 
Schmidt, W. Schulze, Thomas^, v Wilamowitz-Modlendorf und Wdlfflin. Zu 
del vorhegenden Arbeit wurde idi von Hetm Prof. Luders angeregt, dem ich 
fur sein Interesse an don Entstehen der Arbeit und dartiber hinaus fiir meine 
wissenschaftliche Bildung zu grossten Dank verpflichtet bleibe. Die Promo- 
tion^rdfiing bestand ich am 18 Juni 1914 


I. Introduction* 

No methodical study^ has yet been made of the thirteen anonymous 
dramas issued as Nos XV-XVII, XX-XXII, XXVI, XXXIX, and 
XLII of the Trivandrum Sanskrit Senes and ascnbed by their editor. Pandit 
r Gainapati Sastri, to the cdebrated play-wright Bhasa The first attempt 
at a comprdiensive review of the plays — and the only one that has contri- 
buted substantially to our knowledge of them — ^is found in the editor's own 
introductions to the editio princeps of the Svapnavasavadattd and that of 
Pratimdnataka respectively. Opinion may be divided as to whether the learn- 
ed, editor has fully vindicated his claims r^rdmg the age of the dramas 
or the authorship of Bhasa, but it seems unquestionable that the arguments 
brought forward by him in support of his case deserve serious consideration 
Another approach to a study of these dramas is found m the introduction 
to a subsequent edition** of the Svapnavasavadattd by Prof. H. B. Bhide. 
This author rephes to the arguments of a sdiolm who had in the meanwliile 
published an article iiif a vernacular journal callmg mto question the conclu- 
sion of Gapapati Sastri regarding the authorship of Bhasa, and attempts 
to re-establish it by adducing fredi proofs m support of it. Mr Bhide thai 
turns his attention to the question of Bhasa’s age, whidi he endeavours to fix 
by what may be termed a process of successive elimination. Incidentally it 
may be remarked that his arguments lead him to assign the dramas to an 
epoch even earlier than that claimed for them by Gaoapati Sastri.® While 
it would be iilvid-p49TR>us to behtde the work of these pioneers in the 
held and deny them their meed of praise, it must nevertheless be confessed 
that their investigations are characterised by a narrowness of scope and a 
CCTtain perfunctormess of treatment which unfortunately depnve them of all 
clauns to fi,nality. Vast fields of enquiry have beoi left practically untouch- 
ed , and, it need not be pomted out, a study of these neglected questions 
might seriously modify the views on the plays and the playwright based on 
the facts now available^ 

*‘IJA0S 40. 248-59] 

® A complete bibliography of the Uteratuie, Indian (induduig the works in 
vemacutos, of whidi there is a consideiabW number already) and European, bear- 
ing on the subject, will be the theme of a separate article 

4^1 L- Vasavadatta of Bhasa editdd with Introduction, Notes ate. 

’ B ?. Sansiknt Commentary (Hiavanagar, 1916). 

Acwrdmg to Ga^iapati SastbS the author of these dramas Rhsef , ‘must 

th„d or esntray >.c. ; 

to Mr. Bhide, 475 b.c. to 417 B.c. would be the penod of Bhasa, 



Nor have the critics^ of Gaijapati Sastri, who challenge his ascnpticai 
of the plays to Bhasa, attempted — ^perhaps they have not deemed it worth their 
while to attempt— to get below the surface ; their investigations confine them- 
selves to a very restricted field, upon the results of which their conclusions 
are based. Corresponding to the different isolated features of these plays 
selected by them, for emphasis, different values are obtained by them for the 
Qxxh of these dramas ; and havmg shown that these dates are incompatible 
with the probable age of Bhasa, these waters have considered their responsi- 
bility ended. 

Now whatever opinion may be held regarding the age of these plays it 
seems undeniable that they are worthy of very close study Thar discoverv 
has given rise to some complicated literary problemsk which demand elucida- 
tion Their Prakrit, which contains some noteworthy pamliarities, lequires 
analysis ; their tedimque, which differs in a marked manner from that of 
hitherto known dramas, requires careful study , that metre, with its pre- 
ponderance of the 41oka, and their Alathkara of restricted scope, both call 
for minute investigation. The fragment^ Carudatta alone, of which the 
MTcchakatika looks almost like an enlarged version, su^ests a whole liost 
of problems. Some verses (or parts of verses) from these dramas are 
met with, again in different literary works ; we find others referred to in 
critical wwks of different epodis • have they been borrowed or quoted (as 
the case may be) fronii our dramas? If so, what chronologi-{250}-cal 
ccMiclusions follow from these references’ Some of these questions have 
never been dealt with at all before , there are others whose treatment by pre- 
vious writers must be called superficial and unsatisfactory , but aH of them 
merit exhaustive investigation In these Studies I shall try to discuss various 
problems connected with these plays with all the breadth of treatment they 
require. I hope that they will in some measure answer the demand. 

At first I shall devote myself to collation of material ; subsequently, 
when I have a sufficient number of facts at ray disposal, duly tabulated and 
indexed, I shall turn my attention to the question, of the ag^ and the author- 
ship of these dramas, and consider whether, from the material available it is 
possible to deduce' any definite oaiclusions regardmg these topics. Prom the 
nature of the case it may not be possible to find for the question of the 
authorship an answer free from all demwits of uncertainty ; but it is lulled 
that the cumulative evidence of facts gleaned from a review of the plays 

^ Prof PaiTOEYA m the Vernacular periodical SSra|d3, (Vol. 1, No 1), who 
assigns the plays to the lOth century A.D. , and Dr L. D Barnett in JRAS, 1919, 
pp 233 f , who ascribes them to an anonymous poet of about the 7th century 
A D 

® Thereon see my articles ‘ “ Charudafta ” — A Fragment ’ in the Quorto'Jy 
Jourmd of th$ Mythic Society (Bangaloire) , 1919, 


from wxddy different angles will yield some positive result at least regardmg 
their age 

In conclusion it should be made clear that nothing la taken for granted 
regarding the author or the age of these plays It follows, therefore, that 
the choice of the title ‘ Studies m Bhasa,’ or the expression ‘ dramas of Bhasa ’ 
if used in the sequel with reference to them, doe^ not necessanly imply the 
acceptance of the authorship of BhSsa , the use of Bhasa’s name should be 
legarded merely as a matter of convenience, unless the evidence adduced be 
subsequently found to justify or necessitate the assumption involved 

I, On certain archaisms in the Prakrit of these dfamas 

The scope of this article, the first of the senes, is restricted to a con- 
sideration of certam selected words and grammatical forms, occurring in the 
Prakrit of the dramas before us, which arrest our attention by their archaic 
character. There are many other questions relative to the Praknt of thpRp 
plays which await investigation, such as, for example, its general sound- 
system, its varieties, its distnbution, etc they will be dealt with in subsequent 
articles ‘Archaic’ and ‘modem’ are of course relative terms The words 
noticed below are called ‘archaic’ in reference to what may be said to be 
the standard dialect-stage of thq Praknt of the {251} dramas of the cjagftjra i 
period, such as those of KShdSsa No comparative study has yet been 
made of the Prakrit of Kalidasa and his successors with a view to ascertain- 
ing the developmental differences (if any) obtaining between them ; marked 
differences there are none ; and we are constrained, in, the absence of detailed 
study, to regard the Prakrits of the post-KSlidfisa dramas as! static dialect- 
varieties showmg only mmute differences of vocabulary and style 

Methodologically the question whether all these thuteen anonymous 
plays are the works of one and the same author should have been faUm up 
first for investigation But even a cursory examination of these plays is 
enou^ to set at rest all doubts regarding the common authorship ; moreover 
the pomt has already been dealt with m a fairly satisfactory manner by the 
editor of the plays, whose conclusions have not hitherto evoked adverse 
comment The question will, however, in due course receive all the atten 
tion and scmtiny necessary. 

Meanwhile we will turn to the discussion of what I regard as archaisms 
in the Prakrit of these plays. ' 

An Alphabetical (List of Selected Archaisms. 

1. amhath ( = Skt. asmdkam) . 

Svapna, 27 (twice; Ceti), 28 '(CeH) ; Pafica. 21 (Vj-ddhagopalaka) ; Avi. 
25 (DhB.tri), 29 (Vidu^akaL 

Amhdftih is used in the passages just quoted ; but in other places the 
very same characters use the later form anthihfath, which is formed on the 



analogy of the thematic nommal bases : cf. CetS m Svapna. 24, 32 ; Vj-ddha- 
gopalaka in Pafica., 2021 ; and DhaW in Avi 23. The latter form occurs, 
moreover, m C&iu 1 (Sutradhiara) , 34 )(Cet5) The form amha{k)am. 
It may be remarked, is neither moitioned by grammarianss nor found in 
the dramas hitherto known. But PSli, it will be recalled, has still amhakam 
and Aivagho^a’s dramas (Lude5rs 8» 58) have preserved the corresponding 
tumih)ak{,am). Owmg to the simul-p523-taneous occurrence in our dramas 
of both the forms in the speech of one and the same character, we are not 
in a position to decide at this stage whether the amhaam of our tnanuscripts 
IS a genuinely archaic use of the word or whether there is a contamination 
here with the Ski. asmakam It may again be that the promiscuous use of 
the doublets pomts to a period of transition 

2 Root y/arh~ 

Svapna. 7 (XapasS) , Abhi 5 (Tara) . 

Twice) the root appears in Prakrit passages m these dramas with un- 
assimilated conjunct. Once asi a nominal base crM (iSvapna. 7)| and again 
as a verbum fimtum arhadi ’’ (Abhi. 5) . In the latto- case the editor con 
jecturaJly amends the readmg of the manuscripts to arihadi A priori the 
conjunct rh seems hardly admissible m a Prakrit dialect,® and one is tempt- 
ed to follow the editor of the dramas in regarding it as a mistake of the 
scribe. In the iSauraseni of later dramas an epenthetic i divides the con- 
jimct • arih- (Pischel § 140) Of this form wel have two mstances in our 
dramas : arihadi m Pratima 6 (AvaditikS) and amnhdni in Abhi. 15 
(Sita) . In another place, however, the word appears with an epenthetic 
; Abhi 60 (Sita) we have anaruhd^t (instead of anarik^i) in a passage 
which IS otherwise identical with Abhi. 15 quoted above. Thus, an em- 
endation would have seemed mevitable m the two isolated instances contain- 
ing the conjunct, had not the Turfan manuscnpts of A^vaghoea’s dramas, 
with which our manuscnpts will be shown to have a number of points in 
common, testified to the correctness of the reading, by furmshingl a probable 
mstance of the identical orthographic peculianty. In a passage from a 

® Thus, for instance, Markandeya in his PrSkrtasarvasva (ed Grantha- 
pradarsani, Vizagapatam, 1912), IX 95, lays down specifically that the gen plu. 
of the pets. pron. in Sauraseni is amham of amhaijum. 

Here and in sunilar references 'Luders’ stands for Ludebs, Bruchstucke 
Buddhisttseher Bremen (Kleinere Sansknt-Texte, Heft I, Berhn 1911). 

’’ The actual reading of the text is a(,rha'> riha)di, meaning apparently' that 
the MS readmg is arhadi and that the editor would amend to attkadi. 

8 See PiSQHEL, Grammafik d PrakritSprachen (Abbreviated in the sequel 
as ‘Pischel’), § 332. 

8 Pischel (§ 140) remarks that the Devanagaii and South Indian recensions 
of fialfimtnla and MSlavikS, and the PriyadarhlS, have aruhadi in Sauraseni ^ 
acoarding to him it is an incorrect use. 



Speech placed m the mouth either of the Courtesan or the VidOsaka (and 
therefore iSauraseni) occurs a word that is read by Prof. Lijders as arhessi 
(Luraais 49) Unfortunately the portion of the palm-leaf which contains 
the conjunct rk is chipped, and the reading, therefore, {[253} cannot claim tor 
itself absolute certainty. However, that may be. Prof, Ltjders ai^iears to 
have in his mind no doubt regarding the correctness of the reading adopted 
by him. Should this restoration be correct, we should have a precedent for 
our seenungly improbable reading It is not easy to eitplam satisfactorily 
the ongin of this anomaly We can only conjecture, as Prof. Luders does, 
that the conjunct rh was still pronounced without the svarabhakti, or was at 
any rate wntten^® m that manner. Assuming that our! reading of the word 
arh- m both sets of manuscripts is correct, this coincidoioe, which is a proof 
as positive as it is fortuitous of the affinity between our dramas and those 
of A§vaghosa, has an importance which cannot be overrated 

3. ahake ( = Skt aham ) . 

cam. 23 ( SakBra ) . 

Occurs in these dramas only once in the (Magadhi) passage just quoted. 
iSakBra uses only in two other places the nominative case of the pronoun of 
tile first person namdy 0am 12 (which is a verse), and 15 , m both these 
instances, however, ^ elsewhere in our dramas, occurs the ordinary Tatsama 
aha>k. The derivation of ahake is suffidently clear, and since in Sauraseni 
and Magadhi the mMhe- suffix -ka may? be retained unaltered ' (Pisched § 
598), the form is theoretically, at] any rate, perfectly regular. It has more- 
over the sanction of the grammarians, bemg specifically noticed in a Prakrit 
grammar, namdy the PTahtta-prakaSci (11 -9) of Vararnd., which is the 
oldest Prakrit grammar preserved (Pischel § 32) . In his paradigma of 
the 1st pers. proa. Pischel encloses this form in' square brackets, mdicat- 
ing therewith that there are no instances of its use in the available manus- 
cripts. Probably this view represents the actual state of things m Pischel’s 
time. It would be wrong on that account to regard its occurrence here as 
a pedantic use of a speculative form which is nothmg more than a gram- 
marian’s abstraction. For we now have in ASvaghosa’s dramas an authen- 
tic instance of the use of a still older form, ahakam, ini the ‘ dramatic ’ MSr 
gadhS of the Dust {[254} (Bosewicht) ; Luders 36 The ahake of these 
dramas and of Varamci stands midway and supplies the necessary connect- 
ing link between the ahakath of Agvaghoea and the hake, hagi^)^ of later 
grammarians and dramatists The legitimacy and archaism of ahake may , 
therefore, be t^sirded as suffidently established. Incidentally the correspond- 
ence with Vararuci i£( worthy of note The occasion for the use, in this 

10 It would be worth while atamining the Praknt inscriptions to ascatam 
whether they contain any instances of tius usage, and if so to determine its epochal 
and topographical limits. 



instance, of the stronger form dhake?^ instead of the usual aham, appears 
to be that the context requires an emphasis to be laid on the subject of the 
sentence . ahdke dava vafldde . . . ‘ Even 1“ have bem duped . . ’ 
The later forms hake, ha(.g)ie ooair neither in the preserved fragments of 
ASvag^m^a’s dramas m>r in our dramas, a fact which is worthy of remark. 

4 atna 

Svapna. 45 (VidCsaka), 80 (Padmavati) etc. ; Cara. 4 (NaiH), 20 (iSak- 
ara) ; etc. etc. 

An affirmative particle occurring very frequently in these dramas and 
used m all dialects ahke This word, which is met with also in the modem 
Dravidian dialects, where it has preasely the same sense, seems to have 
dropped out of the later Prakrit It need not on that account be set down 
as a late Dravidiarusm introduced into the manuscripts of our dramas by 
South Indian scribes, for its authenticity is sufficiently established by its 
occurrence in PSIi on the one hand and in the Turfan manuscnpts of ASva- 
^oea’s dramas on the other (LOders 46) 

5 kana ( = Skt. krtva). 

Svapna. 52 (Vidiuaaka), 63 (Vasavadatta) , 70 (PratihaiS) ; Pratijna; 10, 11, 
and 15 (Haiiisaka), 41, 45, and 50 (Vidlu^aka) , etc. etc. 

The regular iSauraseni form is kadm (Pkchel 581, 590). But Hema- 
candra (4-272) allows also karia. While this rule of the grammarian is 
confirmed by the sporadic occurrence of kart- [y) a in manuscripts, it is in- 
teresting to remark that it is met with also in a iSaurasena passage in ASva- 
ghosa’s dramas (LItoers 46). pSS} According to Pischel {KB 8. 140, 
quoted by Luders in Brucshtttcke Buddhistischer Dmmen, p 48, footnote 3) 
the use of karia is confined exclusively to the Nagaii and South Indian re- 
censions of l§akuntala. and Malavika. But its occurrence in the Turfan 
manuscripts of A§vagho§a’s dramas ^ows that it is a gienuindy archaic form 
and not a vagary of South Indian or Nagaii manuscripts. -Kadua does not 
occur in our dramas, nor m the preserved fragments of A^vagho^a’s dramas 
Tncidentfiny we may note our plays also fuinidi instances of the use of the 
parallel form gacchia (Skt. gatoa) of which the r^iular (later) iSauraseni 
form is gadua, see Cara. 1, etc. etc. 

6. kissa, kiSSa ( = Skt kasya ) 

An. 16 (Vidiusaka), 20 (Nalinifca), 71 and 73 (TidSfeaka) ; Pratima. 6 
(Sita) ; CSru. 24 (iSak&ra) . 

* The dialects are iSauraseni ikissa) and Magadlu {kiHa). Formally 
theso words represent the gmitive angular of the interrogative prmii^, but 
here as elsewhere they are used exclusively in the sense of the ablative kas 

[EWitorial note : The sufBx ka cannol; m my ojanion, have iMa meaning. 
Here it is very likdy pitying ( " poor unludcy I " ) ; or it may be sv 


mat- ‘ why ? ‘ wherefore ’ ’ . Neither of these words— in this stage of pho- 

netic development — occurs in the Prakrit of the grammarians and other dra- 
matists (with but (Mie exceptitwi), which have k%sa {Idsa) instead (Pischel 
§ 428). idssa occurs frequently m Pali, kUsa is used by the Dusta (‘Bose- 
whict’) m Mvaghcea’s dramas (Luders 36) ; in both tliese mstances tlie 
words have precisely the same sense as here Like ahake (above no 3), 
kissa (ktSSa) corresponds exactly to the theoretical predecessors of forms in 
use in the Prakrit of later dramas. Msa occurs once in these plays also : 
Svapna. 29 (Ceti) . 

Unless a period of transition be assumed, kissa would, appear to be the 
light form to use here. For, iS»sa may repr^ent the spurious correction of 
a learned transcriber , but were kisa {Ma) the original reading in all these 
places, it would be difficult to explain the deliberate substitution of an archaic 
kissa (kiisa) m its place. In other words I assume the principle of progrra 
sive correction, that is the tendency of successive generations of scnbes to 
modernise the Prakrit of older works so as to bring it in Ime with the de- 
velopment of the Prakiit of them own; times. Unless, therefore, as already 
lemaiked, it is assumed that the simultaneous use of the two forms be re- 
garded as indicaUng a period of transition, fussa {fdSSa) would appear to 
be the form proper to the dialect [256} of our dramas. In passing it may 
be pointed out that kksa (kiiia) cannot be arrived at by the Prakritization 
of any Sanskrit form, therefore a question of contamination does not rise 
in' this case. 

7. khu (= Skt. khalu). 

Svapna. 5 (Vasavadatta), 7 (TapaS), 11 (Padmavatl), 12 (Cetl) etc. etc. 

Written almost throughout without the doubling of the imtial. Now 
the rule deduced from an observation of the usage of manuscripts appears 
to be that after short vowels and after e and o (whidi then are ^ortened 
under thosel circumstances) we should have kkhu ; after long vowds, how- 
ever, khu (Pischel 94) . This rule ai^hes to tSaurasenl and kfagadbi alike. 
But in the manuscripts of A^vaghoga’s dramas- the initial is never doubled , 
and in our text of the present plays there ate only two instances of the doub- 
ling, both of which are spurious and due to mistakes of copyists. We will 
turn our attention to these first They are ; — (1) Abhi 23 (Sta) • aho 
aamfo-kklw issara,^ and (2) Pratimfi. 22 '(Sita‘) ' ijaih saha-dkattmut- 
ttriffi-kkhu aham. It is quite evident that the doublmg in these instances, 
whidi takes place after the long finals d and i, is contrary to every rule, and 
13 nothing more than a mistake of scune transcnber. It may therefore be 
assumed that at the stage m which the dialects of our dramas find them- 

^ But note Svapna, TSt (Vasavadatta) * aho akarunS khu issard Of course 
the retention of the intervocalic k is unjustifiable. 



selves the doubling of the initial in khu had not yet taken effect. "We notice 
here, however, the first step taken to its treatment as an enditic. In the 
dramas of ASvaghoea khu remains unaltered throu^out with undoubled ini- 
tial but m our dramas we find frequently hu substituted for it m the com- 
buiatioos na khu and Hm nu khu - Svapna 23 (VasavadattS) 58 (Vidu?a- 
ka), 63 (Vasavaddattfi) , etc. ; Pratijfia 9 (Haansaka) , Pafica. 20 (Vrddha- 
gopalaka); Avi 79 (Nalmikfi), 82 (Kurafi^), 92 (Nalinika) , etc. etc. 
Sporadically khu is retained unaltered even in these combinations.^* 

{257} 8. tava (=SkL taoa). 

Svapna. 17 (TBpasff), 40 (Pladmiavatii), 78 (Dhatri) ; Pratima. 8 (Avada- 
tika,)'; etc. etc. 

This IS the usual form of the word in our plays in all dialects alike ; in 
addition, of course, the old enclitic te {de) is also in use. The Sauraseni of 
Aivaghosa’s dramas furrashes also an example of its use in the Prakrit of 
dramas (Ludebs 46), and it is common enough in PBli On the other hand 
the later forms tu(m)ha, and tujjha are unknown alike tot the ^Prakrit of 
A^vaghoea and these plays Accordmg to Prakrit grammarians and the 
usage of the manuscripts of later dramas tu(m)ha (and not tava) is pr(^)er 
to iSauiaseni ; evidaitly this represents the state of thmgs at a later epoch. 
The use of tava seems later to be restnctsd to M&gadM, Ardbamagadhi, and 
Jaina MaharaBtri (Pischel 421) . 

9. tuva?h ( = Skt. tvam). 

Svapna, 37 (Pladm&vafi), 38 (Vasavadattfi), 53 (Padminikg.), 54 (Padmi- 

mka), 55 (Padminika), Pratijna, 40 (Vidu^aka), 42 (Vidusaka) ; Avi. 

72 (Vidusaka), 77 (Vidusaka), 79 (Kurafi^) , Cru 104 (Durjaya) ; 

Cam 2 (Nati) ; etc, etc 

This form, in which the assunilatbn has not yet taken effect, disappeared 
fiom the Fhraknt of later dramas, iwhich substitute tumont in its place. But 
it is mentioned by Prakrit grammarians (Pischel § 420), and it is the 
regular form of the nominative case of the 2ad pers. pron. in and 
inscripbonal Prakrit It was, moreover, in use still in A^aghosa’s time 
(Lumrs 46), which is significant from our viewpomt. The later form 
tumath occurs sporadically in our dramas also - Svapna. 78 (Dhatri) ; 
Pratijfia 58 (Bhalta and Gatrasevaka), 62 (Bhata) ; Avi, 29 (Vidusaka), 
92 (Vasumitia) . In respect to the references from the PratijfiS. (58, 62) 

Prof. LdSERS does ate °t- kkhu in AiSvaghoSa's dramas, but, as he himself 
points out, it IS far from certam that we have the partide khu before us (Luoxrs 
51, footnote 3) 

w For instance, font nu khu, Svaima. 63 (VSsavadattS) 

See Pischel § 421 for a discussion of the merits and use of the different 
Praknt eqiuvalents of Skt. taiut. 



it should be remarked that the manuscnpts upon which our text is based are 
just at this place defective, and full of mistakes ; consequently the readings 
adopted in the text cannot by any means be looked upon as certain. Twice 
tuoam IS used in the accusative case Cru 105 (Durjaya), Oaru. 71 
(Gaioifca) . pSS} But the usual form of the accusative case m our plays, 
as in later Prakrit, is tumam : e.g Svapna. 27 and 32 (Cet5) . 

10 dissa-, dtisa- ( = Skt.- drSya -) . 

Svapna. 70 (PratJhari), Avi. 22 (Nalmika), 70 (Vidiusaka), Pratijna. 58 
(Bhata) ; BSla ; 50 (Vrddhagopalaka) , Madhyama ; 4 (BtShmani) ; 
tJru. 101 (Gandharl), Abhi. 54 (Sltia) ; Cam 16 ('Sakara) , Pratiina. 
5 (Sta) ; etc. 

In the above instances we have the root-form dissa- On the other 
hand, in a number of other places the later form disa-, with the sunpliiica- 
Uon of the conjunct, has been used. The relation dtssa- : disa- is the ga'mf. 
as that of kissa : kisa discussed in paragraph 6 Accordmg to Pisckel 
dtssa- occurs in the ArdhamSgadhl of the Jama canon, but not in the dramas, 
which substitute disa- instead (Pischel § 541). This later form disc- is 
met wia in our dramas only m : Avi 28 (Vidusaka), 91 (Vaaumitra) ; 
Pratijoa. 54 (Vidusaka), Cam 16 (iSakara). It is worth noting that. in 
one instance (Cam. 16) the two forms occur on the page and are 

placed in the mouth of the same character (SakSia) The remarks made 
in paragraph 6 on the relation of the forms kissa : kisa are also applicable 
here. It is interesting to note that the passive base dissa- is in use not only 
in FSli, but alsol in AiSvaghoea’s dramas (Luissis 58) . 

11 vaath ( = Skt. vayam ) . 

Svapna. 31 (Vidusaka) ; Avi 93 (Vasumitra) ; Cam 49 (Vidusaka) 

In Svapna (p. 31) the word is spelt vayam ; but m conformity with 
the orthography of the manuscripts of our dramas, which ormt the intervoc- 
alic y, the reading vaadi should be adopted also m this instance Tlie fbrtn 
proper to iSauraseni, to which dialect all the above passages belong, is amhe 
(Pischel 419) . But it is mterestmg to note that Vararad (12. 25). and 
Markawjeya 70, accordmg to Pischel § 419, permit the use of va{y)-mh 
m Saura^i. And again in the dramas of ASvaghosa we do actually 
with an instance of the use of vayath m a dialect which is probably) Saura- 
senl (LtiDERS 58). The form amhe does not occur in the preserve! frag- 
ments of ASvaghosa’s dramas And in our plays it occurs, as far as my 

In ^ iiaradigma of the pronoun of the 2nd pern., PisaraL gives the form 
tuvam for the noan. and, act dug., but he endoses it in square brackets 

« It dwuld be reroaiked that, amh- ia the r^uJar base of the obhque 
<rt this pronoun, and that amhe, accqs., is regular ml all dialecte. 


t. introduction 


obaervaticKi goes, only three times : twice, curiously enough, in the sense of 
(the norm- j;259}-native case of)' the dual avam (Abhi.. 48 ; Pratima. 58), 
and once in the accusative” case (Pratim&. 35). «a(y)am may therefore 
be regarded as a form peculiar and proper to the older Piraknts. 


Above have been set forth a number of peculianties of vocabulary and 
grammar in which the Prakrit of our dramas differs from that of the dramas 
of Kalidasa and other classical playwrights. Every caie of these peculiari- 
ties is shared by the Prakrit of Aavagho^a’s dramas In. some mstances the 
archaic and the more modem form are used side by side m our dramas : e.g. 
amhaam and amhanam , tuvam and tumam ktssa and f^a, dtssa- and 
t}isa-; arh-, arih~ and aruh-. But m other instances the ardiaic forms are 
used to the exclusion oif the later forms : for example ahake' (later kaie), 
va{y)am (later amhe Norn. Plu.) tava (later tumha), karia (later kadua), 
and ama (obsolete). The absence of doubling of the imtial of the particle 
khu and e and d may be taken to indicate an epodi when the ^ortening of 
the final e and o had not yet taken effect Worthy of special note are the 
foims ahake and dma, which not only are unknown tot later Praknt, but arc 
not the regular tadbhavas of any Sandrrit words It should also be remem- 
bered that ohake and va{y)am (used in our plays practically to the escIuHon 
of hage and amhe respectively) are noticed m 'Vararud’s Prakrtaprakaca, 
whidi is believed to be the oldest Prakrit grammar ratant 

The affinities with AiSvagho^a’s Prakrit pomted out above have a bear- 
ing on the age of our dramas which will receive our attention in due course. 
Meanwhile it will sufifilce to note that these affinities go far to prove that 
below the accretion of ignorant mistakes and unauthorised corrections, for 
which the successive generations of scribes and ' diaskeuasts ’ shwtld be hdd 
responsible, there lies in the dramas before us a solid bedrock of ardiaic 
Prakrit, which is much older than any we know* from the dramas of the 
so-called classical period of Sansknt literature. 


Tha following notes are the result of an attempt to study intensively 
certain charactefnsttcs of the versification of the metrical portions of these 
dramas which seemmgly distmguish the latter from those of the works of 
the classical period, and which, moreover, appear to suggest points of contact 
with the epic literature The present mvestigation deals mainly with the 
metres and the metrical soleasms of Sanskrit passages. The analysis of the 
metres comprises, besides a review of the metres conducted with special 
reference to the pr^nderanoe of the iSloka, a tabular congiectus of the 
metres (arranged m the order of frequ^y) showmg the number of occur- 
lences of each according to the dramas m which they are foimd, and secondly, 
a list showmg specifically the distnbution of the verses in each metre in the 
several plays The section dealing with the solecisms has a twofold purpose : 
firstly, to ascertam their eicact number and nature, and secondly to discuss 
their significance Other aspects of versification, such as Alhteration, Rhyme, 
and Figures of Speech, will be considered m a separate article dealing with 
the Alatrhkfiras. 

Analysis of Metres 

Specifically, the verses^ in each metre occur in the several plays as 
follows : 

Sloka, Svapna. I. 2, 7, 10, 15 , IV. 5, 7-9 , V. 6-11 ; VI 3, 6. 7, 9, 

11- 14, 16-19 : [Pratijfia. 1. 1, 2, 7, 9, 10, 15-17 , II. 5^-7, 10, 11, 13 ; 

in. 3, 7-9 ; IV. 9, 11, 15. 16, 18. 20-22, 24-26 : Phnca. I. 2, 7, 8, 11, 
12, 15. 16, 24, 26, 32, 33, 35, 36. 41, 42, 44, 48-54 , II. 4, 6, 8, 12-14, 16, 
17. 19-21, 23, 25, 28, 34, 35-38. 41, 47-50, 52. 53, 55-59, 61-69, 71 ; III. 9, 

10. 13. 15, 17-21, 23-26 : Avi I. 4 ; II. 4, 10 ; IV. 7. 14 ; V. 3 ; VI. 3, 6-8, 

12- 14, 17, 22 : BSla. I. 3. 11-13, 15-17, 20, 25-27 ; II. 8. 9. 11, 13-19, 

25 ; III. 7—10, 12, £109]} 13, 16 ; IV. 10, 12 ; V. 14, 16-20 : Madhyama 
2, 7, 1^23, 28-31, 33-40, 42-45, 47, 49, 50 : Dutav. 1, 2, 7, 8, 16, 17, 20, 
25-27, 29-31', 33, 34, 36, 38, 43, 46,. 50, 55, 56 : Diutagh. 6, 7, 15, 17, 18, 
21. 24 ^26, 28, 29, 31, 32, 37-40, 42, 44, 481-50 • Kainja. 2, 7, 12, 25 ; tJru 
33, 37, 41-44, 46, 49, 50, 62, 64, 65 : Abhi. I. 3, 8, 12, 15,i 18-21, 23, 24 ; 

11. 3, 7, 12, 13, 15, 16, 18-20, 23, 24 , III. 5, 6, 8-11, 13-15, 18, 20, 22, 
24-26; IV 4, &-11, 14, 16, 19-22 ; V. 2, 5, 8-10, 12. 14, 17; VI. 8-10, 

♦ UAOS, 41-107-130], 

^ Prakrit verses are marked with an asteridc (*). 
a Li verse 5 of the second Act of the PratijiSL, b is defective. 




1 Slokar:, 

2 VasantatilakaT^t^ 

S Upaiati^ 

4 SSrqulavikridita 

5 MahnL 

6 Pu^itiigra 

7 Vamsastha^,. 

8 Salim 

9 Sikharinl 

10 Fraharsini^ 

11 Arya 

12 Sragdhara,.. . 

13 Harini 

14 VaiSvadevri® 

15 Suvadana? . 

16 Upagiti® 

17 Dandaka^ 


19 Drutavilambita. 

20 IM:hn 

21 BhujamgapraySta^^ 

22 Vaitaliya^^ ..4 o. 

23 » 




































































































































a «• 



• f • 



, , 






















■ .* 









4 « 




• v» 








• 4» 

■ *. 


, , 



• •• 






• •• 







, , 









, , 

i «f 




• •• 









• •• 




, , 



• ■• 





• •• 


»« « 








• •• 






















18, 20, 22, 23, 25-29, 35 : Caru. L 7, 19, 22, 24, 25, 27, 28 ; III^ 12, 14-17, 
19 ; IV. 2, 3, 5, 7 : Pratimfi. I. 4, 6, 9-13, 15-17, 19-21, 23, 24, 26-23, 
31 ; II, 3, 5, 6, 3-12, 15-18, 20 ; III, 4^, 8, 10, 12, 14, 16, 19, 20, 23, 24 ; 

8 Including Indravajra and UpendravajiS. Schema 
^ Schema . v — — v/ — v — 

8 Schema . — v-x-* — 

8 Sdiema : — 

^ Schema ■ 

3 X Schema : a and c 12 morae ; b and d 15 morse, 

® Schema • w/x^x>x-»v^vyH"7 amphimacere. 

18 ^Abbrewated Dai^daka’ (24 syllables); its sdiema : ^ v/ + 6 

amphimacers See below. 

11 Schema- ^ ; or four consecutive bacdiih 

12 See below, footnote 18 

13 Undetennuied Praknt metre. Its schema is . 

and C 13 naorae , b and d 14 morae), 


IV. 3-5, 11. 12. U, 15. 19, 26. 28 , V. 6, 8, 9, 12-15, 20-22 ; VI. 5, 9-11, 
13-15 ; VII. 5, 8, 13, 15. 

Vasantatilaka, Svapna. I. 4, 6, 11 ; IV. 2 ; V. 1-3 ; VI 2, 4, 5, 15 : 
Pratijfii. I. 4, 6 ; II 2. 9 ; III. 4 ; IV. 5. 7, 8 : Panca. I 18, 29, 34, 37, 
39 ; II. 27, 31. 42 ; III. 22 : Avi. I. 2, 6, 11 ; II 1, 2, 7, 13 ; III. 1. 7, 

8, 10. 12, 15-17, 19 ; IV. 1, 5, 8. 13, 18, 22 ; V 2, 7 ; VI. 1, 11, 19, : Mia 

1. 5, 8. 23 ; II. 1-4, 6, 7. 10. 21. 22 ; III. 2, 6. 14 ; IV. 6, S, 11, 13 , V. 1, 

3, 6. 8, 10, 11, 15 : Madhyama. 1. 3, 8, 11, 27, 48 : Dutav 3-5, 11-14, 
23, 41, 42. 44, 49, 54 : Duta^. 1, 5, 11, 14, 23. 35, 45, 52 : Kama 4, 6, 

9, 16, 21, 24 : tJru. 2, 3. 7, 9, 11, 12, 19, 22, 31, 32, 36, 40, 54, 59, 60, 66 : 
Abhi. I. 1, 4. 9, 11 ; III. 21, 27 ; IV 7, 13, 23 ; V. 4. 7, 13, 16 ; VI. 1, 7 : 
Cam. I. 2, 5, 8*, 9. 11. 18 ; III. 1, 2, 5, 10, 18 ; IV. 4 * Pratima^ I. 7, 8. 
22 ; II, 2, 4 ; IV -1. 2, 16, 22, 24 ; V. 10, 11 ; VI. 4, 6, 7, 12 ; VII. 4, 6, 7, 

Upajati (including Indravajra and Upendtavafra) , Svapna. V 5, 13 : 
Pratijfta I. 5, 12 ; II. 1 : IV. 3 ; Pafica I. 1, 10, 13, 19, 23, 27, 31, 40, 

43, 46, 47 : II. 9, 11, 30. 60. 70 ; III. 3, 12, 14 ; Avx. I 3, 9, 10 , II, 8, 9, 

12 ; III. 6, 18 ; IV. 2, 6, 15-17, 21 ; V. 1, 5 ; VI. 2„5, 10, 15, 16, 20, 21 : 
Bala. I. 2, 4, 7. 2P‘, 22, 24, 28 ; II. 5, 12. 20, 23. 24 ; III 4, 6 ; IV. 4. 5, 

9 ; V. 2, 7 ■ Madhyama. 9, 41, 51 ; Dutav, 9, 18, 19, 22, 28, 52, .53 ; 

Dutagh. 2, 9, 10, 16. 19, 30, 36 : Kama 13, 17i® : tJm. 30, 38, 45, 47. 48, 
55 : Abhi. I. 26 ; II 14 ; III. 3, 19 ; IV. 6 ; V. 1, 11 ; VI 14. 21, 32 : 

CBra. fllOJ I. 4, 10*«, 12*. 23* ; III. 3, 7 ; IV. 1 : Pratima. I 1, 29 ; 

III, 15 : IV. 9, 13, 25 ; V. 3-5 ; VI. 16 j VII. 3. 14. 

&SrdSlaoikndita, Svapna. I, 3, 8, 12 ; IV. 1 ; V. 4, 12 : Ptatijfii. I 
8 ; III. 5, 6 ; IV. 13, 17 : Panca, I. 4, 5, 9, 55 ; II. 26. 29, 39 , IIL 6, 7 . 

Avi. III. 3, 20 ; IV. 4, 10, 11 • Bala. I 1 ; III. 3 ; IV. 1, 7 : Madhyama. 

26 ; Dutav^ 24> 32 : Duta^. 3. 8, 12, 22, 27, 34, 41, 51 : Kaijia 10, 15 : 
Cm. 1, 4, 13-18, 21, 23-25, 28, 29, 34, 35, 51-53, 58, 63 : Abhi. I. 5 ; 
II 4, 6, 10. 22 ; III. 1 ; IV. 1, 2 ; V. 6 ; VI 3, 16. 19. 30, 31, 34 i Cam 

I. 6 ; III. 6, 8, 11, 13 ; PratimS. I. 3, 5 , II. 2, 19 ; IV. 23. 27 ; V 1. 16 ; 

VI. 3. 

Mdlim, Pratijfia. L 11, 14 ; II. 3 ; IV. 4, 14 : Pafica. I 38, 45 ; II 
5, 15. 45 ; III. 2, 4 : Avi. II. 5 ; III 2 ; IV. 9 : Bala I 9. lO ; III. 

F9da a of verse 21 of the first Act of the Bfila is a VaiMasitha Ime. 

FSda b of verse 17 of the Kama, ul a Vatbiastha line. 

F§da a of verse 10 of the first Act oS CSru is defective. Perhaps yrt have 
to read mibandhaanti instead of mubandhaanti of the text ; cf. the (Prakiitic) 
loss of the initial of adhi in verSs and that oi api in the compound (a)pihita 
(from api + dhS) even in dassical Sanskrit. Or better still, in view of the portion 
of the oesura, delete the final syllable hi of amehi and read mnh^ aiutbondkaantt, 
amhe being the shorter fonn of the Instr. Plu. , cf. Pischel, Grammatik d, Prakr*h 
Sptachen, 8415, 

versification of metrical portions 


II, 15 , IV. 3 , V. 12 : Maxihyama. 5, 6, 32, 46 : Dutav. 10. 35, 39, 40, 
45, 47, 48 • Datagah 43, 46 : Kainja. 1, 3, 14, 18-20 : tiru. 6, 20, 26, 27, 
39, 56, 57 : Abhi. I. 16, 25 ; II. 8, 9, 21, 26 ; IV. 15 , V. 15 , VI. 4, 6, 11 ; 
Cam. I. 13, 14, 17, 29 : PratuiH. I. 14, 25 ; III 9, 21 ; IV. 10, 21 , 

V. 7 ; VII. 1, 2, 12. 

Pu^pitdgrc, Svapna. I. 5 , VI. 1 : Pratijnla. II. 12 , IV 6, 10 . Panca 
I. 17, 30 ; II 35, 51 ; Avi. II. 11 III, 4, 9,' 11, 13 ; IV. 12, 20 ; V. 4 ; 

VI. 4, 9, 18 . B&la. I. 14 ; V 9 : Madhyama. 4, 24, 25 : Dutav. 6, 37 : 
Abhi. I. 6, 14, 22 ; II. 2, 5, 11, 17, 25 ; III. 2, 16, 23 ; IV, 3, 5, 12, 18 ; 
V. 3 , VI. 2, 12, 13, 17, 24, 33 . Cam I. 16, 20 : •Prating II. 21 , IV. 
18; V. 19; VI. 8. 

Vamiastha, Pratijns III. 2 ; IV. 19, 23 : Pafica. I. 20, 25 , II. 1, 
18, 32, 33. 43, 44 ; HI 1, 8, 11, 16 . Avi. IV. 23 : Bala. I 18 : Madhyama. 
10 . Dutav. 21 : Dutagh. 13, 33 : Kanja. 8, 11, 22, 23 : Gru. 8 : Abhi. 
I. 2 : 0am. I. 3, 15', 26 ; III. 4 : PthUma. III. 13 ; IV. 20 ; VI, 1, 2. 

• Sdkni, Svapna. I. 13 ; IV. 6 ; VI. 10 : Pratijna. I. 13, 18 ; II. 14 ; 

IV. 12 : Paaca. I. 22, 28 , II. 2, 10, 40, 46 : Avi. I. 7; III. 5 ; BSla. 
I. 29 : Dutagh. 20 : Abhi. I. 13 : Cara. Ill, 9 : Pratima. II. 13 ; III. 
IS ; V. 17. 

£111} Sikarttii, Svapna. I. 14, 16 : Pratijna. II. 4 : Paftca. I. 3, 14, 
21 ; II. 7, 22, 24 : Avi. I 5 ; II. 3 ; III. 14 ; Um. 61 • Abhi. IV. 17 : 
Pratima, II 14 ; III, 1, 2, 22 ; IV. 7. 

PraJuff^tn'tf Fahca. II. 3, 54 ; III. 5 : Avi. I. 8 ; IV. 3 : Bala. I. 6 ; 

V. 13 . Dutagh. 4 : Kanja. 5 : Abhi. I. 7, 10, 17 , III. 17 : Cam. IV. 
6 ; Pratima. I. 30 ; IV. 6 ; V. 18, 

Atrya, Svapna. I. 1 ; IV. 3, 4 : iPratijna. IV. 1* ; Bala. I. 19*; III. r ; 
V. 4" : cam. I, 1*. 21 : Pratimia. I 2 ; II. 7. 

Sragdhard, Avi. I. 1, 12 , IV. 19i : Bala IV, 2 : Dutav. 51 : Abhi. 

III. 7, 12 ; Pratima. IV. 17. 

Hanm, Svapna. VI. 8 : Dutagh. 47 : Cm. 5, 10 : Prahma. I, 18 ; 
III. 17 ; IV. 8 ; V. 2. 

V(divadevt, Svapna. I. 9 : Pratijna, I. 3 ; II. 8 : Abhi. II. 1 ; VI. 5. 
—Sui/adana, Pahca. I. 6 : Dutav. 15 : Pratima. III. 7, 11.' — Upagiti, Bala, 
V 5*. — Doifdaka, Avi. V. 6. — ‘Abbreviated’ Daitdaka?-'’, PratioB. III. 3. 
— Drutavilambita, Abhi. Ill 4. — Prthm, Avi, II. 6, — Bhufamgaprayata, 
Abhi. VI. 15. — V^tdUya^^, Ihatijfia III. 1*. — ? (Undermmed Prakrit metre), 
PratijiSa. IV. 2*. 

See p. 112 below. 

IS Read b as: ptdhn-^padedum uvcdfhiid) a. The Vaitiliya stanza shoidd 
have 14 morse m a and c, and 16 in b and d ; all the p3das, moreover, should end 



The lists given above supplement incidentally the data of the metrical 
coUeciaoDS of Stenzler, edited by Kuhnau, ZDMG 44. 1 ff., with the mate 
rial placed at our disposal through the discovery of this important group of 
dramas A comparison of our material with that brought together by Stenz- 
ler shows that, with the exception of what I have called above the ‘ abbre 
viated Daa?,daka’ of twenty-four syllables and an undetermmed Prakrit metre, 
the metres of these dramas are those of the classical poesy. 

In the Hmdu works on Sanskrit prosody we come across a group of 
metres which have this charactenstic in common that they, on analysis, are 
found to consist of six light syllables followed by a senes of amphimacers. 
The best known vanety is the {112} Daipdaka with its sub-classes, consist- 
ing of six light syllables followed by seven or more amphimacers.^® A well- 
known example is MSlaRmMhaoa, V. 23, which is a metre of 54 syllables 
consistmg of six light syllables and sixteen amphimacers. Metres of the 
same scheme consisting of less than twenty-seven syllables are not unknown 
and are cited by prosodists under different names.®® The shortest of tliese, 
formed of twelve syllables (six hght syllables and two amphimacefs®’^), is 
called Gauti m Pingala’s ChmdassuPra. According to the commentator 
Halayudha, there are between the Gauri and the shortest Daindakn (of 
twenty-seven syllables) four other metres formed by the successive addition 
of one ami^imacer, each having a special name. Piiigala mentions the name 
of only one of them, namely, the one which contains four amphimacers®® 
In the different manusenpts of the text and the commentary it is variously 
called VanamSl^ MahlamSlika, Naraca, etc. ; the names of the other three 
have not been handed down Now we have in our dramas an instance (Pra- 
tima. III. 3 : patitam iva sirak pttuff, etc ) of one of the unnamed metres 
referred to in Halayudha’s commentary It has twenty-four syllables con. 
sisting of six light syllables and six amphimaoers. This metre differs from 
the shortest Daiwjaka m contammg only one amphimacer less than the mini- 
mum number reqmsite ; I have accordmgly called it the ‘ abbreviated Dan- 
daka’. It may be noted that the verse ated above is the only instance 
hitherto discovered of this rare metre. Besides the ‘abbreviated Dapdaka’, 
our dramas include also an sample of the fuller form with twenty-seven 
syllables (Avi. V. 6). 

Among the fixed syllabic metres the Vasantatilaka and the Ulpajflti (in- 
cluding the Indravajra and Upendrava^ra) are the favourite metres of tlie 

m an amphimacer followed by an iambus The first part of c is defective, in that 
It measures only five morse instead of the six, which are necessary. Note that the 
close of all the four pSdas answers correctly the requirements of the defimtion 
Vide the Dmjdakas in Stenzler's collections, ZDMG 44. 1 £f. 

®» Pingala 7. 33 ff. (Weber, ISt vol. 8, pp 405 ff) and Pmgala 8 5 (-Webeh, 
I c p. 419), for whidi references I am inddited to Prof Franklin: Edgerton. 

®^ Schema • v 

®® Pingala 8. 17, and Halayudha (Weber, 1 c.) . 


author. Out of a total of 1092 verses (Sanskrit and Praknt) included m 
the dramas there are 179 Vasantatilakas” and 121 Upajatis.®* Among the 
metres of the Sanskrit verses, the five metres Bhujamgaprayata, the 24- 
syllable ‘Dai)4aka’, the 27-syllable Dandaka, Drutavilambita and Pfthvi 
{[1133 occur only once each. Worth noting is perhaps the fact that there 
aie no examples of these five metres in the preserved fragments of A4va- 
ghosa’s dramas^® ; for it shows at any rate that they did not figure very con- 
spicuously m them. 

A metre which deserves special mention is the Suvadana, one of the 
metres which these dramas have in common with the ASva^csa fragments 
Our list includes four iiKtances of this uncommon metre : two in the Pratima 
(III 7, 11) and one each m the Phlfica. (I. 6) and the Dutav (verse 15). 
The SuvadanB.®* (a metre of twenty syllables) differs from the Sragdhara 
(twenty-onfi syllables) only in its final foot ; the first fifteen syllables of 
both have the identical schema ; yet there are far fewer instances of the Suva- 
dana in Sanskrit literature than of the Sragdham Until the discovery of 
the fragments of ASvaghosa’s plays there was only one solitary example 
known of its use in a drama ; that was Mudrdraksasa IV. 16, which, by the 
way, was mistaken by Stenzler®^ for Sragdharfi But now we have besides 
quite a number of instances in AAvaghdsa’s dramas, to which Prof. Luders 
has drawn attention in his remarks on the versification of those plays. 

The Arya, which must on^nally have been a -Prakrit metre, and its 
varieties, are used very sparmgly by our author, though they figure so pro- 
minently in the MrcchakafikS and the dramas of KBhdBsa In our plays 
there are only eleven Aryas (of which five are Prakrit) and one (Prakrit) 
Upagjti Compare with this K&lidBsa’s Vtkramorv<^ which has as many 
as 31 ArySs out of a total of 163 verses, and the Mdlavikdgnimitra with 35 
Aryas out of a total of 96 verses. 

There are in this group of plays thirteen Praknt verses, of whicih five 
arc Aryas, one Upaglti, three UpajBtis, one Vaim&istha, a (defective) Vaita- 
Uya, and lastly an undetermined Praknt measure , the last may be only a 
piece of rhythmic prose. The versification of the Praknt verses does not call 
for any ^>ecial comment 

We shall now turn to the consideration of a unique feature of the versi- 
fication of these dramas, namely, the preponderance of the iSlcka The ana- 
lysis of the metres shows that out of 1092 verses which these dramas contain, 
436 are iSlokas • in other words the iSloka forms nearly forty per cent, of the 
total, which, it will be admitted, is a remarkaWe high proporbon. 

“ TTirhiHing one m Prakrit. vOf which three are m Prakrit 

®® LdsES^ Bruchstiicke Buddhisttcher Dramen, Berlin 1911. 

Da schema is: « — y, ^ ^ ^ ^ 

Kuhanan ZDMG 44. jiff. 




Indeed m many individual draniaa of this group the proportion nses still 
lugher : in some it is as high as fifty per cent, and in a few it is higher still 
In the Svapnavasavadatta there are 26, Slokas out of a total of 57 verses , 
m the DutaghaUrtkaca 22 out of 52 ; in the Paflcaratta 76 out of 152 ; and 
m the one-act play Madhyamavyayoga there are as many as 33 Slokas out 
of a total of 51 veises. Notably the pioportion of this metre is very low m 
the AvunSraka,®® where there are only 15 iSlokas out of a total of 97 verses 

It IS wdl known that works of the epic, Pura|ip.c, devotional, and iSfistric 
or didactic order formed the field par eccellence of the Sloka. The dramatists 
made use of this unpretentious metre rather sparingly , they must have found 
it too commonplace The later fixed syllabic metres with their sonorous and 
complicated rhythms were more suited to their flamboyant style. The greater 
the number of these in a play the greater the camatkfira, the greater the 
skill of the playwright For this reason, it seems to me, the simple iSloka 
epicus lost ground m the drama, where it must have once figured prommently 
in favour of the fancy metres The old Tnstubh of the vedic and epic lite- 
rature, however, maintained its popularity even in the classical period. A 
few figures are quoted to show the actual proportion, m different dramas, of 
the Slokas to the total number of verses*® Bhavabhfiti is the only dramatist 
of the classical penod who employs the Sloka on a large scale in two of the 
three plays attributed to hun Out of a total of 395 verses in the Mahavira- 
cania, 129 are Slokas ; while m the Uttarardmacarita the ratio is 89 : 253 ; 
the .Sloka thus forms about a third of the total number of verses in these 
dramas. This is the highest proportion reached in any one drama or a group 
of dramas by the same author, except the dramas which are the subject of 
these Studies. In the MdlaRmaihova the ratio drops to 14 • 224 In the 
plays of Klahdfisa the iSlokas are few and far between. For the Mdlavikagm- 
miira the figures are 17 *96 , for Sahmtald 36 '230 ; for the Vikramorvasi 
30 : 163. We may further compare the figures for other dramas. In the 
RatnavaH {[115]} the ratio is 9 . 85 , m the Nagananda 24 . 114 ; in the 
Mudrarak?c^a 22 - 163 ; m the VeiitSmihara 53 . 204 ; in the Prcdjadhacan- 
dtodaya 36 : 190 ; in the Mjcchakatikd 85 : 336 . in these dramas the 'Sloka 
thus forms on an average about 20-25 per cent, of the whole These figwds 
make abundantly clear that the prefefence for the Sloka is a featured of metri- 
cal technique in which our plays differ from aU dramas of the classical age. 

As to the structure of the 'Sloka it may be remarkedl that the posterior 
ipfida has invariably the duambic dose ; sometimes even at the sacrifice of 
giammar as in Pratinfi. III. 8 : pratimarh kim na prechase, where thd final 

In the other non-epic dramas of this) group the proportion is not so low • 
in Svapoa rt is 26 : 57 , PratijfSl, 29 . 67 ; Caru. 17 : 55 

The figures have been computed from the data of Srra^ZLER’s collec- 
tions, Joe. eft. They will be of course different for the different recendons and 

versific;ation of metrical portions 99 

is, as a matter of fact, a syllable anceps. The prior pada ends as a rule with 
the pathya foot v, _ _ 3 ; ; occasionally howesver it ends with one of the vipula 
forms Concemmg the vipulas the following particulars will be found to be 
of interest There is a complete absence of the fourth vipula, and comparative 
rarity of the second , noticeable is also a partiahty for the first vipula which 
is used about twice as frequently as the third, variety. In the tlurd vipula 
the caesura is without exception after the fifth syllable, whidi usually follows 

, The precedent foot of the firat vipuli is commonly or 

i V and only occasionally v _ of which latter, as is wdl known, the 

post-epic style has mcreaauigly fewer cases.®® 

The analysis given above shows that the iSloka of our drama is of the 
refined type, not different at all from the classical model. The pacentage 
of vipuB forms m these iSlokas is somewhat lower than m the d^ical Q>ica 
• like the Raghuvamia, Kumarasambhava, Kiratarjuniya and SiSupalavadha 
One reason for the low proportion may be the followmg In epic and lync 
poetry, where the iSlokas (whenever they form the runnmg metre of a whole 
adhyfiya or chapter) follow each other m scores and hundreds, the vipula 
foiros crept in mevitably and may even have been introduced as an agreeable 
change frcMn the monotonous rhythm of an immutable octosyllabic scheme. 
With the lirruted number of the iSlokas occurring in a drama it was compara- 
tively easier to produce a larger proportion of ‘good’ Slokas; moreover 
owing tol the intervening prose and the crinkling of fancy metres the need 
for variation was not as keenly felt. 

In connection with this predilection for the Sloka epicus I {116} may 
draw attention briefly here to certain passages mdmdualised by containing 
shorter or longer runs of Slokas. Herd the prose is unimportant, while the 
verses with fancy metres are mostly lyrical ; the Sloka is in these passages 
the dynamic element A typical instance is the section of the Madhyamavya- 
yoga from verse 12 to verse 45 This passage, containing 34 verses, includes 
as many as 28 Slokas, and only 6 fancy metres. Moreover, it will be noticed, 
the dialogue is carried on in simple unadorned Slokas, the contents of which 
are not at all lyrical but indude just what is necessary for the progress of the 
action of the drama The prose cannot be entirdy dispensed with, but it 
makes the distinct inyiression of being secondary in importance. Another 
such passage is Pafica. Act II from verse 47 to the end It includes 25 
verses of which as many as 21 are Sldras and only four fixed syllabic metres. 
A piece shorter still is PratimS Act I from verse 9 to verse 28, which indudes 
a group of 16 Slokas punctuated with 4 fancy metres These passa^ rather 
suggest to my mind rudimentary attempts at dramatisation which are not 
quite emandpated from the limitations of the epic prototype. 

Jacobi, Das Ramayana , pp 80 ff.; ISt . vd 17. 443 f. 



The following list of set phrases and conventional compansons (the 
number of which can easily be mcreased»i) borrowed by our author directly 
fiom the epics illustrates in a striking manner how deeply he is indebted to 
the epic sources for hisl inspiration 

(0 acirenaiva kilena, Pratima. IV 
26 c; with the variation sucnetfipi 
kalena, ibid. 26 a 

(ii) kampayann iva medinlm, Panca. 
II. 21 

[117] (iii) ^aktih kalantakopama, 
Abhi. VI. 8 

(iv) nayami Yamas&danam, Pratima 
V. 22 

(v) pra^datn kaitum arhasi, Pafica. 

n. 68 

(vi) madasalalit a g s ml nmttarnatangali- 
lah, Abhi. II. 9 ^ and, mattamatan- 
gablat, Abfaj. IV. 15 

(vii) samblKainotphidlalocana, Dutav. 
verse 7 , Oaru. IV. 3 

(viii) audienapi It^ena, Pratuna. IV. 
26 a 

32acirepaiva kalena, MBh. 9, 2. 58 ; 
Ram. 5 26 23 , 6 61. 20, etc 

kampayann iva medinlm, MBh. 2. 29. 
7 , 8 34, 58 , 9 18. 26, etc , R^. 
(Gorr ) 6 37. 101 ; Riam. 6. 56. 13 , 
67, 115 , and variations, MBh, 3. 78 
3,9 30 60 , Ram. (Gorr ) 3 62 31 , 
Ram. i 67. 13 Also compare such 
expressions asi nBdayann iva medinim, 
puiSyann iva medinlm, and daiayann 
iva medinim occurring m the epics 

saksat kal^takopamah, MBh, 3 157. 

50 , Ram 6 88. 2 ; Ram (Gorr ) 6. 
45. 19 Cf. also kalantakayomapa 
mab, MBh 3 22 SI , 27. 25 , 4 33. 
25 , (Gorr.) 3 3^. 5 , 6 49, etc. 

anayad Yamasadanam, MBh 6. 54. 81; 
7, 19. 15 , Bmi (Gorr ) 3 34 31 ; 
75 28. Compare also yiyasur Yama- 
sadanam, MBh. 1. 163. 10-; Ram 
(Gorr.) 6 57 23 

praaadam kartum arhasi, MBh. 9, 35, 
72 , Ram 4 8 19 , Ram, (Gorr ) 
2, 110 7, etc 

mattamatangagaminam, MBh 3 80. 14, 
277, 9 , Ram 2. 3 28 , R^ (Gorr ) 
6 37 61, etc. 

vismayophullalocan^, MBh 1. 136. 1 , 
13. 14 386, Ram. 7. 37. 3, 29, 

Ram*, (Gorr.) 4 63. 10, etc 

(See above the references under na i). 

And lastly (ix) with the following phrases from the bharatavakya 

imam api mahim krtsnam, m PratijSSi, Pafica., Avi , and Abhi , 
mahim dfitapatifinkam, in! Svapna., Bala., and Dutav. , 
iSjfi bhumuii pra^astu nah, Pratinfi ; 

Only such, passages have been enhsted below as occur in both the epics, 
and occur there very frequently. 

82 In this Ust MBh, refers to the Bombay edition of the Mahabhdrata, Rant, 
to the Bombay edition of the GtWRESio's edition is (Mmguidied fn>m 

the latter by the addition of Gorr, in parentheses. 

versifiqation Of metrical Portions 


cocqpare the hemistich from the MahSbbamta . 

ya imam prthivim fcrtsn^ ekacchatiSm pragasti ha.— MBh 12 321. ,134 

In conclusion I shall add a fevf words on the structure of the verses. 
The style of the author is notably simple andi vigorous. The lucidity of the 
verses is due as much to the absehce of long and comphcated compounds 
as to the arrangement of words and phrases chosen with due regard to the 
position of the caesura ; almost invariably the csesara falls at the end of a 
complete word. The half-verse is m general mdependent of the rest of the 
verse in sense ; but often it is connected with it syntactically. Inside the half- 
vetae the padas are sometimes even euphonically independent , for instance, 
Bala. II. 4 there is hiatus between a and b vigahyo ulkam, a phenomenon 
common m the epics®'* but rare m the {[118J works of the classical penod 
On the other hand metre requires the sandhi®* in PafLca. I. Id (a and b) : 
mitrdny acarytam^^ Without the sandhi we should have a superfluous syl- 
lable in a, and a metncally faulty line ; with the sandhi we have a perfect 
Upaiati line Pratima. IV. 24d, which commences with the enclitic tne, shows 
agam that c and d are to be treated as a sm^e sentence ; for, an accentless 
word cannot stand at the beginning of a pada any more than at the begin- 
ning of a sentence. Instances of the sacrifice of grammai are discussed in 
a separate section. Here it will suffice to draw attention to the rhythmic 
lengthening in anukarsa (Pafica. II 7) and the use of the uncommon p&r^tA 
(with tfie long final) in Svapna V. 12 and mauH m tJru verse 59 (see PW. 
s. v.’) ; the fonn par^i, it should be added, is not metrically conditioned. 
Similar lengthening of the stem-vowel is to Be observed in mya& (Pratima. 
I. 21), in the sense ‘destiny’, of which only the form with the short i is 
ated in the dictionaries.*** 

Metrical Solecisms (Sanskrit) 

The list of solecisms in the language of these dramas appended by 
Pandit Gai;iapati SAstrI to his edition of the Pratimgnataka (Trivandrum 
Santioit Series, No. XLII) is a contribution to literary history of which the 
full import appears not to have been generally realised. The significant thing 
is not the fact that some solecisms have been found m these dramas. Every 
Sanskrit work, I suppose, if submitted to a rigorous esamination. by a com- 
petent critic, will yield at least a few grammatical errors, whidi is not to be 
wondered at m view of the! history of the language and the intricades of its 
grammar. The mterest about the solecisms m our dramas lies principally 
in character and their number. I am iiersuaded that it will not be pos- 

See HbPKiN% The Great Epic of India , pp. 197 f. 

Seldom in the Rantayona. ,, , , . u., .v, 

8® Compare a very similar instance in MSlatimSdhava X. 1 (a and bj . v 

§arcmyS»y ace^titard. . ■ j l i«„j— 

«« To the word with the long final, a different meanmg is assigned by lexico- 




studies in bhasa 

sible to name a reputable author of the classical period whose work or works 
could be shown to contain a proportionate number of grammatical ‘ mistakes ’ 
of the same order as those about to be discussed 

{119} The first requisite m this connection was to ascertam exactly the 
points in which the language of these dramas differs from the literary Sans- 
knt of the classical period Admirable as the list prepared by the learned 
Pandit is, it seemed to me that it needed, for the purpose m view, revision 
and rearrangement in certain respects The list of Ganapati SAstr! mdudes, 
on the one hand, certam items which do not strictly belong there ; on the 
other hand, it omits certam others which have an important bearing on the 
subject. For instance, the Praknt examples, to which the rules of Pfinini’s 
grammar cannot be expected to apply, have been palpably misplaced. It 
seemed to me also best to separate the solecisma occurrmg m the verses, of 
which the form is fixed by the metre, from those occurrmg only m the prose 
passages, which are more liable to be mutilated in the course of transmission 
Again, certain details m the Pandit’s hst refei only to metrical®^ irregularities 
and have no connection with grammatical solecisms as such Lastly, catain 
positive solecisms, which were atplamed away by the editor in the footnotes 
of the text editions of the various dramas®® and therefore not considered at 
all subsequently, had to be added to the list. Through these additions and 
omissions a new list resulted This list, appended bdow, indudes only such 
metrical forms as offend against the hterary Sansknt as represented in the 
works of the dassical age. It may be added that the dramas contain a few 
more ur^ulanties m the non-metrical portions, which by their nature are 
not as certain and in their character not as important ; they will be dealt 
with later m another conpectioa. 

Few scholars, if any, will be pr^red to accept Pandit Ganapati SAstrI's 
chronological scheme m which a date is assigned to the author of these 
dramas prior to the period of Papini, for whom the now commonly accepted 
date is ca. 500 b c. The postenonty of these dramas with reference to the 
A^tadhySyi is, I may say, axiomatic. Taking our stand on this assumption 
we have to understand and explain the solecisms as best as we can. It has 
been surmised that when grammar has been sacrificed we have in the vast 
majority of cases to do with metrical necessity , obviously the corresponding 
correct forms would not otherwise have been found in other passages where 
metrical considerations {120} do not interfere. What has perhaps been 
lost sight of is that these sdedsins are not arbitrary, but that they belong 
to a well-defined dass of irregularities, irregularities which are common 
enough in certain branches of Sanskrit hteratuie, but which now, for the first 
time, have been shown to exist in the drama also. 

See PratanS IV. 24 , BSla. II. 4 , Abhi. VI. 30. 

»« See BSla. II. 11, and Svapna. V. 5. 


The category of works m vrtuch similar deviations have hitherto been 
met with are of the epic, Puranic and Sastnc order Tliese works are known 
to contain abundant mstances of ungrammatical and almost promiscuous use 
of the Atmanepada and Parasmaipada forms ; examples of irregular fpminmp 
participles, absolutives and a variety of other abnormahties like those met 
with m our dramas Such violations of (Sansknt) grammar are particularly 
common in the epics ; they have accordingly beoi regarded as forming ‘ epic 
Sanskrit’. The free use of the ‘epic’ solecisms in a drama is, as already 
observed, a new factor in our knowledge of the Hindu drama, and is parti- 
cularly worthy of our attention in connection with the theory concerning the 
part that epic recitatons have apparently played in the evolution of the 
Hindu drama, at least of its epic vanety.’® 

It is plain that our dramatist denves his authority for the use of the 
irregular forms fnxn epic usage Such being the case, the question naturally 
arises whether the author, in ererasing this licence, went so far as to invent 
new and spurious forms as occasion demanded them, or whether he had 
availed himsdf merdy of such solecisms as were sanctioned by ^c usage. 
The correspondence, if proved, would bring to a sharper focus the dependence 
of otir author upon the epic source As the following analysis will show, 
the solecisms of our dramas can indeed, with but insignificant deceptions, be 
spectficdly traced back to the Qiics Quotations from the epic sources have 
been added in order to facilitate reference and comparison. 

The solecisms have been arranged under the following heads : (i) Irrfr 
gular sandhi ; (ii) use of Atmanepada for Parasmaipada, and (lii) vice 
versa, (iv) change of conjugation, (v) irregular feminine participle; (vi) 
irregular absolutive ; (vii) simplex for the causative ; (viii) irregular com- 
pounds ; (is)' irregular syntactical combination ; and (x) anomalous forma- 

j[l2i3 List of Solecisms 
irregular Sandhi 
1 puMth ,+ iti'= puheti 

]fiayat@ltii kasya putreti. — Bala Act II Verse 11. 

Here metri causa the hiatus (hetweai a and i)l required by Skt gram- 
mar has been effaced. The emendation suggested by the editor, putro ’bkiit 
for putreti, is uncalled for. This is a clear case of ‘ epic ’ eandhi. Instances 
of the effacement of the hiatus effected by thd combination of the temmning 
final a with the follawing vowels are exceedingly common in epic Skt. , a 
common example is tatovMO ( = tatak + uvaca), quoted by Whitney, Sans- 

LimESS, Die SaiiWukas. Em Beitiag zur Gesdudite des mdischen Dramas, 
Sttzungsberichte d, konigl, preuss, Akffdetnie d, Wissenschaft^ 1916. 



ktit Grammar, § 176b ; for examples from the RomSyai^a, see Bohtlingk, 

‘ Bemerkenswerthes aus Rfimaiana’.*“ Cf. also no. 2 below. It should be 
noted that this solecism could not be an accidental shp ; it must be the result 
of a cooscioua effort It is needless to add that there are no examples of 
such a sandhi in the prose of the dramas. 

2. Avantydh + adhipate^ = Avantyadhipattlh 

smarSmy AvantySdhipateh sutayah — Svapna. V. 5. 

Here again we have a conscious efifacement of the hiatus between a and 
a The editor tries to circumvent the assumption of a ‘ mistake’ by explain- 
ing Avantyadhipati as a compound of Avmti+d+adkipati, evidently an un- 
satisfactory explanation. Instances of such effacement are exceedmgly com- 
mon in the epics and the earlier texts. See Whitney’s Sanskrit Grammar, 
§ 177b , Holtzmann" cites the mstances from the Mahdbhdrata and Boht- 
LiNcac from the Rimayana,^* which need not be r^noduced htfe This is 
the only instance in these dramas of the effacement of similar hiatus. 

£122} Use of Atmanepada for Phrasmaipada 
3. ^tmisye 

gamisye vibudhlavasam — Bala. V. 19. 

Metri causa the Atm. form is used in order to save a syllable, though, 
as is wdl known, in classical Skt the root g<m is used exdusivedy with 
Parasm. terminations , of course m prose passages where metrical consideia- 
ticwis do not interfere, the Parasm is regularly used by our author. The 
Parasm. form igamisyasi) occurs also in Madhyama. verse 47. In his list 
of Skt roots Whitney marlra gamsyate with E An epic example is 

Ram. 5. 56. 29 : gamisye yatra Vaidehi. 

4 garjase 

kith garjase bhujagato mama govtsendra. — ^Bala III 14. 

As in the preceding instance the Atm form is used metri causa here 
in order to secure a long final. In classical Skt. the root garj, when used as 
root of the first class, takes exclusively Parasm. terminations. PW. quotes 
a number of instances of the use of the middle pres. part, from the epics, 
but not any of the middle pres. ind. Where the pres. part, is used, the mid- 
dle pres. ind. could be used with equal justificatiai, if the necessity arose. 
I therefore eicplaui the solecism on the ground of epic usage. 

For four books of the Ramayana : Berichte d, phit -hist. Cl d konigl 
sSehs. Gesdk d. Wiss. 1887, p, 213. 

« See Holtzmann. Gramntatisches ms dm MdhSbharata, p. 4, 

BdHTLINtK. tyb rit 

versification or metrical portions 105 

5. draksyatei (Active) 

kathara aganitapurvaim draksyate taith narendrah.— Pratijfia I, 11. 

As in the foregoing instance the Atm is used in order to secure a long 
hnal ; in classical Skt. the future la formed exclusively with Plarasm termi- 
nations. Epic examples of the Atm future are 

Earn 1. 46. 13 ; bhratgraih! draksyase tatah, 

Ibid. 2. 6. 23 ‘ EAmath draksySmahe vayam, 

Nala 12. 93 : draksyase vigatajvaram. 

Other examples (cited in PW ) are : MBh 3. 14728 ; 13. 964 ; Hanv. 10735, 
and Ram 2 83. 8 , 3 42. 49 

6 prechase 

strigataim pircchase katham.— Pafica. II. 48. 

pratimSlm kiim na prcchase— Pratima. III. 8. 

In classical Skt the root pracch is ecdusively Parasm. ; the Atm. termi- 
nation is used here in order to have a long final. In £123} the first example 
the length is almost imperative for the sake of the compulsory diiambic close 
of the postenor pada of the Sloka ; in the second it is preferred, notwith- 
standing the fact that the final syllable of the p6da is a syllable anc^s. The 
medium is used only for metrical reasons, as se^ from Pafica. II. 6, which 
offers an example of the iParasm. prcchatt. PW. quotes numerous instances 
of the use of the Atm. from the q>ics, the Bhagavata Pur , and Manu The 
epic examples are 

MBh. 1. 1451 • karmasiddhim aprcchata, 

Ibid^ 3 2588 : Damayantim apprcchata ; 

also MBh 3. 12070 ; 13. 297 

7. bhraiyate 

daivapifimg{nyfid bhraSyate vardhate vfi. — Pratijfia. I. 3. 

This is either the third pers sing, of a root of the fourth dess, or a pas- 
sive form of the root The classical usage knows only bkraSyati and bhrom- 
iaie in the active sense. bhramSate could have been used without prejudice 
to the metre ^ As the form is not metncally fixed, it is difficult to say whether 
the author should be held responable for it ; apparently all three mss. of the 
drama agree in containmg the same reading bhraSyate. There is abundant 
authority m the Qiics for the form bhraSyate, whetho: regarded as active or 
passive. The epic examples are 

MBh. 3 . 603 ; yair naro bhraiyate Sriyah, 

Ibid. 3. 1048 : bhraiyate Sghram aiivaryat ; 

Ram. 3 45. 12 ‘ ye tikEtjam anuvartante bhraiyante saha tena te, 

Ilnd. 6. 76. 36 : Idth dd cabhraiyata svaraib. 


8. ruhyate 

kale kale chidyate ruhyate ca.— Bvapna. VI. 10. 

Here chidyate is jiassive ; but mihyate ( ‘ thrives. ’ ) should be active 
The Hacairai Skt. admits only rohati. Now thd Iwhole phrase chidyate ruh- 
yate ca is parallel to bhraiyate vardhate va, Pratijfia. I. 3. It seems to me 
therefore better to emend the text reading to rohate, for which PW cites 
Brhatsaihhita 54. 95 . rohate sasyam. But the pass, ruhyatd is quoted with 
the mark E against it in Whitney’s list of Skt roots and is therefore not 
absolutely inadmissible. Either form (yrukyate or rohate) is rqjugnant to 
classical usage ; and <rohati is unsuitable here for metrical reasons 

{124} 9. Srasyate 

katham apurusavakyaim §ro§yate siddhavakyah. — ^Pratijfia. I. 11, 

Metii causa for irosyati In classical SkL the root Sru is used exclu- 
sively with Parasm terminations ; but in the epics the Atm. forms are 
r emarkab ly common. The Parasm. form (.Srosyasi) occurs in Avi II. 5 
Epic examples of Atm are 

Ram. (Gcmt.) 5 23, 18 ■ Ramasya dhanusah iabdam & 09 yase ghora- 
nisvanam, i 

Ibid 5 69. 26 : na aiac chrosyase dhvanim. (Note that the final of 
Sto^yase is prosodically long here.) 

Other examples are : MBh. 9. 105, 107 ; 7. 2725 ; 13. 1119 ; 14. 424 ; 
Ram (Gorr.) 2. 120. 22 ; 5. 23. 18. 

Use of Phrasmaipada for Atmanepada 
10. dpTCcha (Imp. 2nd pers. sing.) 

aprccha putrakrtak^ harinan diumatmiS ca. — ^Pratima. V. 11. 

Metri causa for dprccJtasva, the only form possible m cl assical Skt. 
Even in the epics the only iE^rasm. form used is apparently the Imp. 2nd 
pers sing. The epic example quoted in PW. is 

MBh. 14. 403 . aprccha KumSrdflla gamanaih DvSrakairh prati. Svapna. 
16 dpTcchami occurs in a prose iiassage It is to be noted that the saitence 
containing this word rests on the authonty of one ms. only, and is not essen- 
tial to the context ; it may th«refare be corrected or deleted, as deemed ad- 

11. upalapsyati 

taaii hatva ka ihigialapsyati drath svair duskrtair jlvitam — Dutagh. 
verse 8. 

Jn daesical Skt the root upa -i- Ud>h is never used with any but Atm. 

versification of metrical portions 107 

terirunatiais The epics contain examples of Parasm The Mahablarata 
examples are 

MBh 7. 3070 : na te buddhivyabMcarani upalapsyanti Pan^avaJj, 

Ibid. 1. 1046 : tatha yad upalapsySnu 

12-14. pairisvaja, pamvajatt, pansvajami 

(a) gatjham panisvaja sakbe. — Avi. VI. 1. 

(b) djstar na tjpyati panisvajaliva angam — ^Avi III. 17. 

(c) putraith piteva ca pan^vajati prahp?lah — ^Avi. IV. 9. 

(d) panevajami g^itjhaim tv^m. — Bala II. 9 

(123} Examples a, b and d are metrically conditioned ; m example c 
the Parasm.’ appears to have been used on the analogy of the other forms. 
The presoit reading in example c is based on the authority of two mss. 
Compare example d with Madhyama. verse 22 ; parii^vajasva ggdhahi mam, 
where metre does not stand in the way of the Atm. form. 'Only epic ex- 
amples ard available for the use of Parasm. 

MBh. 4. 513 : pan^vajati FlafiiSh madhyamaih Piqidunandanam, 

Rim. 3. 38. 16 : Sita yaiu ca hrsIS parisvajet. 

Change of Conjugation" 

15-16 vijantt; vijantafy (pres, part.) 

snehSl lumpati pallavSn na ca punar vijanti yasySlm bhayS-t 
vijanto malaySnila api karair aspistabBladrumsl'— Ablu III. 1. 

Mjetn causa for classical vij(tyanti and vtjaytmtah, from tnj to fan or 
to oooi by fanning E^ic examples of the use of vtj as a root of the first 
or axth dass are 

Hariv. 13092 : vijanti baiavyajanaih, 

MBh. 7. 307 . jaloiatyarthhaiStaia vijantah ptujyagandhina. 

Irr^lar Feminine Partiaple 

17. rudanti- 

svairfisano Drupadarfijasutalm nidantatn. — ^Dutav. verse 12. 

The dassical form is rudad. But in the epics the form rudmtl is parti- 
cularly common, whenever metrical conditions call for it 
MBh. 2. 2249 : tatha bruvanfaim, kanipam rudantim ; 

R^. 2 40. 29 ’ ^ruve cagratah string rudantainSch mahfisvanah. 
Ibid.* 2. 40. 44 tatha mdantiih Kausalyam. 

Other examples are MBh 3. 2686 ; Ram. 2. 40. 29 ; 3, 51 42 ; 5, 26. 
42. « 

" This may be regarded as the use of the simplex for the causative. 


Irregular Absolutive 
18. grhya 

vyadhSmosraairh grhya capam karana — Dutagh verse 20. 

It is unthinkable that this form could be used by any poet of the 
classical period In the epics, however, it is regularly substituted {126} for 
glhltva whenever metre requires it See Whitney’s Sanskrit Grarttmar, 
§ 990a. Other irregular absolutives like this used m the epics are : arcya, 
IK^ya, u^ya, tyajya, plavya, etc. Of these grhya is the commonest Holtz- 
ftANN cites thuteen eiamples from the Mahabhiarata, adding that there are 
many more ; Bohtlingk (op at ) mentions nearly twenty examples from 
the MmiayaDa. 

Simplex for the Causative 
19 sravati 

fiarais channh m&rgah sravati dhanur ugtam 4arana(Km. — ^Panca. II. 22. 

In ejnc Skt. the simplex is frequoitly used for the causative stem 
Holtzmann (see Whitney’s Sanskrit Grammar, § 1041) mentions vetsyami 
(for vedayi$yam), veda (for vedaya), ramantl (for ramayantl), abhivMata 
(for abJUvadayata), cudita (for codtta), etc. I have not beai able to trace 
a specific use of sravati for siavayati, 

20. vimaktukama 

bhuyah paravyasanam etya vimoktukama. — Avi I. 6. 

Metri causa for vimocayitukdma. See the preceding. Spedfic use is 
not traceable elsewhere. 

Irregular Compounds 
21. sarvarapiah (Aca plu.) 
utaa.dayi§yann iva sarvatajnalji' — ^Mtav veirse 9 
Used irregularly for sarvardjan, though not conditioned metrically The 
reading is based apparently on the authority of three mss. The epics con- 
tain quite a craisiderable number of similar formations Thus, MBh. 4. 
527 Matsyarajnah ] ibid 1. 169 Matsyarajm , ibid. 9. 2756 Yak^arajHa; 
ibid. 14. 19917 Dharmctr’ajfia — Avi. p. 110 we have KMtrajne instead of the 
grammatically correct KaSnajaya. This must be set down as the error of a 
copyist, for we have in the very same play the correct compounds Sauvlra- 
rafena, and Sauviraraja-KdHrafau (Avi p. 11) , and there is nothing, as far 
as I can see, that can he added in justification of the use of an incorrect 
form in a prose passage**®. 

[Except that the language was, to this author, too much ^ liv ing thing 
to be oomprest in & grammarian’s straight-jacket. F. E.] 


£127} 22. vyudhoras- 

vyudhora vajramaddhyo gajavTsabhagatir lambapinairiisabahulj — Madh- 
yama. verse 26. 

Metn causa for vyudhoraska-, which is required accoidmg to FSo 6. 
A. 151, and found used ini Raghu 1. 13 and Kumara. 6. 51, also in the 
MBh. and Ram. But the MBh. supplies itself a precedent for the use of 
the unaugmented stem vyudhmas, cf MBh. 1. 2740. 4553. 

23. tulyadhanm- 

evam Ittos tuls^adharmo vanlanam — Svapna. VI. 10 

All three mss of the drama read tulyadharmo According to PBn. 5 4 
124 dharma at the end of a Bahuvrihi compound becomes dharman, a rule 
which IS strictly observed m classical Skt. But m e^cs dharman is used 
freely also m Tatpurusa compounds and, vice versa, dharma ui Bahuvnhi 
compounds. Holtzmann cites 

MBh. 12 483 iSjan viditadharmo 'si. 

The emendahon Udyadharmd suggested by the editor is uncalled for. 

Irregular Syntactical Combination 
24. Use of yadi iwith cet 

i§taim ced ekacittanlain yady) agnih s&dhayi^yati. — ^Avi IV. 7. 

This pleonasm (of which I have not seen any instances m classical Skt.) 
18 , I think, to be traced also to the Qiics, from which here are two imtances . 

Riam. 2. 48 19 : Kaikayya yadi ced rajyam ; 

MBh 1 4203 : yady aati ced dhanam sarvam. 

This combinatioa of yadi and cet recurs in a prose passage of anothei 
drama of this group (Pratijfia. p 70). And though the readmg of the text 
is based on the concordant readings of three mss., the combination seems 
har^ and hardly appropriate in prose. 

Anomalous Formations 

We shall now proceed to consider certain anomalous formations for 
which there seems to be neither grammatical justification nor literary autho- 

£128} pratydyati 

na pratytiyati kikarta. — ^Abhi II. 24. 

Ganapati SAstrI explains it as prati+d+ayati (from Rt ay to go). To 
me it seems to be merely a oonfusioo. between the simplex pratyeti and the 
causative pratydyayati ; or rather a haplological contraction of pratydyayali 
with the meanmg of the simplex A ^milar ungrammatical contraction 
appears to be the one to be discussed next 


26. samdivasitum 

T ^irikam abhyupaySini bandhusahitali Sitain sam^vasitum. Abhi. 
VI. 19. 

This is a clear case of a poet’s compromise between samaivasitum and 

The irregularity to be discussed next appears to be as arbitrary as the 
last two. 

27, Stem yudh as masc 
maharpavabhe yudhi nasayami, Svapna V. 13 

As the adjective mahdrnavabhe m this pSda shows, the author 
treats the word yudh as a masculine noun But it always appears as a fenun- 
me word in literature, and is quoted as such by lexicographers. 

In additicm to the above. Pandit Ganapati Sasto mentions three other 
rnetncal forms as irregular They are indeed irr^ular m so far that the 
formations are ungrammatical. But they appear to have been accepted in 
the hterary dialect as good San^it The Pandit objects to the Atm use 
of rusyate (Panca II. 45) The Parasm. occurs, as a matter of fact, m 
Panca. I. 38 and II 58, 67 in verse and m Madhyama p 18 in prose , 
moreover m Panca I. 38 the Phrasm. form is not metrically necessary. In 
spite of all this the Atm. form is not wrong Whitney cites it with E+in 
his list of San^t roots, and accordmg to Apte’s dictionary (s. v. r«^) the 
form ru?yate does occur, though ‘rarely’. It ia thus plain that it was a 
current form. The Atm of obhikankse (Pratijfla II 4) is common in the 
epics . but evai for the classical dialect, the dictionaries cite the root as 
Ubhayapadin. The imp. 2nd sing, umdmaya (PratimS IV 16=VII 7) is 
also included by the editor in his hst of solecisms. But namayati is cited 
by Whitney with the mark XJ S + , while PW. quotes both namayati and 
namayati, adding ‘ mit prapp angeblich nur namayati ’. 

{^129} Index of verses that have been shown to contain soleasms ** *** 
Svapna V 5, 13 ; VI. 10 
Pratijfia. I. 3, 11 
Pbfica. II. 22, 48 

Avi. I 6 : III. 17 ; IV 7, 8 ; VI. 1 
Bala II. 9, 11 ; III. 14 ; V. 19 
Madhyama. v. 26 
Dutav. w. 9, 12 

** It diould be noted that the soledsms occur not only in the dramas which 

derive their plot from the ejncs and the PuiSnas, but alsd in the dramas of which 
the plot is drawn from other sources. No soledsms have been found in Karjtja, 
Oth and C5tti_ 



Dvitagh. w. 8, 20 

Abhi. II. 24 ; III. 1 ; VI. 19 

Pratitna. III. 9 , V. 11 

Of the tvirenty-seven solecisms dealt with above, three (nos. 25, 26 and 
27) are anomalous and peculiar to these dramas; two (nos 19 and 20) 
belong to a class not unrepresented m the epics , but the r emainin g twenty- 
two were shown to be specifically traceable to the epics thensdves. Now of 
these twenty-two some may again be nothing more than instances of indivi- 
dual capnce , others may be the results of lapsus memonae, in other words, 
pure and simple blunders. But it would be, m my opimon, quite wrong to 
hold that they are all of a form purely arbitrary. And what is of moment 
IS that for the majority of them it (would be impossible to find authonty in 
classical works. It seems to me beyond all doubt certain that the author de- 
rives his sanction for their use from a class of works different from the 
dramas of the classical epoch , they mvolve the deliberate expose of a liberty 
which may justly be regarded as the prerogative of the rhapsodists. 

Here follows a list of solecisms selected fiom the above and arranged 
in the order correspondmg to the degree of certainty with which it can be 
said of them that they he outside the range of the license enjoyed by classical 
dramatists : the effacement of hiatus m. putreti and Awmtyadhtpateh ; the 
absolutive grhya , the Atmanepada of gamisye ; the compound sarvarajAah ; 
the Atmanepada of prcchase , the Parasamaipala of Spfccha partsvaja(ti), 
and parUvc^Smi ; and the fem. part rudat^in. 

C1301 I am not oblivious of the fact that the classical rule allowed the 
use of nma for masa, provided that metncal norm was observed ; but I am 
fully persuaded that no playwright of the classical age, who aspired not to 
pass for an ignoramus, wonJd, to such a degree, mdulge in a license which 
was little more than an unequivocal confession of inconqietence. If, there- 
fore, we attempted to find for our group of plays a place within the frame 
work of the classical drama, we should first have to account for this apparent 
reaction from the tradition of the classical drama implied by the occurrence 
of the solecisms pomted out above. 


The forgoing investigation leads to the inevitable- conclusion that the 
Sanskrit of the verses included in these dramas, which differs in certain 
mmute particulars from the Sanskrit of the classical drama, reflects a stage 
of literary development preceding the classical drama, which culminates in 
the works of KShdiasa and Bhavabhuti. But our conclusions regarding the 
Prakrit of these dramas, which formed the subject of the first Study, con- 
verged to the same point They revealed in an equally forciWe manner a 
stag^ of development of the Middle Indian dialects older than that preserved 



in the classical drama. While the Prakrit betrays its afiSnities with the 
Prakrit of the fragments of Alva^oi?a's dramas, the Sanskrit of the metrical 
portions of our plays is found to be linked with the language of the epics. 

1 will not venture to draw any defimte chronological conclusions legard- 
. ing the dramas from these divergences and affimties, nor attempt to account 
for them here I shall content myself for the present with having stated the 
facts of the case. 

Post-scnptum. It should have been made clear that the references to 
the Svapnavasavadatta follow the pagination and the text of the second 
edition of the play, Trivandrum 1915. 



* m 

The close correspondence between the anonymous fragment® Caru- 
dalta and the celebrated Mrcchakattka,^ attributed to King iSudraka, inevi- 
tably necessitates the assumption of a genetic relationship, and indisputably 
exdudes the possibihty of mdependent origin. 

It is commonly taken for granted* that the Carudatta is the original of 
the Mrcchakatika, a rdation which does not, however, necessarily and imme- 
diately follow from the terseness or brevity of one, nor from (what amounts 
to the same thing) the length and prolixity of the other ; for, m adaptation, 
abridgment is as common and natural a debermimng pnnciple as amplifica- 
tion.® In view of the mtrinsic importance of the question, it seemed, £60} 
therefore, desirable to undertake an unbiased and exhaustive investigation so 
as to remove (if possible) the haze of uncertainty surrounding the subject 

Only the resemblances between the two plays appear hitherto to have 
attracted any attention ;« the differences between them, are, however, equally 
remarkable and much more instructive A careful comparative study of the 
two versions produces hi^ly valuable text-critical results, which help further 
the understanding of the plays and throw unexpected light on the subject 
of our inquiry. 

Regarding their rdationship there are only two logical possibilities ; 
either, one of the plays has formed directly the basis of the other, or else 
both of them are to be traced to a common source. In the former case we 

* IJAOS 42 59-74]. 

^ A paper presented at the One Hundred Thirty-third Meeting (Baltimore, 
1921) of the Amer. Or Soc , under the title ' The Carudatta and the Mrcchakapka: 
their mutual relationship.’ 

® See thereon my article, ' “ Charudatta ” — A Fragment' in the Quarterly 
Journal of the Mythic Society (Bangalore), 1919 

" Ed N B Godbole, Bombay, 1896 

* For instance, Ganapati SAsxRi m the IntroductiOQ to his editions of the 
Svapnavasavadatia (p xxxvm), and the Carudatta (p. i), LindenAU, Bhasa- 
Siudten (Leipzig, 1918), p 11 , and Barnett (hesitatingly) Bulletin of the School 
of Oriental Studies, vol I, part III (1920), 35ff. 

® Some attempt has already been made in India to — .-.-it r *■ 

of the Carudatta , see, for instance, Rangaraiya Radi ', I ;» ». , 

(Bombay), 1916, and P V. KAi^e, ibid 1920, Bhattanatha Svamin, Indian Anti- 
quary, vd. 46, pp, ).B9S. 

® See particularly Gauapati SAstb!, Svapnavasavadattd, Introduction, 
pp. xxxviu-xIiL 



are called upon to answer the question, which of the two plays is the original ; 
in the latter, which of them is doser to the onginal. 

We cannot be too careful m deciding what is onginal and what is not 
The original may have beai concise and well-proportioned, and later clumsy 
attempts at improvement may have introduced digressions, tiresome repeti- 
tions and insipid elaborations ; on the other hand, the origmal may have 
been prolix and loose, and subsequent revision may have pruned away the 
ledundandes Again, one may feel justified in assuming that the inaccu- 
racies and inconsistencies of the onginal would be corrected m a later revised 
version ; but one must also readily concede that a popular dramatic text like 
the MTCchakatika, after it had been written down, during its migrations 
through centuries over such a vast territory as India, may have imdprgnnft 
occasional distortion and corruption 

Every change, however minute, presupposes a cause ; even the worst 
distortion was ushered m with the best of mtentions, and though it may not 
always be possible to trace a given change to its proper cause, iwe are safe in 
assummg that m a hmited number of favourable instances the mtrmsic 
character of the passages under consideration may spontaneously suggest the 
cause for the change, and readily supply a due to the relative pnonty and 
posteriority of two vanations In isolated { 61 } instances we could say no 
more than that the change m a certain direction appears more probable than 
a change m the contrary direction But the cumulative force of a suflScient 
number of analogous instances, all supporting one aspect of the question, 
would amply justify our giving precedence to that particular altonative and 
tieating it as a working hypothesis . The problem, therefore, before us is to 
collect such instances, m whidi the motive for the change is directly percep- 
tible and capable of objective venflcation. The cumulative effect of the 
mdications of these scattered traces should not fail to give us the correct 
perspective. This digression iwas necessary in order to erplain the metho- 
dology underlying the present mveshgahon 

The textual differences between the two versions comprise a lar^ 
of details of varying importance The sdection presented below, though con- 
ditioned on the one hand by the reqmrements of the present inquiry, is by 
no means exhaustive ; for lack of space, only a few typical pxnmplPB have 
been singed out for discussion 

A Seubction of Significant Textual Differences 

We shall now proceed to a discussion of the textual variations roughly 
classified here under four headmgs : 1 . Techmque ; 2 . Prakrit , 3 . Vercifida- 
tion ; and 4 Dramatic incident. 

1. Technique. 

la pomt of techmque the Carudatta differs from the Mrcchakatiha (as 
from other classical dramas) in two strikmg particulars. In IhTfirst place, 



the usual nandi is missing, m both the available manuscripts of the Caru- 
datta , in the second place, there is no reference to the name of the author 
or the play in the sthapania, which does not contain even the usual address 
to the audience 

The Mrcchakattka, as is well known, begins with two benedictory verses ; 
the name of the play is announced in the opening words of the sutradhSra ; 
then follow five verses which allude to the play, the playwright,^ and other 
details not directly connected with the action 

£623 Elsewhere^ I have tried to show that the Carudatta is a fragment 
I hold, accordingly, that we should not be justified m basing our conclusions 
regardmg the technique of termination on the data of the fragment preserved. 

Worth notmg appears to be the fact that m the stage directions of the 
Carudatta. the hero is never called by his name or his rank, but merely by 
tlie character of the r61e he plays, nayaka Professor Luders® has already 
drawn attention to two other instances of this usage (if it may be called a 
usage), namely, a drama bdooging to the Turfan fragments, and the play 
Ndgananda attnbuted to Harsa a?rof. Luders sees in it an archaism inten- 
tonally copied by the author of the Ndgdnanda At present we can, it 
seems to me, do nothing more than record this third instance of its occurrence 
in a play of uncertain age and authorship. 

2. Pmkrit 

In the first article of this semes, it was shown in a general way that the 
Prakrit of the whole group of plays under consideration was more archaic 
than the Praknt of the dassical plays.^® This statement holds good also m 
tlie particular case of the Carudatta and the MfcchakaUka . A comparison 
of parallel passages m the two plays shows that the Mrcchakattka mvariably 
contains Middle-Prakrit^^ forms in place of the Old-Prakrit forms of the 
Carudatta Here are the examples. 

The Absolutive of the roots gam and fef. C§ru has the Old-Prakrit 
gacchia and kana {kcdta) : Mtccha gadua and kadua Cf. m particular 
Cam. 1 geham gacckta jandmi with the correspruiding passage, ^%ccha. 7 
gehaih gadua jandmt. The form gadua, which never occurs in the Cam, is 
used uniformly in the Mrccha — ^For the absolute of kr kairio f 63J (iSauia 

r The verses in the prologue \duch refer to the death of the alleged author 
. are palpably later additions. This self-evident fact does not however, necessanlj 
justify the assumption that there was) no reference whatsoever to the author ui the 
prologue of the original draft. 

® See above, footnote 2. 

® Bruchstiicke BuddMstischer Dramen (\Klemere Sanskrit-Texte, Heft 1), 
Berlin, 1911, p. 26 

i® Above, vol 40, pp 248ff 

12 See above, vd. 40, p. 254, 

11 lyUPSRS, op. dt., p. 62. 



sen!) Ciaru 46, ka\ia (Magadhi) Cam 23 : kadua (Igauraseni and Magadhi) 
Mrcdia. 53, 212, 213, etc In the CSm kadua never occurs; conversely 
karia is never met with in the Mrccha. 

Pronoun of the 1st Person, nom. sing. CBm. 23 we have the Old- 
MBgadhi otefec” (hut never hage or hagge): Mrccha. (passim) hagig)e 
(but never ahakd). Noteworthy is the following correspondence Cam I. 
12c aham tumam ganhia Mrccha I. 29c ese hage genhia — Nom plu Cam. 
49 has the Old-Praknt vaam Mrccha (passim) amhe The form amhe 
(noro plu.) IS never met with in the CBm, and conversely vaam never 
occurs in the Mrccha. 

' Pronoun of the 2nd Person ; nom smg Cara (passim) we have Old 
Prakrit tuvam Mrccha (passim) tumam. Cf especially C&ni 34 kim 
tuvam. etc, with the corresponding passage Mrccha 79 haiije tumam mae 
saha, etc.' — (Sen sing Cam muformly tava '■* Mrccha sometimes tuha Cf 
in particular Cam 25 tava geham pavittha with Mrccha. 59 tuha geham 

The Neuter plu of nom and aca of thematic stons ends m the CBm. 
invariably in -ot» {-dm in the A^vaghoea fragments) . m the Mrccha. it ends 
in -anh 

Treatment of the assimilated conjunct Retained m Cam. 16 diisadi^’’ 
(as in the Turfan fragments) ; simplified in Mfccha. {64} 41 disanli. The 
root-form diSs- (dtss-) is never met with in the Miccha, which shows uni- 
formly dii- (dis-). 

Vocabulary. CBm uniformly geha (Skt. grha) . Mrccha 39 ghala. Cf 
especially Cam 16 edam tassa geham with Mrccha 39 vamado taiia ghcdcm. 
— ^The Old IPraknt affirmative partide dma,'^^ which occurs m Pali and the 
Turfan fragments and which figures so conspicuoudy ifi Cam (eg pp. 4, 
20, 64, etc.), is never met with in the Mrccha. — ^Therd is one other thing 

“ See above, vcJ. 40, p. 253 Dr Truman Michelson has drawn my attention 
to an article of his (.Indo&jmamsche Farschutigen, vol 23, p. 129) m which he 
points out that the Mlgadlu ahake occurs several times m the Devanagaii recen- 
sion of the SakuntalS. The paragraidi on this word in my article ated above needs 
modifkation m view of this fact. The statanent that ahake is archaic is none the 
less correct 

See above, vol. 40, p. 258 

See abowe, vd. 40, p, 257. In the references under no. 9 the last item 
‘ Cam. 2 (Natl) ’ is a misltake. Here tuvam is used for the acc. smg , and not for 
the nom. smg. as imphed Accorduigly, on the same page, in 1 6 frmn bottom, 
read * thnce ’ instead of ‘ twice,* and add this instance, ' CBru. instances of tuvam 
(w>m ang.) are Canu 34 (GanM), 47 (Ce^), etc, 

1* See above, vol 40, p 257. 

See above vol 40, p 258. — The form dii-, with the conjunct, is 

met with on the same page (CSm. 16), spoken by the same character Sahara, 

« See above^ vol. 40, p. 254, 


to be noted about the difference in the vocabulary of the two versions. While 
the Mrccha contains a number of DeiS words (not found in the Cam ) , the 
vocabulary of the Cam. consists notably of pure tatsamas and tadbhavas 
Here follow some of the DeS words which occur in the Mrccha Mrccha 17 
ch%via, ‘having touched,’ from root chiv (Hem. 4. 182) with the reflexes 
in the Tertiary Pkts., Hindi chvm, Marathi kvan&. ‘ to touch ’ ; Mrccha. 
104 dhakkehi, ‘shut,’ from dhakkai, dhakket, traced! by Pischel {Gramma- 
Uk 221) to a root *sthak, with reflexes m the Tertiary Pkts , Hindi dhdknd, 
Marathi dhdkne, ‘to cover’ ; Mrccha 134 uddhehi, ‘open,’ for which in the 
correspondmg passage of the C&m. (p. 19) we have a tadbhava of the root 
apd + and which for that reason is particularly worthy of note , Mrccha 
207 ksratta-didini, ‘malevolent ogress’ (cf Marathi karatd, a term of abuse, 
and dSkiff, ‘ogress’) 

3 VersifkatKm. 

In the verses common to the two plays the Mrcchakattka almost always 
offers better readings, of which a few are cited below 

For C&m I 3 b yathdndhakardd iva dipad arianam, we have Mrccha. I 
10 b, ghanandhakaresv iva, etc , m which ghana- is substituted for the tauto- 
logous yatha 

Similarly, instead of the Prakrit Ime lOaru. I. 10 b jahd iigS^ vut kukku- 
leht, containing the same fault, we have Mrccha. I. 28 b v(me sidli via kukku- 
lehitii, m which vme takes the place of jahd 

f653 For C&ra. I. 3 c yo yati da&am dandratdm, we have Mrccha I. 
10 c yo ySti neno dartdvatdm. It is correct to say dasdm dandrdm, but 
d(^am daridratdm is clumsy, to say the least 

Oaiu I. 23a begins m M vd&u\ instead, we have Mrccha I. 41a 
CSC ii vasu. The si which takes the place of ht eliminates the expletive hi, 
and adds moreover another sibilant to the row of alliteiating syllables In 
the same verse, for kiijaht kmdaht of the Cam , we have akkosa vikkoSa in 
the Mrccha, which serves better the purpose of the anupr&sa, the dommating 
alamkiara of this verse. Similarly m d, instead of mahesSalam of the Cam , 
we have ^ambhum itvam m the Mrccha., which latter reading contains an 
additional sibilant as well as a pleonasra.^” These are nunor details, but they 
all tend in the same direction 

For 0am. I 25 a akdmd hriyate 'smabhth, we have Mrccha. I 44 a 
sakamdrwisyate 'smdbhihi The reason for the change is not obvious, as in 

The text reading is avdvuda, imp. 2nd sing, which is evidently inCoirect 
What the correct form should be I am unable to say. The mitial letters avavu of 
the word diow urumstakably that the root is apa + vr. 

20 Arwriling to T.a11a DUcduta, commentator of the Mrcchakatika : vyarthai- 
kartham apdrtham bhavati ht vacatutrtt iakarasya (Mrccha 28) . 


the foregoing instances But a closer examination of the context will show 
that the reading of the Mrccha marks a distmct improvement, m so far as 
it implies a more nunute analysis of character In the Caru. the ingenuous 
Vita inculpates sakara and himself by admittmg that they were engaged in 
carrying away forcibly an unwilling maiden. In the Mrccha the artful Vita, 
leadily mventmg a plausible he and explammg that they were followmg a 
girl who was willing, offers undoubtedly a much better excuse 

Cam I 29a describes the moon as khnnakhmjurapaii4u, ‘pale as the 
moistened fmit of the date ’ Mrccha. I 57 a has kaminlgcmdapandu, ‘ pale 
as a maiden’s cheek' The former is onginal and naive, the latter polished 
but hackneyed ; the latter harmomzes better with the sentiment of §rng^a 
which pervades the last scene of the first act, and is more! in keeping with 
the tradition of the later aiervated rasa theory. 

For Cam. III. 3 d vv^a^koilva nimajjamdnd, ‘ like the tip of a tusk 
sinking m the wata ’, the Mrccha. (III. 7 d) has tik'itiam visanagram wava- 
bi^otn, ‘ hbe the sharp tip of a tusk that alone remains visible As far as 
the sense goes there is not much to choose between them ; but the Ime from 
the Cam. {66} contains one serious defect In classical Skt. the root nt-majj 
is used exclusively with Paras, terirunations , ntmajjamana is, in other words, 
nothing less than a gross grammatical blunder.*^ 

With Cam. III. 6 b iauryam na karkasyata, cf. Mrccha. Ill 12 b caur- 
yam na iaurayam hi tat. karkasyata of the Cam is an anomalous word, 
being a double abstract formation The Mrccha eliminates this anomaly 
by substituting instead caurya, whidi, incidentally, rhymes with the succeed- 
ing smrya. 

These few instances'^^ must sufSce to illustiate the statement made above, 
that the Mrccha. verses are largely free from the flaws of the corresponding 
verses of the Cam. It should, however, be remarked that in. a vast number 
of cases it is not possible to assign an adequate reason for the change the 
difierent readings appear to be just arbitrary vanations. 

4, Dramatic Incident 

The Mfcchakafika shows a marked improvonent m the selection and 
arrangement of the inadents of the action 

The action of die CSrudatta bepns with a soliloquy of the Vidusaka 
followed by a lengthy dialogue between the Nayaka and the Vidusaka The 
hero is craiversmg with his friend, deploring his poverty This dialogue is 
brought to an abrupt end by the scene introducing Vasantasenh, who appears 
on the street outeide piUrsued by the iSakSra and the Vita (Cam 10). 

*1 Siimlar soleaslms, met with in other dramas of this group, are discussed 
by me in the second aitide of the senes (dbove, vol. 41, pp. 121 ff,). 

It may be remarked that there are no verses in the second act of the Com- 
iatta, and only seven m the fourth act. 



In the Mjcchokctika (p, 25) the abruptness of the chsnge of scene is 
skilfully avoided by the addition of the following words placed m the mouth 
of Oarudatta ; 

bhavatu | tistha tavat ] aheoh samadhtm nivartaydmi, 

‘ Very well. Wait awhile and I will finish my meditation ’ 

These words of CSrudatta serve adnurahly to adjust the tune relation of the 
different events. The playwright here unmistakably indicates that the suc- 
ceeding scene, which introduces the offers of love by iSakara, their indignant 
lejection by Vasantaseni., and her subsequent escape, devdops during Caru- 
datta’s £673 sam&dhi. Furthennore, as indicated by the subsequent words 
'of CSmdatta (Mrcdia. 43) vayasya samdptajapo ’smi, ‘Fkiend, my medi- 
tation is over’, Vasantasena’s reaching the door of Carudatta’s house co- 
incides exactly m point of time with the emergence of Cfirudatta from his 
sam&dhi. The words of CSrudatta quoted above, which serve to link together 
these various groups of inddents, are missing in the CSfudatta. 

Heie IS another example In the fourth act of the Carudatta (p. 72), 
Sajjalaka comes to the house of the Ganik^ to buy Madamka’s freedom. He 
stands outside the house and calls out for Madanifea Madarnkfi, who is 
waitmg on the herome, hears him and, seemg that her mistress is musing 
on other things, slips away and joins Sajjalaka. The defect of this arrange- 
mait IS obvious • it is inconsistent and illogical. With stolen goods m his 
possession Sajjalaka sneaks to the house of the herome with the object of 
secretly handing over the spoils of his theft to MadanM.. Under these 
circumstances it is the height of indiscretion to stand outside the house of 
the heroine and shout for his mistress at the top of his voice. Agam, if 
MadanikiL is able to hear Sajjalaka, so shcnild Vasantasoifi, who is sitting 
close by, be able to hear him. Apparently she fails to do so owing to her 
preoccupation ; but this is a drcumstance that could not have been foreseen 
even by a scientifio burglar like Sajjalaka The situation in the Mrcchnka^ 
tika (p. 169) is much more realistic On reaching Vasantasena’s house, 
Sarvilaka, instead of calling out for Madanikfi, hangs about outside the house 
waiting his opportunity The meeting of the lovers is brou^t about in the 
following manner. Soon after Sarvalika reaches the house of Vasantasena, 
the latter sends away Madanika on an errand ; on her way back, Madanika 
is discovered by iSstrvilaka, whom she thereupon naturally joins 

One more instance, which is the last A time analysis of the first three 
acts of the Cwrudatta will show that the mcidents developed in these acts 
are supposed to take place on three consecutive days, the sixth, seventh and 
eighth of a certain lunar fortni^t Here are the specific references Caru- 
datta 7, Vidusaka, in speakmg of the NByaka, applies the adjective saftkl- 
Jadadevakayyo to hiin, which inddentally shows that that day was the sixth. 
Later on in the same act (CSru. 30), addressing the Ceti, the Vidueaka says ; 



^;683 saUhte sattamie a dharehi | aharit attamU cnfaddhde dharcAsath. 

The anangement he proposes is that the Ceti should guard the jelwds of the 
Ganika on the sixth and the seventh, and that he should take over the charge 
of them on the eighth In the third act we have a ccMifirmation of the same 
airangement CSru 53, CeH remarks : 

lam suvanmbkandam satthie sattande (parivetthdmi?) \ atthann khu ajja. 

The Ceti, appearing before the Vidu§aka, with the jewels, on the night of the 
eighth, points out that she has guarded them on the sixth and the seventh, 
and adds that that day bang the aghth it is the turn of the VidQsaka. Later 
on in the same act (C&m 65), the BrShmani, the hero’s wife, incidentally 
mentions that she was observing on that day the Fast of the Sixth,®® to which 
the Vidu§aka pomtedly retorts that that day was the eighth and not the 
sixth.®* These various refermces leave no doubt that the events that form 
the action of the first three acts are supposed to take place within the span 
of three consecutive days. 

There are in the plays some further dironolo^cal data, which we must 
also take into consideration. They comprise two lyrical stanzas which des- 
cribe respectively the rising and the setting of the moon. In that elegant 
verse (C&ru. I 29) banning with 

udttyati ht sasaAkah khnnakharjurapaifduh 

the moon is described as rising, late m the evening, after the lapse of a short 
period of darkness following upon sunset, during which Vasantasena escapes 
from the clutches of the evil iSakara. In the third act, on his way home from 
the concert, CSrudatta, in a lyrical mood, redtes another vra-se (C@ru. Ill, 
3) , beginning with 

asau ht dattv& ttmravakaSom 
astmh gala hy ostamapak?acemdrah,^'^ 
and having for its theme the setting moon. 

^693 This is the chronological malarial of the CSrudatta. Let us turn 
for a monoent to the Mrcchakattka and examine its data Here also appar- 
ently the same conditions prevail. Apparently the events of the first three 
acts take place ai three cmisecutive days, but only aj^arently sa There is 
notl^ in the play itself from whidi the duratKxi of the action could be 
precisely computed. 

To begin with, the reference to the sa$thi is missing fram the opening 
words of the Vidusaka in the first act In place of satthikidadevakayya of 

®» The words of the Biihmaui are, noth Stdtbtm uvavasatni. 

®* The Vidusafca observes • atthaim khu ajja. 

Translation ; '-For yonder the Moon of the Eighth, riving place to dark- 
ness, has sunk bdiind the western mount' 



the Carudatta, we have the reading stddhiktdadevakajja, in which siddhi 
takes the place of satthi Likewise we find that all subsequent references! to 
the lunar dates are missing from the succeeding speeches of the Vidiisaka 
and the Servant. An aitirdy different scheme has been adopted for the 
division of labour between the Vidiisaka and the Servant. The Servant ex- 
plains in the third act (Mrccha 137)' the arrangement arnved at as follows : 

ajja mittea edatk teah ^uvamdbhctndfusth mama diva tuha lattim ea, 

‘ Maitreya, here is the golden cadtet, that’s mine by day and yours by night 
no reference here to the softht, sattami and attham of the Carudatta. This 
is not all. The verse from the third act of the C&ru cited above, containing 
a reference to the date, has also been substantially modified. C&ru III. 3 b 
specifically states the date to be eighth : astaih goto hy astamapaksacandrak. 
In the Mrcchakattka versicoi the line reads (Mrccha. Ill 7 b) : astarh vrajaty 
unnatakotir induh The phrase umatakoti has taken place of asfamapaksa, 
which brought in its tram, naturally, the change of goto to a word like vra- 
jati It is true that later on, in the same act of the Mrcchakatika (p 159), 
the Vadhu, C&rudatta’s wife, refers to satthi, saying that she is observing 
the raanasattfS (ratnasastki) But here also a significant omission con- 
fronts us. The Vidugaka, instead of correcting her, accepts her statement 
with the necklace, and there the matter rests. 

£70} As remarked above, apparently the joint duration of the first three 
acts of the MrcchaJiatika is also three days But I have grave doubts whether 
any strict proof can be brought forward to support sudi an assumption. I 
have read the drama carefully and I have failed to find any allusion that 
necessitates such a time scheme Howeva: that may be, it is absolutely cer- 
tain that the specific references of tiie CSrudatta to the lunar dates are cons- 
picuous by thar absence in the other play. 

At this place it may be observed that the tithi-scheme of the CSrudatta 
taVpr> in conjunction with the references to moon-rise and moon-set in the 
verses already cited involves a chronological inconsistency, so minute and so 
latent as to be hardly noticeable. But the inconsistenqr is, nevertheless, an 
nndpniahlft fact For, the rising of the moon late m the evmmg and the 
setting of the moon at or about midnight” are phenomena that inherently 
belong to two different lunar fortni^ts Only m the dark fortnight does the 
moon rise late in the evemng : and only in the bright fortnigjht does the moon 
set at or shortly after tnidmght. In other words, if the moon is seen risu^ 
late in the evening on any particular day, it is nothing less than a physical 

** The present tense vrajati gives better sense than the past goto, in regard 
to the simile contained in Imes c and d 

Instead of the vague sattfn of the Carudatta we have the more specific 

raa^jrasatthi m the Mrcchakattka. j - 

” According to the words of the hero, just preceding the verse asau ht aattv^ 

■ etc. (CSru. HI, 3) *• uparudho ‘rdharatrab (Cam. 50), 


impossibility that after an interval of forty-eight hours the moon should be 
seen setting at or about midnight. 

The general time-scheme of the Carudatta has thus be«i shown to con- 
tain a latent contradiction from which the Mrcckakottka is wholly free owing 
to the absence therein of any speafic references to the days on which the 
action takes place. 

Are t hese, variations arbitrary , or are they directly or mdirectly related ; 
and if so how ? 

Summary and Conclusion 

Briefly summarized, the significant differences between the two ver- 
sions discussed above are the foUowmg Firstly, in point of technique, the 
Carudatta differs consiMCuousIy from the other play in the absence of the 
nandS, and in havmg a rudimaitary sthSpana. Secondly, the Prakrit of the 
Carudatta is more archaic than that of the Mrcchakatika, in so far that the 
f 71} former contains a number of Old-Praknt forms not found in tiie latter. 
Thirdly, as regards versification, the text of the Mrcchakatika marks an 
advance upon the other play in the following directions : rectificaticHi of 
grammatical mistakes ; elimination of redundancies and awkward construc- 
tions ; and introductian of other changes which may be daimed to be im- 
provements in the form and substance of the verses Fourthly and lastly, 
because of suitable additions and omissions the Mrcchakatika presents a text 
free from many of the flaws, such as unrealities and inconsistenaes, in the 
acticMi of the CSrudatta 

These are the facts of the case. Do these facts enable us to decide the 
qaestion of priority and anteriority? 

Let us assume first, for the sake of argument, that the Carudatta con- 
tains older material (at least m respect of the passage discussed above) which 
was worked up later into the Mrcchakatika. 

The differences in the technique neither support nor contradict definitely 
such an assumpticm. The ntndl, for all we can say, may have been lost The 
words nandyante tatah pravisati sutradhSrah do not militate against such a 
supposition : they could be used with or without a nSndi appearing in the 
text. Moreover, we cannot, in the present state of our knowledge, rightly 
evaluate the absence of all reference to the name of the play and the play- 
wright in the sthapana.^^ To say that m pre-dassical times that was the 
practice is begging the question The (mly technique of introduction with 
which we are familiar is the wdl-known dassical model. Again the only 
play which is ddiiutely known to antedate the dassical plays is the Turfan 
fragment of A§vagho^'s dra m a. Unfortunatdy, as the beginning of the 

The refersKes in the text-books of rhetoric and dramaturgy are obscure 
and partly oontiadictary. 



Sanputraprakar(a}a'^'^ is missing, we are not in a position to say wh^er the 
prologue of the dramas of A&vaghosa conformed to the standard of the classi- 
cal dramas, or that of the dramas of the group under coosideration. We 
are therefore bound to admit that at present we have no dear evidence that 
can aid us m placmg with any degree of assurance, {^72} chronologically or 
topographically, a drama with the techracal peculiarity of the Carudatta. 

But the pnority of the Carudatta version would explain, and satisfac- 
torily explain, all the other differences between the two plays. It would ex- 
plam the presence of archaisms in the Praknt of the Carudatta It would 
explain why many of the verses of the Mrcchakatika aie free from the flaws 
of the corresponding verses of the Carudatta ; the grammatical corrections 
one may be justified m regarding as an mdication of an increasingly msistmt 
demand for scrupulous purity of language The hypothesis would lastly 
explain the reason for the differences in the incident of the action of the 
play. All this is legitimate) field of ‘ diaskeuasis ’, and' is readily intdhgible. 

Let us now examine the other possibihty, and try to explain the diverg- 
ences on the assumption of the priority of the Mrcchakatika versicai. 

The question of the techmcal differences between the plays has been 
dealt with already. It was submitted that this part of the evidence was in- 
conclusive , It supported neither one side nor the other 

We will proceed to the next point, the Prakrit On the assumption 
of the priority of the Mrcchakatika version, it is at first sight not quite dear, 
how the Carudatta should happen to ccmtain Prakrit forms older than those 
found in (what is alleged to be) a still older play. But a httle reflection will 
suffiqe to bring home to us the fact that it is not impossible to account for 
this anomaly. We have only to regard the Carudatta as the version of a 
different province or a different hterary tradition, which had not accepted 
the innovations in Praknt that later became prevalent. In other woids we 
have to assume merely that the Prakrit neologisms of the Mrcchakatika are 
unauthorized innovations and that the Carudatta manuscripts have only 
t73j preserved some of the Old-Praknt forms of the ongmal Mrcchakatika.^^ 
This does not, however, necessarily make the Carudatta version older than 
the Mrcchakatika version The Carudatta would become a recension of tlie 
Mrcchakatika with archaic Praknt. Thus the Praknt archaisms of the 

Ed. Luders, Sitzungsbenchte. d. kgl. preuss. Ak. d. TFissi 1911. 

Until we have before us most carefully edited texts, any linguisltic conclu- 
sion based upon xnmute differences in the form of Ret. words, as appealing m the 
text-editions empioyed, must needs be regarded as tentative, a point not sufficiently 
emphasized m my article deahng with Praknt archaisms (above, vol. 40, pp>. 248 fi ). 
It may, however, be pointed out that no amount of aitical editing can disturb the 
general inference that the dramas of this group contain quite a nunfier of Old-Fkt 

Or that the Old-Praknt forms had been s/Ubstituted for the Middle-Prakrit 
forms, because the local tradition demanded the use of Oid-Ptakiit forma 


CatuAatta may be said to be not irreconcilable with the general priority of 
the Mfcchakcdtka version 

It is much more difficult to explain why the Mrcchakc^ika should con- 
sistently offer better readings of the verses Some of the discrepancies could 
perhaps be explained away as the result of misreading and faulty transciipt^ 
but not all. We could not explain, for mstance, why the excellent pada . 
ttkttj.cth vi'jaijagram ivavaHstam should have been discarded, and another, 
vi^anakotlva nimajjammS, be substituted, forsooth with the faulty nimajja- 
mana. Why should there be a change in the first place, and why should the 
change be consistently foT the worse ? We could not reasonably hold the 
copyists guilty of mtroducmg systematically such strange blunders and m- 
excusable distortions. 

Let us combme the archaisms of the Praknt with the imperfections of 
the Sanskrit verses. On the assumption of the postenonty of the Camdatta. 
we are asked to beheve that while the compiler of the Carudatta had care- 
fully copied out from older manuscripts all the Prakrit archaisms, he had 
systematically mutilated the Sandmt verses, which is a reductio ad absurdum’ 

Let us proceed to the fourth point The theory of the priority of the 
Mrcshakafika, which could with difficulty be supported m the case of the 
divergencies already considered, breaks down altogether when we try to 
account for the inccmsistenaes in the action of the Carudatta m general, and 
m particular the presence of the tithi-scheme, which latter serves no purpose,- 
aesthetic or didactic, but on the other hand introduces gratuitously an mdis- 
putable incongruity. The deleting of the whole tithi-acheme admits of a 
simple, self-evident explanation, acceptable to every impartial critic. But, 
assuming {74} that the ongmal play contained no trace of it, can any one 
pretend ta be able to give a satisfactory reason for the deliberate introduc- 
tion of the tithi-scheme ? 

Taking all things into account, we conclude, we can leadily, understand 
the evolution of a Mrcchakafika version from a Carudatta version, but not 
vice versa The speaal appeal of this hypothesis lies in the fact that it 
explains not merely isolated variations, but whole categories of them : it 
iirgiliea the formulation of a sin^e uniform principle to explam divers mani- 

It may be that I have overlooked inconsistoides and flaws in the 
Afrcchakaiika version, absent from the other, which could be better explained 
on the contrary supposition of the priority of the Mrcckakatika version. If 
so, the proUem becomes still more complicated, and will need further 
investigation from a new: angle. I merely daun that I have furnished here 
some pnma fade reasons for hdding that the Carudatta version is on the 
whole older than the Mrcchdkatika version ; hence (as a corollary) if our 
CSfiidatta is not itself the original of the Mrcchakatika, then, we must 
assume, it has preserved a great deal of the original upon which the Mfccha- 
kafika is based. 


Ganapati SAsral and other scholars after him, who uphold the theory 
of the authorship of Bhasa, have sou^t to justify their axription of tlie 
entire group of thirteen dramas to one common author on the strength of 
some stray sunilanties of eiqiression and analogies of thought to which they 
have drawn attrition in their writings “ The evidence that has hitherto 
been adduced must, however, be said to be inadequate to prove the daira in 
its entirety The recurrent and parallel passages collected by them, although 
they show m a general way that this group of thirteen anonymous plays con- 
tains a number of ideas and expressions in common, do not suffice to esta- 
bhsh the common authorship. It has not been realized by these scholars 
that the ascnption of common authorship has to be justified and proved 
rigorously m the case of each drama separately'' Only intensive study of 
the diction and idiosynaacies of the dramas, taken individually, will enable 
us to pronounce an authoritative opinion on the question 

{[168) There la no doubt that every poet or dramatist of note has not 
merely a settled style but a settled diction of his own This is particularly 
true of writers who are natural, and who do not strain after the unnatural, 
poignant and hi^-flown The natural or ordmaxy dement in the diction 
will recur, and recur frequently, in different works by the same author. It 
must, however, be borne in mind that the mere recuirence, m different works, 
of a limited number of ideas, phrases, and expressions, would not necessarily 
be sufSaent evidence of common authorship Such recurrence might after 
all be the result of consaous or even unconscious imitation It is, in final 
analyds, the number and character of these repetitions and analogies that will 
count and arable us to decide the! question 

The period of devdopment of andent Hmdu literature was a period of 
communal art Repetitions and analogies of thought and expression are 
therefore quite commoa Particularly, in the case of -a cramped literary 
form like the Sansknt drama, it is to be expected that works of even dif- 
ferent dramatists wiU betray a certain amount of family resemblance. Here 
we must be prepared for the recurrence of certam stock aimilira and meta- 
phors, and for the reappearance of the familiar figure of the Vidusaka and 
his stock wittiasms We must be pr^^ared for the exploitation of fossilized 

1 lAnnds SORI 4 167-187] 

* Ganapati! SAsral, Introduction to the Svapnavasavadatta (2nd edition), 
p. xix f ; Lindenau, BhaeaSttidien, p. 51 , Winternitz, OstoiiotUcke Zeitsckiift, 
Jg. rx, pp 286 ff. 

Sten KbNOW (Das ittdkhe Drama, p 53) accepts BMsa'si authorship for all 
dmtnas of this group except the Pratima See thereon Wintesnitz, op, cit. p 289, 



poetic conventions regarding certain trees such as the a:loka, regarding cer- 
tain birds such as the cakravaka ; of certain well-wom dramatic situations 
such as the rescue of the hero or the heroine from the clutches of the infuriat- 
ed elephant We must esptect to find here verses and verse-portions culled 
from epic, narrative and even didactic literature of the day that have been 
bodily rqiroduced, or that have been assimilated and have served as the 
nucleus of further inspiration All this was common property, the hterary 
stock in trade of every poet and dramatist who cared to tnalffi use of it. 
This and much else besides will be naturally ruled out as irrelevant in any 
careful exanunation, based on recurrence and analogy merely, of the question 
of the authorship of any anonymous Sanskrit drama. 

{ 169 '} At the time of the discovery of these plays the novd technique 
implied by the position of the stage direction nandyante tatah pravtiaU 
sutradharok at the head of the plays, and the use of the technical term 
sthapanS (employed in these plays instead of prasliavania to denote the pro- 
logue)', were regarded as decisive factors But since then quite a number 
of plays by different authors have been discovered in South India that show 
some of the same tedmical peculiarities® These factors, therefore, lose all 
significance now in this inquiry, and can no longer be adduced as evidence 
of common authorship. 

Intensive study of details is the only ri^t method to be adopted in this 
instance In sudi an intensive study it is evidently essential to make a 
most careful comparison of sentences and turns of eiqiressions, and even of 
words and phrases, occurring in these dramas. The case win not be decided 
on the testimony or the presentation of isolated facts, however matenal tliey 
may appear to be ; hut a conclusion can be made highly probable through 
the preponderance of evidence We have to adopt the rigorous method of 
quantitative analysis, if we are to get reliable results. To facilitate such 
comparison and such investigation, I have prepared the present hst of paral- 
lelisms and recurrences. 

The sccpe of the article has been restricted to the presentation of mate- 
rial, which falls into the following six categories . — 

(«) Entire stanzas ; 

(fr) Ekitire pfidas of verses ; 

(c) Longer prose passages ; 

(tf) Short passages; 

(«} Set phrases and rare words ; 

(/) Echoes of thought 

The list records all instances of recurrence and parallelism that i have 
been able to trace in these dramas ; except that, in order to avoid uimeoessary 
expansion, I have as a rule omitted notice of unimportant expressions recur- 

» See WnqraNnz, Ostasiatische Zettschnft, Jg, ix pp. 285 



ring in one and the same drama £170} but not found elsewhere , for, the 
list was formed pnmarily for the sake of companng the diction of the dif- 
ferent dramas. A seccBid list is appended, which is a conspectus of the same 
material arranged on a different plan • it is m fact an analysis of the fore- 
going hst. Here the recurrences and parallelisms have been arranged under 
the heads of the dramas m which they occur. This supplementary list wdl 
be espeaaUy serviceable when we imdertake a cntical examination of the 
claims of individual plays, a theme which will form the subject of a sub- 
sequent article of the senes 

(i) List of Recurrences and Parallelisms^ 
a. Entire Stanzas. 

J. imam sagarapaiyantam himavadvmdhyakundaldm { 
mahim ek&tapatrankam idjasimhaii piaiastu nah || 

Svapna. VI. 19 ; B^a. V. 20 ; Dutav. v 56 

2. bhavantv arajaso ggvah paracakram pra^myatu | 
imiam api mahim krtmam lajasimhah praSastu nah || 

Prabjfia IV. 26 ; Avi. VI 22 ; Abhi VI. 35 

3. limpatfva tamo’ ng^ varsativaihjanam nabhah | 
asatpurusaseveva djetir niephalataih gata \\^ 

BBla I. 15; CSru. I. 19. 

4. vaksaJh prasaraya kavaltaputapramanam 

aimga marh suvipulena bhujadvayena | 
unnamayiananam idam 4aradmdukalpam 

prahladaya vyasanadagdham idam sariram |I 
Pratuna IV. 16 ; ibid. VII 7 

£171} b. Entire padas of Verses 

5. imam sagaiaparyanlam — Svapna. VT. 19; BBla V 20 Dfltav. 
V. 56. C£. catussagarapaiyantam — ^BSla. IV. 10 (See 1) 

6. imam api mahSm krtsnam | iSjasknhalj praSstu nah— Prafijfia. 
IV. 26 ; Pafica. III. 26 , Avu VI. 22 , Abhi. VI. 35. (See 2 and 17) 

7 katham tisthati yatv iti — ^Parica. II. 58 ; Pratung. IV. 5 (var. 
Usthatu instead of ii?tbati) 

* In the citations, the nxman and the axabic figures refer to the Act and 
the verse respectively A smgle arable figure refers to the page of the edition 
used, unless otherwise indicated The second edition of the Svapna has been 
ated throuihout. 

* Thi^ stanza occurs also ui the Mrcckakaiika (ed. GomOLB, I. 26), and is 
quoted in Dandin’s Kavyadaria (II. 233). See F. W. Thomas, Kavindravacana 
satiMccaya (Btbltatheca Indica, 1911), p. 106, and Pischel's Introducbon to 
Rudrata’s SrngdTatHaka, pp. 16 ff. 



8 kirii vaksyaUti hrdayarfa paniankitam me — Svapna. VI. 4, IS , 
Abhi. IV. 7 

S. gadapatakacagrahe— tJru w. 41, 62 

10. candralddieva §obhate — Dutav v. 7 ; Cara. I 27. Cf. vidyul- 
lekhfiva gobhatei — ^Abhi II 7 (See 120) 

11 tvam pSjndfevanaih kura samvibihlagam — jPafica. I. 31, 47 

12, dharmasnehantare nyasta — ^Pratijna II 7 ; Abhi VI 23 

13 nrpa bhi§madayo bhagnah — ^Panca II. 41, 61 

14. bharatanlam kule jatab — Svapna VI. 16 , Pratijiiia, IV 18 

15. majqflviradtamaubs c&rutamiiayatafc?ah | . mattamatangalilah 

Abhi II 9 , IV 15 

16. yadi te’ sti dhanu^Slagha— Abhi III. 22 ; Pratitn& I. 20 

17. najasimhah praSastu naji— SvajMia. VI 19 , PTatijna. IV 26 ; 
Pafica. III. 26 , Avi. VI. 22 ; BSla. V 20 ; Dutav. v 56 , Abhi. VI 35. 
Cf. bhunum ekah praaastu nah— Kaipja. v 25, and aaja bhumim pra^stu 
nab— Pratima VII 15 (See 1 and 2) 

18 ^trupofcsam upaintya— Abhi. III. 24, 25 (var. upa&ntam mstcad 
of upSiiitya) 

19. sambhramotphuUalocania® — ^Dutav. v. 7; CSru. IV. 3 
e. Longer Prose Passages 

20 atipati karyam idatn \ fighram mvedyat^—Pafica. 24 AWii. 27 
(repeats mvedyatam) 

f 172} 12. ayam akramah | . . . atha kajj kramab— Panca 7 ; Pratima. 

22 apas lavat . yad ajniapayati mahaiiajah ] ni§krainya praviSya; 
... ima apab— Panca 10. 43, Abhi. 11; Pratiiria. 38. Cf. Madhyama. 
20 ; Pratima. 67. (See 28 and 44) 

23. evam aryamiSian vijfiapayfimi i aye kun nu khalu mayi vyfiapana- 
vyagre 4abda iva 4ruyate | anga pa^yami | (nepathye)— in the prologues of 
Svapna Pafi^. BSIa. Madhyama. Dutav. Dutagh Kanja tJtu. and Abhi ; 
of the re m a ining , m Pratijnia Avi. and Pratima, a naS is introduced 

24, kadacid anrtam noktapurvam | raja : a asty etat— Panca 30 ; cf . 
PraUifla, 32 (var. m . . anrtam abhihitapurvam) . Cf. also anpam nabhi- 
hitapflrvam 1 —Bala. 27 

25, kim edam ] . . . eda m | idam ] idam edam | . bhainadu bhapadu 
ayyo bhanadu— Svapna. 50 ; cf. kah edaih j . . . idarh | . . bhanadu bhapadu 
ayyo bhanadu— Pratijna. 17 

® Cf. vismayotihullalocaiia— Mbh. 1, 136 1 , 13 14 386 . Ram. (Bom ed ) 
7. 37. 3 29; !^im. (Gorr.) 4. 63., 10, etc. 

(Concordance op the dramas 129 

26 kiifi guhase | mama khalu piaitiai'h §apitab syah | yadi satyam na 
bruyghl — ^Abhi 37 f. , cf. kism guhase [ svargam gatena mahirajapadamu- 
lena i^pitah sySh | yadi satyam na bruyah — jPratima. 95 

27. kim te bMyah pnyam upahai&nu 1 yadi me bhagavan prasan- 
nab kim atah paramaham icchami — ^Avt. 110 , and (with vanation m) 
Pratijfia 73 , Dutav. 48 , Ablu. 75 , Cf. Bala. 67 (in which the answer to 
the query is in verse form) 

28 kutah khalv apah | . g.pas lavat j hanta sravati | (acamya . ) 

— ^Madhyama. 20 ; Dutav. 43 (See 22 and 44) 

29. gaccha ] bhuyo jfiayaf^ vrttantah — Panca. 30, 31. Bala. 58, 
Abhi. 30, 59 

30. jayatu svanS | jayatu bhadramukhah | jayatu maharajah | jayatu 
lAvapSntakah, — ^Abhi. 73 , cf. Pratima. 113 (var in different order) 

31. (tatah pravigati . ) . ka, iha bhoh kiaficanatoramadvaram 

aMnyam kurute | (pravi^ya) pratMiS : ayya aham vij'aa | kim {173} 
kariadu ) . mvedyatam mvedyatam — Svapna 69 (ms. kha) ; Abhi 26 , 

Pratiml. 90. Cf also Abhi 52 

32. parityajantrva mSm priiQiah | ima gahgSprabhrtayo mahanadyab | 
eta urvaiSyadayo ’ psaraso mSm abhigat^ I esa sahasrahaihsaprayukto 
vlravShl vimanah kalena presito mSm netum agatah | bhavatu | ayam ayam 
^aohami j (svargam gatah) — ^Abhi. 12 (ms. ka) ; cf. speech of the dymg 
Raja in Crru. 114 

33. pravii^ tvam abhyantaram . . . na khalu na khalu prave§tavyam — 
Svapna. 83 ; Panca. 42 

34. fflta : satto si | ravanah . hahaha | aho pativrataya tejah — ^Abhi 
20 , Pratimta. 86 (var. addition of a stage direction) 

35. hi hi ^ultthu naccidarh | sutthu g^idath | java ahaiii pi naccemi— 
Panca. 22 ; cf. 42 (var. hi M ^uptku idem'’ \ sutthu vdidath | sutthu nacct- 
deah j ]ava etc ) . In both cases spoken by the character VyddhagcHialaka 

d. Short Passages 

36 aaiinaih ca (dani) acchari(i)a!m— PratijOa. 14, Cam 7, 49 , and 
apnaith ca idam aodialiam — B&Ia. 36 ; as also, ekattii pi tahiifa acchanarh — 
Avi. 20 

37. atisnigdhaiii anunipaih calbihitata — Svapna. 78 ; Pafica 46. 

38. aho a(k)aru!na (k)kkhu issara— Svapna. 27, 62, Abhi 23 

39 aho balavamS ciayam andhakarah | saiihprati hi— Baa. 7 ; CSru. 17 

40. aho panjagjassa pamndo — Svapna. 63 ; Avi. 54 

’’ Read here also gSidam? 


41. aho hasyam abhidhianam — ^Pratijfia 67, 71 , Pafica 48 ; Dutagh. 66 

42 alalm daaji bhavam adimattaih samtappia— Svapna 72 : Caru 8 
(var samtapidum instead of scuhtappia), cf ma dSnuh bhaifcfa adimattam 
samtaw>iduriv— Svapna 77 , and ajaih adimattam samdavena — Avi 83 

43 aft kasya mahlarajah — ^Pafica. 39 ; Bala. 61 

fl74} 44 apaa tSvat— Pratijfia 21 ; Pafica. 10, 43 ; Madhyama 20 ; 
Dutav 43 , Abhi 11 , Pratimfi 38, 67 (See 22 and 28) 

45 fisramapadavarainatrain apa saihbhavayiayamah — Madhyama 25 ; 
Pratima. 73 (with var m the last word) 

46. ussaraha (ussaxaha) ayya ussaraha— Svapna. 1, 2, 3 ; Pratijfia 
63, 65 ; Cm. 99 

47 ete smo bhoh | ete smafc — ^Cru 88 , Abhi 62 

48. evam iva (Pkt evam via), as whispered m the ear, Svapna. 51 ; 
Pratijfia 17, 71 ; Avi. 40, 51 ; Cfim 38 (twice), 76. [Read Cam. 85 (1 13) 
likewise evam via instead of evam of the text] 

49. esa gacchami mandabhfia— Svapna. 9 , Pratijfia 17 ; Bala. 6 ; Abhi 
6 Cf. Cm. 193 ; gacchami mandabbiB 

50. k& gatih (Pkt. kfi gai) — Svapna 9, 45 , Pratima : 49, 95 

51 ko kSlo — Svapna 27, 32 , Pratijfia 56, 58 ; Cam 49 (See 52) 

52 ko kalo tumalih airoesami — Svapna. 27, 32 ; Pratijfia. 58. (See 51) 

53. gacchatu. . .punardarsanaya (Pkt gacchadu .pcmodaiihsanaa) — 
Svapna. 17, Avi 67 ; Bala. 67; Madhyama 24; Dutav. 48; Cfim. 44. 
It IS followed by . yad 6,jn&payati bhagavSn narfiyanah', in Bala. 67 and 
Dutav. 48 

54. pa me saddha (in the speech of the vidiui 5 aka)^Avi. 72 ; CSm. 
18, 30 

55 tattahodS padumavadi iha fiacchia niggadfi bhave— -Svapna, 40, 59 

56. na 4akonu ro^ajh dharayitum — ^Dutagh. 69 ; Abhi. 19 ; Pratima. 20 

57. pdairii de nivedemi — Svapna. 82 ; Cam 60 

58. badhaih prat hamah kalpah*— Svapna 52 , Pafica 19 ; Avi 26 ; 
Bala. 55 : Madhyama 24 ; Cm. 9 ; Abhi. 69 ; Pra tima 90 

58o bhuyo j'fiayatSiih vrttSntahi— See 29) 

59. mahaiajasya pratyanantaribhavSmah—Cru. 96 ; Abhi. 53 (var 

bkavi^yand) • 

tl75} 60, vaktukamam iva tvstoa lakpaye— Pratijfii. 25 ; Abhi. 43 

61. eanfi hodu ganti hodu anM(iij)aih godhanaisa a— Pafica 20, 21 ; 
Bala. 35 

* AhncBt invariably at the end of a scene or an act. 


S2. sarvalih tavat tisthata (Pkt, sawaih diava atthadu or gawaiix etc.) 
Pratijfia 58 ; Avi. 85 ; Bala. 38 ; CSru 4, 66 

63 sawaiii apedu (bodhl) vajjia bhoanjajn— Svapna 32 ; C&ru. 80 

64. sthito madhyahnah (Pkt. thido majjhaoho)— Svapna. 10 ; Pra- 
tijfia. 46 

63. athiiSknyatam atma— Pratijifi&. 17, 52 

e. Sift Phrases and Rare Words 

66 abhyantaracatu§6ala (Pkt. abbhantaraccaussala) — Svapna. 30 ; 
Cam 20, 26, 53 C£. Mirccha. act VI. (See 74) 

67 as^bhogamalmataya (Pkt asatiibhoamaliioadae) — Cfim 18, 84 

65. ama. An aflirmative particle frequently used in these dramas ; 
see references givai m ]A0S 40 254. Outside the group of these dramas 
found generally in old texts, and a few times in the Turfan dramatic 
fragments. Prof Winternitz draws attention {Ostastatische Zeitschrift Jg 
IX, p 290) to its use in the BjhatkedhaSlokasamgrdha, 5, 114 and 9, 70 

69 dcatapatra — Svapna VI. 19 , Avi. I. 1 , Bala. V. 20 : DQtav. v. 
56 , Pratima. VII, 1 

70 kanakaraata— iE*ratijna. IV 4 ; Abhi. II. 2, 5 ; VI. 6, 11, Cf. 
kanakakhaata — ^Dutav. v. 47 

71. kamalfiyalfikga— BSla V. 15 ; Dutav v. 42. Cf kamalamalayatksa 
--BSla. V 9 j 

72 krtottaiSya— Pratijfia IV. 3 ; Dutav v 3 

73 kaudumbika® (Pkt. kddumbia) — ^Pratijfia 46 ; Cam. 84 

74 catu§fi&la“ (Pkt caussBla) — Svapna 26, 30, 67 ; Avi. 23, 42, 86 ; 
Cam. 20, 26, 53 , Pratima. 96 

{176} 75. tatkaladurlabha (Pkt takkaladuUaha)— Pratijfia. 15 ; Avi. 6 
76. dattamiulya— tim 98 ; Cara. I 4 
77 dahippidapainidaral— Pafica. 22 ; Avi. 28 

78. daraparvataka (Pkt dampawadaa)— Svapna, 39 , Avi 47 

79. dutasairopata (Pkt dOdasaimpada) — Svapna. 6 ; Avi 12 
80 devasuravigraheju — ^Bfila I. 4, 21 

81. parispanda (Pkt. paripphanda) — ^Pafica. 32 ; Cara 45 

82 purobhaada — Svapna 44 ; Cgiu. 32 Not cited in lexica ; 

83 bhavanasya vinyasah — ^Abhi 15 ; bhavanavmyasa — C&ru S7 

84 hb&gyai§ calaih — Svapna . J. 3 , VI 4 

35. maioibhumi(k&) — Svapna. 27 ; Pratijfia 37 

» Generally used in the sense ‘ paterfamilias ' (see sub' voce Aptb’s dictiwj- 
ary) , here probably equivalent to 'servant', 'retainer,' 
m Occurs also often in the Mjcchakafika, 



86, mallaa”— PratijfiS 39, 41, 57 , 7 

87 lalitagambMiakjti — ^Bala 61 ; Dutagh 64 

88. vyaghiSnusSracakitaH- Madhyama v. 3 (read sa) ; CBm I. 9 

89 sajalada— Madhyama. v. 32 , Abhi IV 5. Cf. sajalajaladhara 
— AWu. IV. 3 

90. samudragrhaka (Pkt. samuddagihaka) , ‘oceanic pavilion’ — Svapna. 
54, 56, 57 ; PratuioB. 27 

9J, sumanavaininaa — Svapna 32 ; Avi 20, 40, 54 (twice) 

92 suraA ®suraife— Om v 29 ; sasuiBinatth suianam— Pratima. IV. 17 

f. Echoes of Thought 

93 ajja ewa kila sobhanaith ipakkhattam | ajja evva koduamaOgalaiii — 
kadawaim Svapna. 24 , and adyaiva khalu gufiavlan nakeatram [ adyaiva 
vivaho ’syah pravartatam — ^Panca. (p. 98 of the 1917 edition)’ 

94 ameaa mama bhada hado | anena mama pida. | alD.€ina mama sudo 
mama vaassa— PratijfiB 13 , and asyah karapeija bahavo btuBtarah sutah 
suhrda4 ca ma nihataji — ^Abhi 60 

£177} 95 abhijanassa sadisam mantidam — Svapna. 46 and abhijana- 
yuktam evabhihitam— Avi 106 

96 abhijaipeina ii?a sijcina [ganiS;] — Oaru. 37 ; Cf. jStya lakgaai na eatnu- 
dacarena*— Madhyama. 23 

97. astadnmastakagataii pratisamhrlBmi^uh — ^Abhi. IV. 23 and 

ravir api ca samksiptakiraioah | 
rathaih vyavrtyaaau praviSati ^air astaiikharam 1 1 

— Svapna I. 16 

98. a^rah fcaraajena bahavo bhifitarah, etc. — ^AHii 60. See 94 

99 imam ^garaparyantiam— Svapna VI. 19, etc. See 104 

100. kajScanastambhasad]aa — Madhyama. v. 42 ; and yah kaficanaFtatn- 
bhasamapramSinah— tiru. v. 45 

101. kim etad bho bhayatn nSma | bhavata’ dya maya ^tam — ^Bala 
III. S ; and ^penu satyesna bbayaithi na jane | jSatuiih tad icahBmi bhavat- 
samfpe— Madhyama. v. 41 

102. gajaSvanarendraiiaudha — Oru, v, 3 ; and hayanaganarendrayaudha 
— t)m. V. 12 

103. giritatakathmaiiisav eva bahii mam ai tau 
piaharatoam aparaih tu tvBdiiSim durbalanam | 

—Baa. Ill 11 

“ PW. ates only leadoographical references for the meaning ‘cup whidi W 
required in the present context. 



. . . and 

sahajau me prahara|nam bhujau plnarhsakomalau | 
tav laSntya prayudhyeyam durbalair grhyate dhanuh || 

— EPanca II 55 

104. catussagaraparyantam — ^Bala. IV. 10 , and imam sagaraparyantam 
— Svapna VI. 19, BSla. V. 20; Dutav v 56 

105. c],ttha citthn vailaficaieiDie cittha — Cam. 10, together iwith na^va 
yasi patagendrabhayabhibhiita — C3m. I. 11 , and bhoh biahmana tii^tha 
ti§tha — Madhyana 3 and kdm yasi madbhayavinasitadhairyasaralj — ^Madh* 
yama v. 8. 

106. jatya igksas! na samudac^eina — Madhyama. 23. See 96 

107 tanti!$u ca svaraganan kalahamS ca loke — ^Avi. VI. 11 , and tantris 
ca vairaim ca ghattaySmi>— Bala. I. 4 

£178} 108. devSb sendradayo bhagnah — II . 18 ; and devSb 
sendra jita yena — ^Abhi V. 12 

109 nagaraparidto ’ham rak^nno jfiatcisiaiSh 

timiragahanabhimam vartate dardhanatram | 

— Avi. Ill 2 


paridtatimiig me ^llado^ena ratrih 
bahalatmurakalas dmapuiva vighaittah | 
vipa|nj$u hatale^ rakgijnaih ^k$iino me jp’’ 

— Cam T . 13 

110. nastla sariraih kratubhir dharante — ^PaSca I. 23 , and hate§u dehe^u 
guna dharante— Kaipa. v. 17 

111. naglva ySsi patagendrabhaylabhil^uta — Cam. I. 11. See 105. 

112. paridtatimiia me, etc. — Oara. I 13 (See 109) 

113. paribhraQitoi durad ravir api ca samk^iptakiraoah — Svaima. I. 16. 
See 97 

114 paii^vajami gBdhaitfa tvana— Bala II 9 ; and paiigvajasva gSidhaith 
mam— Madhyama v. 22 

115. bhiimim ekah praUSstu nah — ^Kamja. v 25 ; and raj& bhunuith 
prai§astu naih — Pratimd VII 15 (See also 17) 

116. bho btShmaoa! ti^tha ti^tha, etc.*— Madhyama. 3. See 105 

117. mocayiami na tiajanam | tfismi yaugandhai&sraioah— Pratijfla. I. 
16 ; yadi tarn na hared rSja | nasmi yaugandhanasrainah— ibid III. 8 ; naha- 

Note that both the stanzas are in the Malini metre and that the recurring 
vrords ponata, raksinah, timira and «5fr< [.rStm) otcur in the corresponding halves 
of the respective paxlas. 




ranu nipam caiva | nasntu yaugandharayainah — ibid III 9 , yadi na patas 
bhumau nasmi damodaro ’ham — ^Bala III, 11 

118, raja bhiimiTn praSstu nah — Pratima VII. 15 See 115. 

119. vijaya khalv asi— Pratijfia 17 ; sajjalakab khalv aham— Cam. 57 ■ 
and radaimka khu ahara— Caru. 26 

120 vidyullekheva sobhate— AWtu. II. 7 , and candralekheva lobhate— 
Dutav V. 7, C3aru. I. 27 

j[l79} 121. 4apimi satyena bhayam na jane— Madhyama. v 41 See 101 

122 aajjalakah khalv aham — Caru. 57. (See 119) 

123 sahajau me praharaipam bhujau pinamsakoraalau — Panca. II . 55 
See 103 

124. sopasnehataya vaniantarasyahdiitah khalu ki§kmdhaya bhavitavyam 
— Abhi. 3 , and sopasnehataya yrksanam abhitah khalv ayodhyaya bhavitav- 
yam — 'Pratima. 42 

125 hatesu dehssu guinal dharante— Kama v 17. See 110 

126 hayanaganarendrayaudha — Cm v 12. See 102 

127. ha vatsa sarvajagatam jvarakrt krtastra 
ha vatsa vtsavajid anatavaincakra | 
hi. vatsa vira guruvatsala yuddhaSaiujda 
ha vatsa mam iha vihaya gato ’si kasmBt | [ 

— Abhi. V. 13 


hi vatsa tima jagatim nayanibhirama 
hi vatsa lakemanja salaksamasarvagitra I 
ha sadhvi maitbili patisthitaattavrtte 
ha gatah kila vanam bata me tanidjih || 

Pratima 11. 4 

(ii) A Conspectus of Recurrences and Parali^usms 

This list has been arranged according to the dramas in which the recur- 
rences and parallelisms occur. The italic figures refer to the serial numbers 
of the forcing list. 


Entira Stanzas. 1. Svaima VI. 19= Bala. V, 20; DiQtav. v 56. 
Entire padas. 8. Svapna. VI. 4, 15 ; Abhi IV. 7. — 14. Svapna VI. 
16 : PratijfSi. IV. 18.— 27 Svapna. VI 19 : PratijBa IV, 26 ; Pafica. III. 
26 ; Avi. VI. 22 ; Bila. V. 20 ; Dutav. v. 56 ; Abhi. VI. 35. 

{[180} Longer prose passage. 23. The sthipena of Svapna : Pafica 
Bala. Madhyama. Dutav. Dutagh Kaipja- Cm. and Abhi — 26. Svapna. 50 : 
Pratijfia. 17. — 31. Svajma. 69 (ms. kha) • Abhi. 26 ; Pratima. 90. — 33. 
Svapna, 83 : Pafica. 42. 



Short passages 37. Svapna 78 : Pafica 46.— 3S. Svapna. 27. Abhi. 
23. — Svapna 63 • Avi. 54 — 42, Svapna 72, 77 * Avi. 83 ; Ceru 8. — 
46. Svapna. 1, 2, 3 • Pratijna 63, 65 tJru 99 — 45. Svapna 51 PraPjfia 
17, 71; Avi. 40, 51; Caru. 38 (twice), 76. — 49 Svapna. 9; Pratijfia 
17 ; Bala. 6 ; tJru 103 Abhi 6 — 50 Svapna. 9, 45 fPfatima 49, 95.- 
51. Svapna 27, 32 , Piabjna 56, 58 ; C5ru. 49. — 53. Svapna 17 : Avi. 
67 ; B&la. 67 , Madhyama. 24 , Dutav 48 , Caru 44 — 55 Svapna. 40 • 
ibid 59 . — 57 Svapna 82- Cam 60 — 58 Svapna. 52. Pafica 19; Avi. 
26; Bala 55; Madhyama 24, tJm. 96, Abhi 68, Pratima 90. — 63. 
Svapna 32 • Cam. 80 . — 64 Svapna. 10 • Pratijfifi 46. 

Set phrases etc. 66 Svapna. 30. C&m 20, 26, 53 — 69 Svapna. VI. 
19 Avi. I. 1; Bala V. 20, Dutav. v. 56, Phatimfi. VII. 1 — 74. 
Svapna 26, 30, 67 • Avi. 23, 42, 86 ; Cfim 20, 26, 53 , PratimS. 96. — 78. 
Svapna. 39 : Avi. 47 1 — 79. Svapna 6 • Avi. 12. — 812. Svapna. 44 Cam. 
32. — 84 Svapna. 1.3- ibid VI. 4 — 85. Svapna 27 • PTatijM. 37. — 90. 
Svapna. 54, 56, 57 Pratuna 27. — 91. Svapna 32 Avi. 20, 40, 54 (twice). 

Echoes of thought. 93. Svapna 24 • Pafica (p. 98 of 1917 edition ) — 
95. Svapna. 46 : Avi. 106. — 97. Svapna I. 16 • Abhi. IV. 23. 


Entire Stanzas 2 Pratijfia. IV. 26 = Avi VI 22; Abhi VI. 35. 
Entire padas. 6. Pratijfila. IV. 26* Pafica. III. 26; Avi. VI. 22; 
Abhi VI. 35—12. Prattjfia II. 7. Abhi. VI. 23.— 14. Pratijfia. IV. 
18: Svapna. VI. 16. — 17. Pratijfia. IV. 26: Svapna. VI. 19; Pafica. 
III. 26 ; Avi. VI. 22 ; Blala. V. 20 ; Dutav. v 56 ; Abhi. VI. 35. 

{^181} Long/ifr prose passages 24. Pratijfia. 32 Pafica. 310. — 25 Pratijfia. 
17 : Svapna 50 — 27. Pratijfifi. 73 : Avi. 110 ; Bala 67 , Dutav 48 ; 
Abhi 75. 

Short passages. 36 Pratijfia. 14 : Avi 20 ; Bala. 36 ; Cam 7,' 49. — 41. 
Pratijfia. 67, 71 . Fhfica. 48 ; Duta^ €6.-44. Pratijfia. 21 : Pafica 10, 
43 ; Madhyama 20 ; Dutav. 43 ; Abhi. 11 , Pratum. 38, 67 — 46. Pratijfia. 
63, 65 Svapna. 1, 2, 3 ; tim. 99. — 48. Piratijfia 17, 71 : Svapna< 51 ; Avi. 

40, 51 ; Gam. 38 (twice), 76 . — 49 Pratijfia 17 : Svapna. 9 ; Bfila. 6; Abhi. 
6 ; tJm 108. — 51. Pratijfiia. 56, 58. Svajma 27, 32 ; Cfim. 49. — 60. 
Pratijfia. 25 : Abhi. 43 —62. Pratijfia. 58 : Avi. 85 ; Bala. 38 ; Cam. 4, 
66 — 64. Pratijfia. 46 • Svapna. 10. 

Set phrases, etc 70. Pratijfia IV. 4 . Abhi. II. 2, 5 ; VI. 6, 11. — 72. 
Pratijfia. IV. 3 : Dutav. v. 3. — 73. Pratijfia. 46 : CSm. 84. — 75. 
Pratijfia. 15: Avi. 6.1—85. Pratijfil. 37 : Svapna. 27. — 86. Ftatijfia. 39, 

41, 57 cam. 7. 

Echoes of thougfit...94. Pratijfia. 13 Abhi. 60, — 117. Pratijfia I. 16 ; 
BSla. III. 11.— 119. Pratijfia. 17 : CSiu. 26, 57. 



Entire padas €, Pafica. Ill 26. PratijfSa IV. 26; Avi. VI. 22, 
Abhi. VI. 35 —7. Pafica II. 58 • PratinHa IV. 5.— 2i. Pafica. I. 31 : 
ibid. I. iT —13 Pafica. II. 41- ibid II. 61.— i7 Pafica III. 26: 
Svapna VI. 19, Pratijfia. IV. 26, Avi. VI. 22; Bala. V. 20; Diitav. 
V. 56 , Abhi VI 35 

Longer prose passages 20. Pafica 24 Abhi 27. — 21, Pafica; 7: 
Pratiraa. 35 . — 22 Pafica 10, 43 ; Abhi. 11 , Pratima 38 — 2$. The 
sthapana of Pafica ■ Svapna. Bala. Madhyama. Diitav. Dutagh. Karna. 
Oiu. and Abhi — 24. Pafica 30 • 'Pratijna 32 ; Bala 27 — 29 Pafica. 30 f : 
Bala. 58 ; Abhi. 30, 59 — 33. Pafica 42 , Svapna. 83. — 35. Pafica. 22 : 
Bala. 42. 

Short passages 37. Pafica. 46 Svapna. 78 — 41 Pafica 48 : Pratijfia. 
67, 71 ; Dutagh 66. — 43. Pafica. 39 : Bala. 61 —44. Pafica. 10, 43 • 
Pratijfia 21 ; Madhyama 20 ; Diitav 43 ; Abhi. {182} 11 ; Pratima. 38, 
67. — 58 Pafica. 19 • Svapaia 52; Avi. 26; Bala. 55; Madhyama. 24; 
tiru. 96 , Abhi. 68; Pratima 90 

Set phrases etc. 77'. Pafica 22 Avi. 28. — 81 Pafica 32- Cam. 45. 

Echoes of thought 103. Pafica. II. 55 • B51a. III. 11.— iJO Pafica. 
I. 23 Karoa. v. 17 


Enure stanzas. 2 Avi. VI. 22 = Pratijfia IV. 26; Abhi. VI. 35. 

Entire padas 6 Avi. VI. 22 : Pratijfia IV, 26 ; Pafica. III. 26 , Abhi. 
VI 35.-17. Avi VI. 22 Svapna. VI. 19; Pratijfia. IV. 26, Pafica 
HI. 26; Bala. V. 20; Dutav. v 56; Abhi. VI. 35 

Longer prose passages 27. Avi. 110 Pratijfia. 73 , Dfitav. 48 ; Abhi. 

Short passages. 36. Avi 20* Pratijfia. 14. Bala. 36; Oaru. 7. 49 -- 
40. Avi. 54: Svapna. 63 —48. Avi. 40, 51 Svapna. 51 ; Pratijfia. 17, 
71; cam. 38 (twice), 76— 53. Avi. 67 • Svapna. 17; Bala 67, Madh- 
yama. 24 ; mtav. 48 ; CBm. 44. -^4, Avi. 72 , Cfira 18, 30—53. Avi. 
26 : Svapna. 52 ; Pafica. 19 ; Bala 55 ; Madhyama. 24 ; Dm. 96 ; Abhi. 68 ; 
Pratima. 90.— ^53. Avi. 85 ; Pratijlfig. 58 ; BBla 38 ; Cam. 4. 66. 

Set phrases etc. 69 Avi. I 1 • Svapia. VI. 19 , B51a. V. 20 ; Dutav. 
v. 56 ; Pratima VII. 1.-74. Avi. 23, 42, 86 ; Svapna. 26, 30, 67 , CBru. 
a), 26, 53; Pratima. 96.-75. Avi 6: Pratijfia. 15.— 77. Avi. 28: 
Pafica. 22. — 78. Avi. 47, Svapna. 39. — 79. Avi 12: Svapna. 6. — 91. 
Avi, 20, 40, 54 : Svapna. 32. 

Echoes of thought. 95 Avi. 106* Svapna. 46.— i<?7. Avi. VI. 11. 
Baia. I. 4.— 109. Avi III. 2: CSm. 1. 13. 




Entire stanzas. 1. BBla- V. 20 = Svapna. VI. 19 ; Dutav. v. 56.— 
3. Bala. I. 15'= Cam. I. 19. 

^183] Entire padas 5 Bala. V. 20 : Svapna. VI 19 • Dutav. v. 56. — 
17 Bala. V. 20 : Svapna. VI 19 ; PiatijM. IV. 26 , Pafica. III. 26 : Avi. 
VI . 22 ; Dutav. v. 56 , Abhi VI 35. 

Longer prose passages 23. The sthiipanS, in Bala . Svapna. Panca. 
Madhyama. Dutav. Dutagh. Kama. Cm and Abhi — 24, Billa 27 : PratijM. 
32 , Pafica 30.— 27 Bi51a. 67 Pratijfila. 73 ; Avi 110 ; DiQtav 48 , Abhi. 
75.-29. Bala. 58i : Pafica. 30, 31 ; Abhi. 30, 59.-35. BSla. 42 Pafica. 22. 

Short passages 36 Pratijfifi. 14 , Avi 20 ; CSm 7, 49 —39 Bala 7 • 
Cam. 17 . — 43 B&la 61 : Pafica 39.-49 Bala. 6 ; Svapna 9 , Pratijfia. 
17; Abhi 6; Cm 103. — 53. Bfila. 67. Svapna. 17, Avi. 67; Madhyama 
24 ; Dutav. 48 ; Cam. 44.— 58 Bala 55 Svapna 52 ; Pafica 19 , Avi. 26 ; 
Madhyama 24 , Cm. 96 ; Abhi. 68 . Pratima 90 . — 61 Bala 35 : Pafica. 20, 
21 — 62. Bala. 38 : Pratijfia. 58 ; Avi 85 ; Cam 4, 66 

Set phrases etc. 69 BSla. V 20 Svapna VI. 19 , Avi. I. 1 , Dutav v. 
56 ; Pratima VII !.■ — 71. Bala V. 9, 15 : Dutav. v. 42. — 80. ^^la. I. 4 ; 
ibid. I 21 — 87. Bfila. 61 Dutagh. 64. 

Echoes of thought. 101. Bala. III. 8 : Madhyama. v. 41 .—103. Bala. 
III. 11 ; Pafica II. 55 — 104. Bala. IV. 10 : Svapna VI. 19, Dutav. v. 
55.-107. BBla I 4 : Avi. VI. 11—214 Bala II. 9 . Madhyama. v. 22. 

Madhyama • 

Longer prose passages. 23. The sthBpana of Madhyama : Svapna. 
Pafica. Bala. Dfitav Dfitagh Karpja. Cm and Abhi. — 28. Madhyama 20 : 
Dutav. 43. 

Short passages. 44. Madhyama 20 * Pratijfia. 21 ; Pafica. 10, 43 ; 
Dutav. 43 ; Abhi. 11 ; Pratima. 38, 67 . — 45 Madhyama. 25 : PratimS. 73.— 
53. Madhyama. 24 : Svapna 17 ; Avi. 67 : Bala. 67 • Dutav 48 ; Cfim. 44. 
— 68. Madhyama. 24 . Svapia. 52 , Pafica. 19 , Avi 26 ; Bala. 55 ; Cm. 
96 ; Abhi. 68 ; Ptatima. 90. 

£184} Set phrases etc. 88. Madhyama. v. 3 : CBm. I. 9 — 8i9. Madhy- 
araa. v. 32 : AWii IV. 5 

Echoes of thought 86. Madhyama 23 : CBm 37 . — 100 Madhyama. 
V. 42 : Cm. v. 45. — 101. Madhyama. v. 41 . BBla III. 8. — 105. Madhya- 
ma. 3. and v. 8 : CBm. 10 and I. II . — 114 Madhyama v. 22 : Bala. II. 9. 


Entire stanzas. 1. Dutav. y. 56 = Svapna. VI. 19, BBla. V. 20. 



Entire padas. 10. Dutav. v. 7 ; Cara I 27 — 17 Dutav. v 56 ; Svap- 
na VI. 19 ; Pratijm IV' 26 ; PaSca III. 26 , Avi VI. 22 , Bala V 20 ; 
Abhi VI 35— Dutav. v 7 • Cara. IV 3 

Longer prose passages. 23 The sthaparia of Diiitav • Svapna. Pafica. 
Bala Madhyama. DQtagh Karoa Cm. and Abhi— 27 Dutav. 48 . Piatyfia. 
73 ; Avj 110 , Abhi. 75 —28. Dutav 13 , Madhyama 20 

Short passages 44. Dutav 43 : PratijiSh. 21 ; PaSica. 10, 43 ; Madhya- 
ma. 20 ; Abhi. 11 , Pratima 38 67.-53, Dutav 48 : Svapna. 17 ; Avi. 67 ; 
Bala. 67 ; Madhyama. 24 , Cara 44 

Set phrases etc 69 Diitav. v 56 : Svapna. VI. 19 ; Avi. I. 1 •, Bala 

V 20; Pratima VII. 1—70 DQtav v 47. Pratijlta IV; 4; Abhi II 
2, 5 , VI 6, 11.— 7i. Dutav. v. 42 . BSIa. V 9, 15.— 72. Diitav v. 3 : 
PratijSa. IV. 3 

Echoes of thought 104. Dutav. v. 56 : Svapna VI. 19 . Bala. IV. 10 ; 

V 20 — J20, Dutav. v. 7 • Abhi II 7 ; Oaru I. 27 


Longer prose passages. 23. The sthapanS of Dutagh : Svapna Pafica. 
Bala Madhyama. Diitav Kama. Cru and Abhi. 

Short passages. 41 DQtagh 66 ; PratijfSa. 67, 71 ; Pafica. 49.-55. 

69 . Abhi. 19 ; Pratima, 20. 

Set phrases etc. 87 DQtagh. 64 : B&la. 61. 

tl85} KarisiabhAra 

Entkd padas 17. Karina, v 25 : Svapna VI 19 ; PratijfSa. IV. 26 ; 
Paika. Ill, 26 ; Avi. VI. 22 ; Bala V. 20 ; Dutav. v. 56 ; Abhi VI. 35. 

Longer prose passages. 23. The sthfipai]® in Katpa : Svapna. Pafica. 
Bala. Madhyama Dutav DQtagh Ora. and AWii 

Echoes of thought, lio. Kamja. v, 17 . Pafica I. 23,— Ji5. Karpa. 
V. 25: Pratima. VII. 15 . 


Entire padas. 9 Oru. v. 41 : ibid. v. 62 

Ti>r prose passages. 23. The sthSpana of tJru . Svapna. Pafica. BSla. 

Madhya^ DQtav. DQtagh Kama and Abhi— 32. Cra. 114: Abhi 12 
(ms. ka). 

passages. 46 Cm. 99 : Svapna 1, 2, 3 ; Pratijfia 63, 65 —47. 

AWh R ' 103 : Svapna. 9 ; Pratijfia. 17 ; Bala. 6 ; 

Aora. 6 —55. Cru. 96 • Svapna, 52 ; Pafica. 19 ; Avi. 26 ; B§la. 56 ; Madlv 

7aina. 24 ; AWii. 68 ; Pratinfi. 90.— 55, Cm. 96 ; Abhi. 53. 


concordAno: of the dramas 139 

Set phrases, etc. 76. Cm 98 : CSra. I 4—92. Cm. v. 29 , Pratina. 
IV. 17. 

Echoed of thought 100. Cm v. 45 , Madhyama v 42. 


Entire stanzas. 2 Ablu. VI. 35 = Pratijfia. IV. 26 ; Avi VI. 22. 

Entire padas. 6 Abhi. VI. 35 PratijfSa IV. 26; Panca III. 26; Avi. 
VI 35 —8 Atei. IV. 7 Svapna VI 4, 15— W Abhi. II. 7 Dutav. v 
7 . Cam I 27. 12 Abhi VI. 23 Pratufla. II. 7—15 Abhi II. 9 . ibid. 
IV 15—15. Abhi III 22 Pratima. I 20— 17. Abhi. VI. 35 : Svapna. VI. 
19 , Prati]i55. IV. 26 , PaBca. III. 26 , Avi. VI. 22 Bala V. 20 Diiitav. v. 

|[186} Longer pvose passages 20 Abhi. 27 Psaica 24 — 22 Abhi 11 • 
FaQca. 10, 43 PratimB 38. — 23 The sthapana in Abhi. ; Svapna Pailca 
Bala Madhyama Dutav. Dutagh Kama and Cm. — 26. Abhi. 37 f : Pra- 
tima. 95, — 27. Abhi. 75 : Pratijfia. 73 ; Avi. 110 ; Diiitav. 48 , Bala 67. — 
29. Abhi. 30, 59 : Plafica. 30 f ; Blala 58. — 30 Abhi 73 : Pratima. 113. — 
31. Abhi 26 : Svapna 69 , PratimS. 90 — 32. Abhi. 12 (ms. ka) : Cm 114. 
—34. Abhi. 20 Pratmfi. 86 

. Short pasde^es. 38. Abhi. 23 Svapna. 27. — 44. Abhi. 11 : Ptatijna. 21 ; 
Pafica 10, 43 ; Madhyama 20 ; Dutav. 43 , Ihratinia 38, 67. — 47. Abhi. 62 : 
Cru 88. — 49. Abhi 6 . Svapna. 9 ; Pratijffii. 17 ; Blala 6 ; Cm 103.- -56. 
Abhi. 19 ; Diiitagh. 69 ; Pratimla 20 — 68 Abhi. 69 • Svapna. 52 , Pafica. 
19 ; Avi 26 ; Bfila. 55 ; Madhyama 24 ; Cm. 96 ; Pratima. 90 — 60. Abhi 
43 : Pratijfia, 25. 

Set phrases etc. 70 Abhi II. 2, 5 ; VI 6, 11 : PratijiSa IV. 4; Dutav 
v. 47. -^5 Abhi 15 Cam. 57.^9. Abhi IV. 5 . Madhyama. v. 32. 

Echoes of thought, 94. Abhi. 60 , PratijM. 13.— 97. Abhi IV. 23 : 
Svapna I. 16 — 120. Abhi II 7; Diiitav v. 7. CBra I. 27 — 124. Abhi. 
3 : Pratrafi 42.— J27. Abhi. V. 13 ; Pratima II. 4. 


Entire stanzas. 3. Cam. I. 19 >= Bala. I. 15. 

Entire padas. 10. Cam. I, 27 : Diiitav. v. 7 ; Abhi. II. 7,-29. C&m IV. 

3 : Dutay. v. 7 

Short passages. 36. Cam. 7, 49 . Pratijfia. 14 '-^39. C&m. 17 : Bala.- 
7.-42. C&m. 8 : Svapna. 72, 77.— 48. Cara. 38, 76 Svapna. 51 ; Pratijfia. 
17, 71 ; Avi. 40, 51—52. Cam. 49 ; Svapna. 27# 32 , Pratijfia 56, 58.-^3. 
Cam. 44 . Svapna. 17 ; Avi. 67 ; Bala. 67 , Madhyama. 24.-54 Cam. 18, 
30 • Avi 72.— 57. Cam. €10 . Svapna 82.-52. CBm. 4, 66 ; Pratijfia. 58 ; 
Avi. 85 ; Bata. 38.-55. CBm. 80 . Svapna. 32. 



{187} Set phrases etc. 66 Oara. 20, 26, 5S : Svapna 30.-67. Caru 
18 ; :bid 84 —73 Cam 84 . Pratjna. 46 —74. Oaru. 20, 26, 53 : Svapna 
£6 ; 30, 67 ; Avi. 23, 42, 86 —76 Oaru I 4 ; tJm 99.— 8i| Cam. 45 ; Paflca. 
32. — 82. Cam 32 : Svapna 44. — 83. Cam. 57 • Abhi. 15.' — 86. Cam. 7 : 
Pratijna. 39, 41, 57.— SS Cara I. 9 • Madhyama v 3 

Echoes of thought. 96. Cara. 37 : Madhyama 23.— 105. Gam. 10, and 
I. 11 . Madhyama 3 and v 8 — 109 Cam. I. 13 . Avi III. 2 — 119. Cira. 
57 : Pratijfia. 17.-720. C6m I. 27 • Abhi II. 7 , Dutav v 7. 


Enttrd stomas 4. PratimS. IV. 16=ibid VII. 7 

Entire padas. 7. Pratma. IV. 5 • Pafica. II 58 —16, Pratim&. I 20 : 
Abhi. III. 22. 

Longett prose passages 21. Pratima. 35 : Plalnca. 7. — 22. P*ratirn@. 38, 
67 . Pafica. 10, 43 ; Madhyama. 20 ; Abhi. 11.— 26 Ih^timia. 95 ; Abhi 37 f. 
- -30 Pratima. 113; Abhi. 73. — 31 Pratima 90 : Svapna, 69 (ms. kha) ; 
Abhi. 26, 52 — 34, Pratima 86 .' Abhi. 20. 

Short passages 44 PratimS. 38, 67 . Pratiji5& 21 ; PaSca. 10, 43 ; 
Madhyama 20 ; Dutav 43 ; Abhi. 11 —60. Pratima. 49, 95 ; Svapna 9, 45.— 
66. Pratima. 20 , Dutagh 69 , Abhi. 19 — 58. Pratima 90 : Svapna. 52 ? 
Pafica. 19 ; Avi. 26 , Bfila 55 ; Madhyama. 24 ; tJm 96 , Abhi. 68. 

• Set phrases etc 69 Pratima. VII. 1 Svapna VI 19 , Avi I. 1 ; Bala. 
V. 20 ; Dutav. v. 56.— 74. Pratinfi 96 : Svapna 26, 30, 67 : Avi. 23, 42, 86 ; 
Cam. 20, 26, 53—50. Pratima. 27 : Svapna 54, 56, 57—52. Pratinfi. IV. 
17 : tJm. v. 29. 

Echoes of thought 124 PratimS. 42 , Abhi. 3—727. Prating II. 4 : 
Abhi V. 13. 


The effort to place the group of anonymous plays discovered in South 
India by Pandit Gapapati SAstrI of Travancore has engagied the imagination 
and the pai of Indologists for over a decade ; but no definite solution of 
that problem has yet been logically justified Opmion is divided on more 
than one a^iect of the plays Opmion is sharply divided between those who 
place the dramas in the fifth century bc. and those who place them in the 
tenth century ad, between those who ascnbe them to the ‘far-famed’ 
Bhasa, honoured by Kalidasa and those who ascnbe them to a poetaster 
whose name even is forgotten by postenty , between those who daim 
foi them high literary merit and those who describe them as the miserable 
lucubrations of a plagiarist. These three aspects of the plays, it may be 
added, are not entirely independent of each other ; in fact, the second and 
the third of than are really doedy connected. For while, on the one hand, 
those who support the Bblasa theory invariably claim to be able to recogmze 
high merit in the plays ; on the other hand, those who repudiate that theory 
at the same time deny the plays all real ment. 

Despite 'the divers opinions hdd by scholars regarding the age and 
authorship of the plays, and despite the formidable phalanx of arguments 
advanced by them to support tiieir respective claims, the significant differ- 
ence, it seems to me, has beoi just on file question of the literary and aesthe* 
tic ments and defects of the dramas Back of all the various aspects of the 
discussion seems to lurk, often unnoticed by the disputants themselves, this 
fundamental divergence And aesthetic merit being a vague quality not amen- 
able to exajct measurement or computation, the difference of opinion as 
regards the place of these dramas in the history of Sansknt literature will in 
all likelihood continue to exist, {231} unless another fortuitous discovery 
happens to place in our hands some material which can give an unequivocal 
reply to the question of the age or the author of our dramas 

Although the suspiaon voiced by Barnett® that few Sandmtists ‘ agree 
with the learned editor’s ascnption of them to Bhfisa,’ appears to be utterly 
without foundation, it cannot be demed that a few cntics who had first haded 
the appearance of these plays with ddat and jubdation, have later, on re- 
examining the plays, become indifferent and turned away from them in 
considerable disappointment But Pandit Ganapati SAstrI’s alluring theoiy 
has in the meantime made fre^ conquests and found new adherents 

* [JBBJIAS 26. 230-249.] 

1 The paper was read at a meeting of the Society ield on ^ardi 22nd, 19ffl. 

* /BAS. 1919, p. 238, 



The Bhasa Question is now, it may be emphatically stated, as far away 
from being settled as ever before. The number of writers on the subject is 
steadily increasing and the fidd of research is gradually widening It is there- 
fore highly that all students interested in the question should have, 

even at this stage, a list as complete as possible of the wnters and their vent- 
ings so that they may be able to tdl at a gdance what editions and transla- 
tions are available, what the problems are, and what has been written con- 
cerning them 

Apropos of the remark of Barnett ated above, a few statistics may not 
be out of place Here is a hst of those who have, at one time or other, written 
on the subject, acceptmg the Bhasa theory expliatly or impliatly : Amara- 
natha Sarma, Afie, Asuti Anantacharya, Banerji-Sastri, Baston, 
Beocarini-Crescenzi, Belloni-Filippi, Belvalkar, Bhatta, bhide, Chau- 
DHURi, Deb, Desapande, Dhruva, Gajnapata SAstrI, Gray, Gune, Har- 
prasad, Hertel, Hillebrandt, Jacobi, Janvier, Jayaswal, Jolly, Kale, 
KhuphekAr, Konow, Lac6te, Lesny, Lindenau, Mehendale, Mor- 
(Enstierne, OoffiN, Parma Lall, Paranjape, Pavolini, Pishakoti, Printz, 
Saunders, Suali, Thomas, Urdhwaheshe, and Weller. It must be added 
that the enthusiasm of Sylvain Lfevi has apparently cocAed down considerably 
dnce he penned his ecstatic preface to Easton’s (French) translation of 
Vasavadatta ; and now, I understand, he has j'omed the ranks pf the (^pon- 
rats of the theory, which include the names : Barnett, Bhattanatha Svamin, 
Kane, Mahabal, Rangacharya Raodi, and Ramavatara Sarma. Whether 
the opponents of the Bhasa theory are really so few, or whether they are over- 
modest and of a retiring disposition, it is cer-f232j-tam that the number of 
such as have expressed their views openly is remarkably limited Between the 
two extreme sections he the views of Winternitz and the present writer, 
who, while they recognize that the supporters of the theory have a good prftm 
facie case, that the authorship of Bhasa is a factor within the range of pos- 
sibility, hold, on the other hand, that the evidence hitherto adduced does 
not amount to a condusdve proof of the proposition ; they accept it merely 
tentativdy, as a working hypofliesis. 

In passing it may be pointed out that the doubts propounded by Bar- 
nett, and the intetpretation of the term rajasimha (oocurrmg in the bharata- 
vakyas of the plays) as a rumen ptoprium — ^features of the controversy gene- 
rally assodated with the name of Barnett® — had been made public by 
Pandit Ramavatara Sarma Pandeya in an artide cootnbufed to thq little 
known Sanskrit journal Sarada long before the appearance of Barnett’s note 
in the Jottnvd of the Royal Astatic Soci&y. In 1915 fE^dit Ramavatara 
Sarma wrote expressing his doubts as to the validity of Ganapati SAstrI’s 

® Barnett’s objectioas have been critidsed and refuted severally by Banerji- 
Sastri, EoNDw, F. W. Thomas, and Winternitz, 

A biblio(%aphical Note 


theory, ascribing the dramas to an anonymous court poet of a Kerala inng 
Rajasimha. BARNEifr’s first article on the subject, as far aa I knoiw, did not 
appear till 1919, that is, four years later. 

Of the thirteen dramas comprising this group, the Svapnav^avadalta is 
undoubtedly by far the most popular. Gaijapati SAstrI pubhshed some years 
ago the third edition of the text , and thore are of this drama seven inde- 
pendent translations m five different languages (English, Frmh, German, 
Gujarati and Italian). Like its remarkable namesake ated by Rajai^hara 
in lus Suktmuktdvaii, it may ■well daim to be able to withstand even the 
rigorous ‘ordeal by fire’ Next m populanty stands that interesting little 
one-act q»sode Madhyama, whidi hasPbeen translated four times already and 
which richly deserves to be more widely known The Prcdvma and the Charu- 
datta have been translated twice each, and a new Italian translation of the 
Chdrudatta is, I understand, m course of preparation Of the remammg, the 
five major dramas Abhisheka, Avtmaraka, PaHchardtra, PratijM, and Baia- 
chanta, have been translated once only, while the four one-act Mahabharata 
episodes CVubkanga, Karnabhara, Dutaghatotkacha, and Dutavakya, have 
not attracted serious attaition so far The PrattjM is really an mteresting 
p33} little drama of uiuiuestionable merit ; but its third act (the so-called 
manttanka) presents certain difficulties. That is perhaps the reason why it 
has not yet tempted any translator except the mtrepid Keshavlal Dhbuva. 

Now as to the cntiasm of the dramas A cntical study of the Prakiit 
of the whole group has been made mdependently by two yoimg German 
scholars Lesnv and Printz There is also an unpretentious little contribu- 
tion on the subject by the %>resent writer The rdationship between, the 
Chdrudatta and the Mjichchhakattka has been exhaustively investigated by 
McaiGENSTiERNE. This monograph, taken in conjunction with two othei 
papers dealing iwith the same subject that were almoat simultaneously made 
pubhc dsewhere, seems to establish bdyond all reasonable doubt two facts : 
firstly, that the Chdrudatta is a fragment ; and secondly, that it represents 
a version of the 'theme earber than the Mrichchhakaltka. We have a schol- 
arly contnbution to the study of the source of the Svapna firom the pen of 
Fdix Lac6te, who has made a special study of the literature dustering round 
the Brihatkathd. The lexicographical pecuhanties of the same drama have 
been studied and listed by the American Indologist Ogi®n. The late Dr. 
Gune has left us a small but thoughtful cootribution to a study of the Pra- 
tijnd. A connected ziccount of these dramas will be found in the Introduc- 
tions to Pandit Gaijapati SAstrI’s editions of the Svapna and the Pratima 
respectively ; and in the secbons on BhSsa in Kono'w’ s Das indische Drama 
and WiNTERNiTZ’s Geschichte der indischm Litteratur 

Estimates of the age of these plays vary, as already averred, by about 
fifte«i centuries. They have been assigned to the fifth century B. c by 
Bhu® , third (or second) century by Ganapati SAstrI ; to the first century 



B c. by Javaswal and Chaudhuri ; to the second century a.d. by Konow, 
Lindenau and Suali ; to the third (or fourth) century by Banerji-Sastri, 
Jolly and Jacobi , and to the fourth century by Lesnt* and Winternitz ; 
to the seventh century by Barnett and Nerurkar (on independent grounds) ; 
to the ninth century (or later) by Kane , to the tenth century (or later) by 
Ramavatara Sarma I^deya , to the elevaith century (or later) by Ranga- 
charya Raddi. 

Very, bnefly summarized the arguments for and against the theory are 
the following Among the most important arguments adduced in support of 
the theory are these. (1) The common authorship of the plays follows from 
the similarity £234J of technique, style and thought mformmg these plays, and 
from the abundant mstances of repetition and parallelism. One of these plays 
IS styled the Svapnavdsavadatta, which is the title of a celebrated drama com- 
posed by Bhasa. (2), A technical pecuharily of the prologues of the Bh^a 
dramas has been noticed by Baiqa in his Harshacharita, which pecuharity 
diaracterises also the prologues of our dramas. (3) The name of the author 
IS never mentioned in the rudimentary sthapanla of these plays, which testi- 
fies to their great antiquity, further evidenced by the archaic languagP! and 
the technique of these plays. (4) Owing to their having been well-known 
plays, verses and passages from them have been cited and criticized by rheto- 
ricians such as Bhiamaha, Dainkhn, and Vamana, although th^ do not name 
the source from which these verses and passages have been taTff»n. (5) Apt 
expressions and feheitous similes have been borrowed from these plays by 
celebrated poets like Kahd S s a, Bbavabhiiti and others. Being distinguished 
products of dramatic art, they are m style and*inatter worthy of the fatTw of 
the great Bh^. These are the arguments advanced m support of the theory. 
On the other hand, those who repudiate the Bhasa theory do so mainly for 
the following reasons. (1) This Svapmvasavadatta does not contain the 
verse quoted by a certam rhetorician as from a drama of the satTM> narry^ 
which drama, it is said, is probably the ongmal Svapnavdsavadatta of Bhasa 
Likewise these plays asenbed to the great diamatist do not contam any of 
the verses dted in anthologies as his verses. (2) The stanza quoted from the 
Harshachantet of BSlpa has been grossly misinterpreted, and is quite irrele- 
vant to the discussion. (3) The similarities of ideas and expression between 
these plays and the works of celebrated dramatists like Kiahdasa dearly prove 
that the author has unfalushingly plagiarized from the works of other drama- 
tists. (4) They contain irregularities of technique and a surprising number 
of grammatical blunders, which exdude the possibility of their being the 
works of any reputable author, not to speak of Bhasa. Obviously works of 
mediocre quality, they are in every way unworthy of being ascribed to the 
distinguished dramatist BhSsa. I have singled out here for the purpose of 
this airvey, only the most important arguments advanced on either side , 
None of them appear to me incontrovertible ; the balance seems dehcatdy 

A. bibliographical note 145 

adjusted. It is a question where the emphasis should be laid, aod the answer 
to that question will largely depend on personal predilections 

The Bhasa question has acquired fresh interest and importance through 
the discovery of other dramas such as the Mc««-p 35 }-m/ds 4 ,i which appa- 
rently stand closer to our group than to the classical dramas like those of 
Kalidasa, Bhavabhiuti, and others. It is becoming increasingly evident that 
we have before us dramas, if not of Bhlasa, at least of a distinctly new school 
of dramatic art, and as such they are undeniably mtereating and worthy of 
most careful study There is nothmg to be gained by peevishly brushing 
them aside as the lucubrations of a plagiarist, or as the creations of an inge- 
nious forger (as one learned Indian cntic* has averred), simply because they 
are not exactly what we expect them to be or want them to be. Already the 
study of them has yidded some fruitful result, and it is not too much to 
say that a deeper study of them may throw further light on some of the 
obscure comers of this inteiesting field of inquiry. 


A. Individual Plays. 

Text edition. 

1. The Abhishekanfitaka of Bbasa. Edited with notes by T. 
Ganapati SAstrI. Trivandrum, 1913, pp. 75 -f 3. (= Tn- 
.vandrum Sandmt Senes, Na XXVI.) 


2 Itedian. II dramma della sacra di BIma (Abhi$ekanStaka) com- 

posto dal poeta Bhasa [Translated by] Elena Beccabini- 
Crescenzi. Firoize, 1915, pp. 79 

Reprinted from Giom Soc. Aaat. Ital vol 27 (1915). 

T4et edition. 

3 The Avaimfiraka of Bhasa Eklited with notes by T. Gaivtapati 

SAstrL Tnvandnim, 1912, pp. Ill -t- 2. (■= Trivandrum 

Sanskrit Series, Na XX ) 


4 Italian. L’ “Avunaraka” di Bhasa. [Translated by] Elena 

Beccarini-(2rescenzi Firenze, 1917, pp. 40. 

Reprinted from Giora. Soc. Aaat, Ital. vol. 28 (1916) 

^ Travancore Sanskrit Series, No. 55. 

* Mr. Pandurang Vaman Kane, m.a... ljum., in the Vividha-mSmt-vUtant, 
1920, p. 102. ■ 


TeU edition 
Sad No 35 


Tdxt edition. 

See No 35 


Text editions 

7. The Ch&ndatta of Bhasa. Edited with notes by T. Ganapati 

SAstri Tnvandrum, 1914, pp. ii ,+ 86 4- 2. (i= Trivandrum 
Sanskrit Series, No XXXIX. 

8. The Charudatta, A Misra Prakarana of Bhasa With commen- 

tary by Mahamahopadhyaya T. Gainiapati SAstrI. Trivan- 
drum, 1922, pp. 11 -I- 2 + 104. 

Text reprinted in Mohgenstiebene’s Dissertation (see No. 13). 


9. Bengali ChSTudatta, mahakavi Bhasa pra|nita. [Translated by] 

Saradchandra GhoshAi, In Upasma, (Year) 1325, 
pp. 139-146, 218-230. 

Bengali year 1325 corresponds to a,d. 1918-19. 

10. Norwegian. Et ganunelt indisk drama [Translated by] Sten 

Konow In Edda (Kiistiama), 1916, pp 389-417. 


11. Belloni-fillipi, F. Note critiche ed esegetiche al “ CSrudatta ” 

di Bhasa. In Riviata degli studi orientali, vol 9 (1923), 
pp 581-590. 

12. Belvalkar, S. K. The relation of gudraka’s Mrochakatika to 

the Carudatta of Bhasa In the Proceedings and Transactions 
of the First Onental Conference, vol 2 (1922), pp. 189-204. 

A sumrnary appeared in vol. 1 (1920), p. Ui (See Nos 13, 15 
and 84.) 

13. Mor(^nstierne, Georg Uber das Verhaltnis zwischen C)3rudatta 

und Mrcchakahka. Ldpzig, 1921, pp. 80 -H LXI. 

Seeks to establish on internal evidence that the Mrdichhakatika 
IS an enlarged verson of the fragment Qiarudatta of Bhasa, ina- 
dentally justifying the authorship trf BhSsa (See Nos. 12. 15 
and 84.) 


£236} 5. 

6 . 



14 StJKTHANKAR, V. S “ Chanidatta A Fragmait In The 
Quarterly Journal of th^ Mythic Society, Bangalore, 1919, 
pp. 181-185. 

[2373 15. SuKTHANKAR, V S On the relationship between the Carudatta 
and the Mfcchakatika. InJourn.Amer Or. Soe.vcA. 42 (1922), 
pp. 59-74. ( 1 = Studies m Bhasa III). 

The paper, whidi was presented at the One Hundred Thirty-third 
Meeting (Baltimore, 1921) of the Amer, Or Soc., seeks (like 
Nos. 12 and 13) to estabhsh on mternal evidence that the Mrccha- 
katika is an enlarged version of the fragment Chanidatta See 
Nos 12, 13 and 84. For a mudi earher article on the same sub- 
ject see S M. Paranjape's Chanidatta .ami Mrchchhakatika (No. 
84). Havmg been pubhshed in a httle known Magazme, it ap- 
pears not to have attracted much notice 


T^t edition. 

16. See No 35. 


Text editions 

17. See No 35 

18. The Dfltavdkya of BhSsa. With the commentary of Mah§maho- 

pddhy^ya Pandit T. Ganapati SAstrI Trivandrum, 1918, 
pp. 32. 

In the Preface it ieI called the second edition f it presents a text 
revised with the collations of a new manuscnpt. 


19 WiNTERNiTZ, M. Mahabharata II, 68, 41 ff. utid Bhasas Duta- 
vakya. In Aufsatze zur Kultur- und Sprachgeschichte, vomekm- 
Ifch des Ortents, Ernst Kuhn gum 70 Gehurtstage am 7. 
Februar 1916, gewidmet von Freunden and Schulem. Mdnchen, 
1917, pp. 290-304. 

Seeks to prove that the ated Mbh. stanzas are post-Bhasa mter- 


Text editions 

20. The Panchai^tra of Bhasa. Edited with notes by T Ganapati 
SAstrI. Trivandrum* 1912, pp. 51-1-3. (= Trivandrum 

Sanskrit Senes, No XVII.) 

Rev. Macdoneix, JRAS 1913, pp 186-190. See No. 80. 



21 The PancharStra of Bhasa. With the commentary of Pandit T. 
Ganapati SAstrI. Trivandrum, 1917, pp. 4 + 4 + 2 + 117 
+ 4. 

f238l In the Introduction it is called the second edition ; it pre- 
Lnts a text revisfed with the collations of a new manuscnpt 

22. Pancharatram ol Bhasa. Edited with Introduction, English 

Translation, Notes, Glossary, etc., by Waman Gopal Urdh- 
WARESHE And a Sanskrit commentary and Hindi translation, 
by Knshiniacharya iSastif. Revised by Mdhara iSastri. Indore, 
1920, pp. lii + 16 + 54 + 110 + 72 + 116 + 16 

“ Bbdsa must have lived one or two centimes before Kalidfls ” 

Text editions 

23. The Pratijnayaugandhai^yaina of Bhasa. Edited with notes by 

T. Ganapati SAstrI. Trivandrum, 1912, pp. 73 + 12 +S 
(= Trivandrum Sansknt Series, No. XVI.) 

Rev. Macdonell, JRAS 1913, pp 186-190 See No 80 

24. The Pratijnayaugandharayama of Bhisa With the commentary 

of Mahamahopadhyaya Pandit T Ganapati SAstrI. Third (!) 
edition. Trivandrum, 19210 pp 4i+ 3,+ 4 + 129 + 3. 

In the Introductioa it is called the second edition , it presents a 
text revised with the ooUations of a new manuscnpt. 


25. Gujatati. Pradhan-nl piatijSa. Translated from BhSsa’s Pratijiia- 

yaugandhanayafla by K H. Dhruva. Ahmedabad, 1922, 
pp 40 + 152. 


26. Gone, P. D. Pradyota, Uda]rana and iSrenika — a Jain legend. In 

Amals of the Bhandarkar Institute, 1920-21, pp. 1-21 

A comparisoD of three different versions of the Udayana legend : 
the Kathlsaritsagara, PTatijfia, and KumBrai^apratibodha. 

27. Hertel, Joh. JinakiSti's “Gescfaichte von 3E^a and Gopiala" 

(Bedchte Sachs. Gesdil. Wiss, voi. 69, 1917), pp. 123 If. 


Tdxt edition 

28. The Pratin^nStaka of Bh^.. Edited with notes by T. 

Gai^apati SAstrI. Trivandrum, 1915, pp. 3 + p39j xli + 
32 + 116 + 4 + 4 + 7 + iii. 0= Trivandrum Sanskrit Series, 
No. XUI.) 

A bibliographical note 



29. Engfish. Bh^a’s Pratimanatakam [Trandated] by K Rama 
PiSHAROTi. In the Quarterly Journcd of the Mythic Society, 
vol. 11 (1920-21), pp. 353-366, vol. 12 (1921-22), pp. 58-66, 
375-396, vol 13 (1922-23), pp. 595-606 

A running oonunentary duadates the text and translation. — Only 
the first four acts have appeared so far. 

30 Gujarati. MahSkavi 4ia Bhasa prajiit pratim&natak Gujarati 
translation by Manual Chhabaram Bhatta. Ahmedabad, 
1916, pp 12 H- 80 

For a criticislm on the statues mentioned in this drama see No. 99 

Tdtf edition. 

31. The BSlacharita of BhSsa Edited with notes by T. Ganapati 
SAstrI Tnvandrum, 1912, pp. h-i-68 -1-2. ( = Trivandrum 

Sansknt Series, No. XXI ) 

32 Blalacarita. (Die Abenteur des Knaben Kridina. Schauspiel von 
Bfalfisa. Text herausg^eben von Dr. H. Weller. Leipzig, 
1922, pp. IX + 105. 

A repnnt of the Tnvandrum Test with emendations, chiefly of 
the Praknt passages, and text critical notes — K German transla- 
tion of the drama, by the same author, likewise pubhshed by 
Haessel (Leipzig, 1922), was not available to me 


33. Lesny, V. Bhasovo BSlacantam. In Listy filologicke, vol. 42 

(1915), pp. 437 flf. 

34. WiNTERNiTZ, M. Kr^na-Dramen. (2. Bblsas Balacarita) In 

Zeitschnft d deutsch morgenl GeseU. Band 74 (1920), pp. 125- 

Besides the translation of a niunber of verses, the article contains 
an abstract of the plot of the drama, and a companson of this 
version of the Ejnsbija l^^d with other versions. 

Text edition. 

35. The MadhyamavyHyoga, Dfltavfikya, Dfltaghatotkacha, Kama- 

bhflra and Orubhanga of Bhfisa. Edited with notes by T. 
Gaijiapati SAstrI. Trivandrum, l[24G} 1912, pp. 114 -f 5. 
(|= Tnvandrum Sanskrit Series, No. XXII.) 




36. The Madhyamavyayoga of Bhasa With the commentary of 
Phndit T Ganapati SAstrI Trivandrum, 1917, pp 6i-K4 
+ 43. 

In the introduction it is called the second edition ; it preseits a 
text revised with the collations of a new manuscnpt. 


37 English. The Madhyama Vyayoga. A drama composed by the 
poet Bhasa Translated from the original Sanskrit with Intio- 
duction and Notes by Rev. Ernest Paxton Janvier (Uni- 
versity of Pennsylvania Thesis ) Mysore, 1921, pp 44 

38. Gujarati. Sri mahlkavi Bhasa-krit Madhyamvyayi^ Translated 

by LalSankar Harprasad. With an introducticai by Uttamram 
Ambaram Bombay, 1917, pp 33 . 

39. Madhyam. Translated from Bhasa’s drama Madhyama by K. H. 

Dhruva. Baroda, 1921, pp. 32 + 49. 

40. Italian I drammi Mahabharatiani di Bhasa I. Madhyamavya- 

yoga .[Translated by] P. E. Pavolini. InGiom. Soc. Asiat 
Ital. vol. 29 (1917), pp. 1-27 


Text editions, 

41. The Svajmavzlsavadatta of Bhflsa Edited with notes by T Ga^ja- 

PATi SAstrI. Trivandrum, 1912, pp. xlvii .+ 43 + 77 + 11,+ 5 
[Editio prmceps.']— 2® ed Trivandrum, 1915, pp xlvii + 43 
+ 86 + 10 + 4. 0= Trivandrum Sanskrit Senes, No. XV.). 

Rev. of first edition, Macdoneix, JRAS 1913, pp. 186-190. See 
No 80. 

42 The SvapnavAsavadatta of BhAsa. With commentary by Pandit 
T. GAivfAPATi SAsTRi. Tnvandrum, 1916, pp 20 + 12 + 148 

In the Preface it is called the third edition It does not appear 
that any iiew manuscnpt material has been used in the revision 
of the text 

43. The Svapna Vasavadatta of BhSsa. Edited with Introduction, 
Notes, Sic , &c., by H. B Bqioe. With Sanskrit commentary 
by Narayan Shankar Rajvade Revised by Phndit Shyam- 
sundara Shastri. Bhavnagar, 1916, pp. 2 + 9D + 120,+ 52 + 
VII + 3. 

£2413 introduction the author 'discusses (among other 

Questions) the date of Bhasa, and asdgns him to the fifth cen 
tuiy B.C. 




44 English The Dream Queai A G. Shirrepp & Panna Lall 
Allahabad, 1918, pp IV + 55. 

Free trandation in blank verse. Noticed by S SSstrI in Ind 
Ant. 48 (1919), p 176. 

45. Bhasa’s Svapna Vasavadatta. (Translated specially for the 

Society’s Journal with aitical notes) By K Rama Pisha- 
ROTi. In The Quarterly Journal of the Mythic Society (Banga- 
lore), vol. 10 (1919-20), pp 164-174, 209-220, 372-381 ; vol. 
11 (1920-21), pp. 122-137. 

46. Vasavadatta, bang a literal rendering of Bhasa’s Svapnavasava- 

datta* By V S. Sukthankar. In Shama’s (a Magazine of 
Art, Literature and Philosophy, published in Madras), vol 2 
(1922), pp. 137-169; vol. 3 (1922), pp. 25-45. 

47. French Vlsavadattil Drame en six actes de BhSsa traduit pout 

la premi&e fois du Sanscrit et du jwflcnt par Albert Baston, 
avec une preface de M. Sylvain L6vi. Le th&itre indien avant 
KSUdlsa. Paris, 1914, pp. VI -1- 121. ('= Bibl. Onent 
Elz6vir. No. LXXXVTI.) 

Rev ilteLLONi-FaiEPi, Giorn. Soc Asia!. ltd, vol 27, pp. 159-176 
See No 51 

48. German VBsavadatta Ein altindisches Schauspid von Bhasa. 

trbersetzt von Hermann Jacobi. In Intemationde Monats- 
schrift fur Wtssenschaft, Kunst uni Techmk, 1913, pp 653-690. 

49 Gujarati Sachuih svapna Translated from Bhasa’s Svapna- 
vasavadatta by K H Dhruva. Ahmedabad, 1916, pp 44 + 103 

50. Italian. La Vasavadatta di Bhasa. Dramma. Trad di F. 

Belloni-Filippi. Lanciano, 1916, pp. XXII 4- 142 (Scnt- 
ton Italian! e stranieii. Teatro.) 


51. Belloni-Filippi, F. Una recente traduzione ddla “Vasavadatta” 

di Bh&sa. In Giorn. Soc. Asiat. Ital. vol. 27 (1915), pp. 

Rev. of A. Baston’s translation of the drama See No. 47 

{242]} 52. LacOte FUlix La source de la Vasavadatta da Bhasa. In 
Journal Asiatique (1919), S6r 11, Tome 13, pp. 493-425. 

Compares the versions of the Svapna with those of the Katha 
sarits&gara, Ratn&vali, Pnyadai^ikS, and T^pasavatsaiaja. 

^ A revised version of this tran^tion has smce been published by the Oxford 
University Press. 



53. Ogden, Charues J Lexicograptucal and grammatical notes on the 
Svapnavasavadatta of BhBsa. In Jotern. Amer, Or. Soc. vol. 
35 (1915), pp. 269-272. 

54 Ogi«n, Charles J. Bbasa’a Treatment of the Udayana Legend. 
A paper presented at the 135th Meetmg of the Amencan 
Onental Society, Princeton, 1923. 

B. General Criticism of the Plays. 

55. Amaranatha Sarma. Mahakavir Bhasah.® In SaradS (Allaha- 

bad), vd. 2 (1916). i[In Sanskrit] 

56. Anantacharyya Asuri MahakayiT BhasaJi. In Samskfita- 

bharafi, vd 4 (1922), pp. 35-49. [In Sanskrit] 

57. Apte, Hari NarAyan. BhBs kavichya nataka-kathS Poona, 

1917, pp. 9 + 115. [In Maratln.J 

Tales from Bhasa told in Marathi. 

58. BANERjr-gASTRi, A. The plaj^ of Bhasa. In Journ Roy, As. 

Soc. 1921, iq). 367—382. 

Chiefly cnitiming Barnett’s views on the subject (see Nos 59- 
61), justifies the authorship of Bhasa, and, on Imguistic grounds, 
places BhiSsa between Advag^iosha and KSlidSsa, or between tlie 
thffd and the fifth century a,d. — ^Barnett's reply, /RAS 1921, 
pp. 587-589. See Nos. 61 & 92. 

59. Barnett, L. D. The plays ascribed to Bhasa and the Matta- 

vilasa. In Journ. Roy As. Soc 1919, p. 233 f. 

60. Barnett, L. D The MattaviLfisa and “ Bhasa.” In Bulletin of 

the School of Onental Studies, London Institution, 1920, vol. 1, 
part 3, pp. 35-38. 

61. Barnett, L. D. "BhBsa” In Joum. Roy. As. Soc. 1921, 

pp. 587-589. 

The writer attnbutes the dramas to an anonymous court poet of 
a Flaipdya King BBjasanha of the seventh century A d,, basing his 
arguments diiefly on his inteipiretatioia of the word rajasirhha in 
the bharatavSkya of the dramas, and on the alleged tedmical 
similarity between these plays and the MattavilSsa. No. 61 is a 
rejomder to Banebji 1243} Sastri's ‘The plays of Kiasa.’ 
JRAS, 1921, pp. 367-382. See counter-joinder by Thomas, ibid, 
1922, pp 79-83. For further cntiosm see Nod 58 and 104. 

62. Bhattacharya, D. C. BhSsa and his alleged works. 

Paper stated as read at a meeting of the Asiatic Society of Bengal 
See Ptoc. As. Sot, Bengal, 1917, p. ccxiv.— Apparently not pub- 

* Not available to me, 



63 Bhattanatha SvAMiN Thirteen, newly discovered dramas attri- 
buted to BhBsa. In Ind. Ant. vol. 45 (1916), pp. 189-195. 

Enqihatically rejecting Bhasa's authordnp, assigns the dramas to 
an anonymous i)oet of unknov7n date. 

64. Chaudhxjei, P. The date o4 Bhasa In Modem Reviw 

(Calcutta), vol, 14 (1913), pp. 382-397. 

Supporting Jayaswal (see No. 71), assigns the plays to the teign 
of NaiSyataa Kanva, basing the conclusions on alleged allusions in 
the plays to contemporary history — See rejoinder by Venkaia- 
RAMAN, Modern Review, 1913, p 579 f (No 93). 

65. DeSapande, R. D. BhSs va Dhavak he niraniitle kavi hot. 

iLBhasa and Dhavaka are two different poets] In Vividha- 
jnana-vtstara (Bombay), vol 50 (1919) [In Mara^^hi ] 

66. De§apande, R. D. Bhas a|oi Kahdas.. In VwidhajMna-vxstara 

(Bombay), vol. 51 (1920), pp. 19-28. '[In Marathi] 

E^atimates KahdSsa’s indebtedness to the author of the plays, 

67. Ganapati SAstrI. See Introductions to his editions of the Svapna- 

VBsavadatta (Nos. 41 & 42) and PratimSnEltaka (No. 28) res- 

68. Guleri, Chandradhar. a poem by Bh&sa. In Ind. Ant. vol. 

42 (1913), p 52 f. 

Draws attention to an old gloSs, which refers to a poem called 
Vishnudhanna by a BhSsa , see the editorial note, tbid. p. 53 . 

69 Hau.* F. Fragments of three early Hindu dramatists, Bhasa, 
Ramila, and Somila. In Journ As Soc. Bengal, vol. 28 
(1859), pp. 28-30. 

70, Jacobi, H. See Introduction to his German translation of the 
Svapnavasavadatta (No 48). 

71 Jayaswal, K. P. The plays of Bhasa, and King Dai;§aka of 
Magadha. In Joum. and Proc. As. Soc. Bengal, vol 9 (1913), 
pp, 259-269. 

r244"} Justifies the authorship of BhBsa, and assigns the plays to 
die ragn of Narayaja KSnva (ca. 50 B.c ) on the gjound of 
allied veiled and diacure allusions to him in some of the vwsea 
of the plays Indirectly answered by Venkataraman in Modem 
Review, 1913, p. 579 f. See No 93, 

72, Kaijie, P. V. Kavi Bhas va tad-rachit natkath. In Vividha- 

jrtana-vist&ra (Bombay), vol. 51 (1920), pp. 97-102. 

[In Marathi] 

Supporting the views of RafigachSrya B. RAUPi (see ^- 86) 
assigns the plays to an anonymous idagianst of some period later 
|hnTj the ei^th century aj), 





Khutoekar, B. M. AbhB.a navhe BMsac. In LokaHkshm 
(Poona), vol 5 (1916), pp. 295-298, 324-328. 353-358, 395- 
^2 Marathi.] 

Rejrander to RangSdiarya B Raddi’s ' KiBs Idiii abhSs (No. 86). 
Konow, Sten. Das mdische Drama. Lteipzig, 1920 (>= Grun- 

driss der indo-arischen Philologie nnd Altertumskunde, Band 

2. Heft 2 D ) 

Tthaaa pp 51^6.— 'AsSigns the author of the dramas Bbasa, to 
the idgn of the Kdiatrapa Rudrasimha I , that is, to the end of 
the second century a.d , xnadentally questions Bh^a s authorship 
of the Pratima. 

75. Lesny, V Vyvojovy stupen ndreci Prafcrtskych v dramatech 
Bhasovych a urceni Bhasovy doby [The stage of develop- 
tnpnt of the Prakrit dialects m BMsa’s dramas and the date 
of Bhasa.] Rozpravy cesbd Akademie Qsara Frantiska 
Josefa Tnda 3, aslo 46. Prag, 1917. 

See No. 76. 

76 Lesny, V. Die Entwiddungsstufe des PiSkrits in BhBsa’s Dramen 
und das Zeitaltear Bhasa’s In Zeitschrift d. deutsch. morgenl. 
Gesell. Band 72 (1918), pp. 203-208. 

Summary of his cootnbutioa to the Bohemian Academy of 
Saences, m which the writer assigns, chiedy on linguistic grounds, 
the author of the plays, Bhasa, to the first half of the fourth 
century AJ) 

77. L£vi, Sylvain. Le Theatre Indien, Pans, 1890, vol 1, pp. 157- 
160; vol. 2, pp. 31-32. 

Gves an almost exhaustive resume of literary references to Bhisa 
and his works, known until then. 

78. Livi, Sylvain. Preface to A. Baston’s translation of die Svapna- 
vSsavadatta (See No 47.) 

£2453 Iandenau, Max Bhasa-Studien. Em Beitrag zur Geschichte 
des altindischen Dramas Leipzig, 1918, pp. VI . -1- 51 . 

A oongerie of observations on divers aspects of the plays. — See 
WiNTHSNlTZ, Ostasiatische Zdtschrift, Jg 9, p. 297f 

80. Macdonell, a. a. Three plays of Bhasa in the Trivandrum 

Sanskrit Series. In Jouirn. Roy As. Soc. 1913, pp. 186-190. 
Rev of Nos. 20, 23, and 41. 

81, MahAbax, Bh. B. BhSs va KShdas (dusaii hejO). i[Bhasa and 

Kalidasa : the other side,] In VtvidkajnanehVKtdra, vol. 51 
(1920), pp. 73-80, ,[In Marathi.] 

Rejcmder to No 66. Indignantly rejects the suggestion of Kali- 
cRsa’s indebtedness to these dramas of doubtful ment, 

A bibuocsraphicAl note 


82. Meerwarth, a. M. The dramas of Bhasa. A literary study 

In fourn and Proc. As. Soc \Bdngal. N. S. vol. 13 (1917), 
pp. 261-280. 

An appraisement of the literary and aesthetic merits of the plays, 
without reference to the question of authorship. 

83. Morgenstierne, Georg. Uber das Verhaltnis zwischen Carudatta 

und Mrcchakatika. Leipzig, 1921, pp 80 + LXI. 

See especially pp. 5-21. 

Pandeya See RSmavatara SakmA Pandeya (No. 87) . 

84. Faranjape, S M. Chamdatta Sni Mrichchhakatika In Chitra- 

maya-jagat, 1915, pp. 46 ff. — ^Ptiyadariika ani Nagananda hi 
ko|i 3 &chi, ibid. 1915, pp 576 ff — ^BhSsa vishayl kahi goshfl, tbid. 
1916, pp 91 ff — ^Bhisachi hhavitavyata, ibid 1916, pp. 381 ff. 

85. Printz, Wilhelm. Bhasa’s Prakrit. Frankfurt, AM. 1921, 

pp. 47. 

86. RAppi, Rangacharya B. Bhas kim fibhSs ’ i[Bhasa or his sem- 

blance?] In Vivtdha-jnana-vistara (Bombay), vol. 47 
(1916). [In Marathi.) 

Emphatically rejects the Bhasa theory, chiefly on the ground that 
these plays of questionable worth could not be the works of the 
great Bhasa. 

87 Ramavatara iSarma PSftjideya. Mabakavir Bh&sah, In 'Sarada 
(Allahabad), vol. 1 (Vikrama 1970^), pp. 4-7. [In Sansknt] 

p46J Like (see Nos 59-61), this writer asagns the plays 

to an anonymous court poet of a FSindya king RAjasitnha ; and 
estimates their age ca. 10th century AD 

88. Smith, V. A. Discovery orf the plays of Bhasa, a predecessor 

of Kalidasa In Ind. Ant. vol. 40 (1911), pp. 87-99. 

See No. 107. 

89. SuALI, Luigi. I drammi di Bhgsa. Firenze, 1912, pp 36 . 

Reprmted from darn. Sac. Asiat. Jtal 1912, The article was con- 
tinued m vol. 26 of the Journal, but the contmuation is not avail- 
able to me — ^Assigns the dramas to ca. 2nd century a.d. 

90. SttKTHANKAR, V. S. Studies m Bhasa : 

I. On certain archaisms in the Praknt of these dramas In 
JouH'n Amer. Or Soc vol. 40 (1928), pp 248-259. 

II. On the versification of the metncal portions of the dramas. 
Ibid vol. 41 (1921), pp. 107-130. 

III. On the rdationship between the Cdrudatta and the Mrccha- 
katika. Ibid. vol. 42 (1922), pp. 59-74. 

^ Vikrama 1970 corresponds to ad. 1914-15. 



IV. A concordance of the dramas. In Annals of the Bhandarkar 
Institute, vol. 4 (1923) . 

V A bibliographical note. In Journ. Bombay Branch of 
the Royal As. Soc vol. 26, pp. 230 ff 

91. SxncTHANKAR, V. S. A note on the dramas of Bhiasa. In 

Shamcia (a Magazine of Art, Literature and Philosophy, pub- 
lished in Madras), vol 3 (1922), p 59 f. 

A note of a popular character appended to the translation of the 
SvapnavSsavadatta, appearing in the same Magazme 

92. Thomas, F W The plays of Bhasa In Journ Roy. As. Soc 

1922, pp. 79-83. 

Sets forth fresh reasons in support of the Bhasa theory, bang at 
the same a rejoinder to Barnett, iUd. 1921, pp 587-589 (see 
No. 61). 

93. Venkataeaman, T. L. The Date of Bhasa. In Modern 

Review (Calcutta), vol. 14 (1913), p. 579 f. 

Rejomder to P Chaudhuhi’s ‘The Date of Bhasa,’ Mod Rev. 
vol. 14 (1913), pp 382-387 See No. 64, 

(247} 94. WiNTERNiTZ, M. Der indische Dramendichter Bhasa In 
Ostasiattsche Zeitschnft, Jg. 9 (1922), pp. 282-299. 

Contents • 1 Is Bhasa the author of the dramas attnbuted to 
him? 2 The date of Bhasa 3. Some observations on Max 
Lindenau's " Bhasa-Studien ” (see No. 79). The Appendix em- 
phasises the 'wnter’s view that the ascription of the plays to Bhasa 
is nothing more than a ‘ hypothesis,’ whidi needs further investi- 
gation, and verification 

95. WlNTHiNiTZ, M. Geschichte der indisdien Literatur, Band 3 

(1922), pp. 184-202, 205 f., 209 f., 644-646; see also Index 
s, V. Bhasa. 

C. Incidental References. 

96, Bhattanatha Svamin. Mayuraja. In Ind. Ant. 41 (1912), 

p. 141. 

A propos of M Krishnamacharya's ascription of Kiranavaft and 
other dramas to Bhasa (see No. 105). 

97. Deb, Harit Krishna. Udayana Vatsa-raja. Calcutta. 1918, 

pp. 1-9. 

A brochure pubhshed by the author himself.— Mainly historical 

98, Gray, Louis H. Vasavadatta, a Sanskrit romance by Subandhu, 

translated with an introduction and notes New York, 1913, 

p. I f. ^f the Inteodaction) (•= Columbia Umversity Indo^ 
Iranian Series^, vcd. 8.) 



99. JAYASWAL, K. P, Statues of two SaiSunaka emperors (483-409 
B. C.) In The Journ Bihar and Offzssa Research Soc. vol. 5 
(1919), p. 96 f. 

The Pratima is ated here to establish the ‘ custom of maintain- 
ing a royal gallery of portrait statues,' s/uch as those of the Sata- 
vahana kings at NSnaghat, and of the Saisunaga kings, now pre- 
served in the Indian Museum, Calcutta 

100 Jolly, J Kolldctaneen zum Kautiliya ArthaiSstra In Nach- 
ncht konigl GesdL Wissen zu Gottmgen, 1916, p 353. 

101. Kale, M. R. In the Introduction to his edition of the Ratnavali, 
Bombay, 1921, pp. xvu-xx 

Discusses the validity of the alleged quotation^i from the Kavi- 
vimaida of Rajasekhara, published by Narayan Sastn See also 
the bhmmka to the Vani Vilasa edition of the Pnyadarstka, 
p. xxvii. 

{248} 1102 Konow, Sten. In hia leview of Hillebrandt’s edition of the 
Mudifiifikshasa, Ind, Ant. vol. 43 (1914), pp. 65-67. 

103 Konow, Sten., Zur Fruhgeschichte des indischen Theaters. In 

Aufsatze zur Kultur- und Sprachgeschtchte, vomehmlich des 
Orients, Ernst Kuhn znm 70. Geburtstage am 7 Februar 
1916 gewidmet von Freunden and Schulem, Munchen, 1917, 
pp 106 ff. 

Embodying views subertantially the same as those repressed in lus 
woric ‘Das indisdie Drama* (see No. 74). 

104 Konow, Sten In his review of W Caland’s edition of Gopala- 

kelichandnka, Ind Ant, vol 49 (19210), pp. 233-235. 

Chiefly cntiases Barnett's articles on the subject (see Nos 59 
& 60). 

105. Krishnamacharya, M. A History of the Classical Sanskrit Lite- 

rature, Madras, 1906, p. 67. 

Refers to a tradition whidi asenbes the U-dattarfighava, Svapna- 
vasavadatta, and KiranavaK to Bhasa That paiisage has been 
cntiazed by Bhattanatha Svamin, Ind Ant vol 41, p 141. 

106. IVfeHENDALE, K. C. Date of ifiudi aka’s Mrcchakatika. In Com- 

memoratwe Essays presented to Sir Ramahrishna Gopal 
Bhandarkar, Poona, 1917, pp. 368-370, 374. 

‘ It IS an undoubted fact that the Chdrudatta formed a unit in 
the ndtakachakra of Bhasa The Carudatta printed in the 
Tnvandiuin San^mt Series is evidently a fragment.' 

107. Narasimhachar, R In Archaeological Survey of Mysore, 

Annual Report, 1909-10, p. 46. 

The pertinent passages have been exceipted by V. A. Smith, Ind. 
Ant, vd, 40, p 87f (steeNo 88), 



108 Nerurkar, V. R. In the Introduction to his edition of the 
Mrichchhakatika, Bombay, 1919, pp. 14-19, 

' Charudatta and Mrichdibakatika are probably the productions 
of one and the same author— namely Bhasa. This is not the Pre« 
Kahdasian Bhasa but a BhSsa who was dhavaha washerman 
by ca^te and who flourished in the time of Shn Harsha (7th 
Century— the first half}.' 

109. PiscHEL, R, In Gotting. Gelehrte Anzetgen, 1883, pp. 1229 ff 

£249} 110. Saraswati, A. Rangaswami, The age of Bharavi and Dandin 
or the literary history of the Pallava period In The Quarterly 
Journal of the Mythc Sodety, Bangalore, vol. 13 (1923), 

p. 686 

111. Saunders, Virginia. Some hterary aspects of the absence of 
tragedy in the classical Sanskrit drama. In Journ. Amer, Or. 
Soc. vol 41 (1921), pp 152-156. 


This is a rather belated review of the thesis Bhasa’s Piaknt by Dr. 
Wilhelm Printz, which was accepted by the University of Frankfurt as 
‘ Habihtationsschrift ’ m 1919, but which was not published till 1921.° It is 
undoubtedly the most important contnbution® hitherto made to the study 
of the Praknt of the thirteen anonymous plays attributed to Bhasa, and as 
such It deserves a detailed notice. Moreover, a^ the author of the brochure 
contemplates incorporating the published material in a Fiaknt Lexicon which 
he 18 prepanng,* it appeared desirable that before the material is finally em- 
bodied in the proposed dictionary, the thesis should be cntically examined 
by some one who has made a careful study of these dramas. As I had 
already collected considerable data of a similar kmd in the course of my 
study of the dramas, I was m a position to check without much difficulty 
the statements of Printz by comparing them with my own unpublished nbtes. 
The following revkiw is the outcome of this comparison 

It may be stated at the very outset that the work of PRINTZ represents 
the most painstaking, mmute and comprehensive review, hitherto published, 
of the Praknt of these dramas. As a monument of patient erudition it com- 
mands respect, and as a conscientious piece of labonous work it will be 
valued by every i[104} sctious student not only of the Trivandrum plays but 
also of dramatic Prakrit The searching criticism to which it is here sub- 
jected IS not made in a captious spirit of fault-finding ; it is offered with a 
view to increasmg the value and utihty of the work. 

A defect whidi mars considerably the value of this dissertation is the 
axiomatic finahty with which Printz postulates the authorship of Bhasa ; 
foi though the attribution of the plays to this dramatst may be said not to 
have been satisfactorily disproved , b it cannot be contended any longer, m 
face of the numerous vahd objections raised against the theory, that it has 
been satisfactorily established either® Not only does Printz categoncally 

1 [JBBRAS NS. 1, 103-117.] 

2 Wilhelm Runtz : BMsa’a Prakrit, Frankfurt a. M., 1921, im Sdbstverlag, 
p 47. 

» The Czech cantnbution oULesny to the Btiieniian Academy of Sciences 
is to me, unfortunatdy, a sealed bode. Its resum6, ZDMG 72 (1918), 203 ff. is 
rather scrappy. 

4 See Printz’s Emldtung (p. 3). 

® A Bemedale KEITH, Notes on the Sanskrit drama, BSOS 3, 295 ff. 

« See my Studies in Bhasa V, JBBRAS. 26, 234 ; Pisharoti and Pishaeoti, 
“HiSsa’s Wo^”— -Are they genuine?, BSOS 3, 107 ff., Kuiihan Raja, Bhasa, 
anoth^ .Side, Zeitschr. f. Ind, u. Iran. 2, 247 ff. ; Easnett, BSOS 3, 35 ; and W, E. 
Clark, MOS 44, 101 f. 



atoume Bhasa’s authorship, his methodology seems to imply also that the 
Trivandrum texts have been handed down in an almost unalloyed condition 
since the timp of the supposed author Bhasa ! Printz deals with the Prakrit 
of these plays in the same ooofident way in which Prof. Luders has dealt 
with the Praknt of the Turfan fragments of Buddhist dramas.’^ In doing 
so, Printz has failed to take into account the essential difference of chaiacter 
between the two sets of manuscripts, not to speak of the manner in which 
they have been editwi ; he appears not to appreciate the elementary fact that 
Praknt texts are liable to senous mutilation and corruption in the course of 
transmission through centimes, and that they need most careful editmg 
Printz’s method of arguing is most unscientific. 

Even a cursory examination of the Prakrit of these dramas is sufficient 
to show that the manuscnpts are full of blunders and inconsistencies. Here 
are some a pnort considerations which cast suspicion on the absolute punty 
of the text : the frequent elision in Sauraseni of f m the termination of 3, 
Sing. Pres (-ft) and in the ending of the Part. Perf Pass (Printz 

32, .39) , the umform £1053 change of mtervocahc -fk- to -h- (Printz 16) ; 
the termination of 2 Plu. Indie and Imp. -ha instead of -dAa (Printz 32) ; 
the frequent change, in MS.gadhl, of imtial y- to (Printz 17) ; the (ap- 
parent) retention of -yy- (derived from Skt -ry-) m iSaur^ifi (Printz 21) ; 
evident Dravidianisms® such as Saur -nd- mslead of -nt- (Printz 19) ; uni- 
form cerebralizatiMi of I (imtial as well as double) (Printz 18) , the forms 
attabhavath, tattabhavam^ (Printz 22) ; palpable Sanskritisms like vtssa- 
siht, samassasihi.^o rodidi (PRintz 34), ommtaam (Printz 32) ; and so on 
and so fwth. 

Anothec — and a more senous — defect in this dissertation of Printz 
arises out of the faulty classification, of the Praknts It is extremdy unfortu- 
nate that Printz (p 6) should have thou^t fit to style as Magadhl the 
Prakrit of the Cowherds in the two ICr$na dramas It seems unnecessary to 
point out that a MagadhI in which the Nora. Sing, of thematic R tpma ends 
in -0 is no MagadhI at all ; at least not the MSgadhi we know anything of. 
This curious dialect of the Cowherds in Bala and Pbfica has all the appear- 
ance of being a western or nOTthem dialect, and may, for the saVe of con- 
venience, be styled a variety of iSauraseni, as Weller has done but I fail 
to see how it could be called MagadhI. Again, to bracket together the dialect 
of Indra (in Kaiaja.) and of the Pugilists (in Bala.), and to label as 
Ardhamggadhiw is not merely a ‘NofbeheH' (as Printz calls it), but the 

* LfjDEBS, Bruchstdeke btiddhtstischer Diamen, Berlin 1911 
« PISCHEL 275. 0 Ibid. 298. 10 Ibid, 496. 

Balacanta (Lapeig 1822), Vorwort, p. iii. Banesji-Sastri, 
Oban ; His age and MagadhI, Jourti of the Bikar & Orissa Res. Soc 1923, pp. 1 ff. 
admits under M l gadh i the dialects of Unmattaka and AakSra only 

th. Sr *“■ » 



hei^t of inconsequence and arbitrariness. It seems almost as though Printz 
needed ‘ Bdege ' for Ardhamiagadhl in order to complete his case for Bhasa ; 
and the dialect of the Pugilists was the only one handy besides the few sen- 
tences spoken by Indra. These facts, unfortunately, make Printz's atations 
for MagadhJ and Ardhan^gadhl all but useless. 

Since the appearance of the dissertation of Printz, our knowledge of 
dramatic Praknt has been considerably furthered through £106} the publica- 
tion of the southern texts of other dramas.^^ The additional light thrown by 
these publications on the practice of southen dramatists and southern scnbes 
will necessitate correction m many a hasty generalization of Printz, based 
on an observation of too narrow a field. 

With these prdimmary remarks we may proceed to an examination of 
Printz’s treatment of the grammar of the Prakrit of these plays, twhich begins 
on p. 8 and comprises the major part of the thesis 

Page 8 {Lm& 5) 5. pdaia- (prdkrta-) Avi. 29 has the usual meaning 
' common ' ; pdadagmid means * a common prostitute and therefore it is 
not necessary to stretch pda4a- to mean ‘weggejagt/ as P. does — {Line 6.) 
There is no need to trace back pdkida- Pratijfia 13 to prakfta- since prdkrtah 
(Pkt. pdkido 'a common fellow*) gives a thoroughly satisfactory sense 
without any difficulty, — {Line 12.) vascd)ha- and gova^aha- B^a, 15 are not 
Mag. ; th^ may be said to belong to a sub-vanety of Saur assigned to Cow- 
herds — {Line 13 ) hiaa- Bala 54 is likewise not Mag — Thus the distinction 
that P tnes to draw between the fSaur and Mag treatments of r {line 9) on 
the ground of the instances cited by him in the first paragraph is illusory •— 
{Line 18 ) 8. -uttm ifc. PratijtHa 44 is noteworthy only as an orthographi- 
cal peculianty , for the dision of medial v in these mss cf. Printz 19. The 
V of -vutti’- has been correctly retained in sdhdranavuttim Cam. 7. 5. uttanta- 
{vrttdnta^) Pratijna. 18, Abhi 24 appears not to have even that justification 
Tliere should be no hesitation m correctmg the text reading to vuttanta, 
since the former appears to owe its existence to the influence of such doubt- 
ful forms as pautta- Pratijna. 51, sampauda- Bala 9 — {Line 24.) amida- 
Bala. 39 is £107} taken from the speech of Vfddhagopalaka and is therefore 
not Mag. 

Page 9 {Line 4.) As we find yeva (ie eva with prefixed y-) even in 
the Old Saur of the Turfan fragimeints (Luders 59), the 6aur. e{v)va of 
our mss, would appear to be an orthographical blunder , it is probably nothing 
more than a Sanskntism 1 — {Line 32.) In odaradi via {avatarati iva) CSm. 

^8 Among othersf Mattavilasa (Trivandrum Skt. Series, no. 55), Kalyajjia- 
saugandhika {BSOS 3, 33 ff.), and the prologue of the AScaryacudamam {BSOS 3, 
116 f.), besides the southern recensions of classical and post-classical dramas pub- 
lished in the Trivandrum Skt Seneal and elsewhere — ^For important additions of 
lengthy Magadhi passages unknown to Pischel and perhaps to Printz, see now 
E. Clark, MSgadhi and Ardhamagadhi JAOS 44, 96, footnote 44. 




51 etc. there appears to be a confusion between the use of the enditikon 
-vva (with subsequent dision of one v and compensation lengthening) 
and that of S The alternative forms are odaraditMt (for odaradi- 

vva) and oiaiadi wto; the hybrid forms of our mss. appear to be utterly 
without justification. 

Page 10. {Line 13 ) There is no shortenmg of the end vowel in haddht 
{hd dhik), which is arrived at by a regular elision of the final consonant ; 
on the other hand there is an anomalous lengthening of the end vowel in 
haddht cited by SP. from iSak. — {Line 17 ) As the short final of vocafaves 
of nouns ending in -a alternated frequently with the pluti voWel,i® it is highly 
irnprobaHe that the mitial of khu should be doubled just after a vocative, 
when It IS not doubled in any other position Weluk (ed Bfilacanta, p 38) 
is therefore perfectly justified m emendmg the ms. readmg kkhu (m the four 
isolated cases in) B^a 34 to khu. — [Lme 21 ) dhikkhu=dMk-khalu and not 
dhik khdu, — {Line 30.) As the Old Saur, of Turfan fragments shows yfXaa, 
the form ;S tdisavcmnayyeva {idrsavailnS+eva), condemned by P, appears 
to be correct Pkt. • on the other hand the spunous forms e{v)va, approved 
of by P, have all the appearance of being unauthonzed Sandcritisms^ as 
already remarked.— (ij«e 34.) durattanayyeva Bala 18 is not Mag 

Page 11. {Line 9.) § {-matra-) ifc occurs likewise m Kalyiaqa- 

^gandhika (ed. Barnett, BSOS 3, 37), ettiammatto maggo. {108} If it 
is an archaism, as it appears to be, it is probably raie common to all Mala- 
yalam mss , and not peculiar to the Tavandrum plays Hema. 1 81 cites, 
as a matter of fact, both variants matta- and meita-. —{Line 34.) P. implies 
that the form purusa- is older than purida- It may be so. But Markapjideya, 
Prakrtasarvasva 9.9, aasigna purusa- to Baur and punsa to MahaiastrS 
This suggests that the difference between them is really a view fully 

cndoised by the ground-form *pursa- (Wackernact^l, Alttnd. Gram 1 § 51). 
In the northern mss , the MaharSetiS form punsa- appears to have been 
stereotyped. In our mss, however, pun^a- may be merely an incorrect (or 
accidentally correct) Tadbhava 

Page 12, Une 20 S. anhadi, etc. I adhere to the views expressed in my 
Studies in Bhasa I, JAOS 40, 252 f., despite the remarks of Printz on p 46. 

Page 13, hnt 3. With -puruva- {-purva-) ifc. of our mss. compare 
dmiu^ntruivo] of the Turfan Fragmaits (LibERS 50), not noticed by P. 

Page 14, tme 19 P. mentons oggada- Bala. 9, 12 as an exception to the 
rule that the prepoertion apa appears invariably as ava- , but, as a matter 

1* PteCHtt's observation is that iva becomes -vva after short vowels awradi- 
mta hoU, sad coly tor Mahar., Aidham. an] JaoB Makar. (Craaroa PM. Spt. 

oLOCHi La formation de la larngue TTUnrathe, p 180* 



of fact, it is better to trace oggada- to udgata-^^ than to apagata- (proposed 
by Qi&ya) ; for the instances of the change of short u to short o, see PriN'IZ 

Page 15 (Lme 18.) The explanahoci of sutthu Idem Bala. 42 (proposed 
in the Chaya and accepted by P.) is unsatisfactory In view of sutthn 
gctidam in the parallel passage Fa&ca 22 , either read $uttkti tdath, or 
correct the text to sufthu g&d<m, following Weller, ed Balacanta, p 49. 
There should be really no hesitation m makmg the correction, smce the text 
of the Trivandrum edition is based on one single ms, which swarms with 
mistakes — {Line 11. )i The changie of -th- to -h- (mstead of -dh-) in 6 aur. 
appears to be a characteristic of these Malayalam mss , thus KalyS/jas. (ed. 
Barnett) has kahath (pp 36, 37, Skt katham), nSha- (pp 41, 48, Skt 
natha-), etc. Similarly in the extract from, the Prologue of the AscaryacQJa- 
ma^ii {BSOS. 3, 117) pubhshed by Pisharoti 

{109} Page 16, line 15. P does not give the reference for agham -dhik; 
but I expect that the CMya spells it correctly as dhik 

Page 17. {line 13.) As regards the diange of cch to sc, it should be 
remembered that the rule is seldom f oUowed in the mss. of dramas. Pischel 
admits that the texts have mostly cch, and although he adds that the mss. 
show distinct traces of this rule, he cites only instances from the Miccha. and 
the Com. IVthvSdhara To judge by the dramatic texts published in the 
Trivandrum Sanskrit Series (such as the Mattavilasa, SuUiadiSdbanatnjaya 
and others), the Malayalam mss show uniformly cc/t .” — {Line 14 ) The 
instances F quotes for the retention of y ui Magadhi have been taken mostly 
from the speeches of Cowherds in the two Krspa dramas, and are therefore, 
for reasons already given, perfectly irrelevant. Moreover, the instances atcd 
for the irregular change of y to 7 in Mag are more numerous than for the! 
correct retention of y Conversdy, the instances for the incorrect retention of 
y in iSaur are almost as numerous as those for the regular change of y to In 
fact, the treatment of y-j in the mss. of our dramas is inconsequent to a 
degree, violating all rules of Pkt. grammar, and cannot therefore be made 
the basis of any inference like that drawn by P 

Page 18. {Line 16.) The rule regarding the change of r to 1 is not appU* 
cable to the cases F. has in view, the dialect in question not being M3g. ; so 
there is probably no text qomiption . — {Line 36.) I for | appears to be a 
characteristic of Malayalam mss. ; cf KalySpas (ed. Barnett) p. 41 lak- 
khaadi, p. 42 saggo-laccH, p. 49 bahalattana . — But it is never carried out 
quite so consistently as in the Trivandrum texta My surmise is that the 
editor has normalized the spdlmg and wntten 1 throu^out, irrespective, of 
the ms. spdling. 

i« Apis'S Dictionary gives sub voce ud-gam- the meaning ‘to depart (as life).’ 
W See W. E. Clark, JAOS 44 82-93. 



Paee 20. (Line 13 )' The confusion between the Saur. and Mag treat- 
ments and -««-) of the Skt -jn- is so complete in our mss. and ^sides 

so in all classes of mss. that to my mind it is most uncritical to 

assume that -tin- has ciept mto our texts through contarnmatira with younger 

texts. (Line 15.) The examples yanna Bala 9 and lanfio Bala 10 cited by 

P, as Mag. {110} are not Mag . — (Line 17.) The treatment of Skt. -ny- is 
analogous and P. himself cites a very lUuminatmg example : S dakkUnnada, 
sadakkhinna, sadakkhinr^a, and adakkhima—(Line 33 ) Owmg to the un- 
certainty charactenzmg the ligature -yy- in southern mss , we cannot attach 
much importance to the gielling uyydna- (udydna-) Avi 2, 4 ; it may be 
read as uyyana- or as u}}difo- , see below 

Pfige 21 (line 4 f.) The examples ama-, kimtad, and tfdsa- ated by 
P. from RaJa are not Mag , but, as pointed out often enough above, a variety 
of Saur — (Line 12.) Barnett in his edition of the Kalyanas. (BSOS 3, 36, 
footnote 5) states that in his ms the word ayya is spelt and therefore 

in all likelihood the Trivandrum mss. also follow the same orthography, 
although Ganapafa Sastri is silent on the point It must thus be regarded as 
still uncertain whether the hgature is to be read as -yy- or as or again be 
looked on as representing a sound intermediate between the two (Pischel 
193, 284). P adds that the readmg -yy- is assured, because of the hesitating 
orthcgiaphy in words like myyadedi- madedt, but in this P. is grossly mis- 
taken ; for P admits that is preserved only — or at least mostly — at die 
point of contact in a compound, but is elided generally m the middle of a 
word (Printz 15) ; ntadedi may tiierefore stand for wddedi as well as for 
ifiyadedi, since intervocahc -;- is dropped in the same way as intervocalic -y-, 
cf mieut- (antaja) Avi. 14, puama (pujaniya-) Cam. 34 rdS (rajd) Svapna 
6 etc. Thus it is evident that it is a futile attempt to try to place the treat- 
ment of Skt -ry- in our dramas on the same footing as in the Turfan Frag- 
ments “ 

Page 22, line 13. The Chayla is perfectly right m explaining the com- 
pound sattKhida- as ^asfhikrta- ; see Morgenstiekne, Ueber das Verkaltms 
ztoisehen Cam. u. Mjccha. 30 The rendering of P. is grammatically fault- 
less; unfortunatdy it makes no sense. Expand the compound sastM(sam- 
bandhi)krtadevakarya- (= krtasasthisoahbcmdhidevakarytt-) ‘one who has 
performed the religious duties {111} (pertaining) to the sixth' , for the trans- 
positkKX of the members of a compound, see PIschel 603 , for the significance 
of the sixth, see the discusaon on the tithi scheme and the time analysis 
of the CBm. in my Studies in Bhasa III, JAOS, 42, 67 ff. Lastly, it may be 
pointed out that the usual reflex of -utA-, in our plai^, is -tth- and not -tfA- ; 

a* See also liis footnote to Pishaboti's transhteratiou of the Prologue of tbe 
Ascaryacu^maoi, BSOS 3, 116. 

** Lbsnt (ZDMG 72, 207) has fallen in the same trap, throxlgh the omisisioii 
of the editor to rgmrt about the orthogcai^cal peculiarity of southern mss. 


cf attkavdvdrd {arthavydpdrd) C&m 10, aitha {artha-) Svapna 54, and 

Page 23, line 12. P has failed to notice that nikkhanta- of our dramds 
has a parallel m nikkhanta- of the Turf an Fragments (Luders 61) 

Page 26. {Line 12). The Mag m which the Noon Sing of thema- 
tic stems ends m -o, as already remarked, is no Mag.^o Printz’s treatment of 
the dialect of the Cowherds as Mag has been rightly rejected by Weller, 
ed B^a. Voiwort, p. in f. — {Lme 14 ) Better to correct the text reading to 
I^andagovaputto pasudo B^a. 35 as Weller (op, at p. 4l)0) has done, be- 
cause the construction of a loc abs. with jadappahudi is harsh. — (Line 15.) 
The Ardham. m which the Nom. Smg. of thematic stems ends in -o is no 
Ardham. — {Line 30.) It is a notable observation of P. that in the ploys 
before us there are mstances of Acc. Plu, Masc. ending in in Saur and 
Mag. But his remarks on the subject call forth following comment (1) 
All the examples ated by P but one are from Saur . ; the exception is amhd- 
libakdni Cam 14. (2) With the exceptions of two adjectives, tddtsojyt and 

amhdliiahdni, all the words refer to maiumate objects {kesa, gucchaa, gumhaa, 
guna, padra, mdsaa phana, saadaa and pafaha). (3) In the example tdfii 
ddva sekdlidg!Uimhadni pekkhdini kusumiddrd vd vetti Svapna. 33, gut^iha- 
dni IS Nom. Plu. and not Acc. Plu. P. was evidently misled by the position 
ot pjAkkhdmi and has taken gumhadni as its object The object of pekkhumi, 
however, is not s^mhadnt, but the whole sentence tdni (112'} ddva sehdltiP 
etc. (4) Pkt grammanans (Hema 1. 34) permit the optional forms gmd 
fm) and gvmMm (n.) and therefore the suggested change m C^. 47 is 
quite uncalled for ; P. has here agam been misled by the Chaya The text 
readmg is dkaptmisapakkhavddidd savvagundnam hanti , and P wants to 
correct the text reading gundnam to gundm ; but gundnam is clearly nothing 
more than an mcorrect contraction of gtmd nam, (5) It is questionable 
whether iwe have to correct pdndni (Svapna.) to pdi^d, or to correct pd^d 
(Pratijfia^) to pdndni ; or agam to let them both stand, like so many doub- 
lets in Pkt. (6) With regard to mdsadni, it ^ould be remarked that m Cam. 
5 the Nom Plu has the identical form mdsadni,^^ which: makes it doubtful 
whether m Pratijna the word is used as mas. or as neut. (7) If iakata is 
n., saadaa- could, I think, quite easily be also n. I am not able to check the 

20 The use of the cerebral s is certainly peculiar Though unnoticed by 
Prakrit grammarians it is not altogether unknown to PTaknt orthography The 
Shahbazgarhi, Mansera and Kalsi versions of ASoka's edicts are full of words spelt 
with the cerebral A feW examples chosten at random are : Pock Edict XII Sh 
savropra^amdam, M savrapra^adam, K ^avdpd^idani ; III M parisa , XIH K 
athi amt^aye; ibid, athava^htsita^d devdrtarh ptyasa Piyada^ne Idptus , VIII Sh 
dc^ava^ahhidtQ sato 

21 The text reading is : avia dakkhi^tnds^aa^ bhavhsanti, repeated by the 
Vidu^aka on p. 6 of the text 



example, as P has omitted to give the reference to the text (8) As a refer- 
ence to the Petersb Diet will show, pctdhcL- is used sometimesi as n, and 
what IS more important is that the passage cited (B&la. 62) is not Ardham. 
(9) P has not given a single instance of any of these words being actually 
used in these pktys with tnasc. ending to show that they are used m (the 
Prakrit of these dramas as masc nouns ; on the other hand, he has cited 
rp. 25) a number of cases m which the gender has actually changed from 
m- to n. : mkuSa, purusakdra, guda, naraka, vdsa^ svapna, and tandula, 
some of which are used in both genders indiscriminately. And as pomted 
out above, none of these words (with the exception of the two adjectives) are 
piotected against neutralization by their meaning, as they are all names of 
inanimate objects, (10) Lastly, it is worth remembering that Prof. Luders, 
after a most exhaustive and minute investigation of the entire material, has 
succeeded in establishing this peculiar form for Ardham. and Mag. only , 
for Saur. its propriety is still questionable (Luders, Epigraphische Beitr^e 
III — Siizungsb Preuss. Akad 1913, p. 1009) It should seem then that 
while there is a distinct possibihty that some of the instances dted by P. are 
Acc. Plu Masc. formed with the termination -5m, in others there has most 
probably been a change of gender. The claim of P is justified to £113} a 
certain extent, but it is undeniable that P,. considerably over-shoots the 

Page 27. (Line 23.) The propriety of assummg a Loc. Sing. Fem. in 
*aa7n is questionable ; we should sooner assume an unauthorized Sanskritism 
— (Line 27.) Caru. 79 has been correctly construed in the Chaya as 

Nom. ; PRINTZ has been apparently misled by the text reading vaddanti, 
which IS only a misprmt for vddimti, duly corrected m the second edition 
(p 97 ). — (Line 32) There is no need to correct Ujjattfio to Ujjaiifie in 
Svapna 21, 22 (first ed, pp. 20, 21), since Ujjaimo is not Gen. Sing, but a 
non^al adj. (= Ujjaymika- or Umytnlya-) derived from Ujjayinl ; P. has 
again allowed himsdf to be misled by the Chaya, 

Page 30. (Line 2.) P. has misunderstood the passage cited by him ; 
the subject of bhavissadi is uvdas^xt$h and not tania which is the predicate ^ — 
(Une 3.) tmi Svapna. 33 is not Acc. Plu. Masc. but Nom. Hu, Neut. (see 
above). (Line 35.) It is uncertain whether ima^i Pratijfia 46 should be 
regarded as Masc. or Neut, since masadni Garu. 5, 6 has been used once as 
Nom. Plu. (see above). 

Page 31, tines 29-31 )S, satthif satlctm, and atthami refer to the day of 
the lunar month, and not to the hour of the day , cf atfhanu khu ajja Cfiru, 
53. Further kmaf^tatm Pratijna 50 is not the ‘ black eighth hour/ but the 
eighth day of the daik fortnight of Sravana when Kpgaja was bom, a day also 
known as Kji^ioaiSftarni. 

Page 34, Um 27. The text reading vSdaanti Caru. 79 is only a nusprmt 
as already remarked, for v^antt, corrected in the second edition. The sign 



of the medial i was displaced and knocked off by the superior Devanagaii 
figure 4 The Chaya correctly renders it as vadymte, a fact which should 
have put P. on the right track 

Page 35, line 36 It is not quite dear to me what P. means by future 
forms with thematic -i-, unless he is referring to forms like ukkanthissidt 
Svapna. 17, jivisstdt Dutagh. 54 The Turfan fragments have preserved 
pavva]issiti (Luders 48, footnote 1) 

Page 36, line 23 No need to correct pucchiadi to pacchiadi, if the sen- 
tence IS understood aright , see Belloni-Filippi, Note {114} cribche ed 
esegetiche al “ Carudatta ” di Bhasa, Riv, studi orient 9, 586. 

Page 41, line 16. In explaining aamia Pratyfin 11 as Abs. of rt. gam, 
P. follows the Chaya, and has been misled agam , for by readmg the passage 
himself, he could have seen that agamya in that context does not make any 
sense ; here Sarma is obviously = dcamya, dcamana bemg a ceremony which 
always precedes the pra^ama. The stage direction dcamya is particularly 
frequent in these plays. 

Page 44. (Line 11 ) 6 dma occurs in the Byhatkathadr^asapigraha 5 
114 and 9 70, as pointed out by Winternitz, Ostasiat Zeitsch, 9, 290, and 
in Mattavilasa. — ^(Line 19 ) iS. uvai^hm^a Avi. 79, to judge by the context, 
is not ‘ Waschwasser,’ but some other accessory of the bath, perhaps oint- 
ment — {Line 26.) The reference for kumbhavalda has been left out inadver- 

Page 45 (Line 2.) If tunnid is the same as tunhid of the second edi 
tion (p 21) it will hardly be necessary to assume the improbable meaning 
‘ Schwiegeitochter ’ for an imagmary word tunnid, smce tunhid is a r^ular 
derivative of Skt tusrdkd ‘ silent,’ which gives a thoroughly satisfactory 
sense, see my translation (Oxford University Press 1923), p. 21. — {Dne 
4.) The successive steps by which pcmkhu Bala. 14 is reached appear to 
be these . Skt pdmsu > Pkt pdmsu,^^ pamkhu, peankku ; whether the form 
IS valid and admissible is another question ; about the meaning, however, 
there cannot be any doubt ; see Weller, Die Abenteuer des Knaben Krischna, 
Anmerkungen, p. 94 — {Line 7.) Instead of correcting vadtvassaa- CSru. 1, 
4 to padivassaa^ (as suggested by P.), adopt the reading of ms. kha, padt- 
vcssa- (Skt prativeiya-) CSm 4 footnote — {Line 13.) padisard is, as 
Ganapati Sastri in his commentary to the second edition of the PtatijSa. 
explains, a charmed protective thread worn round the arm {hastadhdryath 
rak$dsutram) ; m support he quotes Keiava : pratisarastu sydd hastasiitre 
nrsandayok \ ,. vranaiuddhau ca kecit tu stnydfh pratisardm viduh 1[. — 
(Line 21 ) For S. landuo, see now Morgenstierne, Ueber das VerhaltnU 
ewischen Cdru u. Mfccha. p. 27 f., who has undoubtedly proposed a very 

23 Wackernagel, AlHnd. Gramm, 1 § 118, 


satisfactory explana-tll5>tion.— (Line 22) For loht also see Morgen- 
STIERNE, op. dt. p 26, who refers to a Divyavadana passage cited by 
IvIONlER-WllxiAMS. — (Lwe 24.) Both the fonii and meaning of honLi- 
Svapna. 59 are quite clear The Chayia nghtly eitplains it as kunkrti , honti 
-huth‘ti for hum itt, lit. ‘huiii’- making, that is, followmg the narrative 
with the ejaculation ‘hiDh,’ m order to show continued attention ; see my 
(O.U.P. ) p. 57 and explanatory note 20 Cf. the analogous derivatives 
jhai-tti, tad-itt, and see exanples in KaSifca to Paajini 6 1. 98. See also 
now Belloni-Filippi, Rw siudt onent 10, 370. 

We will now revert to p . 5 of the thesis, where Printz has presented in 
a collected form the most important peculiarities of the Praknt of these 
dramas, which establish, according to him (p. 47), the antiquity of the 
dramas, as also m a remote manner the authorship of Bhasa. In regard to 
these alleged peculianties,!!^ I have to submit the followmg remarks and 
reservations : (1) metta-Cmatra-) according to P is later than matta-. It 
may be so. But matta- is maitioned by grammanans like Hemacandra and 
occurs in the KalySnas also It cannot therefore be said to be peculiar to 
the Tnvandrum plays. (2) Svarabhakti u m purusa- is correct m iSaur 
according to MSrkandeya (3) (instead of -puvva-Skt. -purva-) ifc 

is found in the Turfan Fragments, and may therefore be regarded as a genume 
archaism. (4), The regular co'ehralizaticHi of 1 is a characteristic of Mala- 
yalam mss., also found in most of the southern editions of claAsical dramas 
recently published ; it is not a peculiarity of the Tnvandrum plays. (5) 
In the hesitaboin between the reflexes -ijn- and -Hn- (Turfan-^-) of jn- I 
see a confusion between the Saur. and MSg forms, an explanation which 
harmonises with the frequent representation of -ny- by -ntr- (the Turfan m-in 
show -Hn-). (6) The alleged change of -<fy-(m ud-y-) and -Vy- to -yy- is 
un-{|ll6^-oertain, since the symbol used m southern mss to represent the 
ligature is ambig^uous. These doubts are only strengthened by the incon- 
sequent treatment of initial y- (7) The change! of -fe§- to -kkh- instead of 
-cch- signifies nothing relative to the age of the plays (8) Some of the 
instances of Acc. Plu. Masc. ending in -aiji cited by P. are valid ; others are 
doubtful or spurious. (9) Nom. Acc. Plu Neut. in -wti appear^ to be a 
Common, if not the regular, form m Malayalam mss (10) The Loc. Sing. 
Fwn, ending in -3aan, as wdl as attaifcoh (for attanoioh'), I r^ard as Sans- 
kritisms, as there is no authority for them anywhere else (11) vaaiti, 
amhaath, tava, and Idssa are true archaisms, as they are docum^tsd by 
actual instances in the Turfan Fragments. But it appears now that they 

Sse also W, E. Clark, JAOS 44, 101 f. — C lare takes exception to my use 
of the ' anhaism,’ but there can be, I think, no question that the forms men- 
by me are ‘archaic’; that is to say they bdong to the '‘Old Praknt’ in 
omtototinrtKm to the rest of the Prakrit of the dramas^ which is mostly ‘Middle 
Piwcnt. That is omdly the sense in which I us© the word ‘archaic,’ 



aie not peculiar to the Trivandrum plays, since they are also found in othei 
Malayalam mss of, in part, very late plays such as the Mattavilasa, NSga- 
nanda and others. (12) kocct I am unable to account for (13) In view of 
the ge)}k- of the Turfan fragments, ganHadi appears to be a miaformation, a 
hybnd Tadbhava (14) The Part. Pres. Pass in -iamdna-, I am inclined 
to regard with suspicion (15) As has be&a observed by P. and other 
writers, these mss contam clear instances of thd inhibition of simplification 
of double consonants and compensation lengthening. Malayalam mss in 
general, as appears from text editions of dramas published in recait years, 
favour this inhibition (16) kana and gacchta are true ardiaisms ; but 
aamia should be deleted from the list, since it is a reflex not of agamya but of 
acamya (17) The use of ma with Imp, Inf or Abs. and the employ- 
ment of Part. Perf. Pass, as nomm actionis ate matters of style and have 
no beanng on the question of the age of the plays 

The more important of the general observations regardmg the Prakrit 
of these plays scattered through the above pages may be convemaitly sum- 
marized as follows Firstly, evai if these plasms be BhSsa dramas (or as 
some scholars think adaptations of Bhasa dramas), the Prakrit they contain 
IS not necessarily BhSstfs Prakrit, smee our mss. are barely 300 years old. 
Secondly, owing to faulty classification Printe’s citations of Mag and 
Ardham. forms are useless for purposes of dialect dififerentiation. Thirdly, 
£117} we cannot be sure that forms like matta (matra)\ pmusa (puru$a), 
eva are archaic, or even legitimate Praknt forms, unless we find corroboration 
from more reliable sources , they may be mere Sanskritisms. Fourthly, tlie 
treatment of the ligatures ny, ry m our mss is confused and inconsequent ; 
hence in regard partly to the near possibility of confusion between Saur and 
Mag. forms, and partly to the ambiguity of the symbol represaitmg the liga- 
ture jj-yy, Printz’s attempt to bring the treatment of these conjuncts in a 
line with their treatment in the Turfan fragments and to base thereon chrcaio- 
logical conclusions regarding the stage of development of Bh&sa’s Prakrit 
may be r^arded as having signally failed Fifthly, the most important con- 
tribution to the subject made by Printz is to have shown that the mss of 
our plays contain some instances of the Acc. Hu Masc. ending in -am though 
the instances are not quite as numerous as Printz supposes them to be 
Sixthly, besides this noteworthy form the mss contain a few more instances 
of genuine Prakrit archaisms , but as these latter are met with also in Mala- 
yalam mss of classical dramas and of even later southern productions, the 
Praknt argument is inconclusive and cannot by itself be safely made the 
basis of dironology Seventhly and lastly, a satisfactory solution of the 
Bhasa question cannot be reached from a study merdy of the Prakrits of 
the plays. 

Jrdy 1924 , 


The previous historyi of the discussion centering round the thirteen 
unonymous drsmus discovered by Pundit Gsnuputi Sastki £Uid attributed by 
him to Bhasa is sufliaently well known, and there is no need to rqjeat it 
here in It will sufSce to observe that many distinguished scholars, 

whose researches in Sanskrit hterature entitle than to speak with authority, 
fully agree with the learned editor of the Tnvandrum Sansknt Series, and 
whole-heartedly support him on attributing these plays to Bhasa. The theory 
has not however won entire satisfaction. Promment among the dissenters are : 
Ramavatara Sarma Pandeya, Barnett, Bhattanatha Svamin, Rangacarya 
Raddi, Kane, and (latterly also) Pisharoti, who all agree m placing the 
dramas after the seventh century a.d., and m regarding them as the work 
of some paltry playwri^t or playwn^ts Between these eartremes lie the 
views of WiNiEBNiTZ and myself. We accepted the Bhasa theory, but not 
without some reserve, while recognizmg that the propotmder and the sup- 
porters of the hypothesis had a strong prima fade case, we held at the same 
time that the evidence adduced did not amount to a conclusive proof (see 
above, vol. 26, p. 232). 

* * • 

One peculianty of the BhSsa problem appeails not to have been clearly 
realized by most previous writers on the subject This peculiarity is that 
there is not a sin^e argumait advanced on either side that may be regarded 
as conclusive and that has not been, or caimot, be, met by an almost equally 
sound argument on the opposite mde 

Let us coisider some individual instances Take the fact that the title 
of the work and the name of the author are not mentioned {127} in the 
rudimentary sthapana of these plays This omission is explained by the sup- 
porters of the theory on the assumption that in pre-dassical times details 
like these were left to the preliminaries and are therefore not found m the 
sthdpand,^ The explanation possesses a certam degree of probability, but 
nothing more since it mvolves an unsupported and unproved, though plausible, 
ass^ptinn. On the other hand those writers who deny the authoriBhip of 
Bliasa etxpl^ the onussion on the ground that the plagiarists or adapters, 
whose handiworks these dramas are, had very obvious reasons to remain 

* [/RJJltAS NS. 1. 126-143.] 

^ Bibliographical aaatenal will be found in my “Studies m Tth^ga (V) ”, 
above vd. 26, pp. 230ff, 

® The Smskrit Drama (Oxford 1924). p. i^i. 


namdess, an assumption, on tlie face of it, not less improbable than the 

Neat take the lack of accord with the rules of theorists like Bharata, as 
seen in the admittance, mto our plays, of stage fights and death scenes, which 
were avoided m the classical drama, and are in part expressly forbidden by 
Bharata This has beai utihzed by the protagonists of the theory as another 
proof of the antiqmty of the plays But this explanation, like the previous 
one, has all the appearance of being another subtle attempt at exploiting our 
Ignorance of pre-dassical technique, being in the last analysis nothing more 
nor less than a deduction from the a proH assumption that the plays in dis- 
pute are pre-dassical. The Mahabhasya passage enlisted by Keith (The 
Skt Drama, p. 110) m this connection does not m any way countenance the 
assumption ; for Weber’s theory of mimic killing of Kaimsa and miauc bind- 
mg of Ball, iwhich has repeatedly been shown to be inadequate, must, uii- 
fortunatdy, be finally abandoned now, after the condusive proofs brought 
forward by Prof. Luders* to show that the Saubhikas and the Granthikas 
were both merdy raconteurs or rhapsodes — ^The conflict with the mles of 
treatises on rhetorics admits of another e^lanation, which must be pro- 
nounced to be quite as plausible as the former, if not still more so. These 
mnovations, it has been urged, have been introduced in quite recent times 
with a view to producmg a more arresting stage effect, to striking a more 
popular note m the presentation of Sanskrit pla 3 rs ; and there is ample evi- 
dence to show that these plays have indeed been very popular, as stage 
{128} plays, m Malayalam, where some of them are even now r^larly 
produced by professicMial, hereditary actors, locally known as Cakyars anJ 
Nangyars (Pisharoti, BSOS 3, 112 f.) 

Then there is the argument based on similarities m diction and ideas 
between these plays and some cdebrated plays such as Sakuntala. These simi- 
larities are dearly equivocal While they can on the one hand be used’ to 
prove that the strikmg ideas of the author of the anonymous plays have been 
ftedy borrowed and amplified by others, they can on the other hand be also 
used, with equal cogency, to support the view that the anonymous compilers 
of these plays have found m the works of dassical di;amatists a splendid 
hunting ground for bans mats and happy thou^ts.® And the protagomsts 
of the theory have to admit that no strict proof of indebtedness is possible. 
Keith (pp. dt p 124) confidently assures us tiiat “the evidence is suffi- 
cient to induce conviction to any one accustomed to weighmg literary evidence 
of borrowing." Yes, but what is the test of one’s being “accustomed to 
wei^ng literary evidence of borrowing ” ? Presumably, tiie susceptibility 
to the conviction being induced ! 

® Pisharoti, BSOS 3, 115 * "Die Saubhikas,” SBAF 1916, 698 ff. 

® See for instance Ganapati Sastri in the Introduction to his edition of SV. 
6 Cf. Raja m Zeitschr. f. Ind. u Iran (.ZII.) 2, 260 


studies in bhasa 

Then there are verses in these dramas that are found cited or criticized 
m different treatises on rhetoncs They have been used by those who favoiu 
the Bhasa theory to corroborate their view that these are works of a very 
considerable writer, who could be no other than Bhasa. The rhetoricians 
bemg mostly silent on the point, we do not know that the verses quoted were 
taken from dramas by Bhasa It cannot however be demed that the view 
can claim for itself a certain degree of plausibility — On the other hand it is 
also not quite impossible that these verses nught have been appropriated for 
then own use by adapters at a moment when the creative faculty, being too 
severely taxed, had refused to function further. 

Great capital has been made by the opponents of the theory out of ceitain 
verses which are cited as Bhasa's in anthologies of Sanskrit verse, but are 
not found in the present plays ’ The ar- {129} -gumeit is not as sound as 
it at first sight appears. It is easy to eicplam their absence on the hypothesis 
that the supposed author had written further plays or poems which may be 
tlic sources of these citations (Keith, op. at p. lOS). And if that does not 
suffice It may, with some plausibility, be urged that these vexses have been 
excerpted from some lost recensions of these dramas We need only recall 
the well-known fact that m the third act of the Bengali recension of iSakuntala 
one scene is four or five times as long as the corresponding porticai in the 
Devanagaii recension , even the names of the dramatis personae are in part 
different in the two recensions.^ As a last resort one may even enlist the 
unquestionable facts that m these anthologies the names of authors are fre- 
quently misquoted, the same verse is attributed to different authors, and 
finally verses attributed evoi to Kalidasa and other edetwrated dramatists ate 
not found in their extant works. 

1 have so far dealt with some of the minor arguments advanced on 
either side and tried to show that they are utterly inconclusive. There are 
however some aiguments that are considered by their propounders as decisive 
m character, and to these we shall now* turn our attention. 

One of these arguments is that our plays are begun by the Sutradbara, in 
contradistinction to the classical plays, and that this characteristic of the 
plays by Bhasa has beat pointedly alluded to by Baoa in the distich in which 
be celduates the great dramatist This argument on which the supporters 
of the theory place so much reliance is doubly fallaaous, and the great effort 
made to find in this fact a proof conclusive of the authorship of Bhasa must 
ddinitdy be pronounced a failure. The veise from the Harsacanta states 
merely that Bhasa’s dramas were begun by the Sutradhara It is the perver- 
sion of all probability to find m this innocuous statemait a distinguishing 
characteristic of BhBsa dramas, because every Sanskrit play we know of, 

^ Cf, Ratnavatara Sarma Pandeya, SSradS, vol. 1, p, 7. 

• Sakuntali. ed. Monikr Woxiams (Oxford 1876), Preface^ p. viL 



all the dramas by Kalidasa, Harsca, Bhavabhuti and other dramatists, were 
likewise begun by the Sutradhara. The latter fact is somewhat obscured by 
the drcumstanoe that mstead of the correct shorter foimula nmdyante 
£130J sutradharah, some northern manuscripts read : 

ndndyante tatah pravUatt sutradharah, 

these words being placed between the benedictory verse (or veises) with 
which all dramatic manuscripts begm, and the introductory prose speech ot 
the Sutradhara. When the stage direction reads merely nandyante sutradha- 
rah, there is no question that the Sutradhara does not enter at the point 
where this stage direction is inserted, and must be supposed to be cm the 
stage already, for the simple reason that the manuscripts contain no stage 
direction announcing his entry. Who recites the nandS follows from the direc- 
tion of the Ndtyaidstra of Bharata (Ed Kavyamala, adh 5, v 98) ; 

sutradharah pathet tatra madhyamam svafam dsntah 
nandim . . 

In view of this dear statement of Bharata, can we legitimately draw any 
condusion other than that the nandi of the classical dramas was recited by 
the Sutradhara hunsdf ? Thus, according to the testimony of the vast majo- 
rity of manuscnpts and conformably to the mles of rhetoricians, the proce- 
dure is that the Sutradhara first redtes the benedictory stanzas (with which 
manuscripts of all dramas commmce) and thm proceeds with the prose 
speech assigned to his role The words nandyante sutradharah of the northern 
manuscripts then mean : “ at the end of the nandf the Sutradhara (continues 
speaking)''. This is the view of the commentator Jagaddhara,® and it appears 
to be perfectly sound. If it is admitted that the plays without exception were 
begun by the Sutradhara with the redtation of benedictory stanzas, it is clear 
that the positicm and the wordmg of the first stage direction has nothing 
whatsoever to do with the question whether the play is begun by the stage- 
diiector or not The only difference betwe^ the manuscnpts of the Tnvan- 
drum plays and the northern manuscripts of classical plays is as regards 
nomenclature, as has been already pointed out by Winternitz {Ostasiat. 
Zeitschr. 9, 285) Such being the case, it cannot any longer be maintained 
that BSna had the intenticHi of drawing atteation to any distinguishing char- 
acteristic of Bhasa’s works by saying that his plasre were sutradhara- 

kjtarambha. Baija's only object is, as Keith (op cit p. 91) has justly re- 
marked, “ to cdebrate Bhasa’s famei and to show his wit by the comparison 
m the same words with some not very obvious object of comparison.' Baca's 
verse is merdy a subhasita, as will now be admitted by every unbiassed 
critic. 'The discussion whether m this verse from the Harjacarita there is an 
allusion to' some technical innovation of Bbfisa in diorteoing the pieUminaries. 

* MSlatimadhaua, Ed. Bombay Skt, Senes, p. 6, 



combining th6 functions of tho Sutmdhara and the Sthapalca, taking the prof^ 
logiie away from the Sthapaka and placing it m the mouth of the Sutra- 
dhara and much othei vague speculation of the kind (Lindenau, ffhasth 
Studien, pp. 10, 37) is mere verbiage. The Tnvandrum plays at any rate 
offer no occasion for the discussion of these questions and, what is more 
important, furnish no answers to them. 

Our conclusuKis on this point may be summarized this . (1) the nandi, 
which us^ to precede all dramatic rq^resentations, being mvanably recited 
by the Siitradhara, all Sanskrit dramas are suttadhdrakrtdrambha , (2) it is 
thus wholly inadmissible to regard this attribute as speafymg a distinguish- 
ing characteristic of Bhasa’s dramas , and therefore (3) the argument which 
sp(>Va m the position and the wording, in our manuscnpts, of the stage direc- 
bon nSndyante etc. a proof conclusive of Bhasa's authorship is utterly devoid 
of cogency Furthermore, it has now been shown that all Malayalam manus- 
cnpta of dramas bepn in the identical manner. If it then still be true (as 
Keith asserts, Ind. Ant. 1923, 60) that “ by this decidedly noteworthy fact ’ 
(namely, that these plays are begun by the Sfitradhara,) they are “eligible 
to be considered Bhasa’s then all Sanskrit dramas are likewise digible to be 
considered Bhasa’s ! 

Several efforts have been made to prove m these dramas traces of later 
date than Kilit^sa; but most of the arguments,^** as has in part already 
been shown, are (giite inadequate to support the condusion. It is also im- 
possible to find cogency in the argument advanced first— to my knowledge — 
by Kane,'^*^ and then repeated recently by Barnett^* that the Nyayasdstra 
of Medbatithi men- {[132^ -tioned in the Fratima is the same as the Manu- 
bhS^ya by Medhdtithi (c. 10th century). The different Sastras have been 
mentioned in the Prabma (v. 8/9) in the following order • the Mdnamya 
Dharm^dstra, the MdheSvara Yogaidstra, the BSrhaspatya Arthasdstra, 
Medhatithi’s NydyaSdstra and lastly the Prdcetasa Srdddkakalpa If the view 
menbooed above be right, we should, in the first place, be unable to explain 
satisfactorily why the Nyayasdstra of Medhatittu should be separated from 
the Dkamu^tras of £/fonu ; then there is the difficulty that the Manubhd^ya 
is, strictly ^leaking, ndther a work oti Nyaya (Logic) not* a §astra (Keith, 
SSOS 3, 295). More impofrtant than these is in my opinion the following 
consideration. There is somdhing so inccmgmous in ating MedMtithi’s com- 
mentary on Manu in juxtaposition with such Astras as the Dharma, Yoga, 
and Artba, and the SrSddhakdlpa, said in this passage to be proclaimed by 
gods and progemtors of the human race like Manu, MaheiSvaraj, Bfha^mti, 
and Praoetas, that, to say the least, the explanation cannot be considered 
very happy. In fact the context compeils the condusion that the Nydya- 

Par instance, PisHa«jn, BSOS i 107£. 

« Vividha-jndm^Sra, voi. SI (1920), p. lOa 

»* PSOS^SS. 



iasira is a saence of the same order as the other sahttas mmtioned in the 
list, and that Medhatithi la an author, real or imaginary, of the same stand- 
ing as the rest of the authonties menticmed by Bavana. Whether such a 
work as MedhS.tithi’s N'yayas^&'a (or at least some notice of it), has come 
down to us or not seems to me immaterial Moreover the boast of Ravaoa, 
the primeval giant, that he has studied Medhatithi’s commentary on Manu 
would be such a ludicrous anachromsm that iwe must refuse to credit even 
an alleged plagiar'ist of the tenth or eleventh century with such an abysmal 
absurdity The only effect of admittmg such an explanation of the Nyaya- 
sasira would be to make the enumeration and the whole boast of Ravaija 
farcical, which is far from being the desired effect It is thus impossible 
to acoqjt the identification of the Medhatithi of the Pratima with the com- 
mentator on the Mamsmrti 

Now finally the Praknt argument. At one time I myself held the view 
that the archaisms in the Praknt of these plays would throw* some light on 
their age; but my antiapations have not been realized It has now been 
shown that m Malayalam manuscnpts of dramas of even Kalidasa and 
Haisa we come across archaisms {1333 of fVPo whidi are daimed to be 
pecuhar to the Praknt of the dramas m dispute : most of these alle^ pecu- 
liarities recur moreover m dramas by southern wnters of Ihe sixth and later 
centuries (Pisharoti, BSOS 3, 1()&). It dwuld seem that the Praknt of 
the dramas) is a factor depebding more on the provenance and the age of 
manuscripts than on the provenance and the age of the dramatist In the 
course of a lengthy review of Bha^s Prakrit (1921) by PRINTZ, pubhshed 
elsewhere,'! have expressed it as my opmion that the Prakrit archaisms can- 
not by themselves be safely made the basis of chronology, and that a> satis- 
factory solution of the Bbasa e^uestion cannot be reached from a study of 
the Prakrit alone (above;, pp. 103 ff.). With ponderous dogmatism Keith 
msista that “ there being evidence of Bhasa’s popularity ” — strictly, speaking, 
only of the plays attributed to BbSLsa — “ with the actors in Malayalam, it is 
only necessary to suppose that they modified the Praknt of the! later plays 
in some measure to accord with the Prakrit of Bhaaa ” (Keith, BSOS 3, 296) . 
The explanaticm would have value if, and only if, all the plays in dispute 
could on mdepaident evidence be confidently attributed to BhSsa ; but such 
is not the case. Keith’s argument only b^ the question. 

« « « 

However desirable it may be to obtain a decisive answer to the mam 
question in thej afl&rmative or negative, it is qmte dear that neither of the 
solutions proposed will stand critical investigation. The problem appears to 
be much more complex than hitherto generally supposed As is only too 
often the case, the daims of both sides seem to be only iiartial truths : in a 
sense these plays— at least some of them, at present quite an mdeterminate 
number — are BlSsa^s plays and in a sense they are not. 



That they are not onginal dramas seems to follow with sufficient cer- 
tainty from the absaice of the name of any author m both the prologue of 
the dramas and the colophon of the manuscripts. The eqilanation that m 
pre-dassical times the name of the author was not mentioned m the prologue 
of the plays mvolves a gratuitous assumption wholly lacking proof Fur- 
ther no satisfactory explanation has so far been offered by those who regard 
all these dramas as Bhasa’s why the name of the author should not have 
been {134J preserved in the colophon of a single manuscnpt of even one of 
these thirteen dramas The Turfan manuscript of one of Mvagho§a’s 
dramas* ‘ has preserved mtact the colcqihon of the last act, recordmg the fact 
theit the drama is the SanpuUaprakarana by Ai§vagho§a. It caimot, there- 
foie, with any plausibility, be urged that the colophons of the oldest manus- 
cripts of dramas did not contam the title of the work or the name of the 
author ; and it would be demandmg too much from probability to expect the 
wholesale and accidental destruction of the colophons of all manuscripts of 
a group of thirteen dramas by one and the same author 

The true character of these plays was partly recognized by Rangacarya 
Raixii and by two Malayalam scholars A K. and K R. Pisharoti. The 
mam thesis of Raddi** was a ne^bve one , it was to prove that the plays 
could not be by Bhasa ; and the whole of lus lengthy artide on the subject 
comprises practically of a destructive cntiasm of the arguments of Ganapati 
Sasxri. He does not however lose sight of the “ possibihty that these plays 
may be abridged versions of the anginal dramas by BMsa, prepared by 
some modem poet or other ” The Pisharotis also lock upem these dramas 
as compilations, regardmg moreover the Ttivandrum SV. as “ an adaptation 
of the original Svapoavasavadatta of Bhasa.” The two scholars were not 
able to support their claims on more solid ground than that there is a living 
tradition, preserved in the circle of Malayalam Pandits, to the effect that 
these “plays are only compilations and adaptations” (Pisharoti, BSOS 
3, 116 ; compare Raja ZII 1923, 264) But a substantial basis for this 
assumption has now been supplied by Sylvam Lfevi’s discovery of certain 
references to Bhasa’s SV , in yet unpublished manuscripts of two treatises on 

In a notice of these manuscripts Lfivi (/A 1923, 197—217) pnhiirfiPR 
certain informatum which throws more light on this perplexing question than 
anything else that has recently been written on the subject ; but lAvi appears 
not to have realized the full significance of his discovery, unless indeed I 
have miaunderstood him, which is easily posable In the article cited above 
L6vi {;i35j draws attention to the mention of the SV. and the Daridracaru- 
dfttta, as also to certain quotatioiis from these dramas in the NMyadarpat^a 

»» SB AW 1911, 388 ff. 

M Vwidharjmut-matara, voL 47 (1%16), pp. 209 ff. 


(ND.) by Rajnacandra and, Gunacandra, and the Nufakalaksana <NL ) by 
S^ranandin One of these quotations differs in a very important particulai 
from all quotations so far adduced We have found verses from our dramas 
ated and cnticized m works on rhetorics but without any mention of tlie 
source , iwe have seen verses ated in anthologies over the name of Bhasa, 
but without mention of the work m which they occur , we have lastly found 
verses quoted as from a SV., but without specification of the author. Either 
the name of the author or that of the work, connected with thef verse cited, 
has hitherto been mvanably in doubt , sometimes both have been m doubt 
Now for the first time we have some datum which connects a verse witli 
Bhasa as also with a specific drama by him , the verse is ated m the ND 
with the specific remark that it is excerpted from the SV. by Bh&sa From 
the fact that this verse is not found m our play, iJtn concludes that the latter 
IS not the ‘authentic’ SV. by Bhasa {JA 1923, 199). 

Let us first make it dear to ourselves what is the exact meaning of the 
little word ‘ unauthentic ’ with which we ate asked to condemn the drama. 
Are our editions of the works of Kalidasa authentic in the same sense as our 
editions of the works of, say, Goethe ? Are they authentic m the sense that 
the text they present is the text exactly as conceived and finally wntten down 
by the reputed author ? No one will be pt^ared to deny that the Prakrit 
of the dramas miay have been gradually modernized in the course of tians- 
mission, or that the Sanskrit portion may have suffered a httle at the hands 
of well-meanmg ‘ diadreuasts,’ or that lastly some few verses and even scenes 
may have been interpolated or oxmtted. As has already been remarked, a 
scene m the third act of the Bengali recension of SakuntalS is four or five 
times as long as the correspondmg part m the Devanagaii version. The play 
Vikranurrva§i has come down to us in two recensions, of which one contains 
a series of Apabhra|m§a verses that are entirdy ignored m the other. Such 
being the case, what is the justification for considering even one of the shorter 
versions, which are apparently older than the other, in every detail an exact 
replica of the original m the form m which it left the hands of the dra-flSC} 
matist who ccmiposed it ? It seems certain that the tradition fluctuated, and 
fluctuated at times considerably.^® Still we do not make sudi a bustle over 
the fact that ‘ authentic ’ worte of KMdasa are no longer available. 

Be that as it may, there is another aspect of this citation that appears 
to have a positive value. The verse reads : 

paddhrdntam puspam sosma cedam Silasamm | 
mtutm kacid ihaslna math dr?tvS sahasa nata || 

(Read gatS.) 

IS Compare Sten KbNOW, Das indhche Drama, p. 66 . Jetzt amd wohl die 
meistea der Ansicht, dass keine der tms vorliegendlen Rez^sionen den Uretxt des 
Dkhtera \vk. KSlidisa] repraeaebtiert” 


The king of Vatsa, regarding a stone bench in the pleasure garden says * 

“ The flowers are trodden under feet. 

The stone bench retains still its heat 
Forsooth some lady who was seated here, 

On seeing me, has dq)arted in haste.” 

Conunenting on this verse Levi remarks that we find in the Trivandrum SV. 
‘ dislocated ’ elements of the scene as written by Bhasa. Such is however 
not the case. There is no dislocaticm at all. All that may have happened 
IS that the ND. verse has dropped out of the text of the Trivandrum version 

The situatictti m our play is this. In the first scene of the fourth act 
Fadmavati and Vasavadatta are promenading m the pleasure garden, ad- 
miring the beauty of ^phahka bushes in blossom. Padmavati’s maid begs 
her to seat heisdf on a stone bench in or near the sephalikfi. bower, and she 
herself departs to pluck flowers. The ladies seat themsdves on the bench 
indicated and indulge m a tflte-a-tflte. F'resently Padmavati, to her conster- 
nation, discovers that the King and the Jester are strolling leisurely m the 
direction of their arbour. She thereupon proposes to her friend that they 
tiiemselves should move away and hide m a neighbouring jessamme pergola. 
The King and the Jester approach the sephalika arbour just vacated by the 
{'137J ladies. At this pomt there is in our play a small hiatus, all but im- 
perceptible. Standing near the bower the Jester abruptly remarks : “ Her 

Ladyship Padmavati must have come here and gone away.” We fail to 
understand why the Jester ^ould make this curious, unmotivated r^nark. 
The missing link is evidently the ND. stanza, which furnishes the requisite 
motive for the remark of the Jester. We are here told that the King, on 
observing that the surface of the stone bsicb is warm, surmises that some 
lady who had been sitting there, on seeing him approach, had hurriedly de- 
parted, crushing under her feet, during a hasty retreat, the flowers lying 
scattered on the ground. The Kmg has no idea who that lady was. But the 
observation of the King sets the Jester thinking, who shrewdly surmises that 
it must have been Padmavati. 

This r>ecapitulation of the situation should make it ciear to the reader 
that there is no great ‘ di^ocation ’ of the diements of die original scene as 
far as it may be sunnised from the quotation in the ND. All that is needed 
to restore tte text is the replacem^t of the new verse at the point wheie 
there is a hiatus in our versiem. 

In the same article I^vi has another quotation which also has some 
bearing on the pte^ question. The oth«- treatise, the Natakalak?ana, gives, 
wit hout any mmtioa of the name of the author, an extract from a SV. to 
ulustrate a device with which the transitioa from the iwceliminaries to Uie 

main of the play is a^aoved and a diaracter is introduced. The 

quotation is: 


nepathye suPadhSrah utsSra^&n irutva paihatt j aye katham tapovanc 
'py utsaraita \ (,vtlokya) katkath manln Yaugandkarayamk 
Vatsair&jasya rdjyapratymayomih kartuhamah 
Padmavatiyajmienotsdryate || 

" The stage director (sutradhlara) on. hearing the order for dispersal 
shouted behind the scenes repeats : ‘ How now ' Even in a hermitage people 
are being ordered to disperse’ (Looking aside) ‘Why, the minister 
Yaugandhaifiyaina, who is seeking to restore to the Kmg of Vatsa his king- 
dom, IS being turned away by the servants of Padmavati ’ ” 

(138} It IS extremely unfortunate that the name of the author of the 
play has not been mentioned in the NL. The omission, depnvmg us of cer- 
tainty, leaves us to surmise that the author is Bhhsa ; but the conclusion is 
mevitahle unless indeed we postulate the existence of three Svapnavasava- 
dattas, parallel to the three KurnSrasambhavas, now famous in the history 
of San^it hterature 

The prologue of the SV. ated by the author of the NL. is evidently 
worded differently from ours The elements revealed by the extract are 
these . there is a stage director, and a dispersal (utsarapa.) of the crowd 
behind the scenes (nepathye). The stage director hears the orders touted 
out by the servants of Padmavati, and sees the crowd being dispersed. In 
that crowd he notices Yaugandhaiayana, who is there to carry out his plans 
for the restoration of the King of Vatsa. The sanoe dements are present in 
our play. Here the stage director, on hearing the noise behmd the scenes, 
announces that he will gp and find out the cause of the commDtion, which he 
does. Behind the scenes is shouted out the order for dispersal (utsaranh). 
The stage director thereujion e:qjlains to the audience that the servants of 
Padmhvaii are disperdng the crowd of hermits. We observe the repetition 
of the idenbcal word utsaranS, and the similanties between the exclamations 
of the stage director in the extract and of YaugandhariSyaipa in the Trivan- 
drum version • 


aye katham tapovane ’py katham ihSpy utsdryate ) 

utsarana | 

Consequently on the evidence of these two extracts, of which one is 
expressly stated to b«| ftmii the SV by Bhasa, and the other is presumably 
from the sapie source, we may safdy assume that though the Trivandrum 
(day is not identical with the drama known td Rhmacandm and Sagaranandin 
in the 12th century, it does not differ from the latter very considerably : the 
two are near enough to each other to be styled differ^t recensions of the 
drama by BhSsa. My own surmise is that the Trivandrum SvapnavSsava- 
datti is an abru^pnent of Bhasa's dmma, with a different prologue and 
epilogue, adapted to the Malayahun stage. 



{ 139 } * 

Here follows a bummary of the important conclusions arnved at above, 
to which are added ceitam auxiliary observations on the character of the 
present group of plays 

Vitally important are the following facts relating to these plays, which 
will throw a deal of light on the subject and which may not be ignored in 
any future investigation of the question, namely, that these plays form a 
part of the repeitoire of a class of hereditary actors m the Kerala country , 
that the manuscripts of these plays are by no means rare, though they ap- 
parently are the jealous preserve of these actors ; and lastly that the lattei 
produce these dramas sometimes as a whole, and sometimes m detached and 
disconnected parts. Cf Pisharoti, BSOS 3, 112 f ; Raja, ZII 1923, 250 f. 

The circumstance that these plays have been traditionally handed down 
without any mention of the name of the author, whether in the prologue of 
the plays or the colophon of the manuscnpts, is an almost plam indication 
that they are abndgements or adaptations made for the stage, and they have 
in fact been regularly used as stage-plays m Malayalam. 

These fflays show admittedly many similarities, verbal, structural, stylis- 
tic and ideological, which surest common authorship. But in the absence 
of more information as to the orii^alst of which these are evidently adapta- 
tions, It would be unsafe to dogmatize and postulate, at this stage, a common 

The colnadences in fomoal technique are almost certainly to be e 3 tplam- 
ed as due to the activity of adapters. It has been already pomted out that 
the professional actors who produce these plays often stage only single acts 
selected from these plays ; and it is reported that in passing from one act 
of some one drama to another act of a different drama, these actors are m the 
habit of prefixing — quite naturally, it seems to me — ^to each act an appro- 
priate introduction consisting of a benedictory stanza and a short prose speech 
or dialogue announcing the character that is about to enter as well as the 
business. Our prologues appear to be sudi introductions, which thus owe 
their similarity merriy to a peculiarity of local histrionic technique. The 
preliminary benedictory stanzas, which {140} are condemned cm all hands 
as bad verses,, have all the appearance of be^ also the handiwork of these 
adiQ^ters ; the short fcmnal bharatavakya seems likewise to be a sort of a 
fonnulistic epilogue It would be a mistake to see in these external co- 
inddences a proof of coanmon autfaordi^ oi the plays In order to ascertain 
whether two or more of these dramas are by the same hand we shall have 
therefore to enqjiloy some cither tests, which have not so far been used by any 
previous wnter on thfe subject The ^peculation r^aiding the idaitity of 
the iSja^thha of the epilogues (KONOW, op. dt. p 51) is wholly withcmt 
meaning ; the esptesdon seems to have been left intentionally vague so that 


the same stanza could be conveniently used on any occasion and at the court 
of any king. Significant is the similarity between our epilogues and the 
hemistich from the MBh. (12. 321. 134) • 

ya imam prtktvtm kitsnam ekacchatram praSasti ha, 
to which I have drawn attention elsewhere {JAOS 41, 117). 

The Praia it archaisms have no probative value for the antiquity or the 
authorship of the dramas It is, however, not impossible that some of the 
plays may have preserved, so to say m fossilized condition, a few leally 
archaic forms inherited from the old prototypes. *Of this character seem to 
be the Prakrit accusative plural masculines m -am, noted first by Printz 
{,Bnds<fs Prakrit, pp. 3, 26 , but see above, p. 111). 

Similarly the metncal portions of the dramas appear to have preserved 
some epic usages {JAOS 41, 107 ff ) It seems unpossible to beUeve that a 
dramatist who normally wrote ©sod Sanskrit could not produce verses gram- 
matically more correct than the followmg • 

smaramy avantyd ’dhtpateh sutdydh (SV. v. 5) 
jnayatam kasya putreti (Bala ii. 11) 
sMgatdm prcchase hath&m (Paiica. u. 48), or 
apfccha putrahtakan (PratimB. v. 11) 

As regards the stage fights and the representation of a death on the stage 
in these plays, a plausible explanation is that they are, as suggested by K. R. 
PiSHAROTl (iBSOS 3, 113), comparatively {141^ modem innovations mtro- 
duced with a view to produemg a more strikmg stage effect But it is still 
an open question whether some of these elements may not be survivals 
denved from an older dramatic technique. This reservation does not hold 
good, however, in the case of a final death scene. The practice of these 
diamas can form no exception to the general rule prohibiting a final catas- 
trophe ; the tlrubhanga is not mtended to be a tragedy in one act It is the 
only survivmg intermediate act of an epic drama This follows from the 
fact that the play has no epilogue, in which particular it resembles the Duta- 
ghafokaca, which m one of its manuscripts, as reported by Pisharoti {The 
Shamda, 4 (1924), 19), is actually and rigjhtly called DtutaghatotkaeSfika. 
Some sh^t confirmation of this surmise we find further in the report of C K. 
Raja (Z7Z 1923, 254) that there is extant in Malabar a dramatized version 
of the Ramayana in 21 acts! Even apart from that, there is no doubt 
that any spectacular representation ending in a death, whethetr of the villain 
or the hero, would be repugnant to Hindu taste, and foragn to Hindu 
genius, unless it be an apotheosis, a canonization of the hero as m the 

The verse Bhasanatacakre ’pi etc, said to be a quotation from the Sukti- 
muktavah of RBjaiieKhara proves by itself little or nothing for Bhasa’s author- 



sliip of SvapnavBsavadattai since the authenticity of the former work and 
quotation is open, to criticism. It is not generally known that the preceding 
.verses make out that Bhasa was not only a contemporary of Har§a (evi- 
dently Harsa iSladitya of Thanesvar) but also a washerman by caste and 
the real author of the tnad, Ratnavali, Nagananda, and PnyadarSika, a 
statement which we have every reason to discredit. That the Pre- K a li dasian 
Bhasa did write a Svapnavasavadatta follows, however, with tolerable cer- 
tainty from the evidence of the ND. by Ramacandra (/A 1923, 197-217) 

The more important reasons for regarding our SV. as closely related to 
Bhasa’s drama of that name are these. To start with there are the name, 
and the style, as also the merits of the play, which has won general recog- 
nition as a work of high order. The rhetoriaan Vamana cites a stanza 
which not only occurs in our play but fits evidently well m the conteirt. It 
contains scenes compat-{l42}-ible with those suggested by the quotations 
horn Bhasa’s drama cited in rhetorical treatises by Ramacandra, ^gara- 
nandi ti , as also by ^radatanaya (cf. Ganapati SAsrai, /RAS 1924, 668). 
From the second of these it follows that Bhasa’s drama opened like ours with 
the entry of Yaugandharayana (accompanied probably by Vasavadatta) 
followed by that of PadnevaC and her retmue— From Saradatanaya’s sum- 
mary It would appear that some scenes are wanting in our version. — ^The 
Dhvanyalokalocatta ates apparently a lost verse, svancttapaksma’’ etc. It 
is a mistake to argue that this verse cannot have a place m our play Even 
if it does refer to VBsavadatta, as it appears to do, it may be easily included 
in a reminiscence of the King — ^The statement of Sarvananda remains, for 
the time bdi^ unexplained, unless we are prepared to adopt the emendation 
suggested by Ganapati Sastri, whidi, it must be admitted, is an a prion 
solution of the difficulty 

Thee is some reason to bdieve that the SV. and the 'Pratijna an? by 
the same author. In the concludmg act of the SV , it will be recalled, there 
is an allusion to the fact that in the nuptial rites cdebrated at Ujjayini after 
the dopement of V^vadatta, the parties to be umted in wedlock were re- 
presented merdy by their portraits There is no reference to this marriage 
“ by proxy ” in the Kothasaritsagara nor in the Brhatkathomanjan, and there- 
fore there was probably no reference to it in the Brhatkatha dther ; it appears 
to be a free inventicn of the dramatist It forms, however, an important 
element in the denouemoit of our SV. ; it is therefore significant that there 
is a dear allusion to it in the conduding act of the PratijfSa also. 

As regards the Gamdatta I have seen no reason to abandon my former 
view (/AOS 42, ^ff.) that our fragment is probably the original of the first 
four acts of the Mrcchakaftka ; but if it is not that, it is suggested, it has 
preserved a great deal of the ori^nal upon vdiich the Mfcchakatika is based. 
My condudoos are only strengthened by Morgrnstierne’s ind^iendent study 
of the reiatiaDa between the two plays. FYom rdereoces in one of tiie new 



ffihit ya works Utilized by Lj6vi it follows that a drama called Dandracdru- 
datta was known to the author of this treatise , the Mrcchakapka is named 
separately, which shows that they were two £143} different dramas ; both 
of them had however evidently the same theme! The Dandracarrudatta liad 
at least mne acts, and the two plays developed to the end on very similar 
lines The rhetoncian does not tell us anything about the author; «so its 
authorship is still uncertain. 

My view of this group of plays may then be briefly summarized as 
follows Our Svapnavasavadatta is a Malayalam recension of Bhasa's 
diama of that name ; the Pratijfiayaugandharayapa may be by the same 
author , but the authorship of the nest of the dramas must be said to be still 
quite uncertain It may be added that Bhiasa’s authoirship of some parti- 
cular drama or dramas of this group is a question wholly independent of the 
homogeneity or heterc^eneity of the group as a whole Indeed the only factor 
which unites these plays into a group is that they form part of the repertoire 
of a class of hereditary actors The Carudatta is the original of the Mrccha- 
katika. The five one-act Mahabharata pieces form a closely related, homo- 
geneous group ; they appear in fact to be sm^e acts detached from a lengdiy 
dramatized version of the complete MBh. saga, — ^a version which may yet 
come to li^t, if a search is made for it The tirubhafiga is not a tragedy 
m one act, but a detached intermediate act of some drama. The present 
prdi^ues and ^ilogues of our plays are all unauthentic and con^ratively 

November, 1924. 


Inked estampagea of the subjoined insaiption, which commemorates the 
construction of a tank, were prepared by the Madras Epigraphical Depart- 
ment in 1903, and it forms No 91 of the Epigraphist’s collection for the 
year 1902-3 It was briefly reviewed m the Annual Report on Epigraphy 
for 1903, and it has also received a short notice in an article by the late Mr. 
Venkayya entitled ; Irngation in Southern India in anaent times.^ The 
record is incised on two slabs, one smaller than the other, set up in front of 
the ruined Bhairava temple at Porumamilla in the Badvel Taluk of the 
Cuddapah District, situated in 15® 1' N. and 79® E The latter district 
being very dry, cultivation is m general possible only with the help of aiti- 
ficial storing of water The irrigation tank at Pbrumarmlla is, accordmg to 
the District Manual, one of the largest in the TMuk The inscription, apart 
from its historical importance, presents vanous other points of intaest, not 
the least important of which is the light it sheds on the tank-building activity 
in andent India. 

As regards orthography, the inscription follows the same system which 
is to be observed in other inscriptions from the Telugu and Kanarese Dis- 
tricts. A superfluous anusvara is insHted (1) before a nasal -t- consonant, 
as in pu^j^ya 11. 13, 46 ; also in 11. 19, 36, 89 : (2) before h -f consonant, 
as in Vije^Sfhhvaya^ 11. 34, 35 ; also m 11. 57, 65 * (3) before nn as in 
vijayothmatSk 1. 36 : (4) before m as in kamima° 1. 82. We find also 
Jhe doubling of a ccoisonant after an anusvara in chaiiKkcha'" 1. 18. As in 
other inscriptions, we notice the mixing up of the two forms of visofsa-sandfa, 
as in in 1 23 ; also in 11. 29, 83, etc. ; and the writing of thth 

for tth in °rththami'* 1. 1100, and of jkjh for jjh in 1. 12. Other 

examines of {98j incorrect orthography are the followmg ; 1. 113 ru for fi ; 
1. 39 ri for ru ; confusion of the sibilants f and s in U. 20. 22, 29, 37, 52, 
54 and 95 ; w for MM m 11. 66 and 83 ; confusion of d and iA in 11 51, 104 
and 109 ; ^xiradic adscript of y to an initial vowel as in yetad (for etad) 
1. 21, yek = diva (for Sk = ma) 1. 105. The aspirates are sometimes dis- 
tinguished from smularly shaped non-aspurates by means of a short vertical 
stroke added below the letters, as in the modem Telugu alphabet There if 

” [Bp. 14. 97-109]. 

See the ESrertor-Qeneral’S Amrnd for 1903-4, I^rt It, pp 202 ff,— A 
lesuin^ of the ocmtents of the inscription is inducted in the pew edition of the 
Cuddapttk Pistrkt Ganeffw, 



inconsequence in the doubling of consonants after r • cf. 11 7, 10„ etc. on the 
one hand, and 11 6, 16, etc on the other Rough r (doubled) is used once 
in the nomen piopnum Devandja (1. 109) It remains to be remarked that 
the letters are incised between equidistant parallel lines miming along the 
breadth of the slabs — It is necessary to add a few words on the language of 
the inscription. Excepting the benedictory words at the begmnmg of the 
record and a few phrases employed further on to introduce some of the 
stanzas, the whole of the inscription is m verse The language is extremely 
meagre San^t, and the verses are devoid of poetic embelhshment. The 
writer is indeed guilty of the gravest mistakes of grammar and syntax, most 
of which are noticed in the foot-notes to the text and translation. To men- 
tion just two of them here in 1 37 disi purve pratishthitah is used for dtit 
purvasydm pralishthdpitah , and in the first sentence of v. 16 the verbum 
actionis is omitted — In respect of lexicography the foUowmg uncommon 
words and expressions deserve notice • kjnti (11 29, 39) = “composition" , 
tataka^matrika (1. 47 = “ tank nounshed ”, on the analogy of nadi-matnhd, 
etc.; bhtama^altt-gaii (11. 69, 90) =" sluice" (?) ; madhya-kurma (1. 73) 
= devated ground in the middle (i*) , bku-vara (1. 79) = “kmg’ ; gdmgeya 
(1. Ill) = “gold.” 

The object of the record is, as remarked above, to commemorate the con- 
struction of the tank at PorumanuUa by king Bha^ra alias Bhavadilra, 
son of Bukka 1. (v 49). The following analysis gives a synopsis of the 
contents of the record . The grant commences with mvocatory and introduc- 
tory verses (w. 1-10) the succeeding stanzas give the genealogy of the 
donor, Bhaskara Bhavadura (11. 11-22) • the next few verses recount the 
merit attaching to the building of a tank (23-27) • then are given the details 
of tank oonstruction and the spejcrfication of the site of the tank, date of its 
construction, etc. (28-45)' • then the usual imprecatory and benedictory 
stanzas (46-^) : and lastly, the spedfication of the adkikann of the tank, 
and the composer of the record (59-62) The only new facts in the history 
of the First Vijaya-nagara Dynasty® with which the mscription furnishes us 
are the following • (1) Bhaskara. alias Bhavadura (a name whidi is not 
known from any other inscnption) iwas the son of Bukka I , and thus the 
brother of Harihara II. Bhaskara was placed m charge of the eastern pro- 
vinces — ^whidi he ruled from the “top of the sublime Udaya-gin” (in the 
Nellore District) , (2) Bukka I. had four brothers, viz. Harihara, Kaifipana, 
Marapa and Muddapa ; and (3) Anantaraja was one of the ministers of 
Bukka I. — ^The odhikann of the tank was Devanajan, son of the minister 
(probably of Bhaskara) Kumaragiri-NS.tfaa (v. 50) . — ^The writer, who was 

* A jhicanct and connected account of the facts in the history of this dynasty, 
gleaned from stone and copper-plate records, is furnished by Rao Saheb Krishna 
Sasiri in his paper oititled " The Fust Vijayanagara Dynasty ; its Viceroys and 
i^ifuiisters.’’ (See the ptrectOT-General’s Amtudl for 1907-8, Part II , pp. 235 ff.) 


epigraphic studies 

rewarded with a gift of land, was the poet Langaya-Machanaiyya (te. 
Machana, son of Limgaya) of the family of the Kautsas, resident of Nimda- 
pura (v. 511 On completion of the tank lands under it were handed over 
to a number of Biahmanas (v. 45). 

The question of the construction of the tank is shrouded in some obscu- 
lity on account of the umntelligibility of the termim used in the description 
of the tank.’ It would be, therefore, as well to start with the facts regard- 
ing the tank as it now stands I have been able to gather the followmg in- 
formation through the kind courtesy of Mr Banerji, the Collector of Cud- 
dapah, {[993 whom I had addressed on the subject* The tank, which is 
situated about two miles (and as the inscription also tells us) to the east of 
the village called Pbniroamdla, is elongated in shape, being some 7 miles 
long and 2J miles broad The bund consists of four natural hills connected 
by three short earthen dams, rivetted with Cuddapah slabs The western 
flank thus consists of practically the range of hills which runs north and 
south between Porum5nulla and Badva. The total length of the artifiaal 
bund is about 4,500 ft , the total length mcludmg the hills is about 14,000 
ft At the deepest section the bund is about 12 ft wide at the top and 150 ft. 
at the bottom, and about 33 ft deep. The tank has two sources of supply, 
one natural and fhe other artifiaal. The latter was constructed about 20 
years ago. The natural feeder is a stream called the Maldevi nver, — ^The 
reservoir is provided with four sluices, two of which have been rqiaired in 
recent times and provided with saew-gear, and there ate five weirs This is 
the actual condition of the tank at present 

From the inscription we leam that in tiie twelfth century of the Christian 
era tank-building was looked upon as one of the seven meritorious acts which 
a man ought to perform during his lifetime The tank at Porumaniilla was 
called Anantaifija-sagara. It is also stated that for two years 1,000 labourers 
were working daily on the tank and the dam , and 100 carts were engaged 
in getting stones for walls which formed a part of the masonry work The 
dam was 5,000 rekka-dat}4css long, includmg the hills, 8 rekha-dandas wide, 
and 7 high. Besides, the author gives us the twdve sMhanas of the Potu- 
mfimilla tank, and six ddshas of tanks in general. Much of this latter is 
clothed in very obscure language Nevertheless with the help of the descrip- 
tion of the tank given in the previous paragraph we are able to get a fairly 
clear notion of what the author wishes to convey,® In the chatur-bhrama- 
jola-gaH we have a reference to the four sluices ; and in the range of hills 

* To Rao Saheb Kriahna Sastri I am indebted for the explanation of several 
of the tedmical expiessionB. 

* Some time ago, when I visited the she of the tank, I took the opportunity 
of verifying and oorrecting the statements in this report. Some few jEresh obeierva- 
tkxis which 1 made on die spot have also been embodied in the succeedmg remarks, 

* See notes 6-8 on p. 108 and 1-2 on p 109. 



fomung the western flank of the tank we must look for an explanation of 
remarkable phrase tad-yoga-khamdo gtnh. Again, as the Maldevi river aids 
at the tank, the tnyojona must needs refer to its length from the source 
to the pomt at which it enters the tank — One fact which may be gjeaned from 
the measuronents of the tank preserved m the record is of no small signi- 
ficance Knowmg as we do the dimensions of the bund, in terms of the 
rekhd-dandas as iwdl as in feet, we are in a position to compute the equiva- 
lent of this standard of Imear measurement current m the Telugu District 
ill the 14th century For this puipose it would be safest to compare the 
values for the height of the bund, which, I imagine, would offer the least 
variation In the above-mentioned Report of the Collector, 33 ft is quoted 
as the hdght at the deepest section of the bund Takmg, now, 20 ft as the 
minimum haght for the bund of any large tank of that size, the average 
height of the bund in feet works out to be (20 -1-33) — 2 = 53l2t ft. This 
must rou^y correspond to the 7 rekha-dandas of the inscription The 
equation will be 5312 = 7, and this gives us roughly li yard as the equivalent 
of the rekha-danda, which, by the iway, corresponds approximately to the 
distance from the top of the shoulder of one arm to the tip of the middle 
finger of the other arm, measured along the chest, for an average man I 
mention the latter fact, as it is well known that m primitive times a stand- 
ardised (but locally varymg) value of the lengths of portions of the human 
body served as units of length ; cf . the Indian measures mguli, hasta, dmida, 
and the European loot « 

One other feature of this record deserves mention here. Early m the 
begmnmg of the inscription (vv 1-3) we find enumerated the characteristics 
of an edict (sasana-lakshat}a) , which include hints on composition and the 
significance of the metrical foot (gana) with [100} which a sasana com- 
mences (v. 2) The latter is a curious specimen of superstition . the gana 

ma ( ), when standing at the beginning of a iasana, secures bliss , na 

( V V vy ) in the same position secures wealth, etc ! Further wel learn that 
in a faultless verse the vtsarga should stand at the end of the complete stanza 
and not at the end of the first half (v. 3) ! 

The inscnpticai is dated on the 14th of the bright half of the month of 
Karttika, m the cychc year Saumya, correspondmg to iSaka 1291® (expired) 
and Kaliyuga 4470 There is some doubt as to the wedk-day. The 
syllables guru in 1. 58, which evidently mtroduce the name of the 
wedc-day, are dear enough , and the followmg lettem must be read as pttshya, 
as the vertical stroke between the aksharas ru and pu is nothing but an 
accidental depression m the stone. In that case it would seem that the 
wedc-day was Thursday, and the nakskatra PUshya But Dewan Bahadur 

® Expressed by the dironogram bhu-namdy-akshya-eka, and in numerical 


epIgraphic studies 

Swamikannu PIliai, whom I had addressed on. the subject, informs me 
that the titht ended on Monday, the 15th October, aid 1369, at about 7 
ghaltkas after mean sunrise ; and the nakshatra for that day was Agvini, 
which came to an end about 47 ghatikas aftei mean sunrise. 

For purposes of onentahon the localities Sri-parvata (le. iSii-iailam), 
Ahobala, Siddhavata, Udaya-gin and Porumamilla are mentioned. Of these 
only Siddha-vata needs to be specially noticed here Its denonunation in 
the inscripticm is desaka, i e subdivision of a country : therefore the name 
could well he, I think, linked with the modem Taluk Siddhavalttam in the 
Cuddapah Distnct, the boundary of which is not far removed from the site 
of the tank. 

A refeience to Hemadn’s Ddnakhanda (which is undoubtedly what is 
meant by HemMri-knh in 1 39) shows that that work enjoyed the reputa- 
tion of an authority in the Tdugu country at the beginning of the 14th cen- 
tury of the Christian era 


[Metres ; v. 1, Anusklubh {:§loka) ; v 2, Sardulavikndita , w. 3-4, 
Anusktubh (Sloka) , vv. 5-7, Sardulavikridtta , v. 8, Mandahrdntd , w 9-10, 
Anushfubh {Sloka) ; v. 11, SardiUavtkridiia ; w 12-13, Anusktubh (Sloka); 
V. 14, Upajati] w. 15-36, Anusktubh. {Sloka) ; w. 37-38, SardulvikrtdUa ; 
VV. 39-40, Upajati , w. 41-43, Anusktubh {Sloka) , 44, Sdrdulavikfidita ; v. 45, 
Anushfubk {Sloka) ; w. 46-47, Salim, w. 48-50, Anushtubk {Sloka) , 
V. 51, SardiSavikn^ita ] v. 52, Anushtubk (SZofeo)! 

First stone. 

1 i[|] sw! i[i] [«*] !! 

4 [%] ^ [^] l(l)« 

[10135 [’i^] sN 

’’ From the odj^nal stone and a set of inked estampages. 

» At the top of the inacription are engraved from left to n^l figures of 
\^ghnSSvam with his vehicle, the mouse, the Ijngam, the Sun and the Moon. 

• Read 

The syllable Jf was inserted later and engraved above the luie in the 

w Read 

M Hera an empty space in the original diowing traces of letters arored out 

porumAmilva. tank inscription 


7 i ^tpi] 5rr5J[*i«t5i5®«Eroi] n [^] u 

8 [l*] ^WI5rf5t^[s3‘'‘[;*] [5n]«st:^® 

sit^L U«W 

9 fitsnwt «5ira: fqrefWt %^«raiw^i- 

10 [i*] vm jwliler? 

11 g wgrt '»iia4<tgtg »TO L5 ra ( !^ ^i^i; ngi?): ( - ) glt: h'^h strofi- 

12 qrawir!ft[^ii]5t i 

13 ^ gg^® «t: [i*] 5(-)«wm§g «wn[g> 


14 III ?nf^ gwunfnro:®^ ggPiir ii^ii [g]an- 

15 g*qfSi^[%] dismai «Kid [g]gi 


16 ljftl5Pnw! w; [l*] ^ «(tPlR?l5RT: a g<THT [^'jURgH 

17 inw. fiisraif^R!fwei*3RW^®5TO*ra: ii»n ^[e^Jqraira- 


19 ^ qi5 g^iH ^[<^(-)wi]®sng?^gfeiiT i- 

20 ’tR»t n^u g«; [i*] *R 


21 w ®’5ra^?n*^TBi nin wRSRg.[q3 

fe[:3 [1^3 • 

22 gSRJJOtftWHW ?POT ^ in«»U [lflr3^^[**l’‘*^?^t^ 

23 I ^ ''1^(0 *i> 

24 «iRi^q ; [i*3®R5f^ *iT?[iif]^ 4t®Ni[!*3 ^- 

[102] 25 ^ [w -*] gflgPf sitOTmRr<i[^:] ill ill g g[»i]im^tqig: 

26 ^ [i*Q gs^rai^ TOrOT[^]si 3[g filial insii 

27 [i*3 

28 ni^n 1^‘nggqsiRRfffi^'' #q[*T]gPRr, 

1* Read 
“ Read ®!i. 
le Read^lf" after * 1 ^. 
“ Read«5j^is^. 

S' ReadoRi^. 
s8 ReadTj3“. 

1* Read 

Read °^ 0 T ftf 5 <{. . 
s» Read^^^^. 

** Read 
s« Readi^". 
s» Riead^_ 
a» Read^, . 

« Read 
18 Read 
SI Read®^?I^\ 

s* Read^rsnR^. 
ST Read^,^,'. 

80 Read^ after 



Swamikannu Pillai, whom I had addressed on the subject, informs me 
that the tithi ended on Monday, the 15th October, ad 1369, at about 7 
ghattkas after mean sunrise, and the nakshatra for that day was ASvim, 
which came to an end about 47 ghattkas after mean sunrise. 

For purposes of orientation the localities Sri-parvata (i.e i§ri-^ailam) , 
Ahbbala, Siddhavata, Udaya-giri and Porunamilla are mentioned Of these 
only Siddha-vata needs to be specially noticed here Its denonaination in 
the inscription is desaka, le subdivision of a country therefore the name 
could well he, I think, linked with the modem Taluk Siddhavalttam m tlie 
Cuddapah District, the boundary of which is not far removed from the site 
uf the tank. 

A reference to Hetnadri’s Ddnakhanda (which is undoubtedly what is 
meant by HemSdrt-kjtU in I 39) shows that that work enj'oyed the reputa- 
tion of an authority in the Tdugu country at the beginning of the 14th cen- 
tury of the Christian era 


[Metres : v. 1, Anusk(ubh (Sloka) , v 2, Sardiilavikridita , w 3-4, 
Anushtubh {Sloka) , w. 5-7, SdrdSlavtkridita , v. 8, Manddhrdntd , w 9-10, 
Anushfubh (.Sloka) ; v 11, Sardulavikrldtta ; vv 12-13, Anushtubh (Sloha) ; 
V. 14, Upajdti , w. 15-36, Anushtubh {Sloka) ;‘vv. 37-38, Sardulvikrldita \ 
w, 39-40, Upajati ; w. 41-43, Anushtubh (Sloka) ; 44, Sdrdulavtkfidita ; v 45, 
Anushfubh (Sldka) ; w. 46-47, Salini , w 48-50, Anushtubh (Sldka) , 
V. 51, Sardiilavikfi4ita\ v. 52, Anushtubh {Sloka.)] 

First stone. 

1 i[i] 5OT; j[i] [n*] !! 

3 51^5 •ar]? [arjffra^ m[ui] arT 

4 [^] ^ [ait] qianf^« ^ ftawsg; ^ |(|)13 

D0135 [ [ ;* ] sN aq 

6 IRH 

a From the oripnal stone and a set of inked estampages, 

® At the top (rf the inscription axe engraved from left to ngfat figures of 
Vighn§£vara with his vehide, the mouse, the Lifigam, the Sun and the Moon. 

® Read 

w The syllable q ivaa inserted later and engraved above the line in the 

w Read 

Here an empty apace in the onguiai showing traces of letteis scored out 


7 \ ?[»^] ?rRt[*l# 55 t^] It [\] II 

8 [i*] %!ira!ras^^5]“[i*] g»rai^ [m]^CT:“ 
^ uaii 

10 ?n[?B}ir«««<^««i?pSd [i*] virai 

11 ^ Hfri« sisJfegfWpRi?® «n?n: srai^ ( - ) «!t: u'^ii sirtst- 

12 i ?^!i^ *nff [^ — *]”grarfSrlt 

13 [i*] ftraw s(-)’'«w^ «iw[¥]^- 


14 ^1 f|Mi»i*fi^:®i ggrfiRT 8>?n[^]^ ll^ll [?r]sm- 

15 ftrsRRtf 'R# nftftw [?i]w 


16 w: [I*] «R*r a 5 ct*j! [%]jrraw[;] 

17 'iiws f ^i^ ^i%q [ «g 5 PT Rt«w*<tw* ; ii»ii 


19 ^ m spiw 5i[ft(-)?*>]53ira?ft®fe?iT ^- 

20 u«M s«! ffe^TOi^ar [»*] w 

IS t aws r - 

21 w ^^ q giqgR g ^ liiii ‘*''^?Rnf^(-5')?T^'WT]sp?if *TR^[q] 

M'-] [•*] 

22 ^iq^jif^ranqasr !R*r q?P*r^ n?o|i [s9V]?iq[si]^isR5Rft ftgg 

23 I ^ g^qqra «ii^(0[^5^ 

24 [i*]?R3^ qgwi qR[qf]^ sftjfts k- 

[102] 25 ^ ?req ^1® -*] imii 

26 f?qi [i*] sssRRt, ?Rj|«i[t]!T g[t firji id^ii 

27 ar^ftHtfatfiw^ <i^ i fi i * n ^ T q a[;] [i*j qra^qrqire^q 

28 ^fiwrarf^wi:*^ WWW ti«i\<4<l w[q]«i^. 

18 Read 
18 Read 

18 Readq^I* after JT^. 
“ Read-ggsl^. 

28 ReadoRRjiT. 

88 Readt^'. 

1* Read g^». 
i» Read "qnifsj^-. 
80 Readgg|jar. 

88 Read 
28 Read i^°, 

88 Read^, 

« Read 
18 Read 

81 Read^qra^", 
** Read^i^i^^l*. 

82 Read^». 
Rfiad^ after 




29 0! I ^ **’’*'• 

30 watt sir^ ♦frws?['^ 0 !3^n# 'i^s- 

31 ^fetli git U1«sU "Tfe^Rt SW ^ [^]l«*f [l’'^] [^]- 

32 ^ w4qtiqi«!iiFii«<i<^i^- 

33 ® «R0fta^ [i*]^(-K 5r?[^]if^j l[l^] [?]»« 0) 

34 [0 ^nw^raft *wr ^- 

35 OTr(-)|{4 IIS [<£]n 5?i5i<wr g?t ( sic ) [ i‘*‘ ; 

si5N*i pi- 

36 - )Tat: « s pf^qt%»r ^ ^!n^[ : i ] 

37 ar(-)w3arRt*wt^ ^tl^srfltfers «^*u s •’5\4^fiiW4iiJ 

38 ^(-K ^ [•*] f^5Rn5T?[^]d f|3r^ 

[«^s *u] 

39 [I*] 

40 •sssg^ IR55H »in^ ^ ^ a^ftrwrapmm- 

41 «r: [i""] ^ sgi^s*® iR^ii '^w- 

42 «RaiTif»!^ H #?r«r{[ i*] ^ 

43 IRVU wn«R» f^[«^3frt^llt[fiG?: [l*] 51®- 

44 ®a ^aw5Hi<oqT(^ m iR'sn snir ^esr ^ g 55*11 qur- 

45 «wwqf [i*]3^iRRSi^ spHl^taaf^ w nq^ii ■g^- 

46 *5*1^ *i['?r]aj«wa^( [ 1 *] sr*t a(-=-)«>*p®® q;j*Ri^: ®- 

47 TOTRPf: IR»« IRPJT'K® [ S|* ] ^- 

48 ®qns®ig^t lamW ^gsia; iR[<]n eRiRT[ii] 

49 irftq#!RW!i( ~ [ 1*3 5 iffiWw*i^- 

50 E&3®r ir^i ^Wqgqw^f ^[qq] ^- 

51 ®^®% [ 1 *] ^qqqtqq^iff^ qt^«q¥® H?i«H f^[®]?i- 

[103352 *q^qTnfiT[®*3 ® q |owi fi <(% [ 1 *] qg r'g t ^*i»i«*:ii 

53 qqqrfiT ®ra% irsii vn^; qu4«i^ <€1 ^ ^ ^ nn% [ 1 *] 


54 TO qjElShjftJf H^qil ^ 

55 spOTqral [ 1*3 q g'' ww«[ -* ] 5 ^qay ra ® [®}- 

56 8% I wgwst!! i[i*‘3««<*o« g?TO^lai®TO[*n] 

57 nuMsq'^tn ®ifiPw(-)5- 

« RAadg^. 

*• Read-fe^. 
*• Readi^^^ 

«* Hfiad^. 
’’ Readgj^, 

Read *€5%, 

«« Read'g^. 
« Readg^jif. 



58 ^ [i*] 

59 ^ ^ m'<n m- 

60 ssR*T 5i?aM^] 5n[!i05ra! D*] ^rif^g^wry- 

61 ’FRRFT 5IW% 11^ ^11 «nTf^ 

62 gs geqg ft fi rai [i*] 

63 5Wl’»i#5t iJif%(:)^- 

64 f§ri^r(^ «%®n 

Second Stone 

65 II 8iT(-)ira®ft«RW^- 

66 ^ 

67 5q!<srf^ar ^1*62^ 'Esn- 

68 ®i OT [I*] 3i^WH?5wn- 









[104] 77 











8Rg?nsi%s2 i5W[:*] 

^51215^ ll^<n 311^ 

sRsjjjRt^ ^qsf^rai-®® 
w [I*] areq^- 

■gra ft ii^<iii ^■ 

'^goRlflRfl^I^ [u]- 
[I*] »n- 


01 ii^oii gERilig^s? ^- 

% [i*] [2r]»Trfirf^%-'‘® 

gff(-)^ j^42WI ' m 5WT 

«lvi4f^iT: [i*] ^iniiTqospT- 

g ftoRt ii«[i]n 

fjoqt [I*] '^gi^»n[5lr]si f?i[ft]- 
ligwugj iiviiii 

The vertical stroke in the facsimile between ^ and 5 appears to be due to an 
accidental depression in the stone at that point 

^ Read Read wrongly mascuhne. 

« Readj^t^ii}®, 47 Read “ Read“^. 

« Read^, *° probably so. Read^%° 

»* Readl^itl-. »8 Read'liJDt®. ®* Read 




89 50afs^: 

90 ^g?rfw3iw*T&:“ a%5a?- 

91 sisff [i*] 

92 $t?i «3|^?w- 

93 ^4iiqs»g<i t ^rwTO^^nt h««m 

94 ^ si- 

95 5H^a?} [1*] aarai^' 

96 iiEHn 

97 H [i*] 

98 awwwtfrffiMiii/aHi? *T^ «n- 

99 % ««[0>(>) 

100 «^RJ>i5>^-. [»*■] ^iRP#n“^ ['6i]wf%?ti<noi®* 

101 ^ qrgs^ [i*] ss^wra; 

102 «nf^: ^ ^ «Tr- 

103 ii»»« qw'iRRf'iW- 

104 [i*] 

105 ^ HE^U ®*^ 

[105] 106 vtpft [i*] *r 

107 «n «i?oii(-)«nr m%\\ 

108 f*mf?iR«n(-) [w] >wwiww^'- 

109 'fttfj [i*] «lR^TOfW!I ^31®!^ IIH«I1 


111 «i sStfewn?: %- 

112 ST*ra: ^Ir [i*] ^tc5R5RiBr- 

113 [%]*r *fef5f[^>n[ilf]*T 

114 (-)iwTn^4na4^f^*n ^ [gj^r «w- 

113 ^ UHsu ^ i^np^ 

1 16 firei «rflw [i*] 1^1]- 

117 »mn*3 

0* Read'='^t^^jg|5j^.. 

« Bsad^, 

«i Read^sj^. 

** Read 
M Read '^. 

^■0 Read %r ’5^“. 

S’- Readl^^”, 

« Read»^. 

<"> Read ‘-spf}^^Y 

« Read-’ort. 

“ Read5{HfS%?jt“. 


6® Read 

“Read “qig® 

6® Read 

hiatus f In the original an i-sign is also added to 



(L 1) May there be freedom from obstacles ! Salutation to the pre- 
ceptors ! Salutation to the (various forms of) Gana-patis^i who have PushiLi 
(for their wives) ! 

(L 2) The characteristics {lakshcm) of an edict (are as follows) 

(V. 1) Out of the'^'^ mystic syllables {prcM^ava) the sacred 
(syllable) should be uttered first- One should avoid (the use of the letters) cha, 
ka, ta, ha in the ntu (6th), adri (7th), 10th, and rudra (11th) (syllables.) 

(V- 2) At the beginnmg of a composition (the gana) ma, consisting of 
tlnee long (syllables and representing) the Earth, brings bliss ; na with many 
(i.e all) short (syllables, which represents) THAT,^^ (brings) wealth ; and 
ya, with the first (syllable) short (and representmg) Water, (brings) gold , 
]a with the middle (syllable) long, (represaatmg) the Sun, (brmgs)' dis- 
ease ; ra with the middle (syllable) short, (r^iresenting) fire, (brings) fear , 
sa with two short (syllables) m the beginnmg, (representing) Wind, (brings) 
detetruction , ta with a short (syllable) at the end, (representmg) Space, 
(bnngs) lordship, (and) dha with a long (syllable) at the beginmng:, (re- 
presentmg) the Moon, (brings) happmess 

(V 3) Not havmg a visa/rga at the end of the first half , absence of com- 
pounded words at the begmmng , and havmg a visarga (at the end as) 
Sekhfira : (these are) the best characteristics of an edict-stanza.^® 

{^106} (V 4) He (alone) should frame an edict whose diction is ele-' 
gant, who is conversant with Sruti, Smrth Purdm, Itiha$a and Agatna (and 
is also) acquamted with the particularities of time and place 

(V 5) May the merciful Sri (Lakshmi), by Whose extreme grace 
Hleraimba (Gane^a) brought about the weavmg togiether (of events) in the en- 
tire world of movables and inamovables,'^® (and) the lotus-bom Creator (Brah- 
ma) by mere volition brought mto existence the gold-bnght mundane egg, rest- 
mg on the expanse of water, always bestow prospenty on the three! worlds. 

(V. 6) May Achyuta (Vishpiu) protect the three worlds, freed from 
misery, who, in the form of a boar, having extracted with the tip of his tusk 

The number of Ga|na-patis is vanoudy given by different authors. Cf. 
Bhandarkar, Vaisncevism, Saivisni, etc. {Oftmdnss d indo-orkchen Philologie, 
Band, III, Heft 6, pp. 149 f.). 

*72 'The dictionaries mention only one prcmava, namely, the syllable dm. 

It IS the mystic letter forming the essential part of the mantra of any deity 
(Apte’s Ehctio-nary) 

The Earth, Water, Sun, etc. mentioned m this vei«se are the eight forms 
of the Ashtamurti Siva ; and seven of these agree with those enumerated by KMi- 
dasa in the introductory verse of the Sakuftfdld namdy, the five mahdbkutas, the 
Sun and the Moon ; consequently the remaining one, whidi is referred to under 
na^gjona with the pronoun asau, must be the samficmg pnest hotfu 

Needless to say, the author himself does not follow the rules of verification 
laid down here. 

Prbbd>ly in his capaaty of Vighne^vara 




the Earth submerged m the ocean, placed (her, le. the Earth) in (a corner 
of) that (ocean) and (on that account) obtained wondrous and matchless 
merit, (evidently) because there was altogether no such (merit) to compare 
with (known till then) as (that accruing) from the establishment of a tank^^ 

(V 7) May the propitious (Siva), the sole bestower of happiness on 
I^erscms lesorting to him, whose nght and left eyes’'® augment the rain and the 
herbs of the three worlds ; on (whose) brow (is Agm), borne of the Waters,^® 

! whose) friend (isj the lord of nches {Kubera), whose chariot is (the 
Earth) with jewels in her interior, (whose) abode (is) iStigin (iSiiisailam), 
(and whose) bow (is) the (Jolden Mount (Meru), protect you !®“ 

(V, 8) May the Earth (bearing) plentiful crops always protect you — 
blie whose form is resplendent with (her) limbs, namely, the seven contments ; 
with Mera for her head ; the npplmg ocean of milk for (her) beauteous 
breasts , decorated with Rohana^^ and other (mountams) ; with the ghttermg 
oceans for (her) sumptuous garments; and beautiful with rivers and lotus 

(V. 9) A son, a literary composition and a tank, (hidden) treasure, 
a Siva temple, a forest (-grove) a Brahmana-viUage : (these) seven (kinds 
of) offsprings are the best*® 

(V 10) A performer of these and other mentonous works was the 
earth-ruler Bhaskara, sumamed Bhavadura.®® His lineage I shall narrate. 

(V. 11) The Moon (was) bom from the Ocean of Milk Saumya (was) 
the son of the Moon. In his race were bom Pururavas and Nahusha , from 
the latter king Yayati His son (wras) Yadu. In this race of the YSdavas 
(was bom) the azure-robed (Balaiama) and the blessed Hari In (the age 
of) Kali was bom in his family the illustrious king Saiiigama. 

What IS meant is that the exploit of Vishnu is qulite insigiuficaitt when 
compared with the sinking of a reservoir, such as the ond sunk by HiSskara, the 
patron of the poet 

i.e. the Sun and the Moon. 

In the sequence of creation, as desaibed in the Upanishads, Water comes 
after Fire , hence the latter is fancifully represented to be the father of the former, 
s# This IS a reference to the legend of 8iva slaying the demon Tnpura 

fidhana is the name of a mountain in Ceylon It is not unlikely that a 
dhvead of the sense droha^jui C hip ’) is also 

In the Gapapesvara Inscription of Gap-pati (.Ep. Ind., Vol. Ill, pp. 88 ff. 
the seven “offsprings’* are thus described . 

SampaditdT yalhavat sutakritimdhanavivahasuragehmh | 

9ata(Ska$T yak sasaptabhtr etedh samtanavan bhavati Jf 
The Vanapalli plates of Anna Verna (Saka 130)) also allude to them as sabta- 
sathttUi {Ep. Ind., VoJ III, p. 61). . 

In Hemddri-kjitummg^ (I. 39) we have again the wrard kfti used in the rphso of 
' oompodtion *, 

»» Aa remaiked by Dr. Kuuizscaa in the report on Epigraphy for 1902-3 (see 
p, 6, para. 15), Bhavadura be a Sanskntised fonn of Bahadur. 


£107} (V 12) This king Samgama, having worshipped (the gods) Han 
and Hara, obtained by their grace a son, king Harihara. 

(V 13) The supreme lords of the Middle Country‘‘ (Madhya-desa) 
e'.tending from the eastern to the western ocean, viz. theceld>rated (kings) of 
the Solar and Lunar races, were occupied in doing “ foot-salutation” to him. 

(V 14) From his causing hostile kings to tremble his uterine brother 
was (known) in the world (as) king Kampana After him his youngei 
brother Bukka, the crest-jewel among kings, the husband of Tjilfabm? was 
the anjoyer of the Earth, who was perfectly constant (to him). 

(V 15) (Then) were bom Marapa and Muddapa, two brothers of king 
Bukka. And these five virtuous sons were incarnations of the Pandavas in the 
age of Kali 

(V. 16) Vasudieva, who loves his worshippers and (who had acted) in 
former times (as) the messenger (and) the chaiioteer of the IPSindavas, 
(having countless incarnations), became also numster of these in the form 
of Anataita,®* 

(V. 17) Through having Anairhtaraja for his minister kmg Bukka raled 
over the whole surface of the earth and acquired the glory of DevSrhdia 

(V 18) His aty, Vijaya by name, (was situated) on the bank of the 
Tuthga-bhadia near (the temple of) the blessed Virupaksha (Siva), well 
propitiated cai account of steady adoration. 

(V. 19) Formerly the number of the sons of iSa^abthdu (was very 
great) m the world That is but a story ’ (?)' The countless sons of king 
Bukka were exalted through triunpih 

(V. 20) Out of these sons king Bukka had placed Bbaskara, exalted 
through the gory of mdependent sovereignty, in the eastern' direction (of his 

(V 21) And he, rulmg from the top of the sublime Udaya-giri®^ the eaith 
freed from the thorns (of enemies), though Blmkara, (le. the Sun, is still) 
the delighter of the Earth and beloved of the Bnahmainasi®® 

As the rarly Vijayanagar kings had no daim to soverdgnty m any part of 
India north of the Vindhya, MadhyadeSa cannot have its usual significance, but 
must refer to the country lying betwem the eastern and the .western ocean, namdy, 
the Dekkhan plateau. 

®® One must supply a verb like krtavm m the firs/t half of the verse 

®e The construction of the first hal f of the verse is not quite dear to me In 
Ch 65 of the Drona-parvan of the Mahabharata, we are told that Sa^indu had 
10,000 wives, on each of whom he begat 1,000| sons These, it is stated, he gave 
away to the BiShinaiias m the ASvamedha sapifice which he performed The 
“ countless sons ” of Bukka are hxs meritorious acts hk© the saptasamtdna men- 
tioned m V. 9. 

®^ This must refer to the fortification on the top of the Udaya-gm hill Even 
now U. is an exoeedmgly strong hiU-fortress. 

®® The pun on the words kuvdaya (‘night-lotus’ and ‘earth’) and dvxja 
(‘BiShmaija and Moon’) is a very common example of the VjrodhabhSsa, 



the Earth Mibmerged m the ocean, placed {her, le the Earth) in (a corner 
of) that f ocean) and (on that account) obtained wondrous and matchless 
merit, (evidently) because there was altogether no such (merit) to con\pare 
with (known till then) as (that accruing) from the estabhshment of a tank/^ 

(V 7) May the propitious (iSiva), the sole bestower of happiness on 
persona resorting to him, whose right and left eyes^® augment the ram and the 
herbs of the three worlds , on (whose) brow (is Agni), borne of the Waters,’® 
' whose t friend (is) the lord of nches {Kub^ra), whose chariot is (the 
Earth) with jewels in her intenor, (whose) abode (is) iSrigin (iSrisailam), 
(and whose; bow (is) the Golden Mount (Mem), protect you 

(V. 8) May the Earth (bearing) plentiful crops always protect you ’■ — 
slic whose form is resplendent with (her) limbs, namely, the seven continents , 
viith Meru for her head , the npplmg ocean of milk for (her) beauteous 
breasts ; decorated with Rohana®^ and other (mountams) ; with the glittering 
ocians for (her) sumptuous garments ; and beautiful with rivers and lotus 

(V 9) A son, a literary composition and a tank, (hidden) treasuie, 
a ^iva temple, a forest(-grove) a Brahmana-village : (these) seven (kinds 
of) offsprings are the best*® 

(V 10) A performer of these and other mentonous works was the 
earth-ruler BhSskara, surnamed Bhavadura.®® His lin pag p I shall narrate. 

(V. 11) The Moon (was) bom from the Ocean of MiUc. Saumya (was) 
the son of the Mocmi. In his race were bom Pururavas and Nahusha , from 
the latter kmg YaySti Hig son (was) Yadu. In this race of the Yadavas 
(was bom) the azure-robed (Balarama) and the blessed Hari In (the age 
of) Kali was bom m his family the illustrious king SamggTnn 

” What IS) meant is that the exploit of Vishnu is qufi.te inagnificarit when 
con^ared with the anking of a reservoir, such as the one! sunk by Bhaskara, the 
patron of the poet 

i e, the Sun and the Moon. 

In the sequence of creation, as desenbed in the Upanishads, Water comes 
after Fire; hence the latter is fancifully represented to be the father of the former 
®» This is a reference to the legend of Siva slaymg the dpm n n Tnpura. 

Rohaoa is the name of a mountain in Ceylon It is not unlikely that a 
dhvam of the sense diSka^ ('hip’) is also mtend^ 

S'* In the Gaioapesvara Inscription of Gap-pati {Ep, Itid., Vol. Ill, pp 88ff 
the seven " offsprings ” are thus desenbed : 

SampSditeffr yathaoat sutakjittmdhanavivdhasuTagehaih | 
satatSkm yah sasaptabhir etmb samtdnavan bhavati || 

The jgates of Anna Verna (Saka 1300) also aUuda to them as sapta- 

Ind,, Voi* III, p. &l). 

In Hem^r^kriti-fnar^e^ (J. 39) we have again the word krti used m the sense of 
cotxipoaUoti \ 

« As rema^ by Dr KuttzscH in the report on Epigraphy for 1902-3 (see 
p. 6, para. 15), Bfaavadura seems. to be a Sanskntised form of Bahadur. 


{1073 (V. 12) This king Satngama, having worshipped (the gods) Hari 
and Kara, obtained by their grace a son, king Harihara 

(V. 13) The supreme lords of the Middle Country*'* (Madhya-desa) 
extending from the eastern to the western ocean, viz the celebrated (kin^) of 
the Solar and Lunar races, were occupied in doing “ foot-salutation ” to him 
(V 14) From his causing hostile kings to tremble his utenne brother 
was (known) in the world (as) long Kawnpana. After him his youi^ei 
brother Bukka, the crest-jewel among kings, the husband of Lakshmi, was 
the onjoyer of the Earth, who was perfectly constant (to him). 

(V. 15) (Then) were bom Marapa and Muddapa, two brothers of king 
Bukka. And these five virtuous sons were incarnations of the Pandavas in the 
age of Kali 

(V 16) Vasudieva, who loves his worshippers and (who had acted) in 
former times (as) the messenger (and) the charioteer of the IPSindavasi, 
(having countless incamaticms), became also mimster of these m the form 
of Analmta.®« 

(V. 17) Through having Anarhtaraja for his minister king Bukka ruled 
over the whole surface of the earth and acquired the glory of Devemdia 
(V 18) His city, Vijaya by namcj (-was situated) on the bank of the 
Tuithga-bhadra near (the temple of) the blessed Vinupaksha (Siva), well 
propitiated on account of steady adoration. 

(V. 19) Formerly the number of the sons of Sa4abimda (was very 
great) in the world. That is but a story ! (’■) The countless sona of king 
Bukka were exalted through triumph «« 

(V 20) Out of these sons king Bukka had placed Bhaskara. exalted 
through the gory of independent sovereignty, m the eastern' direction (of his 
empire) . 

(V 21) And he, rahng from the tcqi of the sublime Udaya-gin®'* the eaith 
freed from the thorns (of enemies), though BhSskara, (le the Sun, is still) 
the delighter of the Earth and beloved of the BrShmainasi®® 

84 Aa the early Vijayanagar lying s had no daim to sovereignty in any part of 
I n dia north of the Vuidhya, MadhyadSia cannot have its usual significance, but 
must refer to the country lying between the eastern and the .western ocean, namdy, 

the Ddrkhan plateau , ,, , u 

8® One must supply a verb like kftavofk m the first half of the vei3e. 
se The construction of the first half of the verse is not quite to me In 
Ch .66 of the Drona-parvan of the MahSbharata, we are told that Sa&butou had 
10,000 wives, on each of whom he begat 1,000| sons The^, rt is stated, ^ve 
away to the BrShmanas in the Afivamedha saicnlice which he performed. The 
“countless sons’’ of Bukka are his mentonous acts like the sapta^amtam men- 


tioned in v. 9. , . 

87 This must refer to the fortification on the top of the Udaya-gm hill. 

now U. la an exoeedin^y strong hill-foitress , , ^ , , * _ i 

The pun on the words kuwdaya (‘ mght-lotus e^ ) and 7 

(‘ Ttifihtnaria and MooD ’) is a Very common example of the Vmodhabhasa. 


epigraphic studies 

iV. 23) Making chanties in various ways in keeping with the treatise 
of Hemadri,"® he heard that the ment attaching to the gift of water was the 
greatest of all 

(V. 23} On the authority of the Vedas . “ Venly all this is water ! ’"'® 
And the Smti says that : “ From water alone is produced Food , (and) Food 
is Brahman ! " 

(V. 24) There can be no doubt (that) Water alone is the seed of the 
world of movables and immovables. Why speak more ? I shall describe the 
bupenonty of water (as follows) — 

{108} (V 25j Even that (great) Siva is the bearer of the (Jamga ; 
Vishnu has the ocean for his abode , Brahma is sprung from the water-born 
(lotus). Haice Water is superior to eveaything (else) 

(V. 26) A shed for distnbutmg water {prapd), a well and a reservoir, 
a canal and a lotus-tank : the merit of (constructing) them is millions and 
millions (of times) higher in succession 

(V 27) As the water of a tank serves td nurture both movable and im- 
movable creation on (this) earth, even the lotus-seated (Brahma) is unable 
to recount the fruit of merit (attaching) to it 

(V. 28) Having thus heard the supreme reward, king Bhavadura, the 
pious soul, commmenced to make the earth tank-nourished {tataka- 

(L 48) Its procedure (was as follows) 

(Vv 29-31) It (ie the tank) is situated in the country to the south of 
Srlparvata (SilSaila), the great sacred place of pilgrimage ; two to 

the east of the sacred place (^Tirtha) called Ahobala , in the division of the 
gentle blessed Siddhavata-natha ; two yojanas to the west of his (capital) 
Udaya-giri ; and to the east of the flourishmg city of ParumSmilla I shall 
(now) describe in this edict the sequence of the period of construction of the 

(Vv. 32-35)®’’ In the second half of the creator’s hfe-time in the Sveta 
varSha Kdpo, m (the age of) the Vaivasvata Manu, and in the 28th Yuga, 

•* Namely, the DSna-khamja. Henmdn was the minister of the two Yadava 
kings Mahadeva (128D-71) and EStoajdiandra (1273-1310) See Bhandarkar, 
Early History of the Dekkhon, pp. 88 f. 

The i^iraae SpS vS, etc. la a part of the mantra with which water is purified. 
Idagurh is the word idam as it is prcmounced by the Yajurvedins in the reatation 
of 'V'edic texts. 

•r With tataka-muifika cf. the terms dhia-mStrika and nadt-matrika in. a 
Tatoag a ydjana to be equal to 9 mflea, Hus distance is only approedmately 


•* CooBtrue : dhStol^ parSdhakSle. * . ,divye tadde§(hbhage rnnmtosya iafSkasya 
doddi^athiam vitkshydmi. 


—in that divine part of the country— in the first quarter of Kali after the 
lapse of four thousand, four hundred and seventy— (in figures) 447(>— years 
of mortals, and also after the (lapse) of Saka years measured by the number 
of the earth (1), the Naimdas (9), the eyes (2j, and one (1)— (m figures) 
1291— m the (cylic) year Saumya, m the month called Karttika, on the 
fourteenth (day) of the bright half, on the auspicious day of Gum combined 
with Pushya, when there iwas Karkataka lagna, under the influencei of well- 
chosen auspicious planets, — 

(V. 36) Of the tank constructed (at the above specified time and place) 
according to (the requirements of) the Sdstra I shall in this edict descnbe 
tlie twelve constituents (amgcfi^) for the benefit of future kings . 

(V 37) (i) a kmg endowed with righteousness, rich, happy (and) desir- 
ous of (acquiring) the permanent wealth of fame, (ii) and Brahmapa learned 
ni Hydrology {pathas-§dstra) , (m) and ground adorned with hard clay, (iv) 
a nver conveying sweet water (and) three yojanas distant (from its source®® 
(v) the hill parts of which are m contact with it, (le the tank),®® (vi) be- 
tween these (portions of the hdl) a dam (built) of a compact-stone wall, 
not too long (but) firm, (vii) two extremes (irmga) (ix)mtmg) away from 
fruit (-giving) and {phdasthird) outside,®^ (viii) the bed extensive and 
deep, (ix) and a quarry contammg straight and long stones, (x) the neigh- 
bounng fields, nch m fruit (and) level, (xi) a water course (i.e the sluices) 
having strong eddies (bhromd) on account of the position of the mountam 
{adri-sthdna) (xii) a gang of men (dolled in the art of) its construction, 
—with these twelve essentials an excellent tank is easily attainable on (this) 

£1093 (V. 39) While (i) water oozing (?) from the dam, (ii) saline 
soil, (ui) (situation) at the boundary of two kingdoms,®® (iv) elevation 
{kurma) m the middle (of the tank) bed,^®° (v) scanty supply of water and 
extensive stretch of land (to be imgated), (vi) and scanty) ground and ex- 
cess of water : (these are) the six faults m this (connection) 

(V. 40) Devoid of faults and adorned with a multitude of good qualities, 

®^ These cemgcts are later on called sadhaMsi (1. 70 of the text) and are no 
doubt identical with the latter. 

®® See above, p. 99. 

®® Thi s must obviously refer to the range of hills which is utilised to form a 
part of the dam 

I.e. below the tank. 

®® This may be taken to be a clumsy description of the fact that at the egress 
the water is led over a dnny bed alongl a tortuous line, so that it issues* whirhng 
round with great force, formmg strong eddies. 

®® Perhaps as, in tins case, the position of the tank mig^t lead to unpleasant 
consequences during a conflict between the neighbouring kingdoms. 

10® The bed ougjit to form a complete valley 


epigrAphic studies 

renowned in the world by the name Anaiiitaitja, this endless ocean, of which 
the water is sweet, was founded by king Bhaskara, 

(V, 41) (There were) one thousand labourers (working) at the tank 
and dam every day, and a hundred carts (were employed) for the masonty 
work of the sluice and wall (bhrama-bhitti) , 

(V. 42) And this most excellait tank was completed in two years. There 
IS, to be sure, no limit at all to the expenditure of money and gram in this 

(V 43) The measurements m teams of rekha-domdasyo^ of the height, 
tl«; width, and the length of the dam together with (the portion of) the hiU 
(included in the) dam, are here given ■ 

(V 44) The dam, having eddying waterducts (i.e, sluices) (and) 
piotected by VighnSa, (Gaaja-pati), Kvara (Siva), Vishpu, Bhairava, ana 
the gieat Durga, is one which has the enormous length of five thousand rekhd- 
aamdas, haght of seven and its width eighth” And the land (is) excellait 
and yields plentiful crops in all seasons and contains groves 

(V. 45) This land was liberally givai for the gratification of gods and 
Brahmanas. Through the merit of this gift of land the tank was made to be 
an ornament (of tanks). 

(V 46) Just as the dam of a reservoir should not be injured, soi like- 
wise the dharma-ABm of the ocean of kings Therefore I, Bhaskara, repeated- 
ly request the kings on earth to protect my chanty 

(Lv 99f.) These are the ancient slokas of entreaty : 

[Then follow three of the customary verses.] 

(V 50) /The Ofiicer-ia-charge {adhikdrm) of this tank is the clever son 
of the minister called Kum&ragiri-natha, Devajj^ja. by name. 

(V. 51) Having obtained from king Raghu gold by the crore, Kautsa 
gave (it as) dakshtna to his preceptor, Varatamtu, who had bestowed on him 
the fourteen branches of knowledge By a descendant of his (sctl Kautsa)^®* 
the illustrious LimgayaMachanaryya,'*®* of Namdapura, best of Bifihmainas 
and a follower of the RgvSda, the auspicious edict was composed 

(V. 62) (One) k/tdri (of land) producing paddy and (cme) khdri of 
black-soil land,— (these) were out of regard given to him by Bhfidcara, pre- 
ceded by a libation (of water). 

A standard c4 linear measurement roughly equal to yard. See p. 99 
For bhrami-jala-gati see note 8 on p 108. [=96 supra.] 

The pronoiin tad in tad-mstara cannot be taken to refer to the noun imme- 
preceding, but must refer to the setu of wfaidi the dimensions are being 
given. See 1 86 of the text 

The sente is that the poet belonged to the Kautsa sotra. There is, how- 
ever, a ccnfuaco in the mind of the poet between Kautsa the patriardi, and ^utsa 
the pupil of Varatantu aDiided to in the St^hu-varhia, Canto 5, w. 1 S. 

Le, son of Uxogaya. 


The copper-plates which bear the subjoined inscnpfaon of the Rashtrakuta 
king Kfi^aiaja I were discovered at Bhandak, TaJpil Ward|da, in the 
Chanda District of the Central Provinces They were forwarded for exami- 
nation by the Commissioner for the Nagpur Division, through Dr D B. 
Spooner, to the Assistant Archseological Supenntaident for Epigraphy, 
Southern Circle. I am now editmg them from the original plates as well as 
a set of impressions kindly placed at my disposal by the latter 

The plates are thiee m number, each measunng roughly lO^ by 6S ms, 
and (Weigh 340 tolas The margins are folded oved and beaten down, so as 
to serve as rims The grant is engraved on the inner side of the first and 
the third plates, and on both sides of the second TTie plates are pierced by 
a circular hole, | m. in diameter, m order to receive the ring and seal, which 
are, however, missing —The engraving is deep, but not neat The letters, which 
are uncouth in shape, vary m size from f to i in The letters dia, pa and ya 
have been mdifferently incised and are consequentiy diflScult to distinguish 
from each other , so also the letters va (ba) and dho The hgatuie nto is 
often so carelessly wntten as to be indistinguishable from tta Some letters, 
agam, sporadically show quite strange forms, as, for example, su in rSjasu 
at the beginning of Ime Sa in Smafishu, 1 3, le m bdena, 1. 28, etc. — 
The characters bedong to the! northern dasa of alphabets, and, likel those of 
the Mulla! plates^^ of the R&shtrakuta Nanda-raja Yuddhasura, represent 
the last phase of the acute-angled variety.® The medial u is marked by a 
short stroke slanting upwards, attached to the matpka at its right lower end ; 
sporadically by a curve opemng to the left, as, for instance, in the iit of iurfo, 
1 3. The sign for the medial e is a short vertical stroke app^ided to the 
top of the mdtrika on the left; and, only very rarely, by a stroke above 
the motrika. The Central bar of }a slants downwards, but is not 
vertical , the lowest bar does not form a double curve, but merely slants 
downwards towards the n^t and only ^radically ends in a small notch 
Those comers of the letters kha, ga and Sa, which later devdop into loops 
or trianglesi, are m our inscnption marked by small projections or notches.® 
The verticals on the right of the letters are short and project but httle bdow 

* [Ep. Ind 14. 121-130.1 

1 BiiHUER, Indische Palaeosraphie, Tafd IV, Col. XX 

2 Buhler, op. cit p. 50. 

® Eg. kha m mukho, 1. 2, mum, 1. 4; ga in gotra, 1. 8, vlga, L 10, fa w 
foasasati, 1. 11, Stkhara^i, 1 12, etc. 

epigraphic studies 


the remaining portion of the signs* These palseographic characteristics are 
sufficient to establish the archaic character of the scnpt and to prove that 
the plates belong actually to the penod to which they refer th€!n)selves> viz 
the third quarter of the ei^th century of the Christian era ^ Our record con 
tains specimens of initial 5 in 1. 2 , initial i in 11. 3, 8 , initial ti* m 1 . 37 ; 
initial e in 1. 53 , a cursive form of ifew in 11 5, 12, 23, etc , and the follow- 
ing ligatures, nka fl22} 1 9, nga 1 36, ngMi 1. 17, ncha 1 6, nda 1 2, ndya 
1, 27, ksha and kshma 1. 14 jna 1 8, Ipha I 16, shtvd 1 4 ; and lastly final t 
(?) in 1. 33 — ^As regards orthography, the only points worthy of notice aie 
tlie following . (1) the use of gha for ha in rdjasighah, 1 4 (cf . also 1, 23) ; 
(2) no distinction is made between fr and n , (3) no rule la followed with 
regard to the use of anusvdra in the middle of a pdda , (4) wrong conversion 
of the musvara into n before a sibilant in ""likhit-dnsa^ 1. 14 ; (5) once the 
use of da lex dha in dadatd, 1. 5 , (6) the use of the vowd ji ior ri in 
tnpishtapa'' (for ''trivtshtapa''), L 7.— The grant commences with a symbol 
representing orit. Then follows the stanza sa v6 = vydd = vedhasd dhdma^ 
etc, which stands at the beginning of, I think, all the early Rashtrakuta 
records. The rest of the composition is also m Sanskrit . the prasasti, the 
benedictory and imprecatory stanzas bemg in verse, the grant proper m prose. 

* In thef yi of ym = eyam (1. 31 we have an instance where the vertical 
stroke is altogether wantmg 

The Saniangad grant of Dantidurga (ed Fleet, Jnd. Ant, Vol. XI, 
pp. UO ff.) and the Aiks grant of Yuva-iSja Govmda (II ) (ed. D. R Bhandarkae, 
Ep Ind , Vol. VI, pp, 208 ff ) are two Rashtrakuta records whidi bear the dates 
575 and 692, i.e* are dated earlier by 19 and 2 years respectively than the 
grant which is the sfubject of this article The palaeographic differences between 
these three grants are worthy of consideration. The alphabet of Govinda^s grant 
15 wholly different horn that of our record. The scnpt is entirdy Dravidian m 
character ; the letters are round m appearance, and are akin to those of the grants 
of the later CSlukyas of BddSmi, the immediate predecesteors of the Rashtrakutas 
ITie difference la clearly a local one and is perfectly consistent with the geographical 
limits over which the southern aljAabet was current Such is not, however, the 
caae with the other inscription. The alphabet of the Stoangad grant belongs to the 
same categpiry as that of our grant, and represents an archaic vanety of the Nagari 
The difference He^ however, in its showing just those peculiarities which characterise 
the scnpt of an epoch some decades later than that to which it refers itself. The 
regular rign for the medial e in this grant is a curved stroke on the top of the letter, 
while the short vertical stroke on the left appears only oocaaonally as representmg 
thie letter. The nght-hand portion of gha, pa and sa shows the development of 
long verticals on the right of these ^gns The letters kha, ga and ia show distinct 
{122} developments (rf loops, where our record has/ Ojnly straight pirojections or 
notches. The middle bar of ;(? approximates more to the vertical, and the lower por- 
tion loam a distinct double curve. These facts are dear indications of a later 
palaeographic ^xxh and raise susp&cionsi against the bona fides of the grant. For this 
and other reasons I am indined to entertain the gravest doubts regarding the authen- 
ticity of the S&mangad gmnt But, as I intend dealing with the question at length in 
a separate ardde devoted to the subject, I do not widi to enter into details her$. 



Most of the verses of this record are repeated with sli^t verbal differences 
in one or other of the following grants • the Sfimlangad grant of Dantidurga,® 
the Alas plates of Govinda II the Pathap® and the KBvi® grants of 
Govinda III. Of these it approaches closest to the first mentioned, vir. the 
SamShgaid grant The four verses 5, 9, 21 and 22 I have not been able to 
traoel anywhere else In recounting the exploits of Dantidurga all the early 
Rashtrakuta grants repeat the two well-known verses, Kmchiia^ and sa- 
t)h‘iuvibhanga,° etc^® Our grant has mstead only one stanza, made up of the 
two half verses belongmg to the two stanzas, a deficiency whidi, I should 
imagine, is merdy due to the negligence of the scribe. In other respects the 
execution is satisfactory. And with the hdp of this text we are placed in 
a posibon to correct the extremely conupt text of the San^gad grant with 
respect to those verses which it has in common with our grant and which do 
not occur dsewhere. 

The grant, as already remarked, is a record of the Rashtrakuta kmg 
Knshina-iiaja I and is of particular importance, being the first record of the 
king to be discovered so far.'^^ Another grant which refers itself to the lagn 
of Krishna I is the Alas grant of his son Govinda II., while yet a yuvor&jct ; 
it was issued m ISaka 692, that is, two years previous to our record.!® The 
genealogy of the RSshtrakutas given in the present grant commences with 
Govinda I., as in all other early grants of this dynasty, excepting the unfinish- 
ed inscnption from the Da^vatara temple at Ellorai!® ; and the ddails regard- 
ing his successors 'Kakka-raja and Indra-iSja accord wdl with what we know 
of them from other records Here agam, as in the Slamangad grant, the 
queen of Indra-iSja is described as being a Chalukya pnncess, tracing her 
descent from the Lunar race on her mother’s side But from the new recoid 
we gather some more mformation about her, which in the mangled version 
of the Samangad grant was distorted beyond recognition. Tlie defective 
anushtubh half-verse. 

e JBBRAS, Vol. II, pp. 371 ff , ed Fleet, Ind. Ant., Vol XI, pp JlOff., 
and Plates 

1 Ep Ind , Vol VI, pp 208 ff, and Plate 
8 Ibid, Vol III, pp 105 ff., and Plate. 

» Ind. Ant. Vol. V, pp. 144ff 

“ The KSvi grant, w. 8, 9 , the ^mfingad, w 18, 17 fin the reverse order!); 
the Alas, w. 5, 6 ; the Paithau, 11 11-14. 

Since writing theate lines I have oomfi to know of the recent discovery of 
anotha record of Krishma-raja, viz the Talegaon (Poona Distnct) plates dated 
in the year Saka 690, vide Progress Report of the Archaeoilogicail Survey of India, 
Western Circle, 1913, p. 54. [The inscription is published in Ep. Ind., Vol. XIII, 
pp. 275-282,— F. W. T ] 

1* Ed D. R Bhandabkar, Ep Ind., Vol VT, pp. 203 ff. 

M Edited by Bhagvanlal Indraji, No 10 (p 91) of the separate panaphlete 
of the ArcM. Survey of West Indkf, 



Srttaad-yuvatt-gatjmam sSdhv'main — apa m{sa) padam I 
of the Sain3ngaid grant, stands for some original like 

5n»iad-Bhavagauia netma sddhvlnam — upcemapcdam 

whidi I translate with ‘ iSrimad Bhava-gana by name, the {very) standard 
cf comparison ^123]} among virtuous and chaste women’ The name of the 
c^uecn was therefore Bhavagana. Di. Fleet translates the corresponding 
half-verse of the Satn&ngaid grant as follows . ' She attained the position of 
honourable young women who are faithful wives ’ On comparmg my trans- 
lation with that of Dn Fleet there will be no .doubt as to which reading is 
to be preferred Coming to Krishna himself, in addition to his birudas 
fiubhatunga and Aloalavaraha, which we know from other inscriptions as well, 
he appears to have also assumed the title ISri-pralayamaha-vaiaha. Besides 
these three birudas this record ccaitams no further historical information, about 
him, and it would therefore appear that it was issued in the early part of 
his rdgn : at any rate, before the event of the construction of the Elloca 
temple, which event is described with such pomp and ceremony m a later 
record of this dynasty 

With regard to the charge brou^t against Krishna by Dr. Fu®t^® that he 
’bad uprooted bis relative Dantidurga, who had resorted to evil ways and 
apprc^riated the kingdom for the benefit of his family,’ I hc^ this record 
of Krishna-tSja himself will have the last word to say and that too m a 
decided negative. The weak points of Dr Fleet’s theory have already been 
pointed out with sufficient clearness and force by Mr. Devadatta R. Bhan- 
DARfCAR recently in his article on the Al^s plates of Govinda II It is here 
suflSdent to point out that Dantidurga was no licentious weakling, but a 
very powerful and, probably, also a popular king In fact, he was the first 
king of his djmasty to assume the title of Rajadknaja-ParantSivara, or, to 
quote the words of Dr. Fleet himself, ‘ he was the real founder of the dy- 
naaty/’^’f In our grant, just as m the S&manga<} grant, he is called the ‘ son 
to the lotus {which was) his family’; both these records lay stress on his de- 
votion to his mother m unmistakable terms. It is, therefore preposterous to 
identify the relative of Knshina ‘ who had taken to evil ways ’ with the founder 
of the dynasty, Dantidurga, who had merited the epithet sva-ktd-ambhaja- 
bhdshara Besides, were KiMuja really guilty of the murder, it is mconceiv- 
able that he should have tolerated the eulogy showered upon the murdered 
unde in a grant oJ his own and coolly added that he ascended the throne 
after the victim of the assassination had gone to heaven ! 

R- G. Brmuxabkar, Early History of the Dekkan (Bombay Gazetteer, 
Vol. I. Part H), p 19&. 

Kanarese Dynasties, p 391. Ep, Jnd., Vol VI, p 209 

*r Eartartse Dynasties, p. 389 

BHANDAK plates of KRISHNARAJA I. §AKA 694 20i 

The formal part of the grant records that tlie long, being encamped at 
Nandi-pura-dvaiS, granted on the occasion of a samkrdf^U, at the request of 
one Madana, the village of Nagajijapuri to the Bhattaraka of the temple of 
Aditya in the town of Udujnvara-manti The condudmg verse gives the name 
of the writer as Vamana-[nS,]ga. 

The grant is dated m the iSaka year 694 eigiired on the third day of the 
dark half of AshSidha, which was, as remarked above, a SaiftkiSnti. Dewan 
Bahadur Swamikannu SPillai, who kmdly examined for me the details of 
the date^ informs me that the titht mentioned m our record ended <mi the 
23id June (Tuesday) ad, 772, at about one ghatikd after sunnse, the day 
was also the first day of the solar month Karkajlaka by the Tamil rule 
The Karkataka Samkranh fdl on June 22nd (i e. on the previous day) at a 
little before midmght We are therefore led to assume that in the present 
case the first day of the avil month was called Saihkranti, though the astro- 
nomical Samkrmii fdl on the previous day, a supposition which is counten- 
anced by the practice actually followed m Southern India in certain wdl- 
known instances. 


[Metres: v. 1, Anushfubh (Sldka) ] w 2-7, Vasontatilaka; v. 8, 
Anushtubh {Sloka) , v, 9, Indramjrd , w 10-16, AniKhtubh (Sloka) ; 
V. 17, VasantatilakS ; V. 18, Sardulemki^dita ; w. 19-24, Ary 5; vv. 25-26, 
Anushtubh iSloka) ; v. 27, defective Arya.] 

1 [«*] « fa [i*] ^ 

£124} 2 

5 fSjai I 



aftWHR ^ \\[n*] aila 

4(m i t^Fa4l aT[-=-*] -*] [»*] »»■ 


18 From the ongmal plates and a set of impressions. 

18 Represented by a Symbol 

so Read 'IlfiRl". Read ** Rea'i 

*8 Read ^aaf. 

*6 Read ‘’Sjftsft'- 

Read Note aa iibsculme > 
Read *' Read 



204 EpigrAphIc STtfWES 

11 g: |[I'^«*] te[-*] lanTiJTfSra^! 

13 5l5i^PWt*W! ll[Sll*] 5re*I sif*i'M'M6-«^»i v't'iPj'i'nHs 

14 I ?»TW: 

Second Plate; Ftrsi Side. 

16 oiT 0*] wsjf !asr5i[?]flr frl^ i[i<£ii*] 

17 ?ftr^rfiT^^a]gn5ii'® ^ [i*] w^sra#.teKU3ft^:“ ?r[-*] 

18 «*>[:*'] n[«\ii*] <ii^3Rn g^«$?^sgt3ai'- 

«#r m!!5i[: I*] 

19 ?Ra ft5aat [ii«oii*] ?«ti75sa?P»tf aw si- 

(125 J 20 [i**^] i[imi*] 


21 aa cTCTf ?r i(i) “*ftd?a4eiaj$t[«r]5ii?raT-^^ 

22 snfifera<?f [iiisn*] i [ sg]^ [i*] sft- 

23 ^Pd44<wK«i[-] i[ni^u*] sRai^*** ?in^isn?i" 


24 ^?reir 

i[r?«ii’'''] [5rT]5r- 

5!ras% »if^ [^] »rai[:*] 

25 Bf^fii E^j ^ fi(3i'ii*H[ 9' ]fia 

26 i[»'<ii*] **»T:g?flifiB[:*] irfimw [i* 

« Read a* Read ^nf". ao Read 

»» Read °!^“. 



fead The t-siga erf is appended to the symbol for | 

K«ad JP*R^. ^ Read '^. 

^ V^n.The i-sign in seems to be appended to the symbol forSH 
*« See note 6 on p, 127. [ = 74 infra.] ^ 

»» Read t«r* « Read R3|f}°. 

« See note 1 OB p. 128. [ = 78 infra.] 

Read'^rraf. ^ « Read%^:“. « Read 



i* Riad 

^ Read ‘'drThj\ 

Read JinT® 



27 *TraT swr^rar ii[UH*‘] 

28 [i*] ^ 

?i[ 3 n> 

Secottd Plate ; Sefiond Side. 

31 3i?i^ 

32 ii[scH^] aRflP<i4 ipn^ «fRra3n?rnsr:** [i*] 

33 srggsftft^iRt: fwiR i ^q^L l[is*tl>^] ^ 

34 [!’*'] «ftfWRTSR«ni[^on] ftqJtg 

35 ^ [i*] ^Rug gg«5R: e*n?i: aoi^ g- 

36 n[^in*3 gsf^ ^ Rg? ffR}®* ^srRi^ «pnwR5» w- 

37 [i*] «i: ?Fa?mRra^OT5i5sg^ ^*mi: i[r^h*] g^- 


38 ®^*n^ spfeisi®4l[ « n §gf Rwagdt 0*] 

1126} 39 ^ i[r^«*] ^tf^awsR [I*] 


40 sraf^ “^EURpitgC-*] n[^«»i*] h ^ <hot«5i»- 

*i?Ri 5 irRra- 





CTR r ^w<T lj ml<wRm^ g d tHi)*<i!aiwar^ g: “«RgRa 
*TOi »Rn *!RttftRRi»Rar i 

Third Plate. 

45 RRR^ [R]^(^)^ 3 rm«n 

46 g^gwRara^ *wRdi(^«waH*(ywW g^gc^* 

« Read 
**• Read 
M Rfiad 3 R;r“. 

®* Read •il*^'’. 

6* The M-sig^ (nieikal)' is 

« Read “gig;. *® ReadgRPt, 

®i Read^gf. ®* Read“ft#lt. 

»* Read€Rr°. '* Rea 3 "gR". 

ST Read“gi“ 

appended to the symbol for Ri, 



47 ofrpigi^5rPR5mit ^(••)'^['^]' 

48 % it q ’ ia t < 3o g^ l 5 g q [ I ?ppi ■«tiyiiHA 

49 ^i[i^] [if [ii*] [II*] 


50 err[i'^] [«*] 3TR«r;[i^] jnjr:i[t*] “^^qraT- 

51 2^®: [l*] 3*qOT^5rer- 

52 W4i*wKi<3r: ?rf^f5ra^' ^ [i*] 

53 1 ?^ ftq^5ra[-*] i[i*] 3 ^ wsrf^:*] 


54 ^ «R*i ^ «i?T 3?r [“%]® «[qHii*] <«[?]- 

■an^'’ ’qi' 

55 ag s qt i 1 qd dsrqf® [f]1^r: H[q^ii*] 


56 3»it®i(?) [ 1 *] “®qq%ifqi9*i«n fedra 
5ftqm[qTf[q] [ir«ii*] 



(Verse 1<) May he (scil. Vishnu), the lotus on whose navd Brahma has 
made (hts) dwelling, protect you ; and Hara (i.e (§iva), whose forehead is 
adorned by the beautiful moon-crescent 

{127]} (V. 2.) There was a kmg called Govinda-raja [I ], a royal lion 
among kings, whose fame reached to the ends of the regions, (and who) 
pure (of conduct), liftmg (hts) samitar (and) facing (them), destroyed his 
enemies in battles, just as the lustrous Mbon, whose glory (te. radtance) 
penetrates to the ends of the r^ons raisuig the tip of (his) orb (above the 
horizon and sending hts rays) straight forward, dispels at night the <iarimPiBg 

(V. 3.) InvanaWy, when he saiw on the battlefield the armies (of the 
enemies) ermfrontmg him, ringing with the loud laughter of warriors, forth- 
with he, biting (Ais) lip (and) knitting (his) brow, elevated (his) swoid, 
(hts) family, (his) heart and (his) pride*® 

(V. 4.) His son, the glorious Kakka-raj’a [I.], was the gem of the 
i Rdshtrakiita) race, a king who was grateful (foe services rendered)', whose 

“ Correct, perhaps^ to qpipil jq". *» Read 

« Readqg". oa Read^FRT'’. Read«S0. 

« Read^^qr. «« Read%rqf. «« Read 

Read qiqi^^^qt^ ®* The last quarter of this Aryil is defective 

«» Better perh^ to take garva in the sense of £Mntfdi= * dignity*, 
‘ importance,* 


extensive glory was famed throughout the world, who stilled the sufferings 
of the distressed, (a«d) possessed the valour and the majesty of the linn, 
{thus) resembhng {Indra), the king of heaven^® 

(V. 5.) ’1 At the mere {sound of the) name of him fell straightway from 
the wives of his enemies, retreating from the laps of {Ihetr) lovers: with 
{their) hair standing on end and trembling {the following three things .) 
tears, armlets and also {thevr.) mmds, which were deranged by the impetuo- 
sity of their fnght. 

(V 6.) While this king was govemmg the earth, the tame peacocks, 
eager for the advent of clouds, used to break out into cries {of delight), when 
they m the evenings caught sight of the turrets of his palaces, which were 
completely gray with the mass of smoke from the oblations of the twice-bom 
(i.e. Brdkmmas) 

(V. 7 ) His son was Indra-raja, as it were the Mount Meru of the 
noble Rflshlratultas, a prmce whose expansive shoulders were bright though 
bemg scratched through the blows from the tusks of elephants from whose 
spht temples trickled down: ichor, and who had destroyed {hts) enemies on 

(V. 8.,)i Every day people walked about in {his) palace ankle-deep 
through the water {sprhkled durmg) unce^mg recitals of f&ift-texts” by 

(V 9 ) ” The wealth of him, whose pair of feet were worshipped by a 
large crowd of princes who came to do homage {to him), was perpetually 
shared by virtuous men’* . . widi unclouded countenances.’® 

(Vv. 10, 11 ) His queen, who had fulfilled (wU) desires {of others ) ; — 
who was piure like the lustrous moonlight {which), filhng (oii) regbns, 
destroys darkness — who was by her mother’s side descended from the 
Moon’® and on her father’s side from the Chulukyas (Ch&lukyas), called 

All the attributes apply to Indra as well. In his case, however, han- 
I’lkrama-dhama-dkSrl is to be understood in the Sense, ‘one who supports the place 
{covered by) the stndes of Han {ie Vishnu)’, referring to the form of the fertter 
striding over the heavens in throe paces None of the earher mteipretations of this 
verse take into account this ilesha 

This verse is not found m any other Rashtrakuta record. 

These are reated for the averting of evil and the paafication of vanous 


’8 This verse does not occur in any otfaar Rashtrakuta record. 

Tri-jagat-pravhuiih does not convey any sense to my mmd, and appears 
to be a mistake of the scnbe who wrote off tri-jagat in place of sranethmg less 
familiar. R ead p^hapd tn-gaw^pravitfoih and translate : ‘ by those versed in the 
triad of duties (viz. 'dhannle, artha and homo) 

’® 10 without brinit made to fed the siAservience. 

’9 'J'hc Ifa-lriiakuids were themsdves also Soma^vamsin. 


EpigrAphic studies 

the glorious Bhavagana'^ — was the standard of comparisons among virtuous 
and chaste ■women : she freed the world from misery by protecting and mam- 
taming {the needy). 

{;i28} (V. 12). Hei, the best of kings, begat a righteous son from her 
like unto material well-bemg (artha) from prudence (.nitt), (a son who was, 
as it were) the future prosperity (dyati) prayed for by the whole of 

(V. 13 ) (Am) who was known as the illustrious king Dantidurga, the 
sun to the lotus (that was) his family, who illuminated the spaces between 
the r^ons by the flood of his effulgence, the lustre of which waa palpable 

(V. 14.) In the battles with this lion of the martial field the affnghted 
dejdiants (which were hts) enemies, having pulled up by the root the posts 
(namely, their), diame!’* have absconded, no one knows where. 

(V. 15 )' Before the burstmg forth of the ‘ sprouts ’ of his prowess and 
(hts) fierce anger the turreted fortresses of (hts) enemies fall down along with 
their hearts. 

(V. 16.) His devotion to (his) mother was demondrated by (the fact 
of his)i mother’s making (charitabU) donations of land m every 'Village m 
(his kingdom of) four hundred thousand villages. 

(V. 17 ) Haiung m no tune conquered Vallabha,®® who was (even) able 
to inflict cmshmg defeats cm the lord of Kianchl, the king of ICerala, the 
Chola, the Ktndya, Sri-Harsha and Vajrata by the prowess of His arm (or 
arms), he acquired the state of the ‘Supreme King of Kings’ and ‘ Supreme 

(V. 18.), Through the power of his valour he brought under one (royal) 
umbrella this earth frcnn the Setu, where the coast-mountain has tossing 
wa'ves flashing along the hne of its large rocks, 'up to the Snowy Mountain 
(Hinfilaya), where the masses of spotless rocks are stained by the snow,®^ 

” The correspoiidmg verse of the San^gad giant contains a varkias tecUoms 
in the first half-verse. See above, p. 122. 

** This half-verse occurs also in the SamSngad grant. In the editio princeps 
appearing in the JBBBAS this line ■was read as rdtavavemtvdSesha jagatah 
pSitSyatiih] i but the editor of the inscnpriom did not translate the phrase 
nitSvavem Dk. Fleet in his article on the same grant some time later (Ind, 

Ant; VoL XI, p 112), after eiamining the plates, oanected the readmg to nita- 
vad1w(ar #A5)i»=,ett, but followed the example of his predecessor in not trans- 
lating the awkward phrase. The amrect reading is evidently supplied by our plates. 
For the meaning of djtefi, cf, Kk&Srjuiiiya 2, 14 ; rahayiUy Spad-upetam ayatih. 

« I>r. Fleet's correction of sofaj/a of his text to satajjdh is obviously wrong, 
as his translation does not give a good sense. 

«« iA the Chilukyan king IGrtivaiinan II. See Fleet’s Kan. Dyn., p. 391. 
»i See above, p. 122. 

For the idea imphed by the word kedaihkita cf. Ealii&a’s KumSra^ambhava, 
Canto) 1, v. 


as far as the boundary line beautified by stretches of the sandy shores of 
the eastern and western oceans, 

(V 19 ) When that Vallabha-raja®^ had gone to heaven, Krishna-raja, 
the son of the illustrious Kakka-raja who relieved the sufferings of (his) 
subjects,®^ became king 

(V. 20,) The career of that glonous Knshna-raja, dunng which the 
circle of his enemies was completely swept away by the proiwesef of his own 
arm, was as stamless as that of Krishna (Vasudevaj, — 

(V. 21.) 86 ^ho IS famed to be of fierce disposition towards the fierce, 
a mighty repository of generosity towards the poor, most dear to women, and 
towards the prostrate Most-Highly-Gracious C&ibha-timga),— 

(V. 22 ) who, constantly showering wealth on fnends, arrows on enemies, 
love on young damsels, protection on the helpless, was famed in the world 
as the Untinifdy-Showerer®® (Akala-varsha), — 

£129} (V. 2S.) by which glonous Great-Deluge-Boar (Pralaya-Maha- 
vat^ha) was rescued the fnghtened Earth, which was sinking in the Kali 
ocean, which had overpassed (lis) boundary. 

(V. 23.) He, seeing (that) life, which is unsteady like the wmd and the 
hghtmng, is witiiout substance, established this brahmadaya, which is parti- 
cularly meritorious on account of (its bemg) a gift of land 

(L 40 ) And this Parama^hattorcdza Mahdrdj-adhiraja Param-eivara, 
the illustnous Akala-varsha, the Lord of Prospenty and the Earth (&ii-pnthv%'' 
udlldbhd), King of Kmgs (Narendra-^deva) commands all the governors of 
kingdoms (rashtra), governors of distncts (vtshaya), governors of divisions 
(bhoga) and others (as follows) : 

(L. 42 ) Be It known unto you that, — Six hundred and ninety-four years 
of the iSaka era having elapsed, on (the occasion of) an eclipse, on the third 
(day) of the dark half of Ashadha, while encamping®^ at Nandi-pura-dvarl, — 

S3 A biruda of the Western QjSlukya kings, probably adopted by the R^- 
trakiitas as their successors 

Buhler's reading kfita-prajcvadha and translation, 'who did not oppress 
his subjects’ (Jnd. Ant,, Vol XII, pp 182, 187), are both imeatisfactory Why 
should he oppress them? As no impression is appended to hia article, it is not 
possible to decide if the readmg is not a mislection. Kshata-prajSnb^ha corresponds 
exactly to the phrase artt-drtU-harm in a previous verse, and does not m the least 
presuppose that his predecessors had oppressed their subjects [In the Talegaon 
Plates (supra, Vol XIII, p. 279) Dr Konow read krtta-prajd-vddhab* Perhaps 
we should here understand "^prajdbddha as ^^prajd-abadha — F. W T.] 

fis This and the following two verses do not occur m any other Raditrakuta 

To be understood in the sense . raining in season and out of season ” 

87 Samdvdsdike can only refer to a dwelling-place, and the preceding word end- 
ing in pur a evidently supports the idea of enc amp ment Nevertheless, the position 
of this word eixpresauig locahty right m the nuddle of others expressing tune, is a 
htUe cunous. 



epigrAphic studies 

m order to increase the religious merit and the glory of (Our) parents and of 
Ourself the village of Nagana-pun, {situated) at a distance of a to 

the east of Udumvara-manti, has been given by Us at the request of Madaiia 
to the Bhaftaiaka of the temple of Aditya erected in the town of Udumvara- 
manti for the {performance of) bait, charu, naivedya, worship and (tepatrs of) 

(L. 48.; Its boundaiies are noted (as under) 
to the east the village Nagama , 
to the south village Umvara , 
to the west the village Antarai ; 
to the north the village Kapiddha, — 

thus determined by (its) four boundanes, excepting foimer gifts to gods and 
BiahmaDas ; and also the river along the boundary of Umvara-manti to the 
north of the DSva-tadfika (and) to the west of the Ra]i| 0 i-tai(J 6 ka. Thus al- 
together one hundred mvarttanas 

(Vv. 25, 26 ) [|Two of the customary verses.] 

(V. 27.) (This) edict was written at the order of Akala-varsha by the 
illustrious Vamana[na,]ga of benevolent and compassionate nature. 

1 • * 

Note by Rai Bahadur Hiralal, Extra Assistant Commissioner, Jubbul- 
pore, CP* 

“ A gavyuti is equal to 4,000 dajudas or two krosas (Monier Williams) . 

* On the Localities mentioned in the Bhandak Plates of KimaiSia I , Saka 694 
by K. N DiKSHiT M A. [Second Or. Con}. Proe 625-2T.] 

The Bhandak plates of the Rastmkuta king I have been edited in the XIVth 
Volume of the Epigraphia Indka Dr. V. S Sukthankab, -with a note by Rai 
Bahadur Hiralal, now Deputy Commissioner in the Central Provinces. The Bhan- 
dak grant is the first record of the RSistiUkutas of hlanyakheta to be discovered so 
far north-east of their domains But the mere fact of the discovery of a copperplate 
in a particular locality cannot ptwe that the tract m question formed part of the 
dominions of the prince, whose record it is. It is necessary that the locahties men- 
tioneld in the grant should be identified with a degree of certainty, before such a 
conduston is amved at. In the present case, Rai Bahadur Hiraslal has proposed 
^ identi^ the places mentioned with, several places in the Amraoti and Wardha 
tXs. of B&w and the Central Rrovmcesi. The identification is however open to 
^ as r did not feel satisfied with it, I consulted a friend of mine from 
Yeotaal, K. Peshpande who has a good knowledge of the localities con- 

cerned and who happened to come here during Xmas With the help and informa- 
ii<m from him I am now able to identify the locaUties add hope that they 

will be found completely satisfactory. ^ 

pie name of the SamSvSsaka, i.e. place vfiiere the long encamped was NaniE- 
pmAytm. pie modem phoneticai equivalent of Nandipura would be Nandura 
There » a Nant^ in Yeotmal Taluq, which is situated on the river TymHa. a 
tabuto^ of ^ nver Wardha, and has still got a camping ground and a modem Dak 
Bungalow. The acaraty of water wHch must have made itself felt m Berar, then as 



First of all I take Udumvaraimatti to be identical with Umravati. Um- 
ravati means the town of Umar (Ficus Glomerata), the same as the Udum- 
Vara of the Sanskrit. That the pronunciation and spelling continue to be 
Umaravati in the vernacular will be seen from a cutting of a Marathi-Eng- 

now, makes it mcumbent on travellers to choose convement sites on the banks of 
nvers, as their camping grounds We can very well imagme therefore, why Kr^na- 
i&ja tounng in the height of summer (—the grant was issued on the 2aiki June) 
encamped at Nanjdipura, situated on the hank of a perennial river. The place sug- 
gested by Rai Bahadur HiralAL was Nandora m Wardha Tahsil, which has to be 
rejected as it has no such natural advantages. 

We then pass on to the object of the grant, the village of Naganar{[626}-piiri, 
situated at the distance of a gavyuti to the east of Uldumvaramanti, the donee being 
the Bhattaraka or enshnned god at the temple of Aditya erected in the town of 
Udumvaramanti The boundaries of the village granted are given as , the village 
Nagama to the east, the village Umvara to the south, the village Antaim to the west 
an|d the village Kapiddha to the north. As Rai Bahadur Hiralal points out, Um- 
raoti IS the modem eqmvalent of Udumvaramanti, but the modern town of Amraoti, 
besides bemg too far from the localities m question, has no pretension to antiquity, 
as Rai Bahadur claims for it. The old town of Udumvaramanti is the modem vil- 
lage of RSni Umraoti in Yeotmal Taluq, about 5 miles to the south-west of Nandura, 
the place of encampment. The prefix Rami was added to this andent village, some 
three centimes ago, when the village came into the i>assession of the Raa^ Rajputs 
from Udaipur, the present descendants of which family, though converted to Muham- 
madanism are still the Deshmukhs of the village The record besides grantmg the 

village of opToijd- mentions ■ 

^ i*e. a hundred nivartanasf of land within the boundanes 

of Umvaramanti, as follows * to the north of Devatatjaka, and to the west of Raji- 
pitaK^ka and (to the south and east of) the nver.”' There is still a rivulet running 
within the boundanes of Rani Umraoti and there are depressions to the south and 
east at some ciistarice, which may be the silted remains! of the tanks mentioned in 
the grant My fnend expects to locate the site of the ancient temple of the Sun 
at Rani Umraoti 

The village of which was a (or two kroSios or 4 miles) 

distant from Udumvaramanti is to be identified with the village Ganori, four mdes 
to the east of Rani Umraoti I cannot recall to my mind any place name beginning 
with ^ and I presume the imtial na of IjTagaaiapuri was either a mistake of the scnbe 
or a pedantic attempt to Sanskntize the name It will be seen that is a 

correct equivalent of Rai Bahadur Hiralal could not identifyl village 

but he tried to identify some of the boundary villages m a locality, 60 miles to the 
east of Amraoti A SavyuPt can never by any stretch of imagination be supposed to 
cover a distance of 60 miles. His identification of only two villages five which 
be h as tried to justify in spite of the discrepancies as regards the direction, have there- 
fore to be completdy rejected ^ \ 

The present boundanes of Gamri are Antargaon (anaent Antaja-{627}-grai^} 
to the west , Umbarda (anaent Umvaiagi^ama) to the south , Naigam (anaent 
Na^ma) to the east , and BSbhulgaon to the north The ancient village of Kapittha 
named after a woodapple tree, which bounded Ganapuri on the north has apparen y 
disappeared, giving place to a village also named after another tree, the Babich 



lish paper herewith enclosed. Exactly to the east of Amraoti at a distance of 
about 60 miles there is a village named Antaragaon in the Wardha Tahsil of 
the Wardha District, to the west of which and contiguous to it is another vil- 
lage Umaragaon. I take th^ to be identical with Antaraigrama and Umvara- 
giama of the mscription. Ijlagasjapuri, the subject of the grant, is not trace- 
able nor Na^ma and Kapiddhagrama, which bounded it on its east and 
north- The Umvaragrama was to the south of Naganapuri and Antarai- 
grama to Its west. If there is no mistake m interpreting the record, I should 
suppose that the villages have, for some reason or other, changed thar sitCb, 
causing a confusion in the directions of their origmal positions 

The donor’s camp was at Nandipuradvari and I take this to be Nandora, 
9 miles south of Antaragaon. 

The village in the vicinity of Antaragaon and Umaragaon are Jhersi, 
Borkhedi, Chargaon, Pipalasenda, Wargaon, Echora, Kamthi, Hirora and 
Giroli, most of them named after trees, as Kapilddhagiama and Na gamag ra- 
ma appear to be, and apparently the last two have disappeared, giving place 
to names derived from trees which later on abounded in the place whene 
Kapiddha and Mgama were situated. 

£130} Since some four villages give the clue for identification, we may 
take it that l^agaipapuri was situated somewhere at 20® 51' N. and 78“ 44' E. 


The subjoined Prakrit record incised in the reign of Sm-Pulumavi, ‘ King 
oi the Satavahana {.family)! was discovered by Mr T Rajarao, Kanarese 
Assistant in the Oflice of the Assistant Archaeological Superintendent for Epi- 
graphy, Southern Circle, during his tour of inspection in the Adoni TMuk of 
the BeiUary District, Madras Presidency The estampages iwere prepared 
under the direct supervision of Rao Sahib H Krishna Sastri and kindly 
placed at my diqxsal by, him for publication I am indebted to him also 
for many valuable suggestions m the matter both of decipherment and of 
interpretation of the record. 

The inscription is engraved on the eastern face of a large natural bouldei 
of reddish gramte, known to the villagers as Ja'ngli Gundu (Jungle Stone). 
The mscnbed rock, which is firmly buned in the soil, hes midway between 
the villages My^adoni and Chmnakadaburu at a distance of about eight 
miles due N from the Taluk Head-quarters The surface of the boulder has 
peeled off at various places, sometimes right up to the depth to which the 
letters were incised The wntten surface, consisting of four lines of the inscrip- 
tion, cavers an area of 8' by 3', and the hei^t of the average letter) is 2J" 
The engraving, though bold and neatly executed, is not very deep , indeed the 
■' ducts ” of the letters are so shallow that a superficial exanunation of the 
rock discloses hardly any traces of the record It is worth notmg that the 
words are separated from each other by small gaps, a circumstance which 
greatly facihtates the deciphermg of the record The fourth and last line of 
the ihscnption is considerably shorter than the rest, and commences much 
farther to the nght than these The dosest inspection of the rock did not 
disclose any distinct trace of letters m the gap at the beginning!, caused by 
the shortness of this hne 

The alphabet resemblesi that of the Jaggayyapeia mscnption of Puiisa- 
data.^ 'Characteristic are the hooks with which the elongated verticals of the 
letters ka, na and ra terminate, as well as the pedantic semi-arcular arc used 
as the sign for the meldial i The signs for medial a and I show a tendency to 
droop downwards at their free ends. In spite of this similarity with the charac- 
teis of the Jaggayyapeta record, there could be, as far as I can see, no objec- 
tion on palaeographic grounds to their being assigned to an epoch earlia: than 
the third century, to which the Jaggayyapdta inscription is hesitatmgly ascrib- 
ed by BuHLEJt, Indische Palssographie, p 44.® 

* \Ep. Jnd 14. 153-55,] 

^ BtJhikr, Indtsche pkaeagraphie, p. 44, and Tafel III, Col XVII, XVIII. 

* See also his remarks in the Arch, Surv, of Southerri India, Vol. 1, p. 111. 
Excepting the tendency of horizontal lines which are unconnected at one end to 



epigraphic studies 

The number of epigraphic records belonging to the Satavahana Dynasty, 
which had succeeded in holding sway over a large part of southern India for 
an unusually prolonged period, is remarkably small In the Madras Presi- 
dency, besides the one I am now editing, there aie only two inscnptions which 
refer themselves directly to the reign of a Satavahana king,^ and these are 
both records from the Krishna district, one of them being certainly a private 
lecoid So IS the inscription under consideration a private record : it registers 
the construction £154} of a tank by a certain householder {gahapatika) The 
only otlier designation besides Pulumavi which the king receives here is Rano 
Satavahananam, ' of the King of the Satavahanas/ One notices here the absence 
of the metronymic with which the names of the Satavahana kings are as a 
rule accompanied, such as Gotairuputa and others.^ Worthy of note also is 
the use of the family name Satavahana, a term of comparatively rare occur- 
rence in mscnptions ® 

With the scanty information we have in our record about this Pulumavi 
an identification is precanous There are in fact at least four kings with 
the name Pulumavi (or its variants) known to history , and the chronology 
of this djmasty is far from being satisfactorily settled Mr, Vincent A 
Smith,® working upon the Puramc matenal supplied by Pargiter, gives us 
tentativdy the following dates for the vanous Pulumavis — 

1, Pulumavi (I.), the fifteenth king of the dynasty, ruled some time 
before a,d 59 

2 Pulumavi (II ), VasitMputa, came to the throne about a.d 135, 
and ruled for something like 23 years. 

3. Pulumavi (III ), came to the throne about a.d 163, and rul?d for 
something like 7 years 

4. Pulumavi (IV ), came to the throne about Ai). 218, and ruled for 
something like 7 years. 

curve downwards, there is no difference between the alidiabet of our record and 
those which are figured in Table III, Col, X-XIII of Buhler’s Tafelfit which would 
justify its being assigned to a later palaeographic epoch. The curving downwards 
of horizontal lutes is, in my opinion, as much an ornamental vanation as the hooks 
at the ends of elongated verticals, which are to be observed as early as m an ins- 
cription of Satakajji I. (Col. X), whith Biihubr himself assigns to 1-2 century Ai> 
With the semi-circular are representing medial i cf. gt and vi in an mscnption of 
Piijmavi (Cd XI) ; di, ru and hi in an inscription of Satakaoi I (Cd X) ; U 
(twice) and dht in those of Ushavadata (GoL VII, VIII) and others much earlier, 
® See Luoe2«s, List of Brdhmi Inscnptions, Nos. J248, 1340 
^ See Rarson, Catoh&te of the coins of the Andhra Dynasty, etc. (London, 
190S, p. clxxxhc. 

® See Rapson, op, ctu. Index V, s. v. Satavahana. 

® Early History of India, 3rd Edition (19141, pp. 216£f. 



For purposes of identification the Puranic king Pulum&vi (L), of whom iwc 
know next to nothing, may be rejected on palaeographic grounds Further, 
if the lengths of reigns allotted to these kings in the list supplied by Mr 
Vmcent A. Smith happen to be correct, thoi the last two Pulumavis will 
also have to be ijeijected, as they are stated to have ruled only seven yeais 
each, while our inscription is dated m the eighth regnal year of the king. 
Froan this point of view the Pulum&vi of our mscnption will have to be 
identified with Wsithlfputa sami^nn PuilurnSvi (II ), the [Svto^Piolemmos. 
of Ptolemy '' A large number of records dated in the reign of this kmg have 
in recent years come to light. The year of his accession to the throne is, 
as remarked above, put down roughly at a.d 135 Assuming a plurality of 
kings with the name Pulumfivi, there is no other criterion in the inscription 
for identifymg him further. 

It was mentioned above that the object of the mscnption was to record 
the sinking of a reservoir italaha) There is, howelver, no reservoir or tank 
to be seen in the neighbourhood, to which the record may apply But it 
may be remarked that the soil in the vicinity of the inscribed rock is alluvial, 
consisting of sand and finely powdered dust ; so that the adjoining land might 
well at one time have formed the bed of a tank 

Among the locahties mentioned m this record Satavaham-hara is parti- 
cularly interesting, as it occurs once again in the IKra-Hadagalli copper-f^te 
inscnption^ of the PaUava king iSivaskandavarman m the slightly altered 
form of S&tahani-raittha I am not aware that the names of places mention- 
ed in this grant of the PaUava have been satisfactorily identified, so that the 
situation of Sataham-rattha has been, as far as I know, a matter of conjec- 
ture. The inscnbed boulder bearing the present record is, however, a sure 
landmark, as far as the situation of the locality is concerned If, now, the 
find place of the grant, Hira-HaidagaUi, which is also situated in the Bdlary 
District, be supposed to be not far removed from the subject of that grant, 
which is descnb|eid as being located in the Satahani-rattha, then thd terri- 
torial £155} division Satavahani-Sfitahani must have compnsed a good por- 
tion of the modem B^aiy District The rdation in iwhidi the nuthasenapatt 
and the gurmha stand to the janapada and the gama which appear along with 
their names, is not explicitly mentioned But, considering the position of 
these persons, one might hazard the guess that these imbtary office's were 
feudal lords of the lands, holding them in the form of jdgirs. 

In conclusion, it may be remarked that the site of the inscnbed rock is 
an important landmark, flxmg definitely a pomt south of the Knshioa to 
which the sway of the Satavajianas extended. 

^ Rapson, cp. ctt., p. xxxix. 
* Ep. Ind., Vd. I, pp. 2ff. 



L ^'*[Si]dha[m] [||'^] Rano Satavahananam S[i]n-Pulum['a]visa sava 8 
hema [2] diva 1 

2. [masaj^i inahasMpat[i]sa Khamda[na,]kasa janapade^^ S[ia]1:avahani- 

3 mikasa^'^ Kumaiadatasa game Vepurafce vathavena gahapatikena 
[Kom]tanam^^ [Sambelna 

talakam khamtam [ 1 1 “* ] 


Success ! On the first day of the first (fortnight of) second Winter-’ in 
the eighth year (of the reign) of Sin-Pulutmvi, King of the Satavahana 
ifannly), the reservoir was sunk by the householder (gahapattka) , . resi- 
dent m the village (of) Vepuraka, belonging to the Captain (gumika) Kuitia- 
ladata (Kumaradatta), m the country (janapada) of S&tavahani-hara,^^ 
belonging to the Great General (tnahdsmapati) Khamdanaka”’ (Skandanaga) . 

® FrcMu the stone and a set of impressions 

Traces of the bracketed syllable are visible on the stone. 

The cons(Hiant signs are almost ceitam ; the vowel signs are all but obli- 
terated, as at this point the rock has peeled off almost to the depth to which the 
letters were incised. Perhaps, we have to read mobi, making with the foregoing nume- 
ncal symbol 1 the word padhamasu 

12 Read jcmcpadL 

la We have probably to restore gumtkasa (fiom Sk^. gaulmika^ ‘captam’), 
which would accord well with the mah^endpatha of the preceding line. 

1* This and the following word must, in my opmion, contain the speafication 
of the gahapattka; the first (geit, plu,) is most probably a tnbal name, and the 
second (mst. sing.) is the personal name. The readmg of the first aksharas of the 
names must however* be looked upon as problematic. 

This is the seasion commenang with the dark fortnight of the month of 

1® This is a dear case of the use of the word hdia in the sense of ‘kingdom 
or district'. See LibEaRS, List of Brdhmt Inscriptions, Appendix, Index of {miscel- 
laneoug terms s.v. a/wrtf~In the Hira-Hada^li copper-plates (Ludess* List, 
Na 1200) this territorial division receives the designation taftka (rdshpa). Thus 
hma must oorre^nd to raff An. 

Cf. the personal name Khaxhdanaga-sataka occurring in a Buddhist inscnp- 
tion at Kaflh^ (LtDBRS‘ List, No. 1021). 


These three Kshatrapa inscriptions, which are now exhibited in the 
Watson Museum of Antiquities at Rajkot, have been published before, at 
different times and different places, but are heie re-edited in order to have 
them properly illustrated and render them easily accessible. A comparison 
of the onginals with the facsimiles of the same inscr^tiona pubhshed in the 
Bhavnagar Collection of Praknt and Sanskrit Inscriptions made us fed the 
special need of placing before scholars reliable facsimiles obtained by purdy 
mechamcal means These, it is hoped, will enable even those scholars who 
are not in a position to acamine the stones personally to reconsider the previ- 
ous readmgs, which, m our opmion, are in many respects defective. Our 
transcnpts, which were in the first instance prepared from mk-impressions 
and squeezes, were subsequently compared with the originals. 



The inscription wag first edited, with a translation, in 1881, by Georg 
Buhler in Ind , Ant Vol. X, pp, 157 f , from an eye-copy and a transcript 
prepared by Pandit Vallabhacbarya Haridatta of Khthiavad and submitted 
to Buhler by Major Watson for pubhcation Nine years later Buhler 
publidied some corrections in Sitzungsber, Wien Akai, Wiss , Phil Hist. Kl , 
Vol CXXII, No. XI, p 46, note 2, which publication was unfortunatdy 
not accessible to the writers of this article. The posthumous papers of Bhag- 
vanlal Indraji edited by Rapson in the Jour Roy As. Soc (1890) contain 
a short note (pp 650 f.) on this inscription In 1895, the text and a transla- 
tion of this epigraph were repubhshed in the Collection of Praknt and 
Sanskrit Inscriptions, Bhavnagar, pp 21 f., No 3 and Plate XVII In 1896 
appeared in the Bombay Gazetteer, Vol I, Part I, p. 42, some corrections 
proposed by Bhagvanlal Imdriaji himself in his earlier readings and mter- 
pretation , Rapson, in Jour. Roy, As Soc, 1899, p 375, also publidied some 
fresh corrections. The Caidogue of the Coins of the Andhra Dynasty, etc. 
(1908), of Rapson indudes (p Ixi a short note on this record, which gives 
reference to the literature on the subject and briefly summarizes the contents 
of the inscnption In 1912 Prof, Luibrs m his Ust of Brahtm Inscriptions 
(Appendix to Epigraphia Indica, Vol X, No. 963) gave a complete biblio- 
graphy of the mscription, a readmg of the date (it cannot be said whether 
from the published! fac Rimilp or directly from an impression of the sbme), 
and a summary of its contents And finally, m 1915, Prof D. R Bhandar- 

* [By Rakhaldas Banerji and Vishnu S. Sukthankar— Ep Ind, 16533-41.1 


KAR published some corrections of previous readings and interpretations in 
Prog. Rep Arch. Surv of India, W. Circle, 1914-15, p 67 

The inscription was discovered in 1880 by Major Watson in an old 
unused well at Gunda m the Halar Disitnct of Nqrth K&thiavad It was 
subsequently removed to the temple of DvlarakSniatha at Jamnagar, where, 
apparently, it was kept until its transference to the Watson Museum 
of Antiquities at Rajkot. 

{234} The epigraph contains five lines of wdl-engraved wnting, cover- 
ing a space of about 2 ft 2 m in width by about 9^ in m height The 
wnting is, on the whole, in an excellent state of preservation , some isolated 
syllables here and there are, however, seriously damaged The average size 
of such letters as n, m, p, and b is about J". 

,The characters present an earher form of the southern variety of the 
Gupta alphabet than that seen m the well-known inscription of Skandagupta 
at Junagadh. It differs in a few minor particulars from the Junagadh edict 
of the Maha-Kshatrapa Rudradaman , to wit, in the form, of y ( subscript 
as well as uncombined ), and in the markmg of the medial vowel m « (I. 3), 
mi and ti (1 5) Subscript consonants, excluding y, are expressed by the 
ordinary full forms of the letters No final consonants occur Of 
imtial vowels the record has only a (I. 4) Medial a has in various 
mstances been left unmarked, evidently through the cardessness of 
the scnbe ; when; engraved — ^it is (like e, and o) denoted by a short horizontal 
Ime appended, generally, to the top of the consonant sign ; as an exception 
we may mention ;(w)e m which the sign of o (whidi is made up of tlie 
signs of a and e) le drawn m continuation of the middle bar of the letter. 
Noteworthy is the form of the medial long i, m the only certain and clear 
instance of that sign in this inscription, in ahasya (1. 3). In mscnptions 
of the same peiiod and locality the long i is generally represented by a 
crescent-shaped arc, with unequal arms and open at the top. In this 
instance, how'ever, the free end of the shorter arm is attached to, or rather 
drawn in contmuation of, one of the upright verticals of the matrikS, a 
peculiarity which gives this letter a somewhat uncommon appearance. This 
mode of drawing i is probably the origin of the ^iral sign of that vowd in 
the southern alphabets of a later epoch. The medial u is marked either by 
a subscript curved line open to the right, as in Sm of -suddhe (1. 3), or by 
one c^ien to the left, as in of -putrasya 1(1. 2), or lastly by a short hori- 
mntal stroke attached to the lower end of a Icmg vertical as in rw of Rtidra® 
(1, 2), Line 3 indudes the numerical symbols for 100 and 3 No sign of 
punctuation occurs . the letters are engraved in a continuous succession 
without a break. 

The language of the inscription is a mixed dialect, and the whole is in 
prose. The ftakritisms are tTiy-uttara- (1. 3), and bamddhSpita° (1. 5), 


and besides, perhaps, such irregulanties of spelling as cannot be put down 
to the negligence of the scnbe ; the rest is in Sanskrit In passing it tnay 
be observed that the Sandhi consonant y which we find here inserted between 
fri and uttara serves very often the same purpose in Prakrit as may be seen 
by leference to Pischel’s Grammattk der ^Prakrit-Sprachen, § 353 [The 
construction of the gemtives is in some cases m all these inscriptions irregu- 
lar, eg maha-kshatra[pas]ya, 1. I of Inscnption No. I. — Ed.] — ^As regards 
orthography, we may notice the sporadic doublmg of the consonant after r 
in muhurtte (1 4), sarwa- (1 5), in sukhdrtham (1 5) the consonant is 
not doubled There is, moreover, no instance of the phonetic doubling m a 
ligature when r forms the second member of the conjunct. The word 
bamddhdpita seems to offer an instance of the doubling of the con sonant 
following! upon an musvafa, but the reading of the ligature is not quite 
certain, and perhaps we have to read the word as bamdhdpita, in which case 
this would be an illustration of the addition of a superfluous anusvdra before a 
nasal, of which there are instances to be met with m mscnptions of all periods. 

The inscription refers itself to the reign of the kmg (and) Kshatrapa 
Lord Rudrasiha ( Rudrasimha ) , and gives the following pedigree of 
the king -—king and Mabi-Kshatrapa Lord Chashtana , his son king 
and Kshatrapa Lord Jayadiaman ; his son king and Maha-Kshatrapa 
Lord Rudradaman , his son king and Kshatrapa Lord Rudrasiha 
(Rudrasimha ). This is a genealogical hst and not a dynastic one , that is 
why the names of several pnnces who had ruled between ChSditana and 
Rudirasimha, but who were not m the direct Ime of descent, have been left 
out The record is dated on the fiftla tithi of the bright fortnight of Vaiaa- 
kha, during the coostdlaticHi of Rohi|ni m the year one hundred and three, 
which number is {235} expressed both in words and numerical ideograms 
There can be bttle doubt that the era to which the year in this inscription 
IS to be referred is the iSaka era Accordingly the inscription may be taken 
to be dated roughly m the year A D 181. It will be remembered that the 
evidence afforded by the dates and the l^ends on the coins of Rudrasimha 
lead us to infer that he ruled! first as Kshatrapa ui the yjear 102-3, then as 
Maha-Kshatrapa from 103 to 110, then again as Kshatrapa from 110 to 
112, and lastly as Mahia-Kshatrapa from 113 to 118 (or 119). According 
to this scheme the present inscription must be taken to refer to the period 
when he was reigning as Kshatrapa for the first time The earliest date we 
have for his reign is the year 102 on a coin belonging to the Cunningham 

The object of the inscnption was to record the digging and ccaistructing, 
at the village of Rasopadra, of a wdl by the general (smapati) Rudra- 
bhliti, son of the general (.smapatt) Bapaka, the Abhira. 

The villeige of Rasopadra, which is the only locality mentioned in this 
record, remains unddentified 


epigraphic studies 

1 Siddha[ihl [||*,] Rajno maha-kshatia[pas]ya svanu-Chashtana- 
prapautraya rajSo kshatrapasya svaini-Jayadama-pautrasya 

2 (sya) ra][fi6(I maha-kshatrapasya sv[ia]aii-Rudradatn»-pvitrasya 
iSjfio kshatrapasya svlairu-Rudra- 

3. Sihasya [vajishe [tn]y-uttaralate 100 3 Vaisakha-^uddhe pamchani 
[i]-dha [tltya-tithau R5[hi]oi-naksha- 

4 tia-niuhurtt( [^] AbhirSpa senapati-Bapakasya putrena seniapati- 
Rudrabh[u,]tna gieme Raso- 

5 [pa,]driye va[pll [kha]iu[t6] [bamddhlapitasi=cha sarvva-satvanam 
hitaf-siikharthanj=iti [ 1 1 "iJ 

Remarks on the Transcript* 

L. 1 GB and BI rajHo vmha-< and sv'amt ; but m our estampage the sign 
of length can be made out in none of these words. L 2 Over ma in maha, 
to its nght, IS to be noticed a Wanting irregular depression, the nature and 
fli g nTfi rance of which IS uncertain L 3. GB day-uttara4ate sa 100 2, which 
18 dearly inadmissible ; BI and L tn-uttara-sate, diffenng from our readmg 
in the second syllable, which is, however, unimstakably yu and not m ; on the 
other hand, it is uncertain whether the first syllable should be read as in or 
tra. GB, BI and L 'Buddha for suddM ; but our estampage shows the sign 
of e quite distinctly. The estampage does not diow any clear trace of the 
sign of the long i in pamcham as read by GB, BI and L. The projection on 
the left of the sign of cha is abnormal. GB, BI and L ■4hmya- ; but an 
examination of the back of the estampage removes all doubt as to the coaect- 
ness of our reading of the second syllable Most probably we have to correct 
dhattya to dhmya', die former gives no sense Mr BANERjf would read 
2Uya regarding the latter as equivalent to asyam or etasydm, and cognate with 
the Pkt. etiya found m Kushan inscriptions GB Sravana^ for Rohtm-. 
L 5. GB padre hradartthe, and BI padre hradafj , ; L accepts the sense, adding 
hrada in brackets with a query DRB speaks of Rasopadnya and garia in 
giving the contents of the inscription The syllable vd is quite clear in the 
estampage, especially on the back of it ; dd or do, which {236} are made quite 
differently, are out of the question ; cf dd m -fayadama- in 1. 1, and -Rudra- 
dSma- in 1 2 The estampage will also show that the reading hi a for the 
first doubtful syllable is utterly impossible. The anusvara in batn'‘ is well 

I From a set of estampages 

■* Explanation of the abbreviations : — GB = George Buhler, Ind., Ant , 
Vol. X, p 157 , BI = Collection of Prakrit and Sanskrit Irtscriptmts, Bhavnagai, 
pp. 21 f. ; L = Luders, Lust of Brahtni Inscriptions, No. 963 ; DRBI = D R Bhan- 
DAHKAR, Prog. Rep, Arch. Surv, of W. Ctrde, 1914-15, p. 67 


marked ; but it is impossible to say with certainty whether we have to read 
’‘mtidha” or '^mddhd^ , the latter se«ns to us more probable. 


Hail ! On the [auspiaous®,] fifth tithi of the bri^t fortnight of Vai^- 
kha during the auspicious period of the constdlation of Rohiin, in the year 
one hundred and three!* — 100 3 — (during the reign) of the king, the Kslia,- 
trapa Lord Rudraalha (Rudrasiithha), the son of the king, the Maha-Ksha- 
trapa Lord RudradSman (and) son’s son of the king, the Kshatrapa Lord 
Jayadaman, (and) grandson’s son of the Idng, the MahlanKshatrapa Lord 
Chashtana, the well was caused* to be dug and embanked by the general 
(iSienSpatt) Rudrabhuti, the son of the general ismapatt) Blapaka,'* the 
AbhSra,® at the village (grama) of Rasopadia, for the wdfare and comfort 
of all living beings. 

II. — GApHA (Jasdan) Inscription of the time of the Maha-Kshatrapa 

The inscription was first edited, with a translation and lithograph, pre- 
pared prciiably from an eye-copy, in 1868, by Dr Bhau Daji m Jour So. 
Br Roy, As Soc., Vol VIII, pp 234 f., and Plate. After that it remained 
unnoticed till 1883, when Hoernle pubhshed a revised transcnpt and trans- 
lation of It m Ind, Vol. XII, pp. 32 f. The posthumous papers of 
Bhagvanlal Indrajh, edited by Rapson in Jour. Roy. As Soc , 1890, p. 652, 
contain a short note on it In 1885 the tejct and a translation, based upon the 
editio pfinceps of Dr Bhau Daji, were republished in the Collectton of Prakrit 
and Sanskrit Inscriptions, Bhavnagar, pp. 22 f.. No. 4, and Plate XVHI. 
•The Bombay Gazetteer, Vd, I, Part I, p. 43, contains a very short note on 
it, originating from the pen of Bhagvanlal Indraji. RApson’s Catalogue of 
the Coins of the Andhra Dynasty, etc, (p Iidi, No. 42), includes a short 
summary of its contents, and a reference to the literature of the subject 
Prof. LiiDBRS m his List of Brahmi Inscriptions (Appendix to Eptgraphia 
Indica, Vol X), No 967, gives a complete bibliograi^iy of the inscription, 

* The rendenug ‘ aoapidous ' piresuKioBes that we have to correct dhattya to 
dhanya ; see the remarks on the transcript above 

* The form tri-y-uttara is a Ptiaknlism , the y is a sandhi consonant ins^ed 
in order to avoid the hiatus ; cf. Pkt. duyahena (dvyahSna)', tiydhena (iryahma) 
in Pzs(3lEL'5 Prakrit Grammatik, § 353. 

* Bapaka is a variant of Bappa(ka), which name occurs in a number of 
later inscnptions. 

® It is worth noting that the Abluras were emjdoyed as generals und^^the 
rggung of the Kshatr^m dynasty Among the inscriptions in the FSiaJu l-enn at 
NSsik we have an insciiptioa referring itsdf to the reign of the Abhira kmg Kvara- 
sSna, wbidi shows that sane of these generals bad eventually succeeded in replacing 
the sword of th$ commander by the sceptre of the sovereign. 

epigraphic sttoies 


a reading of the date (it cannot be said whether from the published facsimiles 
01 directly from an impression of the stone), and a summary of its contents 
Prof D R Bhandahkar refers to the inscnption Prog. Rep Arch Surv. 
0 } India, W. Circle, 1914-15, pp 67-68, and suggests certain corrections. 

The inscription is said to have been found at Gaidh&, about two miles 
north of Jasdan, Kathiavad, engraved on a thick inegular dab standmg 
upright on the margin of a lake Subsequently the inscribed stone was trans- 
ferred to the Watson Museum of Antiquities, Rajkot, where it is now exhi- 

{237} The inscnption contams six irregular lines of varying length and 
uncouth writing, covering a rhomboidal space of about 3 ft 7 in. in width 
by about 1 ft 10 m in height. The engraving, which is shallow, is on the 
whole in a fair state of preservation. The average size of such letters as 
n, m, p, and b is about li". 

The characters of the inscnption are of the same general type as those 
of other Kshatrapa inscnptions, and the above remarks on the palreography 
of the Gunda. mscnption are, with a few exceptions, applicable to this one 
also The letters! of this epigraph lack, however, all r^ulanty and finish ; 
they have a decidedly cursive character Observe, for instance, the form of 
the uncombmed tn, which is sometimea denoted by the older X-diaped form 
of the earlier insciiptions and sometimes by a more advanced form which 
is met with, n^larly, only in the records of the Gupta dynasty. The 
letter t appears to have been drawn with pron^ of unequal length, both 
curved, one of them bemgj dightly kmger than the other. In tra the subscript 
r is marked by drawing the free end of the n^t prong to a little distance 
to the left ; thus this ligature and the uncombined t are indistinguishable 
from each other when either of them is badly drawn. Of initial vowels the 
inscription contains f (1 5) and « (1 6) Subscript consonantst exdudmg r 
and y, are expressed by the ordinary full signs of the letters, as in the 
Gunda, inscriptiooi described above. No final consonants ai^iear m this 
record. Sporadically one notices the flattenmg out of the serif of the letters 
into a thin horizontal toP so that in some cases it becomes ertremely 
difficult to say whether the top-stroke is only a serif or the sign of a, S or b, 
which is marked by a slight prolongation of the s^f The length in si (1. 5) 
is denoted by a distinct giiral, which is a further devdopment of the form 
whidh. was met with in the GundS! inscription. To judge by the instance of 
bhatrabhih (for bhraijibhih) in line 6, no distinction was made in writing 
between the subscript r and the medial ji The diphthong au is marked 
by the addition of an upward stroke, slanting to the nght, to the sign of d. 
On two occasions the engraver has omitted the syllable tra m writmg 
KshatTapasiy)a. Lastly it may be observed that the first hne of the 
inscription contains four numerical symbols, 100, 20, 5 and another about 
the reading of whidi there is some doubt ; it may be either 6 or 7, 



The language of the inscription is a mixed dialect In this specimen 
the Prakntisms are sli^tly more numerous than m the Gunda inscription 
One may notice the frequent use of sa for the termmation of the genitive 
singular in II 1 and 2 in addition to the verbal form utthavitia [?/j] in 1. 6. 
The sense of the record, as it stands, is incomplete, and, to judge by the two 
final syllables svairga] in 1. 6, it should seem that a few syllables or words, 
in an additional line which is now lost, are missmg [Regarding the irregular 
gemtives (e.g mahakshat \r]apastt, 1 1 ) see remarksi on Inscription No. I. 
— ^Ed ] — regards orthography the only point worth drawing attention to 
IS the sporadic doubling of the consonant before r in pap(m-{t]tTasya in 1 2, 
but it should be added that the reading of the ligature is not absolutely 
certain. Of words not found m dictionaries the inscription contains one, 
VIZ saitta, of which, moreover, the meamng is not known. Prof. Luders 
hesitatingly identifies it with the Prakrit word Sata in an inscription from 
the Katihen caves f Luders’ iisf, No 985), for which he, also doubtfully, 
suggests the meaning ‘ seat.’ 

The inscription refers itself to the reign of the kmg (and) Maha- 
Kshatrapa Lord Rudrasena and records the erection of a satra (meaning ?) 
by the brothers of Khara[r]-pattha, the son of PralH&ika of the Manasa 
gotra Previous editors of the mscnption have read m 1 6 Ptanaithafca 
mstead of PratSj§aka and Khara-pautra instead of Kharalfjpattha Our 
reasons for adpptmg the reading which we have will be found in the remarks 
on the transcript, below, p. 238 The inscnption gives the following 
pedigree . king and Maha-Kshatrapa bh(tdr<timukha Lord Ch&stana : his son, 
king (and Kshaitrapa Lord Jayadaman , his son, king and ^238} Maha- 
Kdiatrapa bhadra-mukha Lord Rudrad&man ; his son, kmg and Maha- 
Kdiatrapa bkadromukha Lord Rudrasiha (Rudrasimha) , his son, king and 
Maha-K^trapa Lord Rudrasena This is the longest' pedigree of the 
Kshatrapas of Surashtra and Malava contamdd m a smgle record It will 
be noticed that the title bhadra-mukha, ‘of gracious appearance,’ is added 
before the names of some of the Maha-K^atrapas, but not before the name 
of the only Kshatrapa mentioned in the record, or before that the last Maha- 
Kshatrapa named here, viz Rudrasena, in whose reign the inscription was 
engraved The reason for the omiasion in the last case is not apparent ; it 
would seem, however, that the title was used with the names of Maha- 
Kshatrapa only. The names of DSmaysada I. and JivadSman, who had 
reigned before Rudrasena, but who were not in the diiect line of descent, are 
not mduded in this hst, which is purely genealogical. 

The inscription is dated in the year 127 (or 126) on the fifth tithi of 
the dark half ofl the month of Bhadrapada The era to which the date is 
to be referred is undoubtedly the Saka era ; accordingly the date of the 
record may be takoi to correspond to 127 (or 126) + 78 = ad. 205 (or 

224 epigrAphIc studies 

The record cwitains no geographical name. 


1 fVaJrshe 100 20 [7] IBhajdrapada bahulasa 5 fl*] R[a]jfi6 

mahakshat [ rj apasa 

2 bhadra-mukhasa svam(a] Chashtana-putxa*papau[t]trasya rajno 


3 svami-jayadtajma-putra-pautrasya rajno maha-Kshatrapasya 


4 [ava]ma-Rud[rIadmna-pau[tra]sya lajno ma[ha']-Ksha[tra-*]pa8ya 

bhadra-mukhasya svS.[m]i- 

5 RudrasSha [-putra*jsya rajno maha-Kshatrapasya svramr-Rudra- 

senasya d*] idam Satram 

6 Alanasarsa-got[r]asya Pra[tg,]isaka-putrasya Khara[rjpatthasya 

bhatrabhih utthavitafm] sva[rga] 


Remarks on the Transcript « 

L 1 The reading 7 is unceartain ; :t may be 6 DRB reads 6 L. 2 
D and H -mukhasya svand-. The slanting lines bdow the sa of the first 
word IS an abrasion and not the subscript y L' 3. D and H Jayadama 
bhadra-mukhasya is continued in a slanting direction above the level of the 
same line L , 4. No trace remains of the i m svdmt, if it was marked at all. 
L. 5. D and H mahd-. DRB Sakri (for satfaah), which is vay doubtful. 
L. 6. Hoernle’s reading -mana\.ih]tu T'ti^g5ttas[^]a ig out of the question, 
and need not be discussed here. D pranathaka- (the previous syllable is read 
by him as Su->), and H Pratd[r<^thaka (for PrcdMaka)\ both of which are 
inadmissible. The second syllable may, perhaps, be na ; but the third one 
cannot be iha, as) tha doeh not contain the vertical bar in the centre which 
our letter ^ows ; the shallow stroke at the lower end of the letter is an acci- 
dental mark, of whidi the rock has many. D and H Khara-pautrasya, but 
the fourth syllable is clearly ttha and not tru ; cf. the same hgature in a 
subs^uent word of the same line DRB Kharapitthasya. D and H bhrat- 
fibhih (for bhssrabhth)- It is doubtful if the medal ri would be marked 
£239J differently from the subscript r by the writer of this mscnptian. DRB 
bhSttrabhth D utthavitdsva and H utthai>itast[t\. The top of the fourth 
syllable is no doubt somewhat thick ; nevertheless the sign of the length 

^ From a set of estampages. 

® Explanation of abbreviations D = Bhau Daji, Jour Bo Br Roy. A<f 
Soc., VoL Vlir, pp. 234f , H = Hoerni«, Ind. Ant . Vol. XII, pp. 32 f, ; DRB 
-•D. R :KiandarkAr, Prog, Rep. Arch. Surv. of India. W. Cncle, 1914-15 
pp 67-a 


cannot be looked upon as having been marked. A part of our bracketed 
lrgai\ is lost in the crack and not distinguishable on the facsimile DRB ends 
line 6 with utthavita swa-, and then gives an additional (seventh) line 
Irggasukhartka] , which we were not able to trace wi the stone. 


On the fifth (tithi) of the dark fortnight of Bhadrapada in the yeai 100, 
20 [7J, (during the reign) of the king, the Maha-K^atrapa Lord Rudra- 
sena, [son*] of the king, the Maha-Kshatrapa Lord Rudrasiha (Rudrasimha) 
of auspicious appearance [bhadra-mukhii') ; (and) son’s son of the king, the 
Maha>Kshatrapa Lord Rudradaman of auspicious appearance (bhadro' 
mukha) , (and) grandson of the son of the king, the Kshatrapa Lord Jaya- 
daman , (and) great-grandscxni of the son of the king, the MahS-Kshatrapa 
Lord Chashtana of auspiaous appearance {bhadra-mukha) ; — ^this satrO'’* was 
erected by the brotherg of Khara[r,]pattha, the son of Pratasaka of the Manasa 
goira heaven 

III. — JuNAGADH Inscription of the time of the grandson of the 

Kshatrapa Jayadaman. 

This inscription was first edited, with a translation and a photograph, 
in 1876, by BiiHLE® m Arch. Surv. West. Ind , Vol. Il, pp. 140 f , and Plate 
XX ; the block is rather small and almost useless for purposes of study. In 
1395 Buhler’s text was republished, with a few minor alterations accom- 
pamed by a facsimile of an inked impression, and a tran^ation of the text in 
ilie Collection of Prakrit and Sanskrit Inscriptions, Bhavanagar, p. 17, No. 1, 
and Plate XV. Rapson gives an abridged bibliography of the inset iption, 
and summarizes briefly its contents in his Catalogue of the Coins of the 
Andhra Dynasty, etc., p. Ixi, No 40 The most recent notice is by Prof. 
Loimers in his List of Brahmi Inscriptions (Appendix to Epigraphia Indica, 

® Bkadra-mukha literally means ' lucky-faced,' but is here used speafically 
as the title of some of the Maha-Kshatrapas. 

This word has not been met \rith dsewhere and its meanmg is uncertain 
Bhau Daji renders it with ' tank ’ without asaigmng any reason for doing so , the 
dictionaries do not support this meaning. Hoernle suggests that it is a Praknt 
form of satra, which denotes ' a kind of expensive Soma sacrifice extendmg over 
many days ’ , to Satra of our teirt he assigns accordingly the derivative meanmg of 
‘ libonlity, munifkenoe,’ which does not convince us It was remarked above that 
Prof. Luders refers in this connection to the word sata (’ seat) occurring in a 
Buddhist Cave inscription. Mr R. D Banes J i lodes upon the word as a Prakrit 
form of satra and would translate it as ' almshouse,' which meaning that word has 
111 most of thei dialects of North India. Mr. D. R Bhandarkar reads the word 
as Sakri and, oonnectmg it with the following -manasa-, regards Saknmanasa as the 
gofrff-name, an explanation which does not commend itself to us. It may be noted 
that utthavita clearly imphes that we have here to deal with a sfructure that was 
raised, elevated, erected, and not dug or sunk. 



Vol. X (1912), No 966), where we find a complete bibliography of the inscrip- 
tion, a reading of the date (probably from the facsimile in the Bhavnagar 
Collection of Prakrit and Sanskrit Inscriptions) and a summary of its 

The mscription was discovered, during excavation, in front of one of the 
cells of an extensive complex of caves situated to the east of JunS.gai^h, close 
to a modem monastery known as Bava Py&iS’s Math. Regarding the mis- 
chances that fell to its lot after its discovery we have the following account 
by Burgess While extricating it, he wntes, “the workmen damaged one 
end of it, but, to add to the misfortune, some one earned it off to the palace 
m the city, and m doing so seriously injured it at one comer When I went 
to photc^raph it, I had a difficulty f240j in tracmg it ; at length, however. 
It was found lying in a verandah in the circle in front of the palace.”’-^ For 
some time previous to its transferwice to the Museum the stone used to be 
kept in the State Pnnting Press at Jutiagadh. The misfortunes which have 
fallen to the lot of this stone since its recovery did not end with those des- 
enbed by Burgess. As a result of some fresh acadent, it is now in two 
lialves, probably having split along the fissure which is noticeable in the 
facsimile publishing in the Bhavnagar Collection of Semskrit and Prakrit 
Inscriptions, and referred to in thd letterpress accompanying the facsimile. 

The inscription is engraved on one of the faces of a dressed slab of soft 
calcareous stone about 2 feet each way and 8 inches thick. TTie epigraph con- 
tains four lines of writing, covering a space of about 1 ft. 9 in. in width by 
about 6 m. in hdgjit. The average size of such letters as n, m, p, and b is 
about 5". Much of the writing is seriously damaged. The two middle lines 
are in a fair state of preservation ; but the greater part of line 1 and a good 
bit of line 4 are illegible. Moreover the inscription is fragmentary. The 
slab has lost a large fraction of its length : how much it is not possible to say. 
BUhixr assumes that lines 2-4, at their left ends, are almost intact, only a 
couple of syllables being necessary m each to complete them. This H how- 
ever, far from being certain. As far as we can judge, there is nothing to 
show how much is missing on either side of the preserved portion. We can 
only say that the lost portion of 11. 2 and 3 must have contained, at least, 
the names of the son and grandson of Jayadaman as well as the yeai* in 
which the record was dated, expressed possiMy both in words and numerical 

The characters dosdy resemble those of the Gunda inscrlprion of the 
time of the Kshatrapa Rudrasiihha, which have already been minutely des- 
cribed above. It will, therefore, suffice to draw attention here only to a few 
outstanding features of the alphabet of this inscription. The syllable me in 
1. 3, it will be noticed, shovra that the sign of e in mm was atfa c h p d to the con- 

«■ Arch. Surv. West. ltid„ VoJ. 11, p. 140. 



stricted part of the letter The same hne offers a specimen of the numerical 
figure 5. The sign of the medial m in sm (1. 1) is seen to open towards the 
left; in Su (1. 3), on the other hand, it opens towards the nght, of the 
medial u marked by a short horizontal stroke appended to the long vertical 
of a letter tins insciiptirai contains no speamen. We have here only one 
initial vowel, namdy % (1. 3); it is denoted by three dots, of Iwhich two aie 
placed in a vertical line on the left side of the remaming one. In § the middle 
bar, which is attached only at one end, is jdmost vertical. The letter y shows 
the simple bipartite form [In regard to the language we may note the irrc 
gular genitives (eg. kshia[trapa]sya, 1. 2 as in Inscnptions Nos. I and II. — 
Ed.i ] — As regards orthography the only pomt worth noting is that the inscrip- 
tion offers no instance of the phonetic doubling of consonants. 

The inscription must belong to the reign of a Kshatrapa or Maha-Ksha 
trapa who was the grandson (or rather son’s son) of the king, K^iatrapa 
Lord Jayadaman, and great-grandson of Ch&ditana ; the name of the ruling 
prince is lost with the portion of the record which is missing. This Satrap 
to whose reign the record referred itself was therefore either DBmaysada I 
or Rudrasimha I (the brother and successor of the former). The puiport of 
this fragmentary inscription cannot be determined, as the portion containing 
the object of the record is lost It may be added that from the occurrence 
of the eigwession kevdi-jnSna-samiprapta] (‘who had amved at the know- 
ledge of the kevaltns’) in 1 4 it may be surmised that the inscnption pro- 
bably had somethmg to do with the Jainasi, since the word kevdin occurs 
most frequently in Jama literature 

The mscription is dated on the fifth (5th) day of the light half of 
Chaitra in a year which, like the purport of the record, cannot be ascertained, 
as It is lost in a lacuna of the tert 

f24lj The only locality which the record mentions is ths well-kflown 
Giri-nagara, which was the anaent name of Junagadh, and which survives 
in that of the adjacent hill of Gim& 


1 . . . . . . s = tatha sura-gaiij[a] . [k^tra]i»aih 


2 Chashfanasya prat[pau]trasya iSjfiah ksha- 

[trapa]sya-svami-JayadSma-p[au]trasya lajfio ma[ha,] . . . . 

4 • » « 

3 i[Chai]tra-§uklasya^ divase paihchame 5 i[ha] 

Girinagare dev&sura-n&ga-ya[ksha]-ra[ksha]8-e 

(From a set of estampages. 

epigraphic studies 


4 . . , .... -thap[u]ram = iva .... kevali-[jna,] 

na-saih . . narh . -jara-inaran[a] 

Rem-arks on the Transcript. 

L. 1 GB reads m the first line . . . ktri . raga . . kshatiapo 

... , BI statha suraga . . kshatrapa L. 2. GB adds 

svSmi in square brackets before Chashtanasya The bracketed syllable m 
•■p\ou\tiasya has broken away and become illegible GB potrasya; BI 
pautrasya The medial vowel of the first syllable is quite uncertain. L. 3. 
GB, BI, and L read pckskasya after °sukla The mistake had its origin in 
Buhler’s faulty transcript GB, BI, and L lead panchame for pamchame 
We do not see the ncha , the sign below cha, we believe, is only an abrasion ; 
in any case ncha is by) no means cartain GB and BI ’‘rakshasendri . . . 

. . L. 4 GB. praka('>)tniva pa . ... kevaU-jUdna' 

saihprdptmani jita-jara’inarmanm (’) 


Also . . the divine hosts 

. . . . the first among warriors (kshatra) 

On the fifth (5th) day of the light half of Chaitra in the year .... 

(during the reign of) king Ma[h&-Kshatrapa] . . . 

. , son’s son of the king K^atrapa Lord Jayad&man, the great-grandson of 

.... Cha^tana. Here in Gin nagara the 

gods, asum, nagas, yak^as, and rakshasas .... dty (?).,. 

. . who had arrived at the knowledge of the kivalins old 

age and death 


The copper-plates bearmg the subjoined inscriptions, which are now 
edited for the first time, belong to Mr Subbaya Nagappa Hegdej of Ajjib^l 
in the Sirsi Taluka of the North Kanara District They have been in the 
possession of Mr Hegde’s family for a very long time , so long, in fact, tliat 
nothing is now known as to when and under what arcumstances the plates 
came into the possession of the family I obtained them on loan through 
the good ofBices of Mr. Shankarrao Kaenad, High Court Pleader, Bombay, 
who, at my request, kmdly induced his colleague Mr V G tfcci®, B A , LL.B , 
Sursi (a son-in-law of the owner), to send the plates to me for inflection and 
to allow me to take impressions from them I am thus editmg the grants 
from the original plates, which were on loan with me for about six months 
during 1918, and from a set of inked impressions prepared from them in the 
office of the Superintendent, Archseological Survey, Western Circle The 
annexed facsimiles were subsequently prfiared under the supervision of the 
Government Epigraphist from the impressions supplied by me. The trans- 
cnpt given below has been carefully compared (m manuscnpt) with the 
originals before the latter were returned to the owner My smcere thanks 
are due to Messrs Kahnad and Hegde for this opportunity of offenng here a 
description of these interesting records of the reigns of the Kadamba kings 
Ravivarman and Eridinavarman of VaijayantI (Banavasi). Their chief 
claim to our attention lies ui the regnal years in which they are dated ,The 
grant of Ravivarman was made (if my reading of the date is correct) in 
the thirty-fifth year of his reign, and -that of Krishnavaiman in the nineteenth 


These are three copper-plates, the first and last of which are inscribed 
on one side only, and each of which measures roughly 51" long by 3" broad 
They are quite smooth, their edges being neither fashioned thicker nor raised 
as rims. Although the plates are fairly thin, the aigraving, not bemg very 
deep, does not ^ow through on the reverse sides. The letters diow evident 
traces of the working of the engraver’s tool. The entire inscribed surface of 
the first plate is more or less corroded, but only at a few places has the 
engraving thereby been so far affected as to have become quite rllesgible. The 
seomd plate is, m a sense, in a worse conditum, as three of its edges are 
^ten away ; and with them the greater part of 1 6, about a third of 1 17, 
and some syllables in 11. 11 and 16 ard completdy lost. The tbird plate is 

* [Ep. Ind. 16. 264-72.1 




fortunately quite untouched ; and the engravmg on it is in almost perfect 
state of preservation. The most deplorable part of the havoc wrought on 
these plates by the destructive agency is that m line 11 some of the letters 
comprising the words expressing the date are damaged in such a manner that 
the reading of the date (which is by far the most important element of the 
record; has to be based on a conjectural restoration from which the element 
of unceitainty cannot entirely be eliminated Of no great consequence is, 
on the other hand, the damage to line 6 , for from the preserved fragments 
of letters we may^ I think, safely conclude that the Ime contained nothmg 
more than a eulogistic phrase or two, which, even if restored, would liave 
added nothing of importance to our stock of knowledge concerning the history 
of the Kadambas. The plates are pierced by a circular hole so as to receive 
tlie ring and seal which are attached The weight, mduding the rmg and 
seal, is 38J tolas. The ends of the ring are securely soldered on to the bnrir 
of the seal About an eighth of an inch of the edge of the latter is raised so 
as to form a nm , the recessed space, which is oblong m shape, is devoid of 
legend or emblematic design. 

The characters, which show great umfomuty throughout, belong to the 
southern variety, and have close affinities iwith those of other grants of the 
Kadamba kings, especially with the £265} Hala^ plates of the Kadamba 
Ravivarman, published by the late Dr. Fleet. The letters t and «, alike 
whether used angly or in conjimction with other consonants, are devoid of 
loops : nevertheless they are clearly distmguidiable from each other For 
in » the nght limb of the letter is regularly drawn m contmuation of the 
slantmg (or vertical) stroke , whereas in t the upright stroke is much shorter 
and distinct from the lower part of the letter, which forms a horse-shoe 
(sometimes with unequal arms), and to which the short vertical stroke is 
attached at the top. It may be added that owing to this characteristic even 
the upper half of the letter t is sharply distmgui^ed from, the corresponding 
portion of v, in which the vertical stroke is regularly drawn in! continuation 
of the ri^t limb (as in «), a fact whose importance will be apparent when 
we shall turn our attention to the subject of the readmg of the date of the 
record The difference between the forms of t and v may be studied in the 
following examples : Hatifi’* and protikrit%° in line 3, ’’pati-pratimah 1. 7, 
tithau 1. 12, ^rakshmi 1.19, i)hovati 1. 2i0 ; and ‘vtjaya'" I 1, ^vipvla” 1. 8, 
and '‘vinayfl!* and °«>iiarada 1, 9., In 11. 7 and 10 occurs an initial a ; m 
11. 10, 12 a ; in 1. 20 «; in 1 19 final A ; in 1. 14 final f ; and m 11. 17, 21 final 
tn. For final consonants, as is usual in these records, the full forms are 
used m reduced size^ written on a dightly lower level than the rest of the 
fettas of the line. The medial vowel in is written by bendmg back the 
last downward stroke m an upward direction, e.g. in Imes 2, 3, etc.— The 

1 iMd. Ant., Vol. VI, pp. 25 ff 



language of the mscnption is Sanskrit, and, with the exception of the impre- 
catory and admonitory verses at the end (11. 20-23), the text is in prose The 
document, it may be added, begms and ends somewhat abruptly The grant 
propier is couched in very terse language The preamble does not mention 
any of Ravivarman’s ancestors, and the epithets coupled with the name of 
Ravivarman himsidf, iwhich are of the stereotyped form, are, relativdy speak- 
ing, few m number They contain no new historical information regarding the 
royal donor. In its brevity the record resembles closely the Nilambur= plates 
of the Kadamba kmg of the same name — iThe orthography does not call 
for any particular remarks 

The inscription is one of the Dkarnia-M^Staja Ravivarman of the 
Kadamba family Wd have already the Hal^ and Nilambur plates of a 
Kadamba Ravivarman The highest r^nal year recorded in these grants 
IS the eleventh The present grant records (11. 10-19) that on the fifth tithi 
of the bnght half of the month of K5.rttika in a specified regnal year (the 
leadmg of which is uncertain and will be discussed later on)' Ravivarman 
granted to the Mahladeva temple of his beloved physiaan, the des-Smdtya 
ISulakaotha,® four mvartanas of land in the village of S&re (or Sara), of 
which further specifications will be found in the appended translation In 
this portioin of the record (U. 16, 17) there is a lacuna, in which some further 
details of the donation are lost 

The genealogy of Ravivarman is not given. But, as the writmg of the 
present record does not differ m any essential pomts from that ot the Halsi 
and Nilambur grants of the Kadamba king of the same name, we may on 
palaeographic grounds tentatively identify him with Ravivarman, the son of 
Mrige^avarman and grandson of Swtivarman. 

The reading of the regnal year is, as stated above, uncertain The year 
IS expressed in words only (as in all the records of this dynasty that have 
come under my notice), which I read as paHcha [tritn]iat[t<me], ‘in the 
thirty-fifth ’ The compound indubitably contains the dement pancha-, which 
is dear, and another word, expressing a multiple of ten, which is obhterated. 
The second syllable of this partly defaced word contains again unquestionably 
a s. The choice, therefore, lies between -vithie and or 

and -triihSattame. As, moreover, the sign of e does not appear to have been 
added to i, the mtended akskara must be taken to be fa. This dicoinstance 
further reduces the possible alternatives at our disposal to -vtm^atitame 
and -truhSattame Further, the remnant of the akskara after fa appeals 
most to resemble a deformed t, very faint, indeed, but still distinguishable 
on the plate^ a coodusion which is in harmony with the above supposition 
that the longjer form of the ordinal (vimSatitama or trhhSattania) has been 

2 Above, Vol Vin, p. 147, and Plate 

® See bdow, p. 26S, foot-note 10. 



used here, and not the shorter {vtmsa, tnmsa) Let us now turn our atten- 
tion to the syllable preceding sa The preserved portion appears to consist of 
the medial i and a short vertical stroke added at the top of a mutilated horse- 
shoe. Therefore, from what I have said above regarding the shapes of v 
.md t, it follows that this defaced akskara can only represent a vi and not U. 
Ihis result also fits m with our former observation that the third missmg 
syllable is a deformed ta (and not tt ) , for an imtial t requires the restoration 
-tnmsattame (cootammg ta in the third syllable), while an initial v would 
necessitate the reconstruction, -viimattteme (with ti in the third syllable) . I 
have, therefore!, for my part, no hesitation m readmg the preserved portion 
of the first damaged akskara as ti, and supplemaiting the lost subscript r 
under it The second syllable is, as already remarked, sa beyond doubt. Then I 
read t[t]a, after which there is just suffiaent space for the inclusion of me, 
which syllable, however, is completely obhterated The complete restored 
regnal year would, therefore, be pancka-trtthsattam^ ' in the thirty-fifth year.’ 
It may be added that, if the reading proposed by me is not accepted, the only 
possible alternative is pancha-vtmSatttarm, which in my opinicsi is extremdy 

The village Sane (or Sara), which is the object of the grant and which 
is mentioned without any specification of its whereabouts, remauns unidentified. 


[Metre of two verses in 11. 20-23 : Sloka iAnuskfubk).] 

First Plate. 

1 II ^ lf i m g l ^< T- 

2 ( *ir ? ) ( ii ) ^ ]- 

3 "tl sifttfigsgoif 

4 < fs wfa i u»w° ^ y. 

5 vwtjjfRTsr: jRjpworaH [ ] . .® 

Second Plate , First Side. 

6 1 

* [The farm traycss-trimSattme occurs m a Telugu record from DrakshSiama . 
No. 349 of the Epigraphical Collection (Maxiias) for the year 1893.— H. K. S ] 

^ From the oiighial plates and a set of m^ressions; 

^ Read 

^ Ihe bradceted letters are coiyecturally added; at this point the plate is 
worn almost ta the d^th to which the letters were mnaed. 

® The last two or three syllables of line 5 have worn away and become com- 
pletely illegible 

* The upper edge of this of the middle plate is eaten away , and, with 
it, the upper i>ortiona of the letters in L 6 are either effaced or completdy lost. It 
is needless to add that the vowel signs afe almost all completdy obliterated, and, 
in the reading ^ven above, only conjecturally supplied. 





C2673 g 









*ra1^Praf^5IRS: ^®TOT8n*#?PRJ|S0t- 

[ 'a- ] 

Second plate ; Second Side. 

« [ I ] qawjf snwR:^® 


[% ] ai«RRi«T [ i* ] 

. . . ^ feynii . 

. . . [t] [*!*] q [’i^] . 

Third Plate. 

q^imRqisa® [ u* ] qHSrajrft ^Rsqq’Rgwn® 

[ \\* ] [ i* ] qqrq? qr qt ag- 

Her^ and m other places bdow, the rules of samdhv have not been observed. 

11 The sign of the visarg^i zs defaced. 

W Read vrifiSRSfil”. 

1^ I can make no sense of the syllables Read opf ^ ? 

See below, p. 268 n. 7 [ i= 28 infra] 

1^ The lower portion of all the remainmg letters of this line are more or less 
defaced. Of the bracketed syllables, the preserved portion of the first, I am fully 
];)ersuadjed, can be nothing but ti (see above, pp. 265-6) , the next syllable, 5o, is 
quite distinct and unmistakably both on the plate and in the impression ; further- 
more, I beheve, it is possible to discern on the plate very faint, but unmistakable, 
traces of a diminutive t (which must be a part c& a hgature) and somewhat uncer- 
tain traces of wi I have, therefore, no hesitation in supplymg the nxissing subs- 
cript r below the t% and I may say that I look upon the reading trimia as more or 
less certain. 

The subscript ma is rather faint, anU appears to have left no trace on the 
impression paper. 

i« Read 05 [°. -^r Qr 

18 The final t (for which the full sign is used), written below the line, is 
faint , but It can be made out on the onginal plate quite unmistakably. 

Or 2 

ao The sign qf the medial t in the bracketed syllable appears to have been 
crowned out of its natural position (which is a little more to the left, over the 
hoillow of pa) by the subscript ya of the ligature immediately over the siyU^Dle 
in question. [Possibly the readmg is 

21 A short space is left between ^ and 


^ [ \\^ ] 

[ I* 3 

?sRI ^jj5r: <1^ ^ ^ [ H* ] 


(Lme 1.) Hail * At (the aty of) victory, the glorious Vaijayanti, the 
DhaT 7 narMahaTaja,^^-^(ol the family) of the Kadambas, anointed after medi- 
tating on Svami-Mah^na and the assemblage of the Mothers ; belonging to 
the Manavya gotra , descendants of KEntI : studying the requital (of good and 
evil) as their sacred text,2<^the glonous Ravivarman before whose prowess 
(are) prostrate all-'* similar to the great leader of the armies of Kadam- 
ba^o (the excellence oP’') whose body had been produced by great religious 
merit acquired in numerous births, well-versed in (rules of) statesman^p and 
decorum, highly nghteous and deeply devoted to his father, on the fifth tithi 
of the bright half of the month of Kaxtbka in the [ thirty] -fifthss year, in un- 
interrupted succession, augmenting his life and sovereignty, has given®® 
four nwafttanas (of land) m the plough-land called Bairhdupukro[pi] (or 
Bamdu®) below Dasa-taKjaka (and) above Maimb5 (situated)* in 

the village of Sane or (Sara), to the temple of Mahadeva (Siva) of his be- 
loved physician named Nilakantha, the dti-amatya ^^ , two parts of it (are 
given) for maintenance . . up to the temple . . to Arya-svamin and 




22 Read 

23 Here used as a title Its literal meaning is ’ the Mahatma who is devoted 
to the performance of duty {dharmd) 

2-* I have adopted Kielhorn'S rendering of the difficult phrase pratikrti^-t and 
I may refer the reader to his note on the subject, Ep- I«d., Vol. VI, p. 15, note 3 
The rest of the sentence is lost 

20 Compare the epithet Kadamba-smanl>‘hy%had-anvay(i^vy[d]ma-cluind7amah 
(f the full mcxin in the firmament of the great lineage of the Kadamba leader of 
armies’), applied to KSkuathavarman in the Talagunda pillar inscription of Ka- 
kustihavarman, e<L Kielhc«w, Ep. Ind , Vol. VIII, p. 31 

27 I suppose we have to supplement here some such words as these 

23 See above, p. 267, note 3. 14 sitpra] . 

20 I propose to amend the text and read am[pa*\y^=anupurvya The un- 
intermpted succession refers naturally to the king's regnal years I have not come 
across the phrase dsewhere ; but the emendation gives, in my opinion, quite a satis- 
factory sense 

20 There is a lacuna in the text here. 

8^ The expressions adhastSt and upari may have been used with refer^ce to 
the levd of the field under description. 

DU-amStya literally means ‘the minister of the country (or province)/ 
but it may have a more specific meaning here. Cf. with this expression Ihe modem 
aumames De^imukh, Deshpande, which are undoubtedly derived from original titles 
of funcricwiaries. Or should we take Nilakap-tha as the name of <iountry ? 


PgiSupata bdonging to the Ka§yapa gotra and the Bharadvaja gotra (Res- 
pectively) . 

(Line 19.) He who protects it will have a share in the merit) accrumg 
from it 

(Lme 20.) It has also been said — 

[Here follow two of the customary admonitory verses ] 


These plates, which are m a much better state of preservation than the 
foregoing, are also three m number They measure roughly 6J" long by 23-" 
broad. They are quite smooth, their edges being neither fashioned thicker 
nor raised mto rims. The plates are thin , but the engraving being shallow, 
though otherwise quite good, the letters do not ^ow through on the reverse 
sides at all The letters show the characteristic marks of the working {269} 
of the engraver’s tool The grant is engraved on the inner side of the firfet 
and last plates, and on both sides of the imddle one The plates are pierced 
by a arcular hole m order to receive the nng and seal, which are attached. 
The ends of tiie nng are, as m the case of the plates of Ravivarman, soldered 
on to the back of a seal, which, m this instance, is oval m shape and bears 
a device. The seal has a raised nm, and inside this there is shown in low 
rehef the figure of a quadruped (perhaps a horse) facmg left. The weight 
of the plates, including the nng and seal, is 52 tiolas Each engraved side 
contains four lines of writing ; there are thus sixteen lines m all. Excepting 
isolated letters which are worn away and now become partly iHe&ble, the 
record is m a perfect state of preservation, and can be deciphered without any 

The characters belong to the southan variety, and have close affinities 
with those of other grants of the Kadamba kmgs They differ palpably from 
the characters of the grant of Ravivaiman described above and appear 
to belong to a later palaeographic epoch The vowel a in wJ is written by 
bending back the last downward stroke in an upward direction , e g. in II. 2, 
3, etc. One notices the tendency of the vertical lines to slope, a feature which 
later develops mto the spiral formation of Hala-Kannaija letters. Noteworthy 
is also the doubling of the left limb of g (11 1, 2, 6, 8, etc.) and i (11. 4, 7, etc.) 
This record contains the earhest spedmen hitherto known, in a southern alpha- 
bet, of the ini tial fi (1. 8),. Initial a occurs m 1 5 , initial a in 11. 4, 6 ; 
initial M in 11. 11, 13 ; initial I in 1. 7 ; the agn of final i in 1. 7, and final n 
m 1. 11 One ligature, with the tword containing it, hja remained undeci- 
phered in 1. 10 ; I have never come across the sign anywhere before and can 
suggest no reading for it— The language of the inscription is Sanskrit and, 
with the exception of the imprecatory and admonitory stanzas at the end, 
the text is in prose. The main part of the test (11. l-'ll) forms a single sen- 



tence and states, like the foregoing grant of Ravivarman, without much cir- 
cumlocution the object of the record. The attributes of the donor are of the 
stereotyped form. In its brevity this record resembles the grant of Ravi- 
varman descnbed above. 

The inscription is one of the DharmaAMaharaja Krishijnvarman of the 
Kadamba family. The hitherto known records of the Kadamba dynasty have 
levealed the esdstence of two Kpshijavarmans in the family. And, as the 
present record neither gives the genealogy of this king nor mentions any cir- 
cumstance which would help to establish his identity, it is difficult to affirm 
with certainty whether he is to be identified with either* the one or the other 
Ktishnavarman already known, or whether he is a new kmg altogether , but 
on palseographic considerations this kmg may tentatively be identified with 
the second Kadamba kmg of that name, whose BannahaUi (now 
grant,^® dated in the seventh year of his reign, has already been publisheld. 
The grant proper records (11 6-11) that on the full moon day in the month 
of Kjarttika, in the nineteenth year of his reign, Krishnavarman granted Kal- 
makapalli m the Girigada village {grama) of the Karvannad district {visha- 
ya) to a Bifih ma i ija of the Vai^Un gotra, named Soma-svfimin, who wad ai 
student of the Rig-veda, and a performer of the Soma sacrifice, making the 
village free ftom all taxes and dura 

To the proposed identification of the Kpshnavarman of our record with 
the Krishjjavarman of the Bannaha](li grant it may be objected that the title 
DhaTmarMahar^o, which is here used along with the name of the donor, is 
not found coupled with the name of Kri^mavarman II. m any other record ; 
thus, for instance, in the Bannahalli grant itself, which is dated in the seventh 
year of the reign, only the shorter title Mahar<^a is prefixed to Kpshijavar- 
man’s name. On the other hand, the earher Kpshnavarman is invariably 
styled Dharma-McASraja in the preambles of the later Kadamba grants. The 
objection is not valid ; for it ^ould be ni^sd that Krishnavarman 1. was, ac- 
cording to all accounts, performer of a {270} horse-sacnfice. If our Krish- 
pavatman is to be identified with this kmg, how are we to explain the aiienc** 
of the itecord te^rding the sacrifice said to have been performed by him ? 
On the other hand the expression aSva-midh-&>hishikta, herein applied to 
the Kadambas as a class, shows that in the time of our Kpshnavar man the 
qjithet aSva-midha-yajm had become a hereditary title of the T^arfanih p 
family, a fact which can be explained only <ki the assumption that some pro- 
longed interval of time separates the actual performer of the sacrifice from 
wr Kjridnjavarroan. Moreover, there is at least one other instance of the 
indiscriminate use of the titles Maharaja and Dharmd^Mah&raja, namely, 
in the case of MpggSavaiinan. Both titles are found used m connection with 
tlds king in epigraphic records.®* 

*• Bp, Indi, Vd. VI, p, 18 and plate. 

KlELHOBirs Lkt InscriptUm of Southern India, Nos, 604 and 605 



A word may be added regarding the localities mentioned in the record. 
The object of the grant is stated to be Kamakapalh, situated in the Gin- 
gaida village (.grarm) of the Karvannadga district {vtshaya) None of tVipgp 
places can be identified with certainty Mr Hegde, owing to whose good 
offices the plates were made available for publication, is a resident of Sirsi 
and has favoured me with the following topographical details, which throw 
some light on the question He writes • “ Sirsi tMukd (which used to be call- 
ed Sunda aSIufea) was formerly divided into a number of ni4gane, pgr^h ot 
which consisted of a number of villages One of such went by the 

name of KarOr mdgane, denving its name from Karur, a village mrliid e d in 
the mdgane. Another such village was called Gingad|de Both these villages 
still bear the same names ” The proximity of Girigadde to Siisi favours the 
identification of the former with the Girigada of the plates, whidi, as stated 
above, come from Sirsi itself. Also, in regard to the great and often inex- 
plicable changes which many place-names have undergone, the identification 
bf Karvannadga with Karur is not an impossible proposition. 

TEXT a® 

[Metre of the two verses in 11. 14-15 Sloka {Anustubha) ] 

First Plate 

1 [«* ] Puimfa si qgg qr " " 

2 ongwirai ( sn ? ) [ \ 

Second Plate ; First Side 

5 Hpi [ ;^ ] [5«] ^ J ] hjwt [^]* 

8® From the onginal plates and a set of impressions. 

“ Read s??t. [The author may have meant this word to be in the ablative 
case. Cf. VjUja^t^kanfUiiSvSrSt of other inscriptions. — H. K. S.] 

S’ The length of ma is added at the top of the akshara. 

88 Read 

8® The length of ma is added to the constncted part of the ak^jhara. Read 
^0 Readorf. 

The Ravivarman plates above read Read Here, and 

in other places below, the rules of samdht have not been observed* 

« Read 5Tf. 

^ The insertion of the visarga is an afterthought, 

♦» Read 

epiceiaphic studies 

{2715 7 

8 sR?^ [ t* ] ’mispm- 

Second Plate \ Second Side, 

9 «TO?r ahjwftsr 

10 5JT«<ri^ ytthRgn . . “ 

11 [ «* ] 

12 ?i 3«rR3wnwraRi «iain- 

Third Plate. 

13 qsRfNi^^fg?^ [ ii* ] 3^ [ i* ] [ ;* ] 

14 ^g’sn gw ?i3if5r [ :* ] 5iiwi^ [ n’*' ] gr® 

15 fir [ ;^ ] 3^ iR*r cf^ *>53®* [ II* ] 'i?TtrP wr «it ^ 

16 sig«Rr“ [ I* ] iiTOPsnfii firwn® f^: [ ii* ]. 


(Line 1.) Hail ! At (the dty of) victory, Vaijayanti, the Dharma-Ma- 
hdrSfa ,^^ — (afi the family) of the Kadambas, anointed during a horse-sacri- 
after meditating on Svami-Mahasena and the assemblage of the Mo- 
thers ; belonging to the MSnavya g5^rc ; descendants of Hariti ; studying the 

The final t is mitten below the line Read 

■is Read The length of md is added at the top of the akshara. 

[The last syllable of the name of the district appears to be SF, not ff* — 

K S] 

The last but one ak^ra remains undeaphered ; the very last one of the 
line is dther va or cha, with or without an anusvdra, [In my opimon the unread 
letter is Ika ; and malkava, like hitanya, must be a technical term indicating some 
source of village income. In the Nilambur plates of Ravivarman (text 1 8l) the 
same term occurs in the form malkavu and Mr, T A. Gopmath Rao has taken it as 
the name of a hamlet. — K. Sj] 

The final n is written bdow the line. Red 

®® The ^gn of the secondary 5 seems to have been also added erroneoudy to 


8* Read <5®. »» Read “frf. »« Read 

M Read m Read ‘iqt. »« Read fftr:. 

•0 Here used as a title. Its literal meaning is ' the Maharaja who is devoted 
to performance of duty {dharmd)* 

61 An ancestor of the donor of the present grant is ^ken of as having per- 
formed a horse-sacrifice ; cf. the Bannahalji plates of Kfiahi^avannan II., ed 
Kibphorn, Ep, lnd„ Vol. VI, p. 18, l. 5 

TWO kadamba grants prom SlSSl 


requital (of good and evil) as their sacred text®* , and looking to the Mothers 
of Mankind for protection, — ^the glorious Knshnavamian, who during count- 
less births has accumulated an abundant store of rehgious meat, who has 
gained fame and the fortune of royalty by virtue of successes in many battles, 
in the mneteenth year of his prosperous ^2723 (rdgn) of victory, on the 
full-moon (day) of Karttika,®* for the religious merit of his father and 
mother, has given with pouring-out of water, with gold, {income) and . . . 
(and) with every exemption, Kamakapalh in the village {grama) of Girigada 
in the district {vishaya) of Karwannlajdga to the Soma sacnficer Soma-sva- 
min, bdonging to the V|a!itShi gotra, who has completely studied the Bg-veda 
and who follows (the moral and ethical duties known as) yama and myoma. 

(Line 12 ) He who shall protect this (chanty) will share in the merit 
(attaching to the making of it) ; and he who shall confiscate it will be (guilty) 
of the five great sins. 

[(Here follow two of the customary admomtory verses,] 

I have adopted Kjelhorn’s rendering of the difficult phrase pTatikrtta", and 
may refer the reader to hia note on the subject, Ep, Ind , Vol. VI, p. 15, note 3. 
'[The next attribute has been translated by Mr Gopmath Rao, 

perhaps more correctly, ‘who were {Itke unto) mothers to people {who were) 
depeiident {on them)', above, Vol. VIII, p. 148. — H. K S.] 

88 The full-moon day of Karttika, as a day on which donations were made by 
the Karfamha lon gs , is mmtioned also in the Nilambur plates of Ravivarman {Ep. 
Ind., Vo4. VIII, p. 145) and the HalaE plates of Migedavarman (Ind. Ant , Vol. VT, 
P. 24).. 


Thas inscription, which is now brought to notice for the first time, was 
discovered by my friend Babu Rakhaldas Banerji, Superintendent, Archaeo- 
logical Survey of India, Western Circle, in 1919, dunng one of his touisi of 
inspection in Central India The excellent estampages from which file ac- 
companying blocks have been prepared were made under his direct super- 
viaon, and very kindly placed by him at my disposal for publication. 

The inscription, Mr Banerji tells me, is engraved on a detiached slab 
of stone which he found lying at the bottom of a donga, adjommg a hill called 
Maluha-tongi near Ganj in the Ajayagaijh (Ajaigarh) State in Bundelkhand 
Close by is a ruined stone structure, probably a dam to hold the waters of 
the stream passing along the donga. The find-place of the record is not far 
removed from the ruined city of .Kuthfira, where Cunningham discovered 
in 1883-84 the Nachaneki talal inscription, which was first brought to notice 
by him, in 1885, m Archxological Survey oj India, Vol XXI, pp 97 f and 
re-edited by Fleet in Gupta Inscriptions, pp 233 ff and PI xxxiii B The 
Ganj inscription, like the one discovered by Cunningham, is one of the 
oldest records of the Vakataka dynasty, and as such is worthy of being care- 
fully preserved 

From the subjoined transcript it will be seen that the text of our ins- 
cription is practically identical with that of the Nachane-ld-talai record of 
the reign of Maharaja Prthwishena, edited by Fleet in Gupta Inscriptions ; 
it differs from the latter only in the length and the number of Imes, and in 
tlie spdling O'! a couple of words. But our inscription is in a much better 
state of preservation than that edited by Fleet ; at all events the stone has 
yielded an impression far superior to the one from which the block accom- 
panying Fleet's article was prepared. Consequently we can study the forms 
of the letters m the subjomed facsimile much better than in that of the Na- 
chane-ld-talal version. Moreover, the wnting of this inscription bang per- 
fectly distinct, we can give a traiKcript which is more rehable, and which at 
the same time disdoses certain minor inaccuracies in Fleet’s transcript, errors 
which even then could have been avoided by a more patient study of the 
available material. 

The writing covers a space about 25" broad by 12" high. In the centre 
of the first lira of the inscription there is a sculpture of a wheel, of which only 
a part is visible in the facsimde The average size of such letters a&m, p and 
V is about 2".— The diameters belong to the ‘ southern' variety of alphabets, 

* !£:#>. Ini, 17-12-14.] 

A VAKATAKA inscription fftOM GANJ 241 

of which the distinguishing features, in our inscnption, are the hooks at the 
lower ends of the verticals of k and r. In particular, we may say that the 
letters are a specimen of the Central Indian alphabet of the penod, which on 
account of the peculiar ‘ box-headed ’ tops of the letters is known as the ‘ box- 
headed’ sub-jvanety of the southern alphabet^ In our speamen the boxes 
are very conspicuous, and uniformly hollow The letters are unequal in size 
and uncouth in appearance It may be added that they betray a conscious 
effort to substitute angles for curves in the configuration of letters. The 
letters t and n are sharply distinguished from each other : the latter has al- 
ways a knot at its lower end — The language is Sansknt, and the mscnption 
IS in prose — ^As regards the orthography the only point calling for remark 
is the phonetic doubling of the d of dh, in °d-d{m)ntiddhydto° (1 2), before 
y, and of the t of th, before r, in punyd-rttM (1. 3). 

{^13} The mscnption, which is a record of the reign of Maharaja Prithivi- 
shepa (I ) of the Vakfiltaka family, states maely that a feudatory of his, 
Vyaghradeva by name, had made something or other for the sake of the rdi- 
gious ment of his parents. The exact nature of this act of piety has been 
left unspecified, just as m the other version discovered by Cunningham. 
The silence of these records on the pomt leads us to infer that the slabs on 
which the inscnptians are inscribed must have been built into that the 
njakmg of which they were mtended to record. 

Our information regarding the Vakfitaka dynasty is unfortunately very 
scrappy. All the important events in its history known to us have been su<V 
cmctly summarized by Kielhokn* in his article on the Balaghat plates of 
Piithivishi^ II ; we can even now add nothing of consequence to what has 
been said there. We do not possess exact dates for any of the kings of this 
family, nor can we form any dear idea of the extent of the country ruled over 
by them. Regarding Plthividiena I. we know that he was the son of Rudra- 
sena I and the great-grandson of Pravarasena I., the latter being either the 
very first king or one of the early kmgs of this house. It should seem that 
the VHkHtaka king at whose hands the ‘ lord of Kuntala ’ had suffered de- 
feat, as recorded m the Vakataka stone inscription at AjantS,® was this same 
Prithividiepa. Beyond these few facts we know nothing of much conse- 
quence regarding the king referred to in our record. 

About Vy&ghradieva, the feudatory of Prithivishena, we know still less. 
Indeed, Vyaghra appears as the name of chieftains in several well-known ms- 
criptions but it is not possible to identify our VySghradSva with any otf 

^ See BiiHLER, Indtsche PaJssograplue, p 62. 

® Above, Vol. IX, pp 2681. 

® Arch. Surv. West. Ind„ Vol. IV, p 124, verse 8. 

* Kielhqrn's Ust of InscripUons of Northern Jnita, Nos. 270, 387 and 509. 




Buiileh"' assigns the coppei-plates of tlie Vakataka Pravarasena II , the 
friandson of Prtthivishwia I, to the fifth or sixth century A D , it is not known 
to me on what grounds. I have examined the inscriptions of the Vakataka 
dynasty and compared them with the allied inscriptions engraved during the 
time of the Guptas,* of the kings of Sarabhapura.^ of Tivara.^ of Kosala and 
of the early Kadamba kings,® without bemg able to arrive at any definite 
conclusion regarding the age of the Vakataka inscriptions. Buhler’s date, 
however, appears to me to be far too early My impression is that there can 
be no objectiCMi, on pabeographic grounds, to assigning this lecord of the Va- 
ifafgkas to as late an epoch as the seventh century ad. I conclude this short 
notice by drawing attention here to the remark of Kielhorn that the B^- 
ghat plate of Prithivishena II., who was the son of the great-grandson of the 
Prithivisheija of our inscnption, “ may be assigned with probability to about 
the second half of the eighth century A.D ” 


1 '■‘Vakdtakana maharaja-srii* 

Pj thivishSiia-pad-S (m ) nuddhyato Vyaghrade- 

'S v6 nwtapitiofh''] i^puny-artthe i-ikrtam = iti [U*,] 

{143 translation 

Vyaghradeva, who meditates on the feet of the M^ardja the illustrious 
Epthivishapa, (of the family) of the Vafcatakas, has made (this) for the sake 
of the religious merit of (his) parents. 

1 Indiiche Pai»>trapJtie, pp 62 f. 

« Corpus Inscriptiomtm Indicarum, Vol, I, Nos. 2-3 
r Gupta Inscriptions, Nos 4041. # Ibid , No 81. 

» Ind. Ant., VoL VII, pp. 35-7. i® Above, Vol. IX, p 270 

Flora a set of estampages prepared and kindly lent to me by Mr. R. D. 

Read VakatakSnam Fleet in his transcript has wrongly spelt this word 
with the dental n in Gupta Inscriptions, Nos 53-64, 

•» Read M. 

-Read pu^y-aritke, Hae also Fleet haa wrongly transcribed the word, both 
as regard} the dental « and the case-ending. In Cunningham’s version the word 
is spelt exactly as here. 

“ The ccrartruction is faulty. The verb should be in the acUve voice. 


I edit here two new Valabhi copper-plate grants (one complete and one 
incomplete ) which were presented, m 1918, to the Trustees of the Prince 
of Wales Museum, Bombay, by the Bhavanagar Darbar, which is ever ready 
to further the cause of epigraphic research by placing ungrudgingly the 
matenals, as they are discovered, m the hands of students of Indian history 
for investigation and publication, and, when possible, by havmg them ex- 
hibited m centrally situated hiuseums The plates under reference were 
discovered at the bottom of a small tank outside the ISatrunjaya Gate at 
PalitSna while the tank was bemg drained during the time of the late Thakor 
Saheb of that State ^ 

A— Plates oe Dhruvasena I. ; [Valabhii]-Sam[vatJ 207. 

The plates, which aie mscnbed on one side only, are two m numbei, 
each measuring roughly llj" bioad by 6|" high. The edges are just slightly 
raised m order to protect the writing, which (excepting portions of 11 14) 
is m a state of perfect preservation The plates are of fair thickness ; but 
the letters, bemg deep, show through on the reverse sides. The engraving 
IS well executed Each of the plates has two holes bored iri it A ring of 
copper passing through one pair of them serves to hold the plates together 
at one end. The seal, which is an invariable accompaniment of such plates, 
is missing. The aggregate weight of the plates is about 102 tolas. Each 
plate contains twelve lines of wnting , the last line but one of the second 
plate contains the date 

From the foregomg description of the plates, as well as from the fac- 
similes of them appearing with this article, it will be evident that this record 
does not differ m any stnkmg particular frotn any of the hitherto pubhshed 
records of the same king Only in the portion dealing with the grant proper 
does the text of this msenphon differ, for example, from that of other plates 
of this kmg which were discovered some years back also at PSlitanS, and 
have been edited by Dtr Sten Konow m a former issue of this Journal.* 
The royal donor, Dhruvasena, as well as the dutaka Mammaka and the 
writer Kikkaka, are names well known to the Indian epigraphist It will, 

*\Ep Ind. 17, 105-1,10] 

^ My fnend Pandil Ginjasankar Vallabhji of Rajkot, Curator, of the Pnnee 
of Wales Museum, Bombay, rnforms me that the five PElitSnB plates edited by 
Prof, EONOW (above, Vol. XI, pp. 104 ff.) wa'e discovered at the same place and 
at the same time as the plated here desenbed. 

* Above, Vol. XI, pp, 104 ff. 



therefore, be unnecessary to go here into a minute description of the char- 
acters and orthography of this inscription It will sufi5fee to ob3er\'e that 
the alphabet offers a specimen of final t (1. 15), final w (1. 23) and the 
numerical ideograms 200, 7, and 5, and that the name of the founder of the 
dynasty is spelt a Bhatakka (1 3). At the end of line 12 is to be found 
a hoiiaontal stroke, about i*" long, emdently drawn with a view to fill up 
the empty space remaining at the end. The reason for leaving the space 
vacant appears to be that the writer did not wish to commence, at the end 
of the Ime, a long word the whole of which would not have been' contamed 
in the short space that was left over. 

The inscription is one of the Maharaja Dhruvasena [I.] of the Maitraka 
dynasty, and the grant contamed m it isl issued from the city of Valabhi 
The object of the inscription appeals to be to record the confirmation by 
Dhiuvasena of the donees a Brahmalna, named IViJadhava, of the Sunaka 
gotra, student of the Chhanddga School, and resident of the village of Jyesli- 
thanaka (stated to be Akdiasaraka-pranesya) in the Hastavapra-harani in 
the possession of some {106} land already enjoyed by him in the village of 
which he was a resident. Besides Hastavapra, which is the modem Hathab 
(6 mileSj south of (Sogha in the Bhavnagar State), and Valabhi. which is 
commonly identified with the modem Vala (situated in 21° 52' N and 71°57' 
E.), none of tha places can be located. The date of the record is the year 
207 (given as usual in numerical ideograms), and the 5th (tithi) of the dark 
fortnight of VaiSakha. The year when referred to the Gupta-ValabhS era 
yields a.d. (207 + 320) = A.D. 527. 

There are two expressions! in this inscription, both occumng in the por- 
tion dealing with the grant proper, which deserve some comment they are 
Akihasaraka-praoUya^ (1, 12) and sa-idbesram (1 16). The latter we will 
con^dor first 

Being mentioned along with the well-known technical expressions so- 
hnany-adeyam and sa~bkSta--vdta,'‘ sa-iaibaram must be a term of like 
nature, i.e a technicality of the lawyers ; but what its significance may be I 
am unable to surmise. There can be no question regardmg the correctness 
01 the reading ; the letters are perfectly distinct The word iaibora is not 
to be found in dictionanes ; nor have I (x>me across it elsewhere I can 
only think that it may be, as it stands, a clerical error ; but I am unable to 
suggest any plausible emendation fOT it 

The word praveiya in the other expression referred to above is also one 
that presaits some difficulty to the interpreter. Here it .is used in a com- 
pound with Akshasaraka, evidently a place-name, and serves to locate more 
definitely the village Jyeshthanaka situated in the Hastavapra-Ztin'nw. As far 
as I know, the word praveiya has been, met iwith only twio^ before : once in 
another Valabhi grant, occurring there m a compound with the same place- 


name Akshasaraka, and once agam in the Khanar grant of MahSsudeva, 
compounded with the word Navannaka, which is also a place-name. 

The former record forms one of the five Valabhi grants from Pehtana® 
edited by Prof. Sten Konow, and is a grant of Dhruvasena I , dated in 
Samvttt 210 In that connection Prof. Konow nghtly pomts out that the 
phrase Akshasaraka-pra«^y« of the grant corresponds to the Akshasaraka- 
prapiya m a third Valabhi grant,^ viz. the Ganesgad (Baroda) plates of 
Dhruvasena, dated Samvat 207. Hultzsch, when editing the latter grant, 
tramlated the phrase by ‘ which belongs to the Akshasaraka-prd^ia.’ Prof. 
Konow, who regards pravUya and prdp^ya as synonjrms, rejects Hutzsch’s 
rendering of Akshasaraka-pr^ya and advances the suggestion that prdvesya 
m this connection means the same thmg as in the phrase a-cksta hha^a- 
prdveiya and accordmgly translates the phrase by ‘ which can be entered from 
(i.e. which borders on) Akshasaraka ’ I cannot, in the first place, admit that 
the expressions a-chdta-bhala-pravUya and Akshasaraka-prai/efyo correspond 
exactly. For m the former the first member of the compound comprises the 
logical subject of the verb contained in prdveiya ; but such cannot be the 
case with the second eiqiression, even if we assign to it the meaning which 
Prof. Konow does. Secondly, I do not understand what is meant by say- 
ing that a village could be ‘ entered ’ from such and such a place If, more- 
over, pravUya meant the same thing as ‘bordermg on,’ as Prof Konow 
asserts, I cannot help thinkmg that the wnter would have employed a simple 
word like sarriipa or parsva-vartm, which lie at hand, to express that simple 
idea of proximity rather than use the arcumlocutioo of pravUya or prdpiya. 
Hultzsch, on the other hand, appeajb to me to be -undoubtedly on the 
right track. He looks upon prSpiya as a derivative of pfapa, which he takes 
to be a word denoting a territorial division smaller than an ahSra Similarly 
the analogous term prdveiya should also be looked upon as a taMhiia of 
prdvesa That this derivation is correct may be seen from the Khariar plates 
of Mahnsudeva, m which a viUagsa is described (L 4) as Kshiiimad-ahdriya 
and Navamaka pravUya. No one will dispute that dhafiya is derived from 
dkdra (-district,’ ‘province’)' by the addition of the suffix; -iya. That sup- 
plies us with the due to the explanation of the other words under considera- 
tion here. All these words are derived by the addition of the second- 
ary -{i)ya to the strengthened forms of the roots d-hri, pra-(d-') vtS and pra- 
(a~)dp (‘ bring to,’ ‘ carry to ’), words with only minute differaices of mean- 
ing. I feel, therefore, constramed to reject the interpretation of Prof. Konow 
in favour of the other. Prdpiya I take to be ‘that which bdongs to the 
Prapa’ and prdveiya ‘that which belongs to the prdveia (or praoUa) 
both prdpa and prdvUa I regard as territorial divisions smaller than the 

®A)t)ove, Vol. XI, pp 104 £f„ and Hates. 
‘Above, Vol. Ill, p, 320, and Plate. 

16 a 

246 epigrAphic studies 


Plate A^. 

10 flfRrsiH3[^s 


12 ?i*n f’5rax^Twn*T«iRR*jnt5*i° 

Plate As- 


14 g ^i^ g pit sRtTRwwsnCioiT^ flfflorBra^^is««» <jvrH«i ' •*‘*’^*0) *n?nf^i’ 

17 aapfl^ gpf fa srai^ isft[d“[i*] wf g g^ i^ wsrai'® 


18 sEq«qr«n*rar*^ ft’^ronr qt g f5t!*n- 

19 ^ww^ * ^ ^w^»^Aw^4><4^5^-l^^^^^H^g4^3^^sas*l [n*] (qr) «iaf%3?tn* 

tiProm the original plates, and a set of estampages. 

®Up to this, the text is practically identical with the text of the Baktaiw plate 
of Dhruvasena I- (dated sofjivat 206), published above, Vol. XI, pp. 106* ff. The 
only varioet iecHortes are unimportant mistakes of orthography, which it would be 
unnecessary to register individually as the facsimiles are there for reference. 

TRead 8 Read 

eln the original a short horizontal sitroke after q'. ^^Read 

11 A short vacant apace between q and sq*^ Read 

laRead wm"". 

18 Read '^?5%'5figOT7I»r°. The anusvara is written over the line between ? , and 
The letters puwvS-bhujyarbhujyamanakah have been engraved over some 
i^tly maaed letters. 

i*Read sf, i^Read “Read 5Fq, 

^^Read ^ “Read 

MRead aiRead “qi^. a^Read W* 

a«Read agr. «^Rea<J a^Read q 8g l flgr<iq|(^ ^ 



£108)20 W smuftW: ^ 

[n*3 q § fii<i4gv n 

21 fI3lfiR5lTRr^[: I*] q?*I »i^r: ^T?r 

^ [«’*‘] qr ^ ^?r 

22 [)*] ^rar "qratf^ [ii*] «^- 

qqif qenssir gfW(;) [i*] 

23 Jrfl^ flflqqt ^ [U*] [u*] 

^ qoo o q H [u*] 

24 JUT JT?i?uia[q*]$FRq [ii*] feftrai [II*] 


[U 1-11 contain the usual preamble ; for translation, cf., for instancy 
that of the opening lines of the Paiit&nS plates. No. 1, edited by Prof. KONOW, 
Ep. Ind., Vd. XI, p. 108 ] 

(LI. 12-16 ) Be It known to you that for the purpose of increasing the 
religious merit of (my) mother and father, and for the sake of the attainment 
of the desired reward both in this world and in the next, I have confirmed, 
as brahma-deya, with libation of water, (the enjoyment of) one hundred and 
surty pdddvarttas. on the northern boundary of the Jy6shth&naka villagse bo- 
Imiging to the Akshasaraka-prave^ya in the Hastavapra-Aarasiii, which had 
(formerly) been and are (still) being enjoyed (by the donee^®), for (the 
benefit of) the resident of the same village, (namely,) the Brahman Madhava 
of the igunaka gotra, a student of the Chhandoga School,— to last for the same 
time as the moon, sun, ocean, earth, the rivers and mountains, to be enjoyed 
by the succession of his sons and sons’ sons, — with (?) Smbara, with gold (and) 
ddeya, with bhuta, vata, and (’) surety of holdmg (pratyaya), 

(LI. 17-19.) Wherefore, no enquiry should be made or obstruction caused 
(to hin\) by any one, while he is, according to the proper conditions of a 
brtdima-deya, enjoying, cultivating, or assigning (it to others). And this 
our gift should be assented to by those bom in our lineage, and by future 
good kings, bearing in mind that power is perishable, the life of man is 
uncertain, and that the reward of a gift of land is common. And he who 

a# Read Uap. 

Over SIT there is a peculiar agn, the meaning of which is not apparent. II 
think it ia upadhmaniya . — ^Ed ] 

28 Read q. =» Read «» Read 

•1 Read 

32 The construction of line 14 is somewhat confused ; it is not clear who the 
donee was, or who, at the time of the grant, was in posseaSion of the land which 
is the obiept of the grant. As it st^ds, tbei text does jiot ipak^ any sense; my 
pendenng is oonjectujal, 



confiscates it or assetnts to its confiscation incurs the guilt of the five great 
sins together with the minor ones 

(LI 20-22.) There are also two verses sung by Vyasa abouti this. 

[Here follow two of the customary verses ] 

(L. 23.) The diitaka is the pratihana Mammaka. (I>ated the) 5th 
(lithi) of the dark (foltnight) of Vai^akha (m the) year 200 7 

(L. 24 ) (This is) the sign-manual of me MahSrSja Dhruvasena (I.J 
Written by Kikkaka. 


This plate, which contains only the opeiung portion of a land-grant of 
the Maitraka king Dhruvasena I , is mscnbed on one side only and measures 
roughly 10|" broad by 6Y' high The {109} edges axe just slightly raised, 
in order to protect the wnting, which is in a state of excellent preservation 
throughout The letters, which are deeply inased, show through on the 
leverse side of the plate. The engraving is well eixecuted The plate has a 
pair of holes bored at two adjacent comers and intended for receiving the 
ring and seal, which are missing. Its weight is 56 tolas It contains fifteen 
linis at writing. The letters are of the penod to which the plate refers itself, 
and of the tj^ie met with on other plates of the Maitraka dynasty. In short, 
this record is exactly like any of the large number of grants of Dhmvasena I, 
that have latterly been brouiht to light. A detailed description of the chai- 
acters, language and orthography of these plates, or even an English render- 
ing of the text, seems superfluous. We may take it for granted that the 
diitaka qf this grant was the pratthSra Mammaka, and the writer Kikkaka. 

The grant was issued from ValaWii by the Mahasdmanta MokSfaja 
DhmvasSna [I,] to the Brahmana Santifarman of the Atreya gotra, [a student 
of] the Vaji[saneya] School and a residoit of Nagaraka, dther bestowing 
upon him or confiiitung him in the possession of one hundred pOdavarttas of 
land on the south-eastern boundary of the village of Bhadrepika, situated in 

I am unable to identify Bhadrauika. Nagaraka is prcJiably Vadnagar, 
the home of the NSgar BiShmans 

Plate B. 

12 • • • 

From the odginal plate, and a set of estampages. 

Vp to this the teort is practically identical with the text of the Klitfim Plate 
of Ittimvaa^ L (dated 206), published above, Vol, XI, pp. lOSflE. In 1. 
6, read ^t^pSd>^htp7<njfaffUi° fpr ""t^pabhipta^j^atna ; and Manvodkta for "^4*^ 



14 *wi ia< i ^wt ^F> raiiiw<!q 




Since writing the above I have come across a new Valabhi plate con- 
taining the ccmduding portion of a grant of Dhmvasena dated in sam. 206, 
about which I should like to add a few iwords in continuation of the above 
note on the Bhavnagar plates This new plate was placed m my hands for 
decipherment by Mr J. C CHATirauEE, Dhaxmadhyakdia (Secretary m the 
Ecclesiastical Department) to the Government of His Highness the Gaikwar 
of Baroda It was sent to him, he told me, ofi&cially from Kathiawad for 
decipherment : that is all that I could elicit from him regarding its previous 
history The plate is 111 inches long by 61 inches broad , the edges are raised 
to protect the writing, which is m a state of perfect preservation , and tlie 
characters belong to the period to which the plate refers itself : m one word, 
the grant is similar in every respect to the records of the Valaldii king that 
have hitherto came to hght fllOj The inscription is one of Maharaja Dhru- 
vasena [I ] and records the grant of a villa^ (of which the name must have 
occurred in the rmssmg portion of the grant and is therefore now lost) to a 
Brahmaipa named Rotgharmtra of the Vrajagana gotra, a student of the 
Chhanddga School, and resident of Siimhapura, for the maintenance of certain 
sacrifices. The grant is dated sarh 200 6, Ai§vma 'iukla 3. The sathvat year, 
when referred to the Valabhi ara, yields A.!i>, (206 + 319) 525. The diitaka 
was Mammaka, and the writer Kikkaka, as usual 

The <Mily pomt worthy of notice in this grant is the vllage-name Siihha- 
pura, which is maitioned in it as the residence of the grantee. It is tempt- 
ing to identify it with Sihor in the east of the Kathiawad peninsula, a juncticai 
on the Bhavanagar-Wadhwan Railway, not far from Vala, the ancient Valabhi. 


TEXT ,i 

1 EQnava-kshlti-sarit-parwata-sthiti-samakillnaih putra-pautr-Snvaya- 

bhdjyairh bali- 

2 diaru-vaiSvadev-adyanaith kriyanaJh samutsaippan-artthadi Sidiha- 

pura-vfistavya brShmaaiarRotghamitraya 


The lest of the inscnpbon is trussing 
91 profu the original plate and a stet of impressions. 



3 Vrajagana-sa-gotraya (Ch) Chhandoga-sa-brahmacharine brahma- 

dayarh nispishtam [*] yat5 = sy = ochitaya brahma- 

4 deya-sthitya bhuihjatab krishatah pradi^atah = karshSpayrataS —tiia. 

na kais = chit = svalpapy = abadha vicharana va 

5 karyy = lasmad-va!rhlajairi= agurami^’S-nripatibhig = ch i= anityany 

= aiSvairyyapy = asthiram manushyarii ch = avekshya samanyatii 

6 bhumi-dana-phalam = avagachchhadbhir — ayam = asmad-dayo -- 

numantavyo ya§ = ch = achchhindy&d = S.chchhidyamanatfa 
V = anumodet 

7 sa pafichabhir = mahSpatakais '=■ s-opaplatakais = sarhyuktas = 

sySd = api ch = atra VySsa glitan 96kau 

8 bhavatah d*] shashtiifa [*] varsha-sahasrani! svargge modati 

bhumidahil*] achchhettS. ch = anuman© cha tany =eva narake 

9 vaset (|| *] sva-dattafn para-datta[|m‘',] = wa, yo hareta vasun- 

dharath [IM gavirh 4ata-sahasrasya hantu[h*] prapn ti 

10 kilbishamlll*,] = iti sva-hasto mama mahfiraja-Dhruvasenasya 

I||*i] dutakah pratihiaraMammakah til*] 

11 Iikhitam Kikkakena [||■^]| saih 200 6 ASvayuja su 3 [||*'] 

*• [Read Ed] 



It is many decades since the discovery of certain Brahmi inscnptions, be- 
longing to the early centuries of the Christian era, led to the recogmton of a 
dynasty of kings claiming to belong to the Satavahana family or tnbe » The 
inscriptions themselves yield very httle direct information regarding the home 
or the sphere of influence of this family of ruhng prmces. But it was soon 
discovered that the names that were gleaned from these inscriptions (agreeing 
in part with those inscribed on certain coins which were almost simultaneous- 
ly brought to li^t) had their analogues in the names of certain other kings 
who, in the Puramc geneologies, are called Andhras ® And as there was no a 
prim reason why the Satavahanas should not be Andhras, scholars, who were 
assiduoudy collatmg every scrap of information bearing on the history of this 
dynasty, were not slow in availmg themselves of this help, meagre as it was ; 
straightway they adopted the Puramc nomenclature and labelled these kings 
Andhras. That proved to be a good starting-pomt for further speculation 
regarding their history. The name Andhra suggested at once a connection 
with the land of the Andhras, which roughly corresponds to the modem 
Tdugu country, and with the Andhra people, of whom there are notices in 
yet older mscnptions, and in the chromcles of fordgn travellers and histi>- 
rians. Round this fragile frame-work, connecting the Satavahanas with the 
Andhras, was built up an edifice of Satavahana-tAndhra history, and a dog- 
matic version of it (for mstance, the account of the Andhra dynasty in Vin- 
cent Smith’s Early History of IndiaY was placed before the puUlic as an 
authentic account of the fortunes of {22} the family. So long as a historical 
narration does not contradict the few directions that are enclosed by a limited 
number of indisputable records, and contains no inherent improbabilities, there 
is every chance of its passing muster and of its being accepted as a fact. Thus 
it comes about that after its formulation it has nevar occurred to any one to 
challenge the Andhra affinities of the Satavahanas set forth in the text books, 
based as it is on the flimsiest of foundations Here and there it is quite 

* [AtmUs BORI, 1. 21-42]. 

1 The following list gives all the mscnptions that can with greater or less 
degree of certamty be ascribed to these kings Luders, Airt of Brahmi Inscriptions 
(.Epigiapkia Ihffica, VoL X, Appendix) Nos 346, 987, 994, 1001, 1002, 1024, 1100, 
1105, 1106, 1112-18, 1122-26, 1141, 1144, 1146, 1147, 1248, 1279, 1340, 1341. 

2 Sra Rafson’s Catttlosue of the Coins of the Andhra Dynasty, etc., (B. M. 
1908), Introductirm ; and PahCIthb’s The Parana Text iff the Dynasties of the Kali 
age (Oxford 1913) under the Andhras. 

8 Early History of India (1914), p. 206 ff. 


epigbAphic studies 

patent to the reader of these accounts, that the author is stretching a point ; 
but a little latitude is always allowed to the constructive histjonan for the 
play of his tm^ginatinn. Now and again he conaes across an. unsupported 
assertion that on reflection may be found to fall considerably short of the 
truth : as, for mstanm, Vincent Smith’s view that Sri-Kakulam (on the lower 
course of the Krsna) was the capital of these ‘ Andhra ’ kings, a view which 
IS based on a piece of thoroughly worthless evidence, as is shown by P. T 
Srinivas Ivengar in his article entitled ‘ Misconceptions about the Andhras,’ * 
But there are yet larger disciepanaes which only a ngorous and unbiassed 
ecammation of the entire material— epigraphic, historical, nunusmatic, and 
l^endary—will disclose, such as I had to undertake in connection with the 
editing of a new inscription of Vasisthiputra Sn-Puluniavi,® discovered in 1915 
in a litfle-known village in the Bellan District In that connection I was 
confronted with the question whether the facts of the Satavahana' history 
necessarily demanded that the home of the Satavahanas should be placed, 
as has hitherto been done, in (what was later called) the Andhradesa The 
results of the mvesbgation and the successive steps by which I amved at them 
are set forth in the sequel 

The Hira-Hadagalli copper-plate grant® which was issued by the Pallava 
Siva-Skandavarman for the purpose of confirming and enlarging a donation 
made by the Mahmaja Bappasvanun to certam Brahmanas, inadentally sup- 
plies us with a very interesting place-name, to wit, Sataham-rattha, which rattha 
(province) is there said to include the settlement named CiUareka, of which 
ihe Bmhman donees were bhojakas (le. probably, freeholders), 

{[23J who edited the grant, did not succeed m identifying the localities men- 
tioned in it.'' Indeed the villages remain still unidaitified. But we can now 
claim to be able to locate the province named in the grant, which we tuSI 
enabled to do on account ofi the discovery, already mentioned, of an msenp- 
tion mcised in the reign of Sn-Pulumavi, which contains another place-name 
having evident afiinities with the name imder reference This insenption* of 
Pulumavi (referred to m the sequel as the Myakadoni inscription) is biased 
on a boulder situated midway between the villages of Myakadoni and Cinna- 
Kadabum at a distance of about eight miles frean Adoni in the Bellari Dis- 
trict, The object of the inscription is to record the sinking of a reservoir by 
a certain householder (gahapotika) , who was resident of the village of Ve- 
pudaka situated in the province (janapada) called Satavaharu-hara, a name 
which at once recalls to our mind the Satahani-rattha of the copper-plate 
grant mentiemed above. The inscribed boulder is a perfectly sure landmark 
fixing a point atuated in the ancient province (.janapada) of Satavahani-hara, 

* Indian Antiquary, 1913, pp. 276 ff 

» Episraphia Indka, Vol. XIV, p. 153 ft. 

« LUdehs’ Ust No. 1200 t Epigrapkkt Indica, Vol, I, p. 2ff. 

s For transcript and tranalalion see the Appendix to this article. 


a point which roughly corresponds to the modem Adoni, the headquarters of 
a Taluqa of the same name m the Bdlan District 

From the copper-plates themselves we can elidt nothing regarding the 
situation of the province mentioned m them , for, as already stated, the vil- 
lage-names also, which might have given us an indication regarding the loca- 
tion of the province containing them, have hitherto defied all attempts at 
identification But we shall bring to bear on this question the scrap of topo- 
graphical information gleaned from the Myakadoni inscription, namely, that 
the modem Adorn was included in the province of Satavahani-hara, and see 
what result the comparison yields. 

At the time when the plate were edited the signification of the word 
Satahani-rattha had remamed obscure. But now it is quite evident that 
this name stands m close agreement with the Satavahani-hara of the inscrip- 
tion. Sataham and Satavaham mean one and the same thing : the former 
is only a corruption of the latter Both plactj-names are derived -from the 
tribal name of the so|-called [24]} Andhra kmgs, who, as was stated above^ 
all claimed to belong to the family or tribe of the Satavahanas (or Sfitava- 
hanas ® "Whether the oreos Satavahani-hara and Satahani-rattha are identi- 
cal or not IS a question more difficult to answer Rattha (rostra) is geneially 
used to denote a province^ realm, empire, or even a country (as in Maharastra, 
Surastra, etc.) The word hara (ie., ahara), on the other hand, which is 
often used indiscnminatdy to denote a district or country, applies, as a rule, 
to a smaller territorial division than what we understand by a kingdom or 
province; that is, it generally refers only to a distnct Apparently there- 
fore, the geographical names in the two records, as thqy stand, do not cor- 
respond exactly with each otiier But the precise connotation of the word 
ahara in the Myakadoni iiracription appears to be given by its being speci- 
fically called a janopada ; and a janapada, I think, very nearly conv^s the 
same meaning as rMha (rostra). Moreover, it should be remembered tiiaf 
Hira-Hadagalli, the village where the plates were purchased, is also situated 
m the Bdlari district ; and it is probably not a matter of mere ccMnddence 
that the find-place of the copper-plate charter and the spot where the Myak- 
adoni inscription stands, should both be included within the small compass 
of the Bellari district If the find-place of the grant may now be supposed to 
be not far distant from the object of the grant (which is by no means invari- 
ably the case), then the modern Hira-Hadagalli may be taken to mark ap- 
proximately another point atuated within, or in the nei^bourhood of, Sata- 
vahanihara-Satahanirattba. We should at any rate not go far wrong in 
assuming that the two place-names are toms which, if not synonymous, were 
the names of areas situated within or dose to each other. 

» In the sequel I have adopted the spdling Satavahana, 



patent to the reader of these acsounts, that the author is stretching a point ; 
but a httle latitude is always allowed to the constructive histlonan for the 
play of his imagination. Now and again he comes across an unsupported 
assertion that on reflection may be found to fall considerably short of the 
truth : as, for instance, Vincent Smith’s view that Sn-Kakulam (on the lower 
course of the Krsna) was the capital of these ‘ Andhra’ kings, a view which 
IS based oo a piece of thoroughly worthless evidence, as is shown by P T 
Srinivas Iyengar in his article entitled ‘ Misconceptions about the Andhras.’ * 
But there are yet larger discr^ancies which only a rigorous and unbiassed 
examination of the entire material — epigraphic, historical, numismatic, and 
legendary — ^will disclose, such as I had to undertake m connection with the 
editing of a new inscription of Vasisthiputra Sn-Pulumavi,® discovered in 1915 
in a little-known village in the Behan Distnct. In that connection I was 
confronted with the question whether the facts of the Satavahana history 
necessanly demanded that the home of the Satavahanas should be placed, 
as has hitherto been done, m (what was later called) the Andhradesa. The 
results of the investigation and the successive steps by which I arrived at them 
are set forth in the sequel. 

The Hira-Hadagalli copper-plate giant® which was issued by the Pallava 
Siva-Skandavarman for the purpose of confirming and enlargmg a donation 
made by the Maharaja Bappasvanun to certain Brahmanas, inadentally sup- 
phra us with a very interesting place-name, to wit, Sataham-rattha, which rattha 
(province) is there said to include the settlement named CiUareka, of which 
ihe Brahman donees were bhojakas (le. probably, frediolders) . Buhler, 
{[23} who edited the grant, did not succeed m ideitifying the locahties men- 
tioned in It ^ Indeed the villages remain still umdentified. But we can now 
claim to be able to locate the province named m the grant, whiiii we tarS 
enabled to do on account of the discovery, already mentioned, of an inscrip- 
tion inased in the reign of Sn-Pulumavi, which contains another place-name 
having evident affinities with the name under reference. This inscnption® of 
Pulumavi ( refen ed to in the sequel as the Myakadom mscription) is incised 
on a boulder situated midway between the villages of Myakadoni and Cinna- 
Kadaburu at a distance of about ei^t miles from Adorn in the Bdlari Dis- 
trict The object of the inscnption is to record the sinking of a reservoir by 
a certain householder (.gahatvMtka), who was resident of die village of Ve- 
pudaka situated in the province (.janapada) called Satavaham-hara, a name 
which at once recalls to our mind the Satahani-rattha of the copper-plate 
grant mentioned above. The inscribed boulder is a perfectly sure landmark 
fixing a point situated m the ancient province (jampada) of Satavahani-hara, 

* Indian Antiquary, 1913, pp 276 ff 

® Epigre^kia Indka, Vol. XTV, p, 153 ff. 

a Ludess’ List Na 1200. t Epigrapkia Indtca, Vd. I, p. Zff. 

*> For transcript and translation see the Appendix to this article. 


a point which rou^y corresponds to the modem Adom^ the headquarters of 
a Taluqa of the same name in the Bellari District 

From the copper-plates themsdves we can elicit nothing regarding the 
situation of the province mentioned in them , for, as already stated, the vil- 
lage-names also, which might have given us an indication regardmg the loca- 
tion of the province containing them, have hitherto defied all attempts at 
identification. But we shall bring to bear on this question the scrap of topo- 
graphical information gleaned from the Myakadoni inscription, namdy, that 
the modem Adoni was included in the provmce of Satavaham-hara, and see 
what result the comparison yields. 

At the time when the plate were edited the signification of the word 
Satahani-rattha had ranained obscure But now it is quite evident tha t 
this name stands in dose agreement with the Satavaham-hara of the inscnp- 
tion. Satahani and Satavahani mean one and the same thing ; the former 
IS only a corruption of the latter Both placd-names are derived 'from the 
tribal name of the sof-called (24} Andhra kings, who, as was stated above^ 
all daimed to belong to the family or tribe of the Satavahanas (or SStavS.- 
hanas® Whether the areas Satavahani-hara and Sataham-rattha are identi- 
cal or not IS a question more difficult to answer Raitha (rostra) is generally 
used to denote a province, realm, empire, or even a country (as in Maharastra, 
Surastra, etc) The word kora (i.e, ahara), on the other hand, which is 
often used mdisciiminately to denote a district or country, applies, as a rule, 
to a smaller territonal division than what we understand by a kingdom or 
province ; that is, it generally refers only to a district. Apparently there- 
fore, the geographical names in the two records, as they stand, do not cor- 
respond exactly with each other. But the precise connotation of the word 
ahara in the Myakadoni inscription appears to be given by its bdng sped- 
fically called a janapada ; and a janapada, I think, very nearly conveys the 
same meaning as iMha (rvistra). Moreover, it should be remembered that 
Hira-Hadagalli, the village where the plates were purchased, is also situated 
in the Bdlari distnct ; and it is probably not a matter of mere coinddence 
that the find-place of the copper-plate charter and the spot where the Myak- 
adoni inscription stands, should both be induded within the small ctunpass 
of the Bdlari distnct If the find-place of the grant may now be supposed to 
be not far distant from the object of the grant (which is by no means invari- 
ably the case), then the modern Hira-Hadagalli may be taken to mark ap- 
prcfitimatdy another point situated within, or in the neighbourhood of, Sata- 
vahanihara-Satahanirattha. We diould at any rate not go far wrong in 
assuming that the two place-names are terms which, if not ^onymous, were 
the names of areas situated within or dose to each other. 

9 In the sequel I have adopted the spelling Stavahana, 



However that may be, we have here an unquestionable proof of the 
existenoe of a proviso called after the Satavahanas, a country that extended 
at least as far west as Adorn, and perhaps even further up to the western 
boundaiy of the modem Bellari district. The province must evidently have 
been so called on account of some mtunate connection between the land and 
the people concerned Of what nature can this relation be? A glance at 
any map of [25J ancient India will supply the answer It will sho-w us how 
common at one tune the practice of naming the country after its early in- 
habitants was The Matsya lend their name to the Matsya country, the 
Magadhas to Magadha, the Kosalas to Kosala, the Ratthas (or Rastrakutas) 
to Maharastra The kingdoms of the Kalingas, the Colas, the Pand 3 ms and 
the Keralaputras, which owe their names to the early inhabitants of those 
countries, pieserve these names up to quite modem times Examples of this 
usage may be indefinitely multiplied, not only from the history of India but 
from that of other countnes as well The mtunate connection referred to 
above must, therefore, be one of original occupation. And we may, on groimd 
of the evidence so far considered, not unreasonably surmise that the country 
had taken its name from the Satavahanas because these people had smce very 
early tunes, probably already in the pre-histonc penod, established them- 
selves there As an alternative solution it may, conformably to the theory 
of the Andhra, ongin of the Satavahanas, be proposed that the early kings 
of this dynasty had proceeded thither from that home m (what m mediaeval 
times was known as) the Andhradesa, made themselves master of the countrv, 
andl renamed the land of thar conquest after themselves. The answer to it 
is that there is no precedent to justify such an assumption The Satavahanas 
liad carried thar conquest far and wide in India, both to the north and soulh 
of the Narmada. In the Nasik inscnptioni® of the Bala-Sri, Sn-Satakarm 
is called the king of Surastra, Aparanta, Vidarbha, Akaravanti , many other 
lands and mountains are named besides , but all of them retain thar names 
known to us from other sources The Satavahanas had not ventured to altei 
the names of the countnes of their conquest These considerations lead us to 
look upon the province known then as Satavahani-hara (or m later times as 
Satahani rattha) as the original habitat of the Satavahanas, a conclusion 
which, I fear, will not find favour with scholars as it mihtates strongly against 
the accepted view on the subject. It is customary to interpret the history of 
the Satavahanas as though it were a migration from the east to the west 
Thus Vincent {26} Smith, “ apparently voicing the unanimous verdict of 
scholars on the point, ‘says : ‘ The Andhras [i.e. the Satavahanas] .... 
set up as an independent power under the governmeit of a kmg named 
Simuka. The new dynasty extended its sway with such extraordinary [italics 
mine] rapidity that, in the reign of the second kmg, Kndma (Kanha) , the 
town of Nasik;, near the source of the Godavari in the western ghats, was 

11 Early History, p. 207. 

LttoERS' list No. 1123, 


included in the Andhra dominions, which thus stretched across India.' It may 
not be superfluous to point out that these two heroes, Simuka and Tfanha, 

‘ whom eternal night holds unwept and unhonoured,’ owe the resuscitation of 
their glones purely to the inventive genius of a historian. For, if the truth 
be told, nothing more is known about these kings beyond the bare fact that 
the name of the one occurs m an inscription at Nanaghat and of the other 
at Nasik ! 

Owing to the heterodox nature of the above conclusicm re^rding the 
home of the Satavahanas which is arrived at merely from a consideration of 
certain topographical information supplied by two inscriptions, it wll be 
necessary for me first to refute the established theory of the ‘ Andhra ’ affini- 
ties of the Satavahanas from an independent standpoint This I shall do 
by showing that the hypothesis is m entire disagreement with the other known 
facts about the Satavahana kmgs, facts which fall into their correct perspec- 
tive only when we assume that the Satavahanas formed a tribe which was 
origmally not even remotely connected with the Andhra country.*^* 

In order to avoid eveiy misunderstanding on the point I must state at 
the outset that I am not here concerned with the larger question of the home 
of the Andhra people My contention is merely that the home of Simuka, 
Krsna and their descendants was not the Andhradesa, which is commonly and 
tightly identified with the country of the basms of the Godavan and Krsna. The 
attempt pyj to seek (as one wnter^^ does) m the passage from the AUareya 
Btahmana m which the Andhras, Pundras, Sabaras, and Puhndas are referred 
to as Dasyu tabes living on the fringe of the Aiyan avilisation, an mdication 
of the Andhras’ being a Vmdhyan tnbe appears to me to be a viaous circle 
For, neither do we know the habitat — ^at least not the original habitat— of 
the Pundras, Sabaras, and Pulindas, nor have we any mformation as to the 
exact limits of Aryan domination m those days The Andhras have, on the 
other hand, in literature, been far crftener associated with the Kalingas, Colas 
and Pandyas and as these appear to have from time immemorial ocaipied 
approximately the same geographical positions m which we find them at the 
dawn of history, it is not unlikely that the Andhras might have done like- 
wise, Everythmg points to their having occupied from very early times the 
same place as m the time of Varahamihira^^-^ and Hiuen Tsiang^*'' (ca. a.d. 

Here I emphasise the point that the arguments set forth in the sequel to 
discredit the Puranic statement are abiolutely tndependent of the above hypothesis 
that SatavahanihararSatahanirattha was the home of the Satavahanas and that it 
lay outside the Andhradesa , in no way do they imply or necessitate its araumption. 

IS P. T. Snmvas Ivengae, Indian Antiquary, 1913, pp 276 ff. 

See for instance Sabhaparva, Chapter 31, and the passage quoted by Sir 
R. G. Bhandaescar from the Ramayaita on p. 4 of the Early History of the Deccan. 

1*"- Brhatsamhita (ed. Kern)' Chapter 14, v. 8. 

«i» Beal, Buddhut Records of the Western World (Triibner’s Oriental Series, 
Popular Edition) , u 217 ff. 



630, when for the firtt time we come across a definite statement r^arding the 
siluabon and extent of the Andhra country), but one cannot be absolutely 
certam. It is true that in the Asoka inscriptions the Andhras are once (Edict 
XIII} placed in a class different from their Dravidian neighbours, and reck- 
oned with the Bhojas, Pitinikas, and Pulmdas. That does not hdp us further. 
For, the habitat of the Bhojas is unknown, that of the Rtmikas doubtful, 
and of the Pulmdas (which appears to be a name used vaguely for savage 
hiU-tnbes) uncertain Moreover it ^ould seem that the principle underlying 
the grouping of these people in the passage under consideration is adminis- 
trative, — other words, one depending on the degree of independence enjoy- 
ed by the rulere of these countnes — and not topographical.^'' The classifica- 
tion is therefore for our purpose without significance 

{28} We shall now turn our attaition to the genesis of the assumption 
that the Satavahanas are Andhras The very earliest source that ecameers 
the Satavahanas with the Andhras is the Puramc literature ; and it may be 
added that outside the PUranas tho^e is not a smgle mdependent authoiity 
that asserts, or in any way implies, this relation One thinks confusedly of 
Greek authorities m this connection It may therefore be emphatically stated 
that nothing that the Greek histonans have to say on the matter can be 
looked upon as lending colour to the Puramc statement, as any one who 
takes the trouble of exanunmg the original text may without difficulty con- 
vince hmiself. The fact of the matter is that those passages from Greek 
authors which explicitly mention the Andhrai country and the Andhra people 
contain no reference to the Satavahanas , while on the other hand, those in 
which certam Satavahana kings are mentioned teaches us that the Satavahana 
kings have nothing to say about the Andhras. It is only constructive history 
which teaches us that the Satavahana kings mentioned in one place are the 
same as the Andhras spoken of m another, a fiction at the bottom of which 
lies the very same Puramc authority. Of the Chmese pilgrims, I believe, <Hily 
Hiuen Tsiang describes the Andhra country at length ; but he has nothing to 
say about any one of the so-called Andhra kings, an omission which is imma- 
lenal as it may satisfactonly be explained on the ground that the Buddhist 
pilgrim visited the country more than three centunesi after the extmction of 
tliis line of kings The Andhra (Telugu) literature is also easily disposed of. 
T^e leam with surprise that the Andhras thsnsdves have preserved no 
memory^ not m any shape whatsoever, of those illustrious ‘Andhra’ kinga 
whose dominions stretched across India and who had succeeded in hnirfmg 
sway over a large part of Southern India for the unusually prolonged period 
of nearly four centuries 

Regarding the Puranic material itself a word may be added. ‘ A glance 
at the fomudable hst of veriati leettones published with the text of extra r t s 

See RApstw’s Catdl(y&ie, p. xvi, foot-note 2, 



collected by Pargiter^® will convince anyone of the futility of trying to get 
a reliable and in every way satisfactory text. I shall not dwell on the variant 
lists of kings, nor on the divergent figures given for the lengths of their 
reign , nor, lastly, on the ingenious attempts made by scholars to reconcile 
these discrepancies,^', as it is not necessary for my purpose It is amusing, 
however, to note that there is no unanunity among the Puranas even as to 
the name to be applied to this hne of km^ Some of the Puranas call these 
kings Andhras , others call them Andhrabhrtyas ; and there are others still 
that call them by both names.!® The majority of the Puranas, however, 
distmguish between the Andhras and the Andhrabhrtyas and state that the 
Andhrabhrtyas succeeded the Andhras , most of them agree m applying the 
term bhrtya to them, implying that these kmgs were ongmally feudatories 
of a paramount power. The hopeless confusion on the point whether the 
Satavahanas were Andhras or Andhrabhrtyas will be made still more apparent 
when it is remembered that while, on the one hand. Sir Ramakrishna Bhan- 
DARKAR calls these kmgs Andhrabhrtyas throughout his account of that dy- 
nasty in the Early Hilary of the Dekkhan,^^ on the other hand, Vmcent 
Smith never so much as mentions that name m his Early History Rapson 
is undecided Sometimes he uses the term Andhrabhrtya to denote the main 
branch of the Satavahana family, and sometimes the feudatcmes of these 1 
Thus p. XV footnote 1, while referrmg to the Satavahana dynasty he says 
that It ‘ was called also Andhrabhitya or Satavahana ' : and subsequently 
while speakingj about oertain coins from the Chitaldnig district, he observes 
that these may ' have been struck by the feudatories of the Andhras {Andhra- 
bhrtyah) iwho rose into power in the western and southem districts after the 
reign of Sri-Yajna”'*! 

Now if the term Andhrabhrtya' is taken to mean ' dependents or feuda- 
tories of the Andhras,’ there is evidently a deal of difference m meaning bet- 
ween the epithets Andhra and Andhrabhrtya . the feudatories of the Andhras 
need not necessarily be Andhras But the ambiguity of the espression coveis 
the difference of meaning, as j[30]} the compound may be equally wdl treated 
as a Karmadharaya (as is done by Ramaknshna)®* and thai taken to 
mpqn ‘ Andhras who were feudatories,' naturally of some other power. How- 
ever, tHis ingenious way out of the difficulty does not appear to have apjieal- 
ed to other scholars and with due deference to the veteran Orientalist, it 
must indie^jd be admitted that, though from the point of view of the gram- 

1* Dynasties of the KaU Age, pp. 35 ff. 

Sir R. G Bbandarkae, Early History of the Deacon, (^884), p. 23 u. 
See Farsiter Dynasties of the iKoK A0, 1. c. 

!» Sea p. 17, and passim. See the Index. 

Rapson's Catalosue, p Ixxxiu, foot-note 2. 
ss Opf art., p. 18. 



marian the scduhon proposed by Sir Ramakrishtia is un.excQ)ti<Miable, it would 
be more natural to treat the compound as a Dqsendent Determinative (Tat- 
purusa), especially in regard to the parallel phrase Simgabhrtya*^^ applied to 
the Kanvas and occurring also in the Puramc genealogies, in which Sir Rania- 
krishna=' also sees a pointed reference to the Kanvas being the servants of 
the Sungas 

Having established that the theory of the Andhra ccxinechon of the 
Satavahanas rests upon the uncorroborated, and at the same time equivocal, 
statement of the Puranas, we shall now turn our attention to other facts of 
tlieir history with a view to ascertain if the statement of the Puranas is borne 
out by these facts. 

We shall in the first instance turn to the epigrapAiic mateiial The first 
thing we notice is that m none of the inscnptions (about two dozen m num- 
ber) engraved during the regime of these kings is there any reference to their 
alleged affimty with the Andhras In these records they are invanably re- 
ferred to by then kula name Satavahana or a variant of it The Hathigumplia 
inscription^^ of Kharavela, the Giraar inscnption^* of Rudradaman, and the 
Talagunda inscription'^’' of the Kadamba Kakutsthavarman, which are among 
the contemporary records mentioning various Satavahana km^, never refer 
to them as Andhras If the Andhra nation, whichl wasi reputed ‘ to possess 
a military force second only to that at the command of the king of the Prasii 
Chandragupta Maurya,' [31} had evoked the admiration even of foreign 
chroniders,*® one naturally wonddrs at this conspiracy of silence r^arding 
this illustrious lineage on the part of the contemporary documents The m- 
evitable cotidusKm might still be ignored by some critics on account of the 
negative character of the evidence on which it rests. Let us therefore also 
examine some positive eividence and see what that yields The Hathigumpha 
inscription of Kharavela tells us that the Kalinga king, ‘iwithoiut entertain- 
mg any fear of Satakani, sent a large army to the west,^<> evidently with a view 
to mvade the dominions of his powerful enemy. Were we now to take a ma p 
of India in hand and try to explain why Kharavda should send a large army 
to the west when his enemy, who is allied to be the tang of the Andhra 
country, lay due south of him, the incongruity of the Andhra theory will 
become manifest. Provided that Kharavela was at war with the Andhia 
king, the fate of the mvader who mdulged in the quixotic attempt of sending 
his army to the west, would not have been a matter worthy of glorifinfjtio Ti 
In any such attempt the invader would inevitably have exposed his flank to 

Vayu Parana . catvarah Sungabkrtyas te nrpah Kanvayana dvijah. 

“ Op. at., p 24 25 LiJDERtf List No. 1345. 

LOdbss' List No. 966. 

Ed Kdelhorn, Bp. Ind VIII, p 24ff. 

« Elliot, Ckmts of Southern India, pp 9fif. 
a® LllDBHS' List No. 1345. 



a murderous attack all along the contiguous frontier of the enemy king ; and 
it will not be senously suggested that he could have advanced by a circuitous 
northern route to attack an outlying western possession of his southern neigh- 
bour. That would have been equally disastrous The expedition of Khara- 
vda, I mamtam, can only be explained on the assumption that, in Kharavela’s 
tune at least, the kingdom of the Satavahanas lay entirely, or at any rate 
principally, to the west of the Kalinga country 

The next point to be considered m this connection is the geographical 
distribution of the inscriptions of the Satavahanas. By far the largest num- 
ber of their uiscnptions is at Nasik, where there are eight records eigraved 
in the reigns of different kin^ • there are five at Kanhen, three at Karle, 
two at Amaravati, one large and several very short ones at Nanaghat and 
one each at Bhelsa, Myakadoni, Cma, and Kodavolu Of the nearly two 
dozen records mentioned here there are exactly fmr from the Andhradesa ! 
Thus the (32'} topographical distnhution of the inscnptions hitherto discovered 
supports, m my opinion, emphatically the view that the centre of gravity of 
the power of the Satavahanas lay in the west of India. 

A study of the distribution of these mscnptions in point of tune relative 
to the locahty is still more instrucfave The following is a list f approximately 
dironological) of the mscnptions of the Satavahana dynasty, giving the find 
places and the regnal years. All but three (viz. Nos. 6, 20, and 21) of these 
inscriptions are such as ather were engraved by order of a ruling pnnce of 
this dynasty or refer themselves to the reign of one of them , the three inscnp- 
tions which are mentioned as exceptions were engraved by persons in tlie 
employ of these kmgs. 


Name of king or queen. 


Regnal year 


(The numbers refer to 
LuDERS* List, Ep. Ind. 
VqL X, App,) 

1. Simuka Satavahana 

2. Krana^o 

3. Sri-Satakami , Devi 

Nafpanika, queen 
of Sri-Satakami 
and mother of 
Vedi-sri andSal^- 
srimat (Hakusiri}. 




(No date) 

No. 1113 An image of 
king Simuka. 

No. 1144. 

Nos, 1112, 1,114, and 
1117. Along with these 
sure to be taken the 
inscribed imagea of 
Kiunara Satavahana 
and others SI 

KiSUa appears to have ruled before Sri-Satakami. See Rapson, Op. cil„ 

p. xix. 

These have not been enumerated separately, as it ia not certain whether the 
persona concerned had acLu^y reigned 






Regnal year 

(The numbers refer to 

Name of king or queen 

LUDHRS’ List, Ep. Ind, 

Vol X, App) 

4. Sri Satakami 




No. 346. Ptobably an 
early long His place in 



the chronological list 
is uncertam 

5. Madhaiiputm-svami 

[ Kanheri 


No. 1001. The name is 
read as Sakasena, 

which is probably a 
jmslection.32 The first 
part of this word la 
probably sin. Chro- 

nological place doubt- 

6, Do 

1 « 

(Year lost) 

No. 1002 See the re- 


marksl against No. 5. 

r34T 7. Gautamiputra 

' Nasik 


No 1125 


1 [81 

.. 1105 

8. Do. 


9. Do. 



„ 1126. 

10. Vasisthiputra - Sn- 





„ 1122 

11. Do. 



12. Do. 




13. Do 



Eptgraphia Indtca, VoL 

XIV, p. 153 ff. 

14. Do 



Nos 1123, 1124 

15. Do 


No 1124 

16. Do 



„ 1106 

17. Do, 


(Year lost) 


18. Vaasthipatra - Sri- 


• • • » 

No. 994 8* The king 


mentioned therein is 
not to be identified 
with the previous kmg. 

Ghromological place 


r35l 19. Sri-Sivamaka 


* « - 

No, 1279 


20. Gairt.irTputra Sri- 




Vajna Saln'vtiini 

^ r-+'- Sri- 



„ 1024. 

V,' 'l.l *• ‘.llu* 1 

22 Gautaimputra Sn- 

Qna (Krishna 


„ 1340. 

Yajna Satakanu 


(Year lost) 

23. G— Sri- 

^ 1 1 i 1 I 


„ 987. 

24. Vasisthiputra Sri- 
Chanda (or Chan- 


? 13 

„ 1341. 

dra) &itakami 

32 Rapsdn's {Op. cit., p, xlvu) correction is extremely doubtful. 

33 For transcript and translation of this new inscnption sJee the Appendix at 
the end 

3< The connection m which the queen is named in this inscripticm is not 
appcraot. Vasisthiputra-Sd-Satakarm is coantnonly identified with the famous 
VB^sthiputra-Sri*Puluniavi, see Rapson, Op. cit., p h. But to me it is extremely 
doubtful whether one and the same king can be assumed to bear both the names 
Satakami and Pulumavi ; it to me that the terms are mutually exduwve. 


£36} Of course the discovery of new in&ciiptions in unexpected quarters 
might scatter to the winds all the fine theories based upon our present knov/- 
ledge. But the above list, as it stands, it seems to me, supports my conten- 
tion in an unmistakable manner The earliest inscriptions are ail from 
Western India , and at is not until the time of Vasisthiputra Sn Pulumavi 
the [SiroJ-Ptolemaios of Ptolemy Cca ad 150), that we meet with an 
mscnpition of any king of this dynasty from the Andhradesa Bhelsa, it may 
ba remarked, stands high in the list , then follow Amaravati, Cina t^nsluia 
Dist ) and Kodavolu The Satavahanas had undoubtedly oveirun and con- 
quered the Andhra country , but their earliest possessions were, the insciip- 
tions seem to tell us, in Western India I have suggested above that the 
tribe to which this line of kings belonged must be regarded as the autochi hons 
of the inland province named Satavahani-hara, a tract of land which has 
not yet been identified with certainty but iwhich lay, probably, considerably 
to the west of the Andhra country We may now proceed onel step, furtiier 
and say that the Satavahanas, who were settled in Satavahani-hara, had 
made themselves masters of the northern portion of the western Ghats, and 
even subdued some part of Malava before turning their attention to the con- 
quest of the Andhradesa. 

It might have been expected that m that highly interesting inscnption^"’ 
from the Nasik Pandu Lena of Bala-sn, the mother of Gautanuputra, we 
should find defimte information raiding the extent of the Satavahana domi- 
nions Unfortunately the topographical information to be gleaned from A 
IS very meagre, partly on accooint.of our mability to identify satisfactorily 
all the place-names mentioned in it, and partly because the terms in which 
the relation between the lands named and their overlord is couched leave it 
doubtful whether the list is intended to represent the extent of Gautamiputra’s 
entire p|ossessions or that of his conquests merely. Moreover, on account of 
the ambiguity attaching to the term Daksmapatha, which m its widest signi- 
ficance includes the whole of the penmsula south of the Vmdhyas, and m its 
narrowest the country between the Narmada on the north and ‘ a variable 
line £37} along the course of the Krsna exclusive of the provinces lying to 
the extreme the geographical data of the inscription remains for us 

enveloped in an impenetrable mist of vagueness* Indeed it is not possible to 
identify with certamty even the country round Nasik (where the inscription 
itself) with any of the countnes named m the record unless it be included 
in the capacious folds of Daksmapatha , it is probably intended to be conveyed 
by the mountain name Sahya (Western Ghats) It may, however, be re- 
marked that the kingdoms recognising the suzerainty of this Satavahana king, 
so far as they can be identified, are all in Western India, and include not a 

3s Luders' List No. 1123 

36 Bhandarkar, Early History of the Deccan, pp. 1 ff. 



Single country definitely identifiable with any portion of the Andhradesa, — 
again with the ^ception of the Daksinapatha, which, as remarked above, 
may indeed imply any part, or even the whole, of the Indian Penmsula south 
of the Vmdhyas 

A word may be added here regarding the language of these inscnptions, 
which IS either Sansknt or scane form of Prakrit , noi Satavahana mscnption 
written m a Dravidian tongue has yet come to hght. This fact has the ap- 
pearance of supjporting my contention that the Satavahanas were not Dravi- 
dians Such is however not the case The earliest Telugu epigraphic record 
known, I understand, is an inscription of the Eastern Calukya kmg Jayasiraha 
I and dates from the sixth century ad It may therefore be that, at the 
period under consideration, Telugu was not yet raised to the dignity of a 
literary dialect, a fact which would sufficioitly account for the use of Prakrit 
01 Sanskrit in the mscripCons of the Satavahanas even though the latter had 
been unquestionahly Dravidians. 

The conclusions regardrag the home and the movement of the Sata- 
vahanas to which we are led by a consideration of the epigraphic material 
die corroborated in a remarkable manner by the numismatic evidence. The 
earliest coins of this dynasty, we find, were all picked up in Western India. 
If we cqien Rapson’s Catalogue of the coins of the Andhra dynasty, etc., the 
very first com on the raster {383 will be seen to be that of Stl Sata (identi- 
fied by Rapson with Sri-Satakami of the Nanaghat mscnption) of which 
we are told that it shows the Malava fabric and was picked up ini Wefetem 
India Coin No 2 which is of the same type as com No 1 was found in 
Western India ; No. 3 was also picked up in Western India, but it is doubtful 
it it belongs to this series Then follows a coin (No. 4) which will be dis- 
cussed presently The subsequent coins (Nos 5—32) are like Nos 1 — 3 
from Western India. Thus all the early coins (Nos. 1 — ^32) with the ex- 
ception of No 4 were picked up m Western India and presumably were cur- 
tent only there The exception is a com of— Vira found m the Andhradesa 
Why Rapson should have placed this com here more than anywhere else 
is a mystery, unless the reason be supposed to he in R«>son’s reluctance to 
leave the Andhradesa entirely unpres^ted in the early peinod of the Sata- 
vahana regime The obverse of the com is perfectly plain , on tiie reverse 
is figured lion standing 1 The inscription has not beai completely decipher- 
ed So far as it can be made out it reads . rano [ — — ] vaC>)(rasa, and is 
therefore unlike any Satavahana legend. Vmcent Smith,®® we are told by 
Rapson, ‘attributes these coins provisicxially to Gautamiputra SriYajna 
Satakami ’ ! Further comment seems unnecessary. We can proceed to tiie 
next lot of corns (Nos. 33 — 46), which are stated to be found in the Andhia- 

Rapson, Op. dt. p xxxv, foot-note 4 
®« ZDMG, 1903, p. 625. 



desa ; they are hesitatingly ascribed by Rapson to a king for whose name the 
alternatives Sakasada and Sakasena are proposed. A careful study of Rap- 
sons’ remarks concerning the inscriptions of these coins will repay the trouble 
About coins No 42 — 46 he confesses that it is not always easy to distinguish 
them from certain coins attributed to Sri-Pulumavi That leaves a balance 
of only ten corns of this doubtful species ; in none of them has the iiKcription 
been completdy read. Of these ten, only four coins show ‘ uncertam traces ’ 
of three or four akshoras each, the inscnption on the remainmg six being 
completely illegible With this datum Rapson reads the legend as being 
either Sakasada or Sakasena It will, I think, be generally admitted that 
Sakasada is an incredible name ; and I may add there are reasons for believ- 
ing {39} that the Sakasena of certain Kanheri inscnptions on the stiength of 
which Rapson puts forth the other conjectural reading, is probably a mis , 
lection , however, I do not wish to add a third conjectural reading, especially 
as I personally have seen neither the coin nor the inscription With the 
material at our disposal, namely four coins with uncertain traces of three or 
four ^llables on each and five similar coins which cannot be distinguished 
from certain others attributed to Pulumavi, it would be hazardous, to say 
the least, to attanpt identification In any case it will have to be admitted 
that there is nothing to show that the coins m question have to be attributed 
to any of the early Satavahanas Thus it becomes evident that of the coins 
from the Andhiadesa, the earliest that can with assurance be assigned to a 
known king of this dynasty are those of Vasisthiputra Sn-Pulumavi (Rapson’s 
Catalogue Nos 88 — 89), he is the same ku^ of whose inscriptions it was 
noticed above that they are the earliest of all Satavahana inscriptions to be found 
in the Andhra country, a significant fact which, it should be noted is in entire 
harmony with my surmise arrived at on independait evidence that the fidd 
of activity of the early Satavahanas was confined to the west of India. 

At this stage it may be craiveniently pointed out that the Jainas have 
preserved a very dear recollection of the connection of the early Satavahanas 
with Western India For in Jama legend, Paithan (the anaent Pratisthana) 
on the Godavari in His Exalted Hi^iness the Nizam’s Dominions, is the 
capital of Salivahana and his son Saktikumara, who have been nghtly identi- 
fied witii Sri Satakami and his son Haku-siri of the Nanaghat inscriptions 
We know, moreover, that Paithan continued to be the capital of the Sata- 
vahanas, at least until the time of Sri-Pulumavi. The Greek geographer 
Ptolemy is, as is well known, our authority for this supposition. His words 
ivii, 1. 82) Boithana, Bastletus [sire] — tol4matou can only be taken to con- 
vey’ that Paithan was the capital of Sn-Pulumavi. 

The arguments set forth above and the coodusion to be drawn from 
them may be briefly summarized as follows ; 

so Rapson, Op. cit p. xxxix, 



{^40} Ihe Myakadoni inscnption of the time of Srjf-Pulumavi mentionp 
the janapada Satavahani-hara, and the Hira-Hadagalh copper-plate grant of 
the Pallava Siva-Skandavarman supplies us with the place-name Satahani- 
rattha Tliese places, which are possibly identical, point definitely to the 
existence of a province or kingdom situated in the neighbourhood of the 
modem Bellary District, and named after the Satavahanas, which must have 
been so called on account of its bemg the onginal habitat of this tribei. The 
latter conclusion is at variance with the orthodox view that the Andhradesa 
Is the home of the Satavahanas This view, however, appears to be based 
merely upon the fact that m the Puranic genealogies the kings Simuka, Krsna, 
and others succeeding them are called Andhras — It was also pointed out 
that while some of the Puranas styled these kings Andhras, there are others 
which called them Andhrabhrtyas The latter term is commonly regarded as 
synonymous with Andhra, but may clearly also mean the ‘ feudatories of the 
Andhras,’ which is quite a different thing — The Puranas are, it was sub- 
mitted, our only authority for the assumption of the Andhra ongin of tlie 
kings in question , there is nothing in the writings either of the Greek or 
of the Chinese chroniclers that may be adduced in support of this — ^The 
oft-quoted passage from the Greek geographer Rolemy has undoubtedly 
rightly been interpreted to mean that Paithan (the ancient Pratisthana) was 
the capital of the Satavahana Sn-Pulumavi This statement not only finds 
partial corroboration in the Jama legend that makes Pratisthana the capital 
of the king Salivahana (Satavahan) and his son Saktikumara, but fits in 
better with what we may surmise regardmg the habitat and activity of the 
Satavahanas from a consideration of the geographical distribution of thi>ir 
inscriptions and the provenance of their coins We are, therefore, led to con- 
clude that the connection of this dynasty of kings with the Andhradesa has 
been considerably antedated , properly regarded it is the result of a migra- 
tion jrom the west to the east , the home of the Satavahanas has to be placed 
in the south-western parts of the Dekkan plateau On this a<wiimptinn it 
becomes intdhgible why Kharavela, who boasts that he was not afraid of 
Satakami (evidently a Satavahana king), should send a large army to the 
west : on the same assumption it becomes still clearer why the Maharathis 
{41 J (a western tribe) should be often dosdy connected by family ties 
with the ruling princes of this house. When we place the capital of the 
Satavahanas at Paithan, we can also understand better why the Buddhist 
caityas at Nasik, Nanaghat, Kanheri, and Karle (which on this hypothesis 
would naturally lie m thar home provinces) should monopolise the patro- 
nage of these princes to the exclusicm of Amaravati, the classic twtha of the 
Andhra Buddhists, situated in the heart of the Andhradesa. 

If we admit the above ccaiclusion (le, if thei activity of the early Sata- 

*0 Contra Rapson (.Op. dt. p xxi) and others. 



vahana kings be regarded as being restricted to the south-western and western 
comer of the Deccan plateau, and if they aie supposed to have noi connec- 
tion with the land which, in later times, is called the Andhradesa), how are 
we to reconcile with this view the Puranic statement that the Satavahanas 
were Andhras’ There are two possible ways of answenng the question 
If the Puranic statement be literally correct and the Satavahanas have to be 
looked upon as bdonging to the tribe of the Andhras, then we must assume 
either that this branch had separated itself early from the mam stock of the 
Andhras (which was settled in the region of the ddtas of the Godavari and 
the Krsna), even before the time of Simuka and Satakami, and settled in 
the west or that the Andhras themselves had at first occupied the part of 
the plateau surrounding the province named Satavahani-hara, and then mig- 
lated before the histone epoch, from that centre, towards the west and to- 
wards the east But it is aftdr all conceivable that the Satavahanas may 
not have been Andhras , and it is quite probable — ^this is the alternative expla- 
nation referred to above — ^that the correct designation of this dynasty ’s really 
Andhrabhrtya (which was later wrongly abbreviated by some of the Puranas 
into Andhra), a germ of genuine history being preserved m the appdlation 
Andhra hhr fya In this case, however, the latter compound is properly re- 
garded as a Sasthi-Tatpumsa, and taken to mean ‘the feudatories of the 
Andhras ’ For there is nothing improbable in the assumption that the foun- 
ders of Satavahana dynasty were originally the va&sals of the Andhra sove- 
reigns, of whom it nnay with assurance be affirmed that at or about the time 
of the rise of the Satavahanas they were the most powerful potentates in the 




1 [Si]dha[m] [ll”] Rano Satavahanam S[i]ri-Pulum[a]-visa sava 8 

hema [2] diva 1 

2 [masa] mahasenapat[i]sa Khamdafna]kaaa janapade S[a]-tavaharu 


3. KuHiaradatasa game Vepurake vathavena gahapatikena 
[Komjtanam [Sambelna 

4. talakam khanitam [11'*^]. 


Success ! On the first day of the second [fortnight of] winter m tlie 
eighth year (of the reign) of Siri-Pulumavi King of the Satavaha^ (family) 
the reservoir was sunk by the householder {gdhapattka) readent of the 
village of Vepuraka bdoogmg to the Captain {gumtka) Kumaradata (Kuma- 
radatta). m the province Umapada) of Satavahani-hara belonging to tip 
Great General (.mc^asenapati) Khamdanaka (Skandanaga), 


Ever unce the providential discovery by Sir John Marshall of the 
writing hidden beneath the thick crust of vermilion coveting the shaft of the 
Garuda column of Beaiagar, that little Praknt record has engaged the at- 
tention of a number of distinguished scholare interested in Indian history,^ 
and their patient research has succeeded, it may now be confidently asserted, 
in eluadating completely the import of the inscription. The scholarly edition 
of the text from the pen of Dr. (now Professor) J Ph. Vogel, published in 
an issue of the Amml of the Director-General of Archaeology in India, ^ 
contains a succinct review of the readings and interpretations proposed by 
different scholars in their articles and notes on the subject, and in this 
edition the labours of previous workers in the field may be said to have 
culminated. Respecting the investigations of these scholars it may be re- 
marked that the historical interest centring round the name of the 
Indian king Antialkidas, and the fact of the conversion of a Greek ambas- 
sador in India to the cult of Vasudeva, nreponderates in them so far over 
every other consideration, that the language and textual critician of the 
inscription have not received that share of attentiwi and scrutiny which they 
deserve. It may, therefore, be permitted to me to supply the want by add- 
ing to what has been already accomplished a few observations on this topic, 
and inadentally to elucidate a new aspect of this— m many senses— umque 

The text (A)! given by Dr. Vo($l in the article just alluded to reads 
as follows : t 

1 Devadevasa Valsudejvasa Garudadhvaje ayam 

2. Imnte. ...Heliodorm Bhaga- 

3. vatena Diyasa putrena Takhasilakena 
[(to}4. Yonadutena agatena maharajasa 

5. Amtalikitasa upa[im]ta sakasa[ih] rand 

6. KSsiputasa Blagabhadrasa tratarasa 

7. vasena catudasina rajena vadhatn^asa^ 

* — - - - - ^ - . . _ 

* [Anmh BOm 1. 59-66] 

r See LOmss’ List of Brakm Imaipttom Nos. 669 and 670 (Appendix pp 63, 
64 and 176). To the liteature given there is to be added, as far as I know, only ; 
1912 Vogel, Annual Report of the Architological Survey of India, 1908-9, pp. 126 ff. 
and Plate ; and 191'4 Rapson, Ancient India, p lB6f. and Plate VI. 

-» Above, 1908-9, p. 126 ff. 

t (Dlacntical marics restored from onginal typescript found in Sukthankar’s 
A nufecro— Ed,} 

* Aa the following ranaifcs refer chiefly to this inscription only, it is vmneces- 
sary to reproduce here the Gatha (comnioaly designated as text B) incised below it, 



Now, in the first place, with respect to the reading kdrite (in line 2 of 
the text), which is adopted by most of the previous interpreters and accept- 
ed by Eh Vogel, it may be remarked that it is not altogether free from 
objections According to it, in this dialect the Nominative Smgular of a 
thematic stem would end in — e, a conclusion which is a priori inadmissible 
m the case of a Western dialect Furthermore, the facsimile appended to the 
vanous editions of the inscription all show quite distinctly that the final 
syllable of the word m question is to, as correctly read by Eh Bloch in 
the editio princeps •* the two inked impressions filed in the office of the 
Archaeological Supenntendent, Western Circle, which were examined by me, 
also diow on their reverse sides a deep dent corresponding to the sign of 
length (dhkara) in that dkshara There can be, therefore, no doubt that 
the short honzontal stroke appended jto the nght of the vertical was inten- 
tionally mcised by the engraver, and the correct reading is kdnto In this 
instance the medial ,0 is marked by a zig-zag sign, which repjroduces quite 
faithfully the form of the imhal o Other examples of this usage in the 
msciiptioti before us are do in Heliodorem (Ime 2) and no m rvmo (line 5) ; 
but an instance of the later cursive form, with the bars at a umform height, 
is supphed by Yo in Yonadiitem (Ime 4) the two forms occur here side by 
side as m other early Brahmi mscnptions “ The to of kdnto bemg admit- 
ted, we are constrained to look upon dhvaje (hne 1) as a lapsus phnnse 
for dhvajo , the small stroke correfepondmg to the length which should have 
been added to the loop of ja has either been left out entirely or is not 
traceable on the impresaons. 

In order to accommodate the final vowel of kdnto to that of dhvaje, 
it has been suggested that the small slanting stroke across the sign £61} of 
length in the to of kdrUo is meant to ddete that length To this I have to 
say that to my mind the chances of the engraver’s having omitted to add 
the sign of length to the loop of ja (m which case, the slanting stroke across 
to will have to be looked upon as a fortuitous mark on the stroke) and his 
having incised to by mistake for te are about equally balanced. But in re- 
gard to the locality of the inscription, I am inclined to look upon je as IJhe 
incorrect syllable. The question could have been finally settled by refer- 
ence to another Nominative Singular of a thematic stem in the same ms- 
cription : unfortunately there is no other instance of it in text A But it 
may be noted that the text B which appears to be intimately connected with 
A, supplies at least one dear example of the requisite form, namely^ apfa- 
mddo^ The vowel marks of the final syllable of dama and cdga are not 

* Jour. Roy, As, Soc., 1900, p 1055. 

e BuHLER, Indische Palxographte, p 37. 

B The bar acrass the top of da clearly marks the subsidiary o in that akshara. 
It 18 true that Dr Vogel read damo, cigo and apramada, which readings serve my 
purpose equally well, but see the transcript of the text of the G&tha by Drs. Venis 
and Barnett, Jour Roy As. Soc , 1909, 1910. 



clear : this much is, however, certain that neither of them ends in -e Thus, 
if the inscnptions A and B are to be looked upon as being linguistically 
connected with each other, the reading kdrito and the correction of dhvaje 
to dhvajo become inevitable. 

After karito some scholars read in the succeeding gap a woid ^\a] and 
render the latter by Skt iha In early Brahmi inscnptions the reality of an 
initial i is always attended with some degree of uncertainty, especially when 
the inscribed stone is abraded as ours is , moreover the impressions that I 
have been able to examine fail to show any definite traces of the succeeding 
syllable Thus every reading of a word intervening between kdnio and 
Heliodorma must be looked upon as open to question With reference to 
the identification of \a with iha, I must remark that there are no certain 
cases of the loss of an mtcrvocabhc h in Praknt known to me one {[62} 
would have to restore, in my opinion, tha (less likely idha) rather than %a. 
But, it appears to me, that the arbitrary insertion of a word like iha is in 
this instance utterly unjustified For, what is meant by saying that ' Garuda 

" The alleged examples {Jout Roy Soc 1909, p. 1089) of ta from in- 
K'nptions all occur in the Kharosthi versions of Asoka's Edicts four times m the 
Shahbazgarhi, and once in the Mansera version Two analogous instances of 
' even a shorter form ‘ are ated from the Girnar inscnption The references are 
avS follows — 

Shahbazgarhi Ed VI, 1, 6 %a ca 

{62} Shahbazgarhi Ed IX, 20 ta-loka 

M IX, 20 [h]ta or "possibly” ia 

M IX, 24 to-loka or "possibly" hialoka 

Mansera VI, 31 ia-ra 

Girnar XI, 4 i-loka 

XIII, 12 t-lokika 

The two instances from Gimar need not have been quoted m this connection 
as, m this mstance, there is no question ot any loss of h In t4oka, t is the de- 
monstrative pronoun, and t-^oka or i-loktka is equivalent to " this world ** or " of 
this world” With regard to the examples with uncertain A, it will have to be ad- 
mitted that they cannot be looked upon as cert<dn instances of the alleged loss 
Thus the evidence adduced reduces itself to the statement that out of the scores of 
mstanccvs in which the woid iha or its Prakrit equivalents occurs in Asoka Edicts 
BOhler had read m Kharoshthi records m three presumably certam cases ia in the 
sense of here The evidence is not overwhelming , and in view of the extreme 
similarity of the signs for * and Ai m the Kharoshthi alphabet, I hold that a re- 
examinatic«i of the inscription m sttu will be necessary before a definite opimon 
be pronounced one way or the other. For the present I shall content myself 
^ remarking that the forms current in the northern 'dialects appear to have been 
hia and lAo. The form kia is, however, not an instance of the total disappearance 
of intervocalic A. What has happened here is the foUowmg. Iha forms a 
doublet AiAa with the same adsenpt A as in Asoka hevam for evam. Then, as the 
rules of Indian phonetics do not permit aspirates in consecutive syllables (Giass- 
^nns Ride) one of them (here the second for obvious reasons) is merged : heni^ 
ae fom hia ... .. .As to the allied instances of the loss of intervocalic h in literar^ 
Fnaknt, see Pischel, Grammatik der Praktit^prarhent para 226, 



colmnn was made here by (the order of) Heliodoros ’ Heliodoros was sure- 
ly not a manufacturer of Garuda columns ! Clearly iha can only he used 
with a word like sthdfnta . with karita it is essentially mappropriate 

In line 3 it has been customary to read the last word Takhastlakena. 
The correct readmg dearly is Takhasilakena, as there is no room for the 
sign of length (marked in this inscription with a prominent honzontal stroke) 
between the syllables la and he, whidi, as it is, almost touch each other 
The irregular projection to the right of the vertical of la is no more intend- 
ed to sigmfy the length [633 ® similar appearance in the very first 

akshara of the same line, which for that reason is not read as vS but as it 
ought to be VIZ va. Grammatically the form Takhasilakena is indefensible, 
at least m the sense ‘reddent of Tal^sila,’ whereas Takhasilakena (for 
Takkhasilakem) is a tadbhava of the San^it Taksc^Uaka or TSksaitlaka 
(the form commonly found in Buddhist Sanskrit Literature), m the same 
way as Mdthtcra(ka) is formed from Mathnrd : cf. the Kahka to Panini 
IV, 3, 93 

A glaring fact which has b^ it appears to me, totally overtoloked by 
previous editors is the pecuhar construction of the only sentence of which 
the digraph consists. The sequence of the words in the sentence is indeed 
so thoroughly un-Indian that I cannot account for the fact that ncaie of the 
distmguished mdologues who have commented <m the inscription has re- 
marked upon it The sequence of words in Sanskrit or Prakrit, it must be 
admitted, is, on account of the synthetic form of the languages, theoretically 
quite free. Notwithstandmg, practice has prescribed certain rules regulat- 
ing the rdative position of words in the sentence, which are not departed! 
from tn prosa without sufficient reason ® As a grammatical principle it may 
safely be laid down that the qualifying word vtsesana almost invariably 
precedes the word which it qualifies visdfya the Genitive stands before the 
word which it governs, the dependent noun before the preposition, the object 
and other adjuncts before the verb, etc In particular, participial adjectives, 
should, unless used predicativdy, precede the word they qualify. If we test 
our sentence m the hght of this rule, it wil be found to stand in fiagrant 
contradiction to it. Now one may think what one likes about lllhe position 
of the words Bhdgavatena, Diyasa putrena, and which are 

in apposition to Hehodorena. But there are two instances of wrong se- 
quence in this sentence which must be considered very remarkable, and they 
are firstly, the use of agotena after Y ortadStena, and secondly, the p 06 iti(»i 
of vadhamdnastt (following on BkSgabhadrasa, of whidi it is an attribute) 
at the very end of the sentence A Praknt sentence ending with a presentt 
partidple is an anomaly. Indeed, it was at one time thought that [64} the 

B See Hennann Jacobi’s remarics on tbe subject in the introduction to his 
Ausgewahlte Breahhmgen in MdhdrSffri (leipzig, 1886), 



seventh line was not the last line of the inscription ; but we are now assured 
that the whole of the inscription is before us, and no further lines have been 
worn out or lost. 

It will be remarked that the participle vadhamanasa and dgatena 
occupy the samp position as finite verbs in dependent clauses Does it' not 
appear as though the latter mode of sentence construction is peculiar to the 
style of the writer, a mode which is foreign to the genius of the language in 
which he was writing ^ If so, how are we to account for this anomaly ^ The 
key to the situation is, I think, supplied by the phrase Yonadmena dgatena 
Mahardjasa Amtalikitasa upamtd (lines 4 and 5), the construction of which 
becomes intelligible only when we remark that it is word for word a Prakrit 
rendering of the Greek upd presbeuton elthdntos pard ton megdbasiUos 
Antalkidou (cf. Hdl., 1. 3, 9 , An., 1, 1, 5), which in Gre^ is quite correct 
and natural. The sequence of the words of the above Prakrit phrase being 
settled by this consideration, the chiastic use of the prepositions places the 
noun governed by sakasam (line 7) in the position actually occupied by it, 
with the result that the adjectival phrase beginning with vasena (line 7) is 
pushed to the very end of the sentence. To change the epigraph into cotr 
rect Prakrit we must set it upside down. Prakrit idiom would necessitate 
the following ;» 

22 23 24 25 18 21 19 

vasena catudasena rdjena vadhamanasa ratio trdtarasa kasiputasa 
20 17 14 15 16 13 

Bhdgabhadrasa sdkdsath nuihardjasa AahtcAikitasa upomta dgatena 

12 7 

YonadUtena. . Heliodorena etc Before passing on to other questions I would 
here draw attention to the striking similarity in point of construction between 
Bhdgabhadrasa . rdjena vadhamanasa and the Gredc expression k&rou 
bdsSe&ontos {Cyra regnante') without wishing to say definitely whether the 
latter has been of any moment in determining the shaping of the unfamiliar 
Prakrit phrase. 

If I am nght in explaining the anomalies of construction referred to 
above as Hellenisms, or in other words if we assume that we must look to the 
Greek syntox and Greek mode of thou^t for an explanation of the abnorma- 
lities of constraction, and peculiarities of phraseology (if not of all, at 
least of some of them)', then some other anomalies become also clear. (Zcmsi- 
der, for instance, the use of the biruda trdtdxasa t^otfxos) with the name of 
Bhfigabhadra. I am not aware of any king of Indian extraction who had 
assumed that title, and it would be surprising if Bhagabhadra, whoever he 
may be, provided he was an Indian) had done so Its use would cause no 
surprise, however, if the writer of the inscnption were a Greek ; for, in that 
case, we could very well understand his tackmg on to the name of an Indian 

» The numeral above eadi word indicates its position in the ongmal sentence. 



prince a royal title which was commonly assumed by Greek kings of his time. 
Consider again the position of the verb kdrito. Although the sequence Oa- 
rudadhvajo ayam karito Hehodatem is not faulty, — not by any means : there 
are instances of it to be found m other inscnptions as well — ^the normal Prakrit 
construction would be \ayam Garudadhvajo Heliodorma kanto But in the 
Greek sentence the position of the verb (which would be a verbum fimtum) 
between the subject and the object would call forth no comment Further, 
for Prakrit I find the sequence Hehodorena . Diyasa PutUna somewhat 
harsh , but agam, it would be the most natural mode of expression for a 
Greek who is in the habit of saying parAsatis, e tou kmou meter. These 
facts bear out in an unequivocal manner the correctness of the above hypo- 
thesis explaming the position occupied by agatena and vadhammasa in the 
sentence, which was the starting point of our enquiry. 

The writer of the mscnption who thus on my showing must have been 
a Greek may have been Heliodoros himself For, it does not require any 
great stretch of imagination to bdieve that this worthy Greek who was a 
convert to the Vasudevic cult of Bhakti worship, and who had gone to the 
length of erecting a Garuda ccJumn in honour of VSsudeva, had also ac- 
quired a working knowledge of the local Indian dialect which was, perhaps^ 
for him the language of the scnptures alsa 

The last line of the inscription is dear as regards its import, but not 
quite so in point of its constmction Various attempts have been made to 
explain it, but none of them is entirdy satisfactory. One of the chief diffi- 
culties has been the phrase rcjena vadhamanasa, for, it was not realised by 
any of the previous interpreters that the expression is not merdy gram- 
matically irreproachable, but even highly idiomatic.^'’ The root vrdh (oftai 
combined with distya) is regularly construed with the Instrumental to ex- 
press gratulation, as in the phrase \[d%styS\ mahdrdjo vifoyena vardhate, 
which means literally ‘ Your (His) Maj'esty [fortunatdy] prospers with vic- 
tory.'’ Thus -rdjena vadhamand&a means 'prosperous with reigning,’ or as 
it IS commonly and correctly trandated ‘gloiioudy reigning.* A real diffi- 
culty, however, is presented by the expression vasena chedudctsena. The In- 
strument of Time has m Sanskrit (and I presume also in Prakrit) a speaal 
significancdi whidi, however, cannot have been intended here. On the other 
hand, as this oosus obltqus. does not exist in the Gredr language, that source 
of explanation seems to be closed in this instance. It may be that the writer 
of the epigraph had not fully grasped the force of the Indian Instrumental 
of Time, and wrongly employed it here , or may it be that it is an instance 
of anomalous attraction by the following rajewt? Perhaps some scholar who 

10 T^ius rdiena is wrongly put equal to Sst rajye by Dr. VOGEI,, above 1908-9, 
p. 128. 

Panini, II, 3, 6 ; apavarge tjtiya. 



IS better acquainted with the Greek idiom than I am, may be able to assign 
a reason even for this apparent anomaly 

I append a transcript prepared by me from a set of excellent impres- 
sions fUed in the office of the Archaeological Superintendent, Western Circle, 
embodying the corrections in the reading of the text proposed above. 

Text A. 

1 Devadevasa Va[sude]vasa Garudadhvaje*-* ayam 

2. kinto Heliodoreija Bhaga- 

3 vatena Diyasa putreiija Takhaalakena 

4. Yonadutena agatena maharajasa 

5. Afiraltalifc [ijtasa upamta sakasaih ratio 

6. K^putasa^^ Bhagabhadrasa tratarasa • 

7. vasena cha[tu]dasena lajena vadhamanasa 

Read G<tru4<idhvaj». 

A gap large enou^ to contain two (iksharas. 

Peii^ps we have to read the second member of the compound as putrasa, 
unless the irregular dejsesrion bdow the sign ta is caused by an ^ihrasirm 

of the Bto^ at the pewt. It is worthy of note, boweva, that in thia inscriptiion 
the subscript r is retained in every other case in which it appears aa the 
member of a ligature, to wit putretta (line 3), Bhagabhadrasa and trStarasa (line 6), 


In the Md of the palaeography of Northern India the beginning of the 
sixth century of the Christian era is marked by the advent of a new epoch of 
alphabet, which is chiefly charactensed by the acute angles that show them- 
selves at the right or lower ends of letters, as well as by the wedges which 
are supenmposed on the tops of the vertical or slanting lines, and which is, 
tlierefore, vanously styled as the ‘ Acute-angled ’ or ‘ Nail-headed ’ alphabet.^ 
The epigraphic documents of the penod from the sixth to at least the begm- 
nmg of the eighth caitury form an unbroken record of the use of this alphabet 
in Central and Northern India. This type was in course of time supplanted 
by a nval alphabet ; and the characters of the inscriptions of the next cen- 
tury present the incipient stages of the Northern Nflgari, the fully devdoped 
forms of whidi may be seen in the Kaulthem (Miraj State) copper-plale 
grants of the C&lukya king Vikramaditya V (ad 1009) The distinguishing 
feature of this type is the substitution of horizontal covering strdces in place 
of the wedges, and nght angles in place of the acute angles of the previous 
variety The general course of the evolution of Nagari out of the acute- 
angled alphabet is evidoit enough ,® but the determination of the actual 
period of transition is a problem which naturally presents certain difflculties 
The earliest forms of the transition alphabet are differentiated from those of 
its predecessor merely by the flattening of the above-mentioned wedges. These 
forms are supplied by the characters of the Muliai (Central Provinces) plates 
of the Raslrakuta Nandaraja Yuddhasura dated m the year corresponding to 
A.D. 708-9, and other inscriptions of a later date. In other respects the 
characters of the Multai plates link on directly to the acute-angled {310} 
alphabet, and Buhler was therefote perfectly nght m looking upon them as 
the last phase of this variety* The distmctive pecuhanty of Nagari, it 
must be emphasised, lies in the widerang of the acute angles mto right angles 
as. wdl as the addition of the flat top strdce which, so to say, covers the 
entire breadth of the letters at their upper end. Both these characteristics 
are unmistakably manifest m the Kaijheri inscriptions of the Silahara feuda- 
tories Pullaisakti and Kapardm II ® Thus upto the begmning of the eighth 
century (a.d 708 : the MultM plates) the acute-angled alphabet was stiU 
current in Northern India ; on the other hand, as we see from the Ka(aheri 

* [i?. G. Bkandarkar Commemoration Volume 309-22.] 

1 BOhier, Jndische Pdaeographie, p. 49. 

® Indian Anitguary, Vol 16, pp. 15 ff. 

* Bohleh, op. cit., pp. 50 f. * BuHlER, op. cit., p 50. 

® Inscnption Nos. 15 and 43 ; see Jndiati Antiouary, Vol. 13, p 135. 




inscnptions (A d. 851 and 877), Nagari had come to be used as an epigiaphic 
alphabet by the middle of the ninth century 

We might therefore set ourselves the question, at what period does this 
change set m? Are there any records® writtai m Nagari, of a date eailier 
than the above-mentioned Kanheri inscription? Buhler was inclined to 
hup^iose that the northern Nagari was in use at least since the beginning of 
the ei^th century ’’ The inscnptions which appear to lend support indirectly 
to this view form the following senes *'(1) the Samangad grant’ of Uie 
Rastra-J[31l}-ku(a Dantidurga (Buhler, Paljeographic Tables, Plate IV, 
Col. XXII), bearing a date corresponding to ad 754, from Western India ; 
(ii) the Dighva-Dubaul! plate*® of Mahendrapala I, and (in) the Bengal 
Asiatic Soaety’s plate of VinSyakapala** (of the Impenal Pratihara dynasty), 
believed by Buhler to be dated in the years corresponding to a d 761 and 
794-5 respectively, from Northern India. We shall presently return to a 
detailed consideration of the Samangad grant, but let us first examine the 
characters of the other two records a little more closely. It is true enough 
that we find here distinctly Nagari charactenstics, eg (i) in the above-men- 
tioned right angles of gha (PI IV, col XXI ; 10), pa, (coi XXI ; 27), ma 
(col. XXIII; 31), ya (col XXI, XXIII, 32), and sa (col XXIII; 37), 
(li) in the flat top stroke of pa (col XXIII , 27), 'ma (col. XXIII ; 31), 
ya (col. XXIII , 32). sa (col. XXIII ; 37), and sa (cd XXIII ; 38). Of 
special interest is the form of ja (col XXI, XXIII ; 14) . In the Dighva- 
Dubauli plate it has entirely lost its original charactenstic element of three 
parallel bars as, for instance, m the specimen^* quoted in the immediatdy 
preceding column (XX ; 14) of the same table , but in the other grant the 
transformation is still more stnkmg The lower portion of the letter forms 

» I Wish to exclude therefrom the signatures or facsimiles of signatures of 
Gurjara princes on the a^jper-dates of Kaira (of ad. 628 and 633), of Dabhoi 
(Ad. 642) of (AD. 705) and of ICavi (ad. 736) appended to texts written 

in a southern alftobet. From these royal sign-manuals it does not necessanly fol- 
low that the alphabet m question waS used at that penod commonly for epigraphic 

’ BOhler, op. dt., p 51. 

* The earliest of these is dated AD 754 But Buhler argued that as an 
inscription from the Kanarese country, viz. the Pattadkal pillar mscription of Kirti- 
vartnan II {Epigre^hia Indica, VoL 3, pp. Iff), which was caused to be mdsed 
by a Brahman from Northern India, shows the mixture oil the Nagari and acute- 
angled letters, we could assunae the use of Nagari since the beginning of the eighth 
century. For my part, I must say, I have not been able to trace any Nagari letters 
in this inscription. 

• Edted by Fueet, Indiem Anttqtiary, Vol. 11, p, 105 
” See Flebt, Indam Antitjuaiy, Vol. 15, p. 106 
w Jttdiem Antiqtuxry, VoL 15, p. 140. 
la MultS plates ; see Fuost, Indian Antiquary, Vol. 18, p. 231. 


a clearly developed double curve, while the (originally horizontal) middle 
bar ia all but veittical. 

Now with re^rd to these alleged specimens of early Nagaii the follow- 
ing is to be noted As far as the alphabet of the Dighva-Dubauli plate is 
concet'np^, the term NBgari seems to me to be applied to it with doubtful 
propnetyA® Buhler has classed it ngjitly as an instance of the acute-angled 
variety The absence of the covenng stroke in gka (col. XXI ; 10), pa 
(col. XXI ; 27), ma (col XXI ; 31), ya {[312]} (col. XXI ; 32), sa (cok 
XXI ; 37), and sia (col XXI ; 38)' shows that it has not passed the transition 
stage; while the sporadic acute angles, for instance, m ma (col. XXI ; 31) 
and perhaps sa (col XXI , 37) entitle it to be considered a phase of tiie 
acute-angled alphabet This is, however, only a matter of nomaiclature. 
No such doubt can be enterfeuned with regard to the copper-plate grant of 
the Pratihara king VinSyakapala of Mahodaya, which is certainly one of the 
earliest instances (if, indeed, not the earliest instance) of the use of Nagar! 
forms for epigraphic puiposes as far as Northern and Central India are con- 
cerned. None the less is the conclusion of Buhler regarding the phase mark- 
ed by these two plates in the evolutton of Nigari wrong , the reason is that 
both these records were considerably antedated by him The mistake lay in 
the erroneous interpretation put in his time upon the syllables sathvatsro 
forming part of the date of the record. Here the ligature tsfo (as was first 
pointed out by Dr Hoernije) must be looked upon as consisting of the t 
of samvat and sro, which latter apparently stands for the multiplicative factor 
1(X), a conclusion which has now found general acceptance The numerical 
symbols thus correspond to the figures 955 and 988, which when referred to 
the Vikrama era yield the dates a d. 898 and 931, and, therefore, relegate the 
plates to the end of the mnth and the beginning of the tenth century res- 
pectively, that IS, fully 137 years later than the date assigned to them by 

The expunging of these two records from their place at the end of Plate 
IV of BOhler’s Tables has the effect of breaking up the senes mentioned 
above, and with it disappears a solid block of evidence for the supposition 
that Nagarl forms were commonly in use for epigraphic purposes since the 
beginning of the eighth century. It may be incidoitally remarked that even 
from a omsideration of the advanced foftns of the plate of Vinayakaplala, this 
t313[l is a satisfactory conclusion, as the latter fits in much better in its 

It is called North-Indian Nagarl by Dr Fleet, Indian Antiquary, Vol. 15, 

p. 106 

“ Buhler, op. at., p. 50. 

An independent proof of the correctness of this view has now been supplied 
by the date of the newly discovered I^rtfibgad Ins, noticed by Mr D. R BnAifn^" 
KAR ip the Indian Antiquary, Vpl 45 (1916), p. 12?, 



new place near the Sadoni^® mscnptioos from Central India, the dates of 
which run from a.d. 968, than in juxtaposition with the Multai p’lates and 
the Baijanathi^ inscription It is, however, an extremdy foitunate circum- 
stance that in this instance the palaographic conclusion finds a substantial 
corroboration from an independent source of evidence 

But to return to the question of the earliest use of Nagari, it may be 
observed that the alteration m the reading of the date of the plates of the 
Pratihara grants leaves in the main the thesis of Buhler untouched ; for, 
in assigning the earliest known specimen of Nagan to the middle of the eighth 
century, Buhler*® was relying on the Samangad grant*® of Danbdurga whidi 
purports to bei dated in the year corresponding to a d 754 ; and it must be 
admitted that m these plates iwe find not the shghtest trace of the wedge form- 
ation nor of the acute angles, but, on the other hand, the frequoit use of 
top-strokes (which cover the entire breadth of the letters) and the right 
angles which, as remarked above, are so characteristic of Nagail Moreover, 
as the reading of their date is beyond all doubt certain, the existence of these 
plates is priraa facie evidence in support of Buhler’s view. But, on the 
other hand, one cannot aitirdy ignore the fact that these plates occupy a 
very isolated position in the progressive develc^ment of Nagari For, the 
next earliest records in which iwe again find anything hke Niagaii forms belong 
to the beginning of the foUowing century, viz. the Radhanpur and Vaini 
copper plate grants of the RSistrakuta Govinda III issued in the year cor- 
re^pondmg to a d. 8D8 but it may be pointed out that m this grant of 
(jovinda III, the Nagari characters are used not exclusively as in the alleged 
grant of Dantidurga, {314'} but side by side with others which are distinctly 
acute angled. This distinction is worth noting : and I shall shortly have 
occasion to refer to it again 

As remarked above, there can be no possibility of doubt concerning the 
reading of the date of the Samangad grant * it is given both in words and 
numerical figures which tally with each other admirably. But this circum- 
stance does not exclude the possibility that the plates may not actually belong 
to the year to which they refer themselves ; and, in my opinion, the date is 
too early by at least a hundred yearsi if not more It is true that the space 
of a century often does not make an appreaable difference m palseographic 
matteTB. Moreover, while tracing the minute dianges in the shape of indivi- 
dual letters, even of co-eval documents, ws are by the natura of the circum- 
stances forced to utilise for purposes of companscxi alphabets from whatever 
locality they happen to be preserved, which is not the most satisfactory basis 

*« BfiHLEH, 0 p. c»f., Plate V, col. VII 

w BUhus, op. cit., Plate V, ool. I is Buhler, op. dt., p. 51. 

1® Indkm Antiquary, VoL 11, pp 106 ff., and facsimile, 
to Mm Antiguary, Vd. 6, p. 59 ; Vol. 11, p 158. 

PAlA^ographic notes 


of comparison. We must further reckon with the personal mdiosyncrasies of 
the engraver which are mostly an indeterminate factor. It is therefore nght 
to add here that the following remarks regarding the age of the Samangad 
grant are made with the diffidence which the circumstances call for. 

We shall now turn to the alphabet of this alleged grant of Dantidurga 
and examine it more minutely with a view to determine the standard of deve 
lopment reached by it. From what I have just said it follows that the best 
course would have been to sdect for comparison such documents as belong 
to the same epoch and are executed m the same part of the country I should 
have preferred therefore to cite for comparison two copper-plate charters of 
the Raslrakuta king Kmnaraja I which have recently been brought to light 
the one found at Talegaon (Poona distiict) has been briefly reviewed in Uie 
Progress Report of the Archseologidal Survey, Western Circle-, for the year 
endmg Match 1910 , but the other, found at Bhandak (Chgnda District, 
Central Provinces), has as yet received no further publicity beyond the bate 
mention of its discovery. C315} It is regrettable, therefore, that it is not 
possible to reproduce them here and make them available for the examination 
of the reader, as no desenption can adequatdy take the place of a facsamile 
Out of the plates which have already beas edited and which lend themselves 
for use in this connection, the Daulatabad plates®^ of the RS^trakiita .Sahkaia- 
gaiija (dated in the iSaka year 715 corresponding to A D 793) are as suitable 
as any other. When these two sets of plates are placed side by side, it will 
be noticed at once that there is a wide gap separatmg their alphabets. The 
characters of the S&mangad grant are far in advance of those of the DaulalS- 
bad plates, which are executed nearly forty years later than the alleged date 
of the former giant. The diffeience between them is now the more difficult 
to explain as the advanced types of the Digliva-Dubauli and Vinayakapala 
plates are no longer available for bridgmg over the mtervening gap. 

A comparisem between the alphabets of the two plates reveals the follow- 
ing points of difference betweai them. In place of the covering stroke of 
the letters gha, pa, ma, ya and ;a of the Samangad grant we have ornamental 
protuberances in the other plate As r^ids gha it is worth noting that 
an example of the tnpartite open form (m line 4, twice) of the Daulatabad 
plates can be seen m as late a record as the Pehva Pra§asti,®® which is assign- 
ed by Buhler himself to cir A d 900. Characteristic of a later epoch is the 
form of ja in the SSmangad grant which originafly and even in, the Multai 
plates (AJD 708-9)1 coosiated of three nearly parallel bars connected at one 
end. Subsequait development of the letter is as follows. The lowest bar 
develops a notch at its free end, and the middle inclines downwards. Inci- 
dentally It may be observed that this is the form of ;a in the Bt»ndak plate 

* 21 Ed. D. R. BhandabkaR, Epigraphta Indica, Vol 9, pp. 193 £f., and facsimile. 

28 BumsR, op. dt., Plate V, c(rf. III. 

18 a 



of Kri?i,iaraja I (ad. 772 ^ In the Viniayakaf<^a plate the notch develops 
into a curve, so that the lower portion of the letter forms a {[316} double 
curve, while the (originally horizontal) middle bar is all but vertical. The 
change is perfectly gradual, and is, I think, a good indez of the age of a 
document The ;a of the Daulatabad plates marks an intermediate stage 
between the two limits ’ the lowest bar is slightly bent backwards, while the 
middle bar, though inclined downwards, is near its point of attachment almost 
horizontal. In the Samangad plates, however, the typical ;« shows further 
piogress in so far as the lowest bar is bent double, while the middle bar is 
well on Its way to become veitical. Most noticeable and important are tire 
chaiacteristic acute angles m the Daulatabad plates as, for instance, in mo, 
ya, la and sa In the Samangad grant, on the other hand, the acute angles 
liave widened into right angles Thus with re^ct to the Samangad grant 
the Daulatabad plates of &mkaiaga|na will have to be looked upon as a retro* 
grade type But the latter is no excephon in this respect In fact, an exami- 
nation of the hitherto published records of the century intervenmg between 
the Samangad grant (alleged date A D 754) and the Kanheri inscnptions (cir 
AD. 850) will prove that it is not possible to produce a single instance of an 
inscription which is on the same stage of graphic development as the plates 
of Dantidurga The alphabet of every other inscnption of this period will 
appear aichaic or retrograde in comparison with the SamSngad grant 

It may be at once admitted, that there could be no exception taken to 
the circumstance that an inscription contains some forms which are slightly 
more advanced than those of other records of the same or even slightly later 
period In the above-maitioned grants of Govinda III, for instance, we find 
side by side types with wedges and those £317} with long covermg strokes, 
that is to say, a mixture of the acute-angled and another more advanced 
alphabet. While on the other hand, the Gwalioi inscription*'* of Bhoja, whidi 
Id roughly fifty years later in date, shows forms which are on the whole acute- 
angled This IS quite natural. In the case of the Sam^gad grant, however, 
tlie outstanding consideration for suspecting its authenticity is the arcum- 
ptance that it contains noit merely advanced forms, but that these should be 
used to the entire exclusion of others which must have been current at the 
epoch The use of the advanced forms is not arbitrary . the regulanty with 
which they recur diows that they had become fixed types at the time the 
document was concocted. The consideration that further search may brmg 

a» See for instance, Kielhorn, lAst of ItucnpUons of Southern India, Nos. 
794, 808, 809, 835, 8&7 In these examples it will be found that the lop stroke is 
attached to the left vertical of the letter and does not cover the entire breadth of it 
unless the vowel sign is appended to the letter, in which case the sign was drawn 
in oontinuation of the top sitroke. Another feature is the sporadic presence of acute 
angles in the letta^ glia, pa, ma, ya, etc. 

Buhler, op, at , Plate V, ool II. 



to light othei records which will supply thel missing links seems to me to be 
a futile consolation. At any late I should say that an essential prdimmary 
condition for re-establishmg the impugned authenticity of this grant will be 
the actual discovery of a sufficient number of dated records that will supply 
forms which can bridge over the gulf between the epoch marked by, let 
us say, the MultaE plates and the Samangad grant. Unless and until evi- 
dence of this nature is forthcoming, one might, in my opinion, legitimately 
doubt if the plates belong to the epoch to which they refer themselves. 

Another fact which corroboiates the suspiaon is the following It is t 
matter of common expenence that forged plates are generally very inaccurate 
as regards their orthography The reason for this may be that the text which 
was bemg copied was not familiar to the executois of the forgery Be that 
as it may, if this be any cnteaion, it will have to be admitted that the Saman- 
g£^ grant stands the test very badly, as the text of that record is in a lamen- 
tably corrupt condition. Dr. Fleet’s transcnpt does not show all the mis- 
takes of the original; for instance, the very first syllable of the first vene 
(.line 1 ) Dr Fleet reads as sa ; it is as a matter of fact a clear §a. In 1 8 
the third syllable is va;I>r, Fleet transcribes it with vi. But {318} Uiere 
are worse blunders than these m the text The half-verse beginning witb 
nltavadhe etc (1 17) has been mutilated beyond recognition, as a comparison 
with the Bhandak grant of Kjanaiaja will prove.®® But the most significant 
blunder is the one m the verse beginning with Mmadyuvcf (1. 16), The 
first quarter of this verse must in the ongmal have read something hke 
Mniad-'Bhuvagm0 noma The forger having misread the ligature dbhu as 
dyu, must have added conjecturally ti after va so as to complete the word 
yuvati and then in o.rder to adjust the number of syllabic instants of the 
quarter, proceeded to convert the final ma into an anusvSra, In doing so, 
however, he obhterated completely the word Bhuvagana, the name of the 
queai, a word which the wnter probably did not know at all. Si^ficant is 
also the fact that the Samdngad grant is the only elarly Ra$traku!ta grant, so 
far discovered, m which verses sabhruvibkanga etc., and KancUa etc (11. 23 
ff ) occur in this order ; dsevdiere the latter precedes the former It is un- 
necessary however to labour the point any further. 

, *® Indian Antiquary, Vol 11, pp. U21f. — 

Saimngad (Dr. Fleet's transcript) — 

nitavadhe(ythe)mivaSe?a}aeata}i pSlitayati[h*] [ Dr. Fleet does not translate the 
dubious nitavadhemwa ; the rest ha renders with ‘ who protected the expense of the 
wliole world’! 

Bhandak plates (verse 12) — 

nitavarthamivttieiajanataprSrtMtayatil'm] \ Translatioti : ‘ (From her he ^tained 
a skm) like unto material well-beinig (orfSa) from (i.e. as a result of) righteous 
conduct («*»)', a son who was, (as it were,) the future (prosperity) prayed for 
by the whole of mankind’ 



Lastly, I ^ould like to call attention to the use of decimal figures in 
expressing the date of the Samangad grant. Is this an anachronism ? Thai" 

IS no doubt a difficult question to answer The Samangad grant is certainly 
no longer the earhest known specimen in which the decimal notation comes 
into use, as remarked 6y Dr Fleet thirty-four years ago But it would be, 
if genuine, still one of the few inscriptions of a date earlier than the ninth 
{315>J century in which deamal notation is used As far as the grants of 
the successors of Dantidurga are concerned, it may be noted that m both the 
(unpublished) records of Krspa I, the Alas plates*^ (ad 770) of Yuvaifija 
Govinda II, and all the plates of Govinda III upto the Saka year 735 (i e. 

A D, 813) and the majoiity of his other records,®* the date is given merely 
in words. A noteworthy exception is a record of the Ra?trakiuta KakkanSja 
of Gujarat of the year ad 757, where the date is expressed both in words 
and numeiical figures. In this instance the symbols which are employed, 
be it remembered, are not decimal as in the SamSngad grant, but letter- 
numerals. But with reference to the use of the decimal notation I may add 
that in view of the mode of dating in the Gunara inscription®® of the Kalacuri 
year 346 (a.d 594), of the Valabhl inscription®® of the Gupta year 365 (’) 
(i.e. AD. 685?) and some others, one might surmise that the Gurjaras and 
perhaps their neighbours m Gujaifit had adopted the more advanced system 
of decimal notation much earlier than their contemporaries furtha: south. 
We know, however, so little definite about the early use of this notation m 
India that it would be unwise to formulate a solution which hapjpens to suit 
a particular case, I leave it, therefore, here as an open question whether we 
can legitimately assume the prevalence of the use of decimal notation in tlie 
heart of the Southern Maratha country as early as the ei^th century, especl 
ally in epigraphic records which admittedly affected a certain amount of 
archaisnx Worth noting, however, is the fact that even to BtiHiER the forms 
of the numerals in the ^mangad grant appeared to he ‘ stron^y modified 
cursive forms.’®i But here again we are on shaky ground for want of suffi- 
cient material on which to base a definite conclusion. 

{320} In this connection Mie is irreastibly remmded of the Dhmiki 
plat^® of Jaikadeva of Sauieisltra bearing the date V. [7914 corresponding 
to A.D. [73]7. In this instanoe also, the numerals expressuig the date are 
decimal and the alphabet is a well deivdoped form of NSgaii. The details^ 
of the date, however, leave (in the concurring judgment of Kielhorn and 

®» BriHLEH, op. cit , pp 78 f. 

Ed. D. R Bhandabkab, Epiiraphia Indica, Vol 6, p 209 and plates. 
*> See Kiu-horn, iMt of the Inscriptions of Southern India. 

®» Ed. Dhkdva, Bpigraphia Indica, Vd. 2, pp. 19 ff. and plate. 
eo Journal of the Bengcd As. Soc , Vol 7, p. 968. 

« BiJhier, op. at., p. 79. 

Inditm Antiquary, Vd. 12, p. 155 and plate, 



Dr. Fleet) no doubt as lo its being a forgery®® It is unfortunate, therefore, 
tliat the details of the date of the Sam&ngad grant are not capable of verifica- 

Taking all things into consideration, the balance of evidence points, in 
my opinion, strongly to the conclusion expressed above, viz that the Saman- 
gad grant is spurious This conclusion, if granted, would have the important 
result of taking the epoch of the use of Nlagari in epigraphic documents for- 
ward by at least a hundred years. For, as remarked above, if we leave out 
of consideration the Samangad grant, the next earliest inscriptions which aie 
wntten throughout in Nagai5 are the Kanjhen inscnptiwis of the Silahfiia 
princes PuUaiakh and Kapardin II These nearly co-eval mscripbons exhi- 
bit the regular use of top-strokes covering the entire breadth of the letters, 
as well as rectangles (as opposed to the wedgels and acute angles) m gho 
(BuHLKi’s Palseographic Tables, Hate V, col. V , 13), pa (col. V, 30), ya 
(col V ; 35), sa (col. V ; 40) and sd (col. V , 41). The RSdhanpur and 
Vapi plates of Govinda III , which cemtain a mixture of both the acute-angled 
and transition types, appear now in a different light. They do not represeit 
a retrograde movement but a progressive one. Preceding as they do ny 
about fifty years the earliest known mscriptions m which Nagaif forms are 
exclusively employed, they represent a true transition stage 

As the outcome of the analysis here undertaken, we arrive at the follow- 
ing conclusion The very earliest dated imiriptiaDS hitherto known which 
are wntten throughout in NSgari characters are the mscriptions (cir. AD. 
850) of C321J the Silahara princes, from the Kalpheri Caves m Western 
India. These show (i) the top-stroke covering the entire breadth of the 
letter, and (ii) rectangular comers Transition stages leading upto these 
forms have been already discussed. In these the top stroke never covers the 
entire breadth of the letters, while some characters retain their former acute 
angles ; the form of ja is also a significant index. The subsequent course of 
the devdopment of Nagan in Western India can be traced with the hdp of 
the below-noted insenptions of the RSetrakutas of Malkhed and LSlta bdong- 
mg to the period cir. ad. 850-950. A minute examination of these records 
will also provide further support to the inference that the Kanheri inscrip- 
tions diould be placed at the middle pioint of the evolution of the Nagad 
out of the acutd-angled alphabet. Following are the inscriptions above re- 
ferred to : 

' 1. Kielhorn’s Southern List No. 77, Baka 789 (A.D 867). The 
BagumtS. plates of the Mahasamantadhipati DhruvaiBja II — Dharavarsa- 
Nirupama of Gujaiat — regarding the alphabet of whidi BOhler remarks 

»® See references under Kielhokn. list of the Inscriptions of Northern India, 
No, 8, 



{Ina. A>it Vol 12, p 181) that the letters resonble those of the Samangad 

2. Ibid No 81, Saka 810 888 J The BaguimS plate of the feu- 

datory Rastraku'ta Kisnaraja Akalavarsa of Gujarat, In this instance the 
tcH3-strake covers the entire bieadth of the letter, and rectangular comers are 

3 Ibid Nos 86-87 iSaka 836 (ad 914) The Bagumra plates of the 
Ra^trakuta Maharajadhiraja Indra III Here the development of Nagari 
along the two mam lines indicated above is completed 

4 Ibid No. 91, t§aka 852 (ad 930) The Cambay plates of Uie 
Rajtrakuta Maharajadhiraja Govinda IV This superbly engraved record 
may be looked upon as a standard to which the Nagari of the tenth centuiy 
was tending 

5. Ibid No 92, 'Saka 855 (AD 933). The Sahgli plates of the Rastra- 
Inita Maharajadhiraja Govinda IV the cha-{322) racters of which are of the 
same type as the Bagumra inscnptions of Indra III. 

6. Ibid No. 94, iSaka 867 (ad. 945) The S^otgi (Bijapur District) 
pillar inscnption of the reign of the Ra^trakuta Kisna III, Akalavarsa The 
forms are perhaps somewhat more archaic than those of the plates mentioned 

Additional reference will be found in BOhler’s Indische Palaeogwphie, 
p. 51. 

So much for the earliest use of Nagari m Western India Regarding its 
use in Northern India, I should like to add the following observati<Mi which 
arises directly out of a fact noted above Buhler’s mislection of the date 
of the VinayakapSla pjlate, as we have seen, led him into an error regarding 
the period at which this alphabet became an epigraphic alphabet m Northern 
India. Having thus erroneously dated this instance of the use of Nagari m 
A D. 794-5 he found that the succeeding, that is the ninth, century was prac- 
tically bare of Nagari inscriptions, and had to admit that it was not till the 
middle of the taith century that this alphabet comes again into general use 
m that part of India. Bijhler was, 1 think, substantially nght m saying 
that in Northem and Central India the NSgari appears first in the copper- 
plate grant of Vina-yakapala, but that event has to be dated in AJ) 931, It 
remains to determine the transitional stages during the latter part of the ninth 
and beginning of the tenth century ; but it would appear as if there are no 
Nagari inscriptions belor^mg to the eighth or even the early part of the ninth 
century from Northern India. 




1. At the instance of Mr. Bhandarkar I resumed during the last cold 
season the work of pr^aring an inventory of the ancient monuments of 
Rajputana where it had been left by him m 1911 I had mapped out a 
programme for a five months’ tour of exploration through .Rajputana I 
intended spendmg die fiist two months of the tour in visiting the places of 
antiquarian mterest in the Sirohi State, and then devoting the remaming three 
months to explonng Jasvantpura, J.alor, Jaitaran, Sojat and a number of 
other districts of Jodhpur, a list of which was kindly placed at my disposal 
by Mr. Bhandarkar. Unfortunately the tour had to be cut down consider- 
ably, as on account of pressure of work at the headquarters I could devote 
barely three months to exploration work I was not able to leave Poona 
before the begmmng of December 1916, and I was recalled again in the first 
week of March 1917. As the whole of this period— with two brief mtemip- 
tions— was taken up with Sirohi itself, the rest of the programme had to be 

2. Though Sirohi had been visited twice already by Mr. Bhandarkar, 
my object m placing this State first on the list was to fill up the lacunae in our 
collection of the mscnpitions of the Pararaaras of Abu, most of which, I may 
add, are located withm the confines of the Sirohi State The project was 
completdy successful And we have now in our office an entire set of the 
inscrqitions of this djmasty accessible in Sirohi. "With the matenal I have 
collected this season, along with that which was already on record in this 
office, It is possible to reconstruct a dceleton of the history of this family of 
Rajput chiefs from the middle of the eleventh century to about the middle of 
the fourteenth century a.d , a subject which I intend making the theme of an 
article for the Director-General’s Annual oj Archxology 

3. The most notable discovery of this year’s touring through the Sironi 
State must be called the marble temple at Varmap, dedicated to the Sun god, 
which can easily rank as one of the oldest Surya temples now standing. This 
monumfiat is interesting alike from an architectural and iconographic point of 

* [Pp. 59-72.1 



view, and, though in a state of advanced min, is pre-eminently deserving of 
preservation. It is therefore to be wished that the Sirohi Darbar be appioadi- 
ed to take the requisite steps towards the conservation of this highly interest- 
ing relic of the past and look to its being kept in permanent good repair 


4. The first village to be visited was Or — ^the Ur of maps — ^three miles 
due north-east of Khaiadi (Abu Road). Besides the temple of Vishjiu along 
with its subsidiary shnnes standing on a high platform, the village contains a 
Jain temple, now dedicated to Parivanlatha Both are of a comparatively 
modem date and from the architectural point of view devoid of interest The 
temple of Vi'thalji is a triple shrine, that is, a shrme consisting of three cells 
in a row. The middle cell contains an image of Vishipu, locally called Vithalji , 
while the other two contam siva Lmgas Although this composite structure 
forms a single temple now, there can be no doubt that ongmally there stood 
in its place two separate small shrines at some distance from each other, 
which, at a subsequent period, were connected together by erecting side walls 
so as to form a third central cell. The so-formed tuple shrine was then pio- 
vidfid with a common closed m<t>.tdapa, the outer wall of which is pierced by 
a finely carved doorway of marble On the dedicatory block of the hntd is 
sculptured a figure seated in the conventional attitude of meditation. The 
supenor workmanship of this member, which distinguishes it sharply from 
other parts of the temple building, betrays its ektraneoua on^ and places 
It in a class with those superbly carved fragments which one notices round 
about the village of Chandiavae and whKh are thel sole material remains of 
the once flouiidung capital of the Paramlaras of Arbuda-matnidala. On a 
pilaster of one of the side cdls of the tnple shrine is engraved an inscription 
(in a local dialect) dated in V. 1589 Bhadrapada-sudi 11, recording a money 
grant to the shnne, from which it follows that the latter was in existence 
before that date. The outer walls of the central shrmes are pierced by niches 
which hold various Brahmanical deities. Worthy of note are the well-carved 
alto-rdievoe £60 J of LakuBSa and iSiva (Photo Nos. 4578-9) placed in the 
niches in the walls of the shnne on the right The nature of the object held 
in the lower nght hand of the latter image (Photo No. 4579) is not evident 
at first ; but a little refiechon will suffice to convince one that it can be 
nothing else than a begging bowl. The image is carved out of the same kind 
of stone as the building material ; the decorative dements of the niche are 
also in entire agreement with the style of architecture of the temple There is, 
therefore, no reason to doubt that the unage is contemporaneoius with and 
forms an int^ral part of the original fane. It was stated above that asso- 
ciated with the triple shrine there were smaller shnnes standing on the same 
pktform (Photo No. 4577). These contain marble or stone images of Surya, 
Vi8hiQu.Oiatiirbhuja and iSiva and P&rvati. Near the western edge of the 


platform lies a fiagment of the arch of a tor ana sculptured on both sides, 
the centre of which is occupied by the seated figure of the four-armed Siva 
as Yo^ndra The lower pair of hands are folded together in the lap, while 
the uplifted hands hold a trident and skull. This fragment bears a sinking 
resemblance to the corresponding portion of a torana standing amid the group 
of temples on the bank of the well-known Mandakini Kunda at Adialgatl.— 
The only objectives worthy of notice in the local Jam temple are the two 
inscnptions dated V 1242, engraved on the base of the images of dvaropalas 
which flank the doorway of the closed hall. They contain a village name 
Odagrama, which is undoubtedly identical with that of the village under 
descnption. It thus appears that smee the twelfth century the village has 
continued to be known under the name Od or Or. It also follows from the 
record that formerly the temple was dedicated to Mahavira 


5 Retracing our steps we turned westward in the direction of Giivar, 
which IS eight miles due north-west of Abu Road Girvar is noted for its 
ancient Vi^u temple of P&t-Narayana, a name which imports as much 
sanctity as that of any temple on Mount Abu Mr Bhandarkar has already 
done justice to the antiquities of Girvar in the Progress Report for the year 
ending March 1907, and in particular to the two inscriptions (Inscr. Nos. 
2737-8), one of which throws a flood of light on the obscure history of the 
ParamSras who hdd sway over the country round Abu We might, there- 
fore, pass on to the consideration of a site (Photo No. 4581), from which 
there has beai recently unearthed, among other things, a large ,Siva Linga 
ai^^edestal. The massive dimensions of these objects of iSiva worship may 
blathered from the following figures The diameter of the pi)}di is 15", 
while across the yoni the diameter measures nearly 3' 10"; the height of the 
pindi above the yoni is 20", but above ground level its hd^t must have 
measured aa much as 3^ 6" As remarked above, the whole of this massive 
Liiiga was buried under a mass of debns, leaving only a portion of the polish- 
ed edge of the pindi exposed to view The curiosity of the villagers, at first 
actuated, I suppose, by the evident traces of the ruin of a temple building 
with which the site abounds, led them to dig around the exposed edge of, the 
pif^i. Great was their surprise when they lighted upbn this Linga Shortly 
after the discovery of the latter a wdl-preserved sculpture of Nandin was 
exhumed. These are dear traqela of there having stood on this spot a Siva 
temple, the antiquity of which is attested by the masave proportions of 
the Linga and the style of decoration of the mutilated members that are 
ij ring around in abundance. These fragments are mostly wdl-carved and 
very carefully fiiushed. The Lmga, we are told, had not been moved. And 
as the pranalikS (water channd) j^nte regularly to the north it is possible 
that the T.ifiga is even now nearly m sttu In the dose vicinity of the ate of 



this ruin stands a brick structure which attracted our attention by thei large 
size of Its bricks. They measure on an average 16" X lOJ" X Of the 
same size are the bricks of a face of a wall exposed to view during the exca- 
vations carried out by the villagers, which from its position appears to have 
been the retaming wall of the plinth of the old Siva temple The above 
mentioned bnck stiucture serves now to shelter a number of idols of Brah- 
mamcal goddesses, such as Kali, Indrani, Brahmaini and others It is not 
clear what sort of a itkhara the -Siva temple had, if, indeed, it did have one ; 
for, although I closely examined the exposed fragmoits, I found nothing 
which could with certainty be looked upon as having formed a part of 
j[6l} the spire It mi^t have been worth while explonng systematically the 
site, which within a circumference of about 80 or 90 feet from the Lifiga 
n studded with brick-bats and broken members of buildmgs It is, therefore, 
a great pity that the villagers have already set about buildmg a temple in 
which to enshrine the Svayambhu Siva (-) on the very spot where the Lmga 
was unearthed, thus obliterating all traces of the older temple. 


6. Our next halt was at Datani, a village situated 6 miles north west 
of Girvar. DataijI is believed to be the scene of the battle fought m V. 1640 
between MahSiSo Surtiui of Sirohi and Emperor Akbar, in which the Maba- 
rao was victorious. There are three temples m the village, none of which 
contains any feature of architectural note But the local iSiva temple of 
SiddheSvara contains a sculptured memorial stone which, on examination 
turned out to be the record of a dread nte called kamalpuja performed by 

• one Suja in this very temple m V 1688 Phalguna-sudi 2 The sculpture 
above the inscription represents the figures of a man and his wife stanlfcg 
side by side, with hands folded together m an attitude of wordiip It appears 
that the kanud-puja culminated in the offering of the worshipper's own head, 
which was severed by a crescent-shaped instrument specially devised for the 
purpose. The widow of this martyr immolated herself at the same time on his 
funeral pile The motive of the suicide is not stated. 


7. While at Datani we secured impressions of an inscription (No 2739) 
of the Paramara Dhanavaisha, the importance of which will be made dear 
presently. The inscription is ehgraved on a plain octagonal monobtli of 
marble, about eight feet in height, standmg on the brink of a small pond three 
miles due north-west of DatSiru. The name of the adjoining village is Maka- 
val. There is no other monument in sight ; the pillar stands by itself. The 
inscriiHion is dated V. 1276 ‘Si&vana-sudi 3 Monday The stone is unfortunately 
very much the worse for weather acticm and the inscription is barely legible ; 
and so the purport of thd record cannot be made out Clear, however, are 
the names of Dhfiifiyacshadeva and his ca^utal of 'phandravati, which dgta 


enable us to ascribe the record to the time of the Paramara Dharavarsha, son 
of YaiSodhavala. The Vikrama year in which the record is dated, viz , 1276. 
becomes thus highly important, for this is the latest year so far found for 
Dhanavarsha His earliest record, the one from KayadrS (Sirohi State) is 
dat^ in V. 1220, and as this one bears the date V 1276, it follows that his 
rule extended over the somewhat prolonged period of 56 years if not more. 
It IS unnecessary to point out the importance of this piece of information m 
the matter of settling the chronology of the Paramaras 

8 During the Chnstmas holidays I visited the Dilvada temples on 
Mount Abu and drew up notes on the "work of repair going on there, for the 
guidance of the Supermtendent m anticipation of his annual visit of inspec- 
tion, which had necessarily to be of very short duration as he was at that 
time busy excavating in Sind and had very little time to attend to work else- 


9. About four miles to the north-west of RohedS Station is situated the 
village of Nitoia, which craitains several temples, of which only two are 
deserving of special notice, and they are the shrine of Surya and the temple 
of PanSvanatha The former faces the east and stands along with several 
other shnnes inside an enclosed court Inside the cella tiiere are two marble 
images standmg on an devated platform . one of them is Surya wearing 
Hessian boots, and the other, small in sizei, is the image of a two-armed 
goddess In one hand she holds a lotus and in the other an object w'hich 
looks like a kalasa. On account of the absence of any distinguishing vehide, 
as also the paucity of symbols, it is difficult to identify the goddess. But 
there can be no doubt that she is in some way associated with Surya (and 
may be even the iSakh of Surya) ; for, the images exhibit similar decorative 
details, the same stiff attitude ; in fact, the same technique throughout. It 
is worth noting that the legs of the! goddess are rqiresented as dightly bent 
at the knee as though she were about to kned. In the centre of {62'} the 
small porch of this temple is a king of a chaumukh stand bearing on its top 
a polished droular slab of stone (Photo No 4582). A curious aspect of 
the onaitation of the square pedestal is that its comers point to the cardinal 
dira±ions At Vasa, about a mile and half to the north-east of Roheda, 
there is a large Surya temple, in the porch of which is placed a simUar chttu- 
mikh stand ; but the latter culminates in a “ flat full-Wown lotus.” In the 
TtiHiati Museum, Calcutta, there are, I understand, two or three such pedestals 
with full-blown lotus on top. But in our spedinen it is impossible to ttiink 
that the drcular disc abbve alluded to stands for a lotus. We must look 
for its eiqilanation. in some other attribute of Surya. It will be remembered 
that the car of Surya has only one whefej ; in fact, ekachakra is one of the 
epithets of the Sun god Such being the case I see no reason why the un- 



decorated flat disc should not be looked upon as an emblem of the vehicle 
(v^ana) of Surya (like Garuda of Vish|nu and the Nandin of iSiva) which 
is, in fact, nothing more or less than the visible disc of the lurmnaiy. — ^The 
only other temple m Nitora that is worthy of special mention is the temple 
of PSTsvanatha. The cella contains the typical marble image of a Jam 
lirthathkara. But it is doubtful whether this image was originally meant to 
be one of PSrsvanlatha , for, the emblem engraved on its pedestal is not a 
serpent, as is to be expected in the case of PSrsvanatha, but a wheel ; which 
latter, by the way, finds no mention at all in the canonical list of emblems of 
the four and twenty tvrthaihkaras In the south-west comer of the enclosed 
aisle is a marble image with protuberant belly. Locally the image is known 
under the name Bdblajl. On the mukuta of the figure is carved a repiesenta- 
tion in mimature of a seated ttrthamkara The inscription on the base which 
reads : 

(1) Samvat 1491 varshe Vai^akha-sudi 2 Guru-dine 

(2) Taksha~bava-ki rmrttih . . . Subham bhuvatah || 

shows that it is the representation of a Yaksha, an attmdant of Kubeia 
The image has four hands ; in the uphfted right he holds tndent, m the 
lower right rosary ; while in the uplifted left there is a pasa (noose), and in 
the conesponding lower hand karmndalu (water-pot) 


10 Our next halt was at Kojra, which is situated about 10 miles due 
north-east of Nitora. Opposite the Jain temple of Sambhavanatha stands 
the inscribed slab conveying the gift of the villagei to the Purohita of the 
royal household. The record bdongs to the reign of the Choh^ (Devaija) 
king Suratajna and is dated on the 9th of Ashadha-vadi V. 1634.. It states 
that the gift was made at the request of Dharabai, the queen dowager. In 
the middle of the village stands a temple of Vishnu, locally knoiwn as Para^u- 
lama, which is highly popular with the Kumbhars (potters) We were in 
fact told by a local Bbaf; that the temple was built in V 900 by a Kumbhar 
Although the chromcle may be ngjit as regards the caste (le., profession) 
of the builder of the temple the style of the architecture does not suppprt 
the claim to the alleged antiquity. I was interested to know what sort of 
image they make of Parasurflma and so visited the temple early in the morn- 
ing in order to mspect the god while he was being bathed, as that is the 
only occasion on which the image is undraped. It was disappointmg to find 
that the so-called ParasurSma was no other than the four-armed Visl^uu- 
Chaturbhuja. — Qose by this temples standing on an eminence is the Jam 
temple dedicated to Sambhavanatha The oldest portion of the structure 
dates probably from the twdfth century, but many additions and alterations 
have beai made m the intervraung period Even now the temple is under- 
going elaborate renovation at the hands of the local Bania community. The 


image which at present finds itsdf in the main shrine is undoubtedly that of 
Sambhavanathd as the lafkhkema on the base erf the image is a horse, ?nd 
thus the temple is nghtly called the temple of Salmbhavanatha But an 
inscnption engraved on a pillar of the gudhammdapa tells a different story. 
Only two lines of the inscription (No 2740) are now visible from under tlie 
coat of ckunam with which the pillar has in recent times been covered, and 
they are : 

1. [Sainlvat 1224 Siavana-vadi 14[Somel 

2. Sn Pdrasva-natha-deva-chaite Rand Rava 

As there is every reason to believe that the pillar is m situ it follows that the 
temple was originally consecrated to Parsvanatha. 


11 Ten miles from Kojra and nearly as far from Sirohi lies the popular 
tirtha of BamainvarjS, which attracts pilgrims not only from all over Sirohi 
but even frmn other states of R&jputan& The group of temples stands at 
the foot of one of the small hillocks which lie along tlie mam road between 
Sirohi and Pijnidviada.. A'high wall encloses the temples and the secular build- ' 
mg s built all around for the convenience of the pilgrims. The mam temple, 
which IS dedicated to Mahavira, locally called B&manvSrji, belongs probably 
to the 14th or 15th century, but it has undergone eictensivel additions and 
alterations in the interval The pujan has a fable— too fantastic to be repeat- 
ed here— to tdl in explanation of the word Bamaipvarji But m older mscrip- 
tions the pTaca-name occurs as Bajnbhanavadagrama, containing the element 
vad{a) which occurs at the end of so many other place-names, e.g, Anhilvad, 
Dilvadfi, etc. Curiously enough, in the heart of this sanctum dedicated to 
Jam worship a iSiva Liifiga also finds a place, and, from all accounts, has 
occupied that place from time immemonal. Within recent years the Jam com- 
mumty had attempted to oust this emblem of Siva, but its restoration was 
speedily brought about, we were told, by a royal mandate From an arclu- 
tectural point of view the temple is unimportant An inscnptiOTi (No. 2742) 
on the architrave of one from the row of small shnnes that surround tlie 
courtyard records the dedicaticn of that dinne m V. 1519 by a Pxagvata 
(Porvtd) Bama livmg m Viravataka This village is to be identified witli 
Viravada (Birwara of Quarter Sheet 20 S. E.)' situated a nule to the norA- 
west of BamalnvIarjS In the same inscnption the itrtha itself is called Brah- 
mSnavdda-gT&ma-mahdstkana — ^In the south-west comer of the enclosuie of 
the temple there is a anall dinne dedicated to MahSdeva Outside the diruiR 
stands an inscribed stone (Inscr. No 2743) built into thei outer wall of the 
duins The edges of the stone are chipped and its lower portion is also 
damaged. In places the hollows have been filled in with cement. The record 
which is short, is dated in V. 1249 (a.d. 1192) and refers itself to the reign 




of the Paramara king Dharavarsha. I have alluded to inscriptions of this 
lung dated in V. 1220 and 1276 The inscnption m B^alnvarjl gives us thus 
only an intermediate date. The stone being highly weather-worn, the con- 
tents of the record could not be made out satisfactorily The language ap- 
pears to be a local dialect The name Bambhanavada-grama is clearly dis- 
cernible, and thus probably the record is one of a gift either to this temple 
or some other temple in this village. The inscription begins as follows — 

fl. 1) Om Saihvat 1249 varashe Sri-Dh&ra[va*']rsha saml^a\}e 


(1. 2J navada-gratne . 


12 From Bama'nvarjl we had intended to proceed directly lo Sirohi. 
But on receiving intimation that Baida, a viUage only six miles away, con- 
tained a very old Jain temple, We turned away from our objective and made 
a detour of a couple of miles m order to visit Baida Great was our disap- 
pointment whai on arriving there we found that the “ very old Jam temple ” 
of which we were told was a temple of the 14th or 15th century with no 
pretensions to architectural mterest The temple stands on a high platform 
and consists of the mam shnne, a large hall and an enclosed aisle of cells, 
which latter are empty The mam shrine contains an image of Mahavira 
installed m V 1697. The temple is, however, much older than the image ; 
for, m the mscnption (No 2744) engraved on the lintel of the door opening 
into the cella, it is recorded that m V. 1493 Jyeshtha-sudi 7 Monday Gupa- 
bhadra renovated the temple built by his ancestor Baladeva Also at the 
same time an image of Mahavlra was installed there The sum needed for 
getting the image made was subscribed by several pious Jains whose names 
aie also recorded in the inscnption. 


13. At Sirohi we made a short halt in order to lay in a store of provi- 
sions in anticipation of our protracted sojourn in comparatively wild tracts, 
and also m connection with certain slight differences that had anserf m our 
dcahngs with the local authorities, and that were happily adjusted through 
the kind mediation of the Dewan Saheb. While there, I wanted to inject 
the Vasantagadh inscription of Punpapala (dated V 1099), which was re- 
moved from its find-place and brought to Sirohi for safe-keeping 
However, as no one at the Darbar knew where it had smee been housed, I 
had to abandon the pursuit in exchange of a promise on the part of the 
Secretary to the MusSheb Ala to make further enquiries and let me know. 


14. At the foot of the eastern mountain range running from SiroM 
ftorthwards lie to the east of the cart track the remains of the old township 


of Kolar (about five miles due north-east of Sirohi) which is all but deserted 
now. At present the only habitation there is that of the family of the iujari 
attached to the temple of Adinatha and a few Bamas The t^rnpie has been 
restored m recent times by the Jain community of Plaladi, which lies about 
five miles further north It possesses no architectural features of note. Those 
of the images m the temple that bear dedicatory inscriptions date from the 
18th or 19th century of the Vikrama Era Worthy of notice is the sculptured 
Maktana lintel (Photo No 4583) of the central bay of the pillared corridor 
adjoining the sabhamandapa on the side of the entrance door of the temple. 
On this architrave is carved in high relief the pictorial rgiresentation of a 
legend, probably from the life of a Jama tirthamkara, the full significance of 
which, however, is not evident In the left comer of the sculpture is re- 
presented a queen reclining on a canopied couch Then follow in succession 
from left to right the following fourteen objects . elephant, bull, saidula (or 
horse), Gaja-lakshml (with elephants), an unintelligible symbol, the Sun 
and Moon, ankusa, kalaSa, walled town, nver (to be identified by the fi^ 
and tortoises swimming m its water), temple, sakasr^-linga, and lastly ratha. 
The connection between these fourteen objects and the sleeping queen is 
elucidated by the following legend in Niagari characters of the 12-15th century 
engraved m the left comer of the sculpture . Maharajm Uialddevi chaturdasa 
svapnani p^yati . ‘ MahSila|ta UiSaladevI dreams fourteen dreams.’ The 
space above this picture is carved with a design of pomted leaves separated 
by ornamental chains of beads On the soffit of the architrave is sculptured 
a row of rosettes arranged in narrow panels. This odd architrave of marble 
has been undoubtedly imported from the mms of an older temple It may 
be noted that Makrana does not otherwise come into use in the constniction 
of the temple under desenpdon. — ^Kolar might have been a big-sized town 
at one time Prominent among the nuns are the dilapidated remains of the 
bastions and buttressed walls of a disused fort which overlooks the village 
from the crest of the adjommg hill. 


15 RaladI is a large village situated about ten miles to the north-east 
of Sirohi. My object m making a halt m Paladl was to secure impressions 
of an important Chahamana mscripfion from the local templd of Mahavfra, 
an inscription which by virtue of its position is a piece of direct evidence of 
the gradual oicroachment of the ChahamSnas of Marv&r into the territory 
of the Paramaras of Abu in the beginning of the 13th century A D. The 
temple is in leligious occupation of the community. It faces the north and 
consists of a gudha- and sabhdmatfdapa, an elaborate porch, and an enclosed 
aisle of cells, some of which contain images of tirthomkaras. TTie sabhd- 
nmf4apa has a domical ceiling supplorted as usual on an octagon of pillars. 
Between the monolithic shaft and the capital is inseited a ^ort lengtli of 



ornamental necking consisting of a fluted vase resting on a fluted cushion in 
order to secure the requisite height (Photo No 4584) (Seven of the pillars 
are inscribed with short records giving the name (or names as the case may 
be) of the donor who contiibuted the sum needed for erdcting the column. 
All the insaiptions (Nos. 2745 — 50) are dated on the same day, viz , Fnday 
Ashadha-vadi 1 in V. 1248 In the gudhamandapa there is a marble image 
of standing dvctfapdla on either side of the shrine door,' and smaller images 
of tlrthamkaras are placed in the niches of the eastern and western walls of 
the mam shrine. The msaiption (No 2751) for which iwe came here is 
engraved <xi a stone of the outside wall near the porch of the temple It 
is dated m V 1249 jMagha-sudi 10 Thursday and refers itself to the reign 
of Mahdrdjadhiraja i§rT-Kdha|nadeva and his son Jayatasliha-deva It also 
mentions the latter’s chief mimster Vllhana and another person named Raja- 
deva, son of Sii(Bhu ?)niadBva, who was in some way related to JayatasOia’s 
mmister £65} The abraded condition of the stone leaves unclear what it was 
that Rajadeva did to this temple of Mah5,vira The kmg Mahdrajadhimja 
Kelhainadeva, to whose reign the record refers itself, is undoubtedly the Chaha- 
mkina Kelhaiija, son of Alhalna, whose capital was at Nadol (Naddula) Pandit 
Gauiishankar m his Hindi history of Sirohl (SnoJn-rajya-ka Itthasa, p. 56, 
footnote) gives the date of this record as V. 1239 This is, however, a 
mislection as the details of the date (which were kindly verified for me by 
my learned fnend Dew^ Bah&dur Swanu Kannu Pillai) are correct only 
for the year V. 1249 and not for V. 1239. The date thus corre^nds to 
Thursday, 14th January, a d. 1193. As the name of the Yuvaiaja is men- 
tioned in the! record along with that of his father, it is possible that the 
Yuvaiaja had been enjoying the country around PSladi as his bhuktt. The 
real importance of the inscription lies, however, m the condusioo deduable 
from it that even before the aid of the twdfth century the Chihamanas had 
penetrated into the kingdom of the Paramaras as far as PSlaiE, which is not 
more than forty miles direct distance from Chandiflvati, the capital of the 
' Paramaras. 


16. From Paladi we visited Va^ which is about a mile away from the 
fomver village. The two Jaina temples of stand side by side on an 

eminence surrounded by a high walled enclosure. One of these temples is 
consecrated to AdinUtha’ and the other to 'S&ntinatha 'The plan of both is 
eicactly alike; the temple of AdmSiha is, however, slightly larger than the 
other. The extenor of the temples is devmd of ornamentation excepting a 
narrow horizontal band decorated with lozenge-shaped leaves. Qi;igmally the 
temple of Admatha, whidi may roughly be attributed to the 12th or 13th 
century, appears to have consisted only of the dirine, gudhamOjdoPd 
porch. At some sijbeequictat period the porch was extended by the additiQi) 


of a large open hall with a domed ceiling supported as usual on an octagon 
of pillars and also by a pillared corridor adjorning this hall There are in the 
temple of iSantmatha three mscnptions (Nos. 2752*— 54) : two short records 
on two of the pillars of the original porch are dated both m V. 1264, and 
contain the names of the donors The inscription on the architrave of the 
doorway of the mam shrine is much later than these, being dated m V 1353 
in the reign of Samamtasimha-deva and records the gift of a certam quanuty 
of godhuma (wheat-flour) to the temple for the purpose of defraying the 
expenses incidental to the holding of the annual festival {ydtrd). The name 
of the village is given as {,V dghasif};a) situated in the Nadduladela The word 
dhivada ocournng in the inscription is used m Marvar to denote a small 
arahatta (Persian wheel), imgatmg only about half as much as the latter, 
while sdi is the name for a measure of 16 payalts The importance of the 
inscription lies' in its date V 1359, whidi is later by five years than the 
latest date mentioned for Samalihtasimha by Mr. Bhandarkar in his article 
on the history of the ChShamanas of Marva, which is the most up-to date 
and detailed pronouncement on the subject. I may here add that even V. 
1359 IS ncA the latest date for that kmg, for at 'Othamain (see next paragraph) 
I discovered another inscription referring itself to the reign of Samvatasimha 
which is dated in V. 1362. 


17. Not more than about a mile and a half to the north-east of Paladl 
lies the village now known as tithamain. In the Quarter Sheet No. 20 S. E 
the name is ^own as “ Utan ’’ which stands evidently for tithalp This dif- 
ference of pronunciation does m fact exist ; for even in the local inscriptions 
the name is spelt sometimes as tJthaman and sometimes as Cthaij or tj'thun. 
In the local Jama temple there is an inscription (No. 2755), on the finely 
carved marble pedestal of the image in the main shrine recording the gift of 
jalobattui?) to the temple of P&nSvan&tha by Devadharji, son of Dhapasava 
by his wife Dharamati In this record the temple is referred to as the Cthuna- 
chaitya. In the mscnphon (No. 2756) on the lintel dated V. 1251 the place 
is called Cthaipa. The small shnne standing on a high mound adjoining 
the hiU at the foot of which Othamain itself lies is a temple of Mahader.'a 
locally known as 'Othamesar (trthameSvara). Besides the shrine there is a 
gudhamandapa adjoining the simple pordi On either side of the doorway 
is an inscription (Nos 2757-581): inased on the pilasters of {66} the poich 
dated V. 1256 Jyeshtha-sudi 14 Monday and referring to the reign of Sam- 
vatasilmha. The language is Marvafi. The puipose of the inscriptions is to 
record gran^' to the temple. Just outside the porch there is a third inscrip- 
tion (No 2760) engraved cm the faces of a dwarf pillar square in section 
Ic refers itsdf also to the reign of a Swvatasimha of Naddula and is dated 
in the Vikrama year 1362. This Samvatasimha can be no other than the 



Chohan king of that name, son of Chachigadeva. The date of the inscrip- 
tion IS, as remarked above, the latest hitherto known date of Sairhvatasitnha. 
Unfortunately the stone is highly abraded and the inscription is to a great 
extent illegible, though it seems almost certain that the record consists chiefly 
of an mventory of a large number of donations made to this Siva temple by 
various persons at the same time Both the Jam temple (Photo No 4585) 
and the temple of Crthamesar (Photo' No 4586) must be anterior to the 
inscriptions found there. The massive torus mouldings and the width of 
these and the othei honzontal courses pomt to the 10th or 11th century as 
ihc period of their construction. The curvilinear Hkhara of the Mahadeva 
temple IS made of brick and has been roughly plastered over in recent yeais 
The diaper lelievmg the outer walls of the shrine is preserved and, though 
simple m pattern, is executed with clearness and decision In the Jam 
temple the torus moulding is somewhat more slendei than m the other speci- 
men, and the former temple may perhaps on that account be attnbuted to 
a slightly later period. 


18 Midway between PSladS and Rarabar, which is two miles east of 
Paladi, are to be seen on the bank of the river Suktl two small iSiva temples 
A high wall surrounds the platform on which they stand The locality is 
known as ISmch-devali The name would lead us to expect a group of five 
temples at that place And m fact we are told that at one time some more 
shrines had stood there, some of which iwere earned away dunng the inunda- 
tions of the nveiv However that may be, there are only the two above- 
named fanes to be seen there now Above the doorway of one of the shrines 
there is an inscription, dated in the V 1231 referring itself to the reign of 
[the Chahamana] MahStajadhiraja Kelhanadeva — At some small distance 
from Rarabar, lymg in the folds of the mountams is a shrine dedicated to 
Mahadeva known by the name of Gangupaya. At the annual mela of this 
temple the Bhils and Menas of the neighbourhood assemble m large num- 
bers, make a confession of all their misdoings of the previous year before the 
Great God, and lay down at the same time a portion of the spoils at the 
God’s feet. And woe betide him who hides a secret guilt at this confession, 
for he lives not to see the year out ! The confession is meant only for the 
ears of the God and no royal official dares to disturb the smners m the midst 
of this confidential interview. So we were told by the pujan of the Paihch- 
devali shrines. 


19. Our next halt was at La§, about ten miles due north-west of Pala<Ji, 
with which we reached the northernmost pomt of our tour In LSS there 
are altogether five temples : two Jain and three Hindu. None of them show 
any architectural features of note. One of the Jain temples dedicated to 


Adinatha is undergoing repairs The old sabhamandapa has been completely- 
dismantled and a new one is being erected in its place The only object of 
antiquarian interest of the locality is a much mutilated but well-carved sculp- 
ture standing supported against the outer wall of the Lakshmi-Naiayana 
temple The sculpture represents a standing figure of a four-armed god and 
his consort Only the (proper) left arms of the god ai;e intact , one of tHi>m 
passes round the slim waist of his consort, while in the othei he holds a lotus 
bud. As the emblems m the other hands of the male figure are broken away, 
it is difficult to say with certainty what deities the fragment represents ; but 
the probabihty is that the group is a representation of Lakshmi-Narayana. 
On its base is an inscription (No 2762) dated both in Vikrama year 1344 
and iSaka 1209 and refernng itself to the reign of Mahardjadhtrdja Saihvata- 
si'ihha, the Chohfin king, several of whose inscriptions have been mentioned 
above. In 1 2 we also find the name JHahalipura— the sanskntised form of 
Jialor— which stands in all probability for the capital of the said Chohan 
chief, although on accoruit of the abraded condition of the stone it is not 
possible to affirm this with certainty. 

td?) GOL 

20. From Uas we proceeded to Gol, which is about eight miles due 
south of it Gol contains only two temples worthy of note — ^the temple of 
Amba Mata (Photo No. 4587) and the Chaubhuj temple (Photo Nos 4588- 
89)'. The former stands on a high platform built of ashlar masonry The 
small quadrilateral porch over the entrance steps widens out into an oblong 
hall with a flat roof supported on twelve pillars, two of which are in the 
centre of the hall One of the six bays into which the ceiling is thus divided 
culminates in a richly sculptured domical roof, while the others are merely 
covered by an arrangement of one square slab and three triangular slabs 
resting symmetrically on the architraves The massive mouldings running 
round the outside of the shnne wall indicate an age much higher than that 
of the poorly sculptured image of the goddess inside the shrine, on the base 
of which is biased the date V. 1752. In the niches in the southern and 
western walls there are placed images which are replicas in miniature of the 
principal imag^ inside the shrine Curiously enough on the north wall there 
IS no niche and the transverse courses run straight on undisturbed The 
goddess locally known as Amba MBtB is portrayed standing with a Sardula 
on dther side In the two (proper) right hands she holds a trident and 
rosary and in the lower left a kamattdalti. The object held in the upper left 
hand is not quite dear but looks very mudi like a bell On the right door 
jamb there is an inscription in a local dialect, dated in the V. 1752, like the 
one on the base of the principal image, and records probably repairs to the 
temple and the installation of the image The temple shows signs of having 
again undergone repairs m very recent times. — ^The temple of Chaubhuj 
stands on a high plinth and consists of three cells in a row, all of the same 



Size. The curvilinear hkhara of the central shrine is larger than those of 
the side shrines. In the middle cell there is an undated image ; but the 
image of Surya (on the left) and that of Vislinu (on the right) in the other 
shnnes are both dated in the year 1741 and refer to the reign of Verisala 
[Vairi^a] I, chief of Sirohi The temple is built out of odd members 
belonging to other and older temples. A low parapet runs round the plat- 
form on which the dirine stands. The skew brackets embedded in the front 
wall of the shrine indicate that the open space in front was intended to be 
covered by a domical roof. The niches at the back of the temple contain 
images of Brahma, Vishaju and the Sun ; those of Gajne^a and the Sun aie 
placed in the niches of the lateral walls Lying near the parapet adjoining 
the shnne of Vishnu I noticed a low marble stool, polyhedral in section, 
about a foot in height and eighteen inches in diameter The sides are cover- 
ed with a bold diaper-pattern and the top is carved m the form of a con- 
ventional lotus It IS the lotus emblematic of the Sun-god 

21. Frmn Gol we moved to J&val (Zawal of Quarter-Sheet 20 S. E.) 
and from there we visited Deldar and Mandvaria But none of these places 
contains anything worthy of notice En route from Javfil to SanvadS we 
halted at the village of Kalandri about six miles south-west of Javal. Be^ 
sides a modem temple of Vishnu there is at JavBl a Jain temple dedicated 
to Mahavlra, of perhaps the fourteenth century In the main shrine of 
this temple there is placed a sculptured panel representing in high relief a 
worshipper iupasaka) in the act of feeding a pigeon, a representation which 
has probably reference to some incident in the life of a tirthanikcna. Worthy 
of note is the short record of four lines below the sculpture, which is an 
unequivocal testimony of the prevalaice of religious suicide in the fourteenth 
century The inscription, which is dated V. 1389 Phalguna-sudi 8 Monday, 
records that on that day the whole of the Samgha committed suicide by 
abstaimng from food '{)aru^anena dtvath gatah). The names of the pro- 
minent members of the Saihgha who immolated themselves in this way are 
given. The name of the village is given as KBlathdrahJ 

22. At SanvaKte about fifteen miles due south of Kalandri we halted 
in order to visit the neighbouring villages of Tokara, AsSva, and Devakhetar, 
the antiquities of which will now be described in succession 


23. Tokatfi, which lies about two miles to the south-east of SanvSda, 
cannot now boast of more than a few isolated huts of shepherds, thou^ at 
one {68} time it must have been a large flourishing village. On the brink 
of the rivulet that streams through the hamlet, stands on a natural devation 
a nefdected group of dirines, the principal cme of whidi is dedicated to the 
SonfidhSri Mahfideva. In the south-west comer of the courtyard stands a 
dilapidated little shrine condsting of a cella and porch (Photo' No. 4590), 


Oq the architrave of the door-frame is sculptured the image of Ganapati. 
The curvilinear sikhara of stone is ornamented with a design of 
horse-shoe shaped chaiiya windows The finely carved amdlaka is enhre and 
in position The dinne is empty and serves as a store-room for chunam. 
On the left door jamb of the sanctum is engraved the following inscnption 
(No 2763) : 

1. Saimvatu 1232 Phaguaja-vadi 6 
2 deva-pratishta Rao Vtjadu hard" 

3. pitd. 

It thus attributes the “ establishment of the god ” in that ^ine to one 
Rao Vijadu. Now, among the places of anhquanan mterest described in his 
Hmdi “ History oj Snohl,” Pandit Gaunshankar Ojha has mcluded Tokara 
There Pandit Gaurishankar speaks of a short record incised on a pillar of a 
shnne situated in the enclosure above alluded to. The inscription, he says, 
is dated V. 1333 Phalguna vadi 6 and adds that the shnne was built by 
Rava Bijaida, the Cbahamaina king of that name The Pandit proceeds to 
conclude from this that previous to the said year, viz , V, 1333, the Devadas 
(the Chiahamlana family to which the present rulers of SiroM belong) must 
have eirtended their sway as far as Mount Abu, evidently because (as he 
imagines) the Chohan (Devada) kmg Bijad had erected a shnne at Tokara 
It will be nobced that all the details of the inscnption described by Pandit 
Gaunshankar agree with those given above by me except in the matter of the 
reading of the year. The date as given by the Pandit is 1333 ; while I have 
no doubt that the inscription I found on the very same spot bears the date 1232 
This is a serious discrepancy, what might be possible m 1333 is certainly not 
possible in 1232. Moreover, assuming for the sake of argument that the 
year as given by Gaunshankar is correct, even then, the absence of any royal 
title (such as MahSrSja or Yuvardja, etc.), coupled with the name of the 
donor, would be!, I think, a serious objecbon to any attempt at identifying 
RBo Bijadu of the inscription with any royal personage whateva". It is 
hardly necessary to point out that the abbreviation Rao of the inscnption 
may stand for Raval or Rfithod or any similar clan-name beginning with RS. 
Unless therefore the Pandit has through some oversight confounded the date 
of the inscription described above by me with some other inscription, it must 
be said that the learned Pandit is mistaken in his view regarding the actent 
of the possessions of the Devadas in the thirteenth cwitury. 


24. About two miles to the south-east of SanvSda is the villas of 
As&vft with which hangs a story of the slaughter of Btah m ai p a s and the 
subsequent act of atonement for this transgression. It is narrated ^t Ham- 
nflra, the younger brother of Jagam31a of Sirohl, was both avaricious and 
impetuous Blinded by his greed he attacked and seized by force several 

epigraphic studies 


villages belonging to his brothei During a raid on the village of AsaiS he 
slaughtered several Brahmanas ; whereupon thar widows immolated them- 
selves over the corpses of the victims of this aggression Subsequently through 
the intercession of the relatives of this Hamimra, the village of A^va toge- 
ther with a large part of the adjoining land was handed over in V. 1545 to 
Brahmanas as Brahmadaya, free of taxes and every other due. No royal 
official was ever to enter the village Such is, I beheve, also the purport of 
some Marvaltff inscriptions of the sixteenth century outside the local temple 
of Vishnu. In this same village there is a well-carved image of Hanumat 
'(about 5 feet high) of which the chief interest hes in the fact that it bears 
a date From an inscription on the base we leatn that it was fashioned for 
Bhunu^a, son of Viraslha, in V. 1355 Magha-sudi 10 (Inscr. No. 2764) 
The resinous oil poured over the image by countless devotees has formed 
such a thick crast over it that it is not possible to get a clear idea of the 
contour of the original sculpture (Photo No. 4591). 


25. Not more than a mile away from Asava lies the village of Deva- 
khetar and about two miles to the east of the latter are situated the ruins 
of a group of temples withm an enclosing parapet wall. Devakhetar was 
visited by Mr Bhandarkar in 1906 and its antiquities are described by him 
in the Progress Report for the yeah ending March 1907 I shall, • therefore, 
restrict my remarks to the description of three short inscriptions discoveied 
by me within the temple precincts. One of them, which is inased on a pwllar 
of the sabhcmtmdapp of the large Siva temple, records the obeisance of the 
Sutradhara and incidentally gives the name of the god as Sidhesar (SiddheS- 
vara) One often comes across short records containmg salutations of suira- 
dharas engraved on different portions of religious buildmgs. These names, 
I may add, are not of some sutiadkara or other who had come there on a 
pilgnraage, but of the plarticular sutradhara who bad planned and built the 
edifice For, such records are not confined to religious buildings, but are 
found in connection with secular buildings also, as, for mstance, in the Tower 
of Fame (Kirtham) at Chitorgad. The second inscription which appears 
to be dated V. 1230 (or 1234 ?) is engraved below an image of Siva and 
Parvati. It has suffered severely from thfe ^ects of weather and is almost 
illegible It contains the name Devakhetar, from which it follows that the 
name of the locality has remained unchanged during the intervemng mne 
centuries. The third and last inscription is a fragmentary record in a local 
dialect, engraved on the base of a column pertaining to the porch of the main 
temple. The inscription is of some importance for tiie reconstruction of the 
history of the PaiamSras, it refers itsdf to the reign of the ParamSra king 
SumaSha (Somasnhha) and bears the date 1293, which is the latest date 
hitherto found for him. The earliest date is that which we gather from a 

progress report of archaeological Survey of india, w. c 2y9 

record in the temple of Vastupala on Mount Abu dated m V 1287 cones- 
ponding to ad. 1230 


26 Hathal, called HathaMdi in the plates of the Paramaia Dhara- 
varsha found at the place, is situated two miles north-east of Hanadra, tlie 
headquarteis of the Tahsil On a low mound close to the boundary of the 
village of Hathal stand the rums of the two temples and a highly dilapidated 
small shrine The site, which is covered with a profusion of carved fiagmenis 
of various membeis of temple buildings, is now used as a public latnne ! 
Across the slanting face of the mound cow-dung cakes are put out for drying. 
The above-mentioned shrine appears to have been dedicated to 'Siva but the 
Linga IS imssmg The shrine mcludes a water channel on its north side, while 
the fragment of a Nandin lying close by leaves no doubt as to the god en- 
shrined there The two temples alongside ot this shrine are so hopelessly 
mutilated that it is no longer possible to say with certainty to whom they were 
dedicated. The remams of the lowest courses which are still tn situ show that 
they were both surrounded by an enclosing wall Within living memory one of 
these brines contained an image of Brahma which, having brought ill-luck 
to the village, was, we were told, despatched to the neighbouring village of 
Sakhav where it is still duly worshipped The third temple of the group is 
stated to have contained a representation of Sflrya Among the sculptures 
surviving I noticed the following an image of Brahma broken in two halves, 
a rudely carved image of Surya and a fragment which might have belonged 
to an image of Vishnu. There are several Maivail inscriptions in the village, 
but no impressions were taken of these, as the inscnptions are of a compara- 
tively modem date In records of the 15th century the name of the place is 
given as Brahma-sthana which indicates that the village was held as mam 
by Brahmapas In fact we learnt that it was made over to BrahmaJjas during 
the regime of the Paramaras of Abu about V. 1215 (ad. 1158) From tliat 
time date the stones engraved with representatirais of Siva Lihgas, Sun and 
Moon, cow and calf — all emblematic of Dharma — ^which are to be seen 
buned at intervals along the boundary of the village in order to mark off its 
limits This IS a novd idea I do not remember having seen boundary stones 
of that type elsewhere or even heard of them. 


27. From Hanadra I visited also the small village of Dhandhapur 
situated about two miles south-west of Hanadra. Several Paramara inscrip- 
tions of the twelfth and thirteenth centuries are to be found there, but most 
of th em are so weather-wom as to be absolutely lU^ble. In these old 
records the name of the place is given as Dhandhukapura, probably so called 
after the Param&ia king Dhandu(ka). Restmg agamst the wall of a square 



chkabutra in the village is a memonal stone on which is sculptured m high 
lelief a mounted nder armed with a spear Below is a short record of three 
lines giving the date V 1347 and the name of him m whose memory the stone 
was set up, viz , Arjuna, son of Param&ra Eatala Paramata is a very com- 
mon clan name among the Rajputs. So there does not seem to be sufficient 
reason for identifying this Paramara Patala with the Paramara chieftain 
Pratapasirhha and investing the latter, on the strength of it with a son of the 
name Arjuna, as one antiquarian has done. - 


28 While at Dbandhapur I was told that there were at a distance of 
not more than a couple of miles some mined temples which were well worth 
a visit and so I set out to look for them. After some hours of careful search 
in the (Wilderness I lighted upon the wreckage of two temples. The site is 
covered with lintels, columns, bases and capitals lymg in a wild confusion as 
though the ground on wludi they were standing had been convulsed by an 
earthquake. Portions of the masonry plmth of one temple are still mtact 
and in the debris lying around I discovered the fragment of a large slab 
forming part of the mandovara that was sculptured in high relief with the 
image of a iiTthatnkara seated in an attitude of meditation It may thus be 
that a Jam temple had stood there once upon a time Not many yards away 
lie the remains of another temple, conspicuous amongst which are the deeply 
carved fragments of the stone stkhara The presence of the ISiva Lmga pedes- 
tal is evidence of its bemg once a temple dedicated to Siva worship. From 
the character of letters engraved on the fragments, the temple may be asenbed 
roughly to the eleventh or twdfth century, a conclusion which is not at dis- 
cord with the style of the ornamentation of the sikharas which are deqily 
sculptured (with a design consisting of chcatya wmdows 


29 On leaving Hajnadia we halted at Revadhar so as to be able to 
iiispcct oHivemently some villages in the neighbourhood of the latter town. 
The first village to be visited from here was Jolpur situated about four miles 
from Revadhar. The tsily temple in the village is m a state of advanced ruin. 
It consists of three dirines standing on a sohd masonry plmth. The porch 
IS m a highly dilapidated conditiem. The middle shrme contains a iSiva LingR 
which goes by the name of Kalesar (KBle^vara) . The doorway of this shrine 
is elaborately carved. To judge from the general style of workmandup the 
temples cannot bdoog to a date anterior to the twdfth century.— But about 
a mile or so from Jolpur are standmg the remains of a comply of highly 
dilapidated temples which are several centuries older than the one just des- 
cribed The group, as it stands, consists of two large dirines and three sub- 
sidiary shrines standing at a short distance behind the former. One of the 
larger shrines contains fragments of a massive pedestal of a iSiva T-mga of 


which the yom lies just outside the porch Near the remains of the pedestal 
IS lyir« what appears at first sight to be a dwarf pillar with section changing 
from a square to a cirde through an octagon But as its dimensions agree 
exactly with those of the internal section of thd above-mentioned yont and 
pedestal, there can be no doubt that this shaft was once fitted into the drcular 
opening of the yoni and is the actual Lifiga. The other large shnne sheltered 
one whole and another mutilated image of Mahishasuramardinl The broken 
icon IS fashioned out of granite — a stone not usually met with in these parte 
— ^and though not unlike the other image, which is of soft limestone, exhibits 
much finer workmanship and greater finish of detail Of the three subsidiary 
shrmes, one is completely demolished (only the lowest course rmaining m 
situ), while the other which is better preserved is empty. The third subsidiarj.' 
shrine contains a well-carved image of Suiya, broken in twain across tlie 
knees The stone sikhara of most of these faneS have fallen in ; but wher- 
ever isolated blocks of these are still standing m position, their ornamenta- 
tion IS seen to consist of repetition in miniature of deeply carved chaitya 
windows and facades Many stones of the debris showed the same design. 
The size of the Sikhara may be judged from the fragraentaiy sector of an 
amalaka with a radius of twelve feet six inches For the antiquity of the 
fptnpipi speak the lai^ size of the undeoorated stone blocks [71} of which the 
walls are built, the massive size of the plain moulding and the large clear 
horse-shoe shaped chcdtya windows which form part of the designs of the 
deeply carved Sikhcera (Photo No. 4596-A) . 


30. About five miles to the north-west of Revadhar lies the village of 
Jiraval. In inscriptions of the 14th century in the local Jaina temple situated 
at the foot of a hill to the west of the village thd name of the village occura 
as Jir&iila or Jireila. The sanctum of this temple contains an image of Nemi- 
characterised by his IMckhana of the conch. But it is amply dear 
from the inscriptions (Nos 2773-80) engraved on the jambs and lintds of 
the dooswajte of the subsidiary shrines that the temple was oripnally conse- 
crated to Plar^vaifetha, a fact which' is well known to the inhabitants of the 
village of Jiraval, who give the following reason for the change of denomina- 
tion. They narrate that dunng the r^ime of a Muslim long (whmn they 
called B o ka ds PB.dashah) the tetrode was attacked, desecrated and plundered 
by a band of Muhammadan troops During this iaid the image of PSrSva- 
natha was pulled down and smashed to pieces by the bigotted iconodatts. A 
long time after, when the temple was resuscitated, an image of NemiMtha 
was made and mstalled in the place of the old image. The suhsidiapr ^nnes 
wkch form the enclosed aisle are all empty. The jairiw and hntels of over 
forty of them are inscribed with dedicatory inscriptions giving tte n^ of 
the donors, the quondam pontiffs, etc. The earliest of thepi is dated W 



V 1421 and the latest m V. 1483. The draiors seem to have been all Osval 
Banias ; and this is the class to which the present Bania community of tlie 
village belongs Vlsalanagara and Kalavanagara appear among the names 
of the' places of residence of some of the donors It is well-known that at 
Vlsalanagara there! was a large colony of Osval BaniSs The temple which 
is probably of the same date as the above-mentioned inscriptions, viz, the 
fourteenth century of the Christian era, contains no features of architectural 


31 The village of Varman lies along the main road leading -from Deva- 
dhar to Majidar, a little to the west of the SuMT river which is a tributary of 
the Banas. The place seems to have been at one time of considerable import- 
ance, but now it has lost its grandeur and is reduced to the condition, of an 
insignificant hamlet About a mile to the north of the village there is a gigan- 
tic Vaita (Ficus indtca) tree which together with its off-shoots covers quite 
an acre of land and must be a growth of untold generations. To the south 
of the village stands on a high eminence a Jam temple dedicated to Mahavira. 
The main shnne is empty. But a large image of Mahlavira which is awaiting 
ceremonial installation is placed temporarily m the gudhamandapa, which 
contains also several other smaller images Among these is a finely sculptured 
image of Kubera In the pillared corridor to the east of thel sabhdmandapa 
there is a sculptured ceilmg panel which bears an inscription dated in the 
year V. 1242. The central figure of this slab is Gajalakshma with elephants 
pouring water The original temple is probably not older than this sculpture. 
The sikhara whidi is very high, and therefore a prominent land-mark, was 
erected, we were told, within the last century Quite recently a large sabhor 
mtajidapa was added to the temple, as also a high wall enclosing the latter 
on all sides The columns of the scdfhdmandapa are not unifcsm and betray 
the fact of their having been brought over from the rums of older temples 
In the enclosing parapet are built in on the mside two coar^y sculptured 
panels, in one of which the central figure is that of Neminlatha and in the 
other an unidentifiable tirthomkara The village also contains a Siva temple 
called the! temple of Vartnesar (Photo No. 4617). In the enclosmg wall of 
this temple are built in sculptured figures which had once formed a part of 
the local SQrya temple Noteworthy is a large sculpture of Gajalakshrm 
placed in the courtyard of this temple Water drawn by dwarfs from the 
reservcar is passed on to women sitting on a higher level, who are represented 
as handing it over to elephants standing above them ; these in their tuin 
pass it on to a pair of dephants standing on a still higher level, which empty 
the kcdaSas over the head of the goddess seated on the komdldsana This 
sculpture, the design and execution of which are extremely happy, deserves 
to be transferred to and exhibited in a central musdum. But the object of 


sntiQUsnan int6r6St for which Vanwam is justly famed is the £72 J marble 
temple dedicated to the Sun-god which even now; in its decay is an impos- 
ing structure (Photo Nos 4598-4616) The careful finish of its caivmg, the 
proportion of its members and the parsimonious use of decorative detail, all 
tend to show that the building must have been constructed at a time when 
temple architecture was a vitally hving art The temple, which faces the east, 
consists of the shnne, sabhamandapa, pradakshma and porch. The oblong 
outline of plan is broken by projecting mches and windows from the mat}dupa 
and the circumambulatory (see drawing No 1392), The hkhara of the shnne 
has fallal a|way and the roof of the porch and maij.dapa is also partly de- 
molished Where the foundation has subsided or the pillars given away the 
loosened stones have shd off one another and are lymg scattered round the 
min (Photo No, 4598) In the debns I discovered a standing image of Surya 
(broken clean aicroes the knees)' which must have occupied the main shnne 
(Photo No 4612), I discovered also findy carved but partially mutilated 
images of the navagrahas, and the eight dikpdlas (Photo Nos, 4615-16) The 
finest piece, howevelr, is a mutilated group, the pnnapal figure of whicli is 
the form of the Sun-god called Surya-Narayaiija (Photo No, 4609), To Uiis 
group belongs the pedestal resting m the mche in the western wall of the 
sanctum (Photo No 4600) The pedestal is sculptured in the form of a 
chariot drawn by seven steeds which is a marvdlous piece of realism. Un- 
fortunately most of the sculptures found here are fragmentary, and even tlie 
fragments are highly mutilated Several of the pillars of this temple are 
engraved with inscriptions (Nos 2782-87) in which the god of the temple 
i6> referred to as Brah mSna -sv5min One of the mscriptions belonging to the 
leign of the Phramara Puripaplala, son of Dhandua (Dhandhuka) , states that 
in V, 1099, Jyeshtha-sudi 30 Wednesday, Nochaka, son of Sarama, repaired 
the temple. Another dated V 1076 Chaitra-sudi 7 {jatha-sopttwiV) records 
jthat Sohapa, while on a visit to the temple, presented to the god two fields, 
A third inscription belonging to theJ “ prosperous and victorious reign ” of 
Mahdrdjakula Vikramasiimha, and dated m V, 1356 Jyeshtha-vadi 5 Monday, 
gives the place-name clearly as Brahmaipa-mahasthBna, There can be, there- 
fore, no doubt that BrahmSiija is the sanskritised form of Varmaij, Three 
other records are dated respectively in V. 1315, 1330 and 1342, 

32 As it would not be possible to do justice to larger questiwis of 
aichitectural and iconographic mterest connected with this temple without the 
hdp of drawings and illustrations which cannot be reproduced here, I intend 
contnbuting an illustrated article to the Director-General's Annual of Archseo- 
logy where these subjects will be discussed in detail. 

July 1917, 

Offg. Asststmt Supermtmdent, 
Archaologtcd Survey, Western Circle. 

SURVEY OF INDIA, 1917-18# 



(A) Epigraphy. 

I.-^Hindu and Buddhist Inscriptions. 

1. In December last Dr Sukthankar received from Mr. Subaya 
Nagappa HEca)E of Ajjibal m the Sirsi Taluka of the North Kanara District 
two sets of copper-plates for inspection. The plates have been preserved as 
curiosities in Mr Hegde’s family durmg several generations , so that it is 
not definitely known now how they came into the possession of the family. 
One of the plates refers itself to the reign of the Kadamba king Ravivaiman, 
and the other to that of Kn^avarman (probably the second king of that 
name) belonging to the same family The regnal years in which the grants 
are dated are worthy of particular notice The plate of Ravivarman (if Dr 
Sukthankar’s readmg is correct) is dated in the thirty-fifth year, and that 
of Knshnavarman in the mneteenth year, of the kind’s reign It should be 
remarked that the only other hitherto known grant of Knshnavarman II 
refers itself to the seventh year of his ragn ; while the highest regnal year 
recorded in the copper-plate records of Ravivarman is the eleventh The 
uncertainty regarding the date of Ravivarman's grant is due to the fact that 
the words compnsing the date have been almost completely eaten away. 
We have, therefore, to depend upon a conjectural restoration of the words ; 
but Dr Sukthankar’s conjecture has every probabihty in its favour. Both 
the plates have their nngs and seals attached The seal of Ravivarman’s 
plate is blank, but that of the other plate bears a horse as device. Ravivar- 
man’s grant, which is dated on the fifth tithi of the bnght half of Karttika 
in the [thirty!] -fifth regnal year of the king’s reign, records that on the spea- 
fied day Ravivarman of the Kadamba family granted four nivarttanas of 
land in a village called Sare (or Sara) to the temple of Mahadeva of his 
beloved physician Nilakantha Some further specifications of the donation 
are lost in a lafuna. The grant of Krishnavarman records that on the full* 
mo(Ki day of Karttika in the nineteenth year of the kmg’s reign, he granted 
Kamakapalli in the village of Gingade situated in the Karvannadga District 
From the topographical information supplied to Dr. Sukthankar by Mr. 
Hb(^e, he concludes that the Gingade of the grant is to be identified with 

* [Pp. 3^7.] 

progress report of archaeological survey of indla, w! 

the modem vdlage of Gmgadde m the Sirsi Taluka, while it is ccmjefitjhgj? 
that Kamr, which is the name ot a neighbouring village, may not be uncon- 
nected with the district name Karvannadga of the grant 

2. To the keen mterest taken by Mr. P. B Gothoskah, Librarian of 
the Bombay Branch of the Royal Asiatic Society, in the 
Two Chalukya search of Indian antiquities, we owe the recovery of two 
interesting copper-plate chaiters purporting to be issued 
by the Chaulukya KJamadeva of Anahilapataka It was after a great deal 
of trouble that Mr. Gothoskar succeeded in obtaining the loan of them from 
him for the purjjose of photographing them The negatives have been pur- 
chased by me for this department, and will be filed in my office. It is in- 
tended to contnbute a detailed descriptive note on them to the Journal of 
the Bombay Branch of the Royal Asiatic Soaety. Dr Sukthankah, in 
whose hands the photographs have been placed for publication, in his report 
on them, sasre that both the sets refer themsdves to the reign of the Chaulu- 
kya Rarna and zire dated respectively on Tuesday, the deventh day of the 
bright half of Margasirsha in the Saka year 996, and on the eleventh day of 
the bnght half of Karttika in the Vikrama year 1131 The cunous arcums- 
tance regarding these grants which are dated on different days is that both 
of thpm are made in favour of the same person and convey the very same 
village. The wording of the grants is, however, quite different in the two 
plates. Moreover there is also great difference in the writing • while on one 
the letters (to judge from the photographs) are deeply cut and uniform m 
size, though their shapes are neither good nor neatly fimshed, on the other 
they are diallow and very poorly cut and their diapes are distinctly ill-made 
A gain , while the first grant begms with a salutation to Vasudeva and a 
mangala stanza, the other begins abruptly with the genealogy of the Chaulu- 
kya kings It is as difficult to give a reason why two grants diould have 
been made conveying the same village to the same person as to explain the 
difference in the dates and the wntmg. It does appear though, as if the 
first set, namely, the one that is evidently the better of the two, is the original, 
genuine document ; the other seons to have been made later in imitation of 
it, as a suhafitiite for it The grant was made by the Mahatnandalesvara 
Durlabharaja bplnngmg to a feudatory Chaulukya family of Nagasarika 
(Nausari) which acknowledged the suzerainty of the Gujarat Qhaulukyas of 
Anhilvada. The preamble of one of the grants contams the genealogy of the 
donee up to three generations. Durlabharaja, we are told, was the son of 
Chandraraja, and the grandson of Gamgeya of the anci^t lineage of the 
Chaulukyas. The donee was tbe Brahmana Pandita Mahidhara, son of 
Rudraditya of the Mandavya gotra who had come to Nausan from Madhya- 
The object of the grant was the vill^e Dhamalachchha, situated ap- 
parently in the district of Talabhadnka Thirty-six. The boundaries of the 
village are given as follows ; — to the east, Kalagrama ; to the south, Torana- 



grama , to the west, Avala (or AmvaJa) sati-grama ; to the north, Kachch- 
havaU-grama. The Atlas sheet gives two villages called Dhamadachchha and 
(to its south) Tarangam as situated in the Nausan District In r^ard to 
the fact that the donor was a Jagirdar of Nausan, as also that the plates are 
now stated to bdong to a resident of Dhamadachchha there is no difficulty 
in the way of identifying Dhamalachchha with Dhamadachchha and Torana- 
grama with Tarangam. The other place-names remain unidentified It is 
perhaps worth noting that in the grant which is above held to be the onginal 
document, the portion containing the boundaries is written at the very end 
of the documait and was added secunda mam, which is palpably different 
from that in which the rest of the grant is written, and which rather resem- 
bles the clumsy lettering of the other grant under reference The problems 
raised by this pair of grants cannot thus all be looked upon as solved 

Two Valabhi 

3. A set of two copper-plates was sent to this office for examination by 
the ^avnagar Darbar, which the Darbar has smce pre- 
sented to the Trustees of the Pruice of Wales Museum 
Dr SuKTHANKAE, m whose hands the plates have been 
placed for decipherment and pubhcation, reports that they are dated in 
samvat 210, and were issued by order of the Mahasamanta Maharaja Dhru- 
vasena I, the Maitiaka kmg of Valabhi The seal, which is attached, bears 
the usual Maitraka device and legend The charter records the gift made 
by Dhmvasena of certain lands at the villages of Chhedakapadraka and 
Malakara m the Hastavapra-aharani to a certam Nanna residmg at Vala- 
padra, for the performance of sacrificial ntes. The exact date of the grant 
IS the 13th tithi of the bright fortnight of Sravana m the year 210, which 
year when referred to the Gupta-Valabhi era yidds ad 529-30 as the ap- 
proximate date of the charter. At Baroda, Mr. J. C Chatterji, the 
Dharmadhyaksha of the Baroda State, showed to Dr. Sukthankar a -dnglp 
copper-plate which was sent to the Dharmadhyaksha from Kathiawad for 
decipherment. The plate on examination was found to contain the latter 
half of a Valabhi charter dated samvat 206, Asvma-sulda 3, and issued, like 
the previous one, by order of Dhmvasena I The donee was Botghamitra 
of the Vrajagana gotra, a resident of Simhapura, which place is to be identi- 
fied with Sihor, near Bhavnagar, a junction on the Sihor-Pahtana Railway. 

4. Dr. Sukthankar has in hand for editing two mteresting ins- 
criptions engraved on the pillars of an old grammar 

Sala at Dhar, from the time of 
the Paraanaras of Dhar The inscriptions are known 
as soTpcbofidha, because they are engraved in the form of intertwining ser- 
pents with thfiir bodies twisted lengthwise and crosswise leaving oblong spa- 
ces within for letters. One of the inscnptions is a chart of the Sanskrit 
alphabet and the other of verbal tenninations. The latter are taken from 
a chapter of the Sanskrit gr amma r called the Katantra which was specially 


intended for the instruction of people who did not care to penetrate too 
deeply into the comphcades of the Sansknt grammar It is worth notir® 
here that the first few chapters of this simplified grammar are still leamt 
by heart m the mdigenous vernacular schools of Malwa. Gujarat and som e 
other parts of India. Alongside of one of the tables is engraved a pair of 
stanzas which contam the names of the Paramara Naravarman and Uda- 
yaditya of Malva and imply that the tables were engraved by order of Uda- 
yaditya (cfl A.D 1150) 

5 In August H917 I proceeded to Sanchi to examine and take estam- 
pages of a short stone inscription which was discovered 
inscnp- ^ ^ village near by The inscripfaon proved to be 

tion of the time of n very interesting one. The first hne opens with an 

Svami Jivadaman. eulogy of Skanda the Commander of the celestial army 
and ends with the name of Jivadaman The second 
and third lines record the name of a General or Judge (Mahadandanayaka) 
Sridharavarmman the Scythian (Sake) and the thirteenth year of his reign. 
The object of the mscnption is to be found in the last Ime (in a verse, only 
a part of which is extant), namely, the excavation of a well The record 
IS m a very bad state of preservation ; the first part of 1 1, and the major 
portions of D. 5-6 have entirely disappeared. The last verse is followed 
by two numencal symbols which I read as 200, 1. These symbols are very 
much like those usipd in the dates on the corns of the Western Satraps of 
Saurashtra They are not preceded by any words or symbols that usually 
introduce a date and therefore their import is far from dear There was a 
Svami-Jivadaman whose son Svami-Rudrasimha II succeeded the Kshat- 
rapa Vlsvasena in Saka 226-^7. As his son’s date is S. 226-27, it is quite 
possible that the numerals in the Sanchi inscription denote a date in the 
Saka era. If I am correct then the Sanchi mscription provides a date and 
a locaticxi for Svami-Jivadaman, the father of the founder of the third Dy- 
nasty of Satraps in Saurashtra, who was hitherto known to us from the 
coins of his son only. 

6. During the year under review I was engaged in dedphering a new 
dated mscnption in a small cave near Asoka’s edict at 
ins<^tion Dhauh m Onssa, which records the visit of a pilgrim 
at a 1 , ssa. j^gj^ ^ jmjg named Santikaradeva, who is 

known from another votive record in the Ganesa Cave Khandagiri. The 
only interestmg feature of the inscription is that it is dated ; but the date 
cannot be referred to any known era except that of the Eastern Gangas. I 
was also engaged in editing two copper-plate grants found m the Native State 
of Baudh for the" Bpigraphia Jndica at the request of the Government of 
Bihar and Onssa A summary of their contents by Mr. H. Krishna Sastri 
has already appeared in the Amual Report of the Supenntendent, Arch®- 
ological Survey, Eastern Cirde, for the year 1916-17, 


epigraphic studies 

II.— Muhammadan Inscriptions. 

7. A large number of Muhammadan mscriptions were copied during 
the year. The majority of these are unpublished and some even unnoticed 
(a) Sultans of Malwa— An mscription of Sultan Alauddin Mahmud 
Shah Khilji was discovered by Cunningham m a modem 
Lohangi Pit. Masjid on the top of the hill near Bhilsa railway station 
on which the tomb of Lohangi Pir stands. It records the erection of a 
Masjid by one Khojendi who bore the titles of “ The sword of the State” 
(.Satf-ul-tnulk) and “The Lord of the east” (Mahk-ush-Sharq) during the 
reign of Sultan Alauddin Mahmud Shah Khilji in the year 862 ah. (1457 
AD.), Cunningham read the date as 864 a h. The Jami Masjid at Sipn, 
the summer capital of the Maharaja Scindia of Gwalior, 
^ Masjid, ^25 built a few years earher. The inscription on this 
monument was pomted out to me by Mr M. B. Garde, 
B.A., Inspector of Archaeology, Gwahor State Unlike the majonty of the 
Muhammadan inscnptions of India this record is incised It records the 
erection of the Jami Masjid dunng the reign of Sultan Mahmud Shah Khilji 
m the year eight hundred and forty-five (1441 a.d ) by Muhammad Tarkan 
and Ahmad Tarkan The date of this record is given both in words and in 

(ft) Sultans of Gujarat — ^The only dated inscnptions still to be found 
among the mins of Champaner are those on the two 
citadel Each of these gateways, now 
called Halol and Godhra gateways, bear inscribed slabs 
The inscnptions on both of the gates are identical. The latter half of that 
on the Godhra gate has become illegible in many places, but it has an ad- 
ditional line mcised vertically which gives us the name of the scnbe Both 
of the inscriptions contam the name of Sultan Nasir-ud-din Abul Path Mah- 
mud Shah, son of Muha m mad Shah (II), son of Ahmad Shah (I), son of 
Muhammad Shah (I), son of Muzaffar Shah, and the date, which is the 
month of Zi-l-qada 889, A.H (1484 ad ) 

(c) Sultans of Bijapur — ^The majority of ancient monuments in the hill 
fort Panala are stiU intact and what is still more interesting, the inscriptions 
on almost all of them are still m position The mins on this fort, which was 
the scene of great revolutions m the history of Deccan, are mentioned in 
CousENS Revised List but not in drfait None of them appear to have been 
surveyed prior to my visit in September 1917. Some of the inscnptions are 

* [In the original file copy of the reprint Dr. Sukthankar has entered his 
agnature just before this sacUcm. It is, therefore!, not dear if he is responsible for 
the present section , but it is repiodooed her^ m view of the fact that his author« 
al^ of the section is not improbableif— Cf, his Catalog of Antiquities in the Bija~ 

pHT Ed.] 


mentioned in the Bombayi Gazetteer, but none of them appear to have been 
pioperly noticed or published as they are not mcluded in Dr. Horowitz’s list 
of Muhammadan inscnptions published in the Epigraphta Indo-MosUmica 
They were copied for the first time in 1917. The oldest of is an 
mscnption found in the Somala tank which records the erection of a tank 
(httuz) during the reign of Sultan Mahmud Shah Bahmani by a nobleman 
named Adil Khan Ghazi. Unfortunately the date of the inscription is miss- 
ing which makes it impossible for us to identify this 
^ Adil Khan The rest of the inscriptions were incised 

during the rule of the Adilshahi Sultans of Bijapur and most of them contain 
their names The outer gate of the Tin-Darwaza bears 
m- aza. which records the erection of the Fort of Panala, 

which IS called the Gate of the Kingdom (Dar-tis-saltamt) m the year 954 
AH, (= 1647 AD.), dunng the reign of Ibrahim Adil S hah I. A small 
spring on the hill-side was converted into a walled re- 
Nagjhari servoir by one Daud Aqa, evidently an Abyssinian, in 

the next year (955 a.h. = 1548 ad.) during the same reign. A large tank 
(kauz) was excavated near the site of the Char-Dar- 
waza by one Malik Jafar during the reign of Ibrahim 
Adil Shah I m the year 964 ah. ( = 1556 ad) It is one of the largest 
reservoirs excavated on the top of a hill An inscription built into the 
walls of a modem Hindu temple near the site of the Char-Darwaza gate 
records the erection of a tower {burj) during the reign of Ali Adil Shah I 
by one Shamsuddm Shahaswar in the year 985 a H. ( = 1577 aj).). A large 
inscnbed slab belonging to the Char-Darwaza gate has 
Char-Darwaza. ^ modern tomb in the courtyard of the 

Mausoleum of the local Muhammadan samt, Sa’ad-ud-din, familiarly called 
SadobrL It records the erection of a gate of the fort by one Maqsud during 
the reign of Ibrahim Adil Shah II m the year 994 a.h. (= 1585 a.d.) . The 
readoice of the former Qiladars is now used by the 
Qiladar’s Palace, Kolhapur as a guest house An inscription, 

now in the walls of one of the chambers, records the erection of a 

palace (makal), by one Maqsud Aqa, during the reign of Ibrahim Adil 
Shah II in the year 1000 a.h (= 1591 a.d.). The builder of this place, 
Maqsud Aqa, appears to be the same person as the one who built the Char- 
Darwaza gate of Panala Fort six years previously. 

(B) Nunusmatics. 

8 No coins havmg any special significance were discovered in the Pro- 
vince dunng the year under review The Treasure 
Inscribed Puri- Trove coins sent to me for examination by the Govem- 
Kuahan Com. nipn t of Bihar and Orissa contained some imique coins 

■The most important among these is a copper coin of the type which is called 
Pun-Kushan by Numismatists. This coin with several others of the same 



t)T)e were sent to me for examination by His Honour the Lieutenant Gov- 
ernor of Bihar and Orissa and were found among the collection of the late 
Chief of Baudh. Coins of this type are found in large numbers in Orissa 
but hitherto no inscribed specimen has been discovered. The com found 
in the Baudh State is unique inasmuch as it bears a legend This legend con- 
sists of two syllables only tanka “ a coin The characters belong to the 
north-eastern variety of the Indian alphabet. Incidentally the coin helps us 
to fix the date of the Puri-Kushan coinage for which we had no reliable data 
so far 

A find of 448 silver coins of Farid-ud-dm Sher Shah found in the Shah- 
bad District was sent to me for erammation This find 
New hCnts. contained some specimens from a new mint • Pmduah. 
There is a town of the same name which for sometime was the capital of 
the Mussalman sovereigns of Bengal This town was a mint town durmg 
the reign of the Independent Sultans of Bengal on whose comage it appears 
as FiTUzabad. It is situated a few miles to the noith of Gaur or Laghnauti, 
the ancient capital of Bengal. The same hoard contamed a few specimens 
of the issues of Sher Shah from the mint at Chunar. Half a century ago 
Mr E Thomas had pubhshed an umque coin of Sher 
and types of the Shah of the same mint. On his com the mmt name is 
Chunar but on all specimens m this find it is spent 
Chanarh, which is to be pronounced either as Chanadh 
or as Chanara The find contamed a new type of the Kalpi mint, which 
has a circle of arabesque work on each side instead of the plain circle. 



Ladies and Gentlemen, 

Indology is, as you all know, the discipline which has for its object the 
study of Indian literature, history, philosophy and kindred subjects. Indo- 
logical studies, rri the widest acceptance of the term, may be said to date 
back to the distant penod of hoary antiquity which witnessed the birth of 
those truly remarkable specimens of Imguistic analysis, the Nirukta of 
YIdca, the SiksSs, and -PrStiSildiyas, which contain such a wealth of 
significant phonetic, etymological and grammatical observations on the Vedic 
SatiihitHs. The beginnings of hnguistic study in India must be even older 
than these works ; but the results of those early speculations had probably 
not crystallized into systematic treatises ; and if they had they have certainly 
not been preserved 

The tradition of these scholastic pursuits had been carried on, more or 
less unmterruptedly, during the intervenmg centuries or rather millennia ; 
and we can look back with pride, admiration, and gratitude, on the massive 
literary monuments left to us by our ancestors, by — to mention CHily a few 
well'known names— able grammarians like PSnini, Katyllyana, Patanjali, and 
Hemacandra ; commeitators like A^vaghosa, MalhnStha, and Sayam ; 
rhetoricians like Bharata, Bhimaha, Dandin, and Vamana , historians like 
VSkpati, Bilhana, and Kalhana ; not to speak of the unknown compilers of 
Puranic genealogies, the writers of the history of the Buddhist and Jaina 
/•h^ tyhes , the chroniclers ofl the lives of saints ; and, lastly, the keepers of 
the records of Hindu cloisters and monasteries 

This vast store of fact and fiction, accumulated through the critical, 
exegetical, and historical, activities extending over centuries has been studied 
and re-studied in recent years by succesave generations of schdars. This 
thesaurus has finally { 9 ^ been turned into a searchlight, and made to 
illuminate the obscure periods of the history of our country, and to con- 
tribute its quota to the duddation of problems thrust on the threshold of 
our consciousness by the Memory of a half-forgotten Past, in other words, 
by Communal Memory. In our own times and here in our midst, the torch 
has been kept alight throu^ the zeal for learning of scholars hke Bhag- 
v anlal iNDRAji, Sir Ramakrishna Bhandarkar, and Skams - ul-ultna Dr. 

* [A discourse delivered on 20th August 1923> at a gathentig of the Institute, 

on the 14tb Anniversary of the late Mr. K R. CAMA, — ^Journal, 19^4, pp. 93-104, | 



Jivanji Modi, men who have nobly consecrated their lives to the work of 
unravelluig the history of India, of interpreting Indian life and thought, 
and have worthily upheld the scholarly traditions of this Land of Rais. 

Within the last two decades the domain of Indology has extended by 
leaps and bounds , it has scpanded, so to say, both vertically and horizon- 

The advance made in the comparative study of language literature, 
mythology and art, the exploration and excavation withm and without the 
confines of India, has each served to advance our knowledge of the past 
in multifarious ways. While this advance has happily solved certain old 
riddles, it has in turn given birth to others that are entirely new I have 
proposed to myself to take ypu out this evening for an excursion on what 
may be termed the periphery of Indology, m contradistinction to the centri- 
cal portion which concerns itself with the inteipretation and reconstructiion 
of the past from sources purdy or chiefly mdigenous I propose to acquaint 
you with soma of the results of research and explorations in fields lying im 
the honzon of our cultural influence, indicating briefly those problems that 
have arisen in the wake of the progress of our research. 

I will commence my review with a survey of topics which are not exactly 
new, but which lie out^de the conventional grooves of Indian research, as 
the term is understood in India. 

It has been surmised that long before the commencement of the Chris- 
tian era, the Dravidian races had developed mdependently considerable 
culture of their own. Some of these Southern kingdoms earned on, for a 
protracted penod, a thriving trade first with Western Asia and EgypL and 
then with the Greek and Roman Empires. Literary evidence appears to 
suggest that the Tynans imported from South Indian seaports ivenry, apes 
and peacocks. And we have dso evidence to show that at a still later 
epoch India— 4o a great extent South India— exported nee, spices, precious 
stones, and a large quantity of doth, mudm and silk Who were these 
adventurous traders on the Indian side ? How did they solve the problems 
of tian^rt and exchange’ How long did that trade continue, and what 
stopped it in the end? These are some of the questions one may readily 
adc. They are not however as easily answered. The question of the early 
commerce of India with Babylon was examined at length by Kennedy in 
JRAS., 1896. In recedt years Egypt and the sites of andent Assyria and 
Babylonia have been systematically explored, and objects of antiquarian 
interest have been recovered from those sites on an unprecedented scale. We 
have therefore every reason to hc^ie that when the new material has been 
thoroughly sifted and studied by experts, the results of their investigations 
win confirm the surmises and condusions based on hterary evidence, and 
throw additional light on the obscurities relating to the early intercourse 
between India and the Western World, 



We will next turn to a field where the intercourse between India and a 
foreign country, if not so ancient, was evidently much mcMB extaisive, and, 
l 3 ang as it does within the histoncal period supplies far richer material for 
study and investigation. 

One cannot imagine a more fruitful field for a study of the evolution 
on foreign soil of Indian thou^t, and Indian art and architecture, than the 
little island of Java The ascaidancy of Indians is really the first great 
epoch m the history of Java The Javanese temples which still bear the 
name Cha|n)di BSma, Chandi Kah, Chandi Durgfi, Buro Buddur and the like 
tell their own story, which in part is set down on stone in mddible letters. 
The island abounds m splendid templds and vihlras of the Hmdu period, 
and they are noteworthy examples of an architecture which attained, as in 
India, a high standard without the use of mortar and arches 

The most important of these ruins is the temple of Buro Buddur, which 
has justly been characteriaed as a great picture Bible of the MahSyana creed, 
and which ranks among the architectural marvels of the world. Buro Buddur 
is not really a temple, but rather a hill encased with imposing terraces con- 
structed of hewn lava blocks surmounted by a dagoba, and crowned with sculp- 
tures illustrative of the MahaySna doctrine The subjects treated in the 
lowest endosufe are of the most varied descnption, forming a picture gallery 
of landscapes, scenes of outdoor and domestic life mingled with mythological 
and religious designs. As one proceeds, the subjects grow (96} in d^th 
and complexity It would seem that the architect had intended gradually 
to wean the devotees from things of this world. When they once b^in to 
ascend from stage to stage of the temple hill, they are introduced to the 
realities of religion, and, iby the time they reached the dagoba they had passed 
through a process of instruction and were ready with enlightened eyes to 
enter and behold the image of the Buddha, symbolically left imperfect as 
beyond the power of human art to realize or portray. 

The ruins in Java are by no means exclusively Buddhist There are 
temples devoted to the cult of Sivaism also Here we come across sculptured 
panels representing ISiva as a Yogi and agam as Kila or Time the Destroyer, 
letniniscent of similar pands at Elephanta. 

Col. Yui^ has pomted out that there are distinct traces of a fine coat of 
stucco-covering on the exterior and interior of Javanese buildings, and he 
has compared m this respect the cave walls of Eluxa, the great idols at 
Bamian (a once renowned town of Afghanistan) and the Done order at 
Selinus (an ancient city on the southern, coast of Sdaly) 

The Indo-Javanese remains have been in part photographed and studied 
in recent years by Dutch archaeologists. But it is desirabl^aiid it is Iiigh 
time— that these ramifications of Indian culture should te studied from m 
I ndian view-point by Indian archaeologists, who are familiar with Buddhist 



and Hindu mythological and religious lore, and are conversant with the deve 
lopment of Indian art and architecture. It is needless to emphasize the im- 
portance of these archaeological remains for a study of Buddhism or of the 
ancient and mediaeval Hindu art 

Important results are likely to be obtauned in other fields as well from a 
study of Javanese antiquities. The Mahfibliirata was translated into the 
Kfivi language about 1000 ad. And in this translation we find embedded a 
large number of Sanskrit verses, and hemistiches ; the prose narrative, more- 
over, reproduces very frequently Sanskrit words and phrases. As we happen 
to know the exact date of the translation, it is a very valuable asset in Maha- 
bhfirata criticism. In our gropmgs in the dark recesses of Indian history, 
we have to accept gratefully even such feeble and precanous guidance This 
KSvi version is, I may add, being used with great advantage in the prepara- 
tion of the new and critical edition of the epic undertaken by a sister Institute 

We shall next turn to Iran With Iran our connection dates back 
to prehistoric times We are all familiar with the evidence which estabUshes 
the connection between the Hmdus and the Iranians, through affinities in 
language and tradition, religious beliefs, ritual observances and even 
manners and customs. In more recent times further evidence has i>ppti 
placed m our hands by certain cuneiform inscriptions on clay tablets which 
the German Professor Hugo Winckler discovered m 1907 at Boghaz-koi (die 
ancient Pteria) in iCapp|adoccia I shall not go into the details of this find, 
because the subject was dealt with at length m the course of a learned dis- 
course CHI ‘ Indo-Iranian nugrations m the light of the Mitani tablets,’ deli- 
vered in this very Institute on a similar occasion not many years ago by the 
late Dr. GUNE of Poona I may be permitted, however, to refer to the find 
briefly as it falls within the provmce mapiped out by me for survey. The 
tablets contain a record of treaties concluded between the king of the Hittites 
and the king of the Mitanis about 1400 e c. The only fact that concerns us 
here is that the treaties include the mvocations of the tutelary deities of the 
respective kings for protecting the solemn agreements contemplated ; and 
among the gods called to witness are deities common m part to India and 
Persia. We find here the names Mi-it-ra, U-ru-wna, In-da-ra, Na-sa-at-ti-ia. 
One easily recognizes m them the Vedic gods Mitra, Varuna, Indra and the 
NSsatyfi respectively. The inscriptions, as I have said, date from about 1400 
B.C. and the names appear not in the Iranian form but, so far as we can judge 
from the imperfect orthography of cuneiform inscriptions, m the form whidi 
they ^ow in the hymns of the Bgveda. We may ask : Are the four daties 
invented in these Mitani tablets proto-Iranian or Vedic or Aryan ? Were they 
the gods of a tribe which was on its way to India, or of a tribe which had 
retraced its steps and returned to an earlier home? Or ^ere they again 
merely borrowed gods ? Did the king of the Hittites or the Mitams worship 
Vfiic gods? Unfortunately this tantalizing find suggests many more 

An excursion on the periphery 315 

questions than it answers. There is nothing to be gained by 
dogmatizing over the results of this discovery, though it is galling to realize 
and acknowledge our helpless ignorance All that we have to go upon is that 
in the 15th century B c four gods who figured conspicuously in the Vedic pan- 
theon were in thei region round Boghaz-koi considered, for some reason, 
fit to be invoked as supernatural witnesses to a solemn and important p 
tieaty Thisi information is clearly too flimsy for the erection of any* solid 
superstructure of theory But it may be pointed out that the mention of the 
Vedic gods m these treaties is but the crystallization at one particular point 
{98} of a diffuse complex, which could not have subsisted unsupported, so 
to say, hanging in the air. It necessanly implies reflexes, reactions and rami- 
fications, which it will no doubt be possible to isolate with the increase in our 
knowledge and the refinement of our instruments We may reasonably hope, 
may expect, that these discoveiies are but the first fruits of a nch harvest 
which may be reaped by patienti study and untiring exploration 

We shall pass on to another field whore we shall be on more solid ground. 
There was a time when Pali was regarded as a subject lying on the frmge 
of oriental scholarship. And our worthy University, faithful to tlie old- 
world standards, probably still regards Buddhism as a heterodox xehgiQn. 
And yet in Pali, it may be said, lies enshrined much of — not only Indian — 
but OriQital culture. The spread of Buddhism from India to Central Asia 
and thence to the Far East is probably one of the moat important contribu- 
tions India can daim as havmg made to the general uplift of mankind 
Ilavmg discarded Buddhism as a rdigion, we are apt to overlook the fact that 
this rdigion, which had its birth in Indiai, is still the rehgion of China, Japan, 
Siam, Burma, Ceylon, and Nepal It is still the credo of millions of men 
and'women. Gandharan art, which was adopted by the Buddhists of North- 
ern India as a medium for expressing its ideals in plastic form, was earned 
by Buddhist missionanes m pamtmg and sculpture to the cases of Central 
Asian deserts and thence to Chma, Japan, and Korea 

Outside India Buddhism found the most fertile soil in China. An inti- 
macy with the Chinese language has long come to be regarded as an essential 
pre-requisite for a thorough study of Buddhism ; in fact it may be said that 
nowadays one cannot do justice to Buddhistic studies without a first-hand 
knowledge of Chinese sources. But Buddhism reached the Middle Kingdom 
not directly from the land of its birth but, as I have already hinted, by the 
route of Central Asia. Thus m following the outward and onward march 
of Indian culture, our eyes ate first turned towards Central Asia, and especi- 
ally to the highlands of Pamir, and to the oases of the Gobi and the Talkla- 
makan deserts. 

The first convincing proof that the arid soil of Chinese Turkestan held 
buried valuable archseological treasures was furmshed by the senes of finds 



of Sanskrit manuscripts, of which almost the first was the Bower Manuscript 
discovered accidentally in 1890 Apart from' their philological interest these 
manuscript finds had value m showing that Sanskrit, the sacred language 
{99} of the Brahmans, w^ cultivated, and assiduously cultivated, even in 
those distant regions beyond the Hindukush, at such an early period. The 
f^pptrf-qt if^ns raised by the discovery of these manuscripts' have been amply 
justified. In fact it may he said that dunng the last thirty years no other 
undertakmg has been more fruitful for the study at once of Indian, Iranian, 
and Far Eastern history, has opened out wider vistas for research, laid bare 
higher treasures of anaent cultures, and, lastly, afforded, deeper insight into 
the anaent intercourse between East and West as the archaeological esplora- 
tion of Central Asia. 

An earlyj appreciaticxi of the importance of these finds for philological, 
historical, and archaeological studies led the Russian, French, British, and 
German Governments, as also some learned soaetiesi, to send orgamsed ex- 
peditions— they were peaceful penetrations, the army of explorers being armed 
with nothing more frightful than spades and pickaxes and a plentiful supply 
of writmg matenals and packing cases— to explore those little-known regions, 
and to recover objects of archaeological interest from the sand-buned sites of 
the ancient cities of Chinese and Russian Turkestan Undoubtedly the most 
successful of these explorations have been those financed by the Goveirmnent 
of India, eind earned out under the direction of that patient, thorough and 
indefafagable archaeologist Sir Aurel Stein Throu^ an inborn love of enter- 
piise and adventure, through the rigorous disciphne he imderwent as a stu- 
dent in a German University, through prolonged study of Indian languages, 
tradition and history, he was eminently qualified to undertake the gigantic 
operations mvolved and carry them to a successful termination By laymg 
bare the regions which had served as the mam channel for the interchange 
of the avihsations of India, China and the West, his explorations have once 
for all shattered the illusive barners which it was once thought had separated 
the east and the west 

The task of an explorer in those inhospitable regions is no bed of roses 
His task is very different from that of the scholar, comfortably lodged in his 
snug study, deaphenng manuscripts, examinmg art treasures, discovered by 
the explorer, and weaving his fabric of theories The great ardhxologist tells 
us that just when he vras completing his ^loratory task, by an ascent to the 
ice-dad summit of the main Kun-lim range, at an elevation of about 20^100 
feet, he suffejeed a severe frost bite, whidi cost him the toes of his right 
foot. 'It was as a hdplesa invalid that he had to get himself {100} earned 
somehow over the three hundred nules of rough mountain track on the 
Karakoran route with its high passes reaching to over 18,000 feet before 
medical aid could be obtained. The capital importance of Sir Aurel Stein’s 
services to science was rerogmsed, among other bodies, by the Royal Geo- 


graphical Society in 1909 with the award of the highest distinction m its 
gift, the Founder’s Gold Medal 

However, to return to the spoils of these expeditions. Stein’s excava- 
tions of 1900-1 at the ruined sites in the TaMmakan deseit round Khotan 
establi^ed b^ond all doubt the great historical importance of that ancient 
culture which, as the joint product of Indian, Chinese and Western influences 
once flounced in the oases of Chinese Turkestan. Khotan was but a step- 
ping stone in the march of Indian culture eastwards, but the bygone culture 
of Khotan, as has been irrefutably estabhdied, rested -mainly on Indian 
foundations In the fine statuary exhumed in or near Khotan, it is easy 
to recognise the influence of the same Grasco-Buddhist art whidi was deve- 
loped in Gandhara, and the modem Peshawar valley. In the pictorial rehca 
of those regions we find agam the lading features of that school of Indian 
painbng with which we have been made familiar by the frescoes at Ajanta 
in the Nixam’s Dominions These remains will have a special appeal to 
students of Indian art, smce in India itself little has survived of early Indian 

The discoveries of these evidences of Indian cultural influoice in far 
Turkestan recalls an old tradibon recorded by the Chinese pilgrim Hiuen- 
tsiang, and repeated in old Tibetan teixts, to the effect that the territory 
of Khotan was conquered and colonised about the second century B.C by 
Indian emigrants from TaksaiSiia (Greek Taxilla), that is roughly Peshawar 
District and the Frontier Province This old tradition has now been con- 
firmed throu^ the discovery in that region of manuscnirts in Kharoslithi 
sciipt, which attest the use, for ordinary practical purposesi, of a Middle 
Indian dialect In the r^rt on Stein’s second tour of exploration (1906- 
8}, he tells us that from ruins now situated at a distance of fully 1(X> miles 
from the nearest supply of drinkable water, he recovered conduave evidence 
tliat the use for admimstrative purpbses of the same Indian dialect extend- 
ed in the first century of the Christian era as far as the most remote comer 
of Central Asia 

I will not describe the mass of Qunese, Uigur, and Tibetan manuscripts, 
in part still undeciphered, and the historical and philological interest that 
attadies to them since they he ■{[lOlJ outside -the acope of the present dis- 
course. I wiU rebtnct my remarks to a few important finds, which are of 
special interest to the Indo-Iranian student 

I have already referred to the Bower manuscript. This m a n uscript is 
written in a Central Asian form of Brahmi, the script current in India dur- 
ing the centuries immediately preceding and following the oommencemeid 
of the Christian era. The manuscript contains portama of an Indian medi- 
cal treatise. The Ayurvedic system of medicine appears to have been popu- 
lar in Central Asia in the first mifiepniiun of the ■Christian erai and it would 



not be surpiising if some of the Indian formulse had travdled thence further 
eastwards. Subsequent to the find of this manuscript, portions of another 
medical treatise were discovered by Stein in the Caves of the Thousand 
Buddhas, situated in the most remote comer of Central Asia. This second 
manuscript is even more interesting than the first ; for it contains besides 
the Sandrrit text, a literal translation into a hitherto unknown Iranian dia- 
lect. It has been surmised that this language is the Tobhanan, the language 
of the Tokhan tnbe Other works written m the same dialect have been 
discovered in the same region ; but these works are all fragments of Bud- 
dhist religious and philosophical texts Owing to its secular character the 
medical fragment is helpful for the elucidation of such Tokharian words of 
secular import as are not met with in the translations of Buddhist teictsu 

The majonty of manuscnpts recovered from the rumed sites of Chinese 
Turkestan are however fragments of well-known Buddhist works written in 
vanous languages and dialects, some known and some still unknown * Among 
the manuscnpts discovered by that intrepid and ill-fated French explorer 
Dutreuil du Rhins is a Prakrit version of the Buddhist psalmody Dham- 
mapada. The Prussian Turfan expeditic® succeeded in rescumg from obli- 
vion a iSansknt version of the same test These taken alc«ig with the old 
Pah text furm^ three different versions of that beautiful collection of ea- 
nobling gathas The discovery of bilingual and tnlingual versions of Bud- 
dhist texts has had erne consequence of fat reaching importance;. It has led 
as to perceive that the Sanskrit and the Pah canons are both traceable to 
a common source, and we must therefore conclude that the onginal Buddhist 
canon was written in a third dialect, which must have been an Eastern 
Middle Indian dialect, a Praknt of the province which was the chief scene 
of Buddha's activity 

As I remarked above, the explorations in Chinese Turkestan has brought 
to hght many a language unknown till then, one of j^l02} which I haVe men- 
tioned already I should like to draw your attention to one other which 
is of special interest to Iranian scholars This dialect, by some called the 
North Aryan, appeirs to have been the language of the Sakas of Indian 
tradition and Sakae of the Greek. In this dialect we have iiortions of the 
Buddhist works Vajracchedikfi, PcajnipfLranutA, and Aparamitiyussfitra, and 
possibly others. To the two wdl-known Indo-Aryan dialects, the Indian 
and the Iranian (m other words, the Sanskrit and the Avestan), this new 
dialect is rdated in a peculiar way. Phtmetically the language of the manus- 
cnpts is dearly Iranian, but in the matter of its vocahulaxy it is strongly 
influenced by the Indian branch ; in other words, it is Indianized Iranian. 
Genetically an Iranian dialect, having for centuries stood) under the cultural 
iiduences of Sanskrit, it borrowed the rehgious and philosophical termini 
from the more advanced sisber dialect, whidi is an illummating commqn- 
taiy cm the spread of Buddhism and Buddhist culture^ 



But I suppose the most remarkable manuscnpt find, the last one ttiat 
I am going to speak about this evening, consists of the fragments of Buddhist 
dramas which were found by Dr Von lE CoQ, the Director of the Prussian 
Turfan espedition, m one of the cave temples of Ming-Oi by Kysyl, west 
of Kuja, on the fringe of the TaHamakan desert. Despite the epodx mak- 
ing importance of its contents, the publication of the manuscript has awak- 
ened httle mteirest m India 1 The pgalm leaf fragments were edited in 1911 
by Geheimrat H. Luders of the University of Berlin m a facsimile entitled 
"Fragments of Buddhist Dramas” The largest fragment, which is made 
up of 8 or 9 smaller pieces, is not more than 34-5 cm long. The 
pieces were fitted together with infimte care and patience by Geh. Luders 
with the assistance of his wife, who is an equally ardent student of Tndjan 
Literature The fragments, which number nearly 150, yield a not inconsi- 
derable portion of two Buddhist plays m Sanskrit and Prakrit The cliar- 
acter of the wntmg, which is an incontestable evidenoe of their age and 
authentiaty, is identical with that of the mscnptions of the Northern Kshat- 
rapas and Kushanas , it also clearly diows that the manuscripts were pre- 
pared m India One of the dramas was an allegorical play, introducing as 
characters the personified qualities of Buddhi, Dhrti and Kirh In another 
the author mtroduces the figures of Sinputra, Maudgalyfiyana, two of 
Buddha’s pupils, and of the Elightened One himself among the dramctts 
persons. Evidently they were Buddhist plays. It is interesting to note 
that the charactenstic figure of the Vidfi^aka, the Clown of the Hmdu 
drama, is not absent from these plays. This is not the £103) place to 
enter into the bearing of these plays on, various literary histoocal problems. 
They contam the usual alternation of Sandmt and Prakrit, and the passages 
in prose are punctuated with verses in artifiaal meters In the Prakrit 
passages we can distinguish three dialects, gauraseni, MtgadM, and Ardha- 
mfigadhl. From the linguistic point of view the most important feature of 
these plays is that the Praknt they contain is in a stage much older than that 
which is stereotyped in the dramas of the classical and the post-classical 
age. From a colophon of another fragment, which was discovered a little 
later, we learn that the author of one of the dramas was no less a personage 
than ASvaghosa, that prodigy of learning who has left his mark on every 
branch of literature and philosophy he touched. Some of the plays thus 
belong definitely to the fi.rst century of the Christian ea They supply us 
with an mcontestable proof that in the first century AJ>. the Hindu drama 
had already assumed its characteristic form, a condusicm which has an 
important bearmg on questions relative to the origin of the Hindu drama, 
or at any rate of the Sanskrit drama. These fragmaits, picked up in Turke- 
stan and now housed m a Berlin museum, are ptortions of the oldest Hindu 
drama and almost the oldest Indian maiuiscripts available uptil now. It 



is therefore difficult to overrate their importance for the study of Indian 
palaeography, linguistics, and dramaturgy 

This hurried survey has, I hope, served to give you an idea of the nature 
and scope of some of the problems lying on the fnnge of Indological reseaich 
These are not more important than what I call the centncal problems And, 
of course, no hard and fast Ime can be drawn between them ; they are not 
mutually exclusive, but merely complementary to each other. Both are 
equally important, each m its own way It is to be hoped that the rising 
generation of Indian orientalists will distribute tlieir time and energy evenly 
over the whole field The excellent work done in the past by the K K 
Caraa Institute fills one with the hope that the scholars associated with it 
wiU turn thar attention also to the solution of the new prdilems that have 
arisen with the birth of the present century 

Will the results repay the trouble? A ceitam number of people will 
answer the query with a shrug of shoulders, and some even m the emphatic 
negative To me the study of the Piist seems to be ai categorical imperative 
of civilized life ; I diall not try to justify it otherwise. I am fully persuaded 
tnai under all conditions of civilized life there will always be found people will- 
ing to “ waste ” either their own time and incidentally {104} the time of others 
by applying their energy to a study of the Past, to a study of dead languages, 
buried antiquities, and civilizations by-gone These unsdfish silent workeis 
will be encouraged in their arduous labours by the sincere homage and gene- 
rous appredation of mm like Mr. Damodhardas Sukhadvala. These men 
are not satisfied with acting as spectators m the academic inquiry whether 
the aims and objects of historical research are wise or otherwise , they attest 
their livdy interest in the work of scholars with the seal of material and 
mumficent, assistance. So long as our country produces such generous and 
unsdfidi donors, we need not fear for the futurel of Indological Research in 




Tradition ascribes the Kavyapraka'^ to Mammata and Mammata is 
for all intents and pniposes the single author of the KavyaprakSsa (KR), 
Another tradition reminds us, however, that the KP forms one of the few 
exceptions to the efficacy of the Nandi to ensure the nirvighnapansamapti of 
the undertaken work , in other words, that its author never lived to com- 
plete the work he had begun This last tradition by itself carries some 
weight, in so far as the old Hindus were so tickh^ about confessing to any 
such exceptions, that a rumour of this nature could not possibly acquire the 
currency it hsa, were it not grounded on fact These two conflicting 
traditions are reconciled in hght of the evidence of the author of Nidaifeina 
— one of the older Vyakhyias of the KP — ^who confirms the lattk' statement 
and tells us that up to the Alarakara Pankara the KP. is the work of Mam- 
malta— and that mcludes all the nine chapters, together with nearly two- 
thirds of the tenth and the last chapter , he ascnbes the rest of it to one 
AUata, about whom nothing more is known. In support of the view he 
adduces two verses composed by two different authors, which refer to the 
tradition, according to which M left his work unfimshed The first of 
these is quoted in JhaJakikara’s edihon’^ of the KfivyaprakaSa at p. 852 : 

krtah Mnummatacaryaifaryaih pankaravadhih | 
prdbandha}}, puritah seso vtdhdyaUatasurifjd ||‘ 

* [ZDMG 66 477-490] 

1 KAvyaprakala, a treatise on poetics by Mammata, edited by jHAtAKlKAS^, 
Bombay Santot and Prakrit Senes, Bombay 1901. References throughout this 
article are made with reject to the llgurmgs of tius edition. A single figure fol- 
lowing KP. dmotes the page and double figures denote the numbers of the Ullasa 
arid the Kanlfi respectively. 

2 Petebson first called attention to this stanza UBomBiRAS XVI, p, 23). 
Being misled by an evidently corrupt passage in the commentary of one of the 
manuscript allies of KP., he had acquired, he was led to imagme that the "met- 
ncal portion ”, the Kankas alone, belong to M , while the prose commentary is the 
work of RijSnaka Ananlda. Prof. BfJHLEB’s reconstruction of the corrupt passage 
in question dearly pomted out Peiebson's mistake (Ind Ant. XIII, £478} pp. 30, 
31) . Prof. Buhus remarks m the course of the same arlide : " Though I am 
unable to accept Prof. Peterson's man theory, I think that he has done a ser- 
vice to the history of Sandoit hterature by showing the existence of an old tradi- 
tion, according to which the KP. is the work of two authors I do not see any 
reason for doubting this statement." An independent proof of the common au- 
thorship of the KSnkas and Vrtti is afforded by the Kfinka maid tu piirvavat in 




{[478} This fact — quite probable in itself — ^has, to my knowledge never been 
further investigated ; and the two facts just quoted are the only ones on 
which the theory of the double authorship of KP. so far rests. A compa- 
rison of the Kavyalaipkara (KL.) with, on the one hand, the part of KP 
attributed to M and on the other, that attnbuted to Allata, as I intend to 
show, sets the matter beyond the pale of doubt Such a comparison discloses 
the different sources which the two authors have used While the author 
of the latter end of KP. depends for his whole material practictally on KL., 
and does not heatate to borrow expressions and phrases verbatim from the 
latter, M. himsdf makes use reservedly of the new ideas brought into the 
Alaijifcara^tra by Rudrata and looks for his authorities amongst writers 
older than Rudrata 

From Parisaijikhya on to the end of the portion dealmg with iSuddha 
Arthelaipkaras— which, for convenience of reference, I will name the 
“second'” part of Ullasa 10, in contradistinction to the remaimng portion 
of the same UlSsa which will accordingly be referred to as the “ first ” part 
—there follows a set of new Alamfcaras nearly all of which are borrowed 
from KL ; and in the following I will try to show that the defimtion in 
KP. agree word for word with those in KL., or at best, offer only a para- 
phrase of the latter A companso!i of the number of illusitratioiis in KP 
borrowed from Rudrata’s work shows us that there are in the “ second ” part 
as many as 14 out of a total number of 48 illustrations borrowed from the 
KL., while in the “ first" part there are only 18 out of a total of 378. 

The followmg is a synopsis of the points of agreemoit between KL 7, 72 
to the end of that adhyaya and KP. 10, 118 — 131, cwnprising the nine 
Alaipkaras : 1. Parikara ; 2. Parisatpkhya . 3 KBranamala , 4 Anyonya , 
5. Uttara ; 6. Sara ; 7 MUita ; 8 Ekavall ; 9. Vieama. As, in the KL. one 
whole aryd is devoted to the defimtion of each single Alamk&ra, while in 
the KP. the style of enunciation is much tenser, only the significant portion 
of each will be ated for purposes of comparison : 

1. Pankara (KL. 7, 72 ; KP. 10, 118) : 

KjL definition : sSbhiprdyaik vtie^anaih vastu viskyeta \ 

KP, has sakOtcd^ instead of sabhiprayd^ and the defimtion runs : 
vUesxanaiT yat $&kikmr uktth 1 

{479} Rudrata mentions four varieties of parikara according as the 
vUe^ya is a drauya, guna, knya, or jSH. In KP. it is not further divided 
This is the last verse attnbuted to Mammata 

2. Porismhhya (KL. 7. 17 ; KP. 10, 119) : 

the Alaqikira RiQpaka, whoe purvavat must refer to mglnp nma which has been 
meatkmed in the Vftti on Upema, as it can refer to nothing in the KSti^a, 
themselves, mdlc never being mentioned in them. 



KL definition : 

pTstam aprsUtm sad gunadt yat kathyate kvactt tulyam |. 
anyatm tu taddbhavah praRyate sett pari'^ || 

KP. definition : 

kim ctt prsiam op^s^rn vd kathitam yat prakalpate ] 
tddT&^yavyapohdya pansamkhyd tu sa snt-jtd 1| 

KP tddTg” corresponds to KL. tulyam anyatra tadabhavak-Ksivya- 
pradlpa explains, in fact, tadrg tulyam \ vyapohaya vyavaccheddya | KP. 
illustration 1 is built on die same pattern as KL. illustration 1 ; and KP. 
illustration 3 = KL. illusbation 2. 

3. Kdmmndld (KL. 7, 84 ; KP. 10, 120) : 

KL definition * 

yath&purvam eti karanatdm arthanam pmvarthat | 

KP. definiticm : 

yathottaratfi cet purvasya purvasySrtkasya hetuta | 

KP. illustration jttendnyatvam etc. embodies the same idea as KL 
Ulustration vinayena bhavati etc. Possibly both are made in imitation of a 
common model ; more likely however as KP illustration is quoted by Mam- 
maita again in Ullgsa 7 to illustrate a dasa, the latter is an older verse and 
R has transformed it into an Sryd. 

4. Anyoym (KL. 7, 91 ; KP. 10, 120-121): 

KL. defimtion • 

yatra parasparam ekah karakabhavo 
’bhidheyayoh kiiyayd sat/tjayet | 

KP. definition : 

kriyaya tu parasparatri vastunor jancme | 

Here the resmblance is obvious 

5. Uttara (KL. 7, 93 ; KP. 10, 121—22) : 

KL. definition : 

uUaravacaru^ravanat unnayanarfi yatra 
piirvavacon&tam .. praSndd apt | 

KP. definition : 

uttareSruttmdtratak praimsya unnayanam yatra kriyate ] 
ttttra vd sati || 

(480^ Here again the similiarity is striking. The structure of KP. 
illustration 2 kd vi^amd , is the same as that of KL. illustration 2 kiifi 



6. Sara (KL. 7, 96 . KP. 10, 123) • 

KL. definition ; 

yatra yathasamudayat, yathatkadekaji kramena gunavad iti \ 
nirdharyate paravadht mratisayaifi tad bhavet saram || 

This complicated definition of this simple alamkara of R. is compressed 
into half an arya with the retention of all the significant elements of R ’s 
deflmtion : 

utfarottaram utkarso bhavet sdTafi paravddhih | 

KL. ynthasttmudayat yathaihadesarn gunavat implies the same idea as 
utkar$ah and kramena = uttarottaram. Pardvadki is the same in both. 
Further, KL. illustration = KP illustration 

7. mua (KL. 7, 106, KP. 10, 130) ; 

KL. definition : 

sammacthnena har^akopadi \ 

apare^a tisrasknyate nityenagantukenapi \\ 

KP definitim : 

scfmena laksmand vastu vastuna yan mguhyate | 
mjendgantuna v^pi . || 

In this defimbon, KL samena aknena is the ecact equivalent of KP. 
samena lakfmand, tiraskriyate of niguhyate, rntyenagclntukenapi of nijena- 
gfintuna vdpi. 

Ekavdfi (KL. 7, 109 ; KP. 10, 131) : 

KL. definition : 

ekavoRti seyam yatrSrthaparttmpard yathalabham | 
adhtyati yathotUuraviie^and sthityapohabhyam || 

KP. definition : 

sthSpyate 'pohyate vapi yathdpurvarp, porarnpatam \ 
viSesamtayd yatra vastu saikdvaR smrta \ \ 

Here KL parampora, yathottaravUesana, sthityapohabhyam are exact 
equivalents of KP. param pararai, yatha purvarn viSesoffatayd and sthapyate 
'pohyate v3pi reflectively. 

KP. illustration 1 is taken from Nava^hasikacaritam and illustration 2 
(to which KL. illustraticm 2 is not at all unlike) is from the Bhattikavya 

Here we will also consider 

9. Vi$ama (KL. 7, 47—55 and 9, 45—47, KP. 10, 126—127). 

In its natural sequence it ccanes in both the works after S^a and be- 
fore !MSlita I did not however consider it there, as it differs from 

the other eight b^iniung with Parjsaihkhya, in so far as it is an ^t^karq 



with several varieties, — described by R. once under Vastavya and again 
under Atriaya alarjikaras — all of which have not been adopted in KP. The 
vaneties, however, which are common to the two show as striking points 
of siimlanty as the other eight Only the varieties which are common to 
both are here quoted. 

KP. vanety 1 • kv^Cfctd yad ahvaidharmydn na ileso ghafandmiyat | 
and vrtti to it dv^yor \iatymt(Xv%laks<matayd yad anupapadya mdnaXdr 
ymva yogafk | 

KL. 7, 49 asambkdvyobhdvo vd abhidMyate | which is to be taken in 
conjunction with IQL. 7, 47 vaktd vighafayati kosm apt sambandham I 

The illustrations in both are formed with kva-kva 

KP. vanety 2 • kartuh kriydphaldvdptti natva ncrthai ca yad bhavel j 

KL 7, 54 : yatra kriydvipatter na bhave'd eva kriydphalam tdvad \ 
kartur marthaS ca bhavet - 1 1 

KP vanety 3, 4 gunakriyabhydyn kdryasya kdranasya g^f}akt^ye | 
kramena ca viruddhe yat sa esa vi^mo matak H 

KL 9, 45 : kdryasya ca kdranasya ca yatra virodhab parasparaifi 
gunayoh j tadvat kriyayor atKaod . . . [| 

Further as in KL illustration 1 (9, 46) so in KP. illustration 3 ( — 
Navasilhasikajcaritam) the properties of objects “sword” and “fame'” bear- 
mg the relation of cause and effect, are ccmtradictory to each other KP. 
illustration 4‘= KL. illustration 2 (9, 47). 

These nine Alarpkaras with the exception of Vi§ama follow each other 
in the same order both in the KP. and KL as may be easily verified by 
comparing the numbers indicating the order in which they appear in the 
two works quoted above ; further, there are no other Alatpkaras in the first 
part of the tenth Ullasa, which agree in wording so minutely with the cor- 
responding Alaffpkaras in KL. A companson of the analysis of these nine 
with those immediately preceding them should leave us in no doubt as to 
the difference of authorship of them respectively. 

The above Alaryikaras from 1 — 8 do not follow each other in KL. un- 
interruptedly in the same order. Rudrata mentions six more Alatpkaras bet- 
ween Parikara and Ekavali viz , Panvrtti, Vyatirdca, Avasara ( - KP. 
Udatta) and Hetu, Suk^ma and LeSa (KL. 7, 77, 86, 82, 98, 100, 103)‘, 
which remain to be noticed. Of them the first three have been dealt with 
by Mammata himsdf in the “first” part of the tenth Ullasa (KP. 10, 113, 
105, 115) and do not come properly under our consideration here. In 
passing, however, it may be mentioned, that a companson of the treatment 
of ParivTtti and Vyatirdca in KL. and KP. offers a significant contrast to 



the Alamkaras just examined. In the definition of Paiivrtti althou^ 
Mammata does not bnng us anything new which is not ^482} there already 
in Rudrata’s definition, still the two definitions are utterly unlike each other 
m wordmg. In Vyatireka, moreover, while quoting Rudrata’s own illustra- 
tion (7, 90) of this Alamkara Mammata points out that it has been wrong- 
ly classified by the former ; and in fact, in opposition to Rudrata, he main- 
tains that there can never be in good poetry a superionty {Mhikya) of the 
standard of companson (Upam^a) over the object compared (Upameya). 
Further, he mentions sixteen varieties of Vyatireka against Rudrata’s four 

Hetu, Subsma and LeSa form a characteristic group in the Alamkara- 
®stra Bhamaha uncomproimsmgly rejects them^ , Dandin, on the other 
hand, most emphatically claims great excellence for them* Vamana and 
Udbhata do not mention any of the three Rudrata agam has all three, but 
his Suksma is different frcan that of his predecessors In KP, LeSa is not 
mentioned at all, Hetu is exphatly demed, Suksma alone is recognised. As 
regards Sukjma and Hetu the other author of the KP shows the mflnpncf. 
of M. R.’s Hetu has been identified by the former in the vitti to TCarainamaia 
(10, 120) with KSvyalinga ; but in doing so, he quotes R's illustration’’ to 
Hetu and observes, so to say as an apolo^ to R., that the verse (although 
it is no illustration of Hetu) deserve to rank as good poetry in so far as it 
contains a Komalanupiasa. In his treatment of SOksma both his dofinitmn 
and the vrtti show that our author borrows his material from Dapidin’s de- 
fimtion KD. 2, 260. KP illustration 2 is in mutation of KD. 2, 261 This 
treatment of Hetu, Suksma and Le^ must be looked upon as a characteris- 
tic of the school to which M belonged and be not allowed in any way to 
affect our ccaiclusicms with regard to the remaming Alamkaras. Here ends 
the list of the Vastava Arthalamfcaras of Rudrata from Parikara to the end 
of Adhyaya seven. 

To summarise the results of the foregomg analysis, taking our stand- 
pomt at R.’8 Paxikara all the remainmg fourteen alamkaras have been ac- 
counted for. Of these, eight follow each other m the same general order 
m toth the wo^ ; the defimtions of seven of them have been copied in KP. 
without any significant alteration; three of them have not further been 
notic^ m KP. as they are already dealt with in the « first” part of Ullasa 

ly <hfferently The different numberings of these in the two works depend 
y on these very facts and on the addition of two other Alamkaras 

VwamaandSama. Of OMse Visama has already been noticed ; 1*33 Sama 

™ ” Pr«Sparad,.y.(.bhll,aw, el 

* KBvyiadaida (=KD.) g, 235 

* Cf. Section III of this paper (Heft IV). 


appears for the first time in KP. and is there defined as the converse® of 

Neitt come under our consideration the fifteen remaining Alaipkaras in 
K P — ^ten of which are met with for the first time with Rudrala, two more 
(Vyajokti, Satnadhi) for the first time in KP — at least under these names. 
They are the following : Vyajokti, Asamgati, Samadhi, Adhika, Pratyanika, 
Samarana, Bhifintiman, Pratipa, Samanya, Vi^esa, Tadgujia, Atadguna, 
Vyfighata, Sahspsti, Saipkara. These AlamMras differ in the two works un- 
der consideration from the others earlier examined in so far as they do not 
fcfllow each other in the same sequence in the two works ; in KL. they are 
spread over AdhyUyas 8 and 9 according as they are upameya or 
Ati&iya Alamkgras • on the other hand, in the KP they are jum- 
bled together anyhow It may, however, be noticed, that (1) nearly in 
every doubtful case our author mentions m the vrUt whether the AlaihkSra 
m question is based on an upamd or an dtiiaya ; (2) that our author does 
not borrow wholesale from R. (as he did the VSstava Alamfcaras) but that 
he picks and chooses his material and often freely paraphrases R.’s expres- 
sions. It is, however, noteworthy that out of a total number of 18 new 
AJartikaras introduced by the author of KSvyalamkara in Adhyaya 8 and 9 
eleven find acceptance in this part of Ullasa 10 of KP. in more or less un- 
altered condition 

Of the fifteen AlataikSras above enumerated, four : Vyajokti, Samadhi, 
Atadgtuja^ and Samsrsti are not known to R , Pratyanika, Prafipa, Vya- 
ghata are treated differently in KP. and KL and Samkara is considerably 
elaborated in KP. Out of the remaining, seven agree with each other in the 
two works very closely — ^sometimes even m wording. We will consider first 
these last seven following the order in which they occur in KP. 

10. Asamgati (KL. 9, 48-49; KP. 10, 124): 

KL. definition : 

vispaspe samakalarit kdranam anyatra kSfyam anyatta \ 

KP. definition : 

bhimadeiataya 'tyantarfi haryakarariabhutayol], j 
yugapad dharmayob khyatib 1| 

The samakdlam corresponds to yugapat, anycdra anyatra to bhinna- 
deiatc^, kSranam . . karyam to kdryflh&ranayob- These are all the im- 
portant elements of the definitions The vtltikdra observes that the Alaip- 
kara is based on an attSaya. 

® In the younger Alaipkara^astra, some new Alaqikaras were obtained by 
simply inverting Hie old ones , thus Sama is obviously the converse of Vi?ama, 
Atadguna of Tadgiaja , more remotely Wnokti of Sahokti. 
r Atadguna is the cxmveise of Tadgupa. See note h 



£484) 11. Adkika (KL. 9, 28 ; KP 10, 128) : 

KL. variety 2: 

yotTo ’dhare sumchoty adheyam (tvasthxtiiffi tardyo pi j 
atvncyate kathai)i cit tad adhikam . . 1 1 

KP. definition : 

mahator yan mahtydmsav airitdSrayayoh kramat | 
aSray^ayiv^ syatam tanutve'py adhtkam iu yat jj 

note the vrttt, dsntam adheyam \ a^ayak tadddhdrah. KP illustration 

1 = KD. 2, 219 to Atisaya. 

12. Smarana (KL. 8, 109. 110 ; KP. 10, 132) : 

KL. definition : 

mstu viiesam drstva pratipatta smarati yatra tatsadfsam 1 
kdldfitardnubkutam vastv anantaram tty adah smoratiam || 

KP definition ; 

yctha 'nubhavom arthasya drste tat sadj§e smrtih | mtaranam. 

Here, drstva, drste, smaratt, smrti, tatsadfSam, tatsadrie, kdlantard- 
nubhuUm, yathSnubhavmt form the parallel series in the two. 

13. Bhrdntiman (KL. 8, 87 88 ; KP. 10, 132) . 

KL. definition : 

arthaviSesam paSyann avagacchad anyam eva tat sadriam | 

KP. definition : 

myasamvit tat tidyadariane | 

Arthavisesarri paSyrn and tatsadfiam avagacchet correspond to tulya- 
darsane, onyasc^vtt. The terms, prakaranika and aprakareayika in the vrttt 
show that the Alaipli^ia is based on an upamS and in fact' the vrttihara 
eaqiresdy states that it is not an atUayd : na ca esa mpakarri prathamatt^a- 
yoktir v3 

15. 14. Samdnya and Tadguna • 

To understand properly the rdation of these we must examine the 
genesis of these AlanMras These two figures run mto each other very 
closdy and they appear in the two works considerably mixed up They 
were forably separated by Rudrata and although this separation is not 
accepted without reserve by our author, he betrays Rudrata’s influence quite 
distinctly. The older AlajpkSra writers knew an Atigaya which was, the 
desire to depict some quahty of the matter in hand iprastuta vastu) which 
surpasses the commonly acknowledged limits, cf Dandm, KD. 2, 214 : 

vivakfa yd vi§e§asya lokastmdtivartinak | 
asav at^ayokti)}, syat alaifikarottama yathd |l 



(485"} to whidi the classical illustration was the description of the whiteness 
of the moon, iwhich makes mvisible the white-clad abkisanhas, with white 
garlands, anointed over with candana (KD 2, 215) The same we meet 
with again in Vamana’s Alamkarasutravrtti 4, 3 10 : 

sambhavyadhormatadutkarsakalpana ’tiSayoktih | 

The illustration (which is very likely a quotation) plays on the same 
idea of the moonlight and abhtsdnkds Dandin knows an attiayopatnd, 
which he illustrates but does not define It is based on the idea that (as 
an AtiSayokti) the upamana and the upameya would be utterly undistm- 
guishabkl from each other, but for some accident or for some one tiifling 
property, which is always present ini the upamana or the upameya, cf KD. 
2, 22, where the moon is said to he different from the face only because the 
moon IS to be seen in the sky and her face on hetsdf Rudrata, who has 
an atiiayd and an upama but no atisayopama, sees in Dai^idin’s lUustitation 
to Atiiaya a state of things in which there is a description based on Ati^ya 
of two objects, which when placed ade by side are no longer distmgmshable, 
Ihd same pjroperty being present in each {tadguna) , while in Dajndin’s 
Atisayopama he sees only an eirtreme similarity {.sdmya) and no AtiSaya. 
In KP we find that the KD Tadgiflna vanety 1 ( = Dandm’s Atidaya) cor- 
responds to KP. Samanya illustration 1 and KL Samya variety 2 ( = 
Dandm’a AtiSayopana) corresponds to KP. Samanya illustration 2 ; wliilc 
KL. Tadgma vanety 2. which is a new AlamkSra takes in place in KP, as 

KL Tadguna variety 1 (KL 9, 22) : 

yasminn ekagundnam cathdndm yogalaksyarupoiidm \ 
satftsarge ndndtvam na laksyat^ tadguna sa iti [| 

KP defines it as an aupamya dtarinkara (KP. 10, 134). 

KP defimtion : 

prostutasya yad anyena gmasamyaviviksaya 1 
aikatmyam badhycde yogdt tat samanyam iU S7K{t(m 11 
R’s illustration is an imitation of the old model and KP. illustraticMi = 
Vamana’s illustration to the Sutra above quoted. Further cf. vrtP , prastuto- 
tadanyayor myundUriktatya nibaddhm dhavalatvam ekSmaheiuli. ota eva 
prtha^httvena na tayor upalaksanam, which reminds us of R's definition ; 
ekagunandm 'arthanoiri nSndtvam na laksyaie 

KL Sdmya vanety 2 (KL. 8, 107) : 

sarvdkdram yasmm ubhayor dbhtdhdtum anyatU samyam | 
upMbyatkarsakaram Mrvita viiefom anyat yat \\ 

IS not further defined in KP. ; but cf illustration 2 to 

yrtft, prathomapratipamam Medam m vyadasttum ubldS^-schata 



(„ cannot do away with the antecedent apprehension of identity”). It is 
an upama-alatjtkdra in both. 

The other Tadguha defined by R. is faithfully copied, almost word for 
word, by M’s successor. 

KL. Tadgtit}a variety 2 (9, 24) • 

asamanagunaiji yasmuin attbahalagunena vastund vastu | 
samsystam tadgunatdtfi dhatte 'nyas tadgumff. sa tti |[ 

KP. 10, 137 : 

svatn utsrjya gu)tam yogad atyujjvedagunasya yat | 
vasiu tadgmatam eti bhoifyate sa tu tadgunah |l 

Here we see that the PCL., atibahalagitnena corresponda to KJP ujjvala^ 
gunasya, tadgu^atam e'ti to tadgunatatri dhatte and damsritam to yogat. 

16. Vi^esa (KL. 9, 6-10 ; KP 10, 135 and 136) : 

The three varieties of R are identical with those in KP. 

KL. variety 1 definition • 

kirn cid avaiyadheyam yosnam abhtdheyate nnadharam ] 
tddrg upfdabhyammaiyt vijneyo ’sou vtsesa tti || 

KP. variety 1 definition . 

vmd prasiddhatn cdharam adheyasya vyavasthitth | 

KL illustrabCMi = KP. illustration 1. 

KL. variety 2 defiiution • 

yatratkam anekasmnn adhdre vasiu vidyantanataya | 
yugapad abhidhlyate . || 

KP. Vanety 2 . 

ekdtmd yugapad vrttir ekasymekagocara 1 
KL. illustration embodies the same idea as KP. Praknt illustration- 
KL. variety 3 : 

yatrdnyat kurva^o yugapat karyantaram ca kurvlta j 
kartum aiakyatfi karta vijheyc ’sou viSeso ’nyak 11 
EP. variety 3 ; 

anyat prakurvatah karyam aiakyasyonyavastunah ) 
tathatva karanam cett ... H 

Here the similarity does not need to be pointed out In the vrtit the 
author points out that this AlaipiSra is based on an Ati&iya. 

There remain to be considered the three new MaraiSad^Pratyanika, 
ProHpa and Vydghata which occur both in KL and KP. and which still aie 
difiT'erently treated by the two authors. Thd Fratyainka and Pratipa of KP, 


Miscellaneous notes on kavyaprakaSa 

have indeed some similarities [487} with those of R. ; but their treatment 
IS widely divergent from that of the 16 Alamharas above considered In 
Pifftyanika (KL. 8, 92. 93 , KP. 10, 129) an angry opponent (m KL., the 
upamdna wishmg to conquer the upameya ; in KP. not the upantSm at 
all) persecutes an innocent third party (m KL any third party ; in KP the 
ally of the invincible offending party) In Pratipa both in KL. and KP 
there is disparagement of the uppmam , but the result is arrived at, accord- 
ing to the two authors, m two differait ways In KL (8. 76—78) the 
upameya is censured or pitied, as the case may be, on account of its com- 
parability with the upamana which comparability is made possible c«ily by 
the presence of some temporary flaw obscuring the excessive beanty of the 
upameya On the other hand m KP. (10, 133) it is PraGpa, when the 
upftmana is condenoned as bemg useless, smee the upameya is quite capable 
of serving its purpose or else when the upamdna is turned into an upameya. 
R.’s illustrabon garvam asaupvahya etc (8 78) is indeed quoted in KP. as 
an exampile of the same figure , but the author explains it m a dightly dif- 
ferent way if, as I take it, duravastka is a necessary condition in R.'& defi- 
nition KP has not „ duravastha ” and he sees in the verse only the turn- 
ing of the lotus {upamma) into an upameya which, according to him, con- 
stitutes its condemnation : upemeytkaranam eva vtpalanam anadaroh [ The 
figure Vy3ghata, which we meet for the first time in KL. and which is tlie 
last figure but one mentioned by R (excludmg, of course, the separate chap- 
ter on Slesap whidi does not come here m consideration) is also the last one 
of the Buddhalamkaraa m KP Beyond thd names, however, the two Alaip- 
karas have nothing in common. In KL (9, 52. 53) it is Vya^ta when 
a cause does not produce its [patural,] effect, even when not hindered by 
other causes — ^which would otherwise explam the absence of the effect follow- 
ing that cause The underlying idea is an Atifeiya On the other hand in 
ICP. (10, 138) there are two agents ; and by the very means by which one 
of them accoraplidies an act, the other one undoes it. The underlying idea 
here is Virodha The d^imtion reads • 

yadyathd sddhitam kenapy aparena tadonyatha \ 

tathmva yad tndkiyeta sa vyaghdta ttt smftah {[ 

In the Virtti we find sSdhitavastuvyahatihetutvat vyaghatah, "it is V. 
because it is the cause of the frustration of an end already achieved ” , and 
in my opinion, Bhaftoji quite rightly Explains * karyavaijatye kSranavai- 
jdiyam prayojakam I do not find any of these things in R.’s definition of 
V ; nor have I been able to identify the V. in KP. with any of R.'s Alatp- 

We will now! turn to the “ first ” part of UUasa 10 of KP, The most 
cursory conqiarison of the KarMs 87 to 118 of KP, together with the VrtU 
to them with AdhyS.yas 7, 8. 9 of KL in [488} which R deals with the cor- 



responding Alambaras convinces us that though it would be quite incorrect 
to assume that Mammata ignores Rudrata’s iwork altogether, still we are 
justified in saying that he did not take) the latter for his model. He lias 
indeed borrowed 1^ ’s illustration®, and even adopted some of the Alaipkaras, 
which we meet for the first time with thd author of KL. , but on' the whole 
M. shows an individuality of treatment and even in the cases of the Alarp- 
karas, which are directly borrowed from R, we find them presented m KP. 
in a distinctly different garb 

Rudrata was, so far as we at present can say, the first wnter on Poetics 
who categorically classified all Alamkaras so as to make them finally rest on 
a simple description of Vastu (Adhyaya 7), or on an Upaina (Adhyaya 81, 
or an Atifeiya (Adhyfiya 9) or a fSlesa (Adhyaya' 10) Thus there arises a 
series of parallel® Alamfcaras sometimes bearmg different names which are 
to be regarded as vastava or aupamya accordmg as we look at them as 
implying a coordinate description of two different things which may have 
some common properties — and in that case it is a vastava — or we consider 
it as a descnption of only one of the objects (i e the prastuta) to which the 
other with similar properbes (i.e the aprastuta) is compared This craving 
after an almost mathematically precise analysis charactenses the whole woik 
KL. This IS not the only instance m which R. forsakes the trodden path 
In the KL he introduces a row of new Alamkaras and adds new vaneties to 
the old ones , on the other hand, follows the older school and his work 
betrays the influence of Udbhata, who himsdf was a follower of Bhhmaha. 
He treats KL. in no kindly spirit When he quotes R , it is to show that he 
is wrong, 1“ with the single exception of the verse KL 4, 32 which he quotes 
with approbation naming at the same tune the author. Compare here the 
Alaijifcara Samuocaya, which, as a Vastava Alamklara,' we meet for the firet 
tune iwith R. R. defines three varieties ; M accepts only two of them In 
the Vrtti he specially mentions that those who try to make out that there is 
a third variety are wrong^'i — ^here he must have R.’s classification m minH 
for the reason above mentioned— m so far as that variety is mcluded in his 
fimt That both the authors understand the first variety in the same sense 
follows from M.’s illustrahons and Vjtti to them M ’s definition is different 
from that of R and it must be adnutted that the former is bettra: thaw the 
latter R, defines Samuccaya, KL. 7-19 : 

ycArdkatrmekai^ vastu paratii syat sukhavahady evu j 

« I think there is no doubt about the fact that R. illustrated hisl rules exdu- 
sivdy by examptes composed by himself. Sea further on. 

» cf, Sahokti (KL. 7, 13—18 and 8. 99—102) ? Samuccaya (KL. 7, J9^2Z; 
and S, 103. 104) ; Samyla (KL. 8, 105) and Tadginja (KL 9, 22—23) etc. 
cf KP. 834 838. Samuccaya; KP 784 Vyatoeka 
cf. Section II cf this article. 



£4893 M’s definition is (KP 834)- 

tatsiddhihetav dkasmm yabranyat tatkaram bhavet | 

This latter definition applies to R ’s illustrations equally well. In each 
of the three illustrations in 7, 20, kini atra vo hdsyapade mahad bhayam, 
in 7, 21, sukhom idam etavad, in 7, 22, ashatvam adhdsyan, from the pras- 
tutakarya of M. (see Vrtti) corresponding to katham nu mrakah sodhavyafy 
Here we see that the definition, although brmging no extraneous dement, 

IS worded differently from R.’s definition In variety 2(= KL vanety 3,' 
M’s definition leaves out R.’s vyadhikarane and efeasroin defe— which Eire 
two of the most important elements of R’s defimtion and which in fact 
exactly define the points in which this variety differs from variety 1 — ^wluch 
virtually alters the Alamkara ; the Vrtti justifies the omission giving examples 
of Samuccaya which are not vyadhikarane or ekasmtn dese. This typifiies 
the cases of R’s Alamkaras which are borrowed directly by M. Now we 
will consider the six Alainkaras with which R commences the seventh Adli- 
yaya : Sahokti, Jati, Yath&samkhya, Bhava, Paryaya, Anumana (KL. 7, 
13-18, 30—33, 34-37, 38—41, 42—46, 56—63). They appear in ICP. 
in the following order : Yatbasamkhya KP. 803 (then follow two other 
Alamkaras); Svabhavokti = R’s Jati KP 814 (then one more) Sahokti 
KP. 817 (then follow seven others) Psiryiaya KP. 842 and Anumana KP. 
847. The wordings of these Alamkfiras in KP. and KL. with the exception 
of Yathasaipkhya offer the widest contrasts ; we may again notice here that 
even when M does not add anythmg new to them, he does not simply para- 
phrase R.’s definition. I refer the reader further to the brilhant monogram 
” Beitrage zur dlteren Gesduchte des Alamkara^stra “ (Dissertation, Berlin 
1911) of my friend Dr. Johannes Nobel, in which he has exhaustively analys- 
ed the aght Alamkaras : Dipaka and TulyayogitS Vihhavana and Vigc^okti, 
AprastutaprsiSamsa and Samfisoikti, NidaiSana, and Arthantaranyasa follow- 
ing them succesavdy as th^ appear with Bhamaha, Dam^n, Vamanai, Ud- 
bhata, Rudrata, Mammata and Ruyyaka and particularly to p. 75 where, 
with reference to M’s treatment of Arthantaxanyasa, he says : „Ganz von 
Rudrata ahhangig ist Mammata, was um so beachtenswertdr ist, als er sonst 
wenig auf das KSvyialamkiara Riicksicht nimmt “ and in note 14 Sonst 
folgte Mammata meist Udbhata, wie wir bei den vorangdienden Unter- 
suchungen sahen". 

From a consideration of these facts I consider I am justified m drawing 
the conclusion that although Mammata lies under obligation to Rudrata 
for a great many of his ideas, he has shown a distinctive individuality in 
the treatment of the ideas he has -borrowed and that his work can in no seise 
of the words be called a slavish imitation of Rudrata’s Kavyalamhfira. 

In ^/^nrfiisinn, I may mention a fact which by itsdf would £490} have 
been thoroughly mcondjiavef namdy, that in the “second” part of tb? 



tenth UUasa there have been borrowed six illustrations (out of a total number 
of 84) from the little known Kavya Navasahasikacaritaml^® while of the pre- 
ceding 518 illustrations there is not a single one which is traceable to that 

In view of these facts taken all together, I think we are justified in 
assuming for true the tradition regardmg the two authors of the Kavya- 
pra(bla|§a and I am mchned to think that the statement of the author of 
NidarSana agrees correctly to the very verse, as Parikara is just the hinge 
where the two parts are most liEdy to hei joined together. 


In the following it is intended to point out that a portion of the Vrtti to 
the definition of the AlamkHia Samuccaya, in the KavyaprakSSa, does not 
originate fi:om either Mamnoata or Allata and that it must be regarded as 
a later interpolaticxi Mammata defines two varieties of Samuccaya. The 
definition of the first vanety KP. 10 116 runs as follows : 


iatsiddhthetSv ekasmin yatranyat tcAhattam bhavet samucpayo ’sou | 

“ When there is already one cause for its production (viz of an effect) there 
are also others doing the same (i. e. producmg the same effect) it is S ” 

Vrtti : 

tasya prastuiosya koryasya ekasmtn sadhake sthtte sddhakSntarmji yatra 
sarnbhavonti sa somucpayat). j 

“ When, there being already present one cause of an effect in question other 
causes are present, it is S.”. 

Illustration 1. 

duTvarSh smoramaigandh priyatctmo dike mono ’tyutsukam 
gddhatii pretna navam voyo ’ttkathmaff. pranah kulam mrmcdam \ 
sMtvatfi dhidryavirodht manmathasuhrt kdlah krtSnto ’ksamo 
no sakkyaS caturak hatham nu vtrahcdi sodhaoya titham ||i 

" Irresistible are Madana’s arrows ; the beloved is at a distanrp. , the heart 
is full of longmg, love deep, age young, life painful, family stainless ; woman- 
hood is the reverse of firmness ; the Season is the friend of Madana , Death 
is inacorable ; the friends are not shrewd ! How is this perfidious separation 

to be endured.'’ 

For this data I am dependent on the alphabetical index of the illustrations 
m the KP. at the end of JhalaXIkara’s edition of the work, as the MahSloLvya 
is as yet known only in MS. 

» tZDMG 66. 533-431, 

1 S5mgadharapaddhati 3753, 


C53^ VrtU : 

'Otra virahasahatvarni smaramargcaid eva kurvantt tadupari priyatoma- 
durasthitySdt updttam ) 

“ Here, Madam’s arrows by themselves make the separation unbearable ; 
over and above this (such other causes, as) the fact of the lover being away, 

etc are mentioned ” 

Vrlti : 

e^a eva samuccayah sadyoge 'sadyoge sadasadyoge ca pajyavasatUi na 
prthak laksyate ( tathd hi 1 1 

“This same S includes (that variety), where there is a scidyoga, asadyoga, 
and sadasadyoga and hence the latter is not separately defined by us ; for 

Ttlustrcdion 2. 

ktUant amalinatti bhadta mUrtir matih srutiialini 
bhujaMam (dam sphitS laksrmh prabhutvam okhanditam | 
prakrtisubhaga hy ete bhdpd ambhir ayatti jano 
vrayatt sutcftdrp darpam rajams ta eva tavahkt^dh {| 

“Family stainless, appearance noble, mind enriched with (the knowledge of) 
the iruti, strength of arms adequate, wealth abundant, lordship undivided ; 
these conditions are naturally diarming , though this one owing to them 
becomes conceited, these same, o king, are your goads (which keep you on 

the path of virtue) ” 

Vftti . 

atra tu satdfn yogah | uktoddhaica^e iv asatarfi yoga^ ||i 

“ In this there is a combination of good things {satdtn yogah) , but in the 
example (first) mentioned there is a combination of bad things {asatdm 
yogah) ” 

Illustration 3. 

5 di» divasadhiisaro giditayauvmd kSmini 
saro vigatavdrijeafi mukhatn anak^aratjt svakrtek ] 
prabhur dhemetpa/rdyco^ah satatadurgatak 'sajjamk 
ttrpSngaifagabah khalo mamsi sapta &alydm me |p} 

“ The moon pale during day, a woman who has lost her youth, a pond devoid 
of lotuses, the illiterate mouth of a handsome person, a patron who is entirely 
devoted to tnoney, a good man always in difficulties, an evil man at a king’s 
court ; these are the seven darts in my mind." 

* Bhartrhan's ]N!ti4 45 = Ind. Spr, 6434. This is the only occasion on which 
a verse from the Mti4. is quoted in the KP. 



Vjtti : 

atra S(dmi dhusare Salyantara^ti iobhandsobhanayogak | 

{5351 “ Here the pale moon being already one dart, there are other darts 
as well thus there is a combination of good-bad things {Sobhanasobhanayo- 
gah) ” 

Samuccaya means a “ multitude ”, " collection ”, “ heap ” ; the essence 
of the figure Samuccaya is a heap of causes all (m equal degree) leading 
to the same effect, which latter forms the theme {pmslutaknya)’ Thus far 
it is dear aiough With the words na prthak laksyate, the Vrttikana evi- 
dently wishes to justify the position of the PCarikakara in not admitting a 
further subdivision of a sat-, asat-, and sadasat-StOimuccaya on the ground of 
the latter variety being already included in the defined S ; the following 
three verses apparently illustrate what “ others understand by those terms 
It is essential for our investigation to determine precisely what these three 
terms mean or can be taken to mean, and as the Virttikara does not explain 
them any further, we will next see how the commentators interpret them. 

Govinda the best commentator of the KP. expresses himself thus ; 

kulamth I atra kuladindrri sanucinmam em yogah J durvard ttyady 
uktoddharat^e smceramSfganddinam asamtctTidndm | iasiti j atra sadasator 
yogah I durjanasydsattvat Sasyddindm sattvSt j etac cintyam | purvarn dwd>- 
sfhitySditnSesanenadhusaratvdidind ’trSpy asamyakivam tti. | 

“ (In the verse) kuiam etc | Here, there is a multitude of good things 
only such as kula etc (viz., kula, miirti, matt, which are goodi, honourable, 
desirable). | In the illustration (commencing with) durvdrah, which has been 
mentioned, (there is a multitude) of bad things only, such as smaramargana, 
etc. (viz., smafamSrganah, utsukam manah, navam vayaJ}, etc which are all 
causes of pam, grief, etc.). In the verse saS etc. | Here there is a combma- 
tion of good and bad things {sadasator yogohj) | on account of the wicked- 
ness of the wicked man and the goodness of the moon etc. | this deserves 
consideration. 1 For as m the previous illustration {durvardh etc.) the “bad- 
ness” of the lover etc results on account of (the attnbute), “being at a 
distance etc., so here also (die “badness” of the moon must follow) from 
the state of being dim during day ” 

The Prabhn understands the last sentence m Govinda's Commentary in 
the same way as I do There the commentary runs • 

cintyatve hetum Aha | durvara tty udaharane ity arthah | visesamtdsa- 
myaktvom ity cmvayah \ tatsrapi priyatamdsya sattvam eva durasthttwUesa- 
nena parom asattvam | ihapi svatah sundarasya Sasino dhusarcdvenety asad- 
yoga evety arthah 1 

Nggojibhatta in his Udyota, after distinguishing the Alatpkara under 
conaderation from Samadbi and Kavyalinga and explaimng the illustrations 


1 and 2 in detail thus commentates Govinda’s remark, atra sadasaior . . . 
sattvdt : 

tdam cintyam | evam hi sahacmabhinnatd sydt j sarvatra vi^esyasya 
sobhanatvam viie^anasydsobhmiatvarn ca prakrdntam £536} it% bhagnapra- 
kramatd vd sydt | tasnidn nrpdngcmtm asadyutam tti pdtho yuktah | sadasad 
iti ca karmadhdiayo yuktah | 

This deserves consideration | Thus there will be “ Dissimilarity of the 
Associated" {sahacaiabhinnatd) ^ 1 oi there will be a “breach of the uni- 
formity of expression’' {hhagnapTakramatva) inasmuch as everywhere 
(le, in all the cases except kkala) the object qualified is “good”, and the 
attnbute is “ bad “ | Hence it would be better to read nrpdngmam asadyutam 
1 it IS better to regaid sadasai as a Karmadharaya^ compound {sofitas ca te 
asantas ca, te^dm yogaJj. and interpret it as Conjunction of tilings that are 
both good and bad) ” | 

Thus the Udyota pomts out that if with the Pradipa the compound 
sadasat were taken as a Dvandva, it would occasion the “bieach of uniform- 
ity He therefore proposes to take it as a Karmadharaya, both members 
of which are adjectives and interprets it as a combination of things which 
are both good and bad : good naturally, bad on account of some particular 
quahfymgiattribute Further he points out that the illustrations 1 and 3 are 
really different, inasmuch as, in illustration 3 the objects which by them- 
selves are “ good ” are represented as bemg “ bad ”, while in illustration 1 the 
objects have no goodness at all in so far as they always are causes of pain 
to a woman m separation from her lover Cf also Prabha : 

durvdfrdh §a^ity anayoh katham bheda iti cet zttham | duTvdrd ity atra 
vtrahdsahisnutayd priyatamddindm satdm apy asattvena vwak^S | iha tu 
sobhanasya sato dhusarafvddind asobhanatvam apiti vivak?d \ 

What the Commentators then say is the following : We might under- 
stand sadyoga as that in which there is a combination of all “ good “ things 
— ^things desirable, praiseworthy , csadyoga as that in which there is a combi- 
nation of all “ evil ” things ; and sadasadyoga as a combination of some 
tilings which are "'good”, pleasuro-giving etc. together with other things 
which are “bad”, unpleasant etc. This is logically irreproachable; but 

s An example of Sahacarahhinnata is given in KP. 486 * 

^utem buddhir vyasanena murkhatd madena nan saULena nimnagd | 
nifa iaiankena dhrPdk samddhtnd nayena caUtnxkriyide fmendratd |[ 

Here excdlent things such as iruta are combined with things dissimilar viz* 
vyasana eta 

* Viiesajoobhayapada Karmadharaya P, 2, 1, 57. viie^afunft vUefyerys bahuhm, 
is quotdd by J{HCA:c.ak£kara m support. (?) 




unluckily it offends the canon of the Alamkaradastra and commits the fault 
of sakacarabhtnnata, apart from the fact that the illustrations do not wholly 
justify this interpretation. Thus illustration 2 should be a collection of all 
“good” things and we find accordin^y that kulam amahmid hhadrd mw~ 
ti}f etc. down to prabhutvam akhmdttam are all “good” things Illustra- 
tion 3 should be a multitude of good objects and ^5373 bad objects as well 
The objects mentioned are saiin, harmtii, saras, svdkrti, prabhu, sajjana and 
khala. It can be argued that the first six are “ good " and the khcda bad 
Ipso facto ; therefore we have a combination of “ good ” and “ bad ” thmgs ; 
but this solution fails altogether when we proceed to illustration 1 That 
should be m accordance with our hyixithesis a multitude of “bad” things 
only ; we might explain the durvaralp smaramdrganSh as being an unquah- 
fied misfortune and equally so the pnyatamo dure, but we cannot rationally 
say that gadhamP^^ma, navarfi vaydh, mrmalam kulam, stntvam and sakh- 
yah as being unconditionally “ bad ” We see thus that our first hypothesis 
does not by any manner of means conform to the condition of the illustra- 
tions. The compound sadasadyoga, it is suggested, can, however, be treated 
as a Karmadharaya Compound and may be taken to mean a multitude 
of things which are by themselves “ good ” but which on account of some quali- 
fying attribute are “bad” (dharmavtse^asamparkdd aiohhanah). Then we 
have a more rational explanatitm of illustration 3 ; we have, for example, 
SaHn kdtm& etc “ good ” m themselves, “ bad ” on account of th^ particular 
circumstances with which they are accompanied This explanation comimts, 
however, the fault of the bhagnaprakramotva, m so far as while enumerat- 
ing things which are " good ” by themsdves and “ bad”’ on aoooxmt of some 
casual attending circumstance, we come suddenly to the khda who is “ bad ” 
in himsdf and can be only looked upon as bemg “ good ”, being at the royal 
court— at best not a very satisfying explanation. We proceed, however, to 
illustration 1 and we find that the villainous pnyatama, preman, ktda, which 
we had hypothesised as beu^ “.bad” are so, also m virtue of some casual 
attending circumstance ; tht^ the pnnaples exemplified m illustrations 1 and 
3 are identical. One way of getting over this difftculty has been already con- 
sidered in connection with Nagojlbhatta ; another one will be considered in 
connection with Ruyyaka That the three verses are examples of Samuccaya 
and that they are already included under the definition of the same in the 
KP is dear enough ; what is not dear, and what the commentators have 
not been able to explain, is the fact, how either the terms, sadyoga etc. or 
the illustrations 1, 2, 3 are to be interpreted so as to fit eadi other. Mam- 
mata defines another variety of S., with r^ard to which we only need to 
condder the Vrtti • 

dhurwti ea^ tanute ea ISrtim ityddei, krpdjpapSids ca bhovSn rana- 
ksftau sasSdhtwddaS ca surak surdlaye ityadeS cd darsanad vyadhikaratpe itt 
ekastmn de§e iti ea rut vdcyam | 



“ It should not be said that (S. is possible onlyj when the substiata of the 
Eiraultaneous actions are diffei^nt ; nor (should it be ^id that it is possible 
only) when the region is one and the same , for (such verses as) dhunott 
etc, and krpdnapSn^h etc. are found," 

In Ruyyaka’s Alamkarasarvasva the subject is dealt with on ^538} the 
same Imes as laid down in the KP. , all the five illustrations to the two 
vaneties of S are repeated by him and with illustration 3 he has the same 
difficulty which we saw pointed out by Ckivinda. To the objection, that on 
the supposition that il the sadasadyoga were taken to mean a mulbtude of 
things which are good in themselves and bad only on account of the attend- 
ing circumstances, the illustration durvardh and ioB cannot be differen- 
tiated from each other, he replies “(In it is intended to re- 

present as bad those ^things which are good by themsdves; while, m the 
other example, only such as are wholly bad ; for this reason, in the one it is 
summarised with the words “there are the seven darts in ray mind” on 
account of their causmg pain to the mind even when they have entered the 
mmd as obj'ects of beauty , while in the other case, where the situation is 
summarised with the words, “how can it be endured”, it is mtenriiHi to 
express the idea, that the objects from all points of view are bad ” ’ 

This exposition is more brilliant than convincing , it however, quite 
dear that this mterpretation was not in the mmd of the Vrttikara ; were it so, 
he would have himself mentioned it, as, to say the least of it, it is not very 
obvious To make a roc^ guess at what the Vrttikara did have in. his mind 
I should say that he meant sadasat as a Dvandva Compound and undei- 
stood it in the sense m which Govinda does 

In passing, 1 may mention that S&hityadarpana brings nothing new to 
the subject except some illustrations ; the “ Sadasadyoga ’’ is, however, illus- 
trated by the dassical example from Bhartrhan, which we have already met 
with twice before. The same difficulties are encountered and the author’s ex- 
planations do not throw any more light on this perplexing question 

B^inning with Mammata, we thus see, there is a uniformity in the 
treatment of the Samuccaya Whether we take the Compound sadasat as 
a Dvandva or as a Karmadharaya die logical incoogmency remains ; and 
be it remembered, that this spurious variety is rejected in KP. not on luicount 
of any inherent contradiction, which it involves, but on the ground, that it 
does not need a spedal mention, it being already induded in the defined 
variety. The peraistenily uniform treatment of this subject after the pattern 
of the KP. by the younger writers on AlamkaraSstra suggests to us the fact 
that this is again caie of those cases, where though a commentator did perhaps 
percdve a contradiction m the old teaching, he would not admit the contra- 
diction but would every time interpret it away— certainly without bdng 
convincing for us. I have for thia reason intentionally considered in detail 



the views of the various cominentators, who have done their best to e3g}lain 
away this oontradiction, but who did not possess the key to the solution 
of the puzzle, and who were not honest enough to admit its existence. 

[539} We will now follow the Alamkara to its source. We find, of the 
old Alaipkara writers neither BhSmaha, Da|ndnii Vianiana nor UdWiatai know 
the Vastava Samuccaya. It makes its appearance first with Rudrata who 
has treated it exhaustively in Kavyalamkara 7, 19—29 There we find the 
sadyoga, asadyoga, and sadasadyoga, as well as the vyadhikaram referred 
to towards the end of the Vitti on S in KP There is no doubt that the 
Kiarikakara had adopted the new AlamkSra of Rudrata and that the Vrtti- 
fear a in his polemical remarks means to hit at Rudrata and Rudrata only. 
But there a surprise awaits us Rudrata understands the three terms sad- 
yoga, asadyoga. sadasadyoga^, quite differently from what the VrttihSra re- 
presents him to do ; the difference is, m fact, so great that unless the latter 
t ritP;ntintiaHy intended to misrepresent Rudrata, we must assume that he 
had. thorou^y misunderstood him ; so much so that it appears to me ques- 
tionable whether he knew of Rudrata’s illustrations of the second variety of 
his S at all. 

Rudrata’s ddimtion and illustrations of S 7, 19-29 are as Mows : 


yatraikatranekoiti vastu param syat sukhdvahddy eva \ 
jneyak samttccayo 'sou tredkanyah sadasator yogah H 

“That IS called Samuccaya where several pre-eminent objects, are (found) 
together which cause happiness etc By the joimng together of “ good” and 
“bad” (objects), (we have) another, which is threefold.”* 

Illustrations . 

durga^ tnkutam pankha payomdhth 
prdbhmr daSdyah subhata§ ca rdk^asSh \ 
naro ’bhtyokta sacivcdk piavamgomml!, 
kim atra vo hdsyapade mahad bhayam || 

“ The Tnkuta mountain is the castle, the ocean is the moat, R&vanja is the 
lord, the Raksasas are the soldiers, Man is the enemy with Monk^s for 
ministers ; where is for you any great fear in this matter for laughter ? ” 

Next fdlow three verses which do not specially concern us here ; and 
then a verse which Nanusfidhu introduces with 

» We can jom "good” and “bad” objects in three ways: (1) two good 
objects together ; (2) two bad objects together , or (3) pairs of objects of which 
one is good and the other bed. Cf. KL. S, 23, where R. uses the dual Dvandva 
vyastasamaste for two vyasta varieties and one samasta vanety. 


atha Mtoi yogah — 

samode tnadhit- kusume jananayanmandane sudhd candu \ 
kvaad apt rupavali gma jagatt sunttam vidhatm idem H 

£540^ " Honey is the fragrant bowei, nectar m the moon, the delight of the 

eyes of the world At least m some beautiful things there are virtues ^that 

is well-ordamed of the creator > ” 

athdsataf yogak — 

adingitah karirath ianiyas tapto^apamsunicayena \ 
maruto 'Ukhara grisme kvm ato ’nyad abhadram Oidu marm 

“ The i§ami« trees embraced by the (thorny) Karira creepers ; the exces- 
sively sharp winds (mixed) with clouds of hot sandy dust ! What can be 
more unwelcome than this in a desert in summer ? ” 

atha sadasator yogah — 

kamotavanesu tu^aro rupavilasadtsahtmu jard \ 
ramomsv api dtticarita?]t dhdtur lakfims ca rdce^u [j 

“ Snow amongst lotuses , old age in women possessing beauty, amorous 
charms etc ; wealth with the mean : that is ill-ordained of the creator I ” 

In the first of the last three illustrations there is a samuccaya of (i) 
madhu and kusuma (ii) sudhd and candra , in the second, of (i) kartra and 
Sami and (ii) tapto^apoffisunicaya and maruta ; in the third of (i) kamala^ 
vano and tusSra ; (ii) ramam and jard (ui) Iiafeiuii and nlca. These, with- 
out any extra distortion of the premises, resolve themselves into, 1, two 
pairs of “ good ” objects, 2 two pairs of " had ” objects, and 3. thiee paiis 
of objects, and in each pair one object is “ good ” and the other “ bad 

Rudrata’s Samuccaya 2 is nothmg like that mentioned by the Vrttikara 
of KP. It IS certainly different from Samuccaya 1 in so far as in 1 there is 
a smgle “ heap ” and in 2 there is a double “ heap There is no question 
of “things” which are “good” by themselves and “bad” on account of 
some qualifying circumstance” Nami Sadhu in his Commentary to 7. 24 
has rightly observed : samodakusumddisu madhvddindm satdm yogah : in 
this S there is a union of two good things ; and further on in 7. 25 
bhiitdk. There is an actual mixture, combination, union. In “Sadyoga” 
there is a heap of pairs of good things ; in asadyoga there is a heap of pairs 
of bad things, in sadasadyoga there is a heap of pairs of things, one of 
which is good and the other bad. This is a perfectly logical arrangement ; 
and to any one who knows of Rudrata's treatment of Yamaka and of his 
partiality for j'ust such mathmatical divisions, the explanation offers no 

« The ^ami trees are asat (unpleasant) because of their containing fire. Cf. 
Sakuntala (ed. CAPPSlxElt, p. 42, 1. 17) agmgarbkaijt Stamm iva. 




difficulty It IS indeed questionable whether this vanety desaves to be speci- 
fically distinguished from the first vanety— for it may be argued that if 
there be only a heap required, it may be a “ heap ” of single objects or of 
double obj'ects ; — ^but not for the reasons appeanng in the {541} Virtti m tlie 
K.P, because in the illustrations of the Vrttikara thae is not the slightest 
trace of a reference to the “ double ” nature of the 2nd vanety. The illustra- 
ticms 1, 2, 3 are illustrations of Rudrata’s first vanety and not of his second : 
the adi of sukhavahadi KL 7, 19 includes duhkhdvaha 

Enough has been said in the earlier part of this paper to leave any doubt 
as to the fact that Mammata himself was thoroughly acquamted with 
Rudrata’s Kiavy&laipkara. His successor AUata we have seen is wholly 
dependent for his matenal on Rudrata's work. So it is impossible for ather 
of them to have made this mistake Agam, as it scarcely can be supposed 
that any one would wish maliciously to misrepresent the views of an anony- 
mous person, malice m this matter is out of question. Does not the solution 
rather he in the supjposition that it is a case of simple misunderstanding ; 
and that we ought to look upon the part of the Vrtti beginnmg with tathahi 
to Sobhanasobhamyogah as an interpolation— an interpolation by some one 
who only from hearsay knew of the existence of the varieties “ sad-, asad-, 
and sadasadyoga ” of another school of Rhetoric and nothmg more , for the 
lest, however^ the interpolator had depended upon his own fertile imagina- 
tion as to what they ought to be. This is, in any case, imaginable in the 
case of Rudrata’s work, which has remamed unacknowledged and unhoooured 
by the youn^ sdiool of Alamkara writers. 

As the illustrations 1, 2, 3 in the Vrtti to the KP. have found thar way 
in Ruyyaka’s Alaqikarasarvasva, (Kavyamala 35, p. 161, 162) the mterpo- 
lation must be looked upon as being considerably old , and if the fact of this 
interpolation be admitted, it will have one important consequence : we must 
allow enough elbowroom m the estimation of the chronological relationship 
between Mammata and Ruyyaka, respectively between Rudrata and Ruyyaka, 
to make possible that, in the one case, such a significant mteipolation in 
Mammafa’s work and in the other, such an obvious misrepresentation of 
Rudrata’s work, should have been in Ruyyaka’s time an established fact. 


Prof Kielhorn in an article entitled “On the Jainendra Vyakarajia” 
(Ind. Ant. X. p. 75) pointed out that the names of the grammatical authon- 
ties mentioned in the Jainendra Grammar must not be looked upon as histo- 
rical data and in fact suggested that these names in all probability are wholly 
fictitious. In a later volume of the same journal {Ind. Ant. XVI. p. 25) he 



makes similar statements with regard to the Sakatayana Yyakarana, to quote 
his own words (ibid, p 28) “ The names employed by him [.Sakatayana] are 
given simply pujartham and they by no means prove that Sakalayana m the 
particular instances knew anything whatever of [542} the teachings of the 
scholars whom he mentions.” It appears that the practice of quoting names 
merely pujartham was followed even in later times and was not confined to 
the province of grammar alone. The facts to which I am referring are ad- 
mittedly not of such an assertive nature as those mentioned by Prof Kiehorn, 
still the certainty in this case of their being false gives us a good opportunity 
of observing at leisure the danger of admittmg too readily as historically true 
evidence of scattered facts, m particular, of the names of authors and autho- 
rities, such as occui loosdy in commentanes, and which are not otherwise cor- 
roborated ’ 

The facts in question centre round the verse KP. 860 ■ 
aviralakamalavtkasah sakalaltntadas ca kokilanondah \ 

'tamyo 'yam eft sampraii lokotkanthakarah kSlah || 

quoted in the Vrtti to the KaranamSM together with the following portion of 
the Vrtti : 

ity dtra kavyarupatwjt komdSnuprSsamahimrunva samamnasi^ur na 
punar hetvalamkarakalpemsatayeti purvoktakavyaMgam eva hetuh || 

“ It is only on account of the Komalanupnasa in this veise that its nature 
of a Kavya is presenbed by tradition, and not on account of the presence of 
the Alaipkara Hetu. Hetu is in no way different from the aforementioned 

From this alone if we knew nothing more about the verse, we might be 
led to conclude that it is an “old”® verse, which in the opinion of some 
rhetoricians contains the Alamkara Hetu, but which in the opinion of the 
author of the KP. contains no such Alaipkhra ; however that may be, the 
VTttikSra seems to say, the reputation of the verse as good poetry is left un- 
damaged, it being not wholly without some AlarpkSra. This fact is taken in 
connection with the remark of Sirabodhini — one of fhe older commentaries 
on the KP. 

» In the article, entitled “Rudrata und Rudrabhatta” ZDMG, 42, p. 426 
Prof. Jacobi pleads : " Nun weiss man aber, was auf die Autorit£t diesser Sdirift- 
steller [der Kompilatoren und Kommentatoren] zu geben ist ; da sie kebien literar- 
historisAen Sinn haben, so nennen sie ihre Autoren ohne kngstiliche Priifung, meiat 
so wie se es in ibrer Vorlage fanden. Daher wird nidit selten dersdbe Vers gaoz 
verschiedenen Dichtem zugeschneben We geringeres Gewidit hat dabei die Ver- 
wedislung zweier so ahnlicher Namen wie Rudrata und Rudrabhatta ! ” This is a 
case in point, and I must say I fully concur with Prof, jAOOei in treating sudi evi- 
dence as not ooncludve. 

® e,g. the half verse, goto ’slam arko bhalimdur ymti vwtSya pakpsfab |> Bh&- 
ipahalaipkara 2, 87, and I^v:^dar6a 2, 244, 



vastutas tv ctvirGlakafnalavtkasa ityadisu vaicttryam anubhavastddha^n 
evety Udbhatddimkitaifi samlcinam eveti navindh |, leads us to beheve that this 
verse or probably a similar verse was known to U ; at any rale, we may justly 
conclude, we are on safe ground in assuming that U at least (amongst others) 
looked upon the Alamkara Hetu as a legitimate independent Alamkara. £543} 
Probably dependmg upon the Sarabodhim, Govinda — ^the author of the Pradipa 
to the KP— boldly ascribes the half verse immediately preceding the verse 
above quoted, KP. 859 : 

hetumatd saha hetor abhidhdnam abhedato hetuh [ 
to Udbhata with the words * 

uktas cdyam Bhattodbhdima hetumatd saha hetor . . . 

Another Commentator Mahe§a Candra Nyiayaratna in his Calcutta edition 
(1866) of the KP going back on a good tradition — the same old tradition'^ 
— also attnbutes the verse to Udbhata® The youngest commentator, the 
author of the Bombay Sanskrit Series Edition (1901) remains true to the 
tradition and further drags in the name of Bhamaha^® Now Bhamaha ex- 
plicitly disavows the existence of the Alamkara Hetu as we saw above 
(p. 482) , and m spite of the overwhelming evidence of a succession of Com- 
mentators it IS highly improbable that Udbhata — the Cbmmentator and 
follower of Bhamaha— defined any Hetu ; it is certainly not included in the 
work of his Udbhatalamkarasamgraha, which is preserved for us. It is 
equally impossible that either Bhamaha or Udbhata could have expressed 
an opinion as to what Alamkara the verse should contain , because the half- 
verse hetumatd etc. is Rudrata's Definition (KL. 7, 82) and avirala etc. is 
also his Illustration^^ KL. 7, 83 of the Alamkara Hetu. The mention of the 
names Udbhata and Bhamaha by the Commentators is merely pujartham. 

Supplementary note. 

As the first section of this article (Heft III, p 477—490) had to be 

® At p. 328 his comment on ketvalamkdra in the Vrtti to KP. is, Bkatfod- 

1® KP. 860 (Commentary 1. 3), prdcdm bhamahadindm .. and further on 
1. 18 Bhatnahadaya iti iesah. 

As Rudrata illustrated his rules by verses of his owq composition [ — cf. 
Iittroduction, p. 1!, Rudrata^s Cingtmtjlalm ed. Pischel I must here add that 
with Prof JaIoobt \WZRM II) I firmly disbelieve m the alleged identity of Rud- 
rafa and Rudrabhatta, postulated by Prof. Pischel , nevertheless most of what 
Prof. Pischel tells us m his Introduction, about Rudrata, the author of the Srhgara- 
tilaka, is true not of Rudra, but of Rudrata, the author of K&vy^amkara. — ] it is 
utterly impxdbaWe that the verse in question is an ‘ old * verse, as one might be led 
to imagine from the remarks of AUata in the Vftti. I cannot explain why AUata, 
who must have known the source of the verse perfectly well, refers to it in such 
ambiguous terms. 

THE satavahanas 345 

printed from the second proof for coriection a few errors of print could not 
be avoided. Please make the following coirections : 

p, 479, 1 29 read, samjayeta for samjayet, 1 35, uitara° fol vUara’’, and 
'^Snavanad tor °sravmSt , p. 484, 1. 19, Bhrantimat for BhrantiriMn ; p 485, 
1, 37, ek^matS^ for ekShnef ; and p 480, 1 35, p 481, 1 24, p. 490, 1. 3, 
Navasahasanka® for Navasahasika®. 

Some minor errors of spelling in the English of the text are left here un- 


I am bound to form and express an opinion on the issues raised in the 
article “ The Home of the satavahanas ” published in. a recent number of 
the Quarterly Journal of the Mythic Society (Bangalore), vol 13, pj> 591 if , 
m which the author, Mr T. N. Subramaniam of Kumbakonam, cites and 
criticizes a certain theory regarding the home of the so-called Andhra kings 
which, I believe, I was the first to formulate In 1919 I published a short 
aiticle in the first issue (pp 21-42) of the Annals of the Bhandarkar Insti- 
tute, drawing attention to certain preivailing misconceptions about the Sata- 
vahanas. This artide is the subject of Mr Subramaniam’s ciitidsm. While 
admitting the validity of my mam contention that the Satavahanas are not 
Andhras and that thar original home was not Andhradega, he takes excQition 
to certain statements in the body of the article whidi, according to him, 
contradict the main thesis It is not my mtentioo to go into ddaila and to 
take up space which can probably be ill spared. I shall content myself with 
correcting the erroneous impression created by Mr. Subramaniam’s note, 
which m part misquotes my words and misrepresents my views 

I have nowhere asserted that “ the SStavlahanas have to be looked upon 
as bdoo^g to the tnbe of the Andhras,” as Mr Subramaniam appears to 
tliink (op at p 592) That is only one of the altemativesf considered md 
rejected by me Assuming for the sake of argument that the Puraoic view 
is correct, I wrote : “ If ” (in order to reconcile the PuiS|ijic statement with 
our conclusion)— ‘ If .the Sfi-tavahanas have to be looked upon as bdong- 
mg to the tribe of the Andhras, then” certain consequences will Mow (ABJ. 
1, 41) Further on m the course of the samel paragraph I reject the alter- 
native proposed aa untenable, concluding the paragraph with the words : 
“There is noflung improbable in the assumption that the founders of the 
Satavahana dynasty were ongmally the vassals of the Andhra sovereigns, 
of whom it may, with assurance, be afSrmfid that at or about the time of 
the nse of the SStavShanaa they were the most powerful potmtates in the 


* [JBBRAS, New Senes 1 160-61 ; QfMS 13 776-7] , 



{161] I must frankly admit, however, that the wordmg of the last 
paragraph of my article m question is rather abstruse and apt to confuse and 
mislead a casual reader I welcome therefore this opportunity to restate niy 
old views more luadly as follows. I hold : (1) that no cogent reason 
having been shown for connecting the early Satav&hana kingp with the 
Andhrade^, their activity should be regarded as restricted to the western 
and south-western pprtion of the Deccan plateau ; only later kmgs of this 
dynasty extended their Sway eastwards, so that subsequently even the Andhra- 
de§a was included in the Satavahana donumons , the S&tav&hana migration 
was fratn the west to the east ; (2) that the Satavahanas are different from, 
and should not be confused with, the Andhras mentioned in Greek and 
Chinese chronicles ; (3) that the home (or the early habitat) of the Sata- 
valianas is to be looked for on the western side of the peninsula and is 
perhaps to be located in the province then known as Satavaham-hara— -a 
province of which the situation is unknown or uncertain 

I see at present no reason to alter my views regarding the date of the 
Myakadom inscription, and I am not prepared to accept the date proposed 
for it by Mt. Subeamaniam. I will admit, however, that Mr Sukiamaniam 
hag offered a very happy explanation of the Puiia|nac anomaly He points 
out that even the oldest PliiSnas arc not older than the third century a.d. 
Thus at the period when the earliest PuiSnas were compiled, the Satavahanas 
had been established firmly, for over a century, as a paramount power 
in the Andhrade^a Moreover it is hi^y probable that about that period 
they had been relieved of their possessions in the! west. The Pumnic chro- 
niclers thus knew the Satavahanas only as rulers of the Andhrade&i, and 
probably mistook them for Andhras. This explanation is much simpler and 
more satisfactory, on the whole, than those I have offered m my artide 



The Assyrian clay tablet here presented was discovered in the storeroom 
of a house m Gitgaum, one! of the wards of the city {143] of Bombay 
Through my friend. Dr Robert Zimmerman, S. J., Professor of Indie Philo- 
logy in St Xavier’s College, Bondiay, it came into my hands I recaitly 
had the opportunity to announce the discovery before the Oriental Club 
of New York, and at Dr. J. B. Nies’s suggestion the tablet iwas placed in 
Dr. C. E. Reiser’s hands for deciphetihent His readmg follows Dr Reiser 
notes that of the two women sold by -z^r-ukin eme was his slave and the 
other his daughter ; the sihi and paguiramu officers who are always! meu- 

* IMOS 40. 142-4], 

"charudatta”— A Fragment 


tioned in these slave contracts apparently gave over the document guarantee- 
ing ownership. I may add that it is not known how the lelic reached India. 



1 -z6r-ukin apil-su sa ™d§ainaS-6tir ina hu-ud llb-bi-su 

[^A]-sar-si-i-biti ii ^Ina-blti.p9n-kalam-ma-lu-mur-assu 
■ ■ . -su a-na 16 Siglu kaspu a-na simi ha-ri-is a-na 
. . . -la ( ? )-a apil-su sa “'*Nabfi-zer-ukIn apil “E-gi-bi id-din 
5. [bu-ut] si-hi-i pa-qir-ra-nu sa *A4ar-si-i-btti 

[ii ^InJ a-blti-p&i-kalam-ma-Iu-mur-su m§rtu-su la-ta-nu-su 
.... -z6r-ukin na-si ina a4a-bi sa ^Ku-ut-ta-a assati-su 

apU-su sa ”Sil-la-a 

™‘*Nabd nSdin-sum 


10 -tu 

Capil]-susa “^Lugal-marad-da-ni 

ut sa “Ba-di-ilu 

“•“Sabatu flmu 22’«“ 

14. sattu2»«“ ®^NabQ-kudurri-usur Sar Babili«. 


. . -zgr-ukin, son of Shamash-6tir, m the joy of his heart li.e. of his 

own free will] Asharshi-biti and Ina-bitL-pan-kalammalumurashshu his . . . 
for 16 shdcels of silver, for a fixed price, to . , la, son of Nabtizer-ukia, son 
of Egibi, gave (i e sold). (The document of) the dht (and paqirranu 
ofhcers, which (was taken out over) Asharshi-biti (and) Ina-blti-pln-luinur- 
shu his daughter (and) his slave, . . . -zgr-ukin bears. In the presence 
of Kiittt his wife (Witnesses) , son of SillS ; NabhG- 

nadin-shum ; -tu ; . , son of Lugal-tnarad-j[l44}-dani ; 

of Badi-Uu month Shebet, day 22, year 2 of Nebudia- 

dressar, king of Babylon 

V. S. Suktiiankar 


Pandit Ganapati Sastri of Travancore, to whose indefatigable industry 
we owe the discovery and publication of the drama ChSrudatta of Bhasa,* 
takes evidently for granted that the four Acts of the play published by him 

i[QJMS 1919.) 

a The Chdrudatta of BhSsa edited with notes by T. Gapapati SItetrt (saTri- 
vandrum Sanskrit Series No. XXXIX), Trivandrum, 1914. 



form a diama complete in itself. Indeeld, the assumption is not entirely 
groundless , for, one of the manuscnpts upon which the play pubhshed by 
Ganapati Sastri is based, does conclude with the words : avasitam Charudat^ 
tarn (‘here ends the Chdmdatta’), which is a clear indication that the play 
should end there But the other manuscript (MS. Kh. of Ganapati Sastri) 
contains no such words — a significant difference which clearly needs s(»ne 
explanation. The MS. Kh. is, moreover, as the editor himself tells us in the 
preface (p. i), comparatively frete from errors^' This ought to have roused 
the suspicions of the learned Pandit, but it appaiently did not do so. He 
unhesitatingly follows the MS K. and assumes that the drama ends with 
the fourth Act 

The absence of the words avmtani Chdrudatiam, oi othei words of like 
meanmg is perhaps, after all, not a matter of much consequence Their omis- 
sion may be ascribed to the carelessness of the scribe. Yet another omission 
m the manuscripts under reference, namdy, that of the Bharata-vdkyoi, or 
the benedictory stanza, found at the end of most of the dramas of Bbasa, 
is undoubtedly of a more serious nature. Ndtlier of the manuscripts con- 
tains any such verse. But it may be urged that the absence of the Bkarata- 
vSkya (as of the word avasita) cannot by itself prove that there are more 
Acts to follow. For it is easily imaginable that the benedictory stanza, which 
naturally stands at the fag end of the drama, may have been at first omitted 
by careless copyists and then entirely lost. Against this latter assumption, 
however, may be supported a number of arguments which tend to prove tlie 
tlieory advanced in this article, namdy, that our Charudatta is a fragment ; 
and these we shall now briefly discuss. 

Even a casual reader of the play will notice that the events narrated in 
the four Acts before us are of a very humdrum character and are deficient 
in the organic cimnection between Character and Plot, wanting in that attempt 
at grouping round a passion which la natural to a love-drama. In the first 
Act, Vasantasena, in order td escape from the imdesirahle advances of ISakSra 
and Vita, takes shelter in Ch&rudatta’s house, and utihses the pursuit as a 
very plausible excuse for leavmg with Charudatta for safe-keoring the oma- 
ments which she is wearing In the second Act, Sajh^^aka (CMrudatta’s 
former shampooer, smee dischar^), is rebcued, first, througdi Vasantasenfi’s 
generosity from the clutches of his clamorotis creditors, and then, by lier 
servant from