TECHNICAL TERMS

AND
TECHNIQUE OF
SANSKRIT
GRAMMAR

Kshitish Chandra Chatterji

Pa
wm
= 7
=.
wo
— 7
?)
— 7
o>
—|
&.
me |

)
‘2,
—
pa)
pent
ome
cD
—
 So -te ' | |
7 C , . |
-
od Fe
.
’ r
or q
gh . | f - .
a £
| | .
! 7 * . v7
%
“ oe
2 - ; 4 |
} ! . |
| | | |

oa are | | 3 .
7’
Fi
hae

a
3
f )
te ed,
bane ¥ aa ‘ ‘
yee

i es
ef *

N | _ TE
way

: 1S
|

aon | |
rt
i | be ae:
ThiTECH |
IC
| ze | .
7 2 th | ;
tes - Se
eee 5
|
;
aA

tae
‘a && f

on
-
_ TEC
CHNIQ
4

ae <2 f : | ?
~2=* -{(" ,
=
_ * )

**t
‘J 4
si * ;
rf ¢ s A :
7)
> ‘
' .
~— Ns

[5 —
| |
: _~ .
- |
a5
‘7. ' : |
. a

ar ;
J .

+)
_ >.
+i af d on {
. P
v
! |
“>= ¥ y my bh : ‘
|
ae,
7
. au : “he 2
;
|
s) 7 F

‘230 os ae
+ |
+ ’
ge' ini
-
)
Z . - . é Pets .
|
. : ; a
Gus
s ie iy ce
> Tale
em, iz pa
:
' +s
5 8 e 40d a
“an?
/
- fG a fp +4.
zr \
oan iA we: i
a+-*
,
we ee is y
"Sag
Tao
pis”
“av Z
ats 4
.
a - . u ve 2a
— ee
|
AO
“ tw :
2 oe f a. a - ou fa

tae
io
Liters ee
ta 74
v= >
a 4
mf)
” Pte vit
” fhNoelas
Lag At vs '
é o ait

Pearcy = ve

? View
id , > ee
ars {
ae
* Mie
) .

<€ >
J

=eass
eth
> “t
7 6. - ¥ ae
or
,,
~e . ie ri.
: cu - i fw
"un \ ie FN
. of a *
7 *<* mm 7
1 ® ™
os) an ‘wv,
: .

Fp
=
ate
‘
hE
; |
py
|
2 reENS ether nf ‘
OD c
mS, sed i) +
Pehl a |
x | t ”
: Wiss Wat
;
wey fhe = te}
ence
4,
¥ Gee Wi
© eae ¢ A
Tw “Ne : WPeie lat| |
, | y

«oc 33) Se
Poses)! HE h eA’ | de
Sr
(tia
ay
f, }
oe, 2 ; =] .
‘ Waa
> #3. i
Se e 4- ee ons
oats ee oh Ys -- ~~ 1 | ‘
= ra
3 : ;
8lesgas d ob ie
* (he sue ~ He) i
ate owt rn y ;
eT ai ; ; }

* ts ee a,
» frown,
Musas
-* ows -
.
a. o>

5
=
Ste (
eh de3
. Se oe»
Pe
a
Fstwe
FeeOe
eh |
— ee
ee ow
sass
F Pe eg wT
a
Lex og . \

eo Seley
F Shawren
Cee

~ ANG Heer
<aé ee,
42°?
hdAL |

T 4y and
& RG a>
® AOS oy
«.
4

+86
> > «
Boe es
ow,
ee) ae .
oF Owing -
8é *e~
"Sa 35)
cngOS
**itzece
a,
Atlee 2
>a. ow NS ot
was 2&ns
ae) aoe <2
mr Vine
a = «=F
==
FIs ae he *
al
5Gey- . :
ia Se
Px
122 R ge
‘sa "

Bie
he
we
an

|
be
ee
a
ym,
‘" o
Ghad
Ter

fre
pee.
sow,
Be
Jeyny
=a)

f by
(weatry
tn
*z
a#a@eart
aw:
pao?
e

4
a
Wifes
RAW
43
on-

FW
Oo
~¥>

yn

F ae
*w7 ed
43
.|
ai*
ae cre
' ry~.
Migaiess
oa
OP
Ps
As
7dt
4

¢od a Tad[3
lat mee
IK ‘ a lived
P
ca ‘ema "eree a , Pd ~ . =“ 2 ‘ws . femur ss 3 ‘ al °
~ -- ‘e. , : ‘ore VR ee a We Pur te > 4
8 he ee ee . ¢ f.@2@ «we
aa AM@ncuae — ryt° 4 «rea érc ~ r < “+ were toe ie A
-eis —
ed hl Eenti
- as
ee
=« - >
— eee
- eee. oe _—— - -- angp enaine ~ - c= = - ~
= CD
I~ -
: Meh
_ = “ 9 Yile) - ¢ «)
— eT we
i~ - — - > ©
cea
et - ~ _ = as Ee - — — -
- —— -~= oe —y ~= => —~ — . ” i iia > * ‘as ~ = s
i me : - ous hry oS wt .
Sse@abs - “< .~ ams w . - 9 — ~~ : : . : — ~ a en, = -
— . = ~ - moe - ‘. o th a4 «4
; “se ° =s me ° tr tasercaz : . i > oe - —_— c in. °- > v ’ : o 4 2+ “ ~ = . - ip tetas th > 7 -=— : ~ = ar > +e D. oe
‘ lew ‘
tsven - : ~ 2 . e = . ~ a .
‘ : ie , P.
“=
An
-*
#2 ( : 2
. .
Situae- ciao ice iettmupearizces
= : ad bes
FLAN CARAS ST
. " 2 P 7 a, ~
~ ‘ - ’ “ : ; y om a on .
. 7 : 4 - 1 P A, bed
« . P . =
. ‘ :
- ° Py ; © “
? ? “ = , =
4 4 - vo a ~ ca +
me -,-ece . ; A _—y “@ ‘ove -_xrr
bs J
-* \ . sc > = ’ . . ob “* = . -
* wri 2 4 . P
= om = Chanda eth Gap
ee Ge wa oan a Le ee Pe BP Nee ee oe — =< s
2 *=
“
ag eee eee ee ee ee ne eee eat, Sent a eae ae See
. -- ~~ .
- < ~ “
-
~**
2

ch —
-
te

——se
= ~~ : = . . >
- ~ - = . ~ ~
- - . ~~ . - - - -
2 3 oO & @ ee ee HS ee = = = .
- - 2 = -
~
~~~ ——<«- +,
- “ _ > - aie
- = ce ee DG - » = a“
pa : - -_* + «- - - ~~
. = ~~ ~
= —
Ze i =* » ms So ” , : . acs «
. ee -
" =

eg
2 Su 9 6 a : ’ : : : ‘ . : . : . ‘ ee ote he
. ,

ae
“

2. - :

ne
-. :‘ce
ee

s.2
- ~

:1
; ~ ~~ 4

myo"
=

'»
.
.
.
:
;
r
>
.
as
.
- t, ~| yy - ou

yy
te

.
«
‘

ea
’

°
;
:

7*@
=

“
’

od
”
;
¥

tad
£
”

2
‘
,

+
,

=~

fy
-*
*

a,
:

a
tah

e
a

«
=
>)
’
~
°

.
~

tt

on
Technical Terms and Technique of ‘
Sanskrit Grammar

Kshitish Chandra Chatterji

SANSKRIT PUSTAK BHANDAR

38, Bidhan Sarani, Kolkata 700 006
Published by
Debasish Bhattacharya
SANSKRIT PUSTAK BHANDAR
38, Bidhan Sarani, Kolkata 700 006, India

© Reserved

All rights reserved. No part of this publication is to be produced in

any form or by any means, without written permission of the
publisher.

Book prepared under the auspices of the Department of

Special Assistance (DSA), UGC Sponsored, Sanskrit
Department, Jadavpur Universtiy, Kolkata

First Reprint Edition : 2003

Price: Rs. 150.00

Printer :
Abhinaba Mudran
Kolkata 700 006
 Preface

It is a matter of great pleasure to place the present book in the

hands of scholars and general readers. The Vechnical Terms
and Technique of Sansknt Grammar is an epoch-making creation
of Prof. Kshitish Chandra Chatterji, a scholar of an unique
type, who traversed with ease over the vast and thorny fields of
Sanskrit Grammar and Comparative Philology. We. still
remember the days of our student-life when Prof. Chatterji
was seen devotedly collecting and shaping variegated contents
for his highly acclaimed Sanskrit Journal Manjtisa wherein
many knotty problems of Sanskrit Grammar and Linguistics
were raised and solved in such an interesting and smooth way
that one could not but be wondered at Prof. Chatterji’s
brilliant and sharp intellect. The Calcutta Orental Journal, the
founder-editor of which was Kshitish Ch. Chatterji himself,
earned high encomium even from eminent Indoligists like
A.B. Keith and M. Winternitz. Two highly contemplative
articles, namely, Paninz as a Poet and The Anubandhas of Panini,
full of original ideas, written by Prof. Chatterji in the Calcutta
Oriental Journal (COJ) Vol-I, No.1, October 1933 are
incorporated in this book for the benefit of the readers. We
could not resist our temptation from including these two
thought-provoking pieces of Prof. Chatterji’s creation on the
great Grammarian Panini and his ideas, as the appendices of
the present book.

The Yechnical Terms, etc. is a critical and comparative

study of the technical terms found in the Niruwkta, in the
Pratisakyas and in the different systems of Sanskrit Grammar.
For the treatment of the subject, Prof. Chatterji scientifically
utilises the views of Panini, Katyayana and Patanjali and their
2 Technical Terms and Technique of Sanskrit Grammar

rules have been quoted frequently in this book. In fact, Prof.

Chatterji’s work is of a novel type that has not been
contemplated even by modern, highly enlightened giants of
learning.
We are fortunate enough to have found out some rare
writings of our esteemed National Professor Dr. Suniti Kumar
Chatterjee on the life and activities of Prof. K.C. Chatterjl.
These have been included in the Prefix-part of the present
treatise.
We are indebted to Professor Visvanath Chatterji, the
worthy son of the great father, for, he has given us consent to
our request to bring out the reprint-edition of this learned
book from the Department of Special Assistance (DSA) 1n
Sanskrit, Jadavpur University. Professor Bijoya Goswami has
checked the Introductory and Appendix-portions, and
Professor S. R. Banerjee has on his own gone through the
proof. Professor Banerjee has also given some useful
suggestions for the improvement of this production. Thanks
are due to them. Sri Debasish Bhattacharya of Sanskrit Pustak
Bhandar is also to be thanked for undertaking the
responsibility of bringing out this book in a new and neat
shape.

Manabendu Banerjee
Professor & Co-ordinator, DSA (Sanskrit),
Jadavpur University
Kolkata
 Contents

Preface by Manabendu Banerji 3

Professor Kshitish Chandra Chatterji
by Suniti Kumar Chatterjee 7
sem- faders agree: — eifa gar agrreaa ie
Dr Kshitish Chandra Chatterji — A Brief Sketch 14
Seas — faire VST: 23
Technical Terms and Technique of Sanskrit Grammar 1-452
Appendix :

Panini as a Poet by K. C. Chatterjee 453-479

The Anubandhas of Panini by K.C. Chatterjee 480-499
 Professor Kshitish Chandra Chatterji
(Obituary)

tis with a very heavy heart that we are presenting what we

Lapprehend would be, for the present at least, the last
number of the Manjusa. The Founder-Editor of this Journal,
Prof. Kshitish Chandra Chatterji, had been seeking to serve
the cause of Sanskrit studies in India with a single-minded
devotion for nearly the whole of his career as a Professor.
Among all his serious work in this connexion, we have to give
to Manjusa a very high place. This Journal, one of the dozen
Sanskrit journals that are still alive and about which India
could boast of at the present day, brought in quite a new
atmoshpere and a distinctively modern approach in the study
of Sanskrit. Prof. Chatterji had concentrated on the study of
Sanskrit Grammar, in which he has made some very fruitful
contributions. Grammar and Linguistics are generally
considered to be dry-as-dust subjects, but Prof. Chatterji even
in his Manjusa brought in occasionally a play of his humour
through Sanskrit, and that was not the least thing in ensuring
the popularity of the Manjusa.
Sanskrit studies unfortunately are at a discount in India
at the present day. We had all hoped that, after
Independence, Sanskrit would once again be given something
like the status which is its due. It ought to have been made at
leastone of the Official Languages of India. Surely, for
communication with the outer world, when formalities are to
be observed, it should have been Sanskrit, and no other
Indian language. That was the proper language for our
Ambassadors to present their credentials in. Sanskrit 1s our
mental and spiritual link with the Indo-European speaking
Europe and America. It is also our great link through
 Technical Terms and Technique of Sanskrit Grammar
8

Buddhism and Brahmanism with the countries of Central Asia

of ancient times, and of South-East Asia, Indo-China and
Indonesia, and also with Vietnam, and China, Korea and
Japan. Sanskrit, as Sri Rajagopalachari, the veteran
administrator and political leaderof India, has rightly
described, forms “the symbol of our Seniority among the
Nations of the World”.
Sanskrit is a great discipline, and we very frequently
forget this central fact in our Sanskrit studies. To preserve the
integrity of this discipline, a handful of Sanskrit scholars, who
themselves were erudite and who had a_ whole-hearted
devotion to the cause, have been trying during the last three
generations to maintain the dignity of Sanskrit studies
through their own exertions — through their own
contributions in the domain of Sanskrit research as well as by
trying to inspire students and others. Prof. Kshitish Chandra
Chatterji was emphatically one of this small band of selfless
scholars who had devoted themselves to this task, and
whatever they did was from an inner urge : the great
satisfaction which they derived from these studies formed for
them its only reward. Their work was not for money, but it was
as Tulasidasa said about his own great work, the Ramayana,
translated into his mother-tongue Awadhi, 17:qaM, for the
happiness of his own inner being.
I need not give a sketch of the career of Prof. Kshitish
Chandra Chatterji. This has been done elsewhere in this same
number. I was privileged to have him as an esteemed
colleague for a long number of years in the University of
Calcutta. He had taken his M.A. in Sanskrit, and had studied
Comparative Philology and Linguistics under Dr. I. J. S.°
Taraporewalla, of illustrious memory, who was for a number
of years Professor of Comparative Philology in the University
of Calcutta. Prof. Kshitish Chandra Chatterji spent most of his
life in the University of Calcutta as a Lecturer in the
Department of Comparative Philology, and in that capacity |]
had him as my colleague. We all admired not only his great
erudition but also certain sterling qualities of his character.
 Professor Kshitish Chandra Chatterji (Obituary) 9

He was steadfast in his own convictions, and intellectually he

was absolutely honest. It is very rare to find a man of that
calibre in any department of life.
Coming from not a very rich middle class family, he
himself was never able to make money, and a large proportion
of his slender means as Lecturer in the University of Calcutta
was devoted almost entirely to the dissemination of Sanskrit
learning through Journals. He brought out an Indological
Journal of very high repute, The Calcutta Oriental Journal,
which created a very favourable impression among scholars,
and was responsible in publishing a number of very good
articles on different aspects of Indology by eminent scholars.
Then his next venture was the Mavijiisa, and this Manjiisa he
has been publishing, although with great financial loss, for 16
years and more. With the exception of a small grant of Rs.
2500/- from the Central Government, which came to him just
before his death, all the expenses were borne by him. It was
too much to expect an impecunious scholar, though of great
reputation, to be the financier as well as the editor of a
learned paper of this type. Up to the last Prof. Chatterji was
very optimistic, and what he had closest to his heart was to
make the Manjusa a bigger and a better paper. But
unfortunately this could not happen, as he fell ill, and finally,
as is the fate of all human beings, he joined the majority on
the 8th of November, 1961.
We now find it difficult to combine both the erudition of
the departed scholar, and the wherewithal which is so very
necessary and which he alone could somehow find out. It 1s
therefore with a very sad heart that we are forced to stop the
further publication of the Manjusa. If better times come,
possibly the work may be taken up again. Prof. Chatterji’s only
son is a Lecturer in English in Jadavpur University, near
Calcutta, and he has in him the love for Sanskrit and the
family tradition in Sanskrit studies. But without adequate
financial resources, much as he would have liked to continue
his father’s work, he cannot do it, at least for the present.
We can only hope that those who have liked the Manjusa
10 Technical Terms and Technique of Sanskrit Grammar

— and we flatter ourselves that their number is not smal]

throughout the whole of India — will give an occasional
thought to this Journal which had been so remarkably serving
the cause of Sanskrit, and will have a prayer for the illustrious
founder of this Journal — requiescat in Pace.

Suniti Kumar Chatterji

Calcutta. Emeritus Professor of Comparative

25th January, Philology, University of Calcutta,
1962. and Chairman, West Bengal
Legislative Council.
 eora
stTard m
aaS OTe:
( FyTTATTTT)
a UTR: SOM aad SHAN sa Bsa, aga BAIN
camara agua: wWerisg fl sen: wren: vias 4
amen aaa: dessaaS Wa ae HITT
aitt
earga
fafsaa farsa undad Sepaynerta sid sia aeeetg
ay
ae afasaqel: wendfal area cugerwran: depaafaet srenfa
staf aeeifre amcad wsafa wéaquiagy, ‘Ne’ dere! depa-
awicrfadt Age are aarti fatigues wedaqi fase Fay
ages STASIS: GepdeaEcraaar| THT ae
e
fad)
fasraa: WerqEaaHAa AHO wea areas Pra Arafafa

AO SATA AAA TST |

TUATHA
qraee AHA Het URday Pepadatn: watt a evadi wa
FASANO Ad AAAS APIS AAI aarivayq afta wateta,
atddl URE US TERa aa Ge Wed ae daa
afef ararenfacit
Tae een: wafad a, A GT wale ardiaa asa) eT
Urea uaa Sarg HTT aT) S-gqedta'-snfen ET
aad wera ae aie a ween waaseaicas daa
CePA | eaA-
gri-nenfakadesntamitaansararesey |=|-
See Rd ST -araqs-d-aiha-stamifefua Ge Pepa Aes
Sra) ARs Vata) Ustad SsATA Tatars retreat Stara
atdrpar aq ded wel faa wfaqeseseark Vigeses Wet" St |
tepid fe nefararenfa uaittafata depraratqeree Wao
fuera deataer favre fagfeend gears aera a A
faa dence fagttisdaqeyad aad Sepa Tafenanes PeperTasnaat
aaa <a Ree n
Warner s a SAS STATA ATSa
rea: |
Te, actin Harifasdiest Faratactarearcn: feriersag
 12 ‘Technical Terms and Technique of Sanskrit Grammar

ada tape waste HUOTaeui WaT | TeaTaPaTaSer aT aT

Ted Haan ye Sartore | Sat waref ae. taal deraterat wen
deranect «Aral Serrated $=Waray Sadar ‘tar: ara’ sfa
TAHARI STAT |
AAThS] fara ae eqns a an fequn-
ces red, Ted od Gea afar | plete -fa aaa aay
Pan aREMea HATS cad Bleed Barra eta | W AeA]
HC Reng ari aia were ararhey wa-seie
YeVAI NIA hAS AA «=ATT ays Was wT ay
Taatrere
Sa Gera
ra Wea WTA Ta, ARTA AeTea:
leadiag oor) sean eareeiere yaa seiner
TAh- Waa aida, dea SF Fanaa ae Ueafifarn aerais |
4 hat a4 qa ufeata wif: wag rhs
radish | way faaeg aaa: 8 qeenfa wer aneiqi sflarey
Te Waal GA: aed Wa SyaG|
Wyte S MR Hema Har da @ wees | 4 Hefag a4
Wan! «6tfaarnfaaferereearaa aq fea wet wee
Tea fatawiaae dpi zara) a warefhend
Teatcentacts fea otra ‘fe aera aithavea ote" arr We |
Waereme fagaacpedaiscdiq vanes faenaat faectea
weer fatwafeekr: aq yftuetarel facen:) aa: wR wares
“Fg | afadteredh, sag, at yarwad wyanatesrd <ipcarta Wea:
racrreaiega uRaedarvent weigaeaeatat asi afi faa wast
SR wate Ga:| aew: waaa
quady
veya: WitSeryVUrasaar:

TANTS: GRAM SAR | a eaarea e SETA

wo Femi a ady sau qumdafaatads aassed aa: FT VMAS HE
ATT ATT ATTA WAT, RASTA TRAP RTATATEET ( 28E8 )
7jartiaeatea
water nerd |
area qaretes Gen aeemetle qe area aaa — a
a stan mente aan — SABI Case Pea |
od Ud Gaenaseadseiuhtd Sed ASEM: WAN: | Ay Maa: ward,
eATaSUa: Rea) sea wa WS: afemrarettoahtts
Wea aaanead Sars afaurirais args: | Se A aa stata depd Wem:
 seq -fadiraeaerneaa: (Syfadtyory) 13

Tapas: SPARES! |«W aeavaHy snfist acer aredifa

saree d:
Ferrata fare Ts HAS — Ad AGH GAA |
Fe arn
aq macnieta He e d —
esaatra Sai den fifa
added ThA :
— a deparare
SAA facet aa Haar
sai AMET Fate Seats wieliferd dae: vada Fara ePaata

straint agrareara:

aiacaslal-v, afaam-faafaenaaet Geyer Watdca-Ta aT

Qu WAAR, 22G2 Il sata: warner: Ussag-fauraaeeaas
 Dr Kshitish Chandra Chatterji
A Brief Life Sketch

Born on the 24th of November, 1896, of middle class parents,

in the ancestral home at Jorasanko in Calcutta, Kshitish
Chandra Chatterji (Chattopadhyaya) was educated at Hare
School, Scottish Church College and City College. He took his
M.A. in Sanskrit from the Calcutta University in 1918 and later
received his D. Litt. from the same University for his studies in
Sanskrit grammar.
He was for a short time a Professor of Sanskrit at South
Suburban College (the present Asutosh College), Calcutta.
Then he joined the Post-Graduate Department of the Calcutta
University as a Lecturer in Comparative Philology and Sanskrit
and served the University in that capacity for thirty five years.
A brilliant student of the Calcutta University and one of
its most eminent teachers, Dr. Chatterji’s aim in life was in
consonance with the motto of his alma mater— ‘Advancement
of Learning’. He was a great scholar and teacher who was
enurely selfless in his pursuit of knowledge. He was
Passionately devoted to his work and dedicated his life to the
Cause of learning. He had a many-sided interest, but was
especially interested in literature and philology. As a scholar,
he specialized in the Vedas and Sanskrit grammar. His
knowledge was profound and his scholarship deep. His books,
journals and research articles bear an eloquent testimony to
his erudition.
The Vedic Selections that he edited for his University is a
splendid work. It has, moreover, attained the practical end he
set before himself. It was primarily meant for students and
teachers in the universities and he had elucidated many a
difficult point with remarkable clarity and precision. Every
 Dr Kshitish Chandra Chatterji — A Brief Life Sketch 15

single quotation in the commentary of Sayana he traced to its

source. The section on the Padapatha of Sakalya appended to
the book is an original contribution to Vedic scholarship. His
own commentaries on the Usa Sikta, Devi Sitkta and A No
Bhadriya Stikta unfold the peculiar beauty of each word and
expression in these hymns. The Vedic Selections remains an
indispensable work for the beginners of Vedic studies. In the
Technical Terms and Technique of Sanskrit Grammar, which 1s a
critical, comparative and_ historical study, Dr Chatterji
produced a really epoch-making book. It is a unique work
dealing with and throwing a flood of light on the technical
terms found in the Nirukta, different Pratisakhyas and the
different systems (from Panini to Harindmamrta) of Sanskrit
erammar. It offers for the first time rational explanations of
even such apparently meaningless terms as lat, ghi, ghu, bha. It
gives reasons for the use of particular letters as Anubandhas by
Panini and his successors. This book has been acclaimed as an
outstanding work among modern Sanskrit studies. Candra
Vyakarana is the first Indian edition of the work of
Candragomin, critically edited, and though based upon
Liebich, utilises the quotations in the commentaries of
lexicons and other systems of Sanskrit grammar current In
Bengal and suggests better readings in many cases. For facility
of comparison, the views of Panini, Katyayana and Patanyjali
have been quoted under every rule. Bhoja’s Sarasvati-
kanthabharana Dr Chatterji regarded as the second great
revised edition of Panini (Candra-Vydkarana, according to him,
being the first), and rules from Bhoja also have been included.
This work is one of the most distinct intellectual contributions
made in the series “Sources of Indo-Aryan Lexicography.”
Patanjali’s Mahabhasya (Paspasdhnika) has been presented by
Dr Chatterji with a brief but illuminating introduction, English
translation, explantions in English and Sanskrit, and well-
chosen excerpts from commentaries. Perhaps the best
available introduction to Patanjali‘s great work, the present
edition offers a translation which is remarkable for its lucidity
and sound scholarship. And the content of the book finds a
 16 Technical Terms and Technique of Sanskrit Grammar

thoroughly congenial and competent smlerprever in Dr.

Chatterji. Greek Proverbs (for students of Sanskrit) presents
some fine and pithy Greek sayings In the original, with their
Latin analogues and exhaustive explanauon in English of all
the Greek and Latin forms, parallel proverbs from English,
Sanskrit, Bengali, Hindi and Marathi. Dr. Chatterji was no
mere insular Sanskritist. He was well-versed in many languages,
and this remarkable little book, like that excellent
Abhanakamala series by him in the Manjusa, is a boon for all
beginners of Greek and Latin. Cntical Observations on
Ajayapala’s Nandarthasamgraha shows one more facet of
Dr. Chatterji’s brilliance. His reconstruction of the corrupt
text of Ajayapala’s dictionary is extremely successful and shows
rare lexicographical talent, Sabdakathd is perhaps the first work
of its kind in Bengali. This lively and fascinating study of words
and their use shows not only his erudition and wide reading,
but also his ability so make the dry facts of grammar and
philology eminently readable, often with flashes of delightful
humour. These are some of his masterpieces, but there are
many other works which he has left (some of these are yet to be
published) and all of them show the same excellence of
scholarship and critical acumen. His articles and papers in
different Indological journals are also in point here.
Dr Chatterji had himself been the founder-editor of the
Calcutta Oriental Journal, Manjusa and the Bengali monthly,
Surabharati. All of them won high praise in India and abroad,
the Calcutta Oriental Journal being the most famous. It was
welcomed by the most distinguished Indologists in the West
like Dr Berriedale Keith and Dr M. Winternitz, to name only
two. Dr Keith wrote in an appreciation that the Editor of the
Calcutta Oriental Journal “always throws welcome light on the
topics discussed.” Dr Winternitz wrote about this Journal, “I
have often to refer to it just now in preparing the English
translation of Vol. III of my History of Indian Literature.”
Dr
Chatterji’s single-handed effort to revive the glory
that was
Sanskrit through the Majfijiisé is bound to inspire admirati
on
in every one. It is one of his greatest achievements.
It has
 Dr Kshitish Chandra Chatterji — A Brief Life Sketch 17

recently been described by Professor Louis Renou as a

‘precious’ periodical. Dr Chatterji’s articles in the Manjusa
show not only his wonderful command of the Sanskrit
language, but also his intimate knowledge of the different
branches of Sanskrit literature. His innumerable grammatical
and philological discussions published in the Manjusa deserve
special mention. The Surabharati was the first, and perhaps the
last, Bengali monthly mainly devoted to Sanskrit literature.
Many of the Editor's powerful articles on the different aspects
of Sanskrit literature were published here. Dr Chatterji also
edited, for some years, the Journal of the Sanskrit Sahitya Parisad,
Calcutta, and Oriental Literary Digest, which contains some fine
and trenchant review-articles from his pen.
Dr Chatterji had a hard start in life and financial worries
clouded even his last days. This financial difficulty was mostly
due to the publication expenses of his books and journals and
his unstinted help to poor relations and friends in distress. An
upright man, he had been ever a fighter and his
unconquerable spirit never failed him. Even to the last day of
his life he continued his study — which was a passion with him.
After a brief illness, he passed away peacefully on the 8th of
November, 1961.

Select Bibliography
Books
1. Technical Terms and Technique of Sanskrit Grammar.
2. Upasarga and other Technical Terms.
3. Popular Etymology.
4. Greek Proverbs.
5. Critical Observations on Ajayapala’s “Nanartha-
samgraha’.
6. The Siva Sutras and the Sanskrit Alphabet.
 18 Technical Terms and Technique of Sanskrit Grammar

7. Sabdakatha (in Bengali).

8. Usar Alo (in Bengali).

Books edited
1. Patanjali’s Mahabhasya, Paspasahnika (with Eng.
translation, notes, explanations in English and Sanskrit,
etc.).
9. Candra Vyakarana.
3. Vedic Selections.
4. A No Bhadriyam (with.an original commentary),
Devistkta (with an original commentary).
Stri-Prasgamsa (of Varahamihira, with an original
commentary).
7 Siddhanta Kaumudi with Lilavati.
8. Kiratarjuniya, Cantos I & II, with the commentaries of
Devaraja and Madhava.
9. Raghuvamsa, Canto I]
10. Raghuvamsa, Canto V.
11. Bhattikavya, Canto I. With commentaries
12. Kiratarjuniya, Canto I. critically edited.
13. Bhagavad-Gita, Adhyaya II.
14. Convocation Address in Ancient India.

Readership Lectures

A serofies
lectures, delivered at the University of Calcutta
on the predecessors of Panini.

Journals edited
I. the Calcutta Oriental Journal (monthly) from October
1933 to March, 1936. |
2. The Monthly Journal of the Sanskrit Sahitya Paris
ad from
March, 1930 to April, 1936. |
 Dr Kshitish Chandra Chatterji — A Brief Life Sketch 19

3. The Sanskrit Monthly Manjtisa from February, 1936 to

January, 1937. Revived in 1949*
4. Surabharatt (Bengali monthly) from November, 1935 to
October, 1938.
CST Sanskrit weekly Manjusa@ (for a short time).
6. The Oriental Literary Digest (monthly), Associate Editor
from August 1937 to July 1939, Sole editor for Vol. IV
(1941-42).

* Professor Sunitiv Kumar Chatterji, M.A., D. Litt. (London), Chairman,

West Bengal Legislative Council, wrote in a review :
The Manjtsa is run by Dr. Kshitish Chandra Chatterji of the
University of Calcutta, who is one of our most erudite authorities on the
technical and other niceties of Sanskrit grammar. He is a student of
languages and in addition, he handles Sanskrit with ease. It has been his
love of Sanskrit that has made him carry on the work of editing and
bringing out this paper since its inception by himself. Financially it’ has
been a thankless task. It has brought other disadvantages also — the editing
and financing of a journal without much wide appeal because of its
language has been a source of worry which has had its repercussions on his
health. But he has not given in, because he is an incorrigible optimist when
he is convinced his cause is noble. Prof. Chatterji is quite an institution in
himself. Then, he has been able to enlist a number of disinterested
colleagues who help him with contributions and who are enthusiasts for
Sanskrit like himself. And he himself writes, It is a remarkable thing that
although he finds joy in unravelling the intricacies of Sanskrit grammar, he
is notin his mental make-up a dry-as-dust grammarian. He has a wonderful
sense of humour supported by an inexhaustible fund of anecdotes and
stories which he brings out in racy Sanskrit. So the pages of the Manjtsa at
times sparkle with laughter. Comparative linguistics he brings in frequently
in his articles, and this isa remarkable “special feature” of this litte paper.
Further, it does not confine itself to Sanskrit alone. Proverbs, Bon mots, wise
Sayings, Mahdvakyas, etc. from different languages, English, French,
German, Latin, etc, are rendered appropriately into Sanskrit,
demonstrating the capacity of the language of the Gods to keep pace with
turns of expression peculiar to various language. One can say also that
there isan “English Section” to this magazine, devoted as it is entirely to the
culture of Sanskrit. Thus, the editor himself is bringing out through this
magazine, in instalments, his English translation and commentary on the
Mahabhashya of Patanjali which, when completed, will be a valuable
contribution to the study of Sanskrit grammar through English. A journal
like this deserves a much wider appreciation which is its due.
 20 Technical Terms and Technique of Sanskrit Grammar

Papers and Articles

i- Review of Rajwade, “Words in the Regveda”. Indian

Historical Quarterly, Vol. VIII.
Commentaries of the Ulttararamacarita — HQ. Vol. IX.
Vyosa — IHQ.
Kasakrtsna — IHQ., Vol. VIII.
The Author of the Bhagavrtti. — IHQ.
Misconceptions about Certain Mimasmsa Terms. — IHQ.
Apisali — IHQ.
The Authorship of the Anunydsa. — IHQ.
wo
DP
pp
ON
AT
w Some Technical Terms of Sanskrit Grammar. — THQ.
Vol. [X.
10. Panini as a Poet. — Calcutta Oriental Journal, Vol. 1. No. 1
Tok. The Mugdhabodha System of Grammar. — GO.Je l.. 2,5
Iz: Caitanya and the Pandit — C.O,J. I, 3.
LS. The Anubandhas of Panini. — C.O,]. 1. 3.
14, Savarnya of rand 1!— C.O,J. I. 4.
15. Some Grammatical Notes. — C.O.J. 1. 4
16. Grammatical Doctrines and technicalities in the Kavyas.
—C.OJ. I. 7.
17. The Tarkikas and Sanskrit Grammar — C.O,]J. I. 7.
18. Some Laukika Nyayas. — C.O.]. II. 4.
19, some Rules of Sandhi (a Comparative Study of the
Sandhi in the Pratisakhyas and in all the systems of
Sanskrit grammar) — C.O,]. II, 11.
20. Some English Parallels to Bhavabhati. — C.O.J. I, 11.
21. The Astadhyayi and the Siddhanta Kaumudi— C.OJ. Il,1.
22. A Rule of Panini. — C.O.J. III, 2.
28, Jagannatha and Bhattoji. — C.O.]. II. 5.
24. Sundari and Kamala, the Learned Wives of Ghanasyama
— C.O,J. IT.
Some Technical Terms of Sanskrit Grammar. — C.O.,J. II.
 Dr Kshitssh Chandra Chatterji — A Brief Life Sketch 2]

26. Some Notes on the Naisadha. — C.O,]J. Il.

27. A Stanza from the Bhattikavya. — C.O.J. Il.
28. Some Sanskrit Stanzas. — New Indian Antiquary, VI, 8.
29. ‘Technical Terms of Sanskrit Grammar. — N.I.A., VIII,
2 & 3.
30. The Language of Visnw Purana. Pracya Vani, July 1944.
31. Some Technical Terms of Sanskrit Grammar. — Prdacya
Vani.
32. Greek Words in Sanskrit. — Calcutta Review.
33. Grammar and its Critics. — Calcutta Review, November &
December 1929.
34. The Siva Sutras. — Journal of the Dept. of Letters. Calcutta
University, Vol. XXIV.
35. ‘The Critics of Sanskrit Grammar — Journal of the Dept. of
Letters. Calcutta University, Vol. XXIV.
36. A Latin Quotation in Indranath’s “Kalpataru”.
7. Golden Thoughts from Goethe. — Calcutta Review.
38. Chhatri vs. Chattra. — Calcutta Review.
39. Psychology in Grammar. — Calcutta Review.
40. On the Cases Governed by Some Sanskrit Particles. —
Siddha Bharatt (Visvesvaranand Research Institute).
41. On the Interpretation of a Rule of Panini. — Indian
Linguistics (Chatterji Jubilee Volume).
42. Samprasarana — Vak.
43. On the Text of the Atharva Veda. — Vak.

Sanskrit
1. Kovidananda of Asadharabhatta — Journal of the Sanskrit
Sahitya Parisad, Vol. XIII.
2. Kavyesu Maghah —J.S.S.P., Vol XIII, No. 9.
3. The Siva Siitras —J.S.S.P., Vol XIII. 2.
 22 Technical Terms and Technique of Sanskrit Grammar

4. Panini—J.S.S.P., Vol XIV, 5.

9. Candralekha-Sattaka — J.S.S.P., Vol XIV, 10, 11.
6. Ancient Grammarians — J.S.S.P., Vol XIV, 9.
7. Vyosa —J.S.S.P., Vol XIV, 6, 10, 11.
8. Sarva Sastra Prabodhini (a work on Mimamsa) el ee
Vol XIV, 12.
9. The Samnjrda-section of the Supadma — J.S.S.P., Vol XIV, 9.
10. Yosiddharma — J.S.S.P., Vol XIV, 10.
Il. Vyakaranadhikarane garjante mimamsakah — [anche ay
Vol XIV, 6.
I2. The Story of Sakuntala in different Recensions —
J.9.8.P., Vol XIV, 2.
13. Sabara Bhasya with Commentary. — J.S.S.P., Vol XV.
14, Sahadharmini-carya. — ].5.5.P4 Vol XV, 8.
15. Bhavisya Purana — J.S.S.P., Vol XVI, 12.
16. Meghadita with the Commentary Séroddhdrini (hitherto
unpublished) — J.S.S.P., Vol XVI, 3.
17. Kdatantrasya Vaisistyam
— J.S.S.P., Vol XVI, 5.
18. The First Stanza of the Uttara carita— J.S.S.P., Vol XVI, 5.
19. Linga-Vrtti of Vararuci (published for the first time) —
J.S.S.P., Vol XVI, 12.

Bengali
Bhasay Samkar — Purnima, Puja Number.
2. Sabdakatha— Purnima, Independence Day Number.
Bhasar Upar Sddrsyer =Prabhava =— Pracyavani,
Pravandhavali, saptam puspa.
4. Sonar Patharbati — Masik Vasumati.
D. ‘Sdradiyaé’ and ‘Sarvajanin’ — sanibarer Chithi.
 stealaeh Tard yTa aS OTe:
( eataraaory)
(C86 faeries adfdufead (2303 GSTelaMSaTey 208 fea)
Heat weaned
alctadrid-aeset iniean MA |
fadigas-ETAa:
a cana -hieradacds-adiacs
faen-aitsrt
ecrear erny |
g framed
aictaiaatdaeadad «depart we uw wWergdviagq faced
asaFela gqa (Ser
ary se HaAS) Pepa srdiq sta:
4 afiamiasadas adamant derma das
aaa Fd AAT | wad vafiveenra fastened fae seta Id
OF STRUM Fea dareaies faafeenrcrarefinra: fs fee seqarfy:
TeMeaTaaeleal 4 hac Aictarnasancaaes
Hal oa ark WAS
CAAT ATA | a uUftsayameneae: aryui-careien:
Weeasaete| SA wi farsi fared wan sled faarsaaetqegerd |
qey fess aes WR aie ATTA ARTO ATA I
qe Srawianacrat
hd aa afaeseqaaq fama sedis aren oft:
Whsae Ga A Hal Wen: wear Tae Faas adtadeurentarHe
aesATaeld |
dala-fa afaeetraes Hd WA GEfadl ACA NS TA: WHIHE: | Weiss
Taare SATA od ASM: Fa Sq! wanes
Tea Fe Seow Feqget fared Sremanfaaq! War
We TIeRigegat Vaada ae a waaay) SAPs
CASI TSS S Caspar Srary aa vitared vfaereea traviaafasra
Myatt aAelfad: | Mes Vers acakfeariq
Pals dara afea-
ae3) difsafier ate eee: | Ure Sfemaieerer yemfrenhiferaterdsa|
oar uUfsnitageasicg ‘cata adq sive tah aa
PaePS TAR! VIG AT Tee Cad Aye wa a: 1 Tesh geri
Ceara ha Has eep-danacrt fear: frewe-wiasres-ahtrta-
PIPA-GR SR VMHeTAA-CH-AaaR ASN -ARAT—_ FS -WaP RTOT-
 oN
Grammar
24 Technical Terms and Technique of Sanskrit

eeofs ae, ATTTAATSeTAAT ete -TA-F-

DeTAASH |
|saioeyeen aaa ae eT
WIE
TIS ASTI
ITAC
=Agar H-aAeF Saas ARS
SMIAAT: |

aay feet ue-aeta-arenemi fear Sy TTY Sara Aa:

We: Wate: | adresse Wea! ST:
wages: Wrersardatcrad

TROT) Tasha virsrerrHcoreta Barsigans | eas AAT We:

yf
Wataerneay (cemfery) | a Bee reat
SISA: sneredepadra Waren Arata atazaern: | saa
qWafaner waaufaata A) Meda sare -uifeeryanty
wasntrens
Ayal Sfyreeaqnaned | Sa TOA Pra var eT eT Perea
Tayaenera (sepa Fa way) =i aah
ain aare-gafaat add) eT Meteosat wecit! |aat aai
fasendra fatsqyanty
area) WePATS, AH le MEN aA
SYM: YEA: | St: AAAS a Hees A act Wea wary, Hasty
T wERtehf asd waenrnesar oqafteqarareitatercsry
TRA) was ASAE TA|
Assos Venere AeA
Sess SI: VEMAesreresy fee WeReTATaTea |
ATTA Hleafapar Wer A aa Papas Fadia Wasa | Teer Wat
aera |
qerententraiescrer
WHAM | AFMUWAH CAM WA: | Wee Wega aa
GATS
Utena 4 Hace WSTtSst-aaTE ra araaite ufcaat ay
TR awry wore Tear fsa Gast Maran aa y
grasemen
Premereraga: apereats aera
Praca ta
UW Ta wea vac | Sit Tears vera wettest ay
Targa yen faarges vetting df add
wae sfetenrg wag wereatantrity tary varia fore: aerate |
a sere: ead aaa sitfeewerraraia’ rae Carefeat
ates
neal a) agar “qa” aeatat wart wares ae ued
ea
aaaa wae sti) «carta ‘‘aererarer siitwersmia”’ saree
 TASER PPIs Fed fea: Westar a uy aafwat weer WA

TH eTaTSa ACTA A Hach TATA S Yea <a SACS RSIS ATG Ot UHI

aeuefiag wag weg) yagssqnaa wea: Wd

START
AM AAAI
AGA FASTAUTEN: |
gun teprfecrrateaieer yen (aareath vise) agar
Tifaeiseat| deprirareary favaqusite wares wan Aa Ayer

sifetch cali We atine WEA aaa aes GST Sra

stata fester = Bea WAHT aT,
USA: TSS Vast Sse Gaps Teas! F
Geary, Frenvat 34:1 Saas FAO astesisias
aaa siwateas
Teast = STATI mraerraaeaasa kkee) |OTR
Faas
GISTAH ATAAAAT |
 PREFACE TO THE FIRST EDITION

In the early stages of literature and science the symbols

used are generally self-explanatory, but soon there comes
a time when they becoine obscure. If at this stage there
is a break in the tradition, the obscurity gradually
increases until someone at last succeeds in unravelling
the mystery. Thus in the marriage-hymn in RV. we find :
aia: saat fafa? aeaat fafae sac: |
dant afas afatqctiae wacezan: ||
QA FFI WIAA Waray zrzNaq |
uff a gataraiafanaat arg ||
‘Soma got her first, Gandharva got her next. Agni is
your third husband and your fourth one born of man.
Soma gave to Gaudharva, Gandharva gave to Agni,
Agni gave me wealth and sons and also this wife (x. 85.
40-41).
The sense of these. stanzas must drave been perfectly
clear to the seers of early days, but they became obscure
later on and Sayana inerely gives running paraphrase
without any attempt at explanation. There is hardly any
doubt, however, that three stages in the physical deve-
lopment of woman are referred to here. A stanza in the
Grhya-Samgraha points in-the Sainé direction.
Again, take for instance the following stanza:
acaife ata ofcfaar varfa
atta fagataar & aatfaur: |
er aifin fafear Agata
gut arat ager afta) uceeres
‘Speech 18 measured in four quarters, the Brahmanas
vi PREFACE

Three (of these quarters) are

who are wise, know them. fourth
It is the
hidden and do uot bestir themselves.
speak.
quarter of speech that men
appears
The explanation of this stanza found in SB.
in
to be correct. The Brahmana enumerates four stages
the evolution o f speech, the first being represented by. the
hissing of serpeuts, the second by the notes of birds, the
third by the sounds of be asts and the fourth by the speech
of men:

ataq atid araisfaerd wage agia) wiag

Gua arastrerd 47 qua aga | Aaa GMa aFat-
sfaaad agaife acfea | atad qua wasted afze
ag atrad azfa | Blzi3lee
The various explanations of “catvari padani’ in Nir.
xiii. 9 would lead one to think that in this case also the
real meaning had become obscure, though not to so great
an exteut as in the case of RV. x. 85.40-41. The Nirukta
(Parisgista) says (xill. 7) :

aaatta aif azafc aztt 2

AISI ACATTTRAATIA, |
ATHISIA alTaafaaiagata 7arETALT: |
Tea: Heat ata aAgal zaraarfeatfa ara: |
mat asifa arafa agai carreras ea: |
. ™ a “A

aq an ara Bae ala agai zaraarfraas |

TIT quay aaaratsl SAAT: |
aarfe arent wafa—ara aa ae aqat sraaq—
cag sag atin, wy qian) art ofaeat ara ar
 PREFACE Vil

Ceat | arear
afeaay arat
araaea | oat fete arfral
al gefa en emafsal | asada) adtar annreafisaa
Ti Maras: | aA aa Swat ara agfta-ara
2aTai at a agearorfata|
In the Mahabhasya Patafjali naturally explains
“catvarl padani” as “catvarl padajatani—namakhyate
copasarga-nipatas ca”. According to Bhartrhari the
refereuce is to the para, pasyauti, madhyama and
vaikhari varieties of speech. This is a fairly late idéa
aud uiust be rejected on that account.
Agaiu there is the stanza in the Roveda (iv. 58. 3) :

aca (ft -TST AAT ACT IAT

BMG AA SATA Bez |
faut agl gaat treaifa
net @at weat at fads |
‘Pour horus, three feet are his, two heads, seven hands
are his. Bound threefold, the bull roars. The great God
has entered the mortals.’

here is hardly any doubt that Soma is the great God

Inentioned here. In the first place the hymn is addressed
to Soma. Secondly, Soma is often spoken of as entering
mortals; e.g., “amartyo martya avivesa” viii. 48. 12.
Thirdly, “the sound made by the trickling Soma is often
alluded to, generally in hyperbolical language, with verbs
Iucaning to roar or below or even thunder; he is thus
commonly called bull among waters, which figure as
cows” (Macdonell). |
Vill PREFACE

But the sense appears to have been lost sight of long

ago. According to the Nirukta Parisista the great God
is the sacrifice. According to the Mahabhasya He is
Sabda. According to others He is the Sun.
Let us now take an iustance from another sphere. Most
people are not aware that the bishop’s mitre is a modified
representation of a fish’s head, the fish being a symbol
used by early Christians, for “that which is occult hath
the fish for a symbol; for the fish is hidden in darkness
and silence: he knows the secret places of the carth and
the springs of the hollow sea.”
In exactly the same way the technical ternrs of Sans-
krit grammar became meaningless after a time. Even
such pre-eminently self-explanatory terms as Parasmai-
pada and Atmanepada are dubbed “niranvaya sainjna” by
a commentator of the Katantra of all people (p. 126),of the
Katantra which more than any other syste, preserved
the ancient self-explanatory terms. And commentators
rack their brains in vain for a rational explanation of the
gigantic “Sarvanamasthana” which, by a curious irony
of fate, has been dislodged from the rightful position in
the Katantra system and has found hospitable refuge in
the system of Panini (p. 38). Since in Panini the suffixes
-satr, -kvasu, -iyasun etc. are without n , the naine
Sarvanamasthaua (lit. ‘in which the entire naine is
preserved’) loses all significance in this system, but since
the suffixes are read with nin the Katantra, the name is
particularly suitable for that system, where, however,
it has been replaced by the mono-syllabic “dhut”. Most
of the commentators of Panini have failed to find any
 PREFACE 1X

meaning in the term for the simple reason that they are
without any sense of historical perspective and do not
take into account the fact that names perfectly significant
in one system may be borrowed and become meaningless
in another. The English word ‘accent’is a case in
point. Derived from ad ‘to’ and cantus ‘singing, tone,
melody’ froin cantum past participle of canere “to slug,”
it is a literal translation of Greek prosoidia ‘a souge sung to
an instrumental or vocal accompaniment’ from pros ‘added
to’ and ozde ‘song’, and is therefore applicable to musical
or pitch accent only. But though in Latin and English
the stress accent is predominant, still the word ‘accent’
is
used. At the present moiment many Indian scholars
are
exercised over the problem of finding out a suitable term
for ‘stress accent’ in the Indian languages. With the
example of the English word ‘accent’ before us Wwe
can
easily use “udatta” for “accented” and “nihata” for
‘“unaccented”,.
To take another instance, the word ‘Kaster’ is
derived
from Anglo-Saxon Eostre, Fastre, a pagan godde
ss of light
aud spring whose festival came off on the sprin
g Equinox.
In English the pagan word is used for the
Christian
festival conlimemorating the resurrection of
Christ.
In this book an attempt has been made to trace the
technical terms generally used in the different systems
of
Sanskrit grammar back to their sources and to give
tational explanations of them. Even such apparently
arbitrary technical terms as “lat”, “bha”, ete. have
been
found on examination to admit of acceptable ex-
planations. In Bengal we have been accustomed to
regard the technical terms of the Mugdhabodha as most
po PREFACE

arbitrary, but a careful study of these terms reveals the

fact that they are in most cases borrowed from or
modelled upou those found in the Jainendra Vyakarasa
and ate susceptible of ratioual explanation.
In Sauskrit “ac” aud “hal” have stood for vowels and
consonants respectively at least frum the time of lV’aunini
and they are formed according to the “principal of the
forceps” (sandamSa-nyaya), by taking two letters from the
Sivasutras according to Panini’s rule “adir antyena saheta”’
(i. 1.71), yet the commentators of the Prayogaratna-
mala, following the noble example of the Unadi-sitras
aud uuder the impression that auy possible (or impossible)
derivation may be suggested for naime (“samjna-sabda
yathi-kathaficid vyutpadyah”) explain them thus :
aa may fra) carrntfarat | cio 1 ee aal | cafra|
mauaid waseaitea fer) merarhaRt 717< 1
Owing to the difficulties created by the communal
riots which caused widespread dislocation, this book had
to be printed in a press‘which, though one of the finest
printing presses in Calcutta, was not properly equipped
for the printing of techuical works of this type aud so
the printing leaves much to be desired. Following
the principle “madhvabhave gudam dadyat” ¢ had to
be used for the long 1, n for the guttural nasal,
uretre J for the sonant / and mere h for Visarjaniva;
10.060. diacritical marks could be used with most of
the capital letters and in Greek and Jatin words
e and ohad to be used for both the long and short
varieties. It was, moreover, a capital mistake to have
the book set up in linotype which did uot afford inuch
 PREFACE x1

facility for correction of proofs and this was aggravated

by the work having to be entrusted to compositors not
familiar with this type of work. It is quite possible that
owiug to these and other causes a uumber of errors have
remained uncorrected, specially in the numbering of the
rules of grammars. Only those which have been detected
at a cursory glance have been mentioned in the Errata.
A ‘full list of errors will be appended to part IJ which
will come out next year.

KSHITISH CHANDRA CHATTERI

CALCULTTA UNIVERSITY
September, 1948.
 CONTENTS

Technical terms and Technique of Sanskrit

Grammar -—brevity the soul of grammartvyiews of
different commentators on brevity—brevity of
expression, brevity of sense, brevity of process. l- “]
Names of Moods and Tenses—nothing is arbi-
trary in language—names of moods and tenses in
AB. and KR.—Naigaimi, the older name of the sub-
junctive—meaning of Paftcami and Saptami in K.—
peculiar names in Mu. and PR.—the names of the
ten Mahavidyas stand for the ten JTakaras—the
order of the verbal endings in Sanskrit and Greek—
“vipratisedhe parain karyam.” _ 8-18
Guna, Vrddhi and Samprasarana—views of
Apisali—earlier senses—found first in the Nirukta—
“ep” in J. “nu” in Mu. aud Govinda in HN.—
definitions in different systems—Vrddhi—not found
in the Pratisakhyas, nor in Nir—“aip” in J., “vri”
in Mu. and Vrsnindra in HN.—de finitions—interest-
ing stanzas —Samprasarana— earlier senses—
Prasfrana—“nu" in ST., “ji? in’ Mu. “Sankarsana”
in HN.—Prasairana again—trules. = 19-33
Vrddha—Yainanva in PS., “dn” in J., Adi-
vrsuindra in HN.—rules—interesting stanzas. ... 33-35
Bhitakarana—syllabic augment and temporal
augment—commentators of K. groping in the dark. 36-37
Sarvanamasthana, Pada and Bha—Sarva-
namasthina—views of Jinendrabuddhi, Haradatta,
Bhattoji—Narayana and Srstidhara give correct
XV1 CONTENTS

explanations—brevity once more—Satr and anti—

‘“sancapadi’ in CA. “sat in Pa “ghut’ i K., He.
and PR., “dha” in J.,“ghi’ in Mu., “catur” in‘Sar.—
different rules—Pada—earlier seuses—Padapatha
“da”'in RT., ST. aud Mu.,“Visnupada” in HN.—
“parvan” in VP.—Bha—origin—lucus a non lucendo
—‘pi”’ in Mu., Bhagavat in HN., “pi”, “bhi” and
“ohi” of Mu.—rules. .., _ -_ 3/51

Sarvadhatuka and Ardhadhatuka—fem. in

Apisali—derivation—‘ea” in J., “sit” in HC., “catnr”
in Sain., “ra” in Mu., and Krsnadhatuka in HN.—
rules. eae
|
a
7 cylop

Ti— probable origin—saimsira in HN.—rules. -

oe uiO23

Ghu — probable origin—‘“‘da” in K. and Mu., “bhu”

in J., “D&amodara” in HN.—rules—interesting
stauzas. ae _ 98-60
Gha —“Jha” in J., “na” in Sak.—rules. ... 60-61
Sraddha, Agni and Nadi—‘“‘Krsna” in HN.—
“sraddha” in K. and PR., “Radha” in HN.—‘“agni’
in K. and PR., “evhi” in Panini, Sak. and Su.—‘nadi"
in Panini, K. and PR., “gopi” in HN., “mu” in J. and
“di? in Mu.—rules—interesting stanzas... 62-67
Pratipadika—commentators unable to explain—.
eit —— - " pm
Srstidhara’s views—“linga” in K.—“naman” in Nir.
BD. and the Pratisakhvas—“namakarana” denotes
krt in Nir. and Samjna in PR.--namavibhalkti in
Nir.—namadhatu—nati—four parts of speech—phis
in F.S.—rules. ies _ 67-75
Sarvanaman—early occurrences—Mbh. —W hit-
ney—another sense—“‘sni" in J., “sri” in Mu.,“Krsna-
 CONTENTS XV1l1

naman’ in HN., “sarvadi” in C., Sak. and Hc.—rules

—sarvanaman vs. Gk. antonumia and Lat. pro-
nomen. ron _ _— 75-79
Akhyata—various definitions of the verbs in
Eastern and Western grammars—two senses—a
passage from the Kadambari. ... _ 79-83
Dhatu— original sense—early occurrences in GB.
and BD.—‘“‘dhu” in J. and Mu.—rules—uinteresting
stanzas. see ve se — 83-87
Abhyasa and Abhyasta—in KSS. and Nir.—
abhyasta in Nir.—‘‘ca” and “tha” in J., “Nara” and
“Narayana” in HN.—rules. _ = 88-90

Pratyaya—earlier sense—in RP., VP., Nir., CA.

~-earlier naines, antakarana and upabandha—prakrti
—interesting stanzas—the suffix -bahuc—rules. 91-99
Aprkta—iu the Pratisakhyas and Panini. 99-160
Krt—krtya—technical terms from the root kr—
early occurrences—rules. se _ 101-103
Krtya—explanation in the Prakriyasarvasva—
“vyya” in J. “lya” in Mu., “ghyap” in Sak., “Visnu-
krtya” in HN.—rules—interesting stanzas. ... 103-105
Nistha and Sat—originally nihstha—earlier
sense “ta”— in J., “kta” in Sak., “ktadi? in Sar.,
“Visnunistha” in HN.,—sat, Acyfitabha in HN.—
Adhoksajabha in HN. for the perfect participles. 105-108
Taddhita—significance—“‘ekaparvan” and ‘‘ane-
kaparvan” taddhita—syat in Sak. for fem. suffixes. 108-112
Tadraja—‘‘fiyadi” in C., “dri” in J., ‘fvri” in
Sak,—rules. a sa bs 112-113
XV1l1 CONTENTS

Gotra and Yuvan _ — 113-116

Upagraha—“parasimaipada” and “atmanepada” in
hag , ee _,
Panini—earlier “parasmaibhasa and “atmanebhasa ;
— da” and “ma” in J., “pa” and “ima” in Mu.—
3?

often shortened into “parasmai’ aud “atmane” in k.,,

Samm., etc.—“‘parapada” and “atmapada” in HN.—
explained by Susena, Sripati, Taraporevala—rules. 116-127
Nivrttisthana and Samkrama—in Nir. for
“ounavrddhinivrttisthina” — “atanu” in PR.,—
samkrama in Mbh., knit in Painini—Prthu, Kapila,
Nirguua and Kamsariin HN. ... nis 127-134
Vikarana—interesting stanza. _ 134-139
Gender, Number, Person—vyrsan for masce.-and
yosa for fem.—gender in AB., SB., Nir. and the
Pratisakhyas — Max-Miiller’s view refuted — nap,
kliba—linga in the sense of pratipadika—interesting
stanzas—number — early occurrences — in Gk. and
Lat.—samnkhya—kva, dva and vvain Mu.—rules—
purusa, person—rules. -_ ve 139-156
It and Anubandha—it—explanations of different
anubandhas. _ _ 156-168
Upadha, Udaya and Anusanga—earlier sense—
later use—un. in J. and Mu., Uddhava in HN.—
udaya—anusanga—iivrtti, anusanga, anuvrtti and
adhyahara. _ sk — 168-173
Samhita and Sandhi—Sandhana—sandhyak-
sara—definitions—.interesting stanza. _ 173-182
Different Varieties of Sandhi—anvaksara (anu-
lomna and pratiloma) — anvaksara-sandhi-vaktra—
Sauddhiksara — Praglista — Ksaipra — Bhugna—
Abhinihita — Vivrtti — Dirgha Vivrtti—Dvisandhi
 CONTEN'S x1X

Vivrtti—Antahpada Vivrtti — Padavrtti—Uderdaha

—Udagraha Padavrtti — Udgrahavat— Upadruta
Pracya Padavrtti — Paiicala Padavrtti — Pragrhita-
pada Samhita — Plutopadha Vivritti — Auundsiko-
padha Vivrtti — Anpada Padavitti — Vivrttyabhi-
praya—Samavasa Sandhi (see also p. 314).— Astha-
pita (avasangama and vaSsSangaima), Paripanna.
Auntahpata, Niyata, Repha, Akama, Prasrta, Usma-
sandhi (vyapanna and vikranta), Upacarita, Andanu-
pirvya Samhita—Sparsa-repha-sandhi—Nati. 182-205
Pragrhya and Pragraha—rules—iuteresting
stanzas. ... Te ce ina 205-209
Aksara—varia —sainanaksara and sandhyaksara
“sim” in VP. —rules—interestinug stauzas. ... 209-219
Svara and Vyanjana ea ea 219-295
Hrasva, Dirgha and Pluta—‘“pra’, “di? and “pa”
39

in J.; “sva”, “rgha” and “plu” in Mu.; Vaimana,

Trivikrama and Mahapurusa in HN. _ 2.95 -229
Guru and Laghu—“ghu” and. “ru” in Mu.—
“ohu? in Mbh. means “uttarapada”—iuteresting
stauza. 6 win ve lise 229-233
Namin and Bhavin—nantr — nati — vinaima
bhavin—Isvara in HN. ie _ 933.35
Aghosa and Ghosavat—ghosa—dlivani—uida
aghosa—ghosavat—svasa aud nada—“dhut” of K.—
whether ‘svara’s are included in ‘yhosavat’ or not
—rules. = 5 7 wets 235-245
Usman—Sit in K. aud He., adi in Sam., Hari-
gotra in HN.—sit of PS.—aniismavat, sosmavat,
aghosavat — rules — alpaprana and mahaprana—
interesting stanzas. ... _ , 945-249
ee CONTENTS

Anunasika—ranga—niasikya—uttama. 249-254
Ayogavaha—yogavalha in the Katantra School. 254-258
Anusvara—“nu,” in Mu. 258-263
Visarjaniya — visrsta — “visarea” much later—
“v1” in Mu. vase " i 263-266
Jihvamiliya gas sin as 266-267
Upadhmaniya ... _ se 268-270
Antastha—earlier “antalstha’—‘antastha” much
later — explanation according to Indian and foreign
scholars. _ ox a 270-274
Sparsa - rt iia 274-276
Varga—“kakara-varga” in the early Prati-
sakhyas, ‘‘kavarga” in later ones and k., “ku” in
Panini—origin of the anubandha u. oe 276-279
Varna and Savarna—“arna’ of PR.—“sasthana”™
an. ©, “oyna” ty J, on _ 279-285
Kara—rules—interesting stanzas... 286-290
Repha— origin—iuterestiug stanzas. ... 290-294
Samkhya and Sas—‘“‘du" in Mbh. we 294-296
Avasana—virdia—mahavirania. eae 296-300
Karyin, Karya and Nimitta—rules of Panini and
K. in verse form—na va and v@ in K.—‘asraddha
bho-
ji Brahimanaly’. a +“, 300-316
Adega—agama—vikira (see op. 317)—"Virinci”
and “Visnu” in HN., Upajana earlier name
of Adesa
—case-endings to denote aigama and Adega. 317-321
Sthanin—interesting stanzas. —— 391-325
Upajana, Upapada and Upabandha—“‘vik” for
Upapada in J. _ sibs x 325-330
 CONTENTS XX1

Lopa, Luk and Lup or ae 330-338

Akrtigana—interesting stanzas. -_ 338-34 1
Utsarga and Apavada—interesting stanzas. 341-344
Antaranga and Bahiranga — _ jnteresting
stanzas. = - - 344-346
Samasa pH ia _ 347-366
Karaka in aid _ 367-384
Vibhakti src ee a 385-398
Upasarga a si “ia 398-432
Gati _ -_ i 432-434
Karmapravacanlya ssi 5 ae 434-444
Nipata = — _ 445-449
Avyaya vis - - 449-452
 ABBREVIATIONS

AA—Aitareya Aranyaka. AB.—Aitareya Brabmaua.

A Pr.—-Atharva-Veda-Pratisakhya (ed.Suryakanta). Apr.—
Atharva-Veda-Pratisakhya (ed. Visvabandhu). Arm.—
Armenian. ASS. — AéSvaldyana Srauta Siitra. AV. —
Atharva-Veda. BD.—Brhad-Devata of Saunaka. C.—Can-
dra or Candra Vyakarana, CA.—Caturadhyayika(Atharva-
Veda-Pratigikhya ed. Whitney). Gk.—Greek. Gr. 5u.—
Grhya Siitra. He.—Hemacandra or Haimasabdanusasana.
HN.—Harinamamrta Vyakarana of Jiva Gosvamin.
IE. — Indo-European. J.—Jainendra Vyakarana of
Devanandin (with Sabdarthacandrika). JUB.—Jaiminiya
Upanisad Brahmana. K.—KA&tantra Vyakarana of
Sarvavarman (ed. Liebich). KB.—Kausitaki-Brahmana.
KP. — Katantra-Parisista. Mbh. — Vyakarana-Maha-
bhasya of Patafijali. Mu.—Mugdhabodha Vyakarana of
Vopadeva (edited with the commentaries of Rama Tarka-
vagisa and Durgadasa by Siva NaraAyana Siromani).
MW.—Monier Williams or his Dictionary. Nur.—
Nirukta of Vaska (ed. Sivadatta Sarma). P.—P&Anini.
PR.—Prayoga-Ratnamala Vyakarana of Purusottama
with the commentaties Prabha-prakasika and Gfdha-
prakasika. PS.—Phit-Siitra of S&ntanavacirya (ed.
Keilhorn). RP.—Rk-Pratisakhya (Text and translation
ed. Mangala Deva Sastrin'. RT.—Rktantra Vydkarana
(A Pratigakhya of the Samaveda ed. Suryakanta). RV.—
Rg Veda. Sam.—Samksiptasara Vyakarana of Kramadié-
vara with the Vrtti revised by Jumaranandin and the
commentary of Goyicandra—(ed. Syama Carana Kavl-
 ABBREVIATIONS XX1iV

ratna). Sak—Abhinava Sakatayana or Sakatéyana Vya-

karana (Lazarus & Co). Sar.—Sarasvata Vyakarana
(Nirnaya Sagara Press and MS. of the original Siitra-
patha in the Bhandarkar Institute, Poona). SB.—Satapatha
Brahimana (Venkatesvara Press). SkK.—Siddhianta-
Kaumudi. Skt.—Sanskrit. ST.—Sama-tantra. SV.—
Sima-Veda. TA.—Taittiriya Aranyaka. TB.—Taittiriva
Brahimana. T'P.—Taittiriya Pratisakhya. U.—Upanisad.
VP.—Vajasaney1-Pratisakhya. VS.—Vajasaneyi-Samhita,
YAV.—Younger Avesta.

a ci—aaqazonanera (Ganda denay ) | wa

BRAG AUMFEIN | we I aRofaaeay |) |-RI-- Ha.
aang | at—alfieiaaer | a—aegeqraron |
H—HAeReTRTAA | A ot—ahata-mfamewy | fae
face) oo
w—ofefearscna | = 9—satnTa@RTeT.
aaTRUNA | AAAs aR | aT oT— arstafir-
qfamera | wwe | fh q—fagay
(Hemaray) | a (a )—afaaaroneere;) ar
amaaacna | fa at —fasrntas | ay—aa1-
TFTTNA | WT mT —gafaataaieimey | #—z2fr-
AATTASTIATTA | een araUeTaMaAT
Enelish translations within double Inverted comm
as
of AA., AB., KB. and TS. are from Keith, of RP. from
Sastri, of TP. and CA. from Whitney, of RT. and APr.
froin Siiryakanta, of the Naisadha from Handiqui.
 TECHNICAL TERMS AND TECHNIQUE OF
SANSKRIT GRAMMAR.

The samjiias or technical terms of Panini are divided by

Patanjali* into two classes: krtrima or artificial and akrtrima
or natural. By krtrima he means terms not current in
speech, but coined to meet the peculiar demands of particular
sciences. By akrtrima he means terms that are current in the
language and that are, as a consequence, self-explanatory.
The krtrima terms again fall into two groups: (i) those that
appear to be entirely arbitrary like ti, ghu, bha etc., and
(ii) those that are abbreviations of longer words with or
without certain additions and alterations. And we may add
a third category, though Patafijali would hardly include it
under krtrima—those that are taken as the types of certain
processes, e.g., krt, bahuvrihi,. etc. Now what is self-
explanatory in one stage of the language or to the followers
of one system of grammar becomes obscure in another stage
of the language or to the followers of a different system of
grammar. Thus sarvanamasthana must have been clear as
day to the predecessors of Panini but his successors generally
failed to make any sense out of it.
There is another point which must not be lost sight of in
dealing with Sanskrit grammar. In a highly inflected language
like Sanskrit, grammar is an absolutely essential branch of

* sy fe saracad 2 Ta MR natageran: werefaen: faaa—

agg Zaeq sia, aga: alaafeguasrenta: |
Mahfabhisya, ed, Kielhorn, Vol. I, p. 223 ll. 3-4.
 2 TECHNICAL TERMS OF SANSKRIT GRAMMAR
a~eeewae%e
a
@’>We >

*- =—
-

% aa ae
iebet
Ya =
SS
Sr
ete
Le

a
— study. And though in the early days people were gifted with
ae
ied
e%

prodigious memories, the demands of sacrifices and other

things made them anxious to finish off the study of grammar
proper in as short a period as possible. This gave rise to
the principle of brevity* which was followed more and more
stringently by successive generations of grammarians. It is
not without reason that ASvaghosa describes grammarians as
“aksaracintaka.”’
Under ‘‘matropajiiopakramac-chaye napumsake” vi. 2. 14
the Kasika gives as an example: Apisalyupajiiam guru-
laghavam.t From this it is clear that the principle of brevity
in the domain of grammar was first fully enunciated by
Apisali. It may be mentioned in this connexion that the name
of Apisali is not found in the Nirukta, nor in any of the
Pratisakhyas, nor is Apigali mentioned as the seer of any
of the mantras of any of the Vedas. Another significant fact
is that a fairly large number of the rules of Apigali have been
preserved, though hardly a single one of any other predecessor
bie
se
\-ast
>t=~*23
sAS
«Ew
=

of Panini has come down to us. It would therefore

aes
al
SI aes appear
that the craze for brevity began shortly before the days
;

of
fy
oA

Panini. But this brevity, it must be clearly understood,

>
Tar is
concerned generallywith the forms of rules only and has
nothing to do with grammatical processes. In his
enthu-
ane
aed
=.*2~

siasm for laghava, Apisali regards the root as as

See
tied
ee
ee
st and per-
v
pte
Se

c ey the paribhiasa
° =.=
: SE ICICICE Gl qalqae Ba aqraca:
|
fe

This tendency persists even to the present

=
day and “cold war” is
regarded as an improvement on “war of nerves”
SM because it saves four
—*
34h letters.
1 Under iv. 3. 116 the Kasika says:
Hah
oiaeT
) Ss RAAT TART |
eee
Se
tIn Classical Sanskrit
the root as is conjugated only in the
pre-

Lege
a2paths
a

ae
a:ae
a
Se
 TECHNICAL TERMS OF SANSKRIT GRAMMAR 3

petrates similar other absurdities. But the principle laid

down by him was very sound. Panini appears to steer a
middle course between the two opposite extremes. He does
not use monosyllabic technical terms like jit, mud -etc., as
found in VP., nor does he postulate a root s instead of as;
but he does not disdain to use short technical terms like ti,
ghu, gha, bha,
lat, lot, lan etc., though retaining the older
and bigger terms like pratipadika, sarvandmasthana, karma-
pravacaniya and upasarjana. Of the successors of P&anini,
Sarvavarman follows an older school and generally avoids
meaningless technical terms, though ghut and dhut and Sit
and Paficami and Saptami have made their way into his
work also. Devanandin, the author of the Jainendra Vya-
karana, improved the system initiated probably by Apisali and
elaborated in VP., RT. and ST. But the greatest name in
this connexion is that of Vopadeva, the author of the
Mugdhabodha, who followed the idea to its logical extreme
and substituted monosyllables for all technical terms of more

sent, imperfect, imperative and optative active; of these 36 forms only

16 show the ‘a’, the remaining 20 are without ‘a’. If we leave out
the imperfect which shows ‘4’ in each number and person then of the
27 forms only 7 show the ‘a’. Api§ali, therefore, thought that it would
be more cgnducive to brevity, if we regard the root as s and prescri
be
the augment ‘a’ or ‘a’ for the special cases. In exactly the same way
Panini regards the possessive secondary suffix as -mat(u) or -vat(n)
instead of -mant (u) or -vant(u), the present participial active suffix as
(S)at(r) instead of (S)ant(r), the perfect participial active suffix as
(k) vas(u) instead of (k)vans(u) and so on, because the ‘n’ appears
only in the strong cases and not in the middle and weak ones which
are far more numerous. The K§&tantra and its followers rightly regard
the suffixes as -mant(u), -vant(u), -(S)ant(rn) and -vans(u).
4 TECHNICAL TERMS OF SANSKRIT GRAMMAR

than one syllable. A discordant note was sounded later on

by the author of the Harinamamrta who, intent on using the
names of Hari and His associates as technical terms,
casts to the winds all ideas of verbal brevity and says:
Maiwaaara Tatas zta wisfaneaed |
efrataatere at agen face: ||
i.e., the followers of the Panini school regard the mere sav-
ing of a mora* as tantamount to the joy arising from the
birth of a son. We, however, repudiate this view, since we
(by our prolixity) obtain words designating Hari.
Other sectarian schools also hold the same view.
The general principle on which the abbreviated technical
terms are based is enunciated by Goyicandra in his commen-
tary on Samksiptasara il. 583 :
qzigataty Ge seq, zat SURAT ATH
LAA ARIAT ATTTTTS -RTcal-ATN -M SaeaTEa aE ha
featta foe wea |
This is evidently based on the following passage of the
Mahabhasya (i. 1. 45):
_ _ere fe aeag aeiaga syn, wa a
qenzdt | ATI Mat arPaRtara—-ofaa frahy
[sfrat wen, war fog), ofr adeny [afi ag,
ee aig); Wwe wat saa ae, aera
|
Similarly, Ramananda says in his commentary on Mug-
dhabodha i. 5:
Ta aamrafaar: GAM: Watt eatarsaca aad-
aaracnag, wat waucea «fa Saatcgaraj FEO |
omen ne
* A reference to the dict um : wymarqrata yaiqad amet Saracar: |
 TECHNICAL TERMS OF SANSKRIT GRAMMAR 5

aamgaa a arse: metcafy ofeg: | azar orem

ahuag— aratarasearantafas: = ardt-nrferet-areh-
auretat reo" arfag |
Simuarly, Durgadasa says in his commentary on the
same rule:

dat segticga wa awe: | waaerafy daca:

aTanfieg wafa, sa wa arararara-
aantata cafe: |
Durga in his commentary on K§@tantra i. 1. 15 Says :
fafsfa gat dang fafa gfarta want aq
aga tft dface fafierd, atarrarcr ogee: ae"
Wd TH] teat wieqal fren xearqenaea: |
TRE Jax a Aa dgg
a Hear waqerat BR fH am
ATMA ILAAT: |
Again Durga says under iv. 1. 45 : feearstsarcferrety
wirara_} Under ii. 1. 80 Durga has a fling at Panini:
ve fazag qardet oy-fRreaeret ctrafanomucadtea
WTAE(!) ATLATAT: ATO |
Trilocana, the author of the Pafiji, hits the nail
right on
the head when he says ::
afazeraqa—a ad daraz Rrngyaar, earfy
TiasageNaTeTA FATT | ag qararat arf
ATH CUTE i fafafa ate: ea@frdan:
softer ata | Ua, arate ararfafe | aaaeh:—
fefedfe orad weft—ogeaq eines fr) aart
* Similarly in Varahamihira’s Brhaj-jataka we find Sa, ku, bu, gu,
$u, ca and sa standing respectively for Sanaiscara, kuja, budha, guru,
Sukra, candra and savitr in WH-F-F-T-4-BreM: aA
AMaaat rere
6 TECHNICAL TERMS OF SANSKRIT GRAMMAR

wang wad wrangdead auaiieq, wa: adeatfa

aan sfeagafa | a fe mrrze gag d wag d a Har
sqaqetanl azrat atfaste |
“There are two varieties of brevity—brevity in regard to
sound andl brevity in regard to sense. Of these it is the latter
that is desired by them inasmuch as their purpose is to serve
others. So Sarvavarman also expounds in that way. There
is nothing clever in the use of language in which vr or ksa is
the symbol for vrksa.’
The Panji says in the Samasa-section which it regards as
the work of Sarvavarman :
maga ante weqarad a faxtiaq, aisedtfer-
of
STAT MATA
S AAA |
Even Rama Tarkavagisa, the great commentator of the
Mugdhabodha, is forced to concede that sometimes brevity
of sense is to be preferred to mere brevity of sound. Thus
while explaining why Vopadeva uses ‘aghau’ instead of
‘pau’ in the rule “dhor 4-lopo’cyaghau” 121, Tarkavagiéa
Says:
‘agara dazg =
fH Oe Ne are, ative
BTTATTASTAAISHT (2) FATT: (9) |
Under ‘tad aSisyam samjha-pramanatvat’ i. 2. 53 Pataz-
jali says: kim ya eta krtrimas ti-ghu-bhadi-samjiids _ tat-
pramanyad aSisyam ? netyaha, samjfianam samjfia. From this
Goldstiicker concludes that “‘such terms as ti, ghu and bha
were known and settled before Panini’s grammar but that
nevertheless, they are defined by Panini because they are not
etymological terms”. (Ed. Panini Office, p. 127). We fail
to see how such a conclusion can be drawn from the simple
 TECHNICAL TERMS OF SANSKRIT GRAMMAR q

statement of the Bhasyakara. In the present state of our

knowledge it is very difficult to say whether ti, ghu, gha and
bha existed before Panini. They are not found in the
Nirukta nor in any of the Pratisakhyas, nor in the KaAtantra.
Bumell accepts Goldstiicker’s conclusion and points out
“similar symbols occur in the Katantra and still more in
Kaccayana’s grammar though (as might be expected) not
identical with Panini’s’” (p. 41). But we have not come
across a single technical term corresponding to ti, ghu or
bha in the pratisakhyas, though, of course, monosyllabic
technical terms occur profusely in VP. and RT.
It will be found in the following pages that some tech-
nical terms are defined in some systems, while others are used
without any definition. When certain terms are very well
known in the locality in which the grammarian flourishes or
in the system which he generally follows he does not find it
necessary to define them. Hence Indra says “siddhir
anuktanam riidheh”. The Katantra also says: “lokopacarad
grahana-siddhih” i. 1. 23 which is abbreviated by Hemacandra
into “Jokat” i. 1. 3. It is clear from this that Panini does
not Gann technical terms only when they have become
sainohished in his neighbourhood or in the system he follows,
otherwise he defines termis even though they may have been
used by his predecessors.
 NAMES OF MOODS AND TENSES.

When we were reading in the lower classes of a high

schoola«student
,‘ once asked the teacher of Sanskrit why Lat
meant the Present Tense. The learned teacher grew furious,
gave the boy an inkling into the abusive powers of the
elegant Bengali language and then said, Lat means the pre-
sent tense because the great Panini wanted it to mean the
present tense. If he had intended to use Drum Dhrash or
Chlorate of Potash for the present tense what was there to
prevent him from doing so? You ought to have greater
respect for Panini than to ask such impertinent ‘questions.
The teacher belonged to the usual class of Pandits who regard
grammar as a strictly normative science and who are followers
of the principle:
“Theirs not to reason why,
Theirs not to make reply’,
and at that time we were greatly impressed by the soundness
of his reasoning. But as -we grew older the argument gra-
dually lost all its force for us. I am now firmly convinced
that there is nothing arbitrary in language. To say that a
thing or phenomenon is arbitrary is merely to confess our
ignorance. If we could find out all the links in the chain of
association in the mind of a writer, we could explain what on
the face of it would appear to admit of no explanation. And
this is true of the technical terms of Sanskrit grammar also.
Ma: WL THAAGTATTA: |
In the Sanskrit grammars that have come down to us the
distinction between moods and tenses is conspicuous by its
 NAMES OF MOODS AND TENSES Q

absence. Yet one cannot help feeling that in several pre-

Paninian grammars there must have been a name for moods
also and the Lakaras divided into two classes according as
they expressed tense prominently or ‘mood prominently.
In the Katantra which has fortunately preserved many
ancient self-explanatory terms there is the Adhikara rule aT@
iii. 1.10 and the rules qrat aaah: arefaars: | TTA iii.
1.15-16. It is them of mre that led Panini to use the ten
technical terms@x_foz_ BE_FzZ BE BE_BE fae_ae_and
we (The Kasika says under aq iii. 4. 77: wz fTascaycy
fea: | AACA FATAL TA GEaT Hmed | ) For the Precative
Panini has no special name, he merely uses faerferfe
to denote it. (The Supadma usesw@re for forerfarte.)
It will be noticed that the indicatory letter = is used in
the case of the primary endings and S_ in the case of the
secondary endings. It is interesting to note in this connexion
that the indicatory letter x_ generally denotes that an Agama
will come in at the beginning and the indicatory letter =&
shows that the Adeéa will oust the final. waTearat sfc,
fe<ai. 1. 46, 53. When a child is born it is the head that
comes out first and z_is the first of miirdhanya letters. =
stands at the end of the varga and so it denotes secondary.
It is therefore fit and proper that x should be the Anu-
bandha in the case of the primary ‘endings, because they
come first and @_ the Anubandha in the case of the secon-
dary endings, because they come later and also lose their final.
First of all Panini takes the three primary tenses—pre-
sent, past and future and names them Se, fox. and GE res-
pectively according to the order of the vowel. The simple
vowels as < 3 being used up, SF comes next as denot-
 10 TECHNICAL TERMS OF SANSKRIT GRAMMAR

ing remote future. It comes after z_ because in Sr

there is the added element ey. Next Panini uses the diph-
thongs @ and eft for the two moods with the primary
endings, viz., the Subjunctive ( az_) and the Imperative
(aTz) - And since the Subjunctive, though fluctuating
between the primary and secondary endings, shows fa, f&
etc. intact in many cases, it is placed before the Imperative
in which the endings though primary are a bit modified. Of
the Lakaras with the secondary endings the tense owe
preceded the mood fee exactly as the tense eqz_ and ez.
preceded the moods @r and wiz_. After we (Imper-
fect) one expected ge (Aorist), but Panini now places the
moods and tenses alternately. Hence after the tense x=
we get the mood fers, then comes the tense Ce followed
by the mood @g_ which naturally comes last as being com-
posed of the elements of @e@_ and @z_.
It is a pity the Katantra follows Panini in the matter of
the Lakaras though it retains most of the self-explanatory
names. It uses qqurar for @z_ following Panini who says
qanta cai ii. 2.123. From the Varttikas of Katyayana we
learn that weardl was another name for the same Lakira-
aaah: cragedtssysrartiscaia ii, 3. 1. 11,
cgaenfaca fara wareradarrareiii.2.123.1. But the
still earlier name would appear to be aaa or Raat as in the
Aitareya Brahmana (xx. 1. 3, xxi. 1)* atterg , aaa and
wag represent the future, present and past tenses respec-
tively. In the Kausitaki Brahmana (xxii. 3) weckrvuse is
substituted for krtam. It is in the later Brahmanas and

* KEITH, Rigveda Brahmanas, p. 80.

 NAMES OF MOODS AND TENSES 11

Sankhayana Aranyaka* that the neuter. = is substituted for

the active @ and waa, Afaeaq and wea are used for
these three tenses. WVopadeva uses Waa, Wa and yey for
Panini’s qdara, wala and wfaerq ( waqyaaed fara:
eatat:, Mughdhabodha 933). Abhinava Sskatayana uses
aq forwaq and qa for afasre , thereby effecting the
saving of a syllable im each case.
For fsx (Perfect) the Katantra has qzleq correspond-
ing to Panini’s rule qapat fez_ ‘ii. 2.115. The name qcheat
is found in the Caturadhyayika—a Pratisakhya belonging
to the Atharvaveda (QVqTaea TUT iv. 84) and also
in the Varttikas of Katyayana. It also occurs in the Slokavar-
ttika: STATA Woe, i. 2.18. The Nyasa says under
i. 2.18: qona fafaaenfeos: afe: woRegRaa.
For Sz (Periphrastic Future) we find sqeqat im the
Katantra “after Panini’s rule staqeraa Bz iii. 3. 15. sataa
literally means, ‘belonging to to-morrow’ and occurs in the
Varttika gfi@qa sata afaearant wef iii. 3.15
For zx (Simple Future) the Katantra has apfersqreit
(a term which is used by Katyayana in his Varttikas) in
keeping with Panini’s rule Be sty asi. 3.13 imme-
diately preceded by wfavafa Weatza:.
Bz (Subjunctive), being confined to the Vedas, is ignored
in post-Paninian grammars, but it appears from the Prati-
Sakhyas belonging to the Atharvaveda that the older name
for the Subjunctive was await; from farra, the Vedas.
* LieBICH, p. 14.
+ Sait at aaa YA at Teaafer
wena ra vata TNT UT AT RRL
Sana TER AIA aaa | ATA aH) W HT RS
 12 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Haradatta points out that in some parts of India the

periphrasis qayq WRIT is used for @z_ as the latter means
in several Indian dialects something which can only be men-
tioned in works on anatomny.*
The original name for @tz_ (Imperative) was lost in
the Katantra school which uses qeayt for it, because wy
occupies the fifth place in the P&aninian scheme of moods
and tenses if the Subjunctive, which is confined to Vedic, is
excluded. It is just possible, however, that the ten Lakaras,
like the seven cases, were at one time known as syeqqq, faetrar,
agatar etc. The Prayogaratnamala, an ardent follower of
the Katantra, preferred Panini’s apparently arbitrary BIT
to Sarvavarman’s apparently arbitrary qeary and retained
wiz. The Atharva Pratisikhyas use the word qu} for
the Imperative and this reminds one of Panini’s rule sorta.
AUIMARTSA Hearsay iii. 3.163.
For we (Imperfect) the Katantra has Geta = (lit.
‘belonging to yesterday’). It is the counterpart of SATAY
for com (Periphrastic Future) and in keeping with
Panini’s rule alae we iil. 2.11. Though the word SIA
occurs in the Mahabhasya in connexion with Sma in the
passage way prafagtarahrearnier waresqaa: OTAT Aq erect
qeq. sacda: (ed. Kielhorn, Vol. I, p. 57, Ul. 4-5), still
@etat as the name of a tense occurs neither in the
Varttikas nor in the Bhasya.-
In the case of fee_ (Potential) as in that of sx
* ae-aeg saan qqqfea:, aaa sfaweR fates: |

The
CAS ei eirer
world Lakara itself is used in that sense in Sv. 2401.
+ Durga says under ii. 1. 20: geet exat
Ota Uzafeceyg.
 NAMES OF MOODS AND TENSES 13

the earlier name was lost and fee came to be called qaAt
in the Katantra system, as with the exclusion of the Vedic
Subjunctive, it occupied the seventh place in the system of
Panini. The Prayogaratnamala prefers fae. to Wat.
Since Panini’s rule for lin begins with vidhi (iii. 3. 161) one
is inclined to suppose that the earlier name was ant or
aaray.
For feerfarfa (Precative) of Panini the Katantra has
waran: . Padmanabha, though closely following Panini, uses
BIS_ for festfarfe in his Supadma.
Ge_ is WAAyaqal in the Katantra and Katyayana’s
Varttikas aqaeatay ii. 4.3.2, a ATT Ara, ili. 2.
102. 6 and AT aACaTA vi. 4.114. 3.
@e (Conditional) is termed faratfaafe in the Katantra
system after P&anini’s rule foo fated SS frarfaway
il. 3.159,
It is remarkable that Candra who studiously refrains from
using technical terms that are not self-explanatory retains the
Lakaras of Panini.
Devanandin, the author of the Jainendra Vyadkarana and
the friend, philosopher and guide of Vopadeva, also sees no
reason to reject these terms of the Panini system, though
Vopadeva does not. follow his guide in this particular in-
stance. He takes the consonants beginning with @&, omit-
ting the nasals and palatals for ease of utterance, and adds ¢
to each, both because yardl, steraatt, aqereat etc. end in §
and because the technical terms for the aq faafe in the
Mugdhabodha sft, gt, etc., end in §. Thus @=ua=
qa ATal= Present. ai=fee=aaait=Optative. M=slz=TaAl=
Imperative, fj=ae =Oedet = Imperfect. a= aS_= ayaa=
 14 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Aorist. st=ferx_ =atrear=Perfect. F=sz_=wataait= _Peri-

phrastic Future. @i=feee farfa=arati:= Precative. q=az =
afeeqeit= Future. eft-ae =Pratfaafat= Conditional.
It “was expected that the Harinamamrta would use the
names of the ten Avataras or incarnations of Visnu for the
ten Lakaras, but as the term Dasavatara has already been
used as the name of the first ten vowels (the DaSavataras of
the Harinamamrta corresponding to the Samaniksara of the
earlier systems), it is compelled to resort to other names.
For waz_ it uses sree which means fHaq and as Bz_ is
used of all times the name is very appropriate. For fisg
it uses fafsz (injunction and Brahma) and for lz_ Rratar
(creator, commander). These are very apt names because
fese_ is generally concerned with injunction and TE with
command. For the three past tenses @E_, qe_ and fore it
uses WayaT, Yast and eases. irae and aq literally
mean ‘the lord of creatures’ and contain Jat, the word for
past time.* epyzrsetm means Visnu and bears some resem-
blance to qzyeqt in that both are beyond the range of sense- -per-.
ception and both contain the letter ey, For fae arfarte
the Harinamamrta has Kamapila ‘the fulfiller of one’s desires’.
The two futures SE and @z_ are named Balakalki and
Kalki respsctively, — Kalki will make His appearance in
the distant future. Lastly the conditional aS) is termed
Ajita. Because the action has not been achieved, it is re-
garded as not conquered and thus the Lakdra is designated
Ajjita.

-*Jtis taken as the type and since waaq

is longer than wayq,
the aroist is denoted by wzam and the imperfect by Yaa,
 NAMES OF MOODS AND TENSES 15

I have heard that in a system of grammar belonging to

the Saktas the ten Lakaras are named after the ten Maha-
vidyas—Ka4li, Tara, Sodasi, Bhuvanesvari, Bhairavi, Chinna-
masta, Dhimavati, Vagala, Matangi and Kamala.
The grammarian Vyadi, who must have been anterior to
Panini since his name is mentioned in the Rk Pratisakhya,
probably used ten Ausas for the ten Lakdras. Thus we find
in the Kasika under ii. 4. 21 and vi. 2. 14: saTsa Te zal-
EeHAcoY} which is explained thus in the Nyasa : saTheteara
garnet frat amt a geacott Bar
qfiarfiaaa— ai ga’ aresferfr; «—»«-On._ this the
Tantrapradipa comments thus: qa Tey 7a Jaret
are dhanfeaars | >| azar qa’
qafafa, cancer| qaquaatcutfarafraarat
fedarg [ cian ] cfa fdaa, wer aa geEco-
aiaqaat AsaeI: | TTAAL aT Egencia Real Hoay-
aa tantraa4rs |
As regards the order of the Lakaras it is not possible to
come to any definite conclusion on the point from P&nini’s
Astidhyayi. In the Katantra iii, 1. 24 ff. we find eq
dtArat,
Saal, TAT, Tete, AAaat, core, zaeaat, sari:
wafasqedt, and fratfaafa—an order followed by most
later systems including Hemacandra, who inverts the order of
errstt; and geaeaedt. In K. ergf: is connected, according to
meqaforeata, with both gazaeft and afavqedt, He. wants
to place the two futures in succession, so it has no alternative
but to place wratt: after qzrar. HN. follows Hc. Mu
follows K. The Prayogaratnamala changes the order for

*JIn Kasik&é iv. 3. 15, however, we find: Apiéalam huskaranam.

 16 TECHNICAL TERMS OF SANSKRIT GRAMMAR

convenience of conjugation and we find here: OTATAT

Cc

fas, Giz, aed, zatadt, arei:, wfsersdt, fire.

faqfa, qtrat and agqaat. In the Samksiptasara also
the order is slightly different: @x a@le_ae_, fee ATSHt-
fae , faz, az, a i -e and 7 ae In the
Supadma the order is entirely different, the tanubandhas
coming first and the nanubandhas following them- Or
and @z_, wiz, Tet, GF, GFF, BS and Se,
fers and @Te_. }
It would appear from the example gyrft
SAT RCO found in the Kasika under ii. 4. 21 and vi. 2. 14
and) arforaiaaanress, SAT RCT, found under iy, 3. 115
that Panini was the first to do away with the Kaladhikara in
Sanskrit Grammar. It follows as a corollary that Panini
was the first to introduce the technical terms: lat, lit -“
It is interesting to note in this connexion that in German a
verb is known as Zeit-wort (@TeRreg: ), According to Aris-
totle also “the Verb is distinguished from the Noun as con-
noting time’. Sandys, i. 98. The Mahabhasya also Says :
aTeaen frat aoa faersaraer sasaed,
As regards the order of the endings Devanandin Stands
alone in beginning with the first person—rmip, vas, mas : sip,
thas, tha; tip, tas, jhi; tt, vahi, mahi; thas, atham, dhvam ;
ta, atam, jhan, ii. 4. 6. So it has to use the pratyahara min
for Panini’s tin.
This is entirely opposed to the Indian point of view.
Indian systems of grammar generally begin with the end-
ings for the third person, then pass on to the endings for
the second person and put in the endings for the
first person
last of. all, while European grammarians begin with
the first
 NAMES OF MOODS AND TENSES 17

person and then pass on to the second and third persons.

Thus in Sanskrit we have asti, stah, santi; asi, sthah, stha;
asmnu, svah, smah; but in Greek and Latin we find eini
and sum (asm), esst and es (asi), esti and est (asti), esmmen
and sums (smah), este and estis (stha), enti and surt (Santi).
This is due to a certain extent to the requirements of the
Paribhasa* ‘“‘vipratisedhe param karyam”’’ i. 4. 2 i.e. in case
of conflict the later is to prevail over the earlier, according
to which when there is any doubt in our minds as to the person
of the verb owing to the presence of nominatives of different
persons, we use the first person in preference to the second and
third, and the second in preference to the third. For this the
Katantra has the rule “yugapadvacane parah purusandm”
iii. 1. 4, but such a rule is unnecessary in the Panini system,
because of this arrangement.
In the matter of the mention of verbs also the same
difference between Indian and European grammars is
observed. In Sanskrit a verb is mentioned in the

*Poets have not been slow in making use of this Paribhasa of

Panini for their own purposes. According to them, Radha, torn
between her loyalty to her husband and her love for Krsna, said to
her friend, ‘My husband is my first lover, Krsna the second; what
am I to do?’ Her reply was, ‘Listen, dear friend, to the rule of
Panini: In all cases of conflict, the later prevails over the earlier’.

“fangiaua: wat efifr ta: aofa fa’ fy |

vy af uifufaa’ fanfaga. ox arte 1”
“aquiaaata ay tS aw Hoe far’ ager |
at ofa afufraa’ fanaadeod aria”
“farmafara: naa aeq a eft: fa Bag aa
UY afer oifefrrert fanfeee oe erty
O.P. 129—2
 18 TECHNICAL TERMS OF SANSKRIT GRAMMAR

third person singular (ikStipau dhatunirdese iii. 3. 108.

2), but in Greek and Latin a verb is mentioned in the first
person singular. Thus in Sanskrit we say bharati (to bear),
but in Greek we say phero and in Latin ferd (corresponding
to Sanskrr bharaGmi) in the same sense.
This would appear to throw a flood of light on the res-
pective mentalities of Europeans and Indians.
The former
always look after number one, while the latter generally shrink
from self-assertion and self-aggrandizement and firmly believe
in the doctrine that in seeking the good of others, we find
our own. Hence agqeztaea indicates aaa *
aerate weer warfare.
EAM TILT ATTY TTI: |
qathrararsaeras
aay
rated aafeareafiarferarg: |)"
ATT LULVS
“© daughter of Bhima, just as the root as is transformed
into the root bhu on reaching the aorist (adyatani) which
indicates the past, even so, the creatures of the oceans of the
world, on arriving at this city, attain absorption into Parvati’s
consort, Siva.”

eee
* After the ten Lakaras of Panini, a poet speaks of five lakaras-
saya fae gee gnei gtr |
TqMRIt Ht GAT: GUalGaTENa |
 GUNA VRDDHI AND SAMPRASARANA
Guna.

In Sanskrit grammar the vowels a, e and o are known

as Guna-vowels.* The guna of a is a, of 7 is e and of u is o.
Of the guna is ar and of / al. Of these mere a may be left
out of account, as has actually been done by many later
grammarians. Panini mentions it for ar and probably for
al and for certain grammatical operations peculiar to the
Panini system. Patafjali mentions the different meanings of
guna in the Mahabhasya on “prak kritéc chah” v. 1. 1:

WIMSs AI: | AKte akaqqaig ed2, az war

faagon «sgftfa ) afta gerverde:, ag ger gerard a
rasaa afer ma: aerfa a ada | AeA
ada, aT
qa at aaa waft @ ene—aonget aaa |
ARATATZ aad, ay AIT Wad stan: eysa a:
ararat Hof | wha dent ada az aa dene
Torafreyeaa | Away AMATI Worsted: ARTTTTT Tae |
Of these the first sense of ‘equal part’ or ‘strand’, the
second sense of ‘property’, the third sense of ‘secondary or
subordinate’ and the fifth sense of ‘refinement’ are all appli-
cable here.
Of the eminent Sanskrit Grammarians Panini alone regards
a as a Guna vowel. Though it is hardly logical to speak of
a
ernest
*Sabara appears to regard guna and vrddhi as arbitrary technical
terms. He says :amfe aaa HAMMAaaqUERMangaee al wa,
1 HPeareaeCMeT TCAATATAG ofa | Loieiee
20 TECHNICAL TERMS OF SANSKRIT GRAMMAR

a as the Guna of a, still Panini includes a in the list of Guna

vowels for the sake of brevity and uniformity. He probably
thought that the Gunas of vowels should also be vowels and
so he lays down thaf a is the guna of r, but in this particular
case a is to’ be followed by r* acc. to another rule (ur an
ra-parah i. 1. 51). Again in conjugation a is dropped
before a, e and o of endings. Here Panini has combined a, e
and o under the general name Guna (ato gune vi. 1. 97).
It is interesting to note in this connexion that acc. to
philologists also “‘both a and 4 represent the guna or normal
stage in the gradation of the a-vowels in many roots”
(Macdonell, Vedic Grammar, p. 16, f.n. 5).
The evidence of other cognate languages shows that e
and o were originally pronounced ai and au. The guna vowels
were thus looked upon as having a strand or some property
added to them or as having been refined (Samskrta) by the
addition of some attribute. In other words guna is used in
the sense of strengthening.
Apisali is said to have regarded guna and vrddhi as
Agama or augment. Thus in the Mahabhasya on “sam-ava-
pra-vi-bhyah sthah”’ i. 3.22 we find the examples ‘‘astim saka-
ram Atisthate. Agamau guna-vrddhi Atisthate. Vikdray guna-
vrddhi Atisthate’ under the-Varttika: anah sthah pratijnane.
In the Kasika we find “astim sakara-matram 4tisthate,
Agamau guna-vrddhi 4tisthate.” This is explained by the
Nyasa-kara thus:
anitaiante argaitiafetrena: sfasrenta, aurfr a

eafeuafa, qat aatifa @afae amafyanaisfafaaaa| BIRR?

AyAae |
 GUNA 21
az aifmatta ‘aa ufa xfa ang: | fe afe: ‘a af’
tf avefa| arrataragl enfasa eft) a cami
ange afasa; oa fe a ofasdia |
Similarly Haradatta says in his Padamafijari:
ada waaay TAT ‘a ufa’ wa srg:
qTaa:, setraral fafa aaag&serm faia:, aman
Tea carat aema;, enfeenfiofsar aaa, anay
mora
aft | aarace: alae gearal astmracafreny:|
WNT ATA SITET NAF |
Bhattoji says in his Sabdakaustubha : arfearfate Nay
ufa afa a cesta, freq wararaq) et: aedieeat-
acmqg. | «afta areifeearfefaea go aercrarmrat aff
Braga | aaa TOTS afer
Apigali maintains that guna and vrddhi (a and 4) are
agama in order to form the words, asti, astu, asit etc. from
his root s (which is as in the Paniniya dhatu-patha) and this
is referred to by the Nyasa-kara in explaining the two examples
cited here by the Vrtti-kara to illustrate the Varttika attached
to the rule i. 3. 22.
As the passage is of some importance we have quoted
all the different views above. We have a transcript of
Maitreya Raksita‘s Tantrapradipa but unfortunately this does
not contain anything under i. 3. 22.
What emerges from the above quotations would appear to
be this: Apiséli regarded guna and vrddhi as augments.
Presumably he regarded the augment as a. In the case of
guna the augment was added before the simple vowel, and in the
case of vrddhi before the guna vowel. Or he may have
22 TECHNICAL TERMS OF SANSKRIT GRAMMAR

regarded the augment as a in the case of guna and as a in the

case of vrddhi and framed some rule on the line of Panini’s
Ataéca vi. 1.90. In RP. xi. 10 ‘‘Gunagamad etana-bhavi
cetana’’* “suna” may have been regarded as an augment. Of
course we can take guna as standing for the guna vowel e
and agama in the ordinary sense of ‘advent, appearance.’
Guna may also have been used in the sense of ‘second-
ary quality.’ According to Indian grammarians the simple
vowels a i u are regarded as forming the normal grade and
the strengthened vowels are regarded as secondary ones. As
Rucaka says in his commentary on the Vyakti-viveka.: anya-
siddhy-artham upadiyamanatvam gunasya laksanam.
Guna first occurs
in the Atharva Veda in the sense of
‘fold, times.’ It is found in the Taittirlya Samhita in the
still earlier sense of ‘string’ or ‘thread’. In the Sankhayana
Brahmana xxvi. 4 it is used in the sense of ‘auxiliary act’ and
in the Srauta sitras it means a ‘secondary element, subordinate
or unessential part of any action.’
In its technical sense the word is met with first in the
Nirukta: ta afa qa@ata, faraa dana araqwar: |
aaaraitast featrane: | ferafaerae wath |
x. 21.
The word Seva means happiness. The suffix ‘ya’ has been
added to the root ‘sis.’ It replaces the final of the root and
causes guna optionally. Siva is derived from the same root
with the same suffix.
Sanskrit grammarians of most schools use the term guna

* HUTA THA TAAYay TAA Fdlaq yezq Vaz |

 GUNA 23
which, in the 19th century, became practically international.
Bearing in mind Panini’s rule aden gunah i. 1.2 Candra
uses aden for guna. Thus for Panini’s ad gunah vi 1.87
Candra has ad aden v. 1.82 and for Panini’s ato sune Candra
has ato’den v. 1.101. Jainendra uses ep for guna, taking
the guna vowel e and adding a euphonic p to it. Sakatayana
does not use any technical term corresponding to guna, but
the pratyahara en along with ar is pressed into service. Cf.
ikyenar i, 1.82 juspakyenar iv. 2.16. The Mugdhabodha
tries to improve on Jainendra by taking the consonant of the
second syllable and the vowel of the first and calling it zu.
The Harinamamrta takes the initial consonant of guna and
uses the name of Krsna beginning with g, ‘viz., Govinda.
P. AIS wo: i, 1. 2.
K. at Ta g BMeryat Ay: iv 4. 31.
J, AS 7 & L. L.
He. WNISTAA il. 3 2,
SK. age_yor: i. 1 81.
Sam. FH CSCS 77: i. 69.
Mu. qBstTas_ 1: 8.
Su. @ecatWo FH: i. 1. 21.
PR. WL-QAT
WOT SFAT
att ete: eaecaATT i 34.
TT MWAAUTTATATTTS | ageaacite: | woratfa
ToT: | STATAAAT |
HN. yaaa U, TeUeT Ai, MgAeT AT, TIT As
Tifrqetat wyoretway 1. 180.
Guna is sometimes used by poets also in this sense.
Thus there is the well-known stanza :
24 TECHNICAL TERMS OF SANSKRIT GRAMMAR

ara anaz aT Tat: HAT Tfaceaw |

qearaarears BITAPTAS TTT: ||
The celebrated scholar Vabhata, it is said, had a very
beautiful and accomplished daughter. Seeing her straying
from the path of virtue, the learned scholar remonstrated with
her. Remonstrances proving unavailing, he began to weep
bitterly. At this the daughter said, “Father Vabhata, do not
weep. Such is the course of action. Our virtues lead to vice,
exactly as in the case of the root ‘dus’, ‘guna’ turns it into’
‘dosa’ (fault, vice).”’

add a andrrartraHat Tarat faafs acta |

foarhrant seareracn facts caret atest |)
ATT LoIRg
amaisty qa TIOqTA efaanat: Way a |
araniaga tearFt:ATIAAT F II
Vrddhi.

Vrddhi literally means ‘growth, increase, extension’ and

is used of the lengthened grade vowels ai and au for their
perceptible increase of moras* or for their increase over the
normal vowels with 4 or guna vowels with a. Vrddhi is a
much older word than guna and is found from RV. downwards.
As a technical term, however, it does not occur in the
Nirukta, nor in RP. or TP. It occurs in VP. at the end
of each chapter in the form of vrddam vrddhih, but there it
is used in the sense of prosperity. This Pratisikhya prescribes
ar for ¢ and al for / in the siitras: dram rkaro’prktat,

* 7y aalfafa wa fafa gages) GeREe |

 VRDDHI 25

|karascalkaram iv. 60-61, but vrddhi in its technical sense is

found in caikaksara-vrddhav ,anihite v. 29.

taddhite It
‘s is not found in CA. but the adjective form vwrddha occurs
there in the sense of that which has undergone vrddhi.
Panini’s Astadhyayi begins with the word Vrddhi, Sarva-
varman’s as it has come down to us, ends with
Katantra,
Vrddhi.* Sarvavarman stands alone among Sanskrit gram-
marians in not admitting @ into the category of Vrddhi vowels.
He thought that since @ was a dirgha vowel the term Vrddhi
should be confined to ai, au and @r. In this Sarvavarman
appears to be following the predecessors of Panini. It is
interesting to note that modern philologists agree with these
ancient grammarians. Thus we find in Macdonell’s Vedic
Grammar for Students: The radical vowel, as a rule, takes
Vrddhi (a being lengthened) in the active [ of the root aoristJ.
The corresponding rules in K. are: [sqTqaTat aint afe-
aifratfafraza 1] fafa wet quran) saarrat-
avatar | ART aT TH BIR
RT. appears to use “dvivarna” for ‘“‘vrddhi” in the rule
‘““sandhyam dvivarnam’”” 95 (sandhyaksaram dvivarnam
Apadyate—-Vrtti). In the Apisali-Siksa the Sandhyaksaras are
described as ‘‘dvivarna’—“‘dvivarnam sandhyaksaram”. In
the Puspasitra and its commentary Vrddhi is used in the sense
of Pluta also, and the root wdh is used in the sense of ‘to
become prolated’.
The Katantra uses the term in the chapter on Akhyata and
defines it at the end of the chapter.

egferred Wee | WaHNt Vales | ST eiyiyy

 26 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Of the lesser grammarians

J. alone follows Panini. K.
and others regard a, 9, ar and al as Gunas.
Candra uses adaic for vrddhi taking the hint from: Panini’s
rule. vrddhir ddaic i. 1.1.
Sakatayana uses araic every time. Cf. araijadyacah iv.
2.132.
Jainendra uses aip taking the vrddhi vowel ai and add-
ing a euphonic p to it. In most systems of grammar, how-
ever, aip is the ending for the imperative first person singular
Atmanepada.
Hemacandra defines vrddhi in the third section of the third
chapter of his grammar, consequently he uses this technical
term only in the subsequent portions of the work.
Vopadeva takes the first syllable vr, adds the euphonic
i to it and forms vri. Was r pronounced as ;i in Vopadeva’s
time?
The Harinamamrta, as usual, takes the first syllable and
forms a name of Sri Krsna with it. Thus for vrddhi it has
Vrsnindra.
The other systems follow Panini and use vrddhi,
P. afexa i, Le ly SK. L 82,
K. am gata afe: iv. 4. 35.
J. amgarq i. 1. 16.
He. afgurtata iii. 3. 1.
Sam. afguizenzieaisea: i. 1.
Mu. aq NTS fa:
Su. afguewrcsisn ata i. 1. 21,
PR. ata ay ae t a aang gaz: i. 35,
aatarfa aga: | aeserfenr |
 SAMPRASARANA a7

HN. aguea ar, quer 0, sarer at, weTET ANZ,

Beet ATS aMiegaat afgeeea i. 191.
A poet uses the technical terms guna and vrddhi in the
following stanza quoted in the Kavya-Prakasa :
Shaaizaa: Haye Wa FATIMA |
fapraafa: ahag aa afatec a 2 1
On being asked about his impressions about the charac-
ter of the inhabitants of a country, a visitor said, ‘Some are
devoid of qualities and prosperity, even as the roots ‘didhi’
and ‘vevi’ and the union-vowel ‘i’ do not admit of guna and
vrddhi (acc. to didhi-vevitam P. i. 1. 6); some who are
strangers to good qualities and prosperity are like the suffix
‘kvip’ (of which nothing remains, acc. to “ver aprktasya’’ P.
L. 67).
Agvaghosa uses the word vrddhi in the following stanza of
his Saundarananda (xi. 9):

aaa a fe dat: SAI TE |

srazfsttarena ofsatsercfiras: 1)
‘That perturbation of his was conducive to the growth of
his well-being like the root edh read in the Akhyata section
by syllable-grinders i.e. grammarians.’
The root edh is read in the Dhatupatha in the sense of
‘vrddhi’, and acc. to Panini’s rule “‘etyedhatyiitsu” vi. 1. 89
the ‘e’ of ‘edh’ contracts with the final ‘a’ or ‘a’ of a preced-
ing preposition into the vrddhi vowel ‘ai’

Sam prasarana.

Samprasarana literally means spreading out fully, extension,

distraction. ‘The idea is that ya, va, ra and Ja contain 7, u ,
28 TECHNICAL TERMS OF SANSKRIT GRAMMAR

r and / in a condensed form: when they are fully spread

out and i, u, r and | make their appearance, the phenomenon is
known as samprasarana. Sam-pra-sr causative is used in the
Samhitéas of the Yajurveda in the sense of ‘stretching or
spreading out’ and in the Apastamba Srauta Siitra in the sense
of ‘drawing asunder’.
It will be noticed that two distinct views are held by
Sanskrit grammarians. Acc. to Panini and K., the change of
y, v, r and / into 7, u, r and / is Samprasarana. Acc. to others
the change of the semi-vowels along with the following vowels
into i, u, r and 7 constitutes Samprasarana.
It would appear from the derivation of “‘isti® in AB.
i. 2. 1 that the author was unacquainted with the pheno-
menon of Samprasdrana, since he derives “‘isti” in the sense
of yaga from the root is, and not from the root yaj. But mt
must be remembered that the authors of the Brahmanas were
intent on showing some connexion between the name of the
sacrifice and the result of the sacrifice* and consequently they
often deliberately resorted to fanciful etymologies. So jt
would not be safe to draw any conclusion from the etymologies
found scattered in the Brahmanas.
Prasarana is generally used in the V4rttikas in the sense of
‘samprasarana’. Vyaghrabhiti also uses prasdrana in his
anit-karikas: spferay aeda afe: searcofy (Kasika on
vii. 2.10) but that may be due to the exigencies of metre. It
is also found in CA. iv. 37 Com.

‘amd ed=>
semlad, afafeft: qadarq, afefefe:
= Ey

Pade ateemar-
fafeaq|
See Calcutta Oriental Journal, Vol. I, No. 3, p. 77.
 SAMPRASARANA 29

The Samatantra uses the technical term mu for sam-

prasarana, taking the n of samprasdrana, changing it into 7
because of the absence of r or with a view to making it look
more like an arbitrary term* and adding a euphonic u to round
it off.
The Katantra uses samprasdrana in the Akhyata-section
though in the Nama-section it studiously refrains from using
the technical term. Cf. caturo va-Sabdasyotvam i. 2. 41,
anaduhag ca ii. 2. 42, aghut-svaradau setkasyapi vanser va-
Sabdasyotvam ii. 2. 46, Sva-yuva-maghonam ca ii. 2. 47 ete.
In the fourth section of the third chapter samprasfrana is
used and it is defined in the last section of the fourth chapter,
but commentators are at a loss to explain the high-sounding
word. The last but two rule in the Akhyata-section is
‘‘samprasaranam yvrto’antasthanimittah” i.e., i n and r origi-
nating from semi-vowels are known as samprasarana.
Candra takes the hint from P&anini’s rule ig yanah sam-
prasiranam i. 1. 45 and uses igyanah for samprasdrana.
This is not only self-explanatory but also shorter. Thus for
Panini’s syafiah samprasaranam putra-patyos tatpuruse vi. 1.13
Candra has syafiah pradhanasya putrapatyoh svayor ig yanah
v. 1. 19. Sometimes other devices are also resorted to. Thus
for Panini’s vasoh samprasaranam vi. 4. 131 Candra has vasor

“CE qdeafaa 4 gar; AB.

Cf. Plato’s Cratylus 414c: You must remember that all languages
are in a process of disguise or transition ; and letters are taken out or
put in at pleasure, and twisted and twisted about in the lapse of
ages—sometimes for the sake of euphony.
 30 TECHNICAL TERMS OF SANSKRIT GRAMMAR

va ut v. 3. 128. Candra here took the hint from Panini’s rule

diva ut* vi. 1. 131.
Jainendra is fond of monosyllabic technical terms, but
in this case he thinks that sufficient brevity will be secured
by substituting ik for sac yan and that a new technical term
would involve prolixity. Thus we find litig yanah sfco’r-
yyajyam iv. 3. 17 corresponding to Panini’s lity abhydasa-
syobhayesam vi. 1.17.
Sakatayana follows Jainendra.
Hemacandra uses yvrt in yajadi-vas-vacah Sasvarantastha
yvrt iv. 1. 72 and explains it thus in his Brhad-vrtti on
yvrt sakrt iv. 1. 102: yvrd ity antastha-sthanikanam
ikarokara-rkaranam Sastre’smin vyavaharah.
Vopadeva uses ji with the y of yan and i of ik clearly
showing that y etc. are changed into i etc., only y is here
changed into j. It is just possible that in the days of Vopadeva
initial y was pronounced like j.7+
The Samksiptasara and Supadma also follow the lead of
Jainendra. Cf. vasader. usadih Sam. ij. 257, Vvacl-svapi-
yajadeh saco yana ik kiti Su. iii. 3.134,...... \jadir yajadeh
Sam. ii. 583, adadi-vac-svapor uc-supoh Sam. ii. 584,
vasoh
saco va ut Su. il. 3. 124 etc.
The Samksiptasara uses asamprasdrana only once in
the

*In the case of kvasu, the reduction takes place before weak
endings, hence Panini uses the word
‘Samprasdranam’. In the case of
div the reduction takes place before middle ending
s, hence Paninj
uses ut instead of ‘samprasaranam’.
P grelet 4 Gziet 4 aamnaaeq «|
a: ame ofa fan a aise: oa sil qa: aaaiteafedt fing |
 SAMPRASARANA 31

rule avad yajo’samprasaranam ca iil. 68 on which the com-

mentator Goyicandra remarks :—sTaqrrEarcusata arforetter-
faftatenanrerniafa adannsage: ware, aware
ATE BM Ravages fa wary |
The Prayogaratnamala uses samprasarana.
The Harinamamrta uses Sankarsana, the name of Balarama
beginning with sam and approximating in sound to sampra-
sdrana.
Yaska understood the phenomenon of samprasarana, but
does not use any technical term for it. Thus he says:
aq wa waugarataeranqaaa ac Boadtat enafafr
ofratfea ) ii. 2*
Of the Pratisakhyas CA. or its commentary} uses the word
prasarana in the sense of samprasarana, but here again the
metre may have something to do with the dropping of the
sam :
TET TACO TET Teed ACTS AAT |
qaarane: feat areata Tar |]
In APr. prasdrana- is used evidently in the sense of
‘spreading out’. Thus the absence of the change of d into
dh is spoken of as prasarana in the rule prasdranam dipsatity
ee
interes
een a A
«war at amaafen afeaa acufacmeaa—az aafa) emar-
Watagesee-nAATaAa aad: |--‘ga’ afaay qrat way’ WATE:
‘sama’ samigar S474 Baka, at al gat at, ‘ware’ acaaaT ad:
‘omaig? uigaar asf ‘ay’ urged fenadtai feamatt meri Mae wre
‘sfa’ vag sraral: ‘afenier afaartagienia | «ag aa—aa 2aga-apfaace-
ziaq xf) WAS WRT Raa | BMAIGe aay ‘wea’ ‘gfe! ‘seq’—
Wa! RAAT Valet ; WaMAaTe ya: ‘ger ‘geaq’‘geen’ vara: 1 ga!|
+ Acc. to some what was regarded as the com. by Whitney forms a
part of the text,
 32 TECHNICAL TERMS OF SANSKRIT GRAMMAR

adyacaturtha-pratisedhe 153c. The first syllable of dipsati

undergoes prasirana in that the first letter i.e. d is not changed
into the fourth letter of the varga i.e. dh. Dr. Suryakanta,
however, explains prasdrana as contraction and translates the
rule thus: prasirana takes place in dipsati, the fourth (bh)
is negatived (replaced) by the first (p). In 217 m., however,
prasarin is used in the sense of that which undergoes sampra-
sarana or reduction:

aghagien waqaganaagiea srerfcor |

Both samprasarana and prasarana are found in the
varttikas as also in the Mahabhasya. Thus under 4to’ny-
pasarge kah ili. 2. 3 there are the Varttikas: ka-vidhau sar-
vatra prasaranibhyo dah and ke hi samprasarana-prasangah
and at the end of the Bhasya on the same rule we find -
facet warcony|
In the Bhasya on bandhuni bahuvrihau vi. 1. 14 we find :
mataj-matrka-matrsu syafi prasaryo vibhasaya which appears
to be a part of a Sloka-varttika. Again in a éoka-
varttika under iv. 1.78 we find:
aa Taraa g araafaet which is explained thus by
Patafjali : aut ata arate: araga: tera svercot oaths |
Again under ul. 2.178.2 Patajiijali says :

aqt Me—afanssaicararcet ath THI | aT

afe area, A ama) Teese sereet
afacete|
The term samprasarana is found in the Unadi-sitra
also: syandeh samprasaranm dha ca i. 12, prathi-mradi-
bhrasjam samprasaranam sa-lopas ca i. 29 etc. In parder nit
samprasaranam al-lopas ca iii 80 samprasdrana appears to be
 VRDDHA 33

used in a slightly different sense, for in Sanskrit grammar

it is generally used when r followed by a vowel is changed
into r.

P: QT ANT: AFAATLNA i. L. 45.

K. amar z-aatseaurfafaa: iv. 4. 33.
He. zafearreren rare carrey Test
SPaar.
START: | Brhadvrtti on zqq ahq iv. 1. 102.
Mu. zara fa: 536.
PR. way wacenafieed arraicng i. 36.
HN. aqt-abat-qauronfa-S-B-a: AEAtee:
«|-
i. 602.

VRDDHA

The technical term Vrddha means that which contains a

vrddhi vowel in its first syllable and is used in this sense by
Panini. Thus words of which the yowel in the first sylla-
ble is @, ai or au are vrddha. In RT. vrddha is used in the
sense of pluta or prolated vowel: tisro vrddham RT. 46.
Some predecessors of Panini used “‘vrddha” in the sense of
Panini’s ‘‘gotra”. ““Vrddha-Sabdah piirvacarya-samjiia gotra-
sya,” says the Kasika on “vrddho yiina tal-laksanas ced
eva visesah” i. 2.65, and under iv. 1.166 quotes a rule from
an ancient work on grammar : “‘apatyam antarhitam vrddham’”’
corresponding to P&anini’s “apatyam pautraprabhrti gotram”’,
iv. 1.162. Hemacandra also uses ‘‘vrddha” in this sense.
Candra cannot use vrddhi, so vrddha also is taboo to
him, so for vrddha he has to use Adaijadyac. Thus for
O.P 129—3
34 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Panini’s “‘udicam vrddhad agotrat’”’ iv. 1.157, Candra_ has

“agotrad adaijadyacah” ii. 4.90.
Santanavacdrya uses “Yamanva” for “‘vrddha” in_ his
Phitsutra,
Jainendra uses ‘“‘du” for ‘“vrddha.”’ Sakatayana and
Hemacandra follow Jainendra. “Du” is evidently formed
by taking “‘d’’ from the last syllable of “‘vrddha’’ and add-
ing “u’’ to round it off. Sakatayana would appear to connect
“du’’ with ‘‘Devadatta.”’
Sam. avoids the term vrddha by using Adisthaijat.
Thus for Panini’s ‘“‘avrddhabhyo nadi-manusibhyas — tanna-
mikabhyah” iv. 1.113 and = ‘“‘nityam vrddha-saradibhyah”
iv. 3.144, Sam. has “anadisthaijato nadi-manusi-namnah”
iv. 201 and ‘“‘adisthaijad-ekac-Sarader abhaksyavastrayoh’’ iv,
591 and so on.
Similarly Su. prefers using vrddhadyac to having a new
technical term vrddha.
Thus for ‘“‘udicdm vrddhad agotrat”’
iv. 1.157 Su. has “‘vrddhyadyaco’ gotrat v. 2.107.
Since HN. uses “Vrsnindra” for “‘vrddhi,” its term for
‘‘vrddha”’ is naturally “‘adivrsnindra.”’
In CA.
also vrddha is used in this technical sense. Thus
we have the rule ‘“‘vrddhenaikaksarena svarantena’’ iv. 55 on
which the commentator says :
avagrhyat svarad yas tu taddhito vrddhiman bhavet
ekajvrddhi-svarantesu na caivavagraho bhavet.
P. vrddhir yasyacam 4dis tad vrddham. en pracam
deSe. tyadadini-ca. i. 1.73-75.
J. yasyaksvadyaib duh. dega vain aisike, prag
dese. tyadadih, va khoh j, 1.83-87,
 VRDDHA , 35

Sak. nama duh. tyadadih. yasyaksvadir adaic. deéa

evain chadau. pragdese. 1. 1.17-21.
He. samjna dur va. tyadadih. vrddhir yasya_ svares-
vadih. edod deSa eveyadau. pragdese. vi. 1. 6-10.
PR. tyadadi-vrddhy-akaradi-svara vrddhah pare tu
va. vrddha-samjnakah 1. 1177.
HN. 4-ai-au-rama yasyadi-sarveSvarah sa 4Adivrsni-
ndrasamjnah, tad-yad-adayasS ca. vrddha-samjiia_ ityanye.
Vrtti on Taddhita 263.
The following well-known stanza containing ‘vrddha’ is
found in works on Subhasita:

“adeq 2” safe fart ares saat

“rar qa” ag wftaarg arid aiff: |
Copy Wa” Wate For: APTANA F:
‘et gaa” saafe eat fae Ae sae1"
“Fach has double”— prosperity and adversity are the
inevitable lot of every creature.
‘The elder, when with the younger’—young maids get
friendly, is left in the lurch.
‘One in a family’—there is only one such and he is
the one who is prosperous.
‘Feminine like masculine’’—when this begins, know the
house is ruined.

* This is a Samasydpiirana with P&nini viii. 1. 1; i. 2. 65 > IV.

1. 93: i. 2. 66. S R B reads the first half thus:
aaa © gaiqnna aaciaimeg
wal aa meat gary a: acer frat |
Fhen it reads our b and d in which it has taddhi-for viddhi.
 36 TECHNICAL TERMS OF SANSKRIT GRAMMAR

BHUTA-KARANA
Bhuta-karana is a very apt name given to the augment
in the Pratisakhyas of AV. and in K. In the original Indo-
Européam language the so-called tenses had no time-signifi-
cation. They only denoted differences in the aspect of the
action. The augment* was prefixed to the Imperfect, Plu-
perfect, Aorist and Conditional to express past time. This
is why it is called ““Bhuta-karana’”’ (lit. maker of past sense
or past tenses) in Sanskrit grammar.
Philologists distinguish between two classes of augments.
When in the parent language, it was prefixed to a verb
beginning with a consonant and formed a syllable by itself,

it is called the syllabic augment, e.g., I. E.* e-bher-e-t, Skt.

a-bhar-a-t, Gk. e-pher-e, Arm. e-ber. This is “at” jn Panini
(lun-lan-Irn-ksvad udattah vi. 4.71). a
When the augment was contracted with the initial vowel
of the verb, it is known as the temporal augment. Thus
I. E.* @sm (e-es-m), Skt. asam (a-as-am),
Gk. é@ (€-es-m).
This is “at”? acc. to Panini (ad ajadinam vi. 4. 72).
[t will be noticed that Bhuta-karana is the common name
for both these varieties of augment.
Bopp identified the augment with the privative particle
‘‘a’”” by which elusa was thought to have come to mean ‘I
loosed once’, because it started with the meaning ‘I am not
loosing now’. Bopp’s opponents naturally asked how it was

*The augment is found in Sanskrit, Avesta, Greek, Armeni

an and
Phrygian. It has disappeared from Latin: and other I. E. languages,
 SARVANAMASTHANA, PADA AND BHA 37

possible to believe that when primitive man wished to say ‘I

have seen’, he said. ‘I do not see now’.
Madhava says in his Rgvedanukramani :
FE-SS-BSRASATA FT: FT FATS ATR: |
A AHA TAATATERTATHIL LHS ATT|] ULLAL
CA. bhutakaranasya ca iil. 49.
And (the a) of the augment (with the initial r short or long
of a root becomes ar).
APr. has “bhuta-kara’’ for “‘bhiita-karana.’’ Here, we find
dur-dvayayor bhuta-karasya lopah 49c. i.e., in the two ‘dur’s
the augment has been elided.
K. bhitakarana-vatyas ca ii. 1. 14..
‘The (three) tenses with the augment are used to denote
past time.’
The Vrtti on this rule runs thus: bhitah kalah karanam,
tad vidyate yasam iti parisesyad hyastanyadytani-kriyati-
pattaya eva riidhah. eta bhutakarana-vatyas tisro vibhaktayo’
tite kale bhavanti. akarot, akarsit, akarisyat.
The dissolution of the compound as a Karmadharaya
would go to show, that the meaning of the word had already
beconie obscure by the time of the Vrtti-kara. It will also
be noticed that the author of the Vrtti uses ‘“‘vibhakti’’ in the
sense of ‘“‘lakara’’ of the Paninian system.

SARVANAMASTHANA, PADA AND BHA

Indian grammarians were familiar with the distinction
between strong, middle and weak endings in the declension
of nominal stems from a very early period. In the Asta-
38 TECHNICAL TERMS OF SANSKRIT GRAMMAR

dhyai the strong endings -su, -au, -jas, -am, -au in the case
of masculine and feminine stems and -jas and -Sas in the
case of neuter omes are designated Sarvanamasthana. For the
middle’ and weak endings there are no special names, but
a stem is termed pada before the middle endings (-bhyam,
-bhis, -bhyas and -sup) and bha before the weak endings
(-Sas, -ta, -ne, -Masi, nas, -os and -ni). We shall deal with
these terms one by one.

Sarvanamasthana.

Next to Karmapravacaniya, Sarvanamasthana is the

longest technical term in the Astadhyayi. Annotators and
commentators of the Panini school have been hard put to it
to find a rational explanation of this sesquipedalion word
and they can put forward no acceptable reason for the use of
the word by Panini who often uses monosyllabic technical terms
like ti, ghu, bha etc. Thus the Nyasa says :

qa Tata wed Aa Tata | aT re

aaa A AT ACSIA WAGM AAS wera TeaG-
aaa, ad Ga: eT Gaemaratyaftereig | afz
fe azrvag wafa, TaNeT TTS BA, ATTA |
‘It is the ancient teachers who have coined this big
technical term for no reason whatsoever. The use of the term
in this work is to show the defective character of the gram-
mars composed by them. And that is done to show that our
author’s work is not superfluous, for if the previous works
are defective then only is the composition of a new treatise on
the subject justified, not otherwise.’
 SARVANAMASTHANA, PADA AND BHA 39

Haradatta in his Padamanjari quotes the correct views of

his predecessors and yet fails to see the point. Following in
the footsteps of Jinendrabuddhi he says :

aa: et aad Haeat | «feed g edt fired 2 She.

ate:—aa ata fasafefafa adataenagq) ata orfa-
afragq | BaMegISATAHTLSATT: | AalaTaye aa aA
fagaivrd: | aatrera saat ara fasdtast aaa) fH
fad waft? stgqr reral amant aa eat
faafe: fag: wafa, sere frahrarcomrarcafasate |
ad g afer ofacias fe aafafa a aaa) aang
qataratgareassqrar Aadt cent fara | Padamafijari, Vol. I,
pp. 88-89.
‘Thus it is established that a technical term has to be
coined. But why is such a big one framed? Some say:
the whole Naman remains in it, so it is named Sarvanamas-
thina. Naman means Pratipadika or stem, sarva denotes
the entirety of the parts. So the meaning is: where-the stem
remains with all the parts intact, the implied sense being that
before other endings the stem suffers loss or reduction in some
of its parts. What purpose does this serve? When there
is Samprasarana in words like upeyusah, the disappearance of
the i that has been already brought in is effected (because of
the implied sense). Otherwise (if the term Sarvanamasthana
were not used and consequently if the implied sense were
absent) because of the absence of any ground for disappearance
it would remain. Even so it is not possible to know what. is
wanting in. what Pratipadika. So this big technical term has
been framed to rebuke the older Aciryas.’
 40 TECHNICAL TERMS OF SANSKRIT GRAMMAR

The author of the Prakriyasarvasva mentions the view of

Haradatta :

aaa HATE aAaizre |

faz: werarartsst ater “agate |
Gaara qeraad aq FRHenNgUsaaAN ||
‘The endings are designated Sarvanamasthana, because the
entire word (Sarvanaman) remains here. Before -si and -sut
syncopation etc. do not take place as in the case of ‘daimna’
and ‘papusa’ there is the elision of the ‘i’ of ‘papivas’ and the
reduction of ‘va’ to ‘u’). Haradatta said that the term was
used for holding up the older grammarians to ridicule, as no
useful purpose was served by it.’
Bhattoji in his Sabdakaustubha is unable to explain the
significance of the name and contents
himself with the re-
mark: mahasamjnakaranam pirvacaryanurodhad ‘the for-
mulation of the big technical term is out of deference to older
grammarians’.
Nagesa remarks rightly under ‘‘is-usoh samarthye”’
S: BK; 156:
warracardteag tft at arate, oar.
aiatata AeTaRTa |
Srstidhara, a Bengali grammarian of the 17th century,
gives very nearly the correct explanation—the explanation re-
jected by Haradatta—in his commentary on the BhasAvrttj-
aataertata wedi det atat vat mfaaari
¢ a. ° “aA @ e

*The Madras edition from which this extract is taken reads

‘yapusetivat’ which is evidently due to the similarity of ‘p’ and ‘b’
or ‘v’ in several South Indian scripts.
 SARVANAMASTHANA, PADA AND BHA 41

enafaehi—aa fe aatata nfaafar aeaiarafagea,

srarfeg g aca-cacaifafatasar aanhifa areata |
‘The big term Sarvanamasthana is used, the sense being
there is room for all the Pratipadikas here. All the Prati-
padikas retain their forms intact in the case of these endings.
Before -Sas etc. they undergo syncopation etc. on account of
their becoming bha, pada etc.’
The explanation rejected by Haradatta is to be preferred
to that of Srstidhara, since that does away with the necessity
of vyadhikarana bahuvrihi and is moreover the more apt and
more natural explanation. But the above two explanations
which ultimately amount to the same thing have not found
favour with the grammarians of the Panini school, because in
the artificial system of Panini the suffixes postulated by older
grammarians were discarded in favour of what were con-
sidered simpler or more conducive to brevity (laghava).
According to the predecessors of Panini comparatives are
formed with iyans which is retained in the strong cases but
suffers the loss of the nasal before the weak and middle
endings. Similarly perfect and present participles were formed
with -vans and -ant respectively. Panini, however, found
that these full forms occur only in five cases, whereas the
reduced forms occur in the remaining sixteen cases, so he
postulated the suffixes as -iyas, -vas, -at etc. to secure laghava
(brevity) and prescribed a m as agama in the strong cases of
the stems. So it is not possible for the grammarians of the
Panini school to realise fully how the full form suffered
reduction or weakening before the middle and weak endings and
how it was retained before the strong endings. One really
 42 TECHNICAL TERMS OF SANSKRIT GRAMMAR

expected Panini to use the monosyllabic pratyahdra sut for

the strong endings. But the Acc. pl. of the neuter stood
in his way and moreover ‘sut’ is an augment in the
Astadhyayi.
The Pratisakhyas of AV. use the term Pancapadi for the
strong cases, because they comprise the first five endings.
vasvantasya pancapadyam CA. ji. 88.*
osadher apancapadyam CA. iii. 5.
antahpade’pi paficapadyam CA. iii. 59.
caksusam iti paficapadyam APr. iti. 1. 13b
catvari ksaipraS ca paficapadyam antodattadinyayat
Apr. i. 14.
The Katantra school, ever unwilling to sacrifice simplicity
to brevity, reads -iyansu, -vansu and -Santrn for -lyasun,
-kvasu and -Satr of the Panini school,
so the technical term
sarvanamasthana is significant in the treatises of this school,
yet by a curious irony of fate, this school, though preserving
the ancient self-explanatory terms in most cases, has substj-
tuted the meaningless ghut (pancadau ghut)
in this particular
case, though it has retained the corresponding term sarvadha-
tuka in the Akhyata-section and uses also gunavrddhisthana
in Akhydta 148. ‘Ghut’ is evidently coined after the
pratyahara ‘sut’ of the Panini school. Gh was probably put
in at the beginning as being the fourth consonant of the
first varga even as sarvanamasthana comprises the first four
vibhaktis, counting the two ‘au’s-as one, ‘ghut’ comprising
the four case-endings -su, -au, -jas and -am.

*It is worthy of note that CA. names the suffix -vasu and not
“Vansu.
 SARVANAMASTHANA, PADA AND BHA 43

With the “ghut” of K. cf. Naradi iksa vii. 16:

frat af: wera: Racrafea: awrerhi:
qeaaag wag fasta greta |
Candra uses ‘Si’ and ‘sut’ for sarvanamasthana. Cf. “‘6i-
suti’? v. 3. 7 corresponding to P&anini’s ‘‘sarvanamasthane
cAasambuddhau”’ vi. 4. 8.
Devanandin uses dha (Syanapsut dham i. 1. 32). In the
language of mathematics ‘dhana’ means plus and ‘rna’ minus.
Since an additional element, n for instance, makes its appea-
rance in many cases in the sarvandmasthana vibhaktis, ‘dha’,
the first syllable of “‘dhana’”’, is used to denote them.
Sakatayana uses “‘Syanapsut’ throughout. Cf. Ssyanap
sutyan i. 2. 109.
Hemacandra and Purusottama follow K. and use ‘ghut’.
Kramadigvara uses ‘sut’. Cf. suti vrddhis trnadeh vi. 58.
On this Goyicandra says in his commentary :
afefa suasaaaelzeraa seofafa mena:
Tea: | AMAA HEN) Tada qaufyqulzenrcragaru:
qMreaeract AT HA: |
Sar. uses catursu. Candrakirti and others following Panini
and K. read ‘paficasu’. ‘Catursu’ is used in Sam. in the
sense of sarvadhatuka for ‘catursu lakaresu’.
Vopadeva uses ‘ghi. The ‘gh’ corresponds to ‘catursu’
of Sar. being the fourth consonant, i being added, as in. the
case of ji, for ease of utterance because ‘‘kt’’, the term for
the generic word “vibhakti”, ends in i, or the ‘gh’ might have
been borrowed from ‘ghut’ of K. Panini uses ‘ghi’ in an alto-
gether different sense. Since the nominative dual is iden-
44 TECHNICAL TERMS OF SANSKRIT GRAMMAR’

tica! with the accusative dual in each and every case, Vopa-
deva with his hard clear intellect cannot bring himself to use
‘au’ for the one and ‘aut’ for the other.
Supadma compresses *sarvanamasthana’ into ‘sadisthana’
where ‘sa’ stands either for the Nominative singular ending or
for ‘sarvanaman’. Visnumisra in his Supadma-makaranda
remarks:
niranvayeyam samjia. idam eva sarvanamasthanam ityuk-
tam munina i.e. this technical term is meaningless. This very
term has been designated ‘sarvanamasthana’ by the sage
(Panini).
HIN. uses Krsnasthana for sarvanamasthana, just as it
uses Krsnanaman for sarvanaman and Krsnadhatuka for
Sarvadhatuka. Cf. Bhagavata Purana 1. 3.28:

Ud BMH: FS: SMT ATA Gaz |

qt) fa aaarAetay | BSATaART! $F 1e1e2-3
eT) UgTl gz.) FETA aR) zypiase
Ht) waa ari leas (RI water) aq
= AT ash | ) }
21 fiedz_) gfe ert) neree: a
Heals: | F|Z00
q| wamlsa
fa: | fit: ert cx-c3
a fa: gz aafaer arfzeatay | 33e
T) gz etsran eitg’cen: eg: | frig’ aR | epr9ce
Z| areas: Rea HeMeIA-Par: | qz-dar cas |
AAAACATAGA CAS | IAG"
With the use of ‘sthana’ in Sarvanamasthana we may
 PADA 45

compare ‘nivrttisthana’ in the Nirukta and ‘guna-vrddhisthana’

in K.
Most of the purposes served by sarvanamasthana are
served by sut,- but sut is unfortunate for two reasons: first,
there is the anubandha t of au, secondly, there-is the 4gama sut.

Pada

In RV. Pada means a step. Thus Visnu is said to have

covered the three worlds with his three padas (trini pada
vi cakrame i. 22. 18) and we are often reminded of the ‘‘para-
ma pada” or highest step (RV. 1. 22. 20 etc.) of Visnu. Now
when each quarter of a stanza (or each third in the case of
Gayatri stanzas and each fifth in the case of Panktis) came
to be regarded as a step to the expression of the full mean-
ing of the stanza (pade pade samapyante prayenartha avan-
tarah—Venkatamadhava on RV. vi. 8. 14) it came to be
regarded as a pada (cf. arthah padam V.P. iii. 2). AB.
speaks of ‘‘ekadaSaksarani padani” i. 6 or verses consisting
of eleven syllables.* AA. (ii. 2.2) says: esa vai padam.
esa himani sarvani bhutami padi. sa yad imani sarvani
bhitani padi tasmat padam. tasmat padam ityadcaksata etam
eva santam. ‘‘He is a quarter-verse, for he has entered all
these beings. Because he has entered all these beings, he
is a quarter-verse. Therefore they call him who is (prana) a
quarter-verse.”” When the seers proceeded still further in their

qzme: VISORS Nrees EMa— “aE: GAIA Geq |” CH:

giz: yau:;; Mahfbhisya cd. Kielhom II. 191. 26-27,
 46 ' TECHNICAL TERMS OF SANSKRIT GRAMMAR

analysis and regarded each word composing a sentence as

such a step, words came to be looked upon as padas. In
the case of compounds the sense of each component helps
to express the idea of the whole, so each component was re-
garded as a pada. In the case of certain suffixes and end-
ings the base of a word behaves exactly as it does when it is
used as the previous part of a compound, e.g., vidvas becomes
vidvad before -bhyam, -bhis, -bhyas and -sup exactly as it
becomes vidvad before gana, vrnda etc., so stems came to be
regarded as padas before -bhyam, -bhis, -bhyas, -sup and
taddhita suffixes beginning with consonants except y. Thus
in the Pada-patha of Sakalya not only are the words of the
hymn taken separately, but compounds are often analysed into
two component parts and single words into prakrti and
pratyaya, pratipadika and vibhakti.*
Every system of grammar with the exception of Mu.
and HN. uses the term “‘pada’’. Even J., the guide of
Vopadeva in the matter of technical terms, had not the
temerity to shorten ‘pada’. Vopadeva uses the second syl-
lable ‘da’ for ‘pada’ and HN. prefixes the word Visnu to it
and changes it into Visnupada. But centuries before Vopa-
deva, RT. and ST. had used ‘da’ for ‘pada’ (da mu RT. 68.
de no miirdhanyam 272. Vide Suryakanta, p. 36).
K. takes pains to define Pada at great length: pirva-
parayor arthopalabdhau padam i. 1. 20—a definition which
combines the ancient definition ‘‘arthah padam” with the
more modern one “vibhaktyantam padam’’. As the Vitti
explains: purva-parayoh prakrti-vibhaktyor arthopalabdhau

*Cf. CA. iv 31-32: bhir- bhyAm-bhyim-bhyas-su. sau ca.

 PADA 47

satyam samudayah pada-samjfio bhavati. The ‘Tika says:

purva-parayoh samudayah artha-pratipattihetuh padam. Su-
sena in his commentary on this rule mentions the different
views : arthah padam 4hur Aindrah, vibhakyantam padam
ahur AijSsaliyah, sup-tin-antam padam P4niniyah, ihartho-
palabdhau padam iti Vararucih. ‘‘vibhaktyantam padam”’
no doubt means exactly the same thing as ‘“‘sup-tin-antam
padam,”’ still it serves to show that Apifali generally used
the bigger technical terms and was probably unacquainted with
the pratyaharas sup and tin.
The reference to Vararuci is not clear. It appears to
be an extract from Vararuci’s Vrtti on K.

According to commentators ‘‘parvan” is

used in the
sense of ‘“‘pada’”’ in VP. i. 138: antarena parvani. In the
Nirukta and CA. iv. 77 “parvan’” means ‘a member of a com-
pound’.
Candra uses “pada” in ‘“‘apadadau padad ekavakye”
vi. 3. 15 without defining it. The Vrtti gives the orthodox
explanation : padyate gamyate’ nenartha iti padam.
Similarly Sam. uses the word though it does not define
it. The middle endings -bhyaém, -bhis, -bhyas and -sup it
denotes by bhadi and the weak endings by éasadyac.

aT SAT! AE TAA] IR MaceAyaa: gay] ger

avi ¢ie3-8

Tw) a: et) fafa)

wrt d
wias eanfreray-
ATACATA | -LIBILB-Ro
Bt) GATaTaaigseat gr] ei eiRe
Rl | AHAPTA: THT Roe
 48 TECHNICAL TERMS OF SANSKRIT GRAMMAR

3 | @ftaerd wa) a Fa) fafe) eee

CIRIZ SEARS
Mm)! ge( gifs)
wy) tt fasevara |
age: | wteaad, Aaararsfau afa caaer-ees
a1 (earfeteahe: 219129) azect Tz ATA fear |
a 74, AW Aad! ageisfeu
afl aeectsas |
{IRIRO-R's
a, frag az t
Brag) TTA: |
a) ofasetazq) aiz0y aFa: |
a) Faegt gat 78, sea AAT cx, ARTAAT,
fa: 201, af aat aeasta) eRe
a, afased waq) a: Fast) aqatgasafa | a
aet aead | afa maristytrgs: |) 33 12-"
| agafeq-aftearaiagaan: wa tas
a1 fasufafes fayrzq i cigee
aregifcar | arnare 32 afaq ghey? |
WSAFATTARAS TAR ||
aaiq 1¢ cifid gate arearafaoia: |
qzTagias ard aWagas Tay|| Teeqa ~12 2S
ANAAT; TA | WAT AR! WRAAQa: aE.
faeqradt | ARTATET 217185

Bha.

The technical term ‘‘bha’’* as the name of the stem

before the weak endings is peculiar to Panini, J. and Su.

| *‘Bha’ is also a taddhita suffix in Panini. It is added to “éam"

to form the word “Sambha" according to y. 2. 138.
 BHA 49

‘Bha’ may be the first syllable of “bhan.ga” or. “bhagna”’

because before weak endings the stem often undergoes reduc-
tion or syncopation. Or “‘bha” might have been suggested
to the grammarians by the ‘bha’ found in -bhis, -bhyam
and -bhyas, which are termed ‘bhadi’ by Kramadisvara.
Since in the case of these endings the pratipadika is regard-
ed as a pada, “‘bha”’ came to denote the other class. Lucas
a4 non lucenda! In RT. ‘bha’ stands for ‘stobha’.
K. generally uses a-ghut-svaradi. Cf. “aghut-svare lopam”’
ii. 2. 37, ‘‘aghutsvaradau setkasyapi vanser vaSabda- syotvam”’
ii. 2. 46, etc.
PR. follows K. Cf. ‘“‘a-ghutsvaresu dhatinam dAlopah”’
ii. 216, ‘‘aghut-svaresu lapy’ono’karah samyugvamasthitah”’
ii. 327. |
Candra uses “yac aSsisut” for Panini’s “‘bha’’. Cf. the
adhikara rule ‘‘yacy aSisuti’” v. 3 126 for P&anini’s “‘bhasya’”
vi. 4. 129.
Sam. generally uses “Sasadyac’”’: Cf. ‘‘Sasidyacyato
yugvamah” vi. 193, “dhator 4l-luk Sasadyaci” vi. 49 ete.
It has also the atideSa rule “taddhita-yacau §asadi’’ vi. 217.
HN. uses ‘“‘Bhagavat’”’ the name of God beginning with
‘bha’ for Panini’s “bha”’. It has a separate name Yadu

for gasadi endings (Sasadayo Yadu-samjiah) and another

Vrsni for the nit endings n.e, nasi, nas and ni (n.ito
Vrsnisamjfah i. 181).
Mu. uses ‘“‘pi’’ to cover most cases of “‘bha”. ‘Pi’ ap-
pears to have been formed with the ‘p’ of ‘pada’ with ‘i’ to”
round it off. ‘‘Pi’” is thus defined by Vopadeva : aghyac tacyep
pih 101 (which is thus explained in the Vrtti: ghi-
O.P, 129—4 |
 50 TECHNICAL TERMS OF SANSKRIT GRAMMAR

varja-syader ac ta-samjnav ac-yakarau ip ca—ete pi-samjiah

syuh) i.e., the vowels of the case-terminations except the
strong ones, the vowels and y of taddhita suffixes and the femi-
nine suffix -I are regardeed as “‘pi’’. It must be noted in this
connection that in the system of Panini the stem is designated
‘bha’ before these endings and suffixes, whereas in Mu. it is the
endings and suffixes that are designated “pi’’. It will be
noticed that Mu. has three technical terms ‘‘pi’’, ‘“bhi’’, and
“ghi’” for the case-endings. Vopadeva must have got the
idea of “‘pi” from pada, that of “‘bhi’ from either -bhyam,
-bhis, bhiyas or from Panini’s “‘bha’’, and that of “‘ohi’’ from
K’s “‘ghut’’.
“Bhi” is formed with the bhi contained in ‘bhyam’ and
‘bhis’ and denotes those endings and suffixes before which
Panini prescribes pada-samjna. But Vopadeva’s definition of
‘bhi’ is peculiar. He says that the grammatical operations
prescribed before bhi are to be understood as taking place
when a consonant follows in the case of pratipadikas and
when a consonant except nasals and semi-vowels follows in the
case of roots. The influence of K. is clearly traceable here.
In K. certain grammatical operations are prescribed in the
section on declension before ‘“‘virama’’ and ‘“vyanijana” (K..
i. 3. 44 ff). Many of thése operations are prescribed before
“dhut” and “‘virama’”’ in the section on conjugation (iy.
2. 54 ff).
a) af am, ae aa) sere cae |
R1BIZ?-Ro
41 afa a) agnitsea aft, TU ssaage
Aras TAY | e11RR0-258
 SARVADHATUKA 51

ga afaumarfeena, 2a
Z| WEWeMrapat anaqedaatsa V1 Rw
SARYADHATUKA AND ARDHADNATUKA
Exactly corresponding to Sarvanamasthana in the sec-
tion on declension Panini has Sarvadhatuka in the section
on conjugation. In ancient times the vikarana was regard-
ed as part and parcel of the root, so those endings before
which the vikarana was reserved were known as sarvadha-
tuka endings, because the endings were added to the entire
root, and those that were added to the root without vikarana
were known as Aardhadhatuka endings, because before these
endings, only half the root i.e. a part.of the root was dis-
cernible. The earlier form of the word was sdrvadhatuka*
which was used by ApiSali and which, as an adjective: to
vibhakti, was naturally feminine. It is clear from this that
even at the time of Apisali the word was felt to be self-
explanatory, but at the time of Panini it had become fossilised
and was looked upon. as a mere technical term without any
meaning. So after the analogy of sarvandmasthana, sarva-
dhatuka lost its feminine suffix.
K. which uses “ghut’ for “‘sarvanimasthina” preserves
“sirvadhatuka” in its later neuter form, though its com-
mentators like those of the Panini system, have failed to find
any sense in the word: sarvadhatukam napumsakam lokatah
siddham pirvacarya-samjiia niranvayeyarn Tika sarvadhatukam

*With the 4-stem “sarvadhatuka” cf. “dargapaurnamasika” in

ASS, ii. 17. 2 and “pasubandhika” in ASS. iii. 1. 8.
 52 TECHNICAL TERMS OF SANSKRIT GRAMMAR

iti purvacarya-prasiddheyam samjha niranvaya svabhavato

napumsaklingam Panji.
‘“Sarvadhatuka” appears to have been formed with ‘-thak’
like pyatipadika in
the sense of ‘grhnati’ (sarvadhatum
grhnati) or in the sense of ‘tasyedam’. Though words form-
ed with -thak and -than take I in the feminine, stil] sarva-
dhatuka follows the analogy of prathama, dvitiya, trtiya,
vartamana etc., and takes a. In the commentary (Gidhar-
thaprakasika) on PR. the word is derived thus:

apigtd Aaage: | SIs 1 aa WAT TAT |

qa Ad era wt aM aqgay (012032 ) cearfir apa}
afe: | saat aaaTgy wa wad Wadeq at( 222 )
fa queer ARITA Waa FTAA ATaeasy HOT TeqVY-
am aqn-arman-weatateteea |
As has already been mentioned the endings were known
as ‘“‘sarvadhatuka” in the Apisali system. Thus Vamana
says in the Kasika ApiSsalas tu-ru-stuSamyamah sirvadha-
tukasu cchandasiti pathanti” (vii. 3. 95) on which Jinendra-
buddhi says : “strilinga-sarvadhatuka-sabdasya sam jiiat-
venapiSalina pranitatvat.”” Haradatta also says the same
thing: “tabantam tatra samjnatvena viniyuktam.”’
APr. uses “‘sarvadhatuka” in ‘“‘vaso dadati bhartava iti
sarvadhatuke’” 125. Apr. has ‘“‘dadateh. iti sarvadhatuke”’
i. 4. 1-2.
J. uses “ga” for “‘sarvadhatuka’’ and “a-ga” for ‘“gr-
dhadhatuka.” Was “ga’’ suggested by “gana” or is it ”
part of “‘samagra’’, a synonym of “‘sarva”’?
Candra, as usual, avoids these technical terms, but fol-
 SARVADHATUKA 53

lowing Panini’s rule “‘tinsit sarvadhatukam’”’ iii. 4. 113 uses

‘“tinsit’ instead. For “‘ardhadhatuka’”’ Candra uses ‘‘a-tinsit”’,
even as K. uses *“‘a-sarvadhatuka.”’
Sakatéyana uses “‘slela’” for “‘sérvadhatuka,” ‘s’ stand-
ing tor Dy
sit’, ‘le’ for* ‘let’ and ‘len’ and ‘la’ for ‘lat’ and
‘lan’. Thus for Panini’s “‘sarvadhatuke yak” iii. 1. 67
Sakatayana has “‘slele yak” iv. 3. 19 and for Panini’s “‘sar-
vadhatukam apit”’ i. 2. 4 Sakatayana has ‘“‘Slyablau”’ iv, |. 147.
Sakatayana uses ‘let’ for Panini’s ‘lot’? and ‘len’ for Pa&nini’s
‘lin.’ ‘“‘Lyasi” is used for Ardhadhatuka (iv. 2. 81 ff), “‘li”
standing for Lit and Lin (Panini’s “‘linagisi’) and “‘S” for
66 4°473
sit’.

Hc. uses ‘“‘sit’” for “‘sarvadhatuka’” and ‘‘a-éit’? for

‘‘ardhadhatuka.”’
Kramadisvara uses “‘catur’” or ‘“‘caturladadi’ for “‘sar-
vadhatuka”’ and ‘‘a-catur’ for “‘ardhadhatuka’’ without defin-
ing the terms, as “catur’’ is sufficiently self-explanatory,
referring as it does to the first four lakaras. Cf. gadaé catursu
ii. 85 (ladadisu catursu kartavyesu—Tika). caturladadau éan
bhinyddeh ii. 193 (kramat kathyamanesu catursu lat lot lan
vidhyadilinsu paresu. Vrtti).
Mu. uses ‘ra’ for “sarvadhatuka” and ‘“a-ra” for
“ardhadhatuka”. The “ra” was probably suggested by the
“ra”? of ‘‘sarvadhatuka’’. But unfortunately r occurs in
‘ardhadhatuka’ also.
HN. uses “Krsnadhatuka” for ‘“s§rvadhatuka’” and

* This is the reason why Sak. uses ‘Let’ for Panini’s ‘lot’ and “lan”
for P’s “lin”.
 54 TECHNICAL TERMS OF SANSKRIT GRAMMAR

‘“‘Ramadhatuka”’ for “ardhadhatuka,’’ because Krsna is re-

garded as the purnavatara (full incarnation) and Rama as
the amSavatara (partial incarnation) of the Supreme Being.

aT| ; ferefare, aa TTHA | ATT AT Ta aT: | fexa |

faerfarts | seaeqagat | =3141223-229
AT| Gt: aeagH! 2121384
a fre fea) aasaaq, fez) feerfate
RIZIS4%-Se
21 Gat: fara) a1a1R0
Gat: ada -aant-wayat-eaer: fara: araaeat
afgacat: | Teahat
BZ) wat: fre) 43%
a fagemefeess: aangeq) aeagen-
ATTRA =BlklZ9O-Fe
aaagefafa qedeat qaienaofasie faagr|
aa: ealiadt ayaanfegary ; wR |
9) sen feet a eeaaft a fer fq, ce
adam fas az eet wea ser: fine
dat: wy) agaea cfefa caiat, fie arpargafat
a | afd:|
E] aegaes: wy fast eewagst: | arargar-
aaa | | Ae TTT SCCTASTITRT: ATPATTRA |
8 |Ac8-Rc™,
One of the most ancient rules of grammar preserved in
the Katantra, as it is now extant in Bengal, is “‘sadadyah
sarvadhatukam”’ iil. 1. 34. The previous ten rules mention
Vartamana, Saptami, Paficami, Hyastani, Adyatani, Paroksa,
 SARVADHATUKA 55

Svastani, Asih, Bhavisyanti and Kriyatipatti. One feels,

therefore, naturally, inclined to take the first six wz. Vartamana,
Saptami, Paficami, Hyastani, Adyatani and ,Paroks4 as meant
by the expression “‘sadadyah”. But Paroksé eam by no
stretch of the imagination be regarded as Sarvadhatuka, and
Panini distinctly lays down that Lit is Ardhadhatuka (‘‘ardha-
dhatukam Sesah lit ca.” ili. 4. 114-5). Moreover, if Adyatani
and Paroksa be regarded as Sarvadhatukas we cannot. have the
union-vowel i in the case of their consonantal endings, for
Sarvavarman prescribes the Agama “‘it’” in the case of
a-sarvadhatuka endings (“‘idagamo’ sdrvadhatukasyader
vyanjanader ayakaradeh” iv. 3. 1). Then again the Katantra
lays down that with the exception af the first person of the
Imperative, all the Atmanepada endings are weak in the
case of the Sdarvadhatuka vibhaktis (‘“‘sarvesim Atmane
sarvadhatuke ’nuttame paficamyah” iv. 1. 18), yet it says that
the Atmanepada endings of Paroksa are all weak (“‘sarva-
tratmane’”’ iv. 1. 21). Again K. directs that the duals and
plurals of the Parasmaipada in the case of the Sarvadhatukas
are to be regarded as weak (‘“‘dvitvabahutvayo§ ca parasmai”
iv. 1. 19), yet it says that the duals and plurals of the Paroksa
also are to be regarded as weak. It is clear from all this
that Sarvavarman does not regard Adyatani and Paroksa as
Sarvadhatukas. This obvious anomaly has led the Vrttikara
and subsequent commentators to explain the rule thus:
sannam vibhaktinam adya vartamand-saptami-pajicami-
hyastanyas catasro vibhaktayah sarvadhatuka-samjiia bhavanti.
This explanation can hardly be accepted. In the first place,
we do not generally find Tatpurusa-compounds with numerals
like ‘‘sadadyah”. Secondly, there is no reason why the
 56 TECHNICAL TERMS OF SANSKRIT GRAMMAR

author should go out of his way to mention the first six out
of the ten. Thirdly the plural in ‘‘adyah’’ should normally
mean three and not four according to the Kapifjaladhikarana
in the Mimamsa philosophy (xi. 1. 38-45). We may say like
the orthodox commentators: qaaqey =a anAASica ray
afcarrad, saaraaer a afteit ancora,
wa Tea matraateaa oq fara aren: |
Thus we see that the explanation rejected by the com-
mentators as also the one offered by them are equally un-
satisfactory. The first five are Sarvadahtuka acc. to Panini,
Vopadeva, and others. And we know that Let is also Sarva-
dhatuka in Panini. So by “‘sadadyah sarvadhatukam’’ the
author means the first six viz., the present, the imperfect,
the aorist, the imperative, the optative and the subjunctive.
As the subjunctive is confined to the Vedas and as K. studiously
avoids all reference to Vedic forms, we can only conclude that
this is a rule from an ancient grammar* which placed the Sub-
junctive (Naigami) aomng the first six tenses and moods, most
probably after the optative. Later redactors did not think it
worth their while to change this rule, as “sadadyah”’ could be
easily explained away as “‘satsu adyasS catasrah’’. They,
however, appear to have changed “‘sarvadhatukah” into
“sarvadhatukam” after Panini’s rule ‘“‘tin-éit sarvadha-
tukam’”’ iii. 4. 113. It will be noticed that all the names of
the tenses and moods are feminine in K. and we learn from
commentators like Madhava that Apisali used the feminine
form Sarvadhatuka.

* Bhis ais va’ i. 1. 48 is another rule of this kind.

 TI 57

TI
(“Ti is one of the few monosyllabic technical terms used
by Panini. It stands at the head of the artificial technical
terms used in grammar. It means the last vowel in a word
or stem along with the final consonant or consonants, if any.
It may have been suggested to the ancient grammarians by
the word ‘koti’ which means ‘the end of a bow. It may’ be
noted in this connexion that 4gamas having ‘t’ as an indica-
tory letter come at the beginning (4dyantau takitau i. 1. 46).
TI occurs also in the Unadisiitras; e.g., “‘avates ti-lopas
ca’’ i. 141, “sives ter u ca” iv. 93,
K. uses ‘svara antya’ for ‘ti’. Thus for Panini’s ‘“‘avyaya-
sarvanamnam akac prak teh” v. 3. 71, K. has ‘“avyaya-
sarvanamnah svarad antyat purvo’k kah”’ ij. 2. 64.
C. follows K. and uses ‘antyajadi’ for ‘ti’, Thus for
Panini’s “teh” vi. 4. 143 C. has ‘“‘antyajadeh” v. 3. 138.
J., Sar. and Mu. retain this word as it consists of only one
syllable.
Following in the footsteps of Candra, Sak. uses “antyajadi”
for “ti”. Thus for Panini’s “tur isthemeyassu” and “teh”? vi.
3. 154-155 Sak. has ‘“‘trantyadeh”’ ii. 3, 57.
Hc. also uses “‘svaresvantya” in ‘“‘tyadi-sarvadeh svares-
vantyat purvo’k” vii. 3. 29 and “antya-svaradi” in ‘‘antya-
svaradeh”’ vil. 4. 43.
Sam. uses “‘svara-tacchesa’”’ for “ti”. Thus for Panini’s
“teh” vi. 4. 143 it has “ditt svara-tacchesayor luk” vi. 224.
PR. also avoids “ti” and generally uses “‘antyasvaradi’”’.
HN. uses “samsara’’ for “ti” probably because ‘‘samsara”’
is the end of our wordly existence and after the extinction of
‘samsara’ there is ‘salvation’.
58 TECHNICAL TERMS OF SANSKRIT GRAMMAR

It will be noticed that those grammars which do not

recognise the technical term ‘“‘ti”” generally use ‘‘antyasvaradi”’
or ‘‘antyajadi’’ for it and use “‘svarad antyat’’ or ‘“‘svaresvan-
tyat”’ or “‘aco’ntyat”’ or “‘aksvantyat’”” pirvam or param when
some grammatical operation is prescribed before or after ‘‘ti’’.
qt| warseamt= fF ziziec
H | AleaeTa: 21219¢
ary) vcoeacrfate: 2122
ay) aeersrfate: a2
Z| aatsearfs f 217130
Z| waapaufrant: sardat: 212¢%
GHU
The roots ‘“‘da’”’? and “dha” present many points of re-
semblance in conjugation—points in which they differ frém
other roots of their class. Hence grammarians have found
it easier to have one term denoting both. The earliest term
appears to be “da” which is found in the Katantra system
(ii. 1. 8) and which is taken as an upalaksana.* Panini uses
‘“‘shu’’, presumably a shortened form of “‘laghu’”’. For some
reason or other Devanandin avoids “ghu”’; for ‘‘laghu’’ he
has ‘“‘ghi” (i. 2. 111) and for “ghu” of the Panini system he
uses ‘bhu’ (i. 1. 28). Sak. sees no reason to vary the
generally accepted term and retains “ghu” (i. 1. 23), He.,
as usual, follows K. and uses the upalaksana “da” (iii. 3. 5)
and he is followed by Vopadeva 535.
Since the most important ‘“‘da’”’ and “dha” are read in
the Dhatupatha with an indicatory du, and since ‘da’

saifagzana ofa qaunfanenaq |

 GHU 59

contains a voiced element and ‘“‘dha” an additional aspirate

element, one naturally thinks of dhu as the best designation
for both varieties. But since there is a stparate root dhu,
Panini takes the first voiced aspirate of the ordinary alphabet
according to the principle ‘‘prathamopasthitasya parityage
karanabhavat”’ and adds the vowel u to it. J. substitutes bhu
for ghu, probably because it is closely allied to the root bhi
which represents both itself and as before ardhadhatuka sidiues
and suffixes.
Padamanabha, as usual, ‘follows Panini.
Kramadisvara alone of the well-known grammarians ap-
pears to have framed no samjfa for da and dha (cf. ‘““da-
dhor hali ca’’ i. 345).
PR. follows K.
HN.takes the name of Visnu beginning with da and uses
‘‘Damodara”’ as the technical term for “da” and “dha’.
In the Slokavarttika ‘“‘yatra wrddhir acim des tatraicdv
atra ghor hi sa” vil. 3. 3, “ghu” is used in the sense of
‘uttara-pada’. Thus the Bhasya Says: kim idam ghor iti?
uttara-padasyeti. |
QT! Tt ST taaTe a1g1Ro
HT | ATT_ATA aT gizIe
Hy at at vatig eizire
A} ZS: 21212
ST] at AT SAT 212123
fio wal areal ar ara
Flo AT AT aT Gah
Gl FT AT TTT eka
60 TECHNICAL TERMS OF SANSKRIT GRAMMAR

a) atfrarat eons
ze. amaeqaret far zat areca: e192
HNs says “da ity anye, dur ity eke’ where ‘du’ is evi-
dently a misprint for ‘ghu’.
Sriharsa makes splendid use of this term in his Naisadha
xix. 62:

amiigaa dea warentad Aisomsttat KIA:

aos qealafafgaaagasararnqpargs
ata: |
ad frac ar cafaaufa vai areas at gaat
craeniee daeate sate fare: ofeanese i
‘This dove (whose ‘ghu ghu’ sound greets our ears) must
certainly have been some one well-read in the science (of
grammar) as propounded by the son of Daksi i.e. Panini (in
its former birth). It bears on its neck an ornamental sign,
the remains of the numerous bits of chalk worn out in the
process of showing the formation of hosts of words. Though
it has forgotten everything that it had learnt, even now it
shakes its head, owing to the impressions of a former birth
due to the (constant) reading (of grammatical operations)
from wooden slates, while shouting forth the technical term
“shu” (the only thing) which it has managed to remember at
dawn by a stroke of good luck.’

GHA.

Panini uses the monosyllabic technical term ‘gha’ to denote

the comparative suffix ‘tara’ and the superlative suffix ‘tama’,
 GHA 61

probably because ‘gha’ is the first syllable of ‘ghana’ mean-

ing ‘cube’. J. uses ‘jha’ for these suffixes. Probably J.
changes the guttural into the palatal, because there is the
primary suffix ‘gha’ of which ‘a’ remaihs, there is the
secondary suffix ‘gha’ which is replaced by ‘iya’ and so on.
Sakatayana prefers ‘na’ to ‘jha’. Hc. as a follower of K. has
no term for these suffixes. Sam., Mu, Su., PR. and HN. also
do not use any technical term covering these two suffixes.
It is just possible that since the letter eh has two elements
—voice and aspiration, it was chosen to represent the two
suffixes -tara and -tama which are added to two classes of
words—subanta and tinanta, eha being the first letter of the
alphabet to contain these two elements. J. thought that
though gh was the first voiced aspirate in the ordinary alphabet,
jh was the first in this respect in the Pratyahara-siitras, con-
sequently he used “jha” for Panini’s ‘“‘gha”, Sak. took the
‘“n’? of “‘tin’’ and represented the two suffixes by “na’’ because
they were added to tin also.
Going back to earlier
treatises, the Pratisakhyas do not
appear to have a single name for -tara and -tama, for we find
‘“‘taratamayoh” in CA. iv. 16 .and “‘tara-tamayos catigaye’
daksinapratyasange” in VP. v. 2,

GT | ACT-AATT T IIRR
St) (am: weet freer | fates az: wierego-gee )
ct HT BVVER
at anrerdt eat ( = serait ) aa-ad} geist ei:|afer: |
ar | (fasar wart aA | sates | ATT 318)92-92)
al Si R1ap93
 62 TECHNICAL TERMS OF SANSKRIT GRAMMAR

SRADDHA, AGNI AND NADI*

Just as in conjugation ten roots are taken as typical of
the ten conjugations, so in declension also a particular word
was takeneas the type of each declension. These are known
as Nayaka in the Agni Purana. Thus the type of feminine
4-stems was ‘‘sraddha,’”’ of feminine i and i-stems_ that
changed their i and i to y and v before vowels was ‘‘nadi,”
the type of masculine i-and u-stems was agni’’ and so on.
¢

As the earliest treatises on grammar have become lost beyond

recovery, we can only surmise all this with the help of the
scanty materials preserved in K. And even in very late
times HN. uses ‘“‘Krsna” for a-stems, ‘““Radha” for ‘‘Sraddha’”’,
“Hari” for ‘‘agni’” and “gopi” for “‘nadi’’.
«“Sraddha”, ‘“‘nadi’ and “‘agni’’—all these are found in K.
Panini accepted “‘nadif” but used the monosyllable “ohi”
for “‘agni” and rejected “‘Sraddha” altogether as serving no
useful purpose, representing as it does only feminine 4-stems
which can easily be represented by the suffix -ap which has
the great advantage of being a monosyllable. The “gh” of
“shi”? would appear to be taken from “‘laghava’’ and ‘i’ added
to make it typical of i and u-stems. Masculine and neuter
*For greater details see K. C. Chatterji’s Some Technical Terms
of Sanskrit Grammar, in the Calcutta Oriental Journal, Vol. III
No. 5.
+ Feminine 1 and t-stems which change their finals to ily and uv
respectively are optionally designated “nadi” before the genetive plural
ending. Feminine stems ending in i and u, short or long, are also
optionally designated “nadi” before the Dative, Ablative, Genitive and
Locative singular endings.
 SRADDHA AGNI AND NADI 63

stems in i and u and feminine stems in i and u when they

are not termed nadi are known as “‘ghi”. ‘Pati’, though
fulfilling this condition, is regarded as “ghi’’ only when it
stands at the end of a compound.
Kaccayana in his Pali Grammar uses “jha” for masc. ?-
and i-stems, “la” for masc. u- and g-stems, “pa” for
fem. 7-, i-, u- and a-stems and “gha” for fem. 4-stems.
“Sha” and “la” were probably suggested by Panini’s rule “‘iko
jhal” i. 2. 9, “gha” by P&nini’s ‘“ghi” and “pa” by Panini’s
“nip”’.
For “nadi” C. uses “yin”, Sak. uses “nityat” and He.
uses “‘nitya-dit’”. C. uses “idut” for “ghi’’.
Devanandinuses “su” for ‘‘ghi’”. This is evidently a
nice word ending in u, just as “ghi’? ends in i. For ‘‘nadi’’
he uses “mu” modelled on this “su”? with an initial nasal as
in ‘“‘nadi’’.
Sak. follows Panini in the matter of “ghi’, but does: not
make use of any technical term corresponding to ‘éraddha”
or “nadi’’. For “‘dirgha’”’ he uses ‘‘di’’.
Hc., though an ardent follower of K., rejects all these
terms. Compare “‘iduto’ strer idiit” i. 4. 21 £.
Sam. follows C:
Vopadeva uses “ghi” for P&nini’s ‘sut’ (syamaujas ghih
81), but follows Hc. in not having any. technical terms cor-
responding to K.’s “Ssraddha” and ‘“agni”. For “nadi” he
uses “di”, the second syllable. This is a vast improvement
on J’s “‘mu’’.
Su. follows Panini.
PR. retains all the three terms “éraddha’’, “agni’’ and
‘“nadi” of K,
 64 TECHNICAL TERMS OF SANSKRIT GRAMMAR

In several rules of Panini “‘nadi’’ is used in the sense of

the feminine suffix ‘7’ e.g., ac chi-nadyor num vil. |. 80, Satur
anumo nadyajadi vi. 1. 173. In “nadi bandhuni”’ vi. 2. 109,
‘“‘nadi” is used in the sense of feminine i-stems. All this would
lead one‘ to suppose that originally “‘nadi’’ represented femi-
nine i-stems only and that there was a separate term for
feminine ‘u-stems.
For “ghi’” C. has “idut” (vi. 2. 49). For “nadi” Hc. has
“Nitya-dit’’, 7.e., that which always takes on endings with an
indicatory d. Thus for Pdanini’s “nadyrtaS ca” v. 4. 153
He. “‘rn-nitya-ditah” vii. 3. 171. The fact is He. adds an
indicatory d to the substitutes for the nit endings in the case of
the stems designated ‘‘nadi’’ by Panini and others. The idea
of this d was evidently suggested by “‘nadi’’.

qt) qarentadt) tagaeeraraet 1 afr, feta

saa, asrwafa: oa: aaa ta) wWignsefa
at] <lelaes
at) agafa: | Sq ered ast) arse) zIeI¢-g0
+1 Tame gi att amiga) a) fea
oq (seer) | eafaria) 2igeo-2¢0
a | waeaeeaesTasT | Uxie
ql qed) weg! aft) ages feta,
29-700

gq | sfzdt fa: qeremrercfeasig| edt eig anf

gat adi) amfed: agar wie fer fefa geal
The RRBNSG-2z
TF) Aes eaaeal, TATA: VA: | CTA: | RRca-a
a Saag adt faarg| Re"
 SRADDHA AGNI AND NADI 65

El AUTATA: HHYAT 1 Rat WATKAT ahTae: |

Q129¢ | AaTacaASeAT UTTAAT AGIA A) LRgo]
$ H went dittdar acide a1 ueesl (TREAT,
wemi=anitag ) |
Even such an apparently arbitrary technical term has not
escaped the attention of poets, one of whom has utilised it to
make his heroine go into raptures over her lover in the follow-
ing stanza :
Aas aHnfrage Fafrarrarae |
afy afa@ aarqaal @ afa a caateat ce |
‘O my friend, my heart is set on the lover who is young
and handsome, whose qualities of head and heart command
respect and who is a stranger to troubles, and not on the
husband who is the reverse of all these.” The other sense is
—‘my heart is set on the word ‘upapati’ which has got the
sweet designation ‘gh’ (according to Panini’s rule ‘“‘patih
samasa eva*’’ i. 4. 3), im whose case guna is possible
according to ‘‘gher nit’ vii. 3. 111) and which is the
receptacle of the particle na in one case (i.e. in the Instru-
mental sing. according to ano nastriyam vii. 3. 120)’.
This reminds one of a similar stanza found in works on
Subhasita :

afa af aa: afaaq catia fF a a safe |

amisfaakareaatattte ae catty ararfer |)
‘Do you not know, dear friend, that the husband is com-
mended as a friend?’ ‘And is it not known to you, my friend,

* The corresponding rule in K. is patir asamise ii. 2. 2. which has

been made the subject of the following. samasya-piirana stanza:

O.P. 129—5
 66 TECHNICAL TERMS OF SANSKRIT GRAMMAR

that the lover is commended as a super-friend?’ The other

sense is: ‘Do you not know, O friend, that grammarians
direct that the word ‘pati’ is to be declined like the word
‘sakhi’?’« ‘And do you not know that the word ‘upapati’ is
to be declined like ‘ati-sakhi’ ?’
Fortunately such stanzas are rare in Sanskrit.
Regarding ‘agni’ a stanza has been preserved in the
Vitti-traya-varttika :

qf: ARrAAaaaSAAT AAT | Cc ®

aceelfaaaag RAAT RAAATAT ||

‘Sarvavarman’s view that i (the word i or words ending in
i, and Cupid) should be regarded as fire (agni) is also my view:
otherwise, how does it consume the bodies of those that are
in love, causing exquisite torture ?’
Critics have not been slow in pointing out the paradox of
Panini in connection with the technical term ‘“nadi”, Thus
one scholar says:

aiforea a wat TAT AAA aT Ast ere |

my: carcarana afreste Kote aq|)
Vrttitrayavarttika 80.

‘According to Panini neither the Ganga nor the Yamuna js

a river; it is land that is a river. A master possessed of
a

ayy aia a waqela Aare 2aQ gata y

frasia ania te gfaceara aarhyer |
Subhisitavali, 2396.
Sam. also has the rule. yfazgara afqaeter Wie. 127.
 PRATIPADIKA 67

independence does what he pleases.’ Similary we may say of

Sarvavarman :

ata: UTAH ATA aT STAT ATSARTAT |

aaanad fad arafaeariseae |
¢ ° ¢

Though the word “‘stri” changes its final into ‘iy’ before
vocalic endings, it is designated ‘“‘nadi’’ by grammarians.
Poets have been enthusiastic in according their support to ‘this
view. ‘Thus one -bard -observes :
ait aft afee wat caarafrar
Gat |
aa de ford a Head Hoarty |
Vrttitrayavarttika 109.

‘Woman is river. This is a fact, for wild with passion she

brings ruin on both the families (of her father and her
husband), even as the river filled to over-flowing with water
plays havoc with both her banks.’*

PRATIPADIKA
Commentators of Panini do not generally take the trouble
of explaining this term of five syllables. According to some
it is merely for the sake of: varying the monotony of the
monosyllabic technical terms (pratipadikam iti mahasamjiia-

*The idea is very common in Sanskrit literature. Compare :

aqgunaelag’ faites aafad = orafaga t
HACAT Sra: HAAAUMTSATRT | WHAT LR
araraaafraraataar frat g car aaatshrearane |
gaa ela aaa at Raed angi qdla yw fraraw vs
40a qe BERMAN Heh enh avarz |
aay Z1ay frqraaia vat fe qenft Geni ara: Gea Ree
 68 TECHNICAL TERMS OF SANSKRIT GRAMMAR

karanam_ vaicitryartham—-Tantrapradipa i. 2. 45). “It is

out of deference to earlier grammarians’’—rightly remarks
Bhattoji_in his ‘Sabdakaustubha (mahasamjna-karanam _pra-
cim anurodhit). Others hold that it is Panini’s preat vene--
ration for the Vedas that led him to use this big term, for in
the Gopatha Brahmana belonging to the Atharva Veda we find
the questions: What is a Dhatu and what is a Pratipadika ?
(mahasamjfia-karanam Srutyanurodhat. tatha catharvane
pathyate—ko dhatuh kim pratipadikam kah pratyaya iti).
Srstidhara, a Bengali commentator of the 17th century,
derives the word “pratipadika”’ from pratipada ‘every word’
with the suffix -thak, acc. to the rule “padottarapadam
grhnati’’ iv. 4. 39. So pratipadika means that which embraces
every inflected word i.e. words like vrksa etc., which are
inherent in all the different inflected forms like vrksam,
vrksan etc. (pratipadikam iti mahatyah samjfidyah prati-
padanena pratipadam grhnati’ ti vyutpattya ‘“padottarapadam
grhnati’’* (iv. 4. 39) iti thaki sarvesveva padesu yani yani
vrksadini samavayanti tesam eva samjfia).
It is just possible, however, that thak has been added to
‘pratipada’ in the sense of ‘prayojana’ (v. 1. 109) and the
word “‘pratipadika” means that which is necessary for each
inflected word (‘prayojana’ being tantamount to ‘phala’ here)
since no inflected word (subanta or denominative) can be
formed without the help of the pratipadika.
Monier Williams notes that -‘pratipada’ occurs in the
Sankhayana Srauta Sutra in the sense of ‘forming the commen-
*It will be noticed that Panini makes provision for “pratipadika”
but not for “sarvadhatuka”’, "
 PRATIPADIKA 69

cement.’ This sense is singularly applicable

in our case.
The pratipadika merely forms the beginning of words which
cannot be used as full-fledged words unless and until the
case-termination is added.
Panini defines “‘pratipadika’” thus: arthavad adhatur
apratyayah pratipadikam. krt-taddhita-samasa§ ca i. 2. 45-46,
1.€. that which conveys sense and is neither root. nor suffix
(nor a word ending in a suffix) is a pratipadika. Com-
pounds as well as words ending in primary and_ secondary
suffixes are also pratipadikas. Thus words like ‘dittha,’
‘davittha’ etc., which are without any suffix, ‘kartr’, ‘hartr’
etc., formed with the primary suffix tr, Aupagava, Kapatava
etc. formed with the secondary suffix an and compounds like
raja-purusa, Brahmana-kambala etc. are all pratipadikas.
It is curious Panini does not regard words ending in feminine
suffixes as pratipadikas. The mention of ‘nip’ and ‘ap’ in
the rule “‘nyap-pratipadikat” iv. 1. 1 and of ‘linga’ in the
rule ‘“‘pratipadikartha-linga-parim4na-vacana-matre prathama”’
ii. 3. 46 conclusively proves this point.
K. uses ‘“ndman” in the sense of ‘an inflected word’ in
“namna atmecchayam yin” iii, 2. 5 (syadyantam iha
namestam), but “linge” in “ini — lingasyanekaksara-
syantyasvarader lopah’”. In the Saméasa-section, which is
not the work of Sarvavarman, we find ‘“namnam samaso
yuktarthah’’.
Under Nir. 1. 1. 9 Durga explains “naman” thus-
aAeCaTeaTaAes TINT, Aaa at weiaTerage-
area worareafa Aga|
The Pafiji on K. i. 1. 1 explains “linga” thus:
70 TECHNICAL TERMS OF SANSKRIT GRAMMAR

araide aun) fegaa fagudstaneaart aaa

fe fesq, afaererinaifaegeaa | aa ca arreare-
aaisfy fiagean a vata, cea frafiza-arget-eqrasfanre-
aard aa adfga-aararracaaaeare, fag: dat faa |
In the Kavyalamkara-sutra of Vamana “naman”’ is used as
opposed to “dhatu”’:
TSIGAAAMTAS ATAKTSA AAT} ANI
The word “‘linga’’ is used in the sense of ‘a characteristic’
in the Upanisad. Thus it came naturally to mean ‘gender’ and
then, by synecdoche, substance which is the substratum of
gender. Mu. follows K., but intent on having monosyllabic
technical terms shortens ‘“‘linga’” to “‘li.’ J. uses ‘“‘mrd’’ for
“‘pratipadika’’, evidently taking the ‘m’ of ‘Naman’ and the d
of Pratipadika and shaping it after the pattern of ‘krt’ and ‘hrt’
(Taddhita suffix) and at the same time making up a common
word expressive of meaning. In the Candra system ‘“‘gabda”’
is used in the sense of “‘pratipadika.”” Su. follows Panini and
uses the big term, while PR. follows K. and uses “linga.”
Hc., Sam. and HN. use ‘“‘naman”’ for “‘pratipadika.”’
In the earlier systems of grammar the choice appears to
have lain between “‘pratipadika”’ and “‘naman.” P§nini uses
the former, while the Nirukta, Pratisikhyas, BD. and other
works use the latter. A Slokavarttika also says: nama ca
dhatujam aha Nirukte vyakarane Sakatasya ca tokam. The
Mahabhasya, while enumerating the four parts of speech,
mentions naman, akhyata, upasarga and nipata (“namakhyato-
pasarga-nipatas ca’’™*).
pe
espor earch
*This is the reading in every single printed edition of the
 PRATIPADIKA 71

Naaman is defined in the Nirukta as “sattva-pradhana’”’

(sattva-pradhanani namani). In RP. we find “tan nama
yenabhidadhati sattvam’’ and ‘“‘sattvabhidhayakam nama.”
The same thing is expressed in a different way in a welbknown
stanza of BD. (i. 42):

Tesaancate ta Tet TATTS |

aaerchran wai aeEA ae: |)
From another stand-point BD. defines “naman” thus
(i. 43):
sel wa Sasa avast, fern: |
TAT HAT: MEAs aaafsya: |)
The Mimamsa Sitra defines ‘‘naman’’ from another stand-
point: yesam utpattau sve prayoge rupopalabdhis tani
namani ii. 1. 3. ‘“Those words are nouns, on the utterance
whereof in actual usage, the forms of the objects denoted by
them become apprehended.”
Naman corresponds to Greek onoma and Latin nomen.
The word nama-karana is used in the Nirukta in the sense.

Mahabhasya we have come across. This is also found in the printed

edition of Kautilya’s Arthasastra. VP. reads “namakhyatopasar-
ganipatah” without ‘ca’. This appears to be the correct reading here
also. RP. has namakhyatam upasargo nipatah xii. 17 and CA.
namakhyatopasarga-nipatanadm i. 1. In the Nirukta we find
‘“namakhyate co’pasarganipatas ca” i. 2. which is also possible here.
For the confusion of ‘ca’ we may compare Paiicatantra (N. S. Press)
IVs: 22 3 |
wane @ Tara araraaly aa |
| wfq wa, nama caq noe = i
where the correct reading is undoubtedly pram wath 4.
 72 TECHNICAL TERMS OF SANSKRIT GRAMMAR

of a krt suffix. avater gatyarthasya’sau nama-karanah i. 17,

kakso gahateh, ksa iti nama-karanah ii. 2, gaur iti.... gater
vaukaro nama-karanah il. 5, ghaser vero nama-karana usiram
iti yatka- ii. 5, mithunau kasmat ? minotih érayatikarma thu
iti nama-karanah vii. 29, Seva iti sukhanaha, Sisyater vakaro
namakaranah, arthantaropalingi x. 17. The commentators
explain “‘nama-karana”’ as ‘pratyaya’, but from a consideration
of the above passages it is clear that “ndmakarana’” means
‘krt-pratyaya’. The word for a suffix in general in the
Nirukta would appear to be “‘antakarana.”’
It may be mentioned in passing that ‘“nama-karana”’ means
samjha in PR.
‘“‘Nama-vibhakti’”’ is used in the sense of case-ending in
the Nirukta (vii. 1) andgin BD. (namakhyata-vibhaktisu
BD. i. 94). In the Sabara Bhasya (ii. 1. 3) and Wilieyiivina
Bhasya “‘namiki vibhakti” is used in this sense.
Though
Panini uses “‘pratipadika’’ for ‘‘naman,”’ still his followe
rs use
“‘nama-dhatu”’ for denominatives.
It may be mentioned in passing that namin js entirely
different in meaning,
denoting, as it does, the vowels that
induce ‘nati’ or the change of dental s into cerebral] s (nama-
yanti dantyam santam miurdhanyam kurvanti’ti naminah—
Uvata under RP. i. 65). |
In some systems ‘“‘naman”’ is used in the sense
of samjna.
Cf. J. (Mahavrtti) i. 1. 71, Sak. ii, 1. 53 etc. The
Kasikavivarana-panjika says :

a gai waa agiqarar: wear Fqzq xaaniea-

eaTha dar afzaet, aaa anrazaqarst eE:, 7g a
sisafztafirat warfare amnferaeicraaifarag
 PRATIPADIKA 73

It will thus be seen that the difficulty in regard to “‘linga”’

js that it is used in the sense of gender also and in regard to
‘“‘naman”’ is that it is used in the sense of samjna also. So
the good old word ‘“‘pratipadika”’ with its ‘‘learned I¢ngth and
thundering sound”’ is retained in many later works also.
It is not unlikely that originally there was some difference
between “naman” and “pratipadika’. The Gopatha
Brahmana mentions both in the following well-known passage :
aS Tess: Al arg fe ofacfes fF ararendfF
fog fe aaa ar fafa:#: saa: aH: ac scat facta: |
In Bharata’s Na&tyasastra also we find the division into
naman, akhyata, upasarga and nipata and yet pratipadi-
kartha is mentioned in three consecutive stanzas. From ‘“‘ko
dhatuh kim pratipadikam”’ of the Gopatha Brahmana, *“prati-
padikartha-yuktam dhatvartham utsrjanti ye svarthaih” of the
Natyasastra (Chap. XV, ed. Chowkhamba), and “tam
pratipadikarthan ca dhatvarthan ca pracaksate”’ of the Vakya-
padiya it would appear that “dhatu” was opposed to
“akhyata” and ‘‘pratipadika” to “naman”. Now it is not
mere words (Sabda) but inflected words that are divided into
the four classes—naman, akhyata, upasarga and nipata, so
“‘naman’’* must have meant ‘a name with the case-termina-
tion,” just as “akhyata’’ must have meant ‘a finite verb with
the proper ending.” Then “naman” came to signify a name
alone to the exclusion of the ending, even as “akhyata”’
* Namisadhu says in his commentary on Rudrata’s Kavyalamkara
i... 22
aqata ge 4) famed fremarenaq aarenan: aqwaey-
wentatatad fro: 1 franfaiaatactefaaaqqedt: |
 74 TECHNICAL TERMS OF SANSKRIT GRAMMAR

followed the oppostte direction and meant in later times the

verbal ending itself to the total exclusion of the root.
The word “naman” is mentioned by Panini along with
“akhyata® én ‘“‘....namakhyatat thak” iv. 1. 12, but neither
naman’’ nor “‘akhyata’’ is used by him as a technical term
in his Astadhyayi.
Of course Panini could have invented a monosyllable
for “‘pratipadika”, but inspite of his passion for brevity he
uses technical terms of five syllables like ‘pratipadika’, ‘ardha-
dhatuka’ and ‘sarvadhatuka’, of six syllables like, ‘sarvana-
masthana’, of seven syllables like ‘karmapravacaniya’ out of
deference to his teachers or the school he was following, or it
may be. that these technical terms had taken such a firm hold
on his mind that he could not bring himself to discard them
and consciously or unconsciously he made use of them in his
Astadhyayi to the great discomfiture of his commentators,
Santanavacarya in his Phit-sttra uses “phis” for “pra-
tipadika”’, evidently taking the initial consonant of the com-
mon word ‘phala’ and rounding it off with ‘is’.
qr) waaengureaa: facia | sifeaaqarasy
g12134-B%
or | mafaataasnaatega* 2lele (@ very — clever
combination of the two rules of Panini).
#1 waa sgqer ( aedfga, a-ema)
LILIS-&
* oma aafamangalarafada |
algrafaasearg oftea: ofaitad 1
afmaaatta aq ee
 SARVANAMAN >

afag | sarge ag ere stfaateat J aT B4e-2e

2) eswargfaafrataraara e12126
J) Wardeq ct ev
at, afaaie aa “lg but uses fagia q9yHT in the
Karaka section.
Zi warafamasradq orfacfaq 21212
T| xarswarfeqedar faafweoraata: we|ec
el wagfiayafesnararaa (fqqafaefaafe )
RIA Be

SARVANAMAN

Sarvanaman is a good old term and is found from Apas-

tamba Dharma Siitra downwards. Thus Apastamba says:
“‘sarvanamna striyo rajanya-vaisyau ca, na namna” i. 4. 20,
‘women as well as Ksatriyas and Vaisyas are to be saluted
by means of the name common to all i.e. pronoun (abhiva-
daye’ham) and not by (means of one’s proper) name (Deva-
datto’"ham abhivadaye). The contrast between ‘“‘sarvanaman’’
and “naman’”’ here brings out the original sense of the term
‘““sarvanaman’’*,
The word also occurs. in the Nirukta:
ag faq ca tarey Sat: a fa faq-senent area
fag:) ane afaq ofarfeafaata arama, Zanel-
afefa qagta | ana aes
ferreting | aura afq
feats’ srfaftafirerata aareente at enter ae,
* With K.’s definition of “linga”, cf. Vaiyakarana-bhiisana-sira (ed.
Bombay Sanskrit Series, p. 482) :
 70 TECHNICAL TERMS OF SANSKRIT GRAMMAR

anfy cate attr daarcoaMdaaasaaa AAAI

sala zaeeaag
Medhatithi says under Manu ii. 122:
‘< aat fa aaa aafaaasfarsay |
Does Yaska use “samkhya-naman’”’ for numerals in Nir.
iv. 6: faaetinaat amt aya) aft ar qeqtaaarhy-
td ear| Tad fafea fa aat az: |
Durgadasa says in his com. on Mu.: sarvesim buddhis-
thanam nama. The Prakriya-Sarvasva says:
Aaraal FeeIAl: BSZATATs war |
aaeqaraaran fe aareeate area: |)
<q ofa fafeaaraet saan | aaa | ens
anfafa cae aaanirgztagq | 4123
garaifiaea saagevan:, aff aa as.
aTAT | «(SIX
It occurs in APr. also:

afafa fama:| faaranmy era) aama

gaqa Al ess
Here also the contrast between “‘sarvandman”’ and ‘‘sukha-
naman” arrests attention.
The Mahabhasya (i. 1..27) explains the significance of the
name thus:
acqaaen aa fara | aaa war aq adam I.
aad ane aqe saatgeaate smf, cat arfy
qamanned a a faraafiiafan faaa Fz: qIeST
ada |... aq adaaaaa frat grasa fetes dar a |
Whitney explains the full significance of the term very
clearly under CA. ii. 4: “The term sarva-ndman ‘all-name’,
 SARVANAMAN <3

used to denote a pronoun, is an ingenious and interesting

one....Naman, ‘name’, includes substantives, adjectives and
pronouns; but while the two former, being descriptive of
quality, are restricted in their application to certain objects or
classes of objects, a pronoun may be used of anything in-
differently ; it is a title of universal applicability.”
Under Panini vi. 4. 174. 4 Patafijali says: ‘‘ekaSrutih
svara-sarvanama, yatha napumsakam linga-sarvanama’’ where
sarvanama is used in the sense of ‘general’. Similarly we
find ‘‘sarvanamya”’ used in the sense of ‘universality’ in ‘“‘prakrti-
sarvanamyac ca’ in the Nirukta (vil. 4).
It is to be noted that in sarvandamasthaina, sarvanama is a
Karmadharaya (sarvam nama), whereas as a technical name
for the pronoun it is a Tatpurusa (sarvesim nama).
K. uses ‘“‘sarvanaman” without defining it: sarvanamnas
tu....li. 1. 43, smai sarvanamnah ii. 1. 25. The _ taddhita
form ‘‘sarvanamikman” is used in the rule ‘‘nanyat sarvana-
mikam”’ ii. 1. 33.
J. takes the first sound s of the first element ‘sarva’ and: the
first sound n of the second element ‘naman’ and forms the
technical term “‘snv’.
Candra (ii. 1. 6), Sak. and He. use “‘sarvadi’’? for ‘“‘sar-
vanaman’’. This sometimes leads to curious sentences. Thus
in the vrtti on “‘na sarvadih” Sak. i. 2. 170 we find “‘dvanidva-
samase sarvadih sarvadir na _ bhavati’’ where the first
“sarvadih” means ‘words beginning with ‘sarva’ in the
sarvadi list? and the second means ‘sarvanaman’,
Vopadeva follows in the foot-steps of J. but concentrates
on the first member of the compound from which he extracts
 78 TECHNICAL TERMS OF SANSKRIT GRAMMAR

s and r and like J. adds i, the favourite device for rounding

off a name. So the name for a pronoun, acc. to Vopadeva, is
“‘sri’’. Kramadisvara defines ‘‘sarvanaman”’ in the Karika:
c— . C
aaa: aaaa eararearfest
fa aera: |
dra feta qaifr]efastiateaa: Gat |) @ e1203
Su. and PR also use “sarvanaman’’. As usual HN.
substitutes ““Krsna”’ for “‘sarva’’ acc. to the dictum ‘“‘Krsnastu
bhagavan svayam’’ and uses “Krsnanaman” for ‘‘sarvaniman.’
qt) waaiatfa aaanntfa 212129
a fer: aatfe: ciewe! =( aafe adara—aerata
21413%)
gq) aafaeanin:.--co
q| aadifa ara enatsdarargq 2212
7| Aq a ATARETT:
aarrasa fac fagtatsaieraqear |
AAA: aaaaita AAT TA amttcar: I] RIR&S?Y
Z| aadtifa sears 312 ee
Some modern writers use “‘pratinama’”’ which is not found
in Sanskrit but which appears to have been coined as an
equivalent to ‘pronoun’ after the analogy of ‘pratigabda’,
‘prativakya’, etc. Santanavacarya, the author of the Phit-
stra, uses “‘sit”* for “‘sarvanaman,” taking ‘the first letter s
and adding if to round it off. Similarly some writers use
‘“‘sadi’’ for “‘sarvanaman’’.
Sanskrit sarvandman is more acceptable than Greek
antonumia or Latin. prondmen meaning ‘a word used for or
* Hence the reading “Sit” is to be discarded.
 AKHYATA 79

instead of a noun’—a definition which leaves much to be

desired. As Jespersen says in The Philosophy of Grammar :
“This does not suit all cases, and the definition breaks down
in the very first pronoun; it is very unnatural ~ the un-
sophisticated mind to say that “I see you” stands instead of
“Otto Jespersen sees Mary Brown,” on the contrary most
people will say that in Bellum Gallicum the writer uses the
word Caesar instead of “I”. We may also say “I, Otto
Jespersen, hereby declare....’’, which would be preposterous
if ‘I’? were simply a substitute for the name. And grammati-
cally it is very important that “I” is the first person, and the
name is in the third, as shown in many languages by the
form of the verb. Further: no one doubts that nobody and
the interrogative who are pronouns, but it is not easy to see
what nouns they can be said to be substitutes for.”

AKHYATA
“A-khya-ta” literally means ‘that. which has been des-
cribed fully’, then it came to mean ‘that which describes
fullyt i.e., the finite verb’. The root with the upasarga is
—_—_—

+“The power of expressing Time by those variations which we

call Tense (after an old form of the French word for time), has
attracted notice as the most salient feature about the verb. Aristotle
defined a verb as a word that included the expression of Time. The
established German word for a verb is Zeit-wort, that is to say,
Time-word. Others have thought that the power of expressing
Action is the real and true characteristic of the verb. Ewald, in his
Hebrew Grammar, calls the verb accordingly That-wort, that is to
say, Deed-word. But in these expressions the essential is obscured
-by
that which is more conspicuous. The Danish scholar Madvig, in his
 80 TECHNICAL TERMS OF SANSKRIT GRAMMAR

preserved in akkhna ‘to say’ in Panjabi where we can except a

continuity in Janguage from the earliest period and which,
in fact, preserves many ancient forms not found in other Indian
vernaculafs. In the Nirukta (i. 1) and BD. (ii. 121) Akhyata
is defined as “‘bhava-pradhanam”’ 7.e., ‘having becoming or
action as its fundamental meaning’. RP. xii. 19 says: (tan
nama yenabhidadhati sattvam) tadakhyatam yena bhavam
sadhatu j7.e. that is a verb by means of which one denotes a
becoming or an action and which contains a root. Some
commentators apparently read ‘sa dhatuh’ for ‘sadhatu’
and identify Akhyata with Dhatu. Kautilya defines Akhyata
thus: avisistalingam akhyatam kriyavaci (Arthaéastra,: ij. 10.
28) i.e. the Akhyata shows no distinction of gender and denotes
action. In the printed Natyasastra the definition of akhyata
would appear to be as follows: sampratyatita-kalaprayojitah
kriyadi-sammyogah prathitah xv. 24 which reminds one of
Yaska’s “‘purvapari-bhutam bhavam akhyatendcaste”’ j.e. a
becoming or action* arising from a former to a later state is
denoted by the verb. In the Mimamsa Sutra, Akhyata is
defined thus: yesam tUtpattav arthe sve prayogo na vidyate
tanyakhatani ii. 1. 4. “Those words, on the other hand, are

Latin Grammar, seems to put it in the right light. He designates the

verb as Udsagnsord, that is Outsayings-word; because it ‘udsiger
om en Person eller Ting en Tilstand eller en Virksomhed, outsays,
pronounces, asserts, delivers, about a person or thing a condition or
an action. Jt is the instrument by which the mind expresses its
judgments, or (in modern parlance) makes its deliverances.” FARLE,
The Philosophy of.the English Tongue.
*See under Upocegraha. Cf. also BD. i. 31:
 AKHYATA 81

verbs, on the utterance whereof what is denoted by them is

not in existence.” This reminds one of Bhartrhari’s Karika+
araq fagafad at araaafienae |
aTisamaeqcary fare saqrsae ||
Under “‘sa-vigesanam akhyatam vakyam’’’i. 1, 21 He. says
“tyadyantam padam akhyatam’’ which is explained thus in
the Laghunyadsa: akhydyate’ nena kriya-pradhanatvena
sadhyarthabhidhayitaya vetyakhyatam,
tac ca tyadyantam iti.
The word is not used by Panini as a technical term in the
Astadhyayi though it occurs in P&nini’s rule- dvyaj-rd-
brahmana-rk-prathamadhvara-purascarana-namakhyatat thak
and in the Ganasttra ‘“‘akhyatam Akhydatena kriyasatatye”’.
Yaska uses ‘‘akhyata”’ in the sense of ‘finite verb,’ but
his predecessors appear to have used the word either exclusively
in the sense of ‘root’ or both in the sense of ‘roots’ and ‘finite
verbs’. Thus in “‘tatra namany akhyata-janiti Sakatayano
nairukta-samayas ca”’ i. 12 where Yaska appears to be quoting
from Sakatayana, Akhyata is used in the sense of roots. It is
interesting to note that in later times ‘dhatu’
was substituted
for ‘akhyata’ here and we find in the Mahabhasya : nama ca
dhatujam aha Nirukte vyakarane Sakatasya ca tokam (ili, 1.
1). In sarvanyakhyatajani namani etc., Yaska merely follows
Sakatayana. Again in “‘tribhya Aakhyatebhyo jayata iti
Sakapunih” vii. 14 where the view of Sakapini viz., Agni is
derived from three roots, is quoted, Akhyata seems to have
been used in the sense of roots. In all other places viz. catvari

MRA Wa 4 8 arf fTRE |

araa aaah aang salfa ada: |
O.P. 129--6
82 TECHNICAL TERMS OF SANSKRIT GRAMMAR

padajatani namakhyate copasarga-nipatas ca 1. 1, purvapari-

bhatam bhavam akhyaten4caste i. 1, udattam tvevam akhyatam

abhavisyat vi. 28, atha paroksa-krtah sarvabhir nama-
vibhaktibhir yujyante prathama-purusais cakhyatasya vii. 1,
“Skhyata’’ appears to be used in the sense of ‘finite verb.’
“Akhyata”’ is used in the sense of a finite verb in the
Gana-sitta “akhydtam akhyatena_ kriya-satatye’ (tinantam
,

tinantena samasyate kriya-sdatatye gamyamane—Nyasa) under

‘“mayura-vyamsakadayas ca” il. |. 72.
Jagadisa says in the Sabda-Sakti-prakasika 97: tinakhyatayoh
paryayatvat.
Since in ancient times roots were denoted by means of Stip
“akhyata’’ may very well mean ‘root’ also.
In the Katantra the third chapter is known as Akhyata, but
the itself does not appear
word to have been used ‘by
Sarvavarman. Commentators generally explain the word
thus: akhyayante mukhyataya pratipadyante yani tani
akhyatani tyadini.* According to this view ‘“‘akhyata’’ means
the verbal endings. In the sense of the finite verb, ‘takhyata”’
is explained thus: 4 samyak pradhanyena nirakanksatvena
sadhyatvena va khyatah pratipaditah pacanadiripah kriyah
kala-kartrtradi-rupo’rthaS ca anena_ ityakhyatam paeatiya
dirtpam.
Similarly the chapter on conjugation in HN. is known as
the akhydta-section, and the word “‘akhyata’’ occurs several]
times in the rules though it is not defined anywhere.
_-

eur aaa sala fama wifeat eaifa aq aznaiqa) saa

es _ e ’ . oe.
giania waa aq aigiefaaeiar | |aeaifa: |
 DHATU 83

PR. uses “akhyata’’ in the sense of ‘verbal ending’ as

is clear from its rule: vibhakti-taddhitakhyata-krtag ca pra-
tyaya matah i. 172, #.e., the case-endings, the secondary
suffixes, the verbal endings and the primary suffixes’are known
as pratyayas.
In the Kadambari (para 85) both “akhyata’’ and “‘kriya”’
are mentioned in the same compound: aTHcnfAq TTH-
merase. fenrhaernaare arcana aera fea
TTA ATASTAT |
DHATU.
Dhatu (from dha ‘to lay, put’) originally meant ‘lay-er,
constituent part’, then it meant ‘etement, primitive matter
(the five Mahabhitas—ksiti, ap, tejas, marut, vyoman or the
constituent elements of the body—vayu, pitta and kapha, or
rasa, asrk, mamsa, medas, asthi, majjan and gukra, or the
five jnanendriyas or the five tanmatras viz. gandha, rasa,
ripa,
sparsa, Sabda)’. In grammar it means ‘elements or essenti
al
ingredients’ of words and is first found in the Gopat
ha
Brahmana.* In the Nirukta it is correctly derived from the
root dha: etavantah samana-karmiano dhatavah dhatur
dadhateh, i. 20. It is found several times in BD. (ii.
101—104) :
fag” ag’ faaft a aatq aa aa a
4G 4g Taare Wes ATE Haig SARE II
* Dhatu is found in Chandogya U. vi. 5. 13 ¢f.-
aaaiad au fadaa) aw a: wfag} aga gieq wala, A ae
aMaaa, Tsfusaraa: |
 84 TECHNICAL TERMS OF SANSKRIT GRAMMAR

aTqaiaa ada fag’ afend wa |

aqsarafaaa: carats wax: |
mqreniaranas fears |
qa nataa att az farang ||
MIA AGAsad aAMwtAAg aT |
arpa safaxiot a faaizad gers oq 1)
“Gender, root and inflexion, one should, in their respective
places, adapt (to the sense). Whatever is Vedic in a formula
one should turn into everyday speech (laukika). The
analysis of the secondary elements (guia) may be (effected)
by the aid of all roots (dhatu) which possess a_ traditional
characteristic form (lingua), and the sense of which can be
stated. A word (pada), the definition of which can be
expressed, whether it be derived from two roots, many (roots),
or one root, is one consisting of a sound (sahbda) that contains
root, preposition, members (¢vayava), and secondary ele-
ments (guna). A word may be explained in five ways (viz.)
as derived from a-root, as derived from the derivative of a
root, as derived from a compound meaning (samastartha),
as derived from a sentence (vakya). and as (of) confused
(derivation).”
The term ‘‘dhatu”’ is used by Panini and retained in most
systems of grammar. J. takes the ‘dh’ of the first syllable
and ‘u’ of the second and combines them into ‘dhu’. Mu.,
as usual, follows J.
It may be mentioned in this connexion that roots are
generally represented by their past participial forms in AB.
Thus a formula containing “‘Aapyayasva”’ is described here as
 DHATU 85
“a@pina-vat”, another containing “‘ariirucat” as “rucita-vat”
and so on. No doubt in several cases roots are denoted by
verbal nouns etc., but this is the more usual way. (Liebich,
pp. 15-17). This, as Liebich points out, gave risz to the
suffix ‘ktavatu’ in latér Sanskrit.
Panini does not define ‘“‘dhatu” but merely names two
important roots—-one of the first conjugation and one of the
second, and says that roots are of the nature of these. His
rule is “bhi-vadayo dhatavah”
i, 3. 1. A Sloka-varttika
explains that the v in the rule is for auspiciousness* : bhii-
vadinam vakaro’yam mangalarthah prayujyate iii, 1. 5S,
The second half is found in the Ka&asika: bhuvo vartham
vadantiti bhvartha va vadayah smrtah, i.e., bhiivadi is ™ be
derived with the aunddika agent suffix thus- bhuvo’rtham
vadanti iti.
It is just possible that in some ancient Dhatupatha va stood
at the head of adadi roots and so some predecessor of Panini
framed the rule “bhiivadayo dhatavah” which was taken over
by Panini.
Katyayana says: kriya-vacano dhatuh i. 3 | which is
sufficiently clear. K. says: kriya-bhdvo dhatuh ii. 1, 9. which
commentators explain thus: yah §gabdah kriyam bhavayati
partipadayati sa dhatu-samjiio bhavati. Thus it becomes iden-
tical with Katyayana’s definition. It is just possible, however,
© yaratat aariisd Baa: nyse |
saute eaaagqaarfig y
BA are aeafa TSCA vf |
qarea sla sa waist areasaar j
J. i. 2. 1 foot-note.
 86 TECHNICAL TERMS OF SANSKRIT GRAMMAR

that we have here nouns from the two very general roots—
‘to do’ and ‘to be,’ in the words of Sanskrit grammarians:
parispandana-sadhana-sadhya gamanadi and a-parispandana-
sadhana-eidhya avasthanadi (Rama Tarkavagisa’s commentary
on Mu. 11), and that the author intended us to understand
‘that which denotes action and that which denotes existence
are roots’.
C. uses “kriyartha” for “‘dhatu’. Cf. C. i. 1. 40 with
Panini iii. 1. 22. Sak. often has ‘“‘edhadi’’. because his
Dhatupatha begins with the root edh.
In the Unadisttra ‘“‘mithune’ pi piirvavac ca sarvam” v.
222 ‘mithuna’ is explained by Ujjvaladatta thus : yatropasargo
dhatu-kriyaya yuktas tan mithunam.

QT) warzar ayaa: | 413%

meq) «fRarraat sag: | 21312
mT) |feeaTaaT ATT: 1 BLN
Ht WaTeaT ue
st] feparai Tags) |21g IRe
21 fart arg: 1 31313
ar) varfezarron freeararara rave: | zy
ql warty a: | 2% ( fer=arrag: | )
Bi warhe-qareaeat arearag: | are,
T | GAA AAT: 1 g1ZGe
S| WAAR AAA: | GIRS
The Taddhita suffix “dha” is probably a_ broken down
part of “dhatu.” In RV. ‘tri-dhatu’ means containing three
elements i.e. three-fold.
 DHATU R7

‘““Dhatu”’ in its technical sense is used by Kalidasa in his

Raghuvaméa :
TAM TATA Gat aearefasa |
waar ararcafeaniad |
‘The army which followed him at the command of Rama
served to fulfil his mission, even as the preposition ‘adhi’
prefixed to the root (7) meaning ‘to study’ serves to bring out
its sense’ (xv. 9).

@ gear atte alreaeq? farcaifeaa |

Wat: Ula Taeat arild aaa ||
‘The hero slew Valin and set Sugriva in what he had long
coveted, viz., his brother’s place, like one root substituted in
place of another’ (xii. 58)
After 1. 18, add: In Kuttanimata 12 we have
“qaafzg qa ‘“Aearfecaq” ‘where speaking of dhatus
is found only among grammarians (and not among
alchemists)’.
frase Gat artaat Garadianft goectse |
Pensraas @ TASER HOfarftag adanrqa
Attributed to Bhatti in the Sabdakaustubha

qHscaaarteniaaa aeaacae TTSRTATATA |

WME Wal CTIG saHOra
facafiag WAIT ||
Quoted by Vidyasigara under K. ii. 4. 13.
In the Nandikesvara kasika (Mbh. N. S. vii. vol. I, p. 133)
in used :

AA TAA FAY HAI ATT |

arent’ agqaize’ aftremsefasa. |
where eqytzqey’ would appear to be a send his son for opvegy’
 ABHYASA AND ABHYASTA

“A bhyasa” is formed exactly like samasa from the root ‘as’

preceded by ‘abhi’ meaning ‘to throw towards, repeat’ and
means ‘repetition’ in Katyayana Srauta Sitra: abhydasa-
nupapattau jyotistomah piiranah xxiv. 7. 18 and the Nirukta:
abhyase bhiyamsam artham manyante, yathaho dar§anij-
yaho darSaniyéti x. 42. ‘“‘Abhyasa” is used in the sense of
‘repetition’ (of rk) in ASS. iii. 1. 11 “anabhyadsam eke” (pra-
thamottamayor anabhyasam ity arthah—Com.). See also
ASS. vii. 10 6, viii. 1. 13 etc. In Panini also itis used in
this sense in “‘mithyopapadat krno’bhyase”’ i. 3. 71 where Jaya-
ditya explains: abhyasah punah punah karanam, Avrttih.
From this it easily passes on to its technical sense of ‘redu-
plication’ which is first found in the Nirukta: kakso gahateh,
ksa iti nama-karanah, khyater vanarthako ‘Dhyadsah’ ii, 2,
cakadrati kadratiti sato ‘narthako’ bhyasah ii. 3, Ading-
bhyasenopadham Aadatte. babhastir atti-karma. iv. 12.
‘“‘Abhyasta”’ in the sense of ‘reduplicated’ is also found for
the first time in the Nirukta: rarano ratir abhyastah ii. 12,
erire itirtir upasrsto *bhyastah iv. 23, rarivan ratir abhyastah
iv. 25, sararukam’ va syat sarter abhyastat vi. 3. In the
sense of ‘repeated’ it occurs in ‘“‘ayutam niyutam prayutam
tattad abhyastam”’ iii. 10.
In Panini the term ‘‘abhydsa’’ is restricted to the redu-
plicated portion of the root (purvo’bhyasah vi. 1. 4) and
“abhyasta” to the root along with the reduplicated portion
 ABHYASA AND ABHYASTA 89

(ubhe abhyastam vi. 1. 5.). K. uses the term in exactly

similar senses (11. 3. 4-5).
For “‘abhyasa’’ Devanandin uses ‘ca’ (iv. 3. 11) and for
‘“‘abhyasta”’ ‘tha’ (iv. 3. 9). ‘Ca’ means ‘addition, repetition’
and is a good substitute for “‘abhydsa’’, ‘tha’ probably repre-
sents the final syllable of ‘abhyasta’ in its Prakrit form (cf.
thara for stara).
Taking the hint from Panini’s rule ‘“‘piirvo’bhyasah’’’ vi.
]. 4, Sak. uses ‘purva’ for ‘abhyasa’ and thus saves one syllable.
(Cf. “‘purvasvasve’ciyuvyoh” iv. 1. 76 corresponding to
Panini’s “abhyasasyasavarne” vi. 4. 78). For ‘“‘abhyasta”’
following in the footsteps of Candra he uses ‘‘aj-jhi’? which is
explained thus in the Vrtti on i. 2. 14: ad bhavati jhir yasmat
(acc. to i. 4. 8&9). Thus, for Panini’s ‘*sij-abhyasta-vidibhyas
ca” ii. 4. 109, Sak. has ‘sivid-ajjher abhvah’ i. 4. 107 and
for Panini’s ‘‘Snabhastayor atah” vi. 4. 112 he has ‘“$naj-jher
atah’’ iv 2. 47. Sak. also uses ‘dvyukti’’ for ‘abhyasta’. Cf,
i. 4. 89.
Candra uses “‘dvir-ukta’” for ‘‘abhyasta”’. Thus ‘for
Panini’s “ad abhyastat” vii. 1. 4, Candra_ has. “‘dvir uktad
at” 1. 4. 4. C. uses si-jhi also. Thus for Panini’s APIA
vil. 1. 75. C. has AISA Nitd: V. 4. 32. The reference
here is the rule fRemrzq i. 4. 4. For “abhyasa” he uses
“dvitve purva’’. Cf. “dvitve pirvasydsame” v. 3. 84 cor-
responding to Panini’s “‘abhyasasyadsavarne”’ vi. 4 78.
Vopadeva uses “khi” for ‘‘abhydsa’’. Thus the rule cor-
responding to Panini’s “‘abhyasasyasavarne” vi 4. 78 is “‘kher
yor iyuv arne” in Mu. ‘Khi’ is evidently the ‘kh’ of ‘khanda’
 90 TECHNICAL TERMS OF SANSKRIT GRAMMAR

with a euphonic ‘i’. For ‘abhyasta’ Mu. has ‘dvi’, the first
syllable of ‘dvirukta’.
He. follows Sak. and uses “pirva” for “abhydsa’”’, e.g.,
“ane-hi-haao ho ghah purvat’ iv. 1. 34, ‘Pirvasyasve svare
yvor iyuv” iv. 1. 37. For “‘abhyasta’’
he uses “dvyukta’’,
e.g. “dvyuktopantyasya Siti svare’’ iv. 3. 14.
Su., Sam., and PR. use ‘“‘abhyasa”’ and ‘‘abhyasta’’.
HN. uses “nara’’ for “abhydsa’’ and ‘Narayana’ for
“abhyasta”, probably because “Narayana” contains a larger
number of syllables including ‘na’ which is also the first
syllable.
In the Unadi-sutras “‘abhyasa’’ is used in the sense in
which it is used by Panini; e.g. pr-vrhiam dveruk cabhyas-
asya iv. 55.
From PR. we learn that “‘pran” was another name for
‘“abhyasa’’.

QT) Gareara: | Sa Aare eRe,

I) GReara: | saya] zIBIle-%
H1 FT siaier, &@ 2ale (82 aga waz
ta:—a fa: | )
J | a: qa: fia: | Re ( feam-AaINTT fz: 219% ) |
1) ad Ta cere Heat! eae) enaqear:
qraanaa(s)fa sziarnmreisar: | <0) (akira:
wafgaaa aaa: | afa:) ate HOt raged
Rafat: | caprenrfernt | )
Q 1 TabAT | Te ara: | eae | Gaareare:
qumaectafa sq: | Tha: |
 PRATYAYA

‘“Pratyaya”’ is derived from prati-i ‘to go towards or

against’ and means ‘that which follows’ and is used in the
Pratisakhyas and their commentaries in the sense of ‘subse-
quent word or sound’ (sparse cosah pratyaye pirvapadyah”
RP. i. 80) ‘usah’ forming the first member of a compound
is designated ‘riphita’ when a mute follows: “Medhatithau
Varunanta-vratantau sparsSantasthapratyayat’ nirhrasete’’ RP.
90—in the Medhatithi-hymns compounds ending in ‘Varuna’
and ‘vrata’ are shortened, when a mute or semi-vowel follows;
‘““sparsah sve pratyaye purvantasasvara bhavati’—commentary
on RT. 25). In VP. the word is used both in the sense of
‘following’ and in the sense of ‘a suffix’. Thus in “pratyaya-
savarnam mudi Sakatayanah” iii. 9—‘§akatayana prescribes
the change of a visarjaniya into the homogeneous letter of
the following sound when a sibilant follows’, ‘“‘pratyasavarna”’
is used in the sense of ‘parasavarna’. In ‘“‘hrasva-vyafijana-
bhyam vakaradau vibhakti-pratyaye” v. 13, however, “‘vibhakti-
pratyaye” means ‘a suffix which is a case-ending’. In CA.
‘“‘pratyaya’’ is used in the sense of a suffix. Cf. namika-
rephat pratyaya-sakarasya ii. 86, vyavadhav apratyaye iii.
3. But in the commentary it is used in the sense of subse-
quent also : ,

GaQIel ATA AAR TH |

Teas Aa BARAT GawfAazy y ep0
In TP. “pratyaya’’ is used in the sense of augment in
‘“akurva iti ca pratyayat parah” v. 7. in Panini we find the
 92 TECHNICAL TERMS OF SANSKRIT GRAMMAR

word used in the specialized sense of a suffix exactly as ‘‘aprkta”

is used in reference to a suffix. In the Srauta_ sitras
and Nirukta the word is used in the sense of conception,
assumptian, notion, idea. ‘“‘Artha-pratyaya’”’ occurs in the
Nirukta (i. 15) in the sense of ‘certainty of meaning.’
In the Gopatha Brahmana i. 1. 26 quoted under KRT,
““pratyaya’’ appears to be used in the sense of ‘case-ending’.
From the Pratisakhyas it would appear that ‘“pratyaya”’
at one time included suffixes, prefixes, infixes and augments.
‘“ Pratyaya’’ is used in this sense only once by Panini in
“anudit savarnasya capratyayah” i. 1. 69.
Hence the objector
urges in the Mbh.: atyalpam idam ucyate—apra-tyaya iti,
apratyayadeSatit-kin-mita iti vaktavyam.
The etymological sense is thus shown in Mbh. iii. 1. 2.8%
yas tam artham sampratyayayati sa pratyayah.
‘‘Anta-karana”’ and “upabandha” would appear to be the
earlier names for a “pratyaya.”” Both are met with in the
Nirukta.
While explaining “kim punar dvigu-samjiiad pratyayot-
tarapadayor bhavati’’ under ‘“‘taddhitarthottarapada-samahare
ca” il. 1. 50, Kaiyata says: “‘Kasakrtsnasya ‘pratyayottarapa-
dayoh’ iti sutram.’’ From
this it is clear that KaSakrtsna also
used ‘“‘pratyaya’’ in the sense of a suffix.
All the different systems
Sanskrit. of
grammar use
“pratyaya” for a suffix, with the exception of J. and Mu. J.
uses the second syllable “tya’. for “‘pratyaya” and Mu.
follows J.
K., Sam. and Su. use “pratyaya” without defining it.
That to which the suffix is added is known as the ‘“Prakrti.””
 PRATYAYA 95

‘“Prakrti’ has the sense of ‘original form’ in the Nirukta,*

Pratisakhyas and Panini. Sam. uses ‘Prakrti’ in this sense in
“prakrtivad dvir. gunadi§ ca” ii. 566 etc. In ii. 289 mila-
prakrti is used. Goyicandra explains it thus: Prakrti-pedena
yasmat pratyayotpattis tasyaiva grahane siddhe miila-prakrtiti
krtam—yasmad dhatoh sabdah vyutpaditas tad-grahanartham:
dhatur eva Sabdasya miilaprakrtih, dhatu-jatvat sarvasab-
Raghuvamsa :
It would appear that the words *“Prakrti’”? and Milaprakrti”’
as used in grammar are borrowed from the Samkhya philosophy.
Two of the greatest poets of India‘ have made use of these
two technical terms in their poems. Kalidasa says in his
Raghuvanésa :
at aUPaTgal SIA a area Hartany|
aisaag ATPTaAATTA: saa aaaTafan: |
“The princesses attained the object of their being on being
united with the princes, as did the princes by being joined
together with the princesses. The union of the bride and the
bride-groom was like the combination of the base with the
affix’? (xi. 56).
Bharavi has the following stanza in his Kiratarjuniya :
a wae waned: faaracds arerat: werd |
Fegure aera wet cafaseraafrargara: ||
‘The enemy came for his own overthrow between Siva, the
o-ooOr

"Cf. Nir. i. 2: sala aaa wats wear, famaq aay) ( arai-

Traaqewlaa a: sala: AT waa: | ATE aera a: Fadia: a1 mata: | et )
The Katantra school uses ‘“prakrta” for “anga” of the P&nini
school. |
 94 TECHNICAL TERMS OF SANSKRIT GRAMMAR

sole cause of the end of creation, and Arjuna of the white

steeds, both about to accomplish their object simultaneously,
even as the indicatory letter comes for elision between the
base ahd the suffix which together express the meaning’
(xii. 19).
It may be mentioned in passing that though Bharavi is
indebted to Kalidasa for the simile, his borrowed light is dim
and obscure before the effulgence of the original. In Kalidasa’s
simile the masc. ‘pratyaya’ and the fem. ‘prakrti’’ are the
upamanas for the bride-groom and the _ bride respectively.
In Bharavi’s simile the upamana ‘prakrti’ is fem., but the
upameya ‘bhava’ is masc., and since the sense of the ‘praktti’
is subordinate to that of the ‘pratvaya’ Siva becomes inferjor
to Arjuna. Moreover, in. most cases the anubandha comes
at the end of the suffix and not between the base and the
suffix.
Bhattoji quotes the following stanza in his Praudhamano-
rama under “kim-idambhyam vo ghah” 184] -

sfraafa atfena sea oraaray

Taata frat a oRasta saz |
ate qaqa faa: Tara ze
afzateta fata et zat ofeearsfr |
‘When the supreme knowledge which has its spring in
scripture arose and the visible world created by Maya or
Illusion melted away, lo! there arose an entity which is pure
knowledge. Who, however learned he may be, can now
measure its extent ?’
When the suffix (gha) which comes after the base and
 PRATYAYA 95

which is prescribed in grammar was added (to idam), and

when everything belonging to the base was elided, lo! there
sprang into existence a word which is solely a sufhix. That
word is ‘iyat’. Who, however learned he may be, can
thoroughly understand and explain this ?’
The suffix-vat is added to ‘idam’ and the va of vat is
changed into gha (i.e. 1ya) acc. to “kim-idambhyam vo ghah”’
v. 2. 40, so that we get idam iyat. At this stage idam is
changed into i acc. to idama iS v. 3. 3 so that we get i iyat.
The 1 is now elided acc. to “‘yasyeti ca’? vi. 4. 148 so that
the whole base is elided and only the suffix remains. This
is acc. to the prakriya of Panini. The commonsense view is
that the stem here is i and the suffix yat. Followers of Panini
however, are inclined to regard kiyat, adhuna etc., as mere
pratyayas, just as in English we have ‘bus’ which is an abbre-
viation of omnibus, the dative plural of Latin omnis and
which corresponds to the Sanskrit case-ending -bhyas. Similarly
in the Gathic Avesta we have ‘ta’ (Yas. xlvii. 3) for Sanskrit
‘pita,’ in the Y. Avesta we find tuiryo (Vis. xii. 15) for Skt.
pitrvya.
Panditaraja Jagannatha has a= similar stanza in the
Rasagangadhara :

TTT? Gt afeneatay ta: TATA |

aay afgiccard 3g ae ARTE |
‘We worship that Brahman in whom, the Supreme Being
consisting of pure thought, there is neither any attribute
nor any growth or development, who is known among the
wise as the existent or eternal one.’
 96 TECHNICAL TERMS OF SANSKRIT GRAMMAR

There is a subhasita:

cheats ag aerate art fear ore |

aaa: Tatathrat saan fafraae arte 11
‘People here regard Hari and Hara
as different and
dig their own graves, (vinasa astram) though there is nothing in
the Sastras to warrant this. There nature (root) is the same but
appears different owing to difference in knowledge (suffix).’
Pratyayas, as a rule, come after the base to which they are
added. Hence Panini frames the adhikara-rule “paras ca”’
immediately after the rule “‘pratyayah” iii. 1. 1. Hence Bana-
bhatta uses the parisamkhya “pratyayanam paratvam’’ in his
Kadambari (para. 47). The only exception is the suffix
‘bahuc.” Parini says: ‘“‘vibhasé supo bahuc purastat tu’’ y,
3. 68, but Candra (iv. 3. 54 Vrtti) and his followers refuse
to regard ‘bahuc’ as a_ suffix. ‘Bahu’ can easily have the
secondary sense of ‘to a large extent’ (cf. Germ. fast and
Fr. assez)and both the position of ‘bahu’ and the accent along
with the gender of the word show that it is a compound of
‘bahu and a following word. In the Madhyama Vyayoga
attributed to Bhasa we find “riipam sattvam balam caiva
pitrbhih sadrsam bahu” (39) where ‘bahu’ is used separate-
ly as also “janayati ca mano-jvaram Svaro’yam__ bahu-
sadrso hi dhanafijaya-svarasya” (25) where ‘bahy’ has been

* also faaimreqq Weapon of destruction :

+ Cf. ‘more’ and ‘most’ as forming comparatives and superlatives
in English. Cf. also Mahabhasya iii. |. 2.:
waenela 4 aaa win eat fener gqeq, fae Alay, Are
faatfa | waened 4 afanaem vafian ye eae |
 PRATYAYA 97

compounded with ‘sadrsa.”. Goyicandra in his commentary

on Sam. Taddhita Parisista 1149 tries to controvert this
view, but he is not convincing. See Wackernagel, Altindische
Grammatik. Il. 1. 236.

Tt A | ATT) ATL APTA | ae 12-3

ASMA | wearaadifa sera: |
afs mearaadifa seara:, afiemiat saad a
omifa | a fe & ffiaa senaafia |
ad afe sareaa sfa cea: |
canta aaretar a sista |
Cd AASTIANSAT, RLAIA: HATTA |
ad afe ga Taq aaiasgaiqaa TTT stgqs-
farqaracedad seqrrafa, a ga: ware sf |
AAATTRNT TENA TATA AEA ASsterceaAy
searaafa, 7 a: THIT afe |
HAA ?
“faearteraison” ( Bialeae) 1 Fea n gee
naneqay,| taaraRaifafe fr cera: | firerca)
ATR | A AAA SAA: He” (GIVIR¢) KeTAat @areraray
aq Tag a AT, TAMAR af area) eater ate
TAITTAATATEA: | RULIE-R
BH Ay TTL Bley
TT | cera Hatseesat: | (asnerarafea: | anaes
BALI) Tt geee-ve
| Aca: TAFAT: TAA | apepac
O.P, 129—7
 98 TECHNICAL TERMS OF SANSKRIT GRAMMAR

BJ) Wear} xe
S| Baa: West: TW aIk12
ee -aarfa-werte-faatn- Saq-AheaT: | RATS AT:
wHa: Gt waedifa aiearfsar: | RA a_GMA? wal agar
gagf [4r2lRe3] xf faarraaare |
q| faafe-afgaren-sae Tera AAT: | 21202
z| oanfa: gat) Sere TT |) ee o-g
apzarara sr | ATaTrend-weada a: gap:
en cafe: | aningaet aera a: ceria: ar fafa: | (aefa
THAT TAT wees FIHAT THT 21212 FAA | )
cara: | aera maa fra cfr cafe: ) eee
araragha: | seri cae fara af cafe: | areie
araedr-grare | Sfaasearrfafa ar cafe:
anaaaesfa: | wee a: serat fadtiaa a aza
sayfa: | IIRe"
qrafraciafsat | sera ya feat xfer santa:|
searanata Hal ae Gags a safeties cfegaaa |
SATA: |
genz: | seaarnafa seat a: qaqaraa oe: ar
cafe:, sata fratanarc sera: faa efa ar cafe:
serfs: | anquifacira-caraara-garea-fasearfe|
sata: aT rears Far Nea feRTay |
aaa maaan qoreaaafas: || oferta
THAI: Asa: ATA: |
AMGINAHAA
GAH LT AMA || FT AT velee
 APRKTA 99

afeenirafa saages at faftacar-

nofieaar aafa afana: gafaat |
at HR: arqayl araeta aay Hae
fagar: get verahrafeagefaa: 1) 4 gira

APRKTA.
The word “‘Aprkta’’ is hardly found outside grammatical
literature. But ‘prkta’ is a good old word occurring from RV.
downwards and means ‘mixed or mingled with, brought into
contact with,’ from pre ‘to mix, unite, join.’ Hence ‘“‘aprkta’”’
means ‘not united, uncombined.’ In the Pratisakhyas it is
used in the sense of ‘a word consisting of a single letter.’
Panini restricted it to suffixes consisting of a single letter.
Later grammarians thought it better to use ‘“‘ekal’’ for
‘‘aprkta’’ which, besides having a larger number of syllables,
had become practically obsolete.
RP. does not define ‘“‘aprkta’’ but uses it several times:
ukaras ceti-karanena yukto rakto’prkto draghitah Sakalena
i. 75, aprktam ekaksaram xi. 3. The word is used exactly
in the same way in CA: uka&arasyetav aprktasya i. 72,
nipato’prkto’nakarah 1. 79, trini padany, aprkta-madhyani
iv. 113.
VP. defines “‘aprkta” thus: eka-varnah padam aprktam
i. 151.
TP. has eka-varnah padam. so’prkta-samijfiah* i. 54-55,
—_——_—_—_—_—— ee "ae

* Whitney reads : eka-varnah padam aprktah.

 100 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Panini defines “‘aprkta” thus: aprkta ekal-pratyayah i. 2.

41, ‘a suffix consisting of a single letter is Aprkta’ and uses
the word,generally in the sense of a single consonant: ver
aprktasya vi. 1. 67, hal-nyabbhyo dirghat su-ti-syaprk-
tam hal vi. 1. 68, guno’prkte vii. 3. 91, asti-sico’prkte vii.
3. 96.
It is difficult to determine if the rule defining ‘‘aprkta’’
was composed by Panini himself or borrowed by him from
his predecessors. In the case of ‘“‘vrddhir Adaic’” j. 1. 1
Patafijali apologises for Panini: idam ekam Acaryasya
mangalartham mrsyatam. He does not, however, raise the
question of Vidheyavimarsa in the case of “aprkta_ ekal-
pratyayah”’ i. 2. 141 as in the case of “‘vrddhir A4daic”’ i. 1. 1.
In his Structure of the Astadhyadyi Pawate draws atten-
tion to the fact that in “aprkta ekal-pratyayah” j, 2. ‘4]
‘“‘aprkta’’ is masculine, whereas in “hal-nyabbhyo dirghat
su-ti-syaprktam hal” vi. 1. 68, the word is neuter. But there
is nothing peculiar in this. In the latter rule “aprkta”’ is an
adjective to ‘hal’ and as the pratyaharas are both masculine
and neuter* in the Astadhyayi it is neuter. In “aprktam
ekaksaram advi-yoni” RP. xi. 3 “aprkta’’ is neuter because
‘aksaram’ is neuter, in “eka-varnah padamaprktam’” VP. i.
151, ‘“‘aprkta’” is neuter, because ‘padam’ is neuter. In
ukaro’prktah” VP. 95 ‘“‘aprkta’”’ is masculine because ‘ukarah’
is masculine. In “‘aprktah sau” VP. iii. 110 “aprkta” is used
as a substantive and therefore masculine.
a a a ee ee a a” ae ae Eo eeeereesc ee
* Under “halan tyam” 1. 3. 3. the Balamanora ma remarks: hal-
§abdo mnapumsaka-lingo’pyasti, antyam iti napumsakasamanadhi-
karanyat. In “halo’nantarah samyogah” i. 1. 7 ‘hal’ is masculine.
 KRT.
The name for the primary suffix is ‘krt’ in almost every
system of grammar. It is formed from the root ‘kr’ with the
agent suffix ‘kvip’ and being itself an instance of a word
formed with a primary suffix became a generic name for the
suffix. In exactly the same way ‘krtya’ stands for the gerun-
dive suffixes. In AB. ‘kurvat’ ‘Karisyat’ and ‘krtam’ represent
the present, future and past tenses respectively. The generic
word “‘karaka” as also the names of most of the individual.
karakas are also derived from the root ‘kr’. Similarly ‘“‘nic”’
is known as“karita’’, ‘‘san” as “cikirsita”, ‘“‘yan” as
‘‘cekriyita” and “‘yanluk”” as ‘‘carkarita’’.
In the technical sense ‘‘krt’” occurs for the first time
in
the Gopatha Brahmana: ko dhatur ityapir dhatuh. avatim
apy eke. rupa-samanyad arthasimanyam nediyah. tasmad
Aper
omkarah. sarvam Apnotityarthah. krdantam arthavat pratipadi-
kam. adarsanam_pratyayasya. nama sampadyate. nipatesu
cainam vaiyakarana -udattam samamananti. tad avyay
i-bhitam
anvartha-vaci Sadbo no vyeti kadacaneti.
1. 1. 26,
Next we find in the Gobhila Grhya Sitra
(ii. 8. 14):
‘“krtam nama dadyat’’—a passage similar
to which is quoted
in the introductory chapter of Mbh.: DaSamyuttara-kalam
putrasya jatasya nama _ vidadhyad ghosavad-ddyantar-antas-
tham avrddham tri-purusiniikam anari* pratisthitam. taddhi

* In the corresponding passage in the Carak

a Samhita viz. tatra-
bhiprayikam ghosavadadyantasthantam: Usman
tam va’vrddham tripur-
usanikam anari-pratisthitam, naksatrikam
tu naksatra-devata-samana-
102 TECHNICAL TERMS OF SANSKRIT GRAMMAR

pratisthitatamam bhavati. dvyaksaram catur-aksaram va nama

krtam kuryan na taddhitam.
Next we find in the Nirukta (ii. 2) : athapi bhasikebhyo

dhatubhy6 naigamah krto bhasyante daminah ksetrasadha

iti,“again, from roots of classical Sanskrit are derived Vedic
names ending in primary suffixes as ‘damunah, ksetrasadhah’
etc. Durga explains ‘krtah’ here as “krtpratyayantah
Sabdah”’.
Krt is not found in RP. and TP., but occurs several times
in VP. : tin-krt-taddhita-catustaya-samasah Sabdamayam i. 27,
etc. Both commentators take ‘catustaya’ as an adjective to
‘samasa’ and explain ‘catusprakarah samasa uavyayibhava-
tatpurusa-dvandva-bahuvrihayah_. In K., however. catustaya
means naman, karaka, sandhi and stripratyaya and something
like that may well be the meaning here.
qegtat aad aa HTH MTS faorz: |
aaa AF BA ATV Iaaz |
In the corresponding stanza of the Natyasastra we find -
ahragqagpaiarerrarranifaqta: |
afgeafeatrafatactatgea: mez ge: ||
Following K., the Natyasastra does not mention krt.
Sarvarman, the author of K., did not frame antes for
krtpratyayas. These were supplied by Vararuci who savs in
his introduction :

khyam dvyaksaram caturaksaram va (iv... 8 50), ‘anari-pratisthitam.

has become ‘anavapratisthitam’ in MSS. and printed editions through
the error of scribes who mistook r for vy owing to the very great
similarity of the two letters in Bengali script.
 KRT 103

aatizagat wet: Hisar a He: Her |

aang & ast fagfeqiaad 1
still the name of the suffix was krt according to him also, for
we find ‘‘kartr-karmanoh krti nityam” ii. 4. 41 corresponding
to Panini’s “kartrkarmanoh krtti’” ii. 3. 65..
The Pratisakhyas of AV. also contain the world :

Heed FIT Ta aor fesre: |

aH RATTAATAY cast at Fa 1 i. 1. 10
aT) wBafas_ gzieg
HT) BAST, ree
| earatas_ HT 22183
waniferaa-ae (eigige) feat aaa
wafa | ata: |
El wMgarscate: Ha yey
BT | BGFAE HITUTA 312
Yl HREr Mis 2133
Gl Aas Sieie
T) wa: Rafe vee
€) We aga aq aq wees a: |
aa cdifa fafa agai wa Tea II
Wat: BEES RAFT | &Ig-g
KRTYA
Krtya is a sub-class under krt and is the name of the
gerundive suffixes tavya, tavyat, aniyar, yat, nyat) kyap and
kelima. It is formed from the root kr with the gerundive
 104 TECHNICAL TERMS OF SANSKRIT GRAMMAR

suffix-kyap and, like krt, is taken as the representative of all

gerundive suffixes. As the Prakriya-sarvasva says:

andfe fafa west ani wea fe eeaF |

staemard Hae td RIANA ||
Another very great advantage of the term is that its
form also distinctly shows that it is a sub-class of ‘krt’. C uses
“tavyadi-sat’”” for “‘krtya’’. Thus for the dictum “lumped
avasyamah krtye’’ C. has “‘tavyadi-satke’vasya-mah”’ v. 2. 90,
The suffixes are known as “‘vya’’ (the second syllable of
sé

tavya) in J. In RT. ‘vya’ stands for ‘talavya’.

Vopadeva is compelled to call these “‘lya’” as ‘tvya” has
already been used for “‘avyaya’’. He takes the letter nearest
to ’v’ and adds ‘ya’ to it. In the Panini system ‘lya (p)’
stands for the gerundial suffix ‘ya’, but as the ‘l’ here serves
the purpose of indicating the accent and as Mu., like other
grammars of classical Sanskrit. does not take accentuation
into account, Vopadeva substitutes ‘yap’ for ‘lyap’. (Sak,
uses ‘pya’ for ‘yap’.) |
Sak. uses “ghyap” (1. 3. 168) for “krtya”. ‘“Ghyap” is
a pratyahara formed with the ‘ghya’ of ‘ghyan’ (Panini’s
nyat) and ‘p’ of ‘kyap”’. |
Sam. uses krtya in vy. 38.
HN. uses Visnukrtya for Krtya.
GT | AN BITS
Raa: | HATAMA MSVIAIAAT zyrjerz
H | VaEAT: aIleleey
@\ HAM “leisy
 NISTHA 105

ft Pao deg ala eT FIT Fed ae: Heda wafra |

Jecaia: |
Bl Fear AaTAAMIGAHIAHIS AT] Ace
Bl Re: AS THT vee
q) aeartat feral sag cao Fer ca Rad | Hea
feat: | 2eiee
e. fava asaria-aq-yaq-0aq-afear fasnsea-
Gat UTARANT: | 41Ze,
The term ‘krtya’ is very happily used by Kalhana (iv.
G35) :
faard sane quafstaanfaa: |
Hawise cfwasy feaacy|
‘Wherein lies the difference between His Maijestry King
Jayapida who has fully nerformed his duties and has deve-
loped all good qualities, and Panini who has treated of the
gerundive suffixes fully and has prescribed Guna and
vrddhi?”
In Subandhu’s Vdsuvadatta (Vani Vilasa Press ed., p.
334) occurs the line :eqrararfaa @l-agi-H_ea-age yA‘abound-
ing in the actions of rivers even as grammar abounds with
feminine forms, words technically termed ‘‘nadi’ and
gerundive suffixes.’

NISTHA AND SAT,

Néstha
. a:
“Nistha”, the name given to the past participial suffixes
-kta and -ktavatu by Panini, was originally “nihstha,” even
 106 TECHNICAL TERMS OF SANSKRIT GRAMMAR

as the word ‘‘antastha”’ so familiar to us in grammatical litera-

ture was originally “antahstha”. The visarga of ‘‘nihstha’’
must have been dropped later on according to the Véarttika
‘-vq-sar-prakarane kharpare lopah” vii. 3. 36. 1. ‘‘Nistha”’
is derived from the root stha preceded by the upasarga nir
with the primary suffix an. Nih stha means ‘to stand firm’,
hence ‘nistha’ means ‘firm decision, perfection, completion,’
and thence came to mean ‘a completed action’.
The exactly corresponding English word ‘perfect’ is used in
the same sense in such expressions as ‘present perfect tense.’
The word “‘nistha” goes back to RV. and is analysed in the
Pada-text as “‘nih-sthah’ (ix. 110.9). It is used in the
Chandogya Upanisad (vil. 20. 1) in our sense :
gaa fafageaa aearfa, aff sents, frag
aq seatfa feet ca fafaareaacafa, fast waaay
fafaara afc t
Though the chapter on primary suffixes in K. is by
Vararuci still we learn from a rule in the Karaka-section (na
nisthadisu ii. 4. 42) that Sarvavarman also used the term
‘nistha.”’
RT. uses Nistha in the rule “‘nisthayam it tadau’’ 234.
Candra uses “‘ta-tavat” every time. Thus for P§nini’s
“‘nisthayam seti’’ vi 4. 52 C. has “‘ta-tavatiti” v. 3. 68 and for
“radabhyam nisthato nah piurvasya ca dah” viii. 2, 42 ¢
has “radat ta-tavator das ca” vi. 3. 74.
J. uses ‘‘ta’’ as the upalaksana for ‘‘-ta’’ and ‘-tavat.”
Sak. and Hc. use “kta’’ in the dual for -kta and -ktavaty.
Sam. does not use “‘nistha’” but mentions -kta and
-ktavatu each time.
 NISTHA 107

Mu. follows Sam.

Su. and PR. follow Panini and use ‘‘nistha’’.
HN. prefixes the word Visnu to ‘“‘nistha’” and calls these
suffixes ‘‘Visnu-nistha,”’ even as it calls “pada” ‘Visnu-pada”
and the krtya-suffixes *‘Visnu-krtya.”’
Sar. uses “‘ktadi’’ for “‘nistha.”’
The Prakriyad-sarvasva explains “‘nistha” thus:
Fast ofcearha: caraat a areataaM: |
ar: eafeases froren yerarafe 11
TT) wma fran weiRg
Ai Wma ag) BAT cy
Blo wewad frwr aire
qT w-waed. fast dat Wt 2o]RW9e
€1 adtaral w-wnadq fawyfasrda usec
The Raja-tarangini (iv. 637) uses this technical term also :

pafeorranter yatta: |
taridrszaea orfirtey fang |
‘Wherein lies the difference between His Majesty King
Jayapida who brought down calamity on the Brahmanas and
who put an end to people and Panini who laid down the pre-
position vi and pra and who prescribed the suffixes -kta and
-Atavatu to denote past ume ?’
agiqatataa: Ss wferfrgrazy zat |
afeeaar fasta srry cristae! |
rangaactyot 3"
‘Rama, the best of kings. whose accepted creed is fo!
giveness, who is given to devotion, who is an object of pra!sa
 108 TECH NICAL TERMS OF SANSKRIT GRAMMAR

and who is possessed of all good qualities is all conquering

like the root mrs which with the union vowel ;(f) takes guna
in the past participle in the sense of forgiveness (P. i. 2. 20)’.

Sat.

“Sat” is the name of the present

participial suffixes ~Satr
and -sdnac in the systems of Panini and Devanandin (tau sat
P. iii. 2. 127, J. u. 3. 15). Sat is itself a present participle,
being derived from the roots as to be’ with the suffix -Satr,
and is taken as the type of all present participles, active and
middle. HN. uses “acyutabha” for these suffixes (vartam-
Gnddau Satr-Sanacav acyutabhau phalantaraprayoge parapada-
tmapadayoh v. 3). “‘Acyuta”’ stands for the present (/at) in
this system and so “Acyu-tabha” (literally ‘resembling the
present’) very properly stands for the present participial
suffixes. The other systems find it unnecessary to have a
single name for the two suffixes and generally use Satr-SGnacau,
Satranau etc. for them.
It may be mentioned in this connexion that HN. alone
of all the systems of Sanskrit grammar uses the technical
term “‘adhoksajabha”’ for the perfect participial suffixes -kyasy
and -kdnac and ki (paroksdtite kvasu-ki-kana adhok sajabha-
samjnah parapaddtmapadayoh v. 20).

TADDHITA.
‘““Taddhita”’ is evidently formed from “‘tasmai_ hitam.”’
Probably in the older grammars instead: of “tasmai hitam”
the rule for hitarthaka taddhita was ‘“‘taddhitam’” and since
 TADDHITA
109

the word was auspicious, containing as it did hita. it came to

stand for all the secondary suffixes. In recent times, however,
a class of indigenous scholars gives a wide berth to the
taddhita section on the ground that it is beneficial to others
but not to one’s own self. K. also uses the term “‘taddhita”’
though it deals with the very important suffixes only and
Sarvavarman does not appear to have composed this portion
of the work. The ancient tag has it—Kalapas taddhite mir-
khah. Rajasekhara in his Kavyamimamsa preserves an
ancient adage—taddhitaniidhah Paniniyah which appears to
be absurd on the face of it. The correct reading would
appear to be “taddhita-mtidhah Katantriyah”. J. in its
search for a suitable monosyllabic term takes ‘h’ and ‘t’ of
taddhita and throws a ‘r’ in between to form a word with
some meaning and gets the name ‘hrt’ which evidently follows
the analogy of krt‘ and ‘mrt’, Sakatayanaprakriya uses the
word the “‘taddhita” as the name of the section. Hc. also
uses the term. Sam. includes feminine suffixes under taddhita.
Mu. uses the first syllable ta for taddhita. Su., PR. and even
HN. use the term ‘“‘taddhita’’.
The word occurs in the Gopatha Brahmana, Nirukta,
Dharmasitras, V. P. (i. 27), CA. (ii 83), BD. (ii. 106)* etc.
The Natyasastra explains the significance of the name
arq-qafa-qerr-firrm-atatrarragay |
TSA [Aaa TART RAATA (9) | ge1Re
* faawftas: aa ardely afza )
nfaasd afra are euerey ceez wate |
“In compounds, as well.as in a secondary derivative, explanation
 110 TECHNICAL TERMS OF SANSKRIT GRAMMAR

The Prakriya-sarvasva says:

ata feafafa aut aenra & a afzat: |

The Gudhaprakasika says:

ease fed: gata WATTERS

afea: |
Candra does not use the term. For Panini’s taddhites
vacam adeh. kiti ca.”’ vii. 2. 117-8, Candra has “kitj elipatya
davacam adeh”’ vi. 1. 11.

The Nirukta gives “‘dandya’’ as an instance of ‘ekaparvan

taddhita’ and “‘kaksya” as an instance of ‘anekaparvan
taddhita’. It is clear from this that Yaska included the
feminine suffixes under taddhita. Panini, however. though
treating o1 both in the same chapter, puts in the adhikara-

“23

should proceed from analysis: one should explain after separating

the parts; thus ‘punishable’ (dand-ya) as ‘deserving punishment’
(danda-arha). Macdonell.

This is based on Nirukta ii, 2:

wy afequHl
gaaes
agag a ya WaHgAge HlausE faa
ana |
In BD. i. 3:

afgdiaeiunaaatat aaeieq |
fanrqafa fanaa waifw fafauifa q |
“Taddhita” appears to be used in the sense of ‘contained in them’.
Cf. Nyasa iv. 1. 76: aga: asian ara wa walaAH, Waza aU
famaa—aen fearafgar 1 dizaaa atfaar afeary seer: manual, ani
VqMNdad Awad | AA aaa HAATWIET aa q NqmE gqgaiftat wan,
aaaial aafraraaqua afgaaagaa wala |
 TADDHITA 111

rule ‘“‘taddhitah” at the end o. the feminine .uffixes* Su

Sam. and others follow the older method, while J.. Mu. and
others follow the new method.
Sak. uses the pratyahara “syat’’ for the feminine suffixes
Cf. pratyaya-nyak-syat prakrtyadeh i. 1. $9 where the com.
mentary Cintamani says: syad iti “‘gurapottamasyanarse’-
patye’nimah svan” 1. 3. 2. ityarabhya ‘‘ytinas tit’? 1. 3. 76
ityatas takarena pratyaharah. This pratyahara is formed
after the analogy of sun i! 62 for sup aud tii.
The samasantas are regardei aes Taddhita by Panini Sam.
distinctly says: “‘sa ca taddhitah” vit. 233

Tr | aeat sense
HA | Fal BWR
A

aT} xatarata afsa: eieive

f 1 afeatsonfe aie
bao

@| aratsa: faararcfea: ele

Hl afr: ure
APUSATIEl:
x e
( B28
-
) LARMALIT ART: (Heqaz:) SH:
@ aad: Sard |
ai ated “ie
«wey afgafeait: ataaaiafed ga aan fad > faaa ala wala o
Samay afgarraia aifaaieafafa enanfaafeana | eieiy ] saa SaaeT
a aaa Hala, nat weed [ Bio] saa 4 afgageuE, aaifa a [elsigsc]
TMAACT 4, afea sara feReT! Wad Faq vi fe fRaMa aa
SULIVIG aq ya WAN ay Ug, Sq-AeSine CaARamqass,
‘wafwa’ [ gas j sft nfaterd) sea ocala ‘Maa | ceive ] sf qu:
S| ANA BL P19
 112 TECHNICAL TERMS OF SANSKRIT GRAMMAR

o| faarazfaainfacaraatiany |
fara afzararanaarerag efzq 1
mares afgar ar qateaigeanfem: | wjego-s
Z| waadfears ole2

TADRAJA

‘“Tadraja”’ literally means ‘king of that’ and stands for the

suffixes afl. an, fiyan, nya and in prescribed by Panini in rules
iv. 1. 168-173. Though these suffixes have been enjoined in the
sense of ‘apatya’, still because of the atidega “Ksatriya-
samana-Sabdaj jamapadat lasya rajanyapatyavat” iv. 1. 168.
3 which though not distinctly formulated by Panini must have
occurred in some form or other in the works of his predecessors,
these suffixes are known as tadraje. Thus the Prakriya
Sarvasva Says:

AT TAATATAT
ETAT MAA, |
ATATALAAIAT ATTA ay |
K. has nothing corresponding to tadraja. J. takes “dra”
from the second syllable, adds an ‘i’ to round it off and forms
‘dri’. Sak. changes ‘‘dri” into ‘“‘vri’. ‘The first example given
in the commentaries is Videhanam raja. Sak. evidently takes
the vy of Videha and r of rajan and like J. adds the vowel
7 for
ease of utterance. Candra uses “fyddi’’ for “tadraja”, Cf.
‘““‘nyadinam bahusu luk” iv. 3. 94 corresponding to “‘tadraja
sya
bahusu tenaivastriyam”’ P. ii. 4. 62. He. follows J.
Kramadisvara and Vopadeva do not use the term.
 GOTRA AND YUVAN 113

Padmanabha also does not use tadraja. Cf. ‘‘yaskadi-

janapada-rajanyebhyo bahusvastriyam” v. 2. 126 for Panini’s
‘‘tadrajasya bahusu tenaivastriyam” ii. 4. 62.
PR. also does not use the term: cf.

aalfog Qaarasfasafarer F |
ATEAT ALCP PATATAIR FH I] LILLS
HN. does not use the technical term “‘tadraja’’ but uses
“rajapatyayoh”’ in vii. 312 and says under ‘kekayad va’ vii.
315: ukte rajapatye. atra tadrajasamjiia Paniniyanam.
UT | St ARIAT: BIRIR98 | SaTAETaTAM: PRIA RL
H 1 Ff eres fas aeiges
UT | TT: Aas Tas fr: zie fasifsyer-
ATT WATT IIRC e
Zl Ugethaarl Away creed fees E1212 2%
GTTraAeasy Sat fF: wiaige

GOTRA AND YUVAN.

‘“‘Go-tra”’ literally means ‘that which protects (tra) cows

(go)’ and is used in RV. in the sense of ‘cow-stall, cow-pen,
hurdle’, then it meant ‘family enclosed by the hurdle’, then
it came to signify the family-name. Grammarians use this
word in connexion with ‘apatya’ and distinguish between
‘gotrapatya and ‘yuvapatya” and the use of the former is
restricted to the grandson and his descendants, if no older
offspring of the same ancestor than his grandson lives. If
the son lives, then the fourth descendant is known as “‘yuvan’’,.
O.P. 129—8&
 li4 TFCHINICAL TERMS OF SANSKRIT GRAMMAR

so also if some older offspring lives. ‘‘Gotra’’ was termed

‘vrddha”’ by the older grammarians. Both ‘‘gotra’’ and
‘‘vrddha”’ are sufficiently self-explanatory. P&nini uses the
word “‘gotra’’, as he had to use ‘‘vrddha” in another sense
(vrddhir yasyacam adis tad vrddham i. 1. 73).
J. and Sak. follow Panini.
K. deos not distinguish between ‘gotrapatya’’ and
‘yuvapata.” Mu. follows K.
Candra generally avoids the use of ‘“‘gotra’ by using
‘pautradi’ instead (cf. apatyam pautra-prabhrti gotram P. iv.
1. 162). e.g., ““pautrader astriyam gurvayatte” ii. 4. 18 corres-
ponding to Panini’s “‘gotrad yinyastriyam” iv. 1. 94, “rseh
pautradau” i. 4. 23 and “‘pautradeh striyah kutsite na ca’ ii.
4. 79 corresponding to Panini’s “‘gotra-striyah kutsane na ca)’
iv. 1. 147. “Gotra’” is used by Candra in its ordinary sense
e.g. “bahvadibhyo gotradibhyah” ii. 4. 20 for P§nini’s
~bahvadibhya’ ca” iv. 1. 96. |
Kramadisvara also uses “‘pautradi’” (iv. 276, 280, 302.
350) for ‘gotra’ where necessary. In the rule “pautradi-pratya-
yantad yuni’’ (iv. 283) the term ‘yuvan’ is used. It is explained
thus in the Vrtti: jivaj-jyesthabhratrko’stri yuva. Jivat-
purvah prapautradis ca. Jivan jivad-vrddhatara-bhratryanya-
sapindo va. The next rule which would appear to be a part
of the vrtti is “pautradih ptja-nindayoh”. In the Samasa-
section, however, we find “yuva-gotra-taddhitantayor gotra-
taddhitantah”’ vii. 409 where the commentator Goyicandra
remarks: “‘apatyam pautra-prabhrti gotram iti Sastrantara-
samjhaya vyavahartavyam,”’
 GOTRA AND YUVAN 115

Padmanabha uses ‘“‘gotra’” and “yuvan’’ without defin-

ing them. In the Vrtti on “‘adyad eko gotre” v. 2. 79 we find
‘pautradyapatyam gotram’ and in the Vrtti on the next rule
“gotrad ytnyastriyam” v. 2. 80 we find ‘vaméye Jivati
tadapatyam yuvocyate’’.Under v. 2. 100 we find ‘“gotrapatye
ye pratyayas te yuvasamjnakah.”’
HN. uses “‘gotre’”’ in the rule “‘gotre’’ vii. 290 and explains
it in the Vrtti: “pautra-prabhrtyapatyam gotram”. ‘Yuvan’
is defined in vii. 299.

qt) anet dlamifa way staf

a dategary
arate a savafa) arafera afiee afaca? aft |
SIZIZER-ZE"
tt Ware aa aieiees | afa det aarseft | erat
mate | aa afiresfiraacera |sretisateeat: 31212 24-
ete (Geeralat
gra Eerai sre: dareafserc: eat |
ata: | )
ati - Wate gan, s@lamet-daesnamat: ahr
gar | aqafarssfraeanr at | gags Haar | P1212 3-E
2: Gafe aaa, agariardisitafe saraet
gat) oafres aeernfas stag at wag Hart
aT | Zizie-"4
Bi Maret magi sfaf det qamashr |
frasaray at aaisfae | dea Qa, Tat a farararq |
ARIZ EC-RRE
1) Ware wi astaafvarie gar) mata
qaiataaafargeag | sjgoee
 116 TECHNICAL TERMS OF SANSKRIT GRAMMAR

<. feat safe dared gadaq|) sagarefe

srafa afaga aera afass Ste gar, afga zee,
Hania tay) afga =e

UPAGRAHA.

‘“Upagraha” literally means ‘seizing, taking possession of"

and thence ‘sense’, viz. whether the action is performed for
the benefit of the agent or not. then Atmanepada and Paras-
maipada. It also means ‘that which is seized’, hence ‘addition
of a sound’.
The word is not found in the Nirukta nor in the Prati-
<akhyas, but is used by Katyayana. Patafijali, Sabarasvamin
and others.
Under P. tii. 2. 127 we find the Warttika ““upagrahapratise-
dha§ ca’’ which is thus explained in the Bhasya : upagrahasya ca
pratisedho vaktavyah katiha nighnanah. tananav atmanepadam
ity atmanepadam prapnoti.
The sense is this: If -Sanan, -canas. etc. be regarded as
substitutes for lat, then it will have to be laid down that the
designation Atmanepada attaching to -canaé. -SAnan, etc.,
according to tananav atmanepadam”’ 1. 4. 100 will not apply
io these cases. otherwise in instances like ‘‘katiha nighnanah,”’
‘how many are here who possess the power to kill?’ the root
‘han’ preceded by ‘ni’ being Parasmaipadin cannot admit -canas
which is Atmanepada.
Kaiyata explains in his Pradipa :
upagrhyata atmanepada-samjfhanayety arthah, anyatha para-
smatpadibhyas canas na syat.
 UPAGRAHA 117

The word is probably first explained in the Upagraha-

section of the Prakirnakanda of Bhartrhari's Vakyapadiya :

qT araaqaiga: afarieg nraa |

MUTINY SMEG ATTA |
afaq araaaare afaq ae faaaorg |
araa aa Hats caaarar fader||
frat faraea sift ies faa |
ara: @ feast aretaty frase? |)
saver: afaeqaE: |
aaa weaarfs: ware arfaersy faaaor
y ||
By Upagraha is meant the distinction of meaning that is
sometimes conveyed by the Atmanepada and sometimes by
the Parasmaipada. In some cases it is the sadhana and in
some cases it is an adjective to the sadhana. Of these, sad-
hana signifies karman, kartr and bhava.
All the distinct senses conveyed by the use of Parasmai-
pada and Atmanepada come under “‘upagraha.’’ ‘Thus in the
case of roots with an indicatory svarita vowel or an indicatory
n, the use of the Atmanepada shows that the benefit of the
action accrues to the agent, in the case of Parasmaipadin roots,
ihe Atmanepada shows that the verb: is used in the passive or
neuter voice, in the case of roots like kram etc., it indicates
the peculiar sense of ‘vrtti, sarga’, etc., in the case of many
roots it indicates the sense of “‘vyatihdra” or reciprocity and
sO on.
 118 TECHNICAL TERMS OF SANSKRIT GRAMMAR

In the Bhasya we find under ‘“‘vyatyayo bahulam’”’ iii. 1.

85 the well-known stanza:

qfarqraaegatren HSCS AATHA IST |

srearafaeata mmeaent asta a faxafa ageae |
The author of the rules desires (i.e., approves of) departure
from the normal form in the matter of case-endings, verbal
endings, upagraha (voice), gender, person. tense, consonants,
vowels, kartr i.e., karaka and yan-. Pataiijali says in reference
to “‘upagraha’’™* : Atmanepada-vyatyaya. brahmacarinam
icchate. icchatiti prapte. parasmaipada-vyatyaya, pratipam
anya urmir yudhyati. yudhyate iti prapte.
Nagesa explains “upagraha” thus: upagraho JAdeéa-
vyangyah svarthatvadih. iha tat-pratiti-nimitte parasmaipadat
manepade upagraha-sabdena lJaksanayocyete. ‘Upa-graha (pri-
marily) means whether a thing is being done for one’s own
self or for others—a sense indicated by the substitutes for the
lakaras 7.e., the verbal endings. Here by Laksana or synec-
doche “‘upagraha’”” means Parasmaipada and Atmanepada
which convey whether a thing is being done for one’s own slef
or for others.’
The word also occurs in ‘“‘na nistha-parasyanuprayogena
purusopagrahau visesitau syatam”’ (Mahabhasya iii, |. 40),
Jf the past participle is used after the root to which 4m has
been added, then the person and the distinction of Parasmai-
vada and Atmanepada cannot be indicated.’ We meet with

"Ct. Natya-sastra xiv. 9: fawaa: gufaeaanfa! &: anagem

fasrerqaeaigeaegat |
 UPAGRAHA 119

‘“upagraha™’ once again in the Mahabhasya: in the passage

‘‘tinabhihitena bhavena kala-purusopagraha abhivyajyante,
krdabhihitena punar na vyajyante’* (iil. 1. 67), ‘when an action
is ©xpressed by means of a finite verb, the time. the person,
and the fact whether an action is performed for one’s own self
or for others are indicated, but these are not indicated when
the action 1s expressed by (a root ending in) a primary suffix.’
Kaiyata here says: upagraha is of various kinds, viz., whether
an action is being done for self or others and ‘so on.
The Nyasa says under iii. 1. 85.

aiaaeaga: frafaast yer sane: | ce g agate-

fafracarey WceaazTeAATTaAATa |
The Sabdakaustubha (ed. Chowkhamba, p. 879) explains it
more fully :

ee a - —___—___ 4 _--

* Under ii. 19. 2 Patafijali says: krdabhihito bhavo dravyavad

bhavati. The distinction between the finite verb and the infinite forms
is thus stated in BD. i. 44-45:

frag wwhafaiah a: galataya sea ca |

frenfatras fasta fay srename a aRdare: |
frarfafad fatto: wemuefafead set we |
eantaafnaalayrge wae zufaalae: i
“That notion which, though one, is connected with many actions,
becoming a later from an earlier (i.e., which expresses sequence of
time), and is effected by the development of the action, they call by
the term ‘verb’. When the becoming arising from the fruition of
an action is designated by a term with a primary suffix, and is joined
with number, different case-endings and gender, then it is to be re-
garded as a substantive.”
 +20 TECHNICAL TERMS OF SANSKRIT GRAMMAR

silanes: feararrafasiver: arerqzistcarzarqae-

SexeqT AST: | TATA
qT anraaqae Az: afazeder mraz |
Raat aeareAaed AQTEA |
( ATRITTIT 312212 )
araaqaifafa araarfefa a gat aya | we og
aandifata fad qteag? aaaq? a SUMSRES HATTAT
add |
The Mimamsakas explain “‘upagraha”’ thus:

santaqiitacaaen: | waftatar areata varfca

sofanaa entaat aa fa admfinecd odiad |
quaqe g ‘asta anen: af conrfaneag |
The expression “‘upagraha-visesa’’ occurs several times in
the Sabara Bhasya, and in each case it means ‘a —s
voice’, either active or middle :

moral
a omdafs ) ad ga: HeennfH: 2 SunE-
fasrrsrantq—asiafa | aat Arar atactt qaets fray.
aaiafa | ag ge HA -AaTaATaaISaary Te |
carat aaa: cat: | HAA? Soae-faserq | syHe-
fara fe vwafa—afanretefa | aenaaeariafacege
a fafe:, wafa meaat aaa cfa elei3e
na Gacaq [ sq-e |]svera-aaaishrararsd: «saat
faarga| serdt—saae-featry) oa safe.
craaqe wafa) §691313%
asTATRey Graefes Ay BA ATT; arahegeTi
ATaAeatta | polxtke .
 UPAGRAHA 121

‘““Atmanepada” literally means ‘word for self’ (2tmirtham

padam— Nyasa vi. 3. 7) and denotes the active endings.
‘‘Parasmaipada”’ literally means ‘word for another’ and denotes
the middle endings. The Paniniya Dh&tupatha uses ‘Paras-
maibhasa’’, ““Atmanebhasa”’ and “‘Ubhayatobhasa”’ while treat-
ing of the 1035 roots belonging to the bhii-class. At the
begining of the ad-class “Parasmaipadin” and “‘Atmanepadin”
make their appearance and continue to be used from time. to
time. Most probably the nine ganes from adadi to curadi
were tampered with to a greater extent than the first. In the
Katantraganamala Parasmaibhasa, Atmanebhasa and Ubha-
yatobhasa occur throughout. That Panini was fully aware of
Parasmaibhasa and Atmanebhasi seems evident from his rule
““valyakaranakhyayam” caturthyah” vi. 3. 8 under which
Para-
smaibhasa and Atmanebhdsa are given as examples in the
Kasika. The Bahuvrihi forms Parasmaibhasa and Atmane-
bhasa (corresponding to Parasmaipadin and Atmanepadin)
appear to have been provided for in the Varttika “atmanebhisa-
parasmaibhasyor upasamkhyanam vi. 3. 8. l. It would appéar
from the Bhasya on wi 3. 8. that though Atmanebhasa and
Parasmaibhasa were used by grammarians no grammar
useg
them as technical termis*. Since the two words are found in

sada ae MAT Ha) ermal anacm waeche | fategte: |

THATS ATI aafae Stacy anda Rignfrergay.
Sar 1 oo maAaeay wad aqaedcmaemes Tay WRG Ufa:
Ue TG AITYR TE 1 gz: |
WaTTe ae Sica wh oxarane: | aaAIAaan | oe TTY fitqrer,
wigfaiaranfaat ageat Wiaaaiiem <iyoe
122 TECHNICAL TERMS: OF- SANSKRIT GRAMMAR

Dhatupathas, the Bhasyakara evidently means that they are

rot used in grammar proper but in subsidiary treatises like
the Dhatupatha etc.
K. uses Parasmaipada and Atmanepada, but shortens them
to Parasmai and Atmane, whenever convenient. Thus, we
find: atha parasmaipadani. nava parany atmane iii. 1. 1-2,
atmanepadani bhava-karmanoh ili. 2. 80, in-stha-da-pibati-
bhubhyah sicah parasmai i. 4. 93, sij-aSiso$ catmane iv. 1. 10,
sarvesam atmane sarvadhatuke’ nuttame pafncamyah, dvitva-
bahutvayos ca parasmai iv. 1. 18-19 sarvatratmane, ASisi ca
parasmai iv. 1. 21-22.
J. uses ‘‘da’”’ the las‘ syllable of Parasmaipada for Paras-
maipada itself and “‘ma’’ belonging to the second syllable of
Atmanepada for the latter. Following in the footsteps of
J., Mu. uses ‘‘pa”’ the first syllable of Parasmaipada and ‘‘ma”
from the second syllable of Atmanepada for the two padas
respectively. “‘Pa’’ is, of course, an improvement on J’s “da”,
and the use of “‘ma’’ for Atmanepada cannot be objected to,
as it occurs in ‘Atmanepada’ alone, ‘mai’ occurring in ‘Paras-
maipada.’
Candra uses “‘tanana’’ for Atmanepada (taking the hint
from Panini’s “‘tananav atmanepadam i. 4. 47) and has
no special name for Parasmaipada which he has to express by
means of a periphrasis. Thus for Panini’s ‘‘gamer it para-
smaipadesu” vii. 2. 58, he has to frame two rules: aningamer
it” and ‘‘na tananath” v. 4. 121-122; for Panini’s “‘sicj ca

winage wet sie: we sfa wimsare: | wd gta are’| gem

fagamay) wmgfaraafaa as) Malae 1 wees
 UPAGRAHA 123

parasmaipadesu” vii. 2. 40, Candra has ‘sicyatani’ v. 4. 103.

In C’s Dhatupatha the Parasmaipadin roots are described as
‘“‘atananah, the Atmanepadin roots as ‘tananinah’ and the
ubhayapadin roots as ‘“‘vibhasitah’’.
Sak. uses ‘tan” for ‘‘Atmanepada’’; thus for Panini’s
‘“‘atmanepadesv anatah”’ vii 1. 5, Sak. has “‘tano’natah”’ i. 4.
90. for “‘Parasmaipada”’ he uses ‘‘atan’’.
He. uses the fuller forms in his definition, but in the subse-
quent rules he uses the shorter forms. Thus while erplaining
“navadyani Satr-kvasti ca parasmaipadam” iii. 3. 19 he says
in his Brhad-Vrtti: parasmaipada-samjiid-pradesah ‘‘Ses&t
parasmai” (iil. 3. 100) ityadayah, i.e.,. the rules in which the
technical term Parasmaipada occurs are ““sesat parasmai’’ etc.
Similarly in the Vrtti on the next rule “parani kAndnagau
catmanepadam” he says: dtmanepada-pradesah “‘sij-ASisav
atmane”’ (v. 3. 55) ityadayah. It will be noticed that the rules
quoted to illustrate the use of the terms ‘“Parasmaipada” and
“Atmanepada” contain only the truncated forms ‘“‘parasmai”
and “‘atmane’’.
ét=

Sam. uses “‘parasmai” and ‘“‘atmane” (in ii, 14 etc.) which

are replaced by the fuller forms in the Vrtti which remarks-
parasmai parasmaipadam iti, Atmane atmanepadam _ iti.
sambandhanupapattibhyam éabdalaksand. The longer forms
are also used in a few rules: anrganteg-upantad atmanepada-
hittvartho mahin-nakaérah ii. 689. na snor indtmanepade
kramer atmanepadarhasya kartr-krti ca ii. 717, pusadyldito’g
parasmaipada il. 762 etc. Even ‘“‘Parasmaibhdsa” is used once
in “‘Sanah Site parasmaibhasac ca” iij. 10.
 £24 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Sar. and Su. generally use the longer forms throughout.

In the Sarasvata-sutra, however, “‘at’’ is used for ‘‘Atmane-
pada” and “pa” for “parasmaipada”’ in “‘ad anudAatta-nitah”’
x. 78 and “‘kartari pam ca’’ x. 77. “‘Ubhe”’ is used for both
padas in ‘“‘nitsvariteta ubhe’ x. 80. Su. uses ‘‘atmanebhasa”’
in “‘atmanebhasad aya-dipa-diksa-nito haladeh”’ iv. | 63.
PR. generally uses the fuller forms, but in a few cases it
is forced to use ‘“‘parasmai’ and ‘‘atmane’’ for metrical
| reasons; €.g., San-sya-yogi parasmai ca_ vrt-vrdhii-syandigrdh-
krpah ij. 247, krfah prayoga amantad as-bhuvoh_kartran-
atmane ii. 320 etc.
HN. uses “‘parapada”’ (lit. the highest step or position)
for Parasmaipada and ‘‘atmapada” (lit. the step or position
tem

of the soul) for Atmanepaaa. The commentary says: pata-

pada-sabdena Vaikunthadikam ucyata iti Bhagavan-ndamatd.
atma-Sabdasya Brahma-vacitvat “atmapadam”’ Brahma-
padam iti Bhagavan-namata.

Under K. iii. 1. 79 Susena explains Parasmaipada and

Atmanepada thus:
qua Tat FCAT AT qUATe way |
ACMAATTA ||
arena Ua TA AT
aunraqaadaran faatad aaa }
nd fe arfata arn wear: |)
qanraqaaarar: ATat afafiecaa
sat a afer: aa arad Baa: |
‘That which is had recourse to for the sake of another is
known as Parasmaipada, and that which is resorted to for
 UPAGRAHA L235

one’s own self is, for that reason, known in grammar as

Atmanepada. One would gather from the use of these self-
explanatory technical terms that it is the view of Panini that
is subscribed to by Sarvavarman. This, however, is not the
case. Self-explanatory technical terms are generally used in-
this treatise (and consequently they are not to be taken lite-
rally, nor is any inference to be based upon them). So
Panini’s rule is not sanctioned by Sarvavarman.’
Similarly we find the following Karikas quoted by Durga-
dasa under Mu. 920 and by Gopinatha under Katantra
Parisista vi. 10:

ae olfanaraae saftrarcisfirgens: |
qranaay aafia faa: Bat TzGs
fags: aifirafce: arared adam: |
AAFAATTSa AATAAATEATT |
‘In the popular speech the exceptions (to the rules res-
tricting the use of the padas) are very numerous. If the
restriction be found in Vedic speech, who is going to prevent
it or take exception to it? Panini prescribes the special use
of the Padas. Sarvavarman is in favour of their indiscrimi-
nate use and the whole host of teachers agrees with Sarva-
varman.
The meaning and uses of the two Padas are thus clearly
explained
by Dr. I. J. S. Taraporevala in the Jua Commemora-
tion Volume :

Parasmaipada—Pada for another. When the action is re-

flected back on someone other than the doer thereof, the verb
 126 TECHNICAL TERMS OF SANSKRIT GRAMMAR

should be (in the) Parasmaipada. When it is reflected back

upon the doer himself, the verb is in the Atmanepada, the
pada for oneself. Cf. ram to sport, labh to gain. as to extend,
iy to be master, so mr to die and di to fly. Deponent
verbs in Latin are verbs with passive personal endings and
reflexive or active meaning, morior to die, guerior to com-
plain, patior to suffer, potior to be master, miror to wonder.
The Atmanepada endings have been preserved in the Latin
Passive alone and so the original true Atmarepada verbs that
have survived in the language have been classified as the
regular deponents (George M. Lane, A Latin Grammar.
Sections 785, 79%. 1486). In Greek the distinction between
the two padas 1s very carefully observed in the earlier writings
and even in the later period the active and the middle
senses are clearly indicated. In the Avesta too the distinction
is very clear and in the metrical portion at least (7.e., in the
Gathas and the Yasts) has been carefully observed.

T1 awaadafsa waeaac) ales, eqftafer:

aafama frame, 7x, Rava waft qzeigqed|
21319
et] aa qeagaf) Aa cova | BIR le-2,
[A-A-WAeAAA | als. (AT Baath qegsed
fateqa Gat} =3leie ZTAT)
4A| Aasrat warmsq) aArarazat: |, fea:|
Hl) STAR! SET URRwe-29¢, Shear zy
1 \ule, fort: HAA] URCR, AMI VRIice
at! aiseaguqeng faq aq fie faq uaa faa
 NIVRTTISHANA AND SAMKRAMA 127

qa AG alate aramrafae ate afes_fSfesstiadeva-

qen-fafafar: ereie-z
21 awarnf naaq aaceageq, cof arnam
ATHAGTA | BIRIRS-Re
Ql AAMT: GAT AAR weal! alee Waa a
Gta, UAT AAA! 212%, wafra-frar: ate fiver
HS be WLR MITTAL ILA | VWRo, HAGTAaASeaAaray|
REA
W) wa 7a feasarteeagia, uae
ary, aa qeeagatfa | qeranaatqarfa | ¢ojo2-02
3 offs qais aqeaq aque) |oTima-
ARATE AS_A| zice-ce, Caftafaa: wd: eave firar-
RAT e,
HS 1 Blelee Hale TEATS] alee
T| Ga qa Cary aqay AKAATTA | 2IL2
| faarferaaarat qagarfo qeqzdenfa, sacract-
UATAGTaAHA | E1aCR-3CR

NIVRTTISTHANA AND SAMKRAMA.

In the earlier grammatical liteature there must have

been a fairly large number of technical terms ending in

e quay Taalaanasna amNINe TF aMAlaa fayaaaefa aaa,

aay aftaaqat BWR sqIeIeINEG a) Wau
aa alfa aal 1a Fea Yaaaaa |
we fadiqata wa gata gate
sfa nama ead) yaada1 fequ Aaealeay af safiras et Hale |
fecqaty |
 128 TECHNICAL TERMS OF SANSKRIT GRAMMAR

‘“cthana’’. Of these Nivrttisthana is preserved in the Nirukta,

Sarvanamasthana in Panini and Gunavrddhisthana in_ the
Katantra. Nivrttisthana is opposed o Gunavrddhisthana
and stands for Gunavrddhinivrttisthana. It means _ literally
the places i.e., endings (or suffixes) in which there is the
absence of Guna or Vrddhi, in other words weak endings or
suffixes. Thus we find in the Nirukta: athapy aster nivrttis-
thanesvadilopo bhavati ii. 1, ‘again the initial of the root as
is elided before weak endings.’
The corresponding rule in Panini is ‘“‘Snasor al-lopah”’
vi. 4. 111 where ‘kniti is obtained by anuvrtti from vi 4.
98 and ‘sarvadhatuke’ from vi. 4. 110.
K. has ‘“‘aster adeh’”’ ui. 4. 4] where ‘sarvadhatuke’ and
‘agune’ are obtained by anuvrtti from iil. 4. 39,
C. has “‘gnasor lopah”’ v. 3. 104, ‘apiti’ being obtained by’
anuvrtti from the previous rule.
J. has ‘“‘Snasah kham” iv. 4. 108 where ‘ge’ (meaning
sarvadhatuke) and ‘kniti’ are obtained by anuvrtti from the
preceding rule.
Sak. has “nam-astyor luk” iv. 2. 46 where ‘apah’ (meaning
‘apiti’) is obtained by anuvrtti from iv. 2. 36 and ‘git? (which
is tantamount to ‘sarvadhatuka’) from iv. 2. 43.
He. has “snastyor luk” iv. 2. 90 ‘giti’ by anuvriti from
the previous rule and ‘aviti’ (which corresponds to ‘apiti’ of
other systems) from iv. 2. 97. He. has -tiv, -siv, -miy for
-tip, -Sip, -mip Of most systems.
Sam. has “snastyor at” ii. 254 where ‘anit’ (corresponding
to ‘apit’ of most systems and ‘avit’ of Hc.) is obtained by
 NIVRTTISHANA AND SAMKRAMA 129

anuvrtti from ii. 248 and ‘catursu’ (corresponding to ‘sarva-

dhatuke’ of most systems) from ii. 249. Sam., it may be
mentioned in this connection, has -tin, -sint min etc. for -tip,
-SIp, -7p etc. of most systems. |
Mu. has ‘‘lopo’styasor nidrasor aghyam’’ 692 where
nit’ stands for weak-ending (‘apit’) and ‘ra’ for ‘sarvadhatuka.’
Su. has “Snam-astyor atah” iii. 3. 9. with the anuvrtti
of ‘nit sarvadhatuke’ from the previous rule.
P.R. has “asya lopo’tanadv asteh’” viii. 9. 4. where
‘‘atanau”’ stands for ‘agune’ or ‘kniti.’
According to P.R. the strong or ‘saguna’ or ‘pit’ endings
are “‘tanu,’’ while the weak or aguna’ or ‘apit’ endings are
‘“‘atanu”’ :

froteraget azfafatahranea:x | TRTAT: |

QuUL-t
“Tanu’ means ‘thin’ and one hardly expects to find this as
the term for the strong endings. It is just possible that
Purusottama designates the strong endings ‘‘tanu’’ because
they are unaccented. It is, however, very doubtful _ if
Purusottama was familiar with the Vedic accents. It is,
therefore, more’ probable that our author took the ‘t’. of
‘pit’ and ‘n’ and ‘u’ of ‘saguna’ and formed the word ‘tanuw’
with them. It is well-known that
in Bengal m and n are
pronounced alike. Probably Panini’s “‘ghi” arose in a similar
way from the ‘g’ and ‘i’ of “‘agni’’ and the aspirate element
in “‘sakhi’’.

*In PR. the ending for the Imperfect 3rd singular is di, and that
for the 2nd singular is s. 9. 151.

O.P. 129—9
130 TECHNICAL TERMS OF SANSKRIT GRAMMAR

HN. has ‘‘snam-astyor arama-haro nirgune”’ 1. 658. In

HN. ‘Rama means ‘varna’ and ‘Hara’ means ‘lopa’.
‘Nirgune’ stands, of course, “Nivrttisthane’.
In the Mahabhasya (i. 1. 3) we find Samkrama_ used in
the sense of weak endings in the passage :

area Farac waomal daa faarn afgarud |

qharsfea) oftasifa, wfeasrg, oftansteg) of-
aaa: Weaasigkaraay |
‘Other grammarians prescribe Vrddhi optionally in the
case of the root mrj before endings beginning with vowels.’
The Kasika paraphrases the words of the Bhasya and
explains Samkrama thus:
gacarl dma fare afefeeaa | daa are oer
afgafarataca: |
The Padamanjari gives the following explanation of
Samkrama :

aaTaatsTAAdt wast seatfsta Rear |

Kaiyata merely says:
qoafgofaratssa: oat aa |
‘““Samkrama”’ is also found in one of two* stanzas quoted
in the Kasika on vil. 1. 35, but not occuring in the Maha-
bhasya, nor commented upon in the Nyasa. The Padma-
fijari says: atafe texatata rast afaq qead |
scale few anaaq eneenfafada aa ag a: |
Stfeat Stfaara safaat q TWGAATS 1
das (ewmemeagtaaaamnfata: qa: |
4q aezaSieiai aa asefaniean: 4
 NIVRTTISHANA AND SAMKRAMA 131

Bhattoji says in his Sabdakaustubha:

danraat aTagg? semfefa aqaqren aafestans-

fafaayqahsaraaisa dnAneaisaa, BT: |
Samkrama is derived from the root kram ‘to step’ pre-
ceded by the upasarga sam ‘together’ and has been used in
the Vajasaneyt Samhita in the sense of ‘going or coming
together.” In VP. it is used to denote the meeting of two
words in the krama text (caused by omitting those between)’
It is therefore highly probable that the weak endings.
are known.as Samkrama because they generally come imme-
diately after the root just as it is, without its final vowel being
changed by Guna by or Vrddhi.
The rules corresponding to “mrjer ajadau samkrame
vibhasa vrddhim icchanti’’ from the different systems of
grammar are given below :

att [amafe:) sieges]

AT] ) astt arise: | ee
afar
a) wag, wasfraty ereie-2
31 aaiq, fawefrar) orere-e
*K. uses “gunavrddhisthane” in the rule “srji-drSoragamo’karah
svarat paro dhuti gunavrddhisthine” iii. 4. 25 which corresponds to
Panini’s “srji-drsor jhalyam akiti” vi. 1. 58. For “nivrttisthane” it
uses “agune”. Thus in the very next rule we find “dino’nto yakarah
svaradavagune” ti. 4. 26 corresponding to P&nini’s dino yud _ aci
kniti” vi, 4. 63.
132 TECHNICAL TERMS OF SANSKRIT GRAMMAR

amt) way avatar BiiRas-32

2, was ate | WESC AT] saieey
aq, wensagafa, 22331 Affe at, 2aze
gq asisafeta
frat art) aq
at, wavered tat a afzatea: fafa ferfa-
TAL AT |
aq, aaafe: aa afeeta gar aticc
g) vworsaferfa aie: Sry | eat J at) 2207-2
ge. waaciieg) 3a. warft-ahanel af g
areqaay | ata: | (geortex=aha: | aSal=ae ages )
Panini and his followers generally use knit to denote weak
endings, suffixes etc. Katyayana in his Véarttikas explains
clearly and concisely why two indicatory letters k and 7) are
used instead of one: Before the kit suffixes samprasarana takes
place in the case of certain roots (vaci-svapiyajadindm kiti vi.
1. 15) and before nit suffixes in the case of certain other roots
( grahi-jya-vayi-vyadhi-vasti-vicati-vrscati-prcchati-bhr jjatinam
niti ca vi. 1. 16). Thus we have Samprasarana in the case of
the past participle supta, because -kta is a suffix with an in-
dicatory k, but not in the case of svapitah, present 3rd dual: of
svap, because -tas is regarded as having an indicatory 7 and not
k (sarvadhatukam apit i. 2. 4.). Similarly in the case of jit
suffixes and infixes like nvanip., najin, can, an etc., no Sampra-
sarana takes place because the suffixes are nit and not Kit.
Thus though we have sina from the root svi with the kit
suffix kta we have asiSviyat (reduplicated aorist 3rd singular
of vi) without Samprasarana because the infix js cgp with
 NIVRTTISHANA AND SAMKRAMA 133

the indicatory #. Again the rule

“‘jagro’ vic-in-nal-nitsu. vii.
3. 85 prescribes guna in the case of the root jagr before kit
suffixes and not before nit suffixes, so we have jagaritah before
the kit suffix -kta but jagrtah before the nit ending -tas of the
present third dual*.
Sarvavarman uses the anubandha » for Panini’s 7 because
the anubandha 7 indicates that the root is Atmanepadin.
For the same reason he calls the nit sup endings yiavat instead
of nanubandha. He generally uses “gunin” for the strong
endings and “aguna” for the weak ones.
HN. has separate names for the pit, kit, nit and knit
endings :

fra ga: | firq afae:) fra afie: | ferferstor: |

fener fiver atari: | BIRGe-2
PR. uses “‘gunin’” and “‘aguna” for the strong and weak
endings respectively :

ANN sagas: AIST: | cau

J., Sak., He., Sar., Su. and others: follow Panini and. use
kit and nit. Since Sam. uses tin, sin, etc. for Panini’s tip, sip
etc., the anit of Sam. corresponds to apit of Panini.
Weaksuffixes are indicated by the anubandha n instead of
kK or nin K., probably acc. to the “lucus a non lucendo
”’
principle, since is part of ‘guna’ and before this 7, no guna

* See Varttikas
i. 2. 1. 8-11. Since the feminine suffixe
s nip, nis .

and nin are weak, the indicatory 1, though it

has been used by Panini
for another purpose, it very appropriate in these
cases.
134 TECHNICAL TERMS OF SANSKRIT GRAM MAR:

hardly probable.
takes place. It is also just possible, though
place’.
that m is the disguised form na, meaning ‘no guna takes

VIKARANA.
Both ‘‘vikarana”’ and ‘‘vikara’’ are used as technical terms
‘n Sanskrit grammar. Both are formed from the root kr
preceded by the upasarga vi meaning ‘to make different, trans-
form.’ “‘Vikara” is formed with the suffix ‘ghafi’ in the bhava-
vacya and means ‘change, transformation,’ while ‘‘vikarana”
like ‘“‘bhita-karana‘“, ‘‘hus-karana”’ etc. is formed in the
karana-vacya in the sense of «that by which thé sense or
shape is transformed or modified’, or in the kartr-vacya in the
sense of ‘that which transforms or modifies the sense or shape’.
It is generally used in grammatical literature to denote the con-
jugational characteristic inserted between the root and the suffix
or ending, or between the last vowel and the following con-
sonant of the root. In the system of Panini the Vikaranas
are Sap (iii. 1. 68), Sapo luk (il. 4. 72), Slu (ii. 4. 75), Sore:
(iii. 1. 69), Snu (iii. 1. 73), Sa (ill. 1. 77), Snam (iii. 1.78),
u (iii. 1. 79), $nd (iit.- 1. 81), nic (iil 1. 25), cin and yak
(iii. 1. 66-67), cli (iii. 1. 43 with its substitutes sic 44, ksq 45,
can 48 and an 52), tasi and sya (iil. 1. 33), sip (iii. 1.34), am
(iii. 1. 35), at and af (il. 4. 94)”.
Vikarana in this sense is found in the Paniniya Dhatu-
patha and in the Varttikas and Mahabhasya :

fracarat ata ad, Harhatad facta: BILISS1%,

* The at and at prescribed in vi. 4. 71-72 are Bhitakaranas.

 VIKARANA 135

fearsgaraa fanciteatssfatea: eieiecele, waafanc-

orafafastgen giao af wera: (arnahas)
elaixe, afz arafaaterd faacorer a oretrfa (set) 2121
The word is found fairly frequently in K. :
sa-fancm: wadf 312132 (af a fants argaed
adif a fared, faacores qatarisfasar! st);
ARAMA] ATHTCS FIRTH ypyiro (smear TIA-
Sfafa or eleies ), Ta fearon ava etc.
The Balamanorama says under “‘kartari sap” S. K. 2467:
faf= aq? ardaifafraat saat aamiat fanco-dar
TSA AIaTST |
Cf. Vrtti on K. iii. 2. 36: safaneqaarcutfaata a aee-
aa [=afaarcata] tert Haq |The Tika says: aTeag-
HATAA AEPTATAATTAAT A-AEAA |
K. has no separate Vikarana for juhotyadi roots which are
regarded as included in the adadi group. For tudadi alsa
no separate Vikarana is prescribed in K., an being added to
both bhvadi and tudadi roots. K. regards Gna as a separate
Vikarana for kryadi roots ending in consonants before the
Imperative second singular parasmaipada ending.
Su. has a special section entitled “‘atha dhator vikaranah”
iii. 2. 113 in which sya, sic, ksa, can, an, in, yak, Sap, Syan,
gnu, $a, Snam, u, $na and Snu are mentioned. The Makar-
anda does not take the trouble of explaining “vikarana’’.
The Pafijika says under iii. 2, 113: vikiryanta iti karmani
kirates tan-pratyayanta-ripam. Vikarana is also mentioned
by the author in the Vrtti on “subadyah pratyayah pare” ii. 2.2:
 (36 TECHNICAL TERMS OF SANSKRIT GRAMMAR

sup-sanadi-ladadi-vikarana-krt-taddhitah pratyayasamjnakah pra-

krteh pare bhavantili paribhasitah. It may be noted in this
connexion that Snam comes between the vowel of the root and
the following consonant.
PR. mentions Vikarana in ‘‘an-vikarana bhuvadayah”’ viii.
47 which is explained in Gudha-prakasika thus: ano visesena
karanam yebhyas te bhuvadaya ityarthah. HN. says in the
Vrtti on “Sap Krsnadhatuke” 1. 351: vikaranakhyo’yam.
In philology the Vikaranas are known as Root-determi-
natives. The parent Indo-European language seems to have
possessed no less than 32 such determinatives as enumerated
by Brugmann, and Sanskrit 14.

The substitute ‘cha’ for the finals of the roots is, gam and
yam prescribed by Panini in the rule “‘isu-gami-yamam chah”
vii. 3. 77 as also in the substitute rccha for the eat T ee
cribed in the rule “pa-ghra....”’ viii. 3. 78 is rightly regarded
29

as a Vikarana by Western scholars. The 4m of the periphrastic

perfect is not generally regarded as a Vikarana, but there is
no reason why it should not be counted among the Vikaranas
when cin and cli and its substitutes are looked upon as
Vikaranas.

Some Indian grammarians regard the union-vowel i as a

Vikarana.
Indian professors of Sanskrit belonging to the old school
generally describe a Vikarana as that which comes between
the root and the suffix and is absolutely without any meaning
This is not a logical definition since it excludes énam. but a
mere description.
 VIKARANA 137

In the Astadhyayi of Panini the Vikaranas are read under

the general heading “‘Pratyayah”’ iii. 1. 1. Kaiyata says unden
‘““vikaragamesu ca paravijnanat” iii. 1. 1. 4:
qara GreaTMaashe sqQaETHAl caer vata |
Thus he also regards the Vikarana ‘‘gnam” as a_ suffix.
thougn it does not come at the end of the base.
That these are not regarded as ordinary suffixes is clear
from the fact that the Bhasyakara speaks of bahuc and akac but
ignores Snam in the passage :

fanafaayaaa 9
qu wa ata, Fat ar yfefa |
Aagka aay) afiesf aig, ae a—“firare
BT Fea FraT” [ 4aie¢ ] afer |
ma afe at afeta |
mast afasfa one aq—‘seaz-aaareTHaa OTA
z:” [ agnse ] afar l) & tarearatsea: sepa: Se: AISA
qaa ot og afasadifa aren qeaats |
The question is asked, what purpose is served by the rule
‘paras ca” ili, 1. 2, ‘a suffix comes after a base’? The
answer is given that this prevents the suffix from coming at
the beginning or in the middle of a base. As against this
it is pointed out that when the author wants a suffix to come
in the beginning he says this in so many words, as for instance,
when he lays down that the suffix “bahuc” is added optionally
to an inflected word in the sense of “‘isad asamapta” and is
placed at the beginning of the base. Again when Panini
wants a suffix to come in the middle of a base, he distinctly
 138 TECHNICAL TERMS OF SANSKRIT GRAMMAR

says so, as in the case of the rule: (the suffix) ‘‘akac’’ comes
in before the ti-portion of indeclinables and pronouns. All
suffixes other than these will, therefore, naturally come at the
end and no special rule is needed for that purpose. So it is
no.use laying down that suffixes will come in at the end of the
base.
It would appear from the above discussion that accor-
ding to Patafijali the only suffix that comes in the beginning
is ‘“‘bahuc’’ and the only one that comes in the middle is
“‘akac.”’
Of the followers of Panini, Padmanabha distinctly mentions
the Vikranas in the list of Pratyayas.
The word Vikarana is used probably more frequently in
K. than in any other system of grammar. Here also the
Vikaranas are read under the general heading ‘“‘pratyayah
parah” iii. 2. 1. and are consequently regarded as suffixes.
Panini and other grammarians regard num as an agama
in the case of mucadi roots (Se mucddinam P,. vij 1. 59).
But this 1 is certainly a Vikarana. We find similar thematic
roots with the nasal infix in Latin iu-n-g6, Greek lambané, etc.
‘“Vikarana” is sometimes used in its ordinary sense of
‘modification’ in grammatical literature also. Thus we find
in Ananta Bhatta’s commentary on VP. iii. 136:

Sey maacrafaearan—
Sata Haleanfsacn
aes |
are satanastt aac AIT|
afa | aera —aarcaarcaanracqen faa af sq-
 GENDER 139

witaqayq sara argafart sat) sravafamaaed

aa Hay |
‘‘Akhyata-pada-vikaranah”’ is used in the sense of ‘words
which modify the finite verb, i.e., make it accented.’
In literature Vikarana is found in stanza 317 of the
Kuttanimata :

sfaged afafear: weratr fafea-fanconaan: |

qTQsaaers: THAT La THE afore: |
‘Courtesans are unfathomable like the roots (bases) in
grammar. They approach every man (they are used in all
the three persons), they are intent on serving their own
interest (they have a predilection for the gerundive suffixes),
they are endowed with various emotions and passions (they
take on various Vikaranas or determinatives) and they exact
a vast amount of money (they have various senses).’

GENDER, NUMBER,. PERSON

Gender.
The word ‘“‘linga’”’ is used the Upanisads in the sense
of ‘mark, token, characteristic.’ In the Nirukta (i. 17) also
the word is used in the sense of characteristic. From this, on
the one hand “‘linga’’ came to mean ‘the characteristic of the
male, phallus,’ on the other hand it came to signify ‘gram-
matical gender.’
“Linga’’ is used in the sense of ‘SabdasAmarthya’ in the
Mimamsa-sttra and in Bhartrhari’s well-known stanza:
damit fasanrr mead faorazar |
ard: sant fos” severed afafs:1)
140 TECHNICAL TERMS OF SANSKRIT GRAMMAR

In the next stanza Bhartrhari uses ‘vyakti’ in the sense of

‘gender’ :

aimetafata 2a: aracafheatiza: |

meqiaemiaquse Pratreafaeaa: |
In AV. xu. 3. 1. we find “puman pumso’ dhitistha
carmehi,’”’ ‘mount male from male, the skin.’ Griffith tran-
slates as above and explains in a note: ‘‘Aégvattha with a
masculine name, that has grown on Khadira which jis also a
masculine word.’” ‘This sems clearly to point to grammatical
gender which is here identified with sex.
In AB. (xxvi. 3) we find:

azie:—_fa gaara qaercarafafa | amare a cio

ant a aa a aaa ate | aE: Sere guid aed
cifrada tft ani fa qaeafa
aaaa |
‘Why has the Subrahmanya its name? ‘It js speech,’
he should reply; “speech 1s the holy power and the good
holy power.’ They say, “Why then do they call him that is
male female as it were?’ “Because the Subrahmanya is speech,’
he should reply, ‘for that reason.”
It is clear from the above that the distinction of gender
was perfectly well-known to the author of the Brahmana and
that the neuter was often regarded as merely dticther form
of the masculine. .
Coming now to the words for the different genders
‘vrsan’ and ‘yosa’ are used in AB. and AA. for masculine
and feminine genders respectively: rsd va rsabho yosa
Subrahmanya (AB. xxvi. 3), ‘the bull is male (masculine),
 GENDER 141

the Subrahmanya female (feminine)’*. rsa vai prefkho

yosasandi A A. i. 2.4, the swing is masculine, the seat
feminine.’
All the three words “pums,” “str?” and ‘‘napumsaka”’
found in later treatises on grammar appear to occur for the
first time in the SB:
ittham ha tv evapi tredha vihito yad asmims tredha vihita
istakA. upadhiyante—punnamnyah — stringmnyo napumsaka-
namnyah. tredha vihitanyu evemani purusasyangani—Punna-
mani strinamani, napumsaknamani >x. 5.1. 2...... ta. u
sarva istaka ity evacaksate, nestaka iti nestakam iti vaco
rupena. vag ghyevaitat sarvam yat stri puman napumsakam.
vaca hy evaitat sarvam aptam. tasmad ena angirasvad
dhruva sidety eva sarvah sddayati, nangirasvad dhruvah
sideti, nangirasvad dhruvam sideti. vacam hy evaitam sams-
kurute. x. 5. 1. 3. “But in this respect also it is three-fold,
inasmuch as three kinds of bricks are put into it—those with
masculine names, those with feminine names, and those
with neuter names...... Now all these (bricks) are called
istaka (f.), not istakah (m.), nor istakam (n.): thus (they
are called) after the form of speech (vac f.), for everything
here is speech—whether feminine (famale), masculine (male),
or neuter—for by speech everything here is obtained. There-
fore he settles all the bricks with Angiras-like lie thou steady
(dhrieva, f.)! not with Angiras-like lie thou steady (dhruvah,
m.) ! or with, Angiras-like lie thou steady (dhruvam, n.)! for
it is the speech he. is constructing.”
* Cf. SB.i.3.1.9: Hara aa, Ta aa i. 5. 3, 15: eafeaf
a % am,
feat war }
142 TECHNICAL TERMS OF SANSKRIT GRAM MAR

The idea of gender is also very clearly expressed in the

following akhyayika (SB. i. 5. 4. 6-11) : Devas and Asuras,
both sons of Prajapati, competed for mastery. They fought
with clubs and bows but could not win. Not winning, they
said, ** Ah, let us gain victory in words. He, among us,
who will be unable to follow up our uttered speech by
(making up) a pair, will lose everything and the other party
will win everything.” The Gods agreed and asked Indra to
speak. Indra said “‘eko mama,”
>?
the others said ‘‘asmakam
eka,’’ eka being the corresponding feminine form of eka.
After a few more sentences Indra dropped his atomic bomb
in the form of ‘‘pafica mama” and the Asuras racked their
brains .o find out the feminine form of “‘panca,’’ but in vain,
as “‘pafica” has the same form for all the three genders.
Thus the Asuras lost the battle and the Gods were victorious.

Zara aT aces
setae Teafaze | A avSjahra
saa | & erfsrrata ae:— ard aaa aera fafaaitar-
ne, @ at a art cared faada arahtamra, a af
quis, wa aaferat serferfer | ‘aoa af Fear aa) a
2at gana aa— ‘caret’ tia | | FR Saetta— ‘aa aa’
cfa) aa ‘aermaar tdtlatsrat) ag aferaanar-
fara) Tau fe—apsatat a, ‘al an’ Tae SHAT |
aa ‘aes F ealata
aa | ag afraatarfeega |frre
fa aga) ‘aa aa’ xdtegisadiq) se ‘crema
faa: edatsaaa| ag afeaaaatarfras| frat fx—
aaa faaqa| ‘acat aa’ edieaisaaia | ate ‘aes
ada: tdtatsa at) ag afawararfiears fiat fe—
 GENDER 143

Tata dda | “Wy AA’ gdirgisadla | ad zat faut

Tf, Tad we fraanfe—iq—qy xfadaagaa
nafa | dadtisau: aa Tamed aaa AWUATIA |
AACA ATAA GU FATA |
The Gopatha Brahmana distinctly mentions “‘trisu lingesu’”’
in its definition of “‘avyaya’’ (i. 1. 26).
‘Similarly we find in the Sankhayana Aranyaka (iii. 7):
‘“He says to him, “How dost thou obtain my male names ?’
‘By breath,” he should reply. “How female names?’ By
speech. ‘How neuter (names)? ‘By. mind.’
Here is a very clear reference to grammatical gender
because the words prana, vak and manah are masculine,
feminine and neuter respectively.
The Nirukta (iil. 21) also clearly refers to the three
genders in the following passage :—
ayainetyu padesasya.
aya te agne samidha vidhema (RV. iv 4. 15) iti striyah.
ena no agnim (RV. vil. 16. 1) iti napumsakasya.
end patya tanvam samsrjasva (RV. x. 85. 27) iti pumsah
“Aya and end.are (synonyms) of reference.
‘With this faggot we worship thee, O Agni.’
Here it is in the feminine gender.
‘With this, to us, O Agni.’
Here it is in the neuter gender.
‘With this husband commingle thy body.’
Here it is in the masculine gender.” Sarup.
Coming now to the PratiSikhyas, ‘‘pumspravada”’ is
144 TECHNICAL TERMS OF SANSKRIT GRAMMAR

used in RP. iv. 42* in the sense of ‘masculine. Again in

RP, Xi, 22.
napumsakam yad usmantam tasya bahvabhidhanajah
anusvaro dirgha-purvah sisyantesu padesu sah
VP. ii 32 maho napumsake etc., CA. i. 84 dirgho napumsaka-
bahuvacane, etc. and APr. i. 17 napumsakam dAdyudattam
brahmeti etc., ‘“‘napumsaka’”’ is used in the sense of the
‘neuter gender.’
The generic term “‘linga’’ is used in the sense of gender’ in
VP. iv. 175, APr. i. 22, etc. BD. (ii. 96) has: trinyeva ioke
lingani puman stri ca napumsakam.
“The distinction of genders,” says Max Miller ‘is
the
only point on which the Greeks may claim a priority to
the Hindus. Jt was known in Greece to Protagoras: whereas
in India the Pratisakhyas seem to have passed it over dnd
it appears first in Panini.” (History of Ancient Sanskrit
’

Literature, ed. Panini Office, p. 85.)

Now Protagoras was a Greek philosopher who flourished
about 440 B.C. and was in all probability a contemporary
of Panini. He used} arrén (male, Skt. vrsan) for the mascu-

+ garg fanesarega: ofan met yw arg)”

gaa yaaa) sae
+ Arrena kai thélea kai skeué. ARISTOTLE, Rhetoric. Cf. arsen-
‘kon kai thélukon kai oudeteron. DIONYSIUS HALLICARNASSENSIS.
“He (Protagoras) had also divided nouns into three classes, male,
female, and inanimate (skeué), a classification apparently founded on
4 real or natural, and not on a grammatical basis, ‘male’ and ‘female’
nouns denoting male and female persons, or distinction in sex, whether
in mankind or among animals in general, and things inanimate in-
cluding the names of all other objects, natural and artificial. real and
 GENDER 145

line (Lat. masculinum*) and thélus (female, Skt. yosa) for

the feminine (Lat. femininum) and skeuos
(implement, in-
animate object, things) for the neuter.* We have shown
above that the distinction of genders goes back to the
Brahmana period and that the earliest name for the mascu-
line was ‘“‘vrsan’? and that for the feminine ‘“‘yosa.’”’ We
have also shown that far from gender being passed over in
the Pratigakhyas, most of the Pratigakhyas as well as the
Nirukta show a full acquaintance with grammatical gender.
Some predecessors of Panini use ‘“‘vyakti’} for “‘linga”’
as is clear from P&nini’s rule: lupi yuktavad vyakti-vacane
1. 2. 51 on which Jayaditya says: vyakti-vacane iti ca linga-
samkhyayoh piirvacdrya-nirdegah, tadiyam evedam sitram.
It may be mentioned in this connexion that the word
““vyafijana”’ is derived from the root arij preceded by the
upasarga wi with the primary suffix ana, just as “‘vyakti’’ is
derived from the same root with the suffix fi, and that
———
——— —

abstract. This last class contains many words which are grammati-
cally masculine or feminine, but the classification of Protagoras can
hardly be identified _with a classification of nouns as masculine,
feminine and neuter. Protagoras uses in the sense of ‘classes’ the
Same term (géné), which was afterwards adopted in grammar to
denote ‘genders’” Sandys, A History of Classical Scholarship, i. 91.
*In the middle of the first centrury B.C. the Gender or genus of
a noun or’ nomen substantivum was distinguished by the terms virile,
muliebre and neutrum (masculinum and femininum not occurring
earlier than the second century A.D.)” Sandys, A History of Classical
Scholarship, i. 182.
»f In Naisadha iii. 23 “vyakti” is used in the sense of ‘vibhakti.’
O.P. 129—10 ~
 146 TECHNICAL TERMS OF SANSKRIT GRAMMAR

‘“‘vyafijana”” is used in the Srauta and Grhya Sitras in the

sense of ‘the mark of sex or gender’, ‘the private organs.’
J. shortened ‘‘napumsaka” to ‘‘nap” which, however,
had been in use for a long time. Thus Santanavacarya says in
his Phitsutra : nab-visayasyanisantasya ii. 3.
Sak, uses “‘nap”’ in i. 2. 1 he said ‘‘napumsaka”’ in ii. 1. 69,
154. te.
‘““Kliba”’ is a good old word found for the first time in
AV.: klibam kliba tvakaram vadhre vadhrim tvakaram Vi.
138. 3. It was not, however, used in grammatical literature
till a fairly late period. It is found probably for the first
time in Hc. who generally -uses it, but ‘‘napumsaka’’ is used
in the rule “‘napumsakad va’’ vii. 3. 89 (corresponding to
Panini’s ‘“‘napumsakad anyatarasyam” v. 4. 109). Mu. fol-
lows He. In Su. “napumsaka” is used in the sitras, prob-
ably out of dererence to Panini, while ‘‘kliba’ is used in
the Vrtti. Thus we find the rule ‘“‘svamor luk napumsake”
Su. ul. 3. 59 the Vrtti on which runs thus: klibe svamor luk
syat.
“Kliba” and ‘“‘visarga’’ are used by earlier writers also.
Thus “kliba”’ is found. frequently in the Amarakosa of Amara-
simha, to whom an honoured place among the grammarians
has been given by Vopadeva in his Kavikalpadruma. At the
very beginning of his work Amara mentions the three genders
stri, pums and napumsaka (i. 1. 3), but uses kliba in j. 1. 6.
‘‘Klibata” is found in the Kasika of Jayaditya who, it is interest-
ing to note in this connexion, makes a half-hearted attempt to
defend ‘kutra’” found in Amara. “Klibata” occurs in the
Kasika on ii. 4. 18 in the three Varttikas: (1) punyasudina-
 GENDER 147

bhyam ahnah klibatesyate. (2) pathah samkhyavyayadeh

Klibatesyate. (3). kriyaviSesananam ca klibatesyate. In the
Present state of our knowledge it is difficult to determine
Whether these are Jayadditya’s own Varttikas or borrowed from
some previous work. In the form in which they occur in Mbh.,
C. and J. we find napurhsaka and nap. Similarly “‘visarga’’,
as M. Renou points out, occurs in the Pratijiia-parisistasttra:
aS.a generic term for the three voiceless spirants—visarjaniya,
jihvamilliya and upadhmaniya ; and both visarga and visarjaniya
are found in the Puspasititra. It is probable that visarga was
at first used for all the three voiceless spirants and only latter
came to be restricted to the visarjaniya. The. reason why
Srammarians previous to Hc., bent as they were on securing
laghava, should have fought shy of ‘these two short terms
‘*kliba” and ‘“‘visarga” in their siitras would appear to be that
they were not deemed sufficiently elegant for a noble science
like the Sabdanusdsana, kliba having too‘much to do with sex
and visarga suggesting the voiding of excrement. Hence these
words came to be widely used when the nice nuances of
words had been altogether lost sight of and when one synonym
was deemed as good as another.
HN. use ‘“Purusottamalinga’” for “pumlinga’, “‘Laksmi-
linga” for ‘“‘stri-linga’’ and “‘Brahma-linga’’ for ‘napumsaka-
linga.””
‘“Linga” is used by K. and its followers in the sense:of
“pratipadika.’ |
In the following well-known example of the figure of
speech known as Visama the word “napumsaka” has been
yery cleverly used ;
 148 TECHNICAL TERMS OF SANSKRIT GRAMMAR

agente ara fara ofead aa |

ad aaa cae er: arforfrar aa 1
A swain who had lost his heart to his beloved cries out
in the agony of his soul: ‘I sent my mind as an emissary
to my beloved, knowing it to be neuter. It, however, dallies
in her company. I have to thank P&nini for this calamity.’
The reference here is to the rule “‘as-anto dvyackah” 151 of
the Linganusasana, the authorship of which is attributed to
Panini. According to this rule dissyllabic words ending in as
are neuter.
It is a pity our poet turned to grammarians for advice in
such matters. Had he consulted the Vedas he would have been
disillusioned in no time, for according to the Brahmanas,
mind is male and speech female :
aufe ma: SB. i 4. 4. 3, are fe arm Ib. i. 4. 4. 4.
Another poet supports Panini :
frarge aq dea Tsqrefee
ataag |
a aifa wea ta AAA AGE ||
Subhasitavali 1195.
The fond lover feeling the pangs of separation from
his ladylove laments: “It is because the remembrance of
that face of the beloved with its long beautiful eyes does not
split my mind into a hundred pieces that the mind js (con-
sidered) a enunch (neuter).”’

Be
epsrcinni
ne!come
* am araaa ata aa: ar a quafa |
fa fsa aaa ara SIRT Aaaae |
 NUMBER 149

The following stanza from Murari’s Anargha-Raghava

(vil. 39) will be read with interest in this connexion:

FFAG AA CSS TANT AM frag arcs SAAT |

Hraqaniartasgn Waa: goad wefaagat i
‘May the (Ardha-narisvara) form of Siva belonging to
the fourth sex transcending the distinction of male, female.
and eunuch make you happy, the form in which the decora-
tions of musk and sandal on the round left breast wet with
sweat are dried up by the dust of ashes in the fingers of the
right hand.’
The word “prakrti’ is used in the same sense in the
following ingenious stanza of the Naisadha-carita (xvii. 70) :

Sarat wate: aa asafefa qaaagq|

aqat adefa ana: afracty |)
‘Both sexes should tread the primrose paths of dalliance”
—this is also the view of Panini who- lays down ‘‘apavarge
trtiya’’, #.e., the third sex or eunuch (is entitled to exertion) in
the matter of salvation.
The rule “‘apavarge trtiya” ii. 3. 6 really means that ‘the
accusative of time and place is put in the instrumental when
the completion of the action and the attainment of the fruit
thereof are indicated.’

Number.

A clear reference to Vacana is found in AB. vi. 6: sa

yady eka-devatyah pasuh syan medhapataya iti briiyat, yadi
dvi-devatyo medhapatibhyam iti, yadi bahu-devatyo medha-
patibhya ityeva sthitam. ‘If the victim be for one deity, he
 150 TECHNICAL TERMS OF SANSKRIT GRAMMAR

should say medhapataye ; if for two deities medhapatibhyam ;

if for many deities, medhapatibhyah.’
Similarly we find in TB: ‘na mata vardhate, na pita,”
that is when there are more than one victim, the singular, in
“mata”? and “‘pita’’ in the formula “‘anvenam mata manya-
tam anu pita” is not to be changed into the dual or plural.
Keith points out (Rigveda Brahmanas, p. 80) that ‘bahu’
is used in the sense of plural in AB. xxi. 2,15. The passage
runs thus :

ITA Feat waste zfa aed aehaeqsery, art

q ag, acnegdiaae: | (@ehrag afaengeay afaegecoia
sada waa aad at ag aang) aa fe
qeazaa: aadiar sa ageayq | aga Favarafasniecoiaiai
fagaracaly FEA) at agaetReatgariesiad |
ATA: )
Keith translates: ‘Pouring showers, the Maruts, of
daring might’ is (the hymn) to the Maruts with much to be
recited : what is much is the end; the third day is the end.
‘“Ekavacana’’ and “‘bahuvacana”’ are found first in §B.:
ned ekavacanena bahuvacanam vyavayama xiii. 5. 1. 18,
‘lest we should over-ride the plural by the singular’.
‘““Dvivacana’’ has not been noted. before the time of Yaska:
api va medasas ca pasos ca sattvam dvivacanam syad_ yatra
hyekavacanarthah prasiddham tad bhavati vi 16, saptamya
ekavacananiti Sakapunih iv. 15, vayo ver bahuvacanam iv.
3, ekasya eva pujanarthe bahuvacanam syat xii. 7, tad etad
rbhos ca bahuvacanena somasya ca samstavena bahini
 NUMBER 151

dasatayisu suktani bhavanti xi. 16. Referring to the duals

and plurals in the hymns Ydaska says: ‘‘api dvivad api
bahuvat tad yad dvivat tad uparistad vyakhyasyamah u. 25,
(dvivat dvivacanasamyuktaih mantraih. Durga). ‘‘Ekavat’’,
‘“dvivat’”’ and “‘bahuvat” occur in BD. in sense of ‘in the singular’
‘in the dual’ and ‘in the plural’ respectively: pravadas tatra
drsyate dvivad bahuvad ekavat iv. 107. The generic term!
‘““Vacana” occurs in “bhede vacanalingayoh” i. 43. Panini
follows his predecessors and uses “ekavacana’’, “dvivacana’’
and “‘bahuvacana.” Later grammarians generally follow
Panini, shortening ‘“‘ekavacana,” “dvivacana’” and ‘“‘bahuva-
cana” into “eka,” “dvi,” and “‘bahu’’—a practice followed by
Yaska and Saunaka also.
RP. has ‘‘dvi-vacah” for ‘“‘dvi-vacana” in i. 71. CA. ii.
47 has aantafead al feaaaeae |
K. uses “dvivacana” and ‘“‘bahuvacana” in the first
chapter in the rules “dvivacanam anau” “‘bahuvacanam ami”
i. 3. 2-3, but in the subsequent chapters it generally uses
““ekatva’”’, ““dvitva’” and “‘bahutva;” e.g., vam nau dvitve ii.
3. 2, tvanmador ekatve te me...... li: 3. 3, at paficamya advitve
3. 14, ed bahutve tvi 1. 3. 42; ridhandm bahutve....ii. 4. 5,
dvitva-bahutvayos ca parasmai iv. 1. 19, sano” lopah svare’
bahutve iv. 2, 33. “Dvivacin” is used only once in the rule
“‘vuvavau dvivacisu’”’ u. 3. 7. “Ekavacana” is used in the rules
“kriya-samabhihare sarvakalesu madhyamaikavacanam pafica-
myah’’ iii. 1. 21 and “‘ijatmanepade prathamaikavacane”’ (iii.
2. 29), probably because “‘ekatva’’ does not sound well with
‘“‘madhyama” or “‘prathama.”’
1SZ TECHNICAL TERMS OF SANSKRIT GRAMMAR

In Greek the generic term for Number is arithmos

(number, Samkhya). The name for the singular, dual and
plural are henikos, duikos and pléthuntikos. In Latin, the
generic term for Number is numerus. The name for the
singular is singularis and for the plural multitudinis in Varro
(116-27 B.C.). ‘“‘Pluralis is found later in Quintilian (circa
A.D. 35-95) who represents the teaching of Remmius Palae-
mon) and plurativus in Gellius’” (2nd century A.D.).*
For the generic term ‘“‘vacana” J. uses the popular word
“Samkhya.”’ For “ekavacana,” ‘“‘dvivacana’’ and ‘“‘bahuva-
cana,’ however, J. coined no special term but used ‘‘eka,”’
“dvi” and “‘bahu” like most other grammarians. Vopadeva
does not follow J. in this particular instance, but uses ‘‘dva’’
for “dvivacana” and frames “‘kva” for “ekvacana’’ combin-
ing the ‘k’ of the first component and ‘va’ of the second of
“ekavacana,” and “vva’’ for “bahuvacana”’ after the analogy
of ‘‘dva.” Candra generally uses “ekatva,”’ ‘‘dvitva’ and
“bahutva’” for “ekavacana,” “‘dvivacana”’ and ‘‘bahuvacana”’
respectively. The bigger terms are found in “‘ekavacanasya te
me” vi. 3. 18, ‘“‘id-tid-ed dvivacanam” v. 1. 125, ‘‘mantasya
yuvavau dvivacane” v. 4. 58, “bahuvacanasya vas-nasau”’
Vic Sy 17 -ete.
‘“Vacana’’ means ‘word, expression.’ It is reasonable
to suppose that “vacana’ as a technical terms was used
with ‘‘eka,”’ “dvi” and “bahu” and then separated from these
and used as a generic term.

* Sandys, A History of Classical Scholarship, i. 182.

 PERSON 153

‘““Vacana”’ appears to have been used in the sense of verbal

endings in K. iv. 2. 88: aa tc qaatz: !

| aetnyaa-eaqaa-qeqaaraqnw: VWz12°%
a1 wafeagy eye
SH) GR feral | BWA) Rec a-eee
a un-fe-aal eaiec
Gt WHfE-AET BIBI -
@| wafsagaad saaetanq) ces
ah attra) ¢ ake faraar) apf ah
agra) sa: |
Bl w-s-cartatsaesan: | 23
&: [=faad: ] cat gad warm a-carda’ car
aa a amtafragedie cera | ah |
T\ swHAdl ata qeartasagaaarta |
arta frags raeaay FAST ASML URGE
El qaswaracarad@) fsaaamal) aga
WA] R1e4R TRA |

Person.

Purusa in the grammatical sense of “person” appears first

to have been used by Yaska :

aq qian: walfiatafiafefirasasa senaqea-

ATSATACT |...8T TAHA waagesaineafta aaa
aa |...acuren
afrser stageadin aefafe Alta
TIATAT | 912
 TECHNICAL TERMS OF SANSKRIT GRAMMAR
154

“Of these the indirectly addressed (hymns) are con-

nected with all the cases of nouns and with the third person
of the verb...... Now the directly addressed stanzas are
connected with the second person and with the ‘pronoun
tvaMm...... Now the self-invocations are connected with the
first person and the pronoun aham.”
camfrgs geadar q smiararinetaata aera, i
ATSAAICAT | 2Ge
“Prathama,” ‘“madhyama” and “‘uttama” are often used
in the same order in the Brahmanas : ala WAATAT Teta i

yd wag afrerq—araaced| ala, mantra efa!

ara sat weacaaaqed | atgdarat tefa—ag
alan: | wag atag cfafagta | AMSAARTAIANA 9) 919 «
Panini drops the purusa and uses “prathama,” ‘“‘madhya-
ma’ and ‘“‘uttama’” instead. Later grammarians generally
follow Panini. “Prathama” means first, ‘‘madhyama’’
middle and “uttama” last. This is the reason why in con-
jugation we begin with the third person and end with the
first person.
K. generally uses ‘“‘prathama,” “‘madhyama” and ‘“‘utta-
ma” like Panini, but K. uses the generic term in the rule
‘““wugapad-vacane parah purusanam.”
Candra referes to the order of the endings by ‘‘prathama,”’
‘‘madhyama”’ and “‘uttama.”
It would appear from the Nirukta that purusa or person
was first used with reference to the verb and then extended
to pronouns and then to nouns,
 PERSON 155
_

The three persons are mentioned in their proper order

in the description of the Rajakula in the Kadambari (para 85)
in a passage which has been quoted on p. 69 and which may
be translated thus: happy- on account of vast expenditure in
the matter of making gifts as advised by many officers divided
into the three categories of ordinary, middling and best persons,
eyen as grammar is well-established by the detailed treatment
of the many substitutes for the endings of the first, second and
third persons, of cases, finite verbs, datives, verbs, and
indeclinables.
In the Upanisads and Gita “Uttama Purusa’’? means the
Supreme Spirit or Paramaétman.
Mandra, Madhyama and Uttama are used in ASS. i. 5. 25,
27, 28, RP. xiii. 42 etc. in the sense of the three qualities
of the voice viz., soft, middle and loud respectively.

wT) fasatir att samaeqaraary =2etgog

qaqa amatsact afeafe asa: | wea a
ANTI AeGaeTA CHAM] AATETA: | WT TTA: |
UII e4-Qo?
at) Stfir aif saamearaa | «gas qe:
gear | aif sasaarasta saaq:, geafe wea: |
ASAD TA: | 3813-9
a1 gate aeqAaaa | AAT AAA| cleleee-evo
samad g amrafafed maaimana qacefsaa
atfuacarsasatage | ata:|
| firefarattsenqasnze | Rize
I | sissameagera far aa fa fac aa a faa aa
AE eelk
 156 TECHNICAL TERMS OF SANSKRIT GRAMMAR

| faa see aaa | aeaMTAsfa

aif TART BATeATTA ALTATAAY | 2126-219
A) fé-qeazeafe ware fer | eet
G1 fasten: seammeratrran: | arf sera: | gent
WA; | AMAA!) 21¢130-33
| Fa ag car AwaAaATTARAAT: |
aTfa Sasrarmste sta: geq: Sara: |
Wea TAT NR: ACA AATET |) 122-43
1 aaKT att aif sanaqurages-daanrfa |
RIRRR

IT AND ANUBANDHA.

Anubandha.

‘“Anubandha”’ (from anu ‘after’ and bandh ‘to ‘bind’)

literally means ‘that which is tagged on’ and is used in
grammatical literature to denote a letter or syllable attached
to roots, suffixes, augments, subsitutes etc. to indicate their
accentuation or some grammatical operation in connexion with
them or to distinguish them from similar other things and so
on. ‘Though anu generally means ‘after,’ an anubandha need
not necessarily come at the end of the base, suffix etc. to which
it is attached, it may come and often does come at the beginni
ng
of these.* This led Bharavi to use the simile “prakrti
-
pratyayayor ivanubandhah”’ (Kirat xvii. 19)7.
—_ —
ES

«aie zea) UwMae)

—— ae

Fy) ahaa) Pay

{See pp. 78-79.
 IT 157

Though “‘anubandha” appears to be a very ancient term

it is used first in the Varttikas of Katyayana, e.g., anubandha-
katanarthas ca (Paspasa xvi). It is next found in K. where it
occurs for the first time in the rule ‘‘4gama udanubandhah
svarad antyat parah” ji. 1. 6, but curiosly enough the word is
not defined at the beginning of Chapter II, but towards the end
of the last chapter. This would lead one to suppose that the
word was too well understood in Sarvavarman’s time to need
any explanation, even as ‘“‘upadha”’ was well-known in the days
of RP. At the time of the revision of Sarvavarman’s original
work the word ‘“‘anubandha” had probably been entirely outsted
by the monosyllabic ‘“‘it” and was therefore defined at very
nearly the end of the treatise in the rule ‘“‘yo’nubandho’ prayogi’’.
The commentator Durga explains that the root bandh preceded
by the upasarga anu is used here in the sense of ‘elision’ and
Supports his explanation with the quotation “‘agnisomiyo gaur
anubadhyatam” !!! And Trilocana borrows this brilliant
exposition and embodies it in his Pafiji!
The best explanation of anubandha is probably that offered
by Ksirasvamin :
Tatra a: aatdarasad farar wafia-
TAS STIs: |

It.

Panini and his followers, bent on securing brevity, use

the monsyllabic ‘‘it” for the polysyllabic ‘‘anubandha.” Indian
commentators derive the word from the root i ‘to go’ with
the
help of the agent sufix kvip thus: eti gacchatiti it. Many
A583" TECHNICAL TERMS OF SANSKRIT GRAMMAR

European scholars believe the word to be a corruption of

*iti.”* The fact that in some of the Pratisakhyas letters are
often indicated by putting ‘iti’ after themft lends support to
this view. In VP. iv. 35 and vii. 9* a riphita visarjaniya is
termed “rid visarjaniya.”” Here “rit”?
C68499
would appear to be
merely an abbreviation of “‘riphita’’.
Panini .does not define or explain the word, but uses it
for the first time in his rule “‘upadese’ janundasika it” ji. 3. 2.
and prescribes the elision of the indicatory letters in his
rule “‘tasya lopah”’ 1. 3. 9.
J., Sak., He., Sam., Su. and others use “‘it”’ throughout.
H.N. uses both “it”? and “‘anubandha’’ without defining them;
e.g., ir-anubandhan no va bhuteSa-parapade i. 450 and iramed
dhator num i. 455.
He. uses “‘anubandha” in the Brhadvrtti on i. on i. 1. 17 -
C-AD-2-S-0 AGEN: | (agqrad arqefea cag-
TIT: | Wyrare: )
The Haimaprakasa explains “‘it’’ thus:
AN? ATAGE! A Ws BATH |
C¢ . ~

a 22aa TAT @: Se aAAH: Bsa: ||

anne: re
*“Dr. Biihler has suggested that ‘it’ is for ‘iti,’ a derivation that
at once commends itself.” Burnell, On the Aindra Grammar. It ‘is
interesting to note in this connexion that even now in Bengal orthodox
teachers of Sanskrit grammar speak of “dhat” instead of “dhatu.”
eangraey fez fasta) ata egy) fee faastitey faastahn felted: |
[erec| saa area a: oftafea) «6Ssazi1 wage | fifeastdtmanta
tam) amo: feeanaara fifgafaesttarra | gaz: |
+ faea sfaat) al at tke
w sfasisfa etc. VP. viii. 3 ff.
 IT 159

Sam. uses the word without defining it in its very second

rule: nityakathader adau. The Vrtti says: na id gacchati
yasya sa nit. Sam. has a rule at the begining of the sixth
chapter “‘luk cihnarthasya’”’ which is thus explained in the Vrtti:
faarier faarainaraea ovata) ocefagl aa fed
afaerery | ( amherst ffaa | wedefacre:, agrarc-
ofaaarogha: mearauofasrdsedt fa, afe aa-
Tata fags aTcarey, aa afe HI Ta-
Tat: HAAS AE: TAT |THEATARTIT THe CTHG |
ate STAY TAT FRCTHTUIT A HAT aT:, ART ‘fea:
faa: ‘iaitso ] xfa fina:, ‘aftafrar’ [ 21232 ] sears
areata, ‘frafraecra’ [ aaee] cfa aa a ara
fang garat a qarat arte aera, arfe agra
ofaaaa saat, freq maractagrradserat:, farferter
dfifa a: areneat sf: a acnafafa sfagaredaet
TAT EAT | TWaTaeT: |
The earlier grammarians, Panini, Sarvavarman and presym-
ably their predecessors, appear to have been very careful in
the selection of the letters or syllablnes to be used as anuban-
dhas. Thus the anubandha § indicates sarvadega in the case
of a substitute and sarvadhatuka in the case of a krt suffix.
The palatal sibilant was used as an enubandha in these two
cases, because the word ‘visva’ meaning the same thing as
‘Sarva’ contains §. The s of ‘sarva’ cannot be used for this
purpose, because the anubandha s indicates that bfeore a suffix
containing
the anubandha the stem is to. be regarded as a pada
(siti ca i. 4, 16). Here again s has been selected for the
160 TECHNICAL TERMS OF SANSKRIT GRAMMAR

simple reason that it is the initial letters of sup (cf. “‘sup-tin-

antam padam” i. 4. 14). Again the anubandha 7 in the case
of roots shows that the penult will not undergo lengthening
before the causal suffix (mitam hrasvah P. vi 4. 92, manuban-
dhanam hrasvah. K. iii. 4. 65). Here also m has been chosen
because many roots of this type end in m. Again the
anubandha kh indicates that the nasal m is to be inserted after
stems ending in vowels, because this 7 becomes anusvara which
is represented in writing by a dot or zero and ‘kha’ means
‘Siinya’ or zero. (It may be noted in this connexion that J.
uses that monosyllable ‘kha’ for the dissyllabic ‘lopa’ of Panini
and his followers.) We may mention in passing that the initial
kh of taddhita suffixes is not an anubandha but merely a
grammatical device and stands for in, even as the initial ph
of these suffixes stands for ayan. Since several prominent
words taking the suffix Gyana end in r (cf. Narayana, Carayana
etc.), and since r is known as ‘repha,’ the ph of ‘repha’ was
seized upon by grammarians to represent Gyan. Again since
‘kulina’ is a well-known word formed with Ina, and since it
begins with k, the corresponding aspirate kh was used to
represent in and so on. Again in Panini’s system vu stands
for aka and yu for ana (““yuvor anakau” vii. 1. 1). Possibly
vu was suggested by the common word ‘pavaka’ and yu by
nayana.’
Again, according to many grammarians an anudatta
vowel or n attached to a root indicates Atmanepada (anu-
datta-nita A4tmanepadam P. 1. 3. 12, kartari rucddinanuban-
dhebhyah K. iii. 2. 42). As has been already pointed out
(p. 17) the ancient Indians set great store by the virtue of
 IT 161

self-effacement and consequently the anudatta or unaccented

vowel is used to indicate Atmanepada. In early literature
“pratyan”’ is often used with “‘atman,” it is therefore highly
probable that the 7 of “‘pratyan’’ suggested the anubandha #.
It may also be argued that it is unnecessary to give any pro-
minence to the Atmanepada by means of an accented vowel,
because the Atmanepada has an importance all its own, exactly
as it is unnecessary to accent the finite verb which is the most
important word in the sentence. It may further be argued that
of the letters of the vargas only the gutturals are regarded as
bhoktr, the others are looked upon as bhogya, and so the
final of the first varga is used to denote Atmanepada because
it is the supremest bhoktr. Ubhayapaditva is indicated by
means of the svarita vowel because it is a samahara of the
udatta and the anudatta, and by means of 7 because it is the
nasal which comes next to 72.
The letter ¢ is used as an anubandha after vowels to res-
trict its quantity according to the rule “‘ta-paras tat-kalasya”’
i. 1. 70, because ‘tat’ in this particular rule contains two t’s.
Similarly the anubandha f¢ indicates the svarita accent, because
the word ‘“‘svarita’’ contains ¢t. Again the anubandha p indicates
that a suffix or ending is strong, probably because it is the
initial letter of the word ‘‘prthu’? meaning ‘broad, expansive.
Similarly the anubandha & indicates that the ending or suffix
is weak, probably because it is the initial letter of the word
‘‘ksina”” meaning ‘lean, thin’. Because of the “anu” of “anu-
bandha”’ Panini uses “‘anundsika” to indicate “‘it” (upadeée’j
anunasika it i. 3. 2).

O. P. 129—11
162 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Panini’s ‘‘Satr’’ corresponds to Sarvavarman‘s ‘‘anti’’ which

is most unfortunate, being identical with the third person plural
ending, but with ant Sarvavarman generally prefers 7. Thus
he prescribes the substitute arvanti for arvan in ii. 3. 32.
Panini’s kvasu appears as vansi in Sarvavarman. It was pro-
bably this anubandha / that is responsible for Sayana’s curious
mistake in his Rgveda-bhasya (e.g., on i. 115. 1) where he
often takes vasi in the rule ‘“‘Sasi-vasighasinam ca’? to mean
the suffix kvasu. (See Vedic Selections edited by K. C.
Chatterji, p. 85 f.n.) The comparative suffix iyas is not
mentioned by Sarvavarman, but it appears to be included in
ansi” in li. 4. 50 and so must have been regarded as iyansi.
Again though Panini’s ktavatu is nowhere directly mentioned,
since it is covered by “antu’’ in the rule ‘antv-asantasya
cadhatoh sau” ii. 2. 20, Sarvavarman evidently regarded it
as ‘“‘tavantu.’’ For the same reason matup and vatup appear
to have been mantu and vantu according to Sarvavarman.
The anubandha 7 of Panini’s “‘gatr”’ and yw of Panini’s
‘“matup” and ‘‘vatup”’ indicate the advent of n in the strong
forms and the formation of the feminine with i. Since
Sarvavarman reads the suffixes with n and since he specifically
prescribes 7 in the feminine for words ending in these suffixes,
these anubandhas are unnecessary for his purpose. So follow-
ing-his predecessors (cf. VP.) he uses the vowel i for ease of
utterance in the case of anti, vansi, tyansi and arvanti and u
with mantu, vantu, etc. because of the presence of the Jabials
v and m in the preceding syllables and because of the necessity
of distinguishing them from anti for Purposes of declension.
It may be mentioned in this connexion that acc. to modern
 IT 163

philologists the n of mantu, vantu, tavantu and anti is original,

that of vansi and iyansi analogical.
In the Mahabhasya on ‘“auna Apah’’ vii. 1. 18 we find
two Sloka-varttikas discussing the utility of the anubandha
nm in the rule. In the last verse of these two stanzas we
read :

fazed aaa at ear |

Patanjali explains this thus:
via qagqafaraaa| Gaig a asgaen a a-
Reecaratfar Rrra |
‘Or the ” of aun is due to the mention of aun in the rule
of a predecessor of Panini. And the grammatical operations
due to anubandhas or indicatory letters in the rules of
Panini’s predecessors are not carried out in this system.’
The difficulty arises thus: Since Panini prescribes the
substitution of $i for aun in the case of a-stems, and since by
aun he evidently means the endings of the nom. and acc. du.
read as au and aut respectively in his system, these two
endings have to be regarded as nit. So these endings will
have to admit the agama ya acc. to “yad dpah” vii. 3. 113
which prescribes the 4gama ya before endings with an indi-
catory n in the case of 4-stems. To prevent the advent of
this ya, the author of, the Sloka-varttikas says that au is not
nit acc. to Panini, but acc. to his predecessors, and as Panini
makes use of the older term aun simply to secure ease of
utterance and as the indicatoy letters in his system and in
those of his predecessors do not always coincide, we are not
to import ya acc. to Panini’s rule “‘yad Apah”, because of the
164 TECHNICAL TERMS OF SANSKRIT GRAMMAR

presence of the anubandha 7 in the works of Panini’s

predecessors.
So what Patanjali means is that if an anubandha is present
in the systems of Panini’s predecessors and absent in this own
system, it will have no indicatory value so far as the latter js
concerned. It does not mean that the values of indicatory
letters differ in different systems (though, of course, they
sometimse do differ). As Kaiyata explains:
qaiand 2°aft feaaa feat oa) a ae afiaails-
Termishea | aes a Tafie: | Ar a: qaqEa-
ae cer geot wadifa soran-fediar-frraa-seo-Rafe: |
‘Ancient grammarians have mentioned both’ the dual
endings with an indicatory 7”. In our system, however, the
suffix aun does not occur anywhere. The method of the
ancient grammars has been followed here (in this particular
case) simply to denote both the ‘aun’s. So by aun (in our
rule) is meant the aun which is found in ancient grammars
and therefore both the nominative and the acc. dual endings
are included.’
Jinendra-buddhi quotes a rule from a predessor of Panini,
but unfortunately fails to mention the name of the author.
He says:
qrarainreatg ¢ aaa fraaa feal aera) ante
“arate” of aa gaqs: |
The rule evidently frames the technical term “aun” for the
two endings au (nom. du.) and aut (acc. du.), just as K.
has the technical term “da” for the roots da and dha. So n
here is not an anubandha at all. The author of the Sloka-
 it 165

varttikas, therefore, was probably thinking of some- other

predecessor of Panini
in whose work auri was given as the
ending for the nom. and acc. du. This ancient grammarian
probably used the anubandha ¢ for Panini’s 7 in the case of
the sup-terminations. Thus for Panini’s au and aut he used
aun and for Panini’s ta he used a. The Nyasa says under
vii. 3. 105:
ararat fe qa efefa addaaaaieat gdfte en.
aan adita “ore” ofa fasta aatanaad pe |
Panini uses these sometimes with vocalic endings for
simplicity.
Thereis hardly any doubt that Panini introduced several
new indicatory letters and that the value of some of the old
indicatory letters was changed or augmented, but a careful
comparison of the anubandhas of Panini with those of K. leads
one to the conclusion that most of the older anubandhas were
retained by Panini and that in most cases the indicatory
signifi-
cance is the same in Panini as in the older grammarians.
In RP. and TP. we do not find any indicatory
letter.
VP. uses ‘“‘nu” with
the anubandha xu in its Adhikara rule
“nuh” iii. 133. VP. also reads an with the anuba
ndha n in the
list. of upasargas (vi. 24), and et and ot with the
anubandha 7
(i. 114 etc.). Uvata merely paraphrases it by nakdrah,
but Anantabhatta explains: nur _ iti nakarasya samjia
. We
meet with ‘‘nu’’ again several times in Chap. IV
:
TAATSTA Bl2z, FATT RP vie, SAGARA vee
Similarly VP. uses “si” for s, “mi” for m, and
“yi” for
y. Hence VP. has got the rule “svyarair api’
i. 40.
 166 TECHNICAL TERMS OF SANSKRIT GRAMMAR
The most important anubandha in VP. is probably found
in the word ‘‘vamsau”’ in the rule:

aia a zante TT serTAa 192

‘““Vamsau’’ may be the loc. sing. of ‘vamsi’ or ‘vamsu’ CA.
has ‘vasu’ for ‘kvasu’ in ‘vasvantasya pafica-padyam’”’ i. 88
and ‘vasau hrasvat’” iv. 35. Panini also uses ‘kvasu’ and
‘vasu’ for this suffix. K., however, uses “‘vansi’’ throughout.
In ii. 4. 50 ‘ansi’ is used for ‘vansi’. In the Krdvrtti, however,
the suffix is read as ‘kvansu’ or ‘vansu’ after Panini’s ‘kvasu’.
Similarly Panini’s “‘satr’’ is represented in K. as ‘anti’, though
in the Krdvrtti, following Panini, Vararuci uses gantrn. All
this sets up a strong presumption in favour of ‘vansi’ in VP.
CA. uses a larger number of anubandhas than VP. and
shows a very close approximation to P&anini. As in the
Astadhyayi, we find san (1. 86), vasu (i. 88, iv. 35), sua for
the particle su (il. 97, cf. Panini vill. 3. 107), uA for the
particle u (ill. 4) and tatil (iv. 20). Matu occurs here (iii.
17, iv. 17) for Pamini’s matup and matvartha (iv. 47) is found
in both. It may be mentioned in this connexion that
Panini’s iyasun appears here as iyas, mayat as maya, jatiyar
as jatiya and thal as tha (iv. 15), without anubandha.
Moreover, the suffixes kyac, kyan etc. are referred to here as
yakaradi (iil. 18, iv. 29), showing that in the system of grammar
followed by this Pratisakhya the anubandha ka was wanting
in these suffixes. Roots are generally indicated by an i affixed
to them (muci li. 76, svapi
ii. 86, sadi ii. 99, vyadhi iii, 3.
etc.) as also by means of the sufhx Stip.

GT] TaserswAAthears LT 61312, ACT SII: e1g1e

 tr 167

HI | WyeTAarswaTét sisisie
aagraa tf aafa ay) awaqal afearfaeia aaa,
TM AAA Mera | saqeeenfarary faerfa,
aeqafaatara ofeareat | aedetisaaea: wate’ ee
sTaerifraqraiad | Afaq we-saect
a qafea | zZrHT|
a | atetafarafra cies
HATA 21213
MT | SATAN 121%
Rl ATA eieiZ©
Zl Wats ve
REAR, BATMATATTY att eae: VAT) «AeA
HasGazt: | aa afa wear aera); ease
Searcy: | afta: |*
QT | Rrataq are
Hi ferarehfery ) cat TT | yaIQo-ze
J BIC START: 212 ¢"2
WHATS fear «af ster|
wa fa-sereitercaarfs: | aft: ag wna waa
VE Tagaar:
T | Rae-
| Haft ay) Tarconedfefs
~aarTenataquasraatafa ara: | saaseeaa: Ae-
*Thus Vopadeva does not distinguish between “karyartha anu-
bandha” and “uccdranartha anubandha,” though the commentators of
Panini take great care to distinguish between the two, restricting, as
they do, the terms “it” and “anubandha” to the former. Thus under
“vetio vayih” ii. 4. 41 the Nyasa remarks: ikara uccaéranarthah, nainu-
bandhah. tena num na. bhavatiti bhavah.
 168 TECHNICAL TERMS OF SANSKRIT GRAMMAK.

saifa: | safe traqaafenofraneoe: | agefia:

qd weg: watarewnn, qaaranfe amd
frahetssaa wa o-ergmantermic-enreife: | TAT
TRTSTRT |
e| ufa wesfe a fagdifa ca agqaerd; a ¥
vencorttaarat frosts afar) 21943 Tf |
UPADHA, UDAYA AND ANUSANGA.
U padha.

“Upadha”’ (from upa ‘near’ and dha ‘to place’) means

literally ‘setting or placing near or next to’, then ‘that which
ig set or placed next to, i.e., a preceding letter or word’ and
lastly ‘the last but one letter in a word or stem’.
RP. stands alone among the PratiSikhyas and early
treatises on grammar in not defining Upadha and in using
the word generally in the sense of ‘a preceding letter or word.’
Of course the preceding letter happens to be the penult in
many cases, but at the time of this Pratisakhya the sense does
not appear to have crystallised into the penult. RP. also uses
the derivatives “upadhiyamana”’ and “upahita’* in the sense
of ‘preceded by’.
The term is wanting in TP.,
but in VP. and CA. it has
the same technical sense as in P&nini.
The word occurs by itself only .once in the Nirukta in a
passage which is rather obscure: “vabdhim te haridhana
i
swaraty amgaaisematea aa: wee Geaueaad: eee ed: nae
aqataart: WHT: WRATH 818
 UPADHA 169
upa fjisam jighratam’’ ityapi nigamo bhavati. Adinabhya-
senopahitenopadham 4adatte. babhastir atti-karma. v. 12. This
is explained by Durga thus :. Adav. abhyasenopaglesitenopadham
adatte, dhakarena nimitta-bhiitenopa-dham akaram 4datte
lumpati.
In “upadha-lopa” (“‘athapy upadha-lopo bhavati. jag-
matur jagmur iti” Nirukta ii. 1) and “‘upadha-vikara” (‘“‘athapy
upadha-vikaro bhavati, raja danditi Ib.) it plainly means ‘a
penult’.
Upadha occurs in the Unadi-siitras also in the sense of ‘a
penult’,e.g., caki-ramyor uccopadhayah ii. 14, kameh kid
uccopadhayah iii. 138 etc.
K. agrees with Panini in its use of ‘the term.
J. contracts ‘‘upadha” to “un” with the ‘u’ of the first
syllable and the nasal ‘nh’ to round it off, because ““upanta”’
contains a nasal. Mu. follows J.
C., Sak., Hc., Sam. and others prefer the term “upant(y)a”’
as being self-explanatory. Thus C. has “mad upantac ca
mator vah” vi. 3. 35, Sak. “mAntopantajayo mator mo Vah”’
i. 2. 96 and He. ‘“mavarninto- -pantapaficama-vargan mator
mo
vah” ii. 1. 94 fér Panini’s “mad upadhayas ca mator
vo’
yavadibhyah” and ‘“‘jhayah” vii. 2. 9-10.
Commenting on upantasya varge” Sam. i. 133., Goyicandra
Says :
SOTA SKAS TUES: | RATATAT: |steal
a9 J TSM: ATA AN | Ta TeaeT THTRET TTTART-
APTA THATIMAET
ARTA | Aca ger AHTTET
ST WT
aanteat sreaft |
 170 TECHNICAL TERMS OF SANSKRIT GRAMMAK

HN. takes the name of Hari that sounds like ‘‘upadha’”’ and
its term is Uddhava.

QUST | Meee Ta STAT e13%

ageearftat | aortgreraya STAT LIAR
TW] ABSIT TASTAT Vee
| AAT TASTAT RALILE
H | SIFAGE wense
J Palseea@ES_ex
QM | AeA Ya STAT @123
Gl TSTAAKST: 212122
T, Wika Sasa z2¢%
El arn wena qaat sgade:, swafa gq
ATs LIRR

Udaya.

‘““Udaya”’ literally means ‘that which rises’, then it came

to mean ‘that which follows’. Cf. Naisadha iv. 3: aafed:
a fe at aqaract: It is used in this sense in all the Pratié-
khyas. In the Astadhyadyi the term occurs only once in the
rule ‘“‘nodatta-svaritodayam a-Gargya-KaSyapa-Galavanam’”’
iii. 4. 6—a rule of doubtful authenticity. Vamana says in the
Kasika : udatta-parasyeti vaktavye udaya-grahanam man-
galartham. Haradatta explains:
STUNT: TREAT Taras: srfsarety cfs: |
In RP. “udaya” is often used in contrast with ‘“‘upadha”;
e.g., “anudattodaye punah svaritam svaritopadhe“ iii. 12, ‘in
a coalescence, however, where a following anudatta is preceded
 ANUSANGA 171
by a svarita, the syllable of combination is svarita’. In u.
81 occurs “‘upadhadnibhodayah” ‘having the following vowels
like the preceding ones.’
In viii. 2 ‘‘udaya” is used twice in the expression
“udayodayah” which means ‘which follow words that follow’.
TP. uses “‘udaya’” in this sense in the rule “‘udaya-
Svaradi-sasthano hakara ekesim” ii. 47, “‘according to some
authorities, h has the same position as the beginning of the
following vowel”. In VP. we find ““nodatta-svaritodayam”’
iv. 143, “patyau talavya-svarodaye” iii. 35 where “talavya-
Svarodaya” means ‘ending in a palatal vowel’, etc.
In CA. “udatta-svaritodayanam” occurs in iii. 65. In
the commentary on CA. iii. 27 “A-an-ana hrasvopadhah svare”’
we find :

S-O-AY TaleaT A SITET: KATIA: |

dt faawafisstea sere gafierfa |
In RT. we find ‘‘ca-la-modayam eke” 175, “some autho-
rities prescribe the change of § into ch, when it is followed by
c, fland m’’,

Anusanga.

‘“Anusanga”” (from anu ‘after’ and safij ‘to stick to, to

attach to’) means ‘close connexion’ and is applied to the
penultimate nasals of roots in grammatical literature, probably
because these nasals were felt to be separate entities attached
to the roots, since they appear sometimes with these nasals
and sometimes without them. The word ‘“Anusanga” might
also have been chosen because it contains two nasals. In RP.
 172 TECHNICAL TERMS OF SANSKRIT GRAMMAR -

xiv. 9 we find ‘“‘nasikayos tvanusange’ nundsikam’’, ‘when

the letter is pronounced closely connected with the nose it
becomes nasalized’ (nasikayor yada varno’nusajyate—Uvata).
It is just possible that the technical term Anusanga stands for
“nasikayor anusangah’’. The word is not found in Panini but
occurs once in Katyayana: antyat plrvo masjer anusanga-
samyogadi-lopartham i. 1. 47. 2. The Nydsa defines the word
thus under ii. 1. 47. nakarasyopadhdya anusanga iti purva-
caryaih krta samjia. The Padamaifijari says the same thing.
The Sabdakaustubha also says:

STINTAET AHTLEATATS efa sai aan |

We find the word fully defined for the first time in K.
where immediately after the rule “antyat
piirva upadha”
occurs the rule “vyafjanan no’nusangah” ii. 1. 12. The Vrtti
explains :
foyer Sarat: at aes sreaTe a: gata aeRIT:
aisarsy wate |
Curiously enough Su. uses the word without defining it in
its rule “‘aniditam anusangasya kniti” iii. 3. 73 on which the
commentary Makaranda remarks: |
ay wTgaa arate a fafeat) meaazgqys:?
way | kaa AHUsaIe fa ofa xe aac
yt | aaa agaratsayy sfa araaaae |
HN. calls it “‘uddhava-narama”’ j, 453 (=upadha-

+ CHa HAgeIIsa! af ugg |

SANT: WHY TETRA |
4 Many read yang
 SAMHITA AND SANDHI 173
nakara) of ‘“‘visnujananta dhatu”’ (=vyanjananta dhatu).
In PR. ti. 47 we find “‘antya-vyafijanadi-na.”’
‘‘Anusanga”’ is also used in the sense of the connexion of
a word or expression used in one connexion with another word
or expression.

waft: aerateader gatqaanag |

MA
rATMATGAaIAATE: |
CUTATHALENRT TAST
LI AAIM ARTS:
A |
AAATTTAUATAA
AT ATETT:
T |

SAMHITA AND SANDHI.

“Samhita” in the sense of ‘putting together’ (saz ‘together’

and dha ‘to put’) occurs in RV. As a feminine substantive,
however, it is first met with in the ‘Faittirlyopantsad
in the
sense of ‘conjunction, connexion.’ In its technical sense of
‘the
closed connexion of words’ it is first found in the
Nirukta.
In the Pratigikhyas it is generally used in the
sense of the
Samhita-text as opposed to the Pada-text. (it may
be men-
tioned in this connexion that Nirbhuja in the sense of the
Samhita-text and Pratrnna in the sense of the
Pada-text occur
as early as AA.) From meaning words or sounds in the closest
juxtaposition Samhita came to mean the modifi
cations caused
by such juxtaposition—a sense in which the word “sandhr”
is used.
“Sandhi” is also a good old word going back
to RV. in
the sense of ‘joint’, but it is found from
the Pratié&khyas
downwards in the sense of the ‘modifications
caused by the
 174 TECHNICAL TERMS OF SANSKRIT GRAMMAR

juxtaposition of letters’.* In this sense the word is

conspicuous by its absence in the Nirukta and P&nini. But
the reason for the non-mention of the word in these two
works is not far to seek. The Nirukta has hardly any
occasion to refer to the phenomenon and Panini did not re-
quire a special name for the changes induced by the closest
connexion of words and so gives a wide berth to it, as he
does to the names of the special kinds of euphonic combina-
tions like ‘“‘Anvaksara-sandhi’’, “‘PraSlista sandhi’’, etc.
Following in the foot-steps of Patafijali*, Candra dispenses
with the use of the word Samhita and the Vrtti remarks:
samhitayam iti na vaktavyam, adharasyaupaSlesikatvat v. 1. 70.
Sak. also avoids both “samhita’’ and ‘‘sandhi’. The
commentator of the Sakatayana Prakriya Samgraha says at the
beginning of the Sandhi-section : plrvottaravarnanam avirame-
noccaranam sandhanam sandhih samhiteti yAvat.

* In RP. Sandhi is used in the sense of mere juxtaposition of

letters also. Thus where an initial and a final consonant occur one
after the other and neither of ‘them undergoes any change, the Sandhi
is known as Avasangama (yatra dvayor vyaiijanayor avikarah
so'vasangamah sandbih—Uvata on RP. iv. 1). The Yajiiavalkya Siksa.
however, distinctly says that Sandhi has four varieties viz. elision,
augment, change and the retention of the original form-
afaaqfa a wafa—aiqiaafaame: nafaaray fa |
By Prakrtibhava is evidently meant the absence’ of Sandhi in the
case of Pragrhya vowels, as also the non-mutation of a Visarjaniya into
a siblilant or Jihvamiliya or Upadhmaniya.
4 FT USAMA) RUA) SfeATM ary faraia—aiqag wis-
afaa asfaafafa | mee ames asaisfHamAy wfagadaaca soy AIT |
aml gute [eo] wafer fa. aamita afeaaed afeqiaiaa afaenfa
EtT9e HIDE |
 SAMHITA AND SANDHI 175

J. has the adhikara-rule ‘“‘sandhau”’ iv. 3. 68 corresponding

to Panini’s ‘‘samhitayam’”’ vi. 1. 72.

He. lays down rules of Sandhi in i, 2 and i. 3 and uses

the word ‘‘sandhi” in rules i. 3. 52 and vii. 4. 111. He uses
“samhita” in “‘tayor yvau svare samhitayam”
vii. 4. 103,
corresponding to Panini’s “‘tayor yvav aci samhitayam’’
viii. 2.
108. He distinguishes between Sandhi and Samhita when
he
Says in his Brhadvrtti on ‘na sandhih” i. 3.
52: viramad
anyatra tu samhitayam sandhir eva.
Later grammarians found the uses of “Samhita” super-
fluous and retained the shorter term ‘“Sand
hi” formed from
the same root with the same prefix. K. has
a chapter entitled
“Sandhi-vrtti” or ‘“Sandhi-pada”, and’ the word ‘“‘sandhi’’ is
used in the rule “na visarjaniya-lope punah
sandhih” i. 5. 16.
The derivative “sandheya” occurs in the
rule “na vyanjane
svarah sandheyah”’ i. 2. 18. Durga says
in his commentary on
1,2. 18:

ATs aaa: | Seaet aorta afd eae |

afsaz-
afe ara’ aaraztaife cfakafiraan
, STANT |
Kaviraja says under the same rule

aerat afeaftfa am a gar, award afsa-qaa

fea=at, oofinfenfs afind: afarfthy afaay
atthenft gafeat: afaftepna | act qos ana:
afaftiteqaen wrest) a amfr ay afaa
rgceea
raked, afeater aac p aay, ufeafiaa-
are aftafterrarcrs|
 176 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Again, under i. 1. 13 Kaviraja says :

aaa afear a faneaa | fe afe: aconft dfeart
aztfard afarntaaa |
Sam. has got the sandhipada i. 1 and the word ‘sandhi’
occurs in “‘Jug-yader na sandhih”’ i. 360 where the Vrtti says
“vamanam druta-taroccaranam sandhih, upacarat tatkaryam
ca’’. The last rule of the section also mentions sandhi :
afeqtaad faeat faeat arqraaa: |
gqacaty amt faca: *aareaa faarfia: 1)
In the printed editions this 1s given as a sutra, but it may
very well be a portion of the Vrtti.
Su. has got its sandhi-section consisting of ‘ac-sandhi’,
‘hal-sandhi’ and ‘visarga-sandhi’, and mentions ‘sandhi’ in
the rule ‘‘na sandhir visarga-ya-va-lope”’ i. 2. 27 etc.
PR. speaks of ‘svara-sandhi’, ‘vyafijana-sandhi’ and
‘visarga-sandhi’ and uses the word “‘sandhi” in ‘“‘ayadinam
va padanta-yvor lopah sandhyabhavakrt” i. 75, ‘‘dvivacanam
idude syat sandhi-virodhi svatah paratrapi’? i. 78, etc. and
defines sandhi in 1. 165.
HN. retains the word “sandhi’’ which it defines in i. 45.
It speaks of Sarvesvara-sandhi, Visnujana-sandhi and Visnu-
sarga-sandhi.
Mu. speaks of ‘ac-sandhi’, ‘hal-sandhi’, ‘visarga-sandhi’
and uses the word “‘sandhi’’ in the rule “lupi na sandhya-
dya-vidhi”’ 15.
ge a a a heen nigella
* This may be defended according to Panini’s rule “so’ci lope cet
pada-puranam” vi. 1. 134, but it is evidently taken from a stanza con-
taining the feminine word “samhita.”
 SAMHITA AND SANDHI 177

““Sandhana”’ is used in the sense of ‘euphonic combination’

in RP, X.. 17:

efaqag aoa ye: a: anzafraq|

qqqAedg FATT II
‘Euphonic combination in case of words precéded by iti
(is desired) by old (teachers). But svah should remain un-
combined. One should recite it with an avagraha.”
The word ‘‘sandhya’”’ is used in the Pratigikhyas in the
sense of ‘belonging or pertaining to or resulting from eupho-
nic combination’.
“Sandhyaksara” is found in the Pratisakhyas, Vé@rttika,
Bhasya and K. in the sense of diphthongs (lit. ‘letters of com-
bination’),
Both Sandhi and Samhita are used in the following passage
of AA. iii, 1. 5: |
aa af faadafa,: da ant fantafa, aa mat-
mat faaakt at afrar |
Sayana explains :

afd qaiaca: afaaiaaar feadafa fate ae-

qa) aq mat a fafasa arrfa, “afme”
KaTaAMITIs WRsaTTaea a vata, faeq afta-
saa cacao fide: | ava amat a frrat|
safe af faadafa af se say wa awe WUe BfrarEcts aT
Wee | Taqwa 2 zg
feel angel gafacael afaq dfemeg emg aguas fayaa ee
saiy aaa seems a fated ceafadtaa) area
O.P, 129—12
 178 TECHNICAL TERMS OF SAN.KRIT GRAMMAR

“arg aq aaafyt:” LAT WARS PRTANTT ZeaisHIe

zaaa, dfearnre a vata, feeaand amswea fama: |
Sripatidatta distinguishes between ‘Samhita’ and ‘sandhi’
in his Katantra Parisista. He holds that the coming together
of two letters without any intervening vowel or consonant and
the pronouncing of the immediately after the other without an
interval of even half a mora is Samhita. And when letters
come together in this way “‘sandhi” takes place: dvayoh
susannikarsah i. 95. Here susannikarsah is evidently a para-
phrase of “parah sannikarsah’’ of Panini. Sripati says in the
Vrtti : |

qamninacaafedtg wana: gafaadt waft, a fe

facfaararareaay |waATaHaaraaa: afediea® | quoi
gee aah wat gad arf | dfearnaa aaa: eg: |
fa) ce afias:+
afar aeive
omit) afer |oagnafa: |) | qateare qatfzfa:
aqaeta TAT RSETATIA | zIF-2
qo | ser afearatasonaa “12 ar aaa: dfrar.
aut, araTIgaesaaanT: Tata saa a213
aT | eareenerinnite eifeat itee, aaftraz:
eifea: ess, Tefrstiscfer: ages, afk ae,
an wily afea: 313 ( & aeagentt qaha—eaqeat:
MAT BUAGATA | ata franc: qa: tazt
wafa
ware agaria, sraarfs ar qaifr wafta qaraq waz
aft) saz 1)
b szimgzmafcami ot fag Gaga) argmangag pier
 SAMHITA AND SANDHI 179

Ba} afanw: dfeat co

at, ot: afane: afar viele 02
amay| xeafacm: fear |-.-gtarfaera: afer t--:
qawaARtacaad dfeat) elwizoc, afxat aararafirfe
slafafeaian, a fe afaq afyeedrormae—‘aeneaia
dfaaraivy’ fa) a aa qenafansadia! eivigee
(an azaicadt dfeacraricedawed sfa qoracwrad |
RAZ | )
7) cae: seaeate avast fra |
Teds sae a: Ar: a aferacad i) lee’
fasiamardt g aaa: afaRat wre
Z| BagHcneqra anarafaam: |
ara faarz: afer: ane fawartern: Hfae n ties
afeatard fread fret araretrn: |
afad qafaraseaa arfrerfasaa || eiee
It will be noticed that in grammatical literature three words
belonging to three different genders are used to denote euphonic
combination. They are Sandhi (masc.), Samhita (fem.)
and Sandhana (neut.).* In the Karika “‘sandhir ekapade
nityam”’ there is a contamination of ‘sandhih’ and ‘sandhanam’,
just as in the Karika ending in ‘“‘saivanyatra vibhasitah” there
is the contamination of ‘samhita’ with ‘sandhih’.
The Stktimuktavali quotes two stanzas containing the word
‘“‘sandhi” in its grammatical sense :

* qaarhe aaraaguafaw: afer: | AQWEM, F: eee

 180 TECHNICAL TERMS OF SAN.KRIT GRAMMAR

a en ated fasr at ea aracg

eg faseq |
arend ate sreafa zaraaa ofiear: ||
‘Let them be ignorant of sandhi (peace, euphonic com-
bination), let them be ignorant of vigraha (war, dissolution
of compounds), if kings listen to akhyata (what is told them,
verbs), they will become learned by that only’.

areata raeqy: THAT AT Tear

aq saraoaaratad aaeiitag a |
man a fread aa ad faq ataegeya
aT afearmrarhrat 7 Fee: ara afacareza: |
‘Alliances (euphonic combinations), breaking up of
alliance (absence of euphonic combination), subjects (stems),
knowledge excelling that of the subjects (suffixes added to
the stems)—these, Sire, are common to both Your Majesty
and grammar. In the following respects, however, Your
Majesty differs from grammar: Your commands are never
called in question (a substitute is never prescribed option-
ally), nor is anything done anywhere destroyed (nor does any-
thing prescribed undergo elision), nor does prosperity put an
end to your good qualities (nor does Vrddhi supersede Guna),
nor is your officer mediocre in any business (nor is there any
second person in the case of any verb).’
It has to be remembered in this connexion that jin
grammar we sometimes come across substitute that are
optional, e.g., vibhasi lun-Irnoh ii. 4. 50 which makes
the
substitute g@ for in optional in the aorist and conditional
;
that what is prescribed sometimes undergoes elision, e.g.,
the
 SAMHITA AND SANDHI 181

supersedes
whole of the suffix kvip; that Vrddhi sometimes
‘ad gunah”’ vi.
Guna, e.g., “vrddhir eci” vi. 1. 88 supersedes
1. 87: and that: the second person has an honoured place in
grammar.
In the following stanza, the grammatical sense of the word
“sandhi’? would also appear to be working at the back of the
writer’s mind:

ATA AMET ATSSTHATTA |

ag aaa: afeateraratt a wa 1
The Padyaveni contains the following stanza on Sandhi-
vrtti or the phenomenon of Sandhi :
SAAARMACLAATAAAHT
PEAANITATATAAT |
afeaafatca aouatiat
arfaratate farara Arad ||
‘Krsna’s inamorata whose name consisted of an equal
number of vowels and consonants, and whose complexion,
steps, voice and curves were all that could be desired, even
as the phenomenon of Sandhi’ is designated Svarasandhi and
Vvafijanasandhi and is used in the formation of crude stems
and inflected words by means of nice letters (or in the forma-
re 5 ee

*“Mana” is thus explained in the previous stanza (480) :

Asya ara as: 8 va Marat arr af vata: |
7) QCAACG AIM FT CHAdanwal aFq: |
This reminds one of the following stanza of Govardhana:
omega gata a: frase? @ us gfteta: |
vat.YA: BsTAAa Ha tl
182 TECHNICAL TERMS OF SAN.KRIT GRAMMAR

tion of expressions consisting of words containing nice letters)

spent the whole night in (a fit of) jealous anger (482)’.

It may be incidentally mentioned that since Patafjali defines

Vrtti as “pararthabhidhana” or the conveying of some sense
over and above or beyond that which is expressed by the
constituent elements, it is. hardly proper to speak of Sandhivrtti
which appears to have gained currency after the analogy of
expressions like Krdvrtti, taddhita-vrtti, samasavrtti etc.

DIFFERENT VARIETIES OF SANDHI

Since the followers of Panini read Anusvara and

Visarjaniya after a and also include them in the Pratyahara
$ar, according to them there will be two varieties of Sandhi,,
viz., ac-sandhi or svara-sandhi and’ hal-sandhi or vyajijana-
sandhi. But as the various transformations of the Visarjaniya
occupy a prominent place in the scheme of Sandhi, so many
scholars add a third varjety, viz., WVisarjaniya-sandhi to the
two already mentioned. In K. we find these three varieties
along with Prakrti-bhava which is regarded by some as a fourth
variety, viz., Prakrti Sandhi. These along with the first sec-
tion which deals with Samjhas are regarded as constituting
Pafica. Sandhi,* i.e., the five sandhis. Thus’ the commentary
Prakasa on the Prakriya-Kaumudi says :

* The Katantra school says:

acaias Bagi: aafaafaaa a |

aged faaaa af: are yaaa: |
 DIFFERENT VARIETY OF SANDHI. . 183

ma aag aa HaAfsyaeail: Beagl warg ata} ( Aa-

afea: caafea: | fearafea: | <rfiefea: | ) wapefee-
Ize aadnand damaish afaagqaifa aar-
Taq|
From an early period, however, the Sandhi of case-
terminations was regarded as a separate category as is clear
from the Tolkappiyam, and so in the Ripavatara, Prakriya
Kaumudi, Siddhanta Kaumudi and works of this type we find
a separate section entitled Svadi Sandhi.
The Rutpdvatara regards the insertion of ¢ before ch in
words like pariccheda as a separate variety and mentions six
varieties of Sandhi :

TH wat: cafasaa saa Aa! TT |

dat fasta arf: wraferasaz ||
The ancient classification into lopa, agama, vikaéra and
prakrtibhava (p. 174 f.n.)—a_ classification found also in
Tolkappiyam, a very early treatise on Tamil grammar, would
appear to be quite satisfactory.
In general, however, writers on grammar
are anxious to
have five varietiés of Sandhi, because of the well-known Pafica
Sandhi of Dramaturgy.
RP. has a special name for each of the various classes of
Sandhi. These names remind one of the different varieties of
Upama (e.g., Acikhyasopama Mohopama, etc.) found in
Dandin’s KavyadarSa.
The combination of vowels with consonants is known as
‘‘Anvaksara-Sandhi” i.e., combination according to the
184 TECHNICAL TERMS OF SAN.KRIT GRAMMAR

succession of letters. When vowels precede and consonants

follow the ‘“Anvaksara-Sandhi” is ““Anuloma” i.e., in the
natural order. When consonants precede and vowels follow
the “Anvaksara-Sandhi” is “Pratiloma” i.¢., in the
reverse order.
RP. mentions three cases of Anuloma Anvaksara-sandhi :
(i) vowels followed by consonants (i. 8),
(ii) esah, Syah and sah followed by consonants, the
visarjaniya of esah, syah and sah being dropped (ii. 8),
(iii) im followed by consonants, the m of im being dropped
(iv. 83).
Of Pratiloma Anvaksara-sandhi only one case is specially
mentioned: the first letter of any varga is changed into the
corresponding third letter, when vowels follow (ii. 9-10). .

ag a a a aT Ta
vated saqayat Beez: |
a Seq IAAS TS AT
sfaatang faqaa a aa ||
aa TaArdarsala
ofasiay Praecarataeg |
sear afer Aarti
wawast aA aa atta Tar |) xI/¢-22
tfrarasia agg
Tay Wat aaa aaa Gaz |
aaa faata gar feofta
cafaearqartafeata Gt: |) vice
 DIFFERENT VARIETY OF SANDHI 185

The Anvaksara-sandhi-vaktra or Anvaksara-vaktra-sandhi

and Sauddhaksara Sandhi may be mentioned in this con-
nexion. When a visarjaniya followed by a sibilant followed
by an aghosa is dropped, the Sandhi is known as Anvaksara-
sandhi-vaktra (iv. 37). The commentator Uvata points out
that the real name is Anvaksara-vaktra-sandhi, the positions
of ‘sandhi and ‘vaktra’ being reversed owing to the exigencies
of metre.*

WATTM MTA BUF FZ

adsft aisrqacafeaars: | vlZe-20
a afacqucanda afeaca:) fire: afhaaest
RIG | TAZ: |
It will be noticed that when the visarjaniya of ‘esah’,
‘syah’ and ‘sah’ is dropped before a consonant the euphonic
combination is known as Anvaksara Sandhi, but when the
visarjaniya is dropped before a sibilant followed by an aghosa,
the sandhi is known as Anvaksara-sandhi-vaktra, because it is
“anvaksara sandhi” in the first stage only (“‘anvaksara-sandhi-
vaktra’’ meaning ‘amukhe anvaksara-sandhi’’).
The insertion of a sibilant or r in certain words is known
as Sauddhaksara Sandhi RP. mentions four cases:
(i) the insertion of § between words like puru, prthu,
adhi and candra (iv. 84);

——---~+

* Similarly the Padamafijari says under Kaésika vii. i. 18:

fed faae aufyeumafala aunafien sf faafeaq, anagaaeeia
AlaNe: AIM: |
 186 TECHNICAL TERMS OF SAN.KRIT GRAMMAR

(ii) the insertion of s in words like pariskrnvan, pariskrnv-

anti etc. (iv. 85, 87) ;
(iii) the insertion of s in the aorist form askrta in ‘‘askrto-
sasam”’ RV. x. 127. 3 (iv. 88);
(iv) the insertion of 7 between vana and sada (iv. 86).

Geraragqay wat STsarae |

Bl A TATea Areas GUseaer ||
qufa Ta HAL THT
qtfa tH: ATAET TUT |
afceqoatca araistnarre
MGA: AEIT TI TBH: |] wlee-ce,
Sauddhaksara Sandhi is so called because it is concerned
with the insertion of a single letter (Suddha aksara). It can-
not be said that the name is due to the fact that in this patfti-
cular variety of Sandhi the original letters (‘‘aksara’’) remain
unaltered (‘‘Suddha’’), only a new letter being inserted, for in
the case of ‘vanarsad’ the intrusive r changes the dental sibilant
into the corresponding cerebral. It will be noticed that the
Pratisakhya deals first of all with the insertion of §, the
instances of which are more numerous than all the cases of the
insertion of s, s and r taken together and the word ‘‘gauddha-
ksara”’ also begins with §. The 4gama in the case of Sauddha-
ksara Sandhi is referred to twice in RP. :

SNGIACTAISTT | 201%
Jee WSACAAATTAA | ee BE
The name “‘Sauddhaksara’’ may be explained in another
way also. In the case of “‘anpada etc.” also r and s come in,
 DIFFERENT VARIETY OF SANDHI 187

but in those cases the previous vowel is nasalized ; in the case

of Sauddhaksara Sandhi, however, there is no nasalization.
The word ‘‘Suddha” is used in CA. iv. 121 anunasikah piirvas
ca §uddhah in the sense of ‘free from nasalization’? And we
learn from the Kausitaki-Brahmana (ix. 5, xiv. 3) that when
Pranava was pronounced without the nasal element, it was
known as ‘‘$uddha’’.

Sauddhaksara Sandhi has to be clearly distinguished from

“antahpata sandhi’’. The former is concerned with an intrusive
sibilant or r, whereas the latter is concerned with the develop-
ment of certain glide sounds.
Philologically the § that is inserted before the word ‘candra’
was Once a part and parcel of the word whose earlier form was
‘Scandra’. The initial sibilant of the conjunct was dropped
for ease of utterance, as in the case of ‘taéyu’ for ‘stayu’, ‘tari’
for ‘stari’ etc. But archaic forms are often preserved in
proverbs and compounds, so we find the fuller form in
‘suScandra’, ‘puruscandra’ etc. Similarly ‘vana’ had an old
collateral form ‘van’, the genitive singular of which appears in
the Vedic compounds ‘vanar-sad’ and ‘vanar-gu’ and in the
common classical compound ‘vanaspati’ and the derivative
‘vanara’. The intrusive s in the case of forms of the root kr
is due to the analogy of words like ‘niskrta’, ‘duskrta’ etc.
The combination of vowels with vowels is known as ‘“‘svara
sandhi.” It has many varieties, the most important being
the Praéglista. The word “praglista” is the past participle
passive of the root Slis: (the later form of ris) ‘to clasp
preceded by the upasarga pra ‘forward, forth’ and means
 188 TECHNICAL TERMS OF SAN.KRIT GRAMMAR

‘clasped, entwined’, thence ‘coalescent’. Hence in RP.

PraSlista Sandhi is the name for
(i) contraction of simple vowels, short or long, into the
corresponding long vowels (ii. 15) ;
(ii) coalescence of a and G@ with the simple vowels i
(short or long) and uw (short or long) to the Guna vowels e
and o respectively (ii. 16-17) ;
(iii) coalescence of a and @ with the Guna vowels e and
o to the Vrddhi vowels ai and au respectively (ii. 18-19) :
(iv) absorption of a and @ by the Vrddhi vowels ai and
au (ui. -18-19).

SAAT AAA AIAHAR cacy|

THT THI: GTA |]
SHO AH WesHTC TaN: |
MATT PATNA TRAST AT BEAT: | 212%-Ro
gqor Swactees: Thess: | aT at RIRLE
Hara Thys: | st AT ABI%
carcat: sthase: | aqrearkrar aie
The word ‘“‘praslesa” is used in the sense of contraction
in “‘ikarayoS ca praslese” RP. ili. 13. Other Pratigakhyas
denote the svarita resulting from the contraction of two ‘i’s by
the term Praélista (VP. i. 116, TP. xx. 6). CA. (iii. 56)
uses the word. ‘praslista’ formed from praslista with the suffix
an in the sense of ‘relating to, pertaining to, resulting from’.
Evidently it regards the sandhi as “‘praglista” and the svarita
resulting from this sandhi as “‘pra§lista.”’
“‘Praslesa” is used in the sense of PraSlista Sandhi in VP.
 DIFFERENT VARIETY OF SANDHI 189

v. 33: pragnibhyam anindhau

praslese, ‘there is no avagraha
in the case of Praglista Sandhi with pra and agni with the
exception of the root indhi’.
**Praslista’’? occurs in both the senses of ‘contraction’ and
‘coalescence’ in the Mahabhasya :
sarussta fafa 1 nae ofesfiatsaq, =
AR AH | ARTA 23162
ase sPashaod fasa, sPasfaessaq) a
TS AS | AAT_AS_| AEnaeyT aleve
wPasfargisaq aaa) werner ezic
TaAyAMIA ws wanes sf | wPaswranizay |
ARTATVA F121
aa ounfanisi warnafterd: | sfeerfas
cain fost firfarfaered: | Raz |
Praslista is used of consonants in the following stanza
quoted in the commentary on CA. (i. 39):

WaT a wan os: TPs wat waq |

% &, fa afgastea sent afsat wa: 1
Similarly we find the Siddhanta Kaumudi :
ar’ ta aware ‘Maa? gaa sare Ganz
“Oral sat afes” fa atta aac: tPaead) 63
The ‘name of the Figure of Speech ‘“Slesa” would appear
to be derived from this use of Praglesa.
The next variety of vowel-sandhi is known as Ksaipra.
The word is derived from ksipra ‘quick’ and means ‘pro-
 190 TECHNICAL TERMS OF SAN.KRIT GRAMMAR

duced by speaking quickly’. In RP. it refers to the change of

vowels into semi-vowels before dissimilar vowels (1i. 21-23).
In all the Pratisakhyas the svarita resulting from such com-
binations is also known as “‘ksaipra’’.

GAARA SMART BUTT |

aT CATAL TF TA A AIM: THAI |] 2IRe-23
The change of o and au to av and Gv respectively be-
fore a non-labial vowel is known as Bhugna Sandhi.
‘““‘Bhugna”’ (from bhujati ‘to bend, twist’) means ‘bent,
twisted, made curved, distorted’ and is applied to this class
of sandhi because the o is here regarded as having been bent
or distorted into av.

MGAAAATALE THCISATATHTA: || 132

The elision of a after final e and o is known as Abhinihita
Sandhi. The root dha preceded by the upasargas abhi and
ni means ‘to place upon or into something adjacent,’ hence
“‘abhinihita’’ means ‘put into something adjacent’. Thus when
one sound is put into an immediately preceding or following
sound and is either wholly merged in it or loses much of its
individuality, the phenomenon is known as Abhinihita Sandhi.
In the other Pratisakhyas the term ‘abhinihita’ is used to denote
the svarita resulting from Abhinihita Sandhi.

aahafafica: afmatd: sraactand: |

caraata wafecareasa afere: 1 a AT 213%
The Abhinthita Svarita is known as Abhinihata in TP.
xx. 4. This may be merely an ‘artificial ‘variation’ as Whit-
ney is inclined to think (TP. xx. 4, CA. iii. 55). Or the name
 DIFFERENT VARIETY OF SANDHI 191

“abhinihata’”’” which means literally ‘struck down’ may be

due to the fact that the following vowel is elided or ‘killed’.
It may be mentioned in this connexion that the noun
‘abhinidhana’ is used in much the same sense in the Pratisakhyas.
RP. says:

afafaar sadfrarat
CAAT ATA LHF |
AEM Bact ee
TAMTATAT | R1VS
‘Abhinidhana consists in the reptessing and obscuring of
a sound, and takes place in the case of mutes and semi-
vowels except r, when they are followed by mutes and have
already undergone euphonic combination.’
CA. says: vyaftjana-vidharanam abhinidhanah piditah
Sannatara hina-§vasa-nadah, ‘the holding apart of a consonant
is Abhinidhana; it is pressed, quite weakened and lacking
in breath and voice’.
In TP. Abhinidhana is the name given to the first letter
of a varga which is inserted after a surd spirant when it is
followed by a mute. Here also the name implies that its
articulation is imperfect.
The name ‘abhinihita’ is appropriate acc. to the system
followed by Panini who prescribes ekadega. ‘Abhinihata’ is
apt acc. to the system followed
by K. who prescribes the
elision of the a.
Abhinidhana is. known as Asthapita in CA. (i. 48) and
Bhaksya or Bhukta in the Carayaniya Siksa (viii).
Now we come to hiatus or Vivrtti, The word Vivrtti is
 192 TECHNICAL TERMS OF SAN.KRIT GRAMMAR

derived from vi-vrt ‘to fly in different directions’ and means

‘the opening of two vowels upon each other without blend-
ing, hiatus’.

eaurat X faaha: | aT RR
garg wie afada a eq |
faafama fadar a fafa freatagq |
ATHATFATAAT OY
Several Siksa-treatises divide Vivrtti into four classes:
(i) Where both the vowels are short, the Vivrtti is known as
Pakavati ‘possessed of young ones’; (ii) where both are long,
the Vivrtti is known as Pipilika* ‘ant’; (iii) where the first is
long and the second short, the Vivrtti is known as Vatsanusarini
‘following the calf’; (iv) where the first is short and the second
long, the Vivrtti is known as Vatsanusrta ‘followed by the calf’
(Manduki Siksa, 92-96, Naradiya Siksa ii. 3. 1-2, Yajiiavalkya
Siksa i. 10-12).
When one of the yowels is long the Vivrtti is Dirgha
according to RP. (ii. 79). So Pipilika, Vatsanusarini and
Vatsanusrta will all come under Dirgha Vivrtti.
‘“Vivrttl’’ is used also in the same sense in VP. vii. 6:
hrasva-kanthyopadham — visarjaniyantam aariphitam vivrttyd.

*In exactly
the same way a stanza in which a verse of eight
syllables comes between two verses of twelve syllables each is known
Pipilika-madhyama Anustup. Similarly a stanza in which a verse of
syllables comes between two verses of eleven syllables each is known as
Pipilikamadhya Usnih. Again a stanza in which a verse of eight
syllables comes between two verses of thirteen syllables each is known
as Pipilika-madhyama Brhati.
 DIFFERENT VARIETY OF SANDHI 193

qat Peeaat Tat frarara: 1 ASe

When there are two consecutive hiatuses i.e., there is a
hiatus on each side of an intermediate vowel, as in ‘‘tasma
u adya”” RV. vili. 66.7, the hiatus is known as Dvi-sandhi*
Vivrtti according to RP. (ii. 80).
fEAPIARTIAAATEAT: | 2I1C0
Internal hiatus (i.e., hiatus in the same word, simple or
compound) is known as Antah-pada Vivrtti or inter word
hiatus (ii. 13). RP. enumerates four cases :

quaat faasat sew aasfife: | aeacad fara: | 2123

When ah, ai or au is changed into @ before vowels, the
sandhi is known as Padavriti ‘hiatus between words’, and
the svarita arising from such sandhi is known as Padavrtta
in the later Pratisikhyas. In RP. this accent is known as
Vaivrtta (il. 18). It is clear, therefore that in Padavrtti and
Padavrtta ‘vrtti’ and ‘vrtta’ stand for ‘vivrtti? and ‘vivrtta’
respectively.

faasarastiteat Adgd: <aqctea: |

araTcgam a gt eal ar: wzgaqa! 1
When ah, e, or o is changed into a before vowels, the
sandhi is known as Uderaha (ii. 29). The word “udgraha”’
is derived from ud-grah with the suffix ghan according to
Panini’s rule ‘‘udi grahah”’ iii. 3. 35 and means ‘taking out,

_ OS

s, but ‘catuh-sandhi’ with the dental. In the Susamadi group of Panini

(vill. 3.98) these are all conspicuous by their absence, though ‘su-sandhi’
,
‘duh-sandhi’ and ‘ni-sandhi’ are met with.

OP 129—13
 194 TECHNICAL TERMS OF SANSKRIT GRAMMAR

drawing out, intercepting’. The term ‘“‘udgraha’’ is used as

a designation for these varieties of sandhi, because the
visarjaniya in the one case and the vowels in the two other
cases (e and o being regarded as Sandhyaksara of a i and
a u respectively) have been taken out.

saad asa gal. aterm ad) a 3e-

ATal | RIRG-Re,
When Udgrahia and Padavrtti are combined, i.e., when
ah, e, or o becomes a before a long vowel, the sandhi is known
as Udgraha Padavrtti (ii. 30).
AAI STARIATUAT: || - 2IB0
When a or @ followed by 7; becomes ar, the sandhi is
known as Udgrahavat (ii. 30), because here also there is
hiatus between a and a short vowel and there is the inter-
cepting of the a of @.

WAT ITF HVSMARL ATGHRAT 213%

The Udgrahavat Sandhi would appear to be designated
Upadruta in the Sankhayana érauta Sitra in the following
passage :

fraere TENA SNE SaTea ag faciser TRISTIAT

rat 3 tfa ae, Hata tha ay ees
“Where there is Upadruta, Abhinihita, Praslista or Ksipra-
sandhi at the beginning of the pada in which o is to be said,
that should be skipped and the o said in the second syllable
as in the case of “‘Sva@ no 3” and of ‘“‘kva syo 3”
tts

* Nyunkha is explained clearly by S&yana under

AB. xxi. 3.
 DIFFERENT VARIETIES OF SANDHI 195

The commentator says:

SIEM aa afaatsaarat wag: acreecci—ar ma

Teaeq awarsy sar: xfer |
Cases of the absence of the Abhinihita Sandhi are divided
into two classes: viz., Pracya Padavrtti and Paficala Padavytti
(ii. 33). When initial a remains unchanged after final e, we
have Pracya Padavrtti. When initial @ remains unchanged
after final o or final ah changed into o, have Pajicala Padavrtti.
The names are significant, for we find ah changed into e in the
Eastern Prakrits and into o in the Midland Prakrits.

SEATETN TART IOAHTL

THAT F et. AaSRRATETA |
SA
WTAIS Gae
Ie FAI
qareriarsrgal vated tt Baz
The name Pafictla Padavrtti is also found in the Sankha-
yana Srauta Sitra : ‘Where there is the hiatus known as
Paficala Padavrtti, the suffix is to be subjected to Nyiinkha.’
frags: :
aneafa
Al Fa WaTeaz g ett
vert APTA
Zefa 1 RRueRle |
ay ra eafefa AAA KAITAIA GaTSTTTRITSAT
faafratettfa etary) ATTA
A hiatus in which the preceding sound is a Pragrhya vowel
is known as Pragrhitapada Samhita or Vivrtti (RP.i. 54).
THAR TRI:
SALT BTCA Ta ATTA
ARITATEA, AYRTATET: | Ree-Key
 196 TECHNICAL TERMS OF SANSKRIT GRAMMAR

A hiatus in which the preceding vowel is lengthened is

known as Plutopadha Vivrtti (RP. ii. 67).
A hiatus in which the preceding vowelis nasalized is known
as Anunasikopadha Vivrtti (RP. ii. 67).
When final dn becomes nasalized @ before a vowel, the
phenomenon is known as Anpada Padavrtti i.e., hiatus
between two words with reference to a word with final an
(RP. iv. 67). In CA. iv. 74 @npada or Gnpada is used in the
same sense.
When final dn is changed into nasalized @ before a semi-
vowel, the phenomenon is known as Vivrttyabhipraya or
apparent hiatus (RP. iv. 68).
Metrical lengthening of vowels
is known as Sama-vaga
Sandhi (RP. vili. 1). According to Uvata ‘Sama-vaéa”
means ‘serving for or caused by the equality of the metre’.
It may, however, also mean ‘caused by the necessities of
Saman’, because when a Rk. has to be changed into a Saman
such lengthenings have often to be resorted to.
RP. itself says :

aman ofa Sara ay

Satea 2 arad azapaat: |)
¢ ° ~.
2|/G0
The Samavasa Sandhis which bestow prosperity on verse
and metre are to be regarded as exceptions (to Anuloma
Anvaksara Sandhi).”’
Uvata explains :
Aqua BM) WAR wea: amy) |aronrsnfar-
qaMaataenareasar [ uieieze] wyamiaeearzy |
 DIFFERENT VARIETIES OF’ SANDHI 197

Seal WAI ast wise Bameaqucaaiar 2 arma:

Geaqq: |

Sama,* however; is not read either: in the Udgatradi group

or in the Yuvadi group. Even such a late work as the
Ganaratna-mahodadhi does not mention “sama” in its Yuvadi
group. Nor are the Ganas regarded as Akrtiganas. So it is
preferable to derive the word from ‘“samavaga’’ with - the
adjectival suffix an, “sama” in samavasa being taken in the
senseof ‘samya’ or ‘samatva’ (“‘bhava-pradhano nirdeéah’’).
The word ‘“‘sAma-vaga” would, however, appear to be a com-
pound of the two words “SaAman” and “‘vagéa”’ meaning ‘due
to the exigencies of Saman or chants’. Thus Wat Say arate
is read as QIati FqEq lane |
Next we come to Asthapita Sandhi. The combination of
a mute with a consonant is known as Asthapita. The word
““asthapita” is the past participle passive of d-stha and means
literally ‘made to stand by, arrested’. The Asthapita has two
varieties—Avasgangama and Vagangama. Where a mute
followed by a consonant remains unchanged we have Avagan-
gama Asthapita. i.c., an unyielding.or unsubmissive Asthapita
(iv. 1). When of the mute or the following consonant either
or both undergo some change the sandhi is known: as

* Ch, Sayana under RV. iii. 30. Gal Ya Ua Bla: 1 ama WaAqa qaa—

“8Qa GH: TU aa Anat ei ai? wa vdfafkafsmen aa” [aA

raigit | xfer
Saman is used in AA. iii. 1. 5 in the sense of ‘sandhi’ and in
TS.
vil. 2. 1 in the sense of ‘equalisation’,
198 TECHNICAL TERMSOF SANSKRIT GRAMMAR

Vasangama, i.c., ‘influenced or yielding’ Asthapita, The

following cases of Vaéangama Asthapita are mentioned:
(1) Unaspirate breath consonants followed by voiced con-
sonants become changed into the corresponding voiced con-
sonants (iv. 2);
(2) Unaspirate breath consonants followed by nasals become
nasals (iv. 3) ;
(3) § preceded by an unaspirate breath consonant is changed
into ch (iv. 4);
(4) A preceded by final breath consonant changed into the
corresponding voiced consonant becomes the corresponding
aspirate (iv.-5); :

(5) m followed by a dissimilar mute is changed into the

nasal of the following letter (iv. 6):
(6) m followed by an initial y, v or 1 is.changed into y, v
or | (iv. 7);
(7) n followed by | is changed into / (iv. 8) ;
(8) n followd by § ora palatal mute is changed in to
fi (iv. 9); |
(9) t followed by j or / is changed into j or 17 (iv. 10);
(10) ¢ followed by c, ch or § is changed into c (iv. 11);
(11) § following # or c which are the substitutes of n or ¢,
is changed into ch (iv. 12).

erat: WSSTAYAT-
TITTATAT ATIF TT |
MNIALIT: WTAE AAT,
CUTFTATIUATTAG|
 DIFFERENT VARIETIES. OF SANDHI 199

aa: TIAST TATA:

SHIc: MHeAayaesHi
Ty |
qaTraeaey Tarnyza-
CIT TTAATTAT CATT: |
feera wpa SEA ART:
AAMAN TIAA A, |
AAI MHI TT
at at varhreagerfernt g 1
AUT AHIT ITF BHI
THE VHTL-AHTLTAAT: |
ART HATHA
ATSEASATT STA ATHITA|
SRL ATTA? WHIT
aTMHAST AT AKAs | vle-22
a ganforata saga dart a=ofa, sat ang-
ATARYSIead | TAZ: |
When m is followed by r or a sibilant, it is changed into
anusvara. This sandhi is known as Paripanna from pari
‘around, fully’ and pad ‘to fall, sink down, perish’ (iv. 15).
In RP. xiv. 37 “‘paripadayanti” is used in the sense of ‘cause
the change of 7m into anusvara’. The word “‘paripada” is read
in the “nirudakadi” group (vi. 2. 184) ‘in Panini.
WHAM RTATIABISTNL AT TheT AAT: |
In RP. the “paripanna” is not regarded as a case of
VaSangama Sandhi, but Uvata, the commentator, holds that
200 TECHNICAL TERMS OF SANSKRIT GRAMMAR

it should certainly be so regarded. RP. evidently regards the

Anusvara Sandhi like the Usma Sandhi as a class apart.
TH asain alzayd AHILISaeaITaITa | Aq Aearar
qficamguara: | wagaers fadiat den aratat gat—
‘afa arr afeqeaoga aq” (4a), “aftart orea-
menatery” (24122) gatzat| =‘ augur’ wena
QW FHT: | BeaNIPwIasTIish: | Taz |
The development of glides into fully audible sounds is
known as Antahpata (lit. ‘coming in between’). Three
cases are mentioned in RP. :
(i) the insertion of k between final 7 and breath con-
sonants or sibilants (iv. 16);
(ii) The insertion of ¢ between final ¢ and s, and final n
and s (iv. 17);
(iii) The insertion of c between final # and ¢ (iv. 18)?
SRITSAATAIS
AT
HHL THAR |
ATS? AHUTAAAHIT
ARM WHC AHI ||
ASAUATAT: | BlZS-2e
Now we come to Visarjaniya Sandhi.
The elision of Visarjaniya after @ before a voiced conso-
nant (iv. 17) or the elision of a rhotacised Visarjaniya before r
along with the lengthening of the preceding vowel (iv. 29)
is known as Niyata which meanis literally ‘restrained or sup-
pressed’. These Sandhis are called Niyata because the Visa-
rjaniya is suppressed in these cases.
 DIFFERENT VARIETIES OF SANDHI 201

faasttia SRTCARH Srseqque: |

atart saqaedt aed Preranferd} |) y123-2¢
Change of Visarjaniya to r before vowels and voiced con-
sonants is known as_ Repha Sandhi (iv. 27).
When a Visarjaniya (after a long vowel). is dropped before
r, the sandhi is known as Akama Or reluctant, because people
are unwilling to pronounce two r’s consecutively (iv. 28).
When ah is changed into o before voiced consonants, the
sandhi is known as Praérita ‘bent forward’ (iv. 25).

BA TaI acTATaTc
tH Tat aA Yar Hae: |
MHizar gad aifsarrat
SAA
THTAAAATSA
TAA I 1R9-Ro
When Visarjaniya is changed into Upadhmaniya, Jihva-
miliya or a sibilant, the euphonic combination is known as
Vyapanna Usma Sandhi (iv. 31-35), from WVyapanna
(past participle of vi-d-pad) ‘changed’; when however, the
Visarjaniya remains intact before k, kh, p, ph or a sibilant, the
euphonic combination is known as Vikranta Usma Sandhi
(iv. 35) from ‘vikranta’ (past participle of vi-kram) ‘passed
over’. See p. 203, bottom.

STAT LEAIGT STAT ETT Ta |

ATAAMALATS TATA
TIATAAATIA MTA aeAfot Braz |
sata Saalea: a ferntra: THAT: |] BIZ2-3E
The change of Visarjaniya into s or s before k or p is
 202 TECHNICAL TERMS OF SANSKRIT GRAMMAR

known as Upiacarita Sandhi (iv. 41). Upacara is used in

this sense in RP. xiii. 31, CA. iv. 74 and Mandiki Siksa 108.
ganfee atfrga: vat
aanrcaeaistitha: RAT |
TH A TeTASraAIs FJ
eagararatca: @ afer: 1
Later on the word became Upacdra with the shortening
of the second syllable, just as Sanskrit upadhana ‘pillow’
became upadhana in Bengali. Thus we find in the Varttikas :
ATTA TAHA FSTAACITAT: e2sV1g
TTATTT
ALT B18
1RNS
On this Kasika says :
ffasarrenfanea Aaceararce xf Gar |
Bhattoji in his Sabdakaustubha 1. 1. 47 follows Jayaditya
and says:
fanenfa anager zfa ora dar |
It will be noticed that the explantion is good so far as
it goes, but it does not go far enough. In the Kaégika on
_ viii. 3. 48 VWamana says:
afafeaert STAT: HeATaT TTA: |
Here Vamana regards both s and s as Upacara. The
Nyasa explains :
fatten: AHCI aT Ga |
ferfear|
The Upalekha-stitra gives the correct explanation :
FAT THT: AHL FT HIV Grart-daneaeq ST:
gfe | 4k
 DIFFERENT VARIETIES OF SANDHI 203

Cases of tmesis in RV. are known as Ananupirvya

Samhita or Samhita not following. the proper order (ii. 78).
Thus cit intervenes betwen Sunah and fepam in “unas
cic chepam niditam sahasrat” RV. v. 2. 7, va and ca between
nara and famsam in “nara va Samsam pisatlam agohyam”
RV. x. 64. 3 and “nara ca gamsam daivyam”’ ix. 86 42.

sata farted vcr at ata gaurd |

AT a qe Fa at Hara eaafear tl) V9
We may in this connexion compare BD. u. 100:
frags’ a aaeqa a Hera |
“And one that is far removed one should bring into
juxtaposition, and should (then). arrange the regular sequence
(of the words).”
When final n after vowels except a and a is changed into
r along with the nasalization of the preceding vowel as in
TALR:, Tuy etc., the sandhi is known as Sparsa-repha
Sandhi*, because here r is the substitute for a (nasal) mute
(iv. 72).
Similarly when final m is changed into s and the preced-
ing vowel nasalized before t, th etc., the Sandhi is known as
Sparsosma Sandhi because the mute (sparsa) is changed
‘into the sibilant s (Osman) (iv. 77).
CHa Ma THAaA caeateaat | Taz: |
When final n is changed into visarjaniya and the pre-
ceding vowel nasalized, the Sandhi is known as Vikranta
because this visarjaniya does not undergo any further change.

*Cf. RP, 11. 36: aarcanaaranaay |

 204 TECHNICAL TERMS OF SANSKRIT GRAMMAR

qf: afavat af: cote a: otal cael: oz: |

afeafnra Tae: |] Bsc
The change of a dental into the cerebral is known as
Nati. The word literally means ‘yielding, surrendering,
giving up some of one’s own characteristics and taking on
some of those of others’. It may be paraphrased by: ‘‘vaitasi
vrtti’’, or in the words of Kalidasa ‘“‘vetasa-Iil4-vidambana”’.
So when a dental loses its dental character and becomes a
cerebral we have a case of Nati or Vinadma which means much
the same thing. It may be objected: why is the change of
a dental into a palatal or into / not called Nati? To this our
reply would be: no separate name is necessary for those
phenomena, because they do not attract much attention, but
the conversion of the dental s and n into the corresponding
cerebral sounds is a prominent feature of Sanskrit Grammar
and demands a separate designation. The alterant letter is
known as Nantr and the letter that undergoes the alteration
as Namya.
We cannot explain the word “‘nati’’ with reference to the
position of the vocal organs because a cerebral is pronounced
higher up the mouth than a dental.
Oar Aa: | WAT UES
Gat Aca Aha APTAAAT | ST Lee
armen quearafadfa: | at aT vie
CA. uses Vinama (iv. 34) for Nati. TP. does not use this
technical term.
CA. uses “‘vinata” in the sense of ‘cerebralised’ in ‘‘abhya-
savinatanam ca’’ iv. 82.
 PRAGRHYA AND PRAGRAHA 205

The reduplication of consonants is known as Krama or

Stride (vi. 1):

watigearcafear frat
aaa: a maisfama a |
The lengthening of a short vowel is known as Pluti or
leap. The verbal form plavate is also used in the sense of
‘undergoes lengthening’.

alt Beal AAA ASeT HITE

aufas’ aman: a ates |
aa cofa at atraizer
aifaamfeg ate: |
MSAQH: Tat SAAS TAH | IR—%,

PRAGRHYA AND PRAGRAHA.

“Pragrhya” (from pra-grah ‘to seize, grasp, tighten
reins’) means literally ‘that which is to be restrained (from
entering into sandhi combination with a following vowel)’.
‘“Pragraha’’ literally means ‘bridle, rein’. Hence in gram-
mar “‘pragrhya” refers to certain final vowels which are
not combined acc. to the rules of sandhi with a following
vowel. The word occurs in the Pratisakhyas and Panini. TP.
uses “‘pragraha” for “‘pragrhya”. The Pratisakhyas and
Panini give the name ‘“‘pragrhya’”’ to certain final vowels and
then lay down that they are not liable to changes acc. to the
rules of sandhi. Latter grammarians generally considered these
two steps unnecessay and merely laid down that. certain
final vowels were not to be subjected to changes acc. to the
206 TECHNICAL TERMS OF SANSKRIT GRAMMAR

laws of sandhi. In the Jainendra system the ‘‘pragrhyas” are

known as “di,” doubtless from the “‘di” of “‘dvivacana’’; cf.
“ki? for the vocative singular. In Sak. most cases of pragrhya
are known as “git” (because they have an indicatory g).
In Sak. the substitutes prescribed in the case of the nomina-
tive and accusative duals of i- and u-stems, the nominative
plural of adas etc. have an indicatory g, evidently suggested
by “‘pragrhya’’.
WOT | ArH aaa:
aa: | Ts ATA: | aq
qa | wiz fraatsemarseaar stat: | arafant
aqai| aegRaatacge:| waa aged 4
qa WI! varsafencia gat calsget arf:
MASA | wWec-9'
THA HLONAL YEMT; ATT ATA] NUR
aT | WIG) catarciarr fRraatear: | anita
qzaSaAIHE: | THAISTH: | AT AeA Gl A sara
HAT | WAL IRS
TIE TL | vied
AMT] AM MT | ATE) ele-R1 efaresfe
gat: | Ararcisatfedtsnrceaaarge: | vie-3 | aerATe
RI | AeA | a geafrwated: | vlc-pe---
A GANA | ors
“Tt occupies here a good deal more space, because the
Taitt. Pr. avoids on principle the mention of grammatical
categories in its rules, and is at infinite pains to catalogue,
word by word, what the other treatises dispose of summarily,
by classes.” Whitney on T.P, iv. 1,
 PRAGRHYA AND PRAGRAHA 207

AGCaITH | THTTAaTagHe | Se: wyerg | fara-

HM FT amas, ferred) waz) a TRA
Rafa atari: at agaaaq. farqraisaatsararc:|
AERA | anfiad AaaarT#, eper-ce
ATA THAT 3133
mal fans dfeatt aagi1 zq) apa
at ft | gs-og gat) Sa ae) saad aa) at fir |
S899] WW) AAT] we-coy By e2l
| ta fad seq) agama) 2 fara
THATS | A Agel wHeTetaaant) Ia Ts
teat
a Garret teize-2¢
coat
ata Tre] e1R1gR%
AT) Gzeat aes at faqra: cat smear), fFaa-
Wa | agtaaadt | agufrera) sale-e
a, a coatsfrat: (cgaisravafe a arta |
af: 1) afe ar | egea fraaaq | AT AAT | ATATS |
ary. at aati sar Hi calsaena acer |
MIE ATH: | 3-2 23
IIR
a) tee fe fe)a) sare) tay a aaly
Sa | Fy URS [ TTATHT1]
aTBAvlarere- “222|
ati acocenfrat) fita:) aeeattens: | ata)
al aat| Hast) eeiee-gov
f1 [arerefea1] caatsfrat) <a at) fede
fgqaaq) aagigqat, arfz: catisrs_| glare) a
AIA | HT 21R132-3e
208 TECHNICAL TERMS OF SANSKRIT GRAMMAR

al [ae 4 afer) aaa, afaaer|

gizeo-2¢2 | fsaaafamiardigzarg) azar arfeu::
amaqadararta | att: eqfanaaat: | coaerta
ATtasar | Tal ALTSIUa: | WESATTAHT SAT | 713RR-awr
Al antiscatse_fa: pa, cassia, eaat
Sa_ Tafa TAT | TH ATSSS | WTR] Be-e
gi fa afeafferrasta | 212129] fe@ar feaaaer |
ATE ATA | THRATA
eS TSAaAeT | Vita: | aT aragrfaat |
TAF A) TisTert | eeqraTates: THM) LIRIRS-2e
art amat) zafea) sfama) ca: |
q| fgaaadre? era afeafacifa <caa: oearfi|
Mzcarg fatal ALT aA a Haeragsa | agar facqa’
qgamifa Wa) 21 aera) ense-cx | atafefad-
azeafaal at) zie?
2. [act a afeatteq] atcarararaacaat
maa apart, igeat fraaaey aotierfzasgq | azet-
Ia | ARGENT A) wso-93 ata agfatad-
gal afesat, saeeara® aaa at) sweat
aft 71Q0-8%| Fa Bras Fada: 7122
In the Bhattikavya (vi. 62) the technical term Pragrhya
is used in a very happy simile in the following stanza :

TATA Areal eTBaqiatatnary |

Ayal HaHa tavamataferary |
‘(Rama and Laksmana saw the Sabari) as chaste as a
word ending ina Pragrhya vowel, of transparent form, free from
 AKSARA 209
all perturbation of mind, thoroughly independent because of
absence of all desires, godly and irreproachable.’
Here the Sabari is compared to a Pragrhya word, because
she did not mix with anybody. In Shakespeare’s King Lear
li. 2. 68 we find ‘‘Thou whoreson zed. thou unnesessary letter,”’
and in Love’s Labour’s Lost v. 2. 53 “Quis, quis, thou
consonant’’,
The following extract will be read with interest in this
connexion :
There are altogether eight groups, and the last of them
consisted six gases,....all of them curiously alike—helium
neon, argon, krypton, xenon, and radium emanation. They
are called the inert gases, because they cannot be got to
com-
bine with anything. The German chemists called them
noble gases, because they refuse to have anything to do with
other elements inferior to themselves.” Book of Knowledge,
p. 4222.

AKSARA
‘A-ksara’ literally means ‘imperishable’. Then it came to
mean ‘that which does not admit of further analysis, element’.
Hence when sentences were analysed into words, Aksara
meant a word. When the process of analysis was carried’ still
further and words were split up into syllables, “Aksara’ meant
a syllable. When syllables were further sub-divided into
letters Aksara came to signify letter and vowel as the most
imperishable element in language. (Cf. atom in chemistry. |
O.P. 129—14 ©
 2.10 TECHNICAL TERMS OF SANSKRIT GRAMMAR

‘Atom’ literally means uncut, indivisible, from a@ ‘not’ tomos

‘cut’.)
In RV. i. 164. 39:
Wat Hate TA atlas aftr Far aha faea Treg: |
maa az fraat afterfa a cafseea A aaa II
‘Who does not know the word of the Rk on which in the
higest space all the gods have taken their seats—what will
he do with the Rk? And those who know it here they seat
together’,
‘Aksara’ means ‘word’ or ‘syllable’.
In RV i. 164. 24:
Taam ofa fietet aaah ata aT aa aaAy|
aad ath facet agqeqarata faa aa aren: ||
‘With the Gayatri-foot he measures out the stanza’ of
praise, with the stanza of praise the chant, with the
Trstubh-foot (the stanza of) speech, with the two-footed,
four-footed stanza of speech the symn. With the syllable
they measure out the seven (i.e. the voices of the seven priests
or the seven metres)’,
Aksara signifies a syllable.
In SB vi. 1. 3. 6 aksara is derived from the verb ‘aksarat’
by popular etymology: yad aksarat tasmad aksaram.
In the Nighantu ‘Aksara’ is given as a synonym of ‘vak’
(speech) and of ‘udaka’ (water).
AA. explains ‘Aksara’ thus: °

TT a AAW, TI ara: Awa wera acfa, a aa-

*Cf. RV. i. 164. 42: ga: UTA |

 AKSARA 241

nfceacfea |& wae: wera yer: exch, =edawfraxfe,

TATZATA | AeaTgerclrerCae
aarae
et | BRI.
“He is syllable, for he pours forth gifts to all these beings
and because none can pour forth gifts beyond him. Becaus
e
he pours forth gifts for all those beings, and because none
can pour forth gifts beyond him, therefore he is a syflabk
e.
Therefore they call him who is (prana) a syllable.”
The Mahabhasya explains a éloka-varttiks which gives
several derivations of Aksara thus:

ay fafazercfata >
‘rat vat fiend|
av etre af ercdtfa arerca |
ate at aigerar |
AAMATAMOTH: AAMT | TAA CAAT |
aot" ave: qa 1
AIM Tras qorerrachrfa xer five |
‘Now what is this ‘Aksara’ ?
By ‘aksara’ one is to understand what is not ‘ksara.’
Or,
that which does not decay, does not perish is ‘Aksar
a’. ‘Or,
Aksara is -sara -added to the root af. Or this
ts the unddi
‘Saran’ added to the root ‘as’. ‘Aksara’ is so
named because
it is pervading. Or, a letter is called ‘Aksara’ in a previ
ous
_ sutra. Or in a previous stitra-work the technical term ‘Aksa
ra’
is used for a letter.’

ea 4 Gqz: | 4 9 aradaa, a vas: Waa | Vata ATW VRP

WUAMWA VITA afeaTATT 2 “MWe a eUfe, a Wiad,
zreww
wafa, aaise sia ari’? far: peer
DHT. - TECHNICAL TERMS OF SANSKRIT GRAMMAR

In AA. and Mbh.* ‘aksara-samfmnaya’ is used in the

of
sense.of ‘varna-samamnaya’ and ‘aksara’ has the sense
‘yamna’ (letter) in the well-known verse of the Bhagavadgita :
Aksaranam ‘akaro’ smi. We learn from Bhattara Hari-

candra’s commentary on the Caraka Samhita that in the

Aindra Vyakarana ‘varna’ was. used instead of ‘aksara” as
the grammar began with “atha vama-samihah”’ (cf. siddho
-varna-samamnayah”’ of the Katantra system which is said to
preserve many relics of the Aindra school).
RP. tells us in chap. I that the vowels are Aksaras:

Sal Bat: AAAA: TIA) Ae rat; Ta

CAAATIAAT | VPRG-22
Later on it amplifies this definition and says: “‘savyafijanah
sanusvarah Suddho vapi svaro’ ’ksaram” xviii. 32. Here
‘aksara’ has the sense of ‘syllable’. TP. also appears to use
‘aksara’ in the sense of a vowel: samakaravakaram tvaksaram
yatra svaryate....xx. 3. In xxiv. 2 ‘aksara-vyafijana’ 1s
used in the sense of vowels and consonants. In xxiv. 2 ‘pada-
samhita aksara-samhita varna-samhita anga-samhita ceti’,
‘aksara’ appears to be used in the sense of a “‘syllable’. In
CA. also ‘aksara” is defined as a vowel: “‘svaro’ksaram”’ 1.
93. In “‘thanekaksarena’’ CA. iv. 15 ‘anekaksara’ means poly-

e@egay sf aadlaaawaraa w, award «te:) vara

TIA VUVZ
Waa aaa: gia: aeapzaxaay ufaafes afeaa
Wei | wateguaaiarara mat wafa aafwat ae ai ak aaa
aera, wien rt (wy qeafeaat qawofeermara; | am cizys )
 AKSARA 213

syable. VP. says: ‘“svaro’ksaram. sahadyair vyafijanaih.

uttarai§ cavasitaih”’ i. 99-101.
In RT. also ‘aksara’ is used in the sense of ‘leter’ :
TANT SAAN: AATHAL | TATTA HAZ] AST
Teeqaa tara, seeqfafeeawest vcgisa, vcsist wie,
Ia cay: | d wfaaneaceaaraafraraaa |
In RT. 46 ‘aksara’, appears to be used in the sense of vowel.
Later grammarians do not generally use the word. They
perfer ‘varna’ as containing a lesser number of syllables.
Thus the very first rule of K., as has already been pointed
out, says: “‘siddho varmasamamnayah.” So does ‘TP. . ‘‘atha
varna-samamnayah” 1. l. |
K., however, uses “anekaksara’ in the rule ‘“anekaksarayos
tvasamyogad yavau” ii. 2. 59 in the sense of polysyllable.
The commentator Durga says:
a acfa a tad aachia free, aq qateranfeai
ma am careaia aa afaeranfrarg afsar carafe
an Hats | AATATACaATATa staan xfs ofrar-
mest a a aca a aadifa Hart at s=az gatas: |
The Panji says :
a acta a aodifa walaaraat ar seqa |
‘Aksara’ in the sense of vowel is divided in to two classes
—“samanaksara” ‘simple vowels’ (lit, ‘homogeneous sound’)
and “‘sandhyaksara” ‘diphthongs’ (literally ‘sounds of com-
bination’).
K. shortens “samanaksara” to “samana”. VP. goes a
step further, takes the ‘s’ and ‘m’ of ‘samana’ and forms the
technical term ‘‘sim”’, |
 214 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Panini uses ‘ak’ for “samanaksara’”’ or ‘samaéna’ and so do

his followers of the Candra, Jainendra, Mugdhabodha, Sam-
ksiptasara and Supadma schools. HN. calls these ‘dasdva-
taras’: Dasa daSivatarah i. 3. Since the vowels are called
Sarvesvara, the first ten are regarded as the avataras or in-
carnations of Sarveévara.
Both “‘samandksara” and ‘“‘sandhyaksara’’ occur in the
Varttikas of Katyayana :

array areas gf aT | «fat |]Ble

qRA Mites Aq Beat anrawrcfays: |
fr & wie, |
The word ‘sandhyaksara’ occurs fitst in the Pratisakhyas.
The formation is thus explained in the Mahabhasya :

CAAA AATTLAN ATATATST AAAI: |

“The diphthongs are composed of combined vowels; their
treatment is that of a simple vowel.”? Whitney.
RT. shortens “‘sandhyaksara”’ to ‘‘sandhya.”
In the Panini, Jainendra Candra, Sakatayana, Samksip-
tasatra, Mugdhabodha and Supadama schools as also in the
Agni Purana these are known as ‘“‘ec’’.
VATA | AT ATA ARATTATA | G111e
arvsamernantyd | ssafaal zeae aa adi
TAT SAHUT | 20184.) THafats aaa awwa—
afea saat afa feat tf adtaq1 2o1¢219
sfireterrafrrg smerny | area AEATTATT PRU?
aga Aelad ae! wat =F AHA
azercfafa TreTATaRs 1 VIReIR
 AKSARA 215

Ranzacfafa | aqacaetaafa) elyzic

oT) sel aararacwafsa: ) aasrafe aeeaerct-
TACT | V-R |
ARTA THIN THe wR alata aa aa
aia sacar fasqaed af aaa | wea |
avai, fararfeatsst cacrong |aeraat Te] else-v&
aq ot| wa aaa: aaratarcrfor |, eR
aa aarracoiifa edi da fRadarfsat » agsaa—
aq Heda saaeaatacia = |
GUAATAR FART ATTA:
AT AUNT | ALNTAKAAAa Awaq, agarara
CIVGATCASa Ada |... CHILITA: SU GqAcaoheaqieqnyas
muguafeywareae, AMA AAAAI wR za
SU: AAAS | + AAAI aay WaTTTVaT-
AAUAAAME | AIA AAG— Halal aan wus fey
STANTASTA VISAARMATAST Gat at Gata aaraTaarefe |
Afaracny |
agealfrar | deere deavanictaaniagha: gio
( BaTaAATeT GATT Fa: B1B2 )
TAM | 2 ZA BUAgen fafa 9 zg
AAA 1 VIR
Al | SRA AATAT: | VtVla, THs fstGreaercrftr | eieic
al AZT
8B Saar), T 8 at al |qezq-
aazcifoT | eGR
2) SrA BATA | et al qe] ezho-c
Saagzaeatcaawrafaafarsa anagy As TAT
frara aamiea fa aa get art ofan cfeset ar
216 TECHNICAL TERMS: OF SANSKRIT GRAMMAR

aati 2 aaa: weafaasanar fag afe t---anal

afa vat. ae) anf aaner wacat Gz-aeaNaAIe:,
CHHCRAAIT:, AAOeT Sacha ArHIT:, MaHITHITIAT-
MA: | BAAS YAT: |
qocaneifeet fireat | faqudant secean: |
T) Saga Bard War: QTHA Al tie,
marsagqanl a: are az) elec
SMTA aeqafiT| Tt a HI-
ATHEAT BT Bl HST STAY’ TTA: |
Te-am deacda favnety Heagar
TASTE QATAR | TATRA RTT
Zio FA FATA: 212
aaaZA AN! TANAAIT-ATATAaT Aafea y---Ga AATAT
aR UTA | Pa: |
omBatt ttATA | kkk
araef car) aft) aaetenta argza-
agin-ceraifregl
Sead ete aaerarrat |)=ztaT|
Sandhyaksara does not occur in TP.
Sandhis of vowels with consonants and of consonants with
vowels are known as ‘anvaksara’ in RP. The former is
termed ‘anuloma anvaksarasandhi’ and the latter ‘pratiloma
anvaksarasandhi.’
Varna makes its first appearance in the sense of ‘sound’
or ‘letter’ in AB. xxv. 7:

afa gare | |seasfrateasat ant senara

HIT THC AATEC Efa | MARAT ARATE | azazrferfa|
“He brooded over them pure ones: from them when
 AKSARA 217

brooded over the three sounds were bom; the letter a, the
letter u, and the letter m. Them he brought together; that
made (the word) om,”
He we have not only clear ideas of sandhi, not only the
use of ‘“‘varna’’ in the sense of ‘letter’, but also the use of
‘‘-kara”’ to denote a ‘letter’ after the analogy of vasatkara.
As Keith points out, “in KB. xxvi. 5 pada and varna
denote ‘word’ and ‘letter’ respectively.”
See also under “Kara”.
“Varna” is sometimes used in the sense of ‘syllable due
to contamination with “aksara’’, as in the following stanza
of Panditaraja Jagannatha:

ay Tradhrat wanaigae fagheay

frenarafaanfamoanaferniafaraea: |
RS UHRA eSATesze: rat
at faaftrerat fasraa wena anise1
wifaifiarea ZI1¢
In the immediately preceding stanza Aksara is similarly
used :

adie cfa fan aati ena fart oa:

mamea garerats afrey carat: wafcea: |
aed a fe agra sia waar yat wa seat
emieacan wafeaiet: aay fara: 1
MATIN SAK SKS? ATATARAT S|
ACRIAT: rat ararefasaaiza: rat 20133
* wad 4& a ae Sai wantaGamlnen “MEK Siren aait
RatawaR wae
 218 TECHNICAL TERMS OF SANSKRIT GRAMMAR

‘Of letters I am a, of compounds I am Dvandva; J am

Time Imperishable, I, the Creator Universal’.
Cf. “I am Alpha and Omega”. Christ.
Bhaskarasena makes use of the first quarter of the above
stanza in his panegyric to his parton :

AACA Setar Fawr: eae TA |

WAM STH A AAA SIH |
Subhasitavali, 2460.
‘Visnu himself said, I am a (akadra) To among letters.
tell the truth, Your Majesty has cast. Him into the shade by
your (splendid) form (akara) (even as before the long 4,
the vowel a appears short).’
This reminds one of the well-known example of Slesa_in
works on Rhetoric:

ccathragay: anf Geaafrae |

FIM HIT FT HATATH ITA: AZ: |]
‘A miser differs from a sword merely in external
appearance (Arpana differs from krpana because of the a).
Both are (1) “drdhatara-nibaddha-musti’” (a) very close-
fisted, (b) held in a very tight grip, (2) ‘“‘kosa-nisanna”’
(a) sitting in the treasury, (b) resting in its scabbard,
(3) “sahaja-malina” black by nature’
In the following stanza from the Naisadha-carita (vy. 121)
there would appear to be a blend of both senses of the word

SL ——

*"Akaratah” For a similar Slesa, cf. “A doctor avers that

women are eating too little. Figures prove it,”
 SVARK AND VYANJANA 219

“‘aksara’’, for negation is expressed: not by the letter n “but by

the syllable na :
aTreretior gear far sere: eae ofsarste 1
CVUAVATAIV SSA MZ BHT AHIT ||
‘Was it not read by him while he was studying the
alphabet or has it, been forgotten even though he did read
it?—Thus did the letter na disport’ itself in the swing of
doubts in the minds of the hosts of your supplicants.’

SVARA AND VYANJANA.

“Svara” is derived from “‘svr” to sound’ and is used in
RV. in the sense of ‘sound’. Then it must have meant ‘the
voice which utters sound’, and then the ‘tone of the voice or
accent’. In this sense it is found in AB.:
Tet gtafa) @ Aca eran seer: |; carat 4
CATT:, TAA: TTT: | TEATAGSST | RIS
He recites the Pragatha; it should be recited in accented
words ; sound is. cattle, the Pragatha is cattle; (it is) for
winning cattle.’
Next the word “svara” came to mean vowels which are
always ‘voiced’. In this sense it probably appears first in
AA. :
Wal XR qedlaea wera, ae af saath
TSUT AT MNT: A ATA, TIA: TW amt: | sqeata
Talcrgata, atch |
‘This is produced as a thousand Brhatis. Of these
the
consonants are the body, the vowels (voiced conson
ants, acc.
 220 TECHNICAL TERMS OF SANSKRIT* GRAMMAR

to Sayana) the soul, the sibilants the breath....They make up

the nights by the consonants, the days by the vowels.’
It would appear from the above extract that ‘“‘ghosa”
was the earlier name for the vowels, ‘‘vyafijana’’ for the mutes,
and “‘usman” for the semi-vowels, the sibilants and the
aspirate h. ‘“‘Sparsa” is used later on for the mutes, but
“asman”’’ appearts to have been used throughout for the semi-
vowels and the sibilants and the aspirate h.
It is interesting to note that the vowels are practically
compared to days and the consonants to nights. The com-
parisons remind one of Tennyson’s ‘the ‘greater light to rule the
day and the Iesser light to rule the night,’ and points to the
greater perceptibility of the vowels. The poet probably had
at the back of his mind the popular etymology of ‘svara’, viz.
“svayam rajanta iti svarah’’.

In the Chandogya Upanisad ii. 22. 3. 5 we find:

Vt CU VARMA, AF FSA: TAIAUKATA:, GA

CT WNT: | ITT Arad TAR AMET...
aq ssarnisaen afacen faaar aaa) aq wat
saaratatateat THAT: |
‘All the vowels are the limbs of India, all the spirants the
limbs of Prajapati, all the mutes the limbs of Death... .all the
vowels are to be pronounced voiced and strong....all the
sibilants are to be distinctly articulated, are not to be thrown
out with a jerk and are to be pronounced open....all the
mutes are to be pronounced without being slurred over in the
least,’
 SVARA. AND. VYANJANA 221

It is usual for grammarians from Patafijali downwards to

explain ‘‘svara’” as ‘syayam rajante’ and “‘vyafijana” as
‘anvag bhavati’. Thus Patafijali says in his Bhasya on
‘uccair udattah’ and ‘nicair anudattah’ i. 2. 29-30:
[ aratarfe saqaaa oda ( Seraarey: )ToT sear,
qT yacadma agree wars | J
aqy weafy fadad wrt cared crc sea wafa
saaatate|
(ara aadtfa| APTA: | fireeaararcsé
fasaaafafkagq, a aa aneredt fafa) aa at
afacft sqsaue: | fafei waqaamaefa aaa-
frart:, vafarata areratenfafea: | sacTer qa-
acrrafnaastr qtorat eatwate a yer | RAZ: |
qd una”of arsqer ‘Serafeaqst’ cea) wg-
Tasdta | angufafs ae) a aa anaresafafa|
fadaa fe aaeradniiaminaste qrareaaaena::|
SqaTA: |
This is amplified by Uvata in his commentary on TP.
xxi. 1:
FAS BUT TWP ECA ASAHI: |
FAS SAY ARSTA TSA AT: 1 [ATaeTeT-FAAT Lee]
faq freat-caTacat
a wag usta a J ACA Taaile: |
softenftar aa sage sagagea |
‘Just as a powerful king wrests away the state of a weak
one, even so the powérful vowel carries away the weak
consonant, Patafijali.calls that vowel (syara) which shines by
 222 TECHNICAL TERMS OF SANSKRIT GRAMMAR

itself, that which is manifested by the vowel placed after it is

a consonant’.
The Vrtti-traya-varttika says:
mataatste Usted AAA: CITT FE |
saaarna Paar: Gaara: ||
waa: weafa fadaat tad crea ef eat: (2)|
aaah AU ad WaT AAT:
‘Spiritual souls, like vowels, shine even by themselves.
Spiritless fellows follow others like consonants. The deriva-
tion is also significant: they shine by themselves and hence
they are known as svara (vowel), since the consonants follow
others, and the vowels never do so.’
Under RP. i. 3 Uvata gives the true etymology of svara:
svaryante Sabdyanta iti svarah, ‘those which are sounded
are svaras’. Under RP. i. 6 he says: vyafijayanti prakati-
kurvanty’arthan iti vyaiijanani.
Panini has no occasion to use the word ‘svara’ in his
Astadhyayi, as he donotes the vowels by means of the pratya-
hara ‘ac’. Candra, J., Sak., Sam., Mu., Su.* and others
follow Panini. K., Sar., and He. retain the older term.
PR. uses both. HN. calls the vowels “Sarveévara” because
the consonants have to depend on them for their pronuncia-

*In “vargya-svarau sajatiyaih Savarnau” i. J. 15, Su. uses ‘svara ?

- also. The author says in the Paiijika:

FAVA ATA RACY faaay ? werremagey, qae save < |
Similarly in i 2. 52 we find “jhasas trtiyo jhasi ca” where the
Panjika says:
AMAA faraasly ahaqed Yay qe gens 4 |
 SVARA AND VYANJANA 223

tion: kadinam uccdranam caisam adhinam iti SarveSvarah.

Vrtti on HN. i. 2.
The consonants are termed ‘“‘vyafijana”. The word is
derived from vi-afij ‘to manifest?’ and means ‘that which is
manifested’. ‘The word appears to be used first in its techni-
cal sense in AA.,
Panini and his followers use the pratyahara ‘hal’ for
consonants. Mu. has ‘has’. HN. uses ‘Visnujana’ for
‘vyafijana” owing to the similarity of sound and explains:
Visnoh sarva-vyapakataya SarveSvarasya jana iva tasya’
dhina ity’arthah.
In the Bhasya on “‘ajader dvitiyasya” vi. 1. 2. Pataiijali
compares the consonants to actors’ wives:
aaa grasa wafeat ag an azrai fear
Uy Wal
at a: gestaeA HATA? aad
qa adage: | Te saaaeae wel Aes: HaAEaa d
AAT | '
“The consonants, however, are like actors’ wives. Thus,
when actors’ wives go to the stage, to whomsoever asks them,
‘whose are you;’ they reply ‘yours’, even so consonants also
cling to those yowels which are: subject of some grammatical
operation.’
Under “‘afico’napadane” viii. 2. 48 Patafijali explains
“vyafijana” meaning ‘sauce’ thus:
[wet Ca aad ST TTT |“ores feet”eee,
aad fad aad aaq carafe] ams sigan ©
SAGA TRINA | TAL HAT TILT aTaSiHararfafezae
aiamafa saqernd a uerearaaq) weal @eafy
faqaaq—‘cagasaafa sagag’ xf |
224 TECHNICAL TERMS OF SANSKRIT GRAMMAR

‘Vyafijana’ is derived from the root ‘afj’ (to manifest),

thus ‘vyafijana’ means ‘manifesting’. The bringing back of
the senses rendered torpid by oily substances and sweetness
to their own natural state is known as raga, or, in other
words, ‘sauce’. And there is also the significant derivation:
that by which (rasa) is manifested is ‘vyafijana’.
In the Katantratika ‘svara’ and ‘vyafijana’ are explained
thus: svayam rajanta iti svarah, ekakinopy’arthapratipadana-
samarthah. This is thus explained in the Pafiji :
anf fsemafraraaa aaemaacusfaae
armreagraaad, aa aaararfarta |
‘Vyafijana’ is thus explained :

asaa faftfa aaah, aa qaidiraizay,

aaareate sata taceaeer
The Paniji explains:

aaata ofaftfe soanf, acomasfagiza sq-

RICHI AAT Taare
OT) Bl aararceasa: |) acacafe aeqerct-
omar |TA TAT! | R-R |A: TN SAAN | Ue
ATH) AM MAUI TIAA ¢1R2 Bea casas | clo...
Jat. weurfsa eau: | AM sash | vy-s
BT) Aa agaMal aT |Ree | Breer eaATA
aka hs ,
2) abeeae AT | rete | STC CUE orf 212120
sasaa 9scaHcifmadsaisaafa SISA , CATLIN THT

TTAICHA,, TA GUAT | WYATT: |

 HRASVA, DIRGHA AND PLUTA 225

T)
wnicifsarnera aAswaada |
ARATAHCI He ATS: ATT AT yp Lle-9
HHUPREHILAT AMAA THA Ty] V1
wy TEI TAIa: Aa | Sa cise efa ary “aararsfy
a” [gsjger] ata Si) “watsean—” [eweec] geatfear
faqra: | ea wal fats) |..atecita asad hh
sassattea | faqaigsrartitaz | vet g casafte eqcra-
Tada fageatfa acarfsearza gears: | cantare |
Ei) Aaa Aqsa AAT: | LILI
Od aU Ba GATT, | Oe RaeaT STOTT: | ATTA
yay Gia atafata Baeaczs: |
HUTA fesse: | VIRILS

HRASVA, DIRGHA AND PLUTA.

‘“Hrasva’”
is derived from the root Aras ‘to become short
or small’ and means originally ‘short, small, dwarfish’. In
the Srautasitras it is used in the sense of later laghu i.e. pro-
sodically or metrically short. In the Pratigakhyas it is used
in the sense of a ‘short vowel’. The root hras is used in the
Nirukta and RP. (*nirhrasitopasarga Nir. ii. 17, vi. 11:
sparsantastha-pratyayau nirhrasete RP. iv. 90) in the sense
of shortening of a vowel. In the third chapter of AA.

* “Prakrtinirhrasa” occurs in Kasika iv. 3. 1000:

als: @ afaceifa vafa-fast@ aa aca:
(rafafasté wa sfa1 saan naefaaia farlAsmawed sae!) ar
sfa\ are nafantey ae TaraI vaio wala | |arimafaacanfeat | )
O.P. 129—15
 226 TECHNICAL TERMS OF SANSKRIT GRAMMAR

where we find speculations on phonetics, “hrasva’’ is used

as an adjective to Mandtkeya in the sense of ‘dwarfish’
but not as an adjective to svara in the passage (iii. 1. 5) :
aq ad ant franca fa e sare seat aveae:,
qaraen TRI, SATHATAT aisaHa: FaRrTATRa
aratn aa afed faadafa, aa caqcreat fasiafa, aa
mamat
farsa art dfeate |
‘Hrasva Mandukeya says: We that recite the Nirbhuja
say that. the former syllable is the former half and the latter
syllable the latter half, but that the union is the space
between the former and the latter halves in so far as thereby
one produces the union and distinguishes accented and un-
accented and separates the mora and what is not.’
‘“Dirgha” is an older word than ‘‘hrasva’’ and goes back
to RV. It originally means long in space and in time. In
the sense of a long vowel it is used in Katyayana Srauta
Stra, Gobhila Grhya Sitra ii. 8. 15, Sankhayana Srauta
Sutra 1. 2. 17, x. 5. 28, etc.
‘““Pluta’’* comes from plu” ‘to float, swim’.
e6
Thus on
3:3 ‘

the one hand the root means ‘to vanish away’, on the other
‘to swell’ i.e., ‘to be lengthened or prolated’: ““maksvityukarah
plavate (is lengthened) sarvatrapy apadanta-bhak” RP. vii. 5,
SS rer el a
* Pandit T. Venkatacarya says in the Journal of the Sri Venkatesvara
| Institute, Vol. viii. No. 2, p. 16: Pa Wada alan: aq
gaaetiaaanfane fanaa Taeatia aa: | da wae AlaaHaAqEa az
ada a faaiatead fiteq gam: cenfanfane asda ofeanfera: | saan
gad
= Bala WT aei=faniaa’ apf saatq yan: fanaa sais
way,
All this is rather fanciful.
 HRASVA DIRGHA AND PLUTA Dor

“saiva plutih” (lengthening), RP.vii. 2, “saiva plutir iti

veditavya ya harsvasya dirghata”—Uvata. Again “pluti” is
used in the sense of ‘prolation’ in ‘“‘adhah svid dai 3d upari
svid asi 3d arthe plutir bhir iva vindati3 trih” RP. i. 31,is used
in the sense of a lengthened vowel in ‘“‘adyaksaram plutam
tesam’’ RP. ix. 32 and in the sense of ‘a prolated vowel’ in
““i-paro dirgha-vat plutah’ RP. i. 4.
RT. uses “‘sva” for “‘hrasva’’ in the rule “‘sparsah sve” 25
and “bhe sve mantasthi’” 150. Similarly for ‘‘dirgha” it
uses “gha” in “gham” 50 (dirgham ca guru-samjiiam
bhavati—Vivrti) ;. “rau gham’’ 93 (svarau dirgham apadyete—
Vivrti).

Sak., Kramadigvara and Padmandabha use“ hrasva’’.

“dirgha” and “‘pluta” without defining them.
J. uses “di” for “‘dirgha” and “pa” for “pluta” and to
make out a word with meaning uses “pra” instead of ‘‘hra”
for “hrasva’’. Thus with the symbols for hrasva, dirgha and
pluta the nice word ‘“‘pradipa’” was formed, the science of
grammar being regarded as “pradipah sarvavidyanam.”’
Vopadeva uses the second syllable ‘‘sva” and “rgha” for
“hrasva” and ‘“‘dirgha’’ respectively. For “‘pluta” he should
have used “ta”, but as the latter is required
for ‘“‘taddhita’’,
the claims of which are far greater inasmuch as it both begins
and ends with ta, the first syllable “plu” is used for ‘‘pluta’’.
HN. calls the ‘hrasvas’ Vamana and the ‘“dirghas”
Trivikrama. Now Trivikrama being used up for the dirghas,
the plutas are called Mahapurusa. From HN. we learn that
the short vowels were also called ‘“‘nirhrasva” in several
 228 TECHNICAL TERMS OF SANSKRIT GRAMMAR

systems. The name Trivikrama would have been much

more appropriate for Pluta. “Urugaya’’ or ‘‘Urukrama”’
might have been used for Dirgha.
ATT | |AAT RAT: AAATeaT: TATION | AFI SAT:|
Sat sercifir) eee-2e (ate fawar: carr wa
axaqdat afzacat: aanqaeat:; wat aw:zs eas
Taz: )
aat} marcenrd aeal) same), aa a ana
areeat: |agearea | faearara dra: | fa: cpa: ) eiae-s
‘ATaT| aaraeatt set} ise) fear ete:|
cafe: | gitxo-4¢
aqcatera | TRATat sea: | fearat
ata: | faara:
Cod: | 2IKS-&F
Ma) sealers gatgea: | 3
T| HASSAGAAT Fa: | LAIR
Ul Tal Se] TA TT ereie-9
HH | TSTSAL TAT ke
a) wa aan sat tte: cga efa faa fra: | ca
atfant ata: | faarfaat tre: | famfan: qa: ) sacra
2E, BI-BR
*A poet-aster has framed the following Samasyapirana stanza
with these rules:
yal ea: 9G Ste: gai TS farcarea |
saat faadag gat de: gat aE: 1
‘The friendship of good men is slight at first and constantly grows
great afterwards; that of bad men is just the reverse, great at first
and slight afterwards.’
 GURU AND LAGHU 229

21 wafsfaatat atadiscgat: | eit

A) MATa) &
T| ACIRWSAl: YSrws HAsAsne |
AeA equ Hadar arasaa feats: ||
WATS SAT UAMIART: FAT 1 UZe-zR
Zl Gata: | wheats: faamt wages: i
214-9
Tea | wa auger fatgera
a ca
GAA |
war geaeaat way stat faraca: ||
VBIRV-RB

GURU AND LAGHU

“Guru” is a very old word occurring in RV. in the sense

of ‘heavy’. In its technical sense it means ‘a vowel long by
nature or by position’, a vowel which is long both by nature
and by position is known as “‘Garlyah’’. It its technical sense
Guru occurs first in RP. The opposite of “‘guru” is “laghu’”,
the younger form of “raghu’. “‘Laghu’”’ literally means
‘light’ and is used of a vowel which is metrically short. A
short vowel which stands alone without any consonant is known
as “‘laghiyah”. RT. and ST.. use ‘ghu’ for ‘laghu’, e.g.,
“vugmam ghu” RT. 236. “u ghor ghuni ghosadih” 237. J.
uses ‘ru’ for ‘guru’. In the system of Panini ‘ru’ represents a
‘visarjaniya which stands for an original s’. Sak. uses “guru”
and HN. has no objection to the use of the. word in rules like
 230 TECHNICAL TERMS OF SANSKRIT GRAMMAR

“‘rccha-varjita-gurvisvarader 4m adhoksaje’’ 1. 481. Candra uses

‘laghu’ and ‘guru’ in his rules, e.g., “‘laghor upantyasya”’ vi.
2. 4, etc., but is silent about these terms in his Varna-Sutras,
though he defines ‘hrasva’ and ‘dirgha’ there. Kramadisvara
does not use ‘guru’ in the rule “‘dirghopanto yuktantas ca
dhatuh”’ i. 74 where Goyicandra says: gurupanto dhatur iti na
krtam vaicitryartham. The word “‘guru”’ is, however, used later
on in ‘“‘gurvijader anrcchornoh” ii. 1725 and in “anrser
antajdvitiyagurvacah” iv. 13. J. uses ‘ghi’ instead of ‘ghu’ for
‘laghu.’ It seems to have some objection to ‘ghu’ as it not
only avoids ‘ghu’ here, but uses ‘bhu’ for Panini’s ‘ghu’. In
most systems, however, ‘ghu’ stands for the roots da and dha.
In several early systems of grammar ‘ghu’ meant ‘uttara-pada’.
Thus in a Slokavarttika under P. vi. 4. 149 we find ghau (v. I
dhau):

a) smistartaa wren Qseaase* |

q-medaraced garqraifagiead arenfafa | age
ACTS LTA: | HAT |
It is just possible that ‘dyu’ is the correct reading here, as
in J. an uttarapada is technically designated ‘dyu’: uttara-
padam dyu 1.3.104 (se i.e. samase yad uttarapadam tad dyu-
samjnam bhavati. Mahavrtti). This ‘dyu’ is evidently composed
of ‘d’ of ‘pada’ and ‘uw’ of ‘uttara’ with a ‘y’ thrown in between,
to distinguish the resulting word from ‘du’ which, in the
Jainendra system, is a technical- terms for Vrddha of the

* What actually occurs in AV. is “ye anti ye ca dirake”. Kat. and

Pat. evidently read ‘antiye ca dirake”.
 GURU AND LAGHU 231

Panini system (see J. 1. 1.68 ff.). Sak. uses ‘uttarapada” and

is thus of no help to us in determining the correct meaning.
Vopadeva uses ‘ghu’ for laghu’ and ‘rw’ for guru’.
It is difficult to. understand why K. and its followers use
Laghu and Guru without defining them. They evidently think
that the words are too well-known to require any explanation.
But Hrasva and Dirgha are still better known and yet they
have been defined. The Vrtti says under i. 1. 6:

eral Byala Tefeaqarename waa, cat dat ata

sesh ae:, “Gendtsass.” [ aeiee ] afa asters |
The Tika says:

squedanht Sarva [ ekg ] kare se BY-

frente|...7g af geatiatafe ated aa fal) ae,
aeafafrqeaiaa fe al, adt fe co-ed ofiqaed
TTATZ; |
Acc. to works on Siksa, y and v.are “‘guru’”’ when initial,
‘‘laghu’”” when medial and “‘laghitara”’ when final.

AIT! Taf aif adatat aargencacifir

Aika | RIRo-2Q
GA SA TA AT GAT FAT: | Vc ge-ge
ae Dea aierg ae aera wT |
BT Asgsaad FI TAA SASMATE | Vcier-wy
TA) ag casaard ag aft 8 dard’ a ang-
apaTRz|
carta aatfn wef faraoradiseanfa aat wate II
AAAs TMAAAATIL AAT |
232 TECHNICAL TERMS OF SANSKRIT GRAMMAR

AGATA Aq AGA Frag || RRLV-z%

aa) aaaTqesaraasaanar: eat fear |
B] Zoe,
ageafyar | ga aradnit, dead) wafer
CA ed
Mal De afr, aq eiee-so (af=aqai|
T=ZT | )
qt) sadaa, aaitigqe, Ar ay cvigo-g2
A) St fra, the) A eeeee-27e (q=eea|
Roda | F=ats | )
Surana: | Tat frase GAVIA WE: |
RaaTTT seat fraser foRat a: ||
F | aat oh ME] wvo-uge | ( <=8e,
g=31e, M=AANT | )
Gl salsa | Tat We: | searar Par) w7124-2¢
1 | Sane FW aT Te aa AeA | tk
agate a aa a yeas
Tvetrar aaa fa are ties
a | aaat wal: | fafamar Je: | Aq-asgiTd qat

ammatsht Tei) wee, ( qaAA=sea| faferaazis

ALAK=AaArT | )
“Guru-laghava” is very happily used in the following stanza:

qe Wearad A woTATASs a aha

qd aa Wad A get Tag arat wa: |
 NAMIN AND BHAVIN 233

ae aragra faafa agal tarasezraci-

“aguante: we: garfasnca geafrar: |
WATATENAITSTTIT, €2139E
cafradia Hat wareTrey MATT: |
aaraagat gaat a fase wet Fat:
SENT 943
NAMIN AND BHAVIN

‘“‘Namin”’ is the name given to vowels with the exception

of a (short, long and prolated). The word is devived from the
root mam with the suffix nini and means much the same thing
as ““‘nantr’’ (RP. i. 66), viz., that which induces “‘nati’’. It has
already been mentioned that the change of a dental (especially
of s and nm) into the corresponding cerebral is known as Nati,
probaly because the dental bows to, i.e., yields to the cerebral.
Hence those vowels after which the dental s is changed into
the cerebral 5 are known as Namin. Thus R.P. says:

MAU Tq Asa: AT: TAT Arar aay AFAATY |

CGn-& av

varia zed aed qaed Haedifa afta) wars €

TRaU
a fa, yar
VP. defines Nati in the rule ‘“‘dantyasya miurdhanydpattir
natih” i. 42, but uses the term “‘bhavin”’ for ‘‘namin’” :

ARIAT UAT | wee

RTARTA Tifa acai arfi-der wafer)
SAX: |
 234 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Neither Uvata nor Anantabhatta takes the trouble of ex-

plaining “‘bhavin’’, but the term appears to have reference to
the phenomenon of Nati. It is derived from the root bhai
with the suffix nini and means that which causes mutation of
a dental s into the corresponding cerebral.*
TP. has no technical terms corresponding to Namin.. CA.
uses the word in namyupadhasya rephah”’ ii. 42 etc., but does
not define it.
In ASS. i. 5. 10 ‘“‘anatyaksara’’ is used in the sense of
Namin.
P4nini and his followers use the Pratyahara “ik” for Namin.
Katyayana uses “‘vindma” in his V4rttikas :
syen Argefafa-aren-farargarseny |foWe vize
ata afm faara-atafa-ofama: | caieele
Patafijali uses “‘vinata” in the sense of a dental s turned
into cerebral :

aq wiifa fRawqe adtat, at faaasacra:»

afaaa frat ata) gaat ciaige
But neither Katyayana nor Patanjali uses “‘namin’’.
K. has “‘svaro’varnavarjo nami” i. l. 7. By the time of
the commentators the original sense was entirely forgotten.
Hence Durga says in his commentary :

ama wm, atsaredifa ara, aa ai cafag e-ara-

Aza waa og fact, a ara’ eafa |

* Monier Williams says : prob. as liable to become the corresponding

semi-vowel. Weber explains: yo bhavayati, i.e., namayati, that which
causes bhgva, i.e., mati.
 AGHOSA AND GHOSAVAT. 235

Kulacandra says :

aifiat am: aa afaematardmnad, a cafaar

tfa ara: | aaoteg LAATASTAATTT |
Hc. has the rule “anavarno nami’ i. 1. 6, but neither he
nor his commentators explain ‘‘namin’’. Sar., as usual follows
K. and frames the rules “‘a-varja naminah” i. 5. Purusottama
lays down :
CHILE TAH ALAS ACA THAT 1 TUL
aaa wa: aAyseardifa awit: zara ag cafa-
free: 1) artsranrfirent |
HN. calls these vowels ISvara: a-d-varjitah sarvesvara
isvarah i. 8, probably because people bow down before God,
and also because the word “ISvara” is part of the word
*“Sarvesvara.”’

AGHOSA AND GHOSAVAT.

Sabda, Dhvani, Ghosa, Nada—all mean ‘sound’. Of these
Sabda was seized upon by grammarians and used in the sense
of ‘sound, uninflected word with a conventional meaning.’
Thus we find in the Mahabhasya :

aq mfterst &: eeg? fa aq aa arene eaeae-

aqufyagqeed a ez: 2
acale— set AAA |
ad ate afafed afsd fatafad a mea: 2
acaTe— finat ATA AT |
ay ate aaa ate: SU: afte: aaa cfs a weg: »
236 TECHNICAL TERMS OF SANSKRIT GRAMMAR

ACIS —IINT ATA A: |

aq ae ag fartcafiedt ferteaiest ararend a
eq. 7
acare-—arataata Ar |
HRS NE: ?
Take Ua sez -at-fasfrat aaa
waft @ Wer: | Awa Adiagaiat wR wafe: wer
yaa | AZ waN—aes He, aT west H:, wears
moran: cf waft aaetaqead | deme vata: mez: |
‘Now when we say “gauh’’, what is this ‘‘gauh’’—what is
the concept behind this word? Does it mean a thing possessed
of dewlap, tail, hump, hoofs and horns? No, says the
writer, for that is merely the substance. It is then a bundle of
various movements indicating internal feelings (such as licking
the calf), or voluntary (such as walking) or reflex (such as
movements of the eyes, ears etc)? No, replies the writer, for
that is merely activity (or motion). Is the word then some
colour like white or dark, dun or spotted ? No, says the writer,
for that is merely a quality. Is it than something which cons-
titutes the generality which remains undifferentiated though the
individuals might differ, which is not destroyed though the
individuals might be destroyed (is it the general conception of
gotva ‘cowhood’) ? No, replies the author, for that is merely
the genus.
Then what is ‘word’? That-is ‘word’ which when uttered
brings us the knowledge of an individual possessing dewlap,
tail, hump, hoofs.and horns, or, in other words, the sound
 AGHOSA AND GHOSAVAT 237

which conventionally conveys a particular sense is properly

called a word. Thus one making a sound is spoken to and
spoken of as follows : make a sound, do not make any sound,
this boy is making a sound.’
A little further on Patafijali says :

Tasted fata: settonficaisa areata:

Req: |
‘Sound is that which is perceived
by the ear, which is
throughly grasped by the intellect, which is lit up by by letters
and which is (nothing but the vibration of) ether.’
‘““Dhvani’’ is taken possession of by rhetoricians and made
to denote suggested serise more charming than the literal
meaning (vacyatisayini vyangye dhvanih). Thus in the well-
known stanza :

Tanta faeqeqaa aeat Aleta: Hraarcarftr |

frngara frraraganifs Sergeants i
‘Having sold kairava-buds with her face unveiled, the
gardener’s daughter, desirous of selling full-blown lotuses,
covers her face with the border of her garment.’ The suggested
sense which is much more charming than the literal meaning
is that the girl’s face is the moon. The kairava buds open
out in ‘the presence of the moon, while the lotus closes its
petals. Hence she uncovers her face while selling the kairava-
buds but veils it at the time of selling the full-blown lotuses,
lest they become shrunk into buds. |
Phoneticians took charge of “ghosa” and “nada” and
used them in the sense of ‘‘voice”’, We tread in RP. that
 238 TECHNICAL TERMS OF SANSKRIT GRAMMAR,

“évasa” is the material of the surd sounds and “‘nada’’ of the

sonants (Svaso’ ghosandm. itaresim tu nadah xiii. 4-5).
Similarly we find in TP. ‘“‘naido ’nupradanam svara-ghosa-
vatsu’” ii. 8 and ‘“‘aghosesu évasah” ii. 10 ie. in vowels and
sonant consonants the after-effort is sound, in surd consonants
it is breath. Similarly we find: in CA.: Svaso* ’ghosesv-
anupradanah nado ghosavat-svaresu i. 12-13.
As Whitney says: The Pratisakhya here lays down with
entire correctness the distinction between the surd and’ sonant
sounds, which consists in the different nature of the material
furnished in the two classes to the mouth organs by the lungs
and throat; in the one class it is mere breath, simple uninto-
nated air: in the other class, it is breadth made sonant by the
vocal chords on its passage through the throat and thus con-
verted into sound (Atharva-Veda Pratisakhya, p. 17).
“Ghosa” in its technical’ sense occurs for the first time in
AA. -where it appears to have been used in two distinct senses.
In ii. 2. 2 after mentioning the etymologies of Sukta, Rk.,
Ardharca, Pada and Aksara, the Aranyaka says :

at ata: wat wa: ad jar: ad der aga enefa:

sro wey, srr ara zeta fee |
Keith rightly translates: “Therefore one should know
that all these verses, all these Vedas, all sounds are one word,
prana, and that prana is all verses.” There can hardly be any
doubt that “‘ghosa” has been used in the sense of ‘phoneme’
here. This would appear to be the earliest technical sense
——— —_~

*Cf. English surd, originally ‘deaf’, then mute, voiceless,

 AGHOSA AND GHOSAVAT 239

of the word. In the next passage in the same chapter “‘ghosa”’

is evidently used in the sense of vowel :
dat a¢ gedtaga’ araaq, ae anfa saga
ASSUIT ,AT ATT: ST AAT, F HCA: HT MNT: | -AyRWw
With this we may compare Chandogya-Upanisad ii. 22:

Ba AT ATA TSA EAT AMAT...Ty HSA AAT

abated fagar ameat...aa erat Batarafafatear amean: |
Curiously enough the sense of “ghosa’”” was next shifted
from ‘vowels’ to ‘voiced consonants’ as is clear from the
Pratisakhyas.
The ghosavat letters are vowels,.g, gh, a, ], jh, n, d, dh,
n, d, dh, n, b, bh, m, y, r, 1, v and h; the rest are aghosa.
Acc. to Indian Phoneticians, the current of air expelled
from the lungs by the will of the speaker coming to the
mouth organ becomes vasa or nada, acc. as the glottis is
open or narrowed; when it becomés Svasa it gives rise to
aghosa sounds and when it becomes nada
it gives rise to
ghosavat sounds. It will appear from this that évasa and
nada refer to the speaker and ghosa and aghosa to the hearer:
in other words, ghosa and aghosa are the acoustic effects of
Svasa and nada respectively.
RP. xiii. 1 says:

WY: I: Heer Rwer @ fat daa ar |

ATTTS area ATTAT AT TAEPTTA||
He. quotes from the Apiéali-Siks4 in his Brhad-vrtti on
bs dee FP:
 240 TECHNICAL TERMS OF SANSKRIT GRAMMAR

@ fata on ata arqedararad

aha sfreat faa:
qigafaera | aa arasfaeama afer fagaene
faaqrt:, daacate az: | aa aat atefad faad vafa,
ag aT Ae, AAA Fats: | ATasAAIgIAATaAAA | are
GAA | A*aeztayq agearat aerafaeiad | ad wat
TARA AAS “Aa aaisgasiaa, aa arrafa-daig
TN Sad, aT T savaisaastaa, aa saracafadant-
Taal SAA |
VP. uses “‘jit’” for “‘aghosa’” and “‘dhi” for “‘ghosavat’’.
Jit is used, probably because the letters stand at the begin-
ning of each varga and are consequently regarded as victo-
rious. In ‘dhi’ dh is taken as the type of the alghsas and a’
euphonic 7 is added for ease of utterance.
[Similarly in “‘dhut” (of K. and its followers) standing for
all the consonants except the nasals and the semivowels,
dh is taken as the type of aspirate voiced consonants. ‘Thus
“dhut’? stands for the aspirate voiced consonants, aspirate
breath consonants, unaspirate voiced consonants, unaspirate
breadth donsonants, sibilants and #. It is just possible that in
some form of the Sivastitras the aspirate voiced consonants
began with dha instead of jha. The commentators of K. make
no attempt to etymologize over this term, but the Haimapra-
kasa explains “‘dhut’’ thus :
aha Pats aedivt) geroarg:) aed afr gz
fra Wea: |
If we accept this etymology the Loc. sig. will be dhusi,
but the form met with in grammar is dhuti.]
 AGHOSA AND GHOSAVAT. 241

Ghosavat occurs in most of the Grhyasttras in the

technical sense in the nama-karana section, e.g. ghosavad
adyantar-antastham abhinistanantam dvyaksaram catura
ksaram va. Asvalayana Gr. Su. ft. 13. 5..
Panini never has occasion to use either “ghosavat” or
“‘aghosa.”” For the former he uses-the pratyahdra ‘has’ and
for the latter ‘khar’. C., J., Sak., Sarasvata, Vopadeva and
others follow Panini. HN. uses ‘“‘Gopala” for ‘“ghosavat”
and ““Yadava” for “‘aghosa’’. This is rather unfortunate since
both Gopala and Yddava begin with ‘“‘ghosavat’’ letters.
The author of HN. thought that the relations of Visnu were
either Gopalas or Yadavas, consequently he divided the letters
into Gopalas and Yadavas.
In the same way Sam. which uses the Pratyaihdras for the
vowels only takes k and g as the types of the surds and
sonants respectively and uses ‘‘kadi” for ‘“‘aghosa”’ and ‘‘gadi’’
for “‘ghosavat”. For “svara-ghosa” of K. it has ‘“‘aj-gadi’”
(i. 187).
The vowels are not generally included in the list of ghosavat
sounds in the Pratisakhyas. Thus RP. mentions “‘svara’”’ along
with “‘ghosavat”’ in:

aarrasfifha qaqa
faasiata ereativaat: | eE9
aaaied TATTLE T | 2123
RP., however, quotes a view according to which the sound
a constitutes the “‘ghosa” of the ghosavat sounds :
Awa NCTATATHTTAB | LAIR ( AIAaT oer raATe
SATA BTATAT: | )
OP, 129—16
 242 TECHNICAL TERMS OF SANSKRIT GRAMMAR

K. also uses “ghosavat” along with “‘svara’”’ in its rules:

TITTAT: TATA: ATTA TAAL | L212
ATTAATIT: | e1Blea
But it is clear from the Mahabhasya that both Katydyana
and Patafijali regarded the vowels as ghosavat. Thus under
““‘parah sannikarsah samhita” i. 4. 109 we find the seventh
Varttika :
araTfaca waar safeaacedfrar |
Patanjali explains:
srafaca waaay ater datsafeenarq <dfkar-
aa a oatta—apaz: face frafirhr |*
faasad dar xa) aa aaa:—aqadifa »
Th: TIC Sae Tess | az Te sat TEs
US A ARTY Sear | archers frat Stesaeang 1
STA |
Jf hradaviramah or continuity of ghosa be regarded as
the correct definition of Samhita then the name Samhita is
not applicable to cases of conjunction of mutes and aghosas,
e.g., kukkutah, pippaka, pittam, because there is the cessation
of ghosa. If it be objected. Why do you speak of conjunc-
tion? What about those cases where there is no conjunct but
only one sound as in pacati? Our reply will be, a single
sound is coloured by the ghosa of the preceding and
=>"iggusaieu SAIS n= euneemeee
mnerepeem e
*The correct reading would appear to. be gaz:
farqet faa Which,
as Kielhorn notes, is reading in Goldstiicker’s Ms. under I 26. 20
we find in the printed editions :gaged aap
feugargan fafya fafa |
aez: face: Galata |
|
 AGHOSA AND GHOSAVAT 243

following sounds. Thus a white cloth between two red cloths

is perceived as red, a polished mirror in a_ basket of plums
takes on the colour of the plums.’
Kaiyata explains :

SACHA TAA AH TATA BA |

‘Because the two a-sounds which are ghosavat the c-sound
appears as ghosavat.’
Again we find in the Bhasva on ‘“supam su-luk....”
vii. 1. 39:
qq ATAU | AZ AAA [++ AT anh qaaar-
aTMaq faafafs > a faeafa, aaa Geant: ereraiat
THT TAIT |
Patafijali says that as an instance of the elision of a case-
ending we may mention ‘ardre carman’. It cannot be argued
that ‘carman’ here can well be justified by regarding the
purva-savarna as the substitute for the case-ending, for since
the case-ending is i the savarna of n as the substitute for it
would be d. What Patafijali
means is that since the vowel
is alpa-prana (unaspirate) and ghosavat (voiced) the substitute
for it must also. be alpa-prana and ghosavat according to
the Paribhasaé: sthane ’ntaratamah i. 1. 50. From this it is
quite clear that Patafijali regards the vowel i as ghosavat.

WOT | ST ait a ceATaaTNy | 2122

arat: at at cwat fq) sear ees) Fe
fa Sta: | Q]'x0-43 |
aot) sen-firestate-saa-fiettiar asta: | 2122,
AGATA ATATL| e122
244 TECHNICAL TERMS OF SANSKRIT GRAMMAR

aqcafaa | ewaTasseqgaa: | al Weraq-

TACT | URR-2R
saga araq-aaaraer 2: 72 aa |
AVSHIA Wika AAMT BT Zeaz |) Da
[ast tarafa arraq-eatg 1) 3132J
Ma | SAAT AGT: | alee | 2k
mes | aT TAfEdiar faaaHTer saTaTsWaTAat
aM | ..gaiaaqul: dganwr aeTwgyaar aTry-
Feds | 221%
atl amet saafeedtar: acearayrat: | aaaeatsea |
21LIR 2-22, Wearaaaaraaigaraag! 2W12123
arZaael | at ah aU g aman MTA aatT? |
ANW WMiaedeg adisea thratfaan: 11
ay a Barer gt aon marafwal |
aart af 4 aeat aaa: daeifaat: ||
TAISARASTTUZAIATAAAIS
a Aa MNaT |
HEASITaAITIHAIAl afta aiteaqearan: eI: |
CSIR I-AA, US
aate-Rrat | arto seat at grarearores gatsan: |
freeing wards giot Fada: eran: i
@ 1 aeatgeia waa Aaa: | AEM ATA |e171e3-28
afaqart att 8am, wat age Hea |---enaT-
qaarareaeay , tat Aorafanredt Me: Aseacqsrercarheanr
aad fad: | oye) a faae dre: aetfrgizaz
argeSat
arSav
en a; | RATATAT|
 USMAN 245

S| saafsdiat ao: sveauaearrdan: eq) ase

TTS TWTA* AGT aAGaeatshy | eR’
& | eferar-eftan-afcfrarta SY MGS | Azar
AA] URY-BR
aafe azarat facet ander anf aaaae-
aa TaTST HUTA aTza xfa Hat ferafera: | Aer

OSMAN
“Wsman’”’ is found in AV. and the Brahmanas in the sense
of ‘heat, hot vapour, vapour.’ Then it came to signify ‘those
letters or sounds in the pronunciation of which there is the
emission of hot breath.’ Acc. to RP. the three sibilants §, s, s
along with h, anusvara, visarjaniya, jihvamiilliya and upadh-
maniya are ‘“tisman”. ‘TP. excludes anusvara and visarjaniya
from the list. VP. goes a step farther and excludes jihvamiliya
and upadhmaniya also. CA. mentions “iisman”’ sevetal
times, but it is not clear which particular sounds are meant.
Acc. to Whitney, visarjaniya, jihvamiiliya, upadhmaniya and
the spirants constitute the “asmans”. Grammarians accept the
view that found favour with the author of VP. and regard
S. §, s and h as “iisman” Later on the initial vowel was
shortened and ‘“‘iisman” became “‘usman”. The aspirates are
termed “‘sosman”. Later on they are called “hamapraina”’.

* PR. uses ‘ghosa’ in the sense of ‘ghosavat’ in “ghosat tu na i”

iv. 489,
246 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Panini uses “‘s4l’’ for “‘iisman’’ and “‘jhas” for the sonant
‘‘sosmans’”. The word “‘usman” is explained in several ways :

(9) SPAT AIy:, AGATA aM: FeatA | =AAT

ATT 212%
(2) FASAMAAIAMMTSAN FMLA A FT A AAA-
fidrnfeg, wera) afxanrco (FT Ue ) 1
(3) SHAAN Tea eelsaea | TUTAT | IA
aqnannfefa | sarc at ae acta a catsaaa: |) ott
RT. and ST. abbrenate “‘usma’’ to “sma”’.
¢ 33

As has already been pointed out the tismans originally

included the antahsthas. ‘“‘Antahstha” occurs only once
in AA,
K. calls the spirants “‘tisman”, but in the akhydata-sec-
tion Sit is used several times. The Vrtti points out ““isman”
name, whereas “Sit” and ‘‘sadi’’ are
is a self-explanatory
contracted terms, apparently without any meaning. But “sit
CT eae eB

is formed with the first letter of the series, viz., § with 7

(abbreviation of ‘iti?) for ease of utterance and a final f¢ to
round it off, ‘Sadi’? means the same thing and is found in
Sam. in “ha-varjam Sadi$ ca halisyate i. 206 etc. Sit is
found in the Phit Sutras in the sense of ‘pronoun’. Cf.
svanga-sitam adantanam PS. ii. 6. Here ‘sit’ evidently stands
for ‘sit’, formed with the initial s of of ‘“‘sarvanaman.’’ Hc.
uses ‘‘Sit”” in the sense of the spirants.
HN. calls these Harigotras, because Samba begins with §,
the first of the Usmans.
“The term usman, literally ‘heat, hot vapour steam’, is, in
 USMAN 247

the grammatical language applied to designate all those sounds

which are produced by a rush of unintonated breath through
an open position of the mouth organs, or whose utterance
has a certain similarly to the escape of steam through a pipe :
they are sibilants and aspirations or breathings. In the term
sosman, ‘aspirated mute,’ and in its correlative aniisman,
‘unaspirated mute’, usman is to be understood, not in thiis
specific sense, but in that of ‘rush of air, expulsion of uninto-
nated breath.”” Whitney on Atharva Veda Pratisakhya i. 10.

AT | SALAAM:
WaT MT a acerg
a aft 1 gigs
AAMT Sareea sfeacar: | Taz |
att | srerrearet: |fafa fife fata fefa 1 cie3-e0
The sibilants are known as “mut” in VP. from the m
of “Usman” with ‘ut’ instead of “‘it’’, because mm is a labial.
And because the ismans begin with §, they are known as
Sit or Sadi in K. etc.

@OT | GPISTATT | Ue
AeaqAEIVT: YT A AM: qza Sada water | wt HO
ga-qerfay afer: |
aittarfs-fireat | TA TAT: | eps
TP. uses “aniismavat” instead of “aniisman” in “vyut-
purva ananudatto’ niismavati”’ iii. 16,
“Sosmavan” occurs in RP. xi, 25, but the reading is
doubtful.
Aghosavat is used for aghosa in K. i. 5. 9. Some use
ghosa for ghosavat or ghosin. Cf. Upalekha Siitra (i. 7) :
248 TECHNICAL TERMS OF SANSKRIT GRAMMAR

sera-faettat aarorrgearna EHCIA ANTIK: |

ST) HN: W-G-A-AT | eR":
farfefe ara: | Bly
2) tux rewn-var fr) ii
af) fe cerita n-9-a-erai TEMA 21c3 TA: |
St fer-egenT: S-A-S-ET Ta Tae: | VIR
fre-car-ne fieisargienal, sada g ofiqeag-
trang | Taatitear) sadane Magafeaad—
warafadaariata | Raarftataat a—
sayger arate arracieeztiar |
afearat Wt get weg eagiaret: |
TH UIA | THT sat: | wETacafaeg: |
afaaren Set | AAU TS gscarer xfa isarcaea
eer: |
TSTHTAHT |
El) U-Ta-ar eranantA) wre
seary:fez: tera | aft: | attiaa afer eft.
?Wareea HeId-airaies
sad ofa aaa) Zar |
The earlier name for the aspirate was ‘“‘sosman”’.
Mahaprana would appear to have come into use later, the
expression occurring for the first time in the Apisali-Siksa.
| BA ATMA) 723
art | fedta-agah: area: | ike
ageatfaen | fédta-agqat! area: } ize
anfunfefiren | aefa ara wet: | eee aTaTaeT-
TTY: | ATIAYARATY
AT | ARTA
CaTSSAC | 123-272
 _ ANUNASIKA 249.

RAIA Thaw Fak

SAGA SIAISHY ATT: |
asrRer
sort ce fates
QE RAANITT Teta ||
Subhasitavali, 2485.
‘Your enemies, though bereft of all heat, deprived of all
prosperity, removed {from all contact (with allies) and humble
of heart, attain dignity when they crouch at your feet, even as
the syllables of a stanza, though there be no anusvara or the
like, no lengthening, no conjunct, though their vowels proper
be short, become long by position when they come at the end
of a verse (pada).’

ANUNASIKA

‘‘Anunasika” literally means ‘accompanied by a nose

sound, pronounced through the nose’ and, like anu-svara etc.,
appeats to be a learned scholarly word created by gram-
marians or phoneticians. The term “anundsika” is used in
the sense of (i) the nasal mutes and anusvara as also in that
of (ii) the nasalization of vowels or semi-vowels. RP.
uses
the technical term ‘“‘Rakta” (lit. ‘coloured’, being pronounced
through the nose) for “‘anunasika”. RT. also uses “Rakta’’*

* Iwalmaumaans cafagqraa 1 ant five ere

* Ranga is used in the Siksis in the sense of nasaliz
ation :
amt datea art aa valine |
US UAT: RAMEN: WT WT va Beay 4
Uifufafire |
250 TECHNICAL TERMS OF SANSKRIT GRAMMAR

in “a raktah” 114. J. uses the first nasal 7% as upalaksana for

all the five nasals. He. uses the word “anunasika”’ without
defining it. Sam. sometimes uses “‘varganta’’ as in “‘varge va
tadvargantah” i. 131 for the nasal.mutes and “‘sanunasika”’ in
the sense of ‘nasalized’. Vopadeva oases the pratyahara “nam”’
for the nasal mutes, but speaks of the two kinds of ‘‘yar’—
niranunasika” and ‘“‘sanundsika’’ i.e. non-nasal and nasal
semi-vowels. Su. uses “‘anundsika”’ in the sense of ‘nasal’ and
in the rule “‘ano’pragrhyasyanunasikah” i. 2. 47 the word is
used in the sense of ‘nasalized’. HN. calls these ‘“‘Harivenu”’
or the flute of Hari, because of the nasal sounds of flutes.
It is to be noted that Panini never uses the -Pratyahara
‘‘Aam’’ in the Astadhyayi, but has always recourse to the self-
explanatory term ‘“‘anundsika.”” We are understand his using
‘““anunasika”’ in a rule like “upadese’j) anunasika it’ i, 3. 2
or ‘‘ano’nundsikasS chandasi” vi. 1. 126 in the sense of
‘nasalized’, but “‘fiam’’ could casily have been used in ‘‘cchvoh
fad anunasike ca” vi. 4. 19, “‘anudattopadega....” vi. 4. 37,
‘“‘anunasikasya kvi-jhaloh kniti” vi. 4. 15 etc. ‘‘fiam’’, however,
is found in the Undadi-sutra “‘namantad dah” i. 111. It may
be noted in this connexion that the Agnipurana does not
mention “‘fiam’”’ in its list of pratyaharas.
ee ee

aufear Mery: gawats amt sfa aaa ae |

AU BHM ASq RMN WA HU: WY aya wT: |
anfefareat-ztaat
wey Aarafwalfa fe Aaaraq |
aaa afaaqay gm: asta wreagay aaa: 1
fHAIT-GIS: |
 ANUNASIKA 251

In VP. “‘uttama’’ is used in the sense of ‘anunasika’ ¢.g.

“uttamantam uttamena’’ vii. 11. ‘“‘Uttama” literally means
‘highest’, ‘topmost? and so means ‘the last of the series’.
Panini uses ‘“‘uttama” in the sense of ‘last’ in his rule “utta-
maikabhyafi ca” v. 4. 90 and “‘upottama’’* in the sense of
‘last but one’ in “anifior anarsayor guripottamayoh syan
gotre” iv. 1. 738, “yopadhad gurtipottamad vu” v. I. 132
etc. Hc. cannot use “‘uttama’’ for the nasal as the term is
reserved for the first person. Dandin in his Kavyadarsa uses
‘“‘nasikya”’? for nasal.*
“Nasikya” appears to occur for the first time in CU. i.
2. 2 “te ha ndsikyam pranam udgitham updsificakrire’’ where
‘“nasikya’’ evidently means ‘being in, or coming from, the
nose’. So it must have meant first any nasal sound. Pro-
bably it is used in this sense in the Mahabhasya (nasikyo
varnah vi. 1. 63). Next it came to be used in opposition to
‘“‘anunasika”” and was restricted to the nasal glide or the
transition sound between A and a nasal (represented by Aum
because of its connexion with fA and nasal), the Yamas and
Anusvara. It is used in this sense in:
ST ArSasTaay AThaseara | WAT evo
aifararat atferat | aacearfat ais
atfasa athena
aaraeart Sea |
aa eataarante
wie aa A war: || stat
*RP. uses “uttama” and “upottama” in exactly the same senses.
* alfeaaen ofcaaqauta sian |
wfa aifaq gt gerezanler say tes
(AMAT! TEGECTE
GT )
 252 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Next it was further restricted to the nasal glide from hf to

a nasal and it is this sense that the word is generally used
in the Pratisakhyas.

apaeTaATgAeAnrsy | WAT Vee

AATAAAHT APAPISAT: | ATIT zico ( atfaat-
gaa am: | Sengsiageare TAGeA | 21¢2-¢3, agafa-
Raa: | eie%) |
“What the sound may be which is thus taught to form
the step of transition from the aspiration to a following nasal,
it is hard to say with confidence. I can only conjecture it
to be a brief expulsion of .surd breath through the nose,
as continuation of the /#, before the expulsion of the sonant
breath which constitutes the nasal. The pure aspiration h
is a corresponding surd to all the sonant vowels, semi-vowels,
and nasals of the alphabet: that is to say, it is produced
by an expulsion of breath through the mouth organs in any
of the positions in which those letters are uttered; it has no
distinctive position of its own, but is determined in its mode
of pronunciation by the letter with which it is most nearly
connected. Thus the h’s of ha, of hi, of hu, and those heard
before the semi-vowels w and y in the English words when
and Aue, for instance, are all different in position, correspond-
ing in each case with the following vowel or.semi-vowel. H
is usually initial in a word or syllable, and is governed by the
letter which succeeds, and not by that which precedes it:
but where it occurs before another consonant in the middle
of a word—which is always its position in the Vedas before
a nasal—the question may arise whether it shall adopt the
 ANUNASIKA 253

mode of utterance of the letter before or after it: whether in

brahma, for example, we divide,brahema, and pronounce the
h in the position of the a, or brachma,and in the position of
the m, through the nose. According to the Hindu method
of syllabication, the former is the proper division, and the
Hindu phonetists doubtless, regarded the h as belonging with
and uttered like the a; and noticing at the same time the
utterance, scarcely to be avoided, of at least a part of the
A in the position of the m, they took account of it as a
separate element, and called it nasikya.” Whitney under CA.
i. 100.
OT | agarfe eige,
rn TH-e
ts atsa
eaafase:
| elas
atsot, qaatfaaracaisanfas:; ense
G0T | sta haat
RaTetgatheae
rat: | RgZ0
agceartiar | Saar agarfeet: 1 ce
Mat) & waaarfan:: 212
Tl qe-afan-aqatgatas: 1) etree
arttrafe fae | S-A-0-S-at Gena: atfeenrenara
R12’
BU Agarkeat SA T-A-AT | eres
ag Tare wear scr can Kergarfaa: |
ag vad fearin a agafeer
g : | sae say
a m— Aaa aftersft| ster) ad
Faery Searcy waranty) Tat |
SH oarfaaat sy tens
atfaaral wat aot weet wate) s-a-s-oran:|
STS: |
 254 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Qi avearsaarerar 1 eee
| GAT ATA S-T-OT-A-AT | IR
Zl B-o-a-aT: Ehranras | eR

AYOGAVAHA.
The word “Ayogavaha’” does not occur in the earliest
Pratisakhyas. It is not found in Panini, but occurs in several
later systems of grammar. The term is explained to mean
‘“‘sounds) which occur (in the actua! language) without being
given (by grammarians) together with (the other letters of
the alphabet), a term for Anusvara, Visarjaniya, Upa-
dhmaniya, Jihvamuliya and the Yamas.” (M.W.).
“Yoga” is used in grammatical literature in the sense
of ‘rule. Cf. “Yoga-vibhaga”’, ‘tna hyekam udaharanam
yogarambham prayojayati” etc., etc. The ‘“Ayogavahas” are
not read in the Sivasitras nor in the Varna-samamnaya of other
schools, hence the name. Uvata appears to think that
these can only be pronounced with the help of ‘“‘a’’ etc., hence
they are called “‘a-yogavaha”’ :
ware Waa ateat:
laa at gefa ana
wnt staafd SATA: | ATT | ete ATER |
The sense of the word was forgotton in later times and
so ““Yogavaha’”’ came to be used instead. Even in VP. viii.
18 the commentator Ananta Bhatta reads “‘yoga-vaha”’ instead
of “‘ayoga-vaha.”’
 AYOGAVAHA 255

The Mahabhasya (ed. Kielhorn, vol. I, p. 28) explains

the term thus :.

wasanrarer a afagafread, et Al sat Rraty

STN: Heise | F GATT TaEr: 9 frastata-frergeiare-
araTyeaTcaraaT: | RAGATANTSTET: 2 TTA
qefa, waafzera wat) (ager cers eanwTaq—
aqifgerefa r | sant: senerceaita, qenazearay
rae: | saga fefra:—are:, semen seat ay
a fefraisctet areiert:) age Bad) qatsageer-
Saisaat ger) afar aaet a qeat, saz
TaAMITaATAa He: Beal gebta ont frateaediteqata-
aTeENeret apqrhag frat | sear: ) |
Katyayana uses Ayogavaha in the Varttika-
AATTATETATALT TT | fo Go we
Bhattoji says in his Sabda-Kaustabha :

STATA: GT: TRITEITT Aeaa: Hat A RANTT:, TAT:

face softs waeiacare TATEITAPIOITEAT «LRT |
areata faateafta satarfef aren: ) eetarea
@ SET
afar aia: | oeafted see cried a aft STATOT
Keay: |
The Katantra does not use the word “ayogava
ha,” but its
commentator Durga says under i, 1. 19.:

AMARA naiica isaac frasittaaedetcan-

rary mat ae erTearararc hire framerate |
Cat Saat CE Arran qarsazey | TENO- arya: Farry:
256 TECHNICAL TERMS OF SANSKRIT GRAMMAR

efa | ad GATUTaTET: 2 AGHA FEeAAT-

_epprrarer
fora Fad |
The commentary known. as Kaviraja says:

mugs gat: arag seediera: | sitafaed

RqaTTATsaag watacant aeedifa aerarer: | He-
ara dara BTA araed TErctifer |
The Vrtti-traya-varttika, a work of the Katantra ‘school
existing only in MS., says:

quaa ax HTT aaa cad |

CAPAC Hat x HO A AAT
arraret fafafeer sa ca aeiifeh: |
qat Bet A AMAT AA- N
In the Paniniya Siksa these are known .as “Ayogavaha” :

samy sem a fangetaaheas |

sarrarer faxat serrearerarhere: ||
(a fara arr: darn anicater Aut a saan: |
eA GF AAAI Ast | AMT AdAraTaEWes:
qeareafuarara sfemeqisaanniag afeaca: } ofan 1)
In the Varna-pradipa-Siksi AmareSa calls them
“samyogavaha” :
maqearcy feerrar afer TET |
faarqeagrean @ att A: TPM: |
darraie: waa freac-frafsar |
QAeny IRAs ST TT Teg |
 AYOGAVAHA 257

Sripatidatta says in his Katantra-Parisista:

sai fe saqaat ateaea eater oar atrareray |

XT Tomas saqaderafifiaa: | Faaraaa-
ACNE ATeaATaT TEratfa AVTATET AANTTA: | LIZe
The commentator explains:

ai ageaifa savaq) ann fe faaaair raqeaa|

aaa ry
aaearey frases staat yaar |
fRargaiarreray TTA
af | soareficae canto denaa watcagqaaiet ares
aafa, Qa wqcard MU AQAA cagqaieaaaed, waa
Saratezrs ara | wearercatfea: ga: sare.
BAHAY A BTAMAPA Awl adedifa seairaret Tar-
acai fet |
Neither ‘“‘yogavaha”’ nor “ayogavaha” occurs in RP. or
TP. In VP. we find the “ayogavahas”’ clearly enumerated at
the end:

AMAA X FH fa frerger: 149 LATA: |

aw garam, a; tf feastttr) eg rf area: |
% @ a ofa war: | <iRe
Under ‘urahkena brmhanam trmhanam iti _ svara-
dharmatvad visarganusvarayoh” Samksiptasdravrtti i. 298,
Goyicandra quotes:
framiaearc gq areeMANAAATISAT |TTATET: |
eqUTAl: |
O.P. 129—17
 258° TECHNICAL TERMS OF SANSKRIT GRAMMAR

The introduction to the Rktantra enumerates the vowels

and consonants and then says:
afa aware: | aartrrane: | a: aa freasteta:|
a tfa fagrgeta: | —eeqaenrtia: | g° ofa arf: |
aq qa, & ta @ sfa a efa g cfa aan.
MTG a’ at cergqeard |

ANUSVARA.

Panini uses “‘Anusyéra” without defining or describing

it. The word literally means ‘after-sound’, and appears to
have been specially coined by grammarians for their own
purposes. Lexicons do not know any use of the word anterior
to the Srauta Sitras.* The Atharva-veda Pratisakhyas do -
not mention ‘‘anusvara”’, though CA. distinctly describes the
same phenomenon which is specified by other grammarians
with the name of ‘‘Anusvara” (Varma) and it is included in
the term “Anu-nasika.” The other Pratisakhyas and the
later systems of grammar generally use the word. Mu., with
its craze for monosyllabic technical terms, calls it by its
second syllable ‘‘nu’’, and HN. calls it “Visnucakra”’ from
its appearance. Some grammarians call it “Bindu’’
also.
(mo bindur avasine va iti Govinda Bhattah—R4ma
Tarkavagisa’s commentary on Mu.: ‘no’ntaé cachor bindu-

sara: | Vy aT AMAR) «weeny af aTaqafana) tay-

QT | GVTMIT UwRME|-wc
Anu-svr is used in JUB. iv. 8. 3. 2:
Peggy THAN aaARTA: |
 ANUSVARA 259

purvam sam tathoh sam tathostu sam’”’ PR. i. 99). It appears

from HN. that it is also called ‘“‘Lava’”’ (lit. ‘a point’, Cf. Gk.
iota) in some grammars. J. uses “fan” for ‘anusvara.’ ‘Thus
J. has “‘nugmo’n” y. 4. 11, for Panini’s “‘mo’nusvaralk ”” vill.
2
VP be|

3,23. Su. also represents anusvara by “an” in i. 1. 14; later

on it uses the word ‘‘Anusvara’’.
Now as regards the nature of “‘Anusvara,’”’ three distinct
views have been held by phoneticians :—
(i) “Anusvara’’ is merely a nasalized vowel. ‘This is
the view of CA. which, however, does not use the term
anusvara but speaks of Anundsika or Nasikya. “The
Pratisakhya of the AV. recognizes this nasal vowel, called
_anunasika, ‘accompanied by a nasal’, alone, ignoring Anu-
svara, which with the preceding vowel, represents two dis-
tinct successive sounds.” Macdonell, Vedic Grammar p. 53.
Thus according to “‘nakarasya ca” CA. ii. 34 in the interior
of a word n is elided before spirants, and according to
“nakara-makaryor lope purvasyanunasikah” i. 67, the preced-
ing vowel is nasalized. So for “‘saramsi’?
will- have CA.
“‘sarasi’’, TP. identifies anusvara with anundsika in v. 31, x.
lis xv. 1, xxn. 14.
(ii) According to the second view represented by RP.,
anusvara is either a vowel or a consonant (anusvaro vyafijanam
va svaro va RP. i. 5). Similarly the commentators of Panini
hold that Anusvara, Visarjaniya, Jihvamuliya upadhmaniya and
Yamas are to be regerded as having been read after ‘‘a” and
also included in ‘‘Sar” (anusvara-visarga-jihvamiliyopadhma-
nlyayamanam akaropari Sarsu ca pathasyopasamkhyatatvena)
and consequently they are both vowels and consonants. When
260 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Uvata says in his commentary on RP.: ‘tasmat svara-

vyafijana-vyatiriktam anyad varnantaram etad ity etat-
khyapana-param evaitat sitram,’’ he merely means that just
as Nrsimha is neither man nor lion, but partakes of the nature
of both, so anusvara is neither vowel nor consonant but par-
takes of the nature of both.
(iii) The third view which is expounded by the Vaidika-
bharana. Yajusabhiisana etc. holds that the anusvara is
merely a consonant and consists of half ‘‘g’. Thus the
Vaidikabharana says under TP. ii. 30:
ASA aga aT tact ar [ war eis ] zfa qeaay |
afatiaragerd | AGA SyaAaS eaAqaRaA ATE aT,
TATAITS TAT |
The Parisiksa says:

AACTGANT LENT Ta AIT ATA HTT: |

The Yajusabhusana says:
APITS WITT ACATATAA LATTATT: |
(Quoted by Varma, p. 151).
For greater details and bibliography, see Varma, Phonetic
Observations of Indian Grammarians, Chap. IX and Surya
Kanta, Rktantra notes pp. 4-5.
“That the anusvara in OFA. was not a mere nasaliza-
tion, but a nasal continuation of, or a nasal’ glide following
a pure vowel is borne out by MIA. and also by the tradi-
tional pronunciations of Sanskrit in different parts of India,
which has altered of OIA. to a definite nasal [n, n. w]”.
S. K. Chatterji, Origin and Development of the Bengali
Language, p. 244.
 ANUSVARA 261

The modern Bengali pronunciation of anusvara is merely

a development of the pronunciation described in (iii)—a nasal
g’ naturally becomes n. Cf. Mahabhasya :
£¢. 9?

IT Tala WAT aaah) argarfersei

ATTA ART WT: | VIS
LT] AGAR saga ar eat ar) gis, |ea
SAA AIAUSA | Beat sagqaregqacey | WEwaearc-
Prestateat ) 1ia2-Re, ( frasistargendd ada qinercy|
qclae ) aUgearcraoegs Feary | 23122
RATATAT AATATATEALENTATATETS |
agent aadafas a seated Ags agg) 7318-33
aca AAAI: | satfeaifasanaarfas ary
RAIRE-BO
AT | AAATT AAT AGATA: | 2Ig0, ataacaTG-
aifaeraganrmarata deans: | ad adafr alect-
Ga: | waearesfrafe arta | g912-3
BIST) HM RAFAT: | IRL, AACATTT aeagai sens.
Aral Tat areaTafs | Blgse
Mat | staged i a at reread |
—

* Here the dual stands for short and long “anusvara”,

There is a good deal of difference of opinion as to the quantity
of the anusvara. Acc. to TP. i. 34 it has the quantity of a short
vowel. Acc. to VP. iv. 150-151, an anusvara preceded by a short
vowel consists
of a mora and a half, when preceded by a long
vowel, it consists of half a mora. Regarding the anusyara before 4.
 262 TECHNICAL TERMS OF SANSKRIT GRAMMAR

BT] AH ATTRA | RRS

aHIe Taraicare: xe farsaiat atigent-ag wala |
afa:; agaqa dad wat sagen: |) getat!
aaaeg: azae | Taada dhafedear nq aut saws |
BACs: |
2) a a: agearefreatt 1 eee
aarcaacora | a’ afa area an: | ata: |
Z) ear gat 2
aa sered:, farcfaftegaral soll sateafredst
a: | SFA:|
q| wn faercqencfaaag amarmaqt 71k
Wy TA aWieacacat waa sarasarfafa-----
ager: | samatant,; oH fa faeghaang,
aa cat
feegugeart: enfefa Shaq) aa, cHTer aaeaTe: |
frag wat ger: sfeat aot face: faeces, ager:
agenda: aq gaa: |... fesaannmfataay a-
arqafrcearmayred|
zi weafamaaq eee
wat wanoe:, fargeqadl ait FesoreqeReTAT|
agra faresaa |
S$, Ss, h and r the Pratidkhya-Pradipa-Siksa quotes from Katyayana
Siksa :
VAAN ax Were qa: |
WAay Waa 4fanaa
Saree StaleMW At | ata) Ta! |
BACT GU Veleratsi mage |
 VISARJANIYA 263

AGA AMA (7) ATA AS AAT: |

aft ag’ wag(2) avast Gea: ATT II
AISA STH
VISARJANTYA.
‘““Visarjaniya’’ means literally ‘concerned with the letting
out (of the breath)’ and is, like anusvara etc., confined to
grammatical literature. The earlier name seems to have been
‘‘abhinistana’’ as would appear from “‘visarjaniyo’bhinistanah”’.
CA. 1. 42 and from Aévalayana Grhya Siitra i. 13. .5:
‘‘Nama dadyur ghosa-vad adyantar-antastham abhinistanantam
dvyaksaram catur-aksaram va’. Panini says: “‘abhi-
nisastanah Sabda-samjfiyam”’ viii. 3. 86 and his com-
mentators explain ‘‘abhinistana” as ‘“‘visarjaniya’. Normally
‘visarjaniya’ should have been found in the Pratisakhyas and
Katantra and ‘visarga’ in P&nini. This is, however, not the
case. “Visarga’’ is a very late word and does not occur in the
Pratisakhyas, Panini and Katantra. Hemacandra appears to
be the first to use “‘Visarga,” and Kramadiévara and others
follow him. Padmanabha uses both “visarjaniya” and
“visarga.”” Thus we find “‘rusoh padante visarjaniyah”’ i. 3. 1
where the Vrtti has “visarga’’—padante ro rephasya sakarasya
ca visurgo bhavati” and “ickuro num-visarga-vyavaye’pi” i.
4. |. J. simplifies matters by using the symbol “ah”? for
“visarjaniya”. Thus for “khar-avasinayor visarjaniyah”
vii. 3. 15, J. has “‘ante"h” v. 4. 21. Mu. uses the first syllable
“vi” for “visarjaniya” and H.N. calls it “Visnusarga” (adding
“snu’” after the first syllable of ‘‘visarga’’), just as it uses
‘Visnubhakti” for ‘‘vibhakti’’.
264 TECHNICAL TERMS OF SANSKRIT GRAMMAR

In ‘‘visarjaniya” it was thought that the suffix -aniyar had

been added in the karmavacya, so “visarga’’ was formed with
ghan in the karmavacya and two syllable were saved. K. uses
‘“‘visrsta’’ for ‘visarga’ in “‘ra-sakarayor visrstah”’ iv. 4. 2
for the same reason. The corresponding rule in the nama-
section is “repha-sor visarjaniyah”’ ii. 3. 63.
“It is called visarjaniya either from its liability to be
‘rejected’, or from its being pronounced with full ‘emission’
of breath, or from its usually appearing at the ‘end’ of a word
or sentence’, says Monier Williams in his Dictionary. Of
these the second explanation appears to be the correct one.
“There are three voiceless spirants appearing only as
finals. The usual one, called Visarjaniya in the Pratisakhyas,
is accoding to the Taittiriya Pratisakhya articulated in the
same place as the end of the preceding vowels [ purvanta-
sasthano visarjaniyah TP. ii. 48]. Its place may be
taken by Jihvamuliya before the voiceless initial gutturals
k, kh; and by Upadhmaniya before the voiceless initial
labials p, ph ‘These two are regarded by the RV. Pratiéa-
khya as forming the second half of the voiceless aspirates
kh and ph respectively [“sosmatam ca sosmanim ismanahuh
sasthanena”’” RP. xu. 16], (just as h forms the second half
of gh, bh etc.) [“‘ghosindm ghosinaiva” RP. xiii. 17].
They are therefore the guttural spirant (Greek) x and the
bilabial spirant £ respectively.” Macdonell, Vedic Grammar
for Students, p. 18.
In Sanskrit works Visarjaniya is described thus:
AFKIE WAATAN Hara: CABIAA |
aaa eager a rad eft tI: ||
 VISARJANIYA 265

arat, a: af faster: 1 iee

Wa) 3 efa frase) aie
el) a: tfa faasaia: 1 eee
ant xetaremg: | 3 eft HMderagarafadit
frastaiadat wate) afta: |
ei a a agen-faat) eee
frasad faces aff faa: | adacatasert aaq |
aa feast faastata cafe aeret eee) |SEITE: |
Jl AM saat ec
T) grataedt feegaa fear fa mah) cee
wade fraidiat fae mai, sf:
ani“ faasra facaasafa, sqaraa aarcerfarhs
Rafer & fae) maenrfien|
=| a afa favre | ies
facggararr aot fagyestrat) fradt fastest:
faaetsfrfrerca 1 aha: 1 et aft: | fem: at fasyest
afa faayerraczeared: | amarante TNS
3 -Tae Halas
sararcat faageer stfehe: | Star |
According to many commentators Sriharsa refers to the
Visarjaniya in the following stanza of the Naisadha (x. 87) :;
faceat ataarafeean:
HIAASUATAT |
Hea wetat feraar: afsean:
aa Ta farce az: 1
‘On whose body was formed the pair of ear-rings with
the best portion of the circular terminal script (ie. the
266 TECHNICAL TERMS OF SANSKRIT GRAMMAR

visarjaniya), the fingers of whose hands with the best of

golden pens, the tresses of whose hair with the essence of
ink and the sheen of whose smile with the best of chalks.’
Anusvara and visarjaniya are referred to in the following
well-known stanza :
faregraatigamacth:
nat farcitarsa
maararaerad
farsa & Asa aati aA: |

Wea: marta frgoaheca aaah |

‘I bow to those who construe sentences by regarding that
which has its forward path undulated by a couple of dots as
the subject and that which has a dot on its head as the
object. Bilhana Kavi arms himself against those alone who
compose poems with flow of diction from which all alloy has
been removed by rubbing in the touchstone of thousands of
books.’

JIHVAMULIYA.

“Jihvamiliya” literally means ‘belonging or pertaining

to i.e. formed at or uttered from the root of the tongue’.
The word ‘‘Jihvamiliya” occurs as early as the Atharva Veda
(i. 34. 2), but as a technical term “Jihvamuliya” ts found in
grammatical literature only. Panini makes special provision
for the derivation of the word in his rule “Jihvamulangules
chah"” iv. 3. 62, ‘the suffix cha (iya) is added to jihvaunula
and anguli in the sense of ‘tatra bhava’. The word in ils
narrower sense of the peculiar pronunciation of the visarjaniya
before k and kh is not found in RP., but the sonants 7, / and
 JIHVAMULIA 267

the sixth of the Usman letters or spirants as well as the first

varga are described there as Jihvamilliya, i.e. formed at the
root of the tongue: “rkira-lkarév atha sastha iisma
jhvamiliyah prathamas ca vargah” i. 41. The commentary
on CA. i. 20 “jihvamiliyanim hanumilam”’ also mentions the
same sounds as Jihvamiliya. TP. mentions “‘jihvamilliya” in
its specialised sense in the rule:

a Rrastattr-frengenatrearharaearca
Raza
WIR
(fared frees wearer | senaia: saenta
aang | afreanrcer| )
VP.
frames the rule “r-kau jihvamiile” i. 65 excluding
1 from the list and states the sthana to be Jihvamila or the
base
of the tongue. In i. 83 ‘“jihvamiliyanusvara hanumilena”
the karana is stated to be the base of the jaws. TP. reverse
s
the sthana and karana and says: ‘Hanu-mile jihvamiilena
Ka-varge sparsayati’ ii. 35 and ““sparsasthanestismana
anupUrvyena”’ ii. 44. Whitney points out that CA.
nowhere
mentions J ihvamiliya and Upadhmaniya, but they are implied
in the rule “‘visarjaniyasya para-sasthano’ghose”’ ii, 40.
According to RP. ii. 33 Visarjaniya Optionally remain
s
unaltered before guttural and labial sounds- ‘“prat
hamot-
tama-vargiye sparse va”. From TP. ix. 4 we learn that
Agnivesya and Valmiki did not accept the change of visar-
janiya into jihvamiiliya.
Vopadeva calls the jihvamiliya “mi” from its middle
syllable.
268 TECHNICAL TERMS OF SANSKRIT GRAMMAR.

UPADHMANTIYA.
‘“Upadhmaniya” appears to be formed from “upadhmana,”
‘breathing or blowing upon,’ and literally means ‘on-breathing’,
and is, like similar other words, confined to grammatical
literature. It is the bilabial spirant produced by the sound
of the Visarjaniya before p or ph. The word does not occur
in RP. or the Pratisakhyas belonging to the Atharva Veda.
This is first found in TP. and VP. The Katantra school also
uses the term, and so does RT. Panini does not use “‘jihva-
miliya’”’, or “upadhmaniya’’, but uses the symbol — “‘ardha-
visarga” in his rule; kupvo~k~ pau ca” vill. 3. 3. 7,
though he uses “‘anusvara’’ and “‘visarjaniya”’ each time. C.,
J. and Sak. follow Panini,, Hc. and PR.: follow K. Krama-
disvara, as usual, uses “jihvamiliya” and “upadhmaniya”’
without defining them in his rules: “‘jihvamiliyopadhmaniyau”’
i. 217 and “pa-phayor upadhmaniyah” i. 378, and the commen-
tator says under ka-khayor jihvamiliyo va visargah ii. 376 :
aqanigafandaas ferquiarraiiadantt are
afagat Aasat |
Vopadeva calls “upadhmaniya” “‘ni’’ from the pen-
ultimate syllable, and describes, after Hc., Jihvamiiliya as
“‘vajrakrti and
Upadhmaniya as “gaja-kumbhakrti”. Su,
evidently reads
“‘upadhmaniya” for ‘“‘upadhmaniya” in its
Vrtti on i. 1. 15. The author himself explains the word thus
in the Panijika :

STAR ATA RT! SUT wa: |

Tartzara [“aiee[ Sa: | Saar fa ance afz
TT VATATATAMATA; |
 _

UPADH MANIYA 269

AT ST | — o SeaeM aA | <}Ro
atl x Hata faerie: | oa seqaearaia:|
URIASs-e¢
PHIT Tet: | x ata agafasul faaetada
wate | watt gears: | Safa aseearefadot sa
variaeat vafa | afa:) soagatq cat aeeridarray:
fa ardt: waradianaa:—sq amie eat aaa efa|
Tat) aafe aensft so ame ened aur afameeaa
sdifatcfa | adie fa qnarfifa fae: ) afc: |
21 eux Acer ae Mr, epee
agarc faa agreferragearafers aor rang faz
Sa Aafa | ARTCATRTL-THT TIOTA: | TEAST
aoteagfeaarety x H TaTaaaAT | aegfa: | farz-az-
Ta GRA x AL wittanrefefraest
euMage «zurante—anrafafefa | asaioaiier a
aat, wapraanfcarfaaer atsfe aur, Rarcqard
aTaqg: Glaengaid | Waa qwaragda) wadt a
aca, afy qaarcarageacapate | THeMAY RIH-
alarmda aa. saimecieraq, sa aq aafy
araTaaaceaseaeanins fafsaeaata Fm: wa: |
AT RAAT ATA, ATAAPATH Tawar | weary, Tae
AA Ata tameng aieafafs:) a aq qoigada
aTreahe araqamigdhaq, aaa) az ant
*In “uvopopadhma- osthe” i. 70 “upadhma” is used for
“upadhmaniya”.
270 TECHNICAL TERMS OF SANSKRIT GRAMMAR

aiaenaa ufagaefa, afasagra) wakarat agaaca

quad, Aenea A ATAHATAAT {ale —agaaa-
fafa ag xa craetardatt ganna, aq wet a:
ae a facra? seaa) tnerfaca ceagqadadifa a
facie: |
q | (faetsarfeanarig ngea
cag farran : 1]
|21%23
sguarafaasiefaaa frat 1 afa: 1 |faetsareantar
TAMATATH: |. Ae] S AMAKAT WIAA AT |
AS
ql xh = qi qat} e
fagiges variasel fargeia: | TTenataatENncT-
GARI: | AVATAAST MIL |
“sat qin araat geal gq Torfrat |
ACMUSAMALAT: TAHATTAT AAT: |
aa: aa fauisaead Tae Waa |)”
EMT |
‘The lip is the organ of pronunciation of the Upadhma-
niya, and the pronunciation is like the breath of the serpent.
Anusvara and Visarjaniya are connected with the preceding
letter and Jihvamtliya and Upadhmaniya with the follow-
ing. These four “‘ayogavahas” are regarded as vowels in the
matter of the mutation of n into n. The vowels shine by
themselves, while the consonants cling to others for support.’

ANTAHSTHA.

‘“‘Antahstha”” was
originally an 4a-stem evidently formed
with the agent suffix vic from the root stha preceded by antar.
 ANTAHSTHA. 271
The word means ‘standing in between, occupying an interme
-
diate position’. In grammatical literature the semi-v
owels
y, r, / and v are called ‘“antahstha’’, because they
stand
between the mutes and the spirants, or (according to
Western
scholars), because they stand midway between the
vowels and
consonants. The word was originally either masc.
or fem.,
like viSvapa, sankhadhma etc. in the later
language. Thus
we find in SB. i. 4. 3. 8:
Ta MAR: oT aaa aes | AT SoreaeeT
AMAL, AAT A TFT ATO, eAISHASATY: | HaCaT =
vata, aaeMad Wert, T CAAT eA Az
In the above extract the fem. sa appears
to stand for
the fem. samidheni (saisa “Agnim. ditam vrnimahe”’ iti
samidheni—S.), so it is not possible to
conclude that since
“antahstha” is fem., sa and esd are fem.
(vidheyapradhanyat
strilinga-nirdegah, as in Kalidasa’s well-known
line “‘Saityam
hi yat sa prakrtir jalasya’’). Similarly “ya evam etiam
antastham pranandm veda’’ is inconclusive, because here also
“etam antastham prananam” refers to
‘‘samidheni’”’ which is
fem. On the other hand “antastha”
is evidently masculine
in the expression. ‘‘antastham enam manya
nte’”’.
The word occurs for the first time in its
grammatical sense
in AA. iii. 2. 1:
AARATAA: ITT SABI, sreeltr CTIA, ASIA:
Sta, ae wtfeay wtaqearagdaaeneny cht = ane
SEA ATTSRT: |
It is not clear from this passage if the
word is masc. or
fem, Since, however, most words endi
ng in 4 are feminine,
OT2; TECHNICAL TERMS OF SANSKRIT GRAMMAR

it soon came to be regarded as belonging to the feminine

gender and in the Pratisakhyas and K. and its followers the

word is uniformly fem.*
In ASS. i. 2. 17 we find: antasthasu tam tam anu-
nasikam.
After mentioning the mutes RP. says: catasro ‘ntasthastatah
i. 12 and follows it up with: uttare’ stav Usmanah. Hence
the commentator Uvata says: sparsosmanam antar madhye
tisthantity antahsthah.
In TP. catasro’ntahsthah” i. 8 comes between
“paras
“Adya paficavimSatih sparsah” i. 7 and ‘“‘pare sad iismanah”’.
The commentary WVaidikabharana says: jihva-madhya-

prabhrtinam karananam antair janyatvad ya-ra-va antastha

ity akhyayante. The commentator evidently reads ‘antastha’

for ‘antahstha’.
VP. viii. 14-15 and RT. (introduction) have: athanta-

sthah. Yiti riti liti viti, RT. uses “stha” for ‘‘antastha”’
hayam
in the rule “‘ranam api sthayam” 181 (karanam api antast
apad yate—Vriti).
Panini and his followers use the Pratyahara “yan” for
Patafijali, however, is not averse to the use of
‘ontastha”.
the word.

ag feagaaarmngqes: fra | fit: a4

ee CAAA HLT | LITITACACAMATT | FIER
sofa fagiat aa aifaws nyqsaa |
dar-faaaa aaraaa Sia Gay 1
amifaatfad Pie
 UPADH MANIYA 273

The tradition of the Pratisakhyas had continued upto Hc.

The visarga of ‘‘antastha” was dropped very early
according to RP. iv. 36 “tsmanyaghosodaye lupyate pare
nate’pi” (corresponding to the Véarttika ‘va éar-prakarane
Khar-pare lopah” viii. 3. 36. 1), even as the visarga of
““nihstha’’ was dropped. The gave rise to the erroneous idea
in the minds of later commentators that the word was com-
posed of ‘anta’ and ‘stha’.
HN. calls these Harimitra. The commentary does not
explain why they are called Harimitra. The fact is that the
word Yadava begins with the semi-vowel y and the Yadavas
are the relations (“‘mitras’’) of Hari, hence the semi-vowels are
designated Harimitra.

AT | AeA B-T-B-AT | dl aiee

Te ET tareated fagediaeatem sews | zTHTC-
EMST:, THY FATA, SHA TT, TH TeAleq: ) ae-
aes ara aft) fegafies STRAT
afa anfaanngqeaeacames: faa Tad, wa Host
feerafy
ATaAA | TET eraTaey aefasascacaraey: atfay
TI Tal) Anugta aearafa, arendifa, aeasaeana
viaraas «fageditagare | eaedt aot eft farteq:,
azvarmatrare—eqararfeta | afar: |
S| AL-S-AT AeaeAT | g1RT Ee
fegafiet ctarmearam xf aotfateoeaaensne:
Sitesi IEeRtt, araqfreaminrary. aeerafasy srr: |
SUA: | AeA TESA MtfayR:, aera faz
O.P. 129—18
274 TECHNICAL TERMS OF SANSKRIT GRAMMAR

qeeqay frame Ged wid, “aragaifatasar-

aTanaara”’ [F213] fa fosTaareatagaane| SATAN: |
Ql AaATSTANT! wW2le
T) PAT ATHATSeaAeMT | UWVWIRwY
ANA Mey Aaa at aed fawedifa “ere”
[esice ] aff o, “stgaisaemufsate:” cares: |
TAIRA |
@| aveareftfaarfar) wars
aeaet wea | afa:| eftfha-aeea acarfresaa|
‘afagq ataraq” sfa cacao) adtset aTaargat |
TVAT|

SPARSA.

“Sparsa” (from sprs ‘to touch’) literally means ‘contact’.

In grammatical literature it denotes the mutes (including
nasals), because in their case the contact of the sthana with
the karana is complete. The word appears to have been used
for the first time in this sense in AA. where in the first two
passages in which it occurs it is used in opposition to Usman
which evidently includes the semi-vowels. In the sense of
mute “‘sparsa”’ is found in the Srauta sutras, RP., T.P.,. CA
and PR.

aera § eat ar, aa equtcafacdsamar at

qatar vata) Bary 213%
TAACATAA: ATT Terea, Keats Ey, ASNT:
a tC
 SPARSA 275

wateqg, ara otfaq retaeeqegedacaensaiafa &

TATE SAT ATTSHA? | VATl Re
ofeat aa waut: seafterenteamn:,
fea: war) at
SG wTat:, araracate:, sift equ: | wearer ad
ATA:, THATSATSATN:, ATHATET KITT: | TASTY SE TNT,
TACASATT:, AA: LACT: | ATTTET BI TAT: AITTACATSATT:,
SaTTET CATT? | QRS
WAI Ai AAAAaAHI | MIASAT eRIkee
VP. uses ‘“‘sparsa” in ‘“‘sparSesveva samkhya” i, 49 etc.
and defines it in chapter viii which is regarded as a later
addition. Similarly CA. uses the word in the rule ‘“‘sparsah
prathamottamah” i. 6 etc. without defining it.
Similarly RT. uses “‘sparsa” in the rules “sparse rgasya’”’
13 and ‘“‘sparsah sve” 25 and defines it in the introductory
chapter which is distinctly a later addition.
Panini has no occasion to speak of the mutes and so he
neither uses the term “‘sparsa” nor any Pratyahara to denote
the mutes. Katyayana uses it once in the Varttika:
slater wratenrdainsatarrca
frag | eisigos
Patanijali uses.the term in the sentence :
ETD’ TAMA BTU V1VILo
K. has no occasion to use the word, hence it is conspi-
cuous by its absence in K.
Mu. uses the Pratyahara fap for Sparsa.
SEAT | Sara: eae | ans
ast sagan aren aah eraidanr affacar: | ae-
HCI: MINT: HITAY ATEAT: |
276 TECHNICAL TERMS OF SANSKRIT GRAMMAR

aot) ara qegfasta:

erat: 1) eps
aTST | (a a 212 ) fafa fate firfa fafa fefa waz: 1
fafa fifa fafa finfe fafa aan: | fefe fafa fefa feta
ficfa zat: | fafa fafa fafa farfa fafa aa) fafa
fafa fafa fafa fafa fafa vat: | xfaeamt) ie-2
NT) aaa: evatean: afrareeaateha: | exe
‘““Sparsa”’ is used in its technical sense in the following
stanza of the Bhagavata Purina :
a faraaa gertanatea-
warring fanfed vat fara: |
Wag aq Texaantaa
faasgarat ao wet fag: 1 RIRe
‘One day, while meditating, hé heard in the water close
by a word of two syllables uttered twice—that which is the
sixteenth and that which is the twenty-first among the mutes
(constituting the two syllables, and the whole word meaning)
O King, that which is regarded as the wealth of the
penniless’. |

VARGA.

‘Varga’ literally means group or series. In_ the

Srauta Sutras and Pratisakhyas the word is used to denote a
class or series of mutes or consonants. Thus we find in ASS. :
sparsesu svavargyam uttamam i..2. 16, ‘before a mute, m is
changed into the last consonant of the series of - that mute’.
In RP. we find “‘varge varge ca prathamav aghosau”’ i. 12,
 VARGA 277
‘m each series (of mutes) the first two letters are
aghosa’,
rkara-lkarav atha sastha usma jihvamiliyah pratha
mas ca
vargah” i. 41. ‘r, 1, the sixth Usmam and the first
series of
mutes are produced at the root of the tongue’. It
is worthy
of note in this connexion that PR. invariably.
adds -kdra to
the name of the sound and then puts in ‘varga’
, thus it
speaks of cakara-varga (i. 42., xii. 1), takara
-varga (i. 43,
v. 11), and takara-varga (i. 44, y. 11).
In v. 42 etc.*
RP. speaks of ‘sparga-varga.” TP. and CA. invariably use
the shorter forms ka-varga (TP. ii. 35, CA. i. 20 com.),
ca-varga (TP. ii. 36. CA. i. 7), fa-varga (TP. ii.
37, CA.
li. 14), ta-varga (TP. ii. 38, CA. ii. 26) and
pa-varga (TP. iii.
39, CA. i. 25 com.). VP. uses both takara-varga and
ta-varga:

TRTARA SHC ViRs, THT gla, aR

TGTA wigs, ( Cale aTHHA aT eva wie) |
TP. distinctly says: prathamo ‘Vvargottaro vargakhya i.
2/7. ‘a first mute followed by the word varga
is the name of
the series’. VP. is not so clear, it says
: prathama-grahane
vargam i. 64, ‘by the mention of the first (mute
) of a series
(‘followed by the word varga’ Uvata) is
to be understood the
entire series’.
Panini goes a step farther and shortens ““varga” into ‘‘u”
FS
rat
er

‘ame Bnaiegaq (a4 sala aarasarz. CEFR HTH: |T-

aia afeaq | watl-zat-aaat aera) gaz) )
Waa May eR, TAMA TeTAT! Wa}: GAT | 223
fafa mamaTsIEE NA aT! AENTT tee
278 TECHNICAL TERMS OF SANSKRIT GRAMMAR

which is merely the Samprasérana form of “‘va’, the first

syllable of ‘“‘varga.” He distinctly lays down that a letter
with an indicatory ‘u’ stands for both itself and its homo-
geneous letters (‘‘an udit savarnasya capratyayah i. 1. 69).
In exactly the same way an indicatory long & in the Dhatupattha
denotes that the root admits the union-vowel i optionally (P.
vit. 2. 44), and an indicatory short u denotes that it does so
before the suffix -ktvd (P. vii. 2. 56), because uw and @ are the
reduced forms of va denoting optionality.
RT. goes farther still and uses the first letter of each
varga to denote the varga, and in its rule 13 uses ‘rga’ for
‘varga’.
| K., Hce., Su. and PR. use “‘varga’’.
J., Sak. and Mu. follow Panini.
HN., as usual, prefixes Visnu to ‘‘varga” and uses
Visnuvarga as the generic term for ‘“‘varga’’. In the case
of the specific vargas, however, it uses ‘ta-varga’”’ etc. (i. 102).

aT) [aware eae: |] wey A ay aM: | xe

qt, eMataraTagedor qeyqsy ani: | vanfserraaia-
AGMA: | ULe-2e, AMAT AMAL BWiaT |} —-e1a9
aot | AMAA aT) eey
amiat Sane RT wea | saz; anet
TAA AT I: Waa: | AAT |
[agqrearfiar) a zane aait arofagnt: |---2139]
mT | dat wy wy TT] eIRIRe
TINS: SHAT SATA | TET |
21 THAT AT 1 reer
 VARNA AND SAVARNA 279.

AAA-AAAY Ve, Ba an RavifeAzarsar aai-

ama tafentamsrs fag: | aaa qypdemda
sqaaataate aa | SIRTTE: |
AAT KATY H-A--T-T-ATHT: |
qsafa: gaphraataat: qey sattarat: |) FATT: |
B( e-a-2-a-qMat wyarat avant seaflaag | 21218
area |
3) atfeaiea qaasgrataar ardarat: | ire
Zl Bt ateat: gy aay faeyant: | eize

VARNA AND SAVARNA.

“Varna”? primarily means ‘colour’, then it came _ to
mean ‘sound, letter of the alphabet’, probably because the
letters when written were covered with a coating of colour*.
Dr. Batakrishna Ghosh says in his article on Aspects of Pre-
Paninean Sanskrit Grammar (B. C. Law Com. Vol. p. 338) :
“This meaning of the word varna should have been deve-
loped first in the Brahmanas of the Samaveda in which we
constantly come across locutions like rathantaravarna rc
‘verse which gets the colour of Rathantara Saman in chant’.
In these passages the word varna is visibly changing its
meaning from ‘colour’ to ‘sound’ of melody. Thus gradually
the ‘sound of melody’. became ‘sound .in general’.” It may
be mentioned in this connexion that “‘varna” generally comes
immediately after “‘svara” and works like the Naradi Siksa
ni eee
*In ancient Egypt all sacred manuscripts were written in colours.
 280 ECH NICAL TERMS OF SANSKRIT GRAMMAR

mention the different colours of different melodies.* As has

alredy been pointed out varna is sometimes used in the sense
of ‘syllable’ also.
In Tantra literature ‘‘arna’’ is often used in the sense of
varna’. Ther of “‘varna’’ was probably first dropped due to
the exigencies of metre and thus “‘arna’’ came to be used
for “varna’”. Writers probably thought that since udaka,
daka and ka mean the same thing why should not ‘‘varna”’
and ‘‘arna’”’ have the same meaning? Gradually ‘“‘arna”
came to be used for “‘varna’”’ in all its senses. Thus in the
Bhagavata Purana (ii. 6. 45) ‘“tadbhutarna” is used in the
sense of “ascaryaripa’.
In PR, “arna” is a special technical term denoting any
letter except th, y and vowels of suffixes. Thus ‘‘arna”
corresponds roughly to padanta of Panini and other systefns
(according to Sanskrit grammarians members of compounds
are regarded as pada, even though they have dropped their
case-endings). “Arna” would thus appear to be merely a
disguised form of “‘anta’’..

TATAATTAIST: BIT | 1 e132

TATIBN- TATA TATA: LITT | A A
vafrerc: vafararc: cafe sraasy get | aha: |

* UNdaae: FSH Wa: TafdRr: |

i ee

SAMA WAIN aayA: HeaTA: |

TeAG Raq HU: Maa Wad fae: |
fame: ara: efearan: exawan: q
tat frat | =cy 2-2
On Varna see Goldstiicker’s Panini, pp. 25 ff.
 VARNA AND SAVARNA 281

PR. has very cleverly added th to y and ac of P&nini.

Because the great grammarian does not take into account
th in framing the rule “‘yaci bham” i. 4. 18, he has to pres-—
cribe the agama th in his rule “sat-kati-katipaya-caturam
thuk” v. 2. 51, instead of the suffix tha as in the imme-
diately preceding rule “that ca cchandasi’, for if tha be
prescribed as a suffix in the case of catur and sas they will
be regarded as padas before the suffix according to “‘svadis-
vasarvanamasthane” i. 4. 17 and the resulting forms would
be ‘catustha’ and ‘saddha’.*
““Savarna”” goes back to RV. where it is used in the
sense of ‘having the same colour or appearance, similar’.
In grammatical literature the word means “having the same,
i.e., Similar, sound, homophone’. Panini makes provision
for the change of “samana” to “sa” in a Bahuvnhi com-
pound in the rule- “‘jyotir-janapada—” vi. 3. 85. Later
commentators, like Durga, think that Katmadharaya will
serve our purpose.
RP. (i. 55) defines Savarna thus:

QAI WTA CATION

SATIN SATA Eqvit ||

* my fray gegen vam fda 4 es Bea TT eT aaa at?

Beas: TRACY Gerafafy-nfatarag |
aeaal: Trace faa qerafafanfaa way eran germfaiuee |
quam cee wal) Ta ema) gamef aaA an eq: - ware
afer! ae) Gee ae A aq) ose aay sff qermahfa
faa ar ayfeta ) aeraTE WIRIVe
 282 TECHNICAL TERMS OF SANSKRIT GRAMMAR

‘In the statement regarding the place of articulation as

also of contraction of vowels, we are to understand the homo-
geneous vowels—long and short, even when only the short
vowel is mentioned.’
TP. (i. 3) is very simple in its definition :

SF WaT gazes|
TaAqaIaA | Matted ara aezaATaa | fauwacaa |
VP. is very explicit and appears to include for the first
time consonants also within the scope of “‘savarna”’ :

AAMAMARLMATAA: AaqN: | 2183

‘Those sounds of which the place of articulation, the
producing organ, and the effort in the mouth are identical are
homogeneous.’
In: “‘pratyaya-savarnam mudi Sakatayanah” VP. iii. ‘9,
‘a visarjaniya followed by a sibilant is changed into the
savarna of the following sound’, the ‘savarna’ is identical with
the original sound.
CA. uses the word without defining it. The commen-
tary (i. 27), however, evidently follows Panini and says:

gaara 3 sort seat Asgarfeaa: |

AAMC 4 F Bai zfa ewan: |
Panini’s “‘tulyasya-prayatnam savarnam” i. 1. 9 would
appear to be an abbreviation of the definition as found in
VP. In VP. ‘karana” is superfluous, and in Panini we
miss “sthana”. Panini evidently takes “‘isya” in the
of sense in the ‘“‘dsyasthana’’ and “‘prayatna” sense of
“abhyantara yatna”. “Samana” he paraphrases by “‘tulya’’,
 VARNA AND SAVARNA 283

saving one syllable.* In the Mahabhasya various explanations

are offered of ‘‘tulyasya-prayatnam’”’, but the one put forward
first viz., “‘tulaya sammitam tulyam, asyam ca prayatnas ca
asya-prayatnam. tulyasyam ca tulya-prayatnam ca savarna-
samjfham bhavati’ appears to be the most plausible
explanation.
C. uses “‘sasthana” for “savarna”’, cf. “halo jharam
jhari sasthane lopo va” vi. 4. 155 for Panini’s ‘jharo jhari
Savarne”’ vill, 4. 65. For “‘asavarna’”’ he uses “‘asama’’; cf.
“‘dvitve piirvasyasame” v. 3. 84 corresponding to Panini’s
“abhyasasyasavarne” vi. 4. 78. |
K. follows TP. whose definition it amplifies thus:
aot al graetissrer Aah | gig12
aaa aa: aat:, @ gates aaa framaicty
Seaztagreqqeya | Zar |
Since K. restricts savarnatva to vowels it uses ““savarga’”’ ‘of
the same group of consonants’ for the consonants.
J. paraphrases P&nini’s rule and lays down “sasthanakriyam
svam’’ i. 1. 4 using “‘sa’’ for “‘tulya’”, saving one syllable and

*Panini uses “samana” in its Vedic sense of ‘identical’ (cf.

Samanam cid ratham atasthivamsa nana havete RV. ii. 12. 8), €.2.,
samana-kartrka iii. 1. 7, samana-karmaka iii. 4. 48, samana-tirtha iv.
4. 107, samana-pada viii. 4. 1, samana-pada viii. 3. 9, samana-Sabda iv.
3. 100, samanadhikarana i. 2. 42 etc. “Tulya” is explained by Panini
as ‘tulaya sammita’ (iv. 4. 91) ‘measured out equally by the balance’
and used in the sense of ‘similar’; e.g., tulyartha ii. 3. 72, tulyd-kriya
lil. 1. 87, tulya-yoga ii. 2. 28. It is curious therefore, that Panini
Should use “tulya” here instead of “samana” in the sense of ‘identical’,
284 TECHNICAL TERMS OF SANSKRIT GRAMMAR

““sva’’ for “‘savarma’’ saving two syllables. It will be noticed

that J.’s term “‘sva” is made up of s from the first syllable and
va from the second syllable of ‘“‘savama’. The word
is very appropriate since ‘“‘sva” literally means ‘belonging
to one’s self, one’s kin’ and thus practically means
the same thing as savarana, but J. has been anticipated
by RT. which uses “‘sva” in very nearly the same sense in the
rule “sparsah sve’ 25. For “‘asya’’ J. substitutes the more
definite “‘sthana” and “prayatna” is replaced by ‘“‘kriya”’ with
the saving of one syllable. Sak. has ‘“‘savah sthandsyaikye”’
1. 1. 6 where “asya”’ is used in the sense of “‘dsya-prayatna”’.
The commentary says:

ae Fa Ae Tia wh HHO | da Aa
tye fsa AHA |
Hc. improves on his predecessor by framing the rule ‘‘tulya-
sthanasya-prayatnah svah” i. |]. 17. Both “‘sthina”’ and ‘“‘asya-
prayatna”’ are explicitly mentioned here.
Sam. uses “savarna’’ in the rule ‘“‘savarnenag dirghas ca”
i. 85 without defining it. The Vrtti says:

Bae TAMIA fet: Bani: |

Goyicandra explains-:

SATAY
Ft: AA: | AAA: ASM: | sna
tai seTNAy
ga: we guint a ae: ae aanfeaa wa) sari
at: ae gaat a ae: ae aaqaka a af arde-
farcrariare—anrceaeane |...z)at eat Gant wardt-
aE) «TAMIA, Taormina «aaorcaqy-
waaay, ta gach ae Bavraratcacat|
 VARNA AND SAVARNA. 285
Intent on having a monosyllable like J. and unable to use.
“sva’’ which is reserved for “‘hrasva’’, Vopadeva: uses the last
syllable “‘rna’’ for “‘savarna”’ in his rule “flapo’k samo ma rk
ca’’ 6 and since he specifically lays down that similar mutes are
homogeneous. with one another and similar. simple vowels are
homogeneous with one.another, he defines ‘similarity’ as ‘having
the same place of articulation’ (simyan tvekasthanatvam)..
As a follower of K. Sar. restricts savarnatva to vowels but
goes a step farther in admitting the Plutas within the fold of
“‘savarna’’ and says:

sadiegaan: aan: | eer

Padmanabha’s definition is similar to that of Vopadeva :

TAA AAAA AA | ies

PR., as usual, follows K. and restricts the term ““savarna’””
to the simple vowels:

VATA, ATCT VAT ATITGTSATA: |g 1912.

HN. also follows K., but is not satisfied with mere *““varna’”’
but wants to drag in Atman (soul). so its term for ““savarna”’
is ““ekatmaka”. Thus after mentioning the ten simple vowels
(daSavatara) HN. says.

asi atartraret 1 tiers

Originally “savarna” appears to have been formed after
“‘samanaksara”’ and was, therefore, restricted to the simple
vowels.
For ‘‘varna’”’ HN. uses RAma.
 286 TECHNICAL TERMS OF SANSKRIT GRAMMAR

KARA.

The element kara appears to occur for the first time with
letters of the alphabet in AB. xxv. 7:
Pasha At AAIcd—aARIa THT WHIT
ta | aaa aaa) adcazfta|
“From them when brooded over the three sounds were
corn : the letter a, the letter u, and the letter m. Them he
brought together; that made (the word) om.’’ KEITH.
This -k@ra has evidently been formed after the analogy of
kara in words like ‘“‘vasat-kara, him-kara, om-kara”’ etc. In
the case of these latter words forms of the root kr are also
used with vasat, him, om etc.; thus we have vasatkaroti,,
vasat-krta, him-karoti, him-krta, anom-krta etc. So when
we find kara used with vasat, him, om etc., we are inclined
to regard kara as formed with the agent suffix -an as in the
case of the well-known words ‘“‘kumbha-kara’’, “‘nagara-kaéra”’
etc., vasat-kara meaning ‘the maker of the sound vasat’, thence
‘the sound vasat’. Though we never come across forms of
the root kr used with the letters of the alphabet still the element
kara was extended by analogy to these cases also, a-kara
meaning the letter which products the a-sound, ma-kara the letter
which produces the m-sound and so on. It is, however, better
to regard -kara in these cases as formed with the verbal suffix
-ghafi in the sense of action, pronunciation, sound, and then
compounded with vasat, him, om, .a, ma etc., so that ‘vasat-
kara’ means the sound ‘vasat’, ‘a-kara’ means the sound a and
so on. Thus -kaéra came gradually to be regarded as a pleona-
stic suffix and so we find in VP. i. 18-19:
 KARA 287

TE aT | sterHre ATEAT |
There is also the well-known stanza about the meaning of
the particle eva:
AMMA AAA = |
~aataafa arey wane aa 1
Svaha-kara occurs in AB. xxiv. 1: “tanto vai svaha-karah”’.
With this we may compare “‘anto vai svar’ Ib.
After laying down “upadista varnah” i. 34 VP. Says :
Fazer afar | arta a | eeaafeta crepe | e13e-3¢
In TP. (i. 16-17) we find similar’ rules :
AT: BATA FUITAT | SH ITeTIaT SAAATATA | 212 G-29
In the introductory chapter of RT. we find:
AT Eds Ya: HITIAILATANY |SAAMHRTCTa | HTT |
fafa 2: |
Vyasa Siksa 13 lays down that ‘‘varna” is added to simple
vowels. In TP. i. 20 we find:

BLA ATATSANTA |
‘A short vowel with varna after it is the name of the three
(varieties—short, long and prolated).’ Acc. to Whitney
‘‘varna”’ in this case, indicates only the ‘colour’ or phonetic
complexion of the vowel, without regard to its length. As
Whitney points out since “TP. acknowledges no protracted y,
und neither a long nor a protracted /, it does not admit the
compounds rvarna and Ivarna.”
Katyayana has the Varttika “varnat karah” iii, 3. 108. 3.
On this Kaiyata remarks :
anratfaat aoigacurfeeres: | agave afar
 288 TECHNICAL TERMS OF SANSKRIT GRAMMAR

wafa, ‘se zal’ [oa]efa aati aa afag anr-

aqaagacngy—aaait tfa | AIT HATTA
dat satnaratara aafa| ad BU HIT | AKT AIT:
anit xan faxafa | aazfa, cafe fanaa anc
auufaans a fart) aicaaed wefae [ 212123]
seafaantce sicatafeacdste Hadst | adtaeraem otfa-
qncatsragaata: |
Nagesa says:

aqdaata | wagentfracs |
The suffix -kdra is not added to the Ayogavahas because
they are not read in the varnasamamnaya :
argent-aa-faasttta-
fagQAA TATA: | AT AT epBe
Though -kara is added in the sense of ‘varna’ there is some
distinction between the two. KdGra can be added to any letter,
vowel or consonant, short or long but varna is used only with
the letters read in the Siva-sttras. Thus we can have akara,
akara, ikara, ikara etc., but we can only have avarna, ivarna
etc. and not 4varna, ivarna etc. Then again ‘‘akara’’ means
the short a, but ‘‘avarna”’ means all a’s, short, long, and prolated.
See Goldstiicker’s Panini, pp. 26ff.

SH) AMT AT | Byaies

AT | ANITA: | Blelere
agefieta |ta ‘afafreia care) a aafs, cane
carat
atwafa ta efa “fer: my” fa fee, cant
saqisad | facaraftr:|
2] AMSTATS TART HIT! | IRIENE
 KARA 289

TNA STAVA TABATAMT

AIT: SITS Mca werk|
ART: | CHS | BHT: | Bare) | RRrerfreaqar
SSTCOTT: | TTS | ATT) | IRTRTS | TTT: |
ATH: | AHL | AERTS | aS | FART: |
TART | SHIT | RAT | AlaHt:
| Gant: | AT
aa egfa: | qiatd-qqway daqedaq fa aafa acer
ateda ardd aaa tere afacafia, aera
fareq ‘agra vercafe’ arate afrad’ ‘gent acd
cate a farafa, cwaaa tfa fea) a: feo,
e:ata,
R: AM, GAH, Al sa avfseqra, crema, ea
fara rasacaai a aafe sritsqaacerarfe aafir |
WA TF BACHE: | HEAATASTT: | Teahe:|
Aw iwANadfa | WT H rerer are xf Freqaa
aat
atfata acre feasts a aged) saree: |
Sa sof aaniishy sia: Ht a Fas: | Rava: |
Bil ANT RT | ewzi22e,
EL MOTTA: | ART TE RaR-4
NARI TIA: | L139

TRA Wal AIT ATER Trae aa: | aehaafuean-

Sara: Tar area ECA eae: | aA ee eerie orftra:|
Rae rate a neve, awrancua corte ware
CUT | Mag TeaeyWH fe |
O.P, 129—i9
 290 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Afesaara aa: aitfesd ared a

facaafe aut at ea ararrarfag |
aty agfe qa Sfefaenraaeat-
ATAHATAMAR AAS AMRIT: ||
caTfritfacre 2133
aa aiafiararar
sored afiresiag|
ageragel arat ware: refers: erat 1)
AMA ACI AA Aaa: ATAAE: |
AU GHAIAMNT: GH: AA TSA: ||
ATAT
HStAa ATAGST ABTA: |}
AATAT
gfcat44 Gard AHI: Ty TST:
WaT IHU Hat Arta AAT AGT |
Al AZ) AH AKA ARI TW AYN: |)

REPHA.

‘‘Repha”’ is the name of the letter r in grammatical literature.

It is found in the Srauta Sitras, Pratisakhyas and works on
grammar posterior to Panini. The word is formed from the
root riph ‘to snarl’ (ripha katthana-yuddha-ninda-himsadanesu
—Dhatupatha vi. 26). As Whitney points out “‘it is
the sole
example of a specific name for an alphabetic element
of its
class”. The derivative “riphita” is found in VS.
in the sense
of ‘pronounced with a guttural roll as the letter r’, In the
Pratisakhyas the word means ‘rhotacized’.
“Viriphita”’ occurs in AB, several times;
faftfrae miagisef agqdeaat STL] 2218, HE Aa
cn: ataetf an a
aad fafifed faftfaaer we-
agacett aquearat BIT] eI"
The sense of “‘viriphita’’ in the above passage is not clear.
Keith translates: that which is by Vimada, that which. is
sounded, that which has various meters,...... these are the
symbols of the fourth day... .‘With Offerings for ourselves,
Agni’ is the Ajya of the fourth day: it is by Vimada and js
sounded, being of the seer who is sounded, on the fourth day
it is the symbol of the fourth day... .‘There is Indra famed,
in what to-day ?’ is the hymn by Vimada. which is Sounded
;
being of the seer who is sounded, on the fourth day it isa
symbol of the fourth day.” ‘‘Viriphita”’ appears to have been
used here in the sense of ‘without the sound 7’. (Cf. “‘ariphita’”’
‘not rhotacized’ found in the Pratisikhyas.) The root riph is
used by ASvalayana in the rule: visarjaniyo ‘natyaksaropadho
riphyate i. 5. 10, ‘a visarjaniya preseded by any vowel
except
a and 4 is rhotacized.’ In the next rule “rephin”’ is
used in
the sense of ‘rajata visarjaniya’ : itarag ca rephi, ‘and the
other
visarjaniya (i.e., a visarjaniya preceded by avarna) is also
rhotacized, provided it is a ‘rephin’ or ‘fa-jata visarjaniya’.
VP. derives “repha” from r with the suffix epha: ra
ephena ca i. 39. The corresponding rule in TP. is “ephas tu
rasya”’ 1. 19. Katyayana has a Varttika “nid iphah”’ iii. 3.
108. 4. VP. and TP. add the suffix to r, hence the suffix
is
epha; the V4rttikakara adds it to the a-stem rg. hence the
292 TECH NICAL .TERMS OF SANSKRIT GRAMMAR

suffix is ipha. And because the Varttika is found. under

“rogakhyayam nvul bahulah”
2a
iii. 3 108, the suffix is regarded
4

as a krt suffix.
aot, [aut Hera anieat| awaited eayar-
AT Ulgiee | THE TET) eee
UC LUM AGA ABTIAH WHET AAT | TAT...
fraaaa.-frera famzaa sorfeaenafagenia xfa
tH: | aaa asardafaasdtaearenfarnepanaa:—
fasataisteaagn frraa xa, aaramea wIt-
qeareaaaaeataral a ander zfa gqueqet| al
afm: [ 2laieccis ] caanrafeaea <caqanrfanacat
fadtaa, a gq aafsta afer |
aot, [fara faa) ain a, acafaa
sae | BE-3¢
1] eatehe | Tahaa) epse-go |
arearad | [satfact arafea | ata ac: |] ofr:
BIBIZCI |
Zl TWH: 9121229"
teaenaeaat at vafa | tH) sTaaaate THIT
qeate|
@) cen: | afrarfcfirs exue
Tema aod afacar aK: eae) af: |
Bl TH 1 VWvRZe
T) WH?) GIVVaz
tH TeaT, | aAMisamA) cHrerdif arate |
afa:| sean xfer tafaarrot eqs req: |...
 REPHA : 293

carafe amar areal aa aay |

aq: Feat aft TAATATAEAt || TETRA |
In Raja-tarangini vi. 39 “repha” is used in its ordinary
sense :

an fravqareia agerdosgy: |
th ant afar afta fafa a |
‘From the fact that a large sum of money had been
paid as
fee to him who deserved a limited amount, the king conclu
ded
that the merchant had got the letter r changed into
s.’
In the following two stanzas from the Yogavasistha
Ramayana “‘vepha” appearss to have been used in its
primary
sense of ‘a burring sound’.

AR F HASTA
H ITETAT |
AMAT Tete fags aaa fSes | TMT yy20
(ator seater ameta coed: Mar fear ce TTR
AATH TPN AAAS A FIT arearwat | Aa 1)
‘Held by his mother on the chin marked with the soft
downs of fresh youth—his mother who was intent on the
utterance of incoherent words ringing with the burring sounds
of loud lamentation.’

ana agataeal ATIGaS

tea |
Ten TogHrataalrs GAH yy TTT wee
With the graceful movements of his hand delicate like fresh
foliage stirring with a soft sound the strings of the lute vocal
with music and so appearing like a tree flapping away with a
soft sound the strings of humming bees.’
 294 TECHNICAL TERMS OF SANSKRIT GRAMMAR

In the following stanza from the Bhagavata Purana ‘“‘repha’”’

appears to have been used in the sense of svara or note of
music.
Gig aH waaaTz-
MAY MAY AACA |
Rasy areactierent
Hes AT AAA AAMTHA || CRA’
(ay etal! BAR Hest wear | sthatenf | )

SAMKHYA AND SAS

Acc. to Panini, not only the numerals ‘eka, dvi’ etc., but
also the words‘bahu’ and ‘gana’, as also words ending in the
Taddhita suffixes -vatu and -dati are comprised within -the
term “‘samkhya’’. Of these again numerals ending in s and a,
as also words ending in the taddhita suffix dati are known as
“sas,” sas being taken as the type of such words. Though
K. uses technical terms like ‘‘Sraddha”’, ‘‘agni’’, ‘‘nadi’’, etc., it
has got no terms corresponding to Panini’s Samkhya and sas. So
far as declension is concerned, sas is required practically for one
rule only and K thought it would be “wasteful and ridiculous
excess” to have a new technical term for a single rule. So having
in mind a rule like Panini’s snanta sat’ K. frames the rule “‘sam-
khyavah snantayah”’ ii. 1. 75. The fact that K. brings together
numerals ending in s and n in this rule and has a separate
rule “‘caturah” for the numeral ending in r which is excluded
by Panini and his followers from the purview of “sas”? would
lead one to suppose that the technical terms ‘‘sas”’ was used by
 SAMKHYA AND SAS 295
the predecessors of Panini also. C follows K.
and avoids these
two technical terms using “samkhya 5 n’’
everytime. Following
the Bhasya* C. also uses the atidega-rules : kati-
ganau tadvat.
vatoh. iv. 1. 33-34. J. follows Panini and:
uses ‘“‘syi” contain-
ing both s and y of samkhya” with a
euphonic i for
“samkhya”. For “sas” J. has “il”, Could “i”
have been
Suggested by dati, the i being placed first with the
Vedic and
Prakrit inter-vocal form of d added to it? Sak.
uses the
technical terms “‘samkhya” only, ignoring ‘‘sas”
which is
necessary for only a very few cases in declension. Hc. has
certain atidesa rules so separate technical terms are un-
necessary. “Vopadveva follows Hc. Sam. uses panicadi (vi
400)
for Panini’s “‘sas” (i. 1. 24.) Padmanabha does not follow
Panini in this particular instance. ‘PR. follows
Panini and
uses both the terms “samkhya” and “sas. HN. has no special
technical terms corresponding to these.
“Du” appears to have been another ancient
term for ““sas’”.
In the Mahabhasya we find: ki punar
du-samjiia (i. 4. 1)?
“Du” has evidently been formed with the d of
dati along with
a euphonic u.

Tl aarrag-sfa der) wren qz) sft ay

RAR | (qaigtat fared dearer fread |
eater|)
mt) aft-noaqai demaq) ata ase
St fe dem)
~”
agi 8) gaserd_q) sed-
HL Te Shas, MET FHL eieiaa-ae
* waa Re Gyray) |awelaeaisaq|
WIIVIGSAT: Wanaxantta |
 296 TECHNICAL TERMS OF SANSKRIT GRAMMAR

mm) sects dear, aero 82) w-aareserd: |

ameiqaqay SLI LVIR-RR
g | [ Hara: Tatar: Wgqy: |) «tlelae ] Fad eqeqt-

a7 | be az | H-AATASeaT: | HATH: qT: |

VIRIAS-BR
ty oSeararfardzar, afmangicfas case,
Yl awras saga AT) yor
a) der onfgagucfaararacat Aat: | Rex
SUA: AAHUaA; ASIN: TLARHE WAT: | xR

AVASANA.

A rather remarkable technical term in the Panini system is

Avasana which is defined as ‘virama’ in the last rule of the fitst
chapter of the Astadhyayi. That the term was not coined by
Panini but had been current before his time is clear from its
occurrences in Pratisakhy literature. And in the sense of pause
it goes back to the Kausitaki-Brahmana. The word is derived
from ava down and _so to bind (ava-so meaning originally to
unbind, unharness) and its primary meaning is a place where
horses are unharnessed, resting place, then it came to mean
conclusion, termination, end, pause. In grammar it means
the end of a word, verse or sentence. Katyadyana points out
that the definitionsof the technical terms Samhita and Avasana
(parah sannikarsah samhita, virimo’ vasinami. 4. 109-110)
are unnecessary, as the words are too well known in current
speech to need any elucidation (samhitavasanayor PORANIGILAL VA
siddhami. 110. 8)..
 AVASANA 297

The word is found in the sense of ‘the end of word’ in RP.

i. 15: “tasmad anyam avasane trtiyam Gargya sparsam’’ where
‘avasane’ is explained by Uvata by as paddvasGne:vartamanam’.
In VP. “parav avasane”’ iii. 31 (avasane sthite parityetasmin
pade pratyaye—Uvata, avasane sthite pdri-Sabde—Ananta),
‘“supadavasana-varjam”’ iv. 22 etc., the word is used in the
sense of ‘pause’. The whole of chapter VII of this Pratisikhya
treats of words in pausa and begins with the adhikdra-rule:
“athavasanani”. TP. has “‘avasane ra-visarjaniya-jihvamuliyo-
padhmaniyah”’ xiv 15 which is explained by the Tribhasyaratna
thus : padavasane vartamano varno rephah etc. In ‘“‘avasitam
purvasya”’ xxi, 3 ‘avasita is used in the sense of ‘a letter at the
end’ and the rule is explained thus by Mahiseya : yad antyam
vyanjanam padavasanavartityarthah, tat purvasya_ svara-
syangam bhavati. The same sense of “‘avasita’ is found in
RP. also. Cf. navasitam vi 7 (na khalu padavasane vartamanam
rephat param vyafijanam kramati i.e. dvitavam Apadyate.
Uvata). Cf. upadha and upahita upasarga and upasrta. The
‘avasana’ also occurs in the Pratisakhyas belonging to the
Atharva school. Thus we find in CA. purusa 4 babhiva
ityavasane i. 70: (‘In the passage purusa & babhiiva the
vowel is nasal before the pause’.) In the commentary on CA.
i. 8 ‘Avasinika’ (at the end) is distinguished from ‘padya’ (at
the end of pada). In APr. we find ‘“‘ityavasinesu cavaiva
vayutani cavayogad va’’ 14c where ‘avasdnesu’ means ‘in words
standing in pausa.

In RP. ‘avasana’ is used in the sense of ‘pause’ in ‘‘madhye’

vasanam tu catuspadanam”’ xviii. 47. ‘In the middle, however,
 298 TECHNICAL TERMS OF SANSKRIT GRAMMAR

is the pause in the case of stanzas with four padas. Similarly

ava-so is used in the sense of making a pause in dvabhyam
avasyet tripadasu purvam padena pascat kvacid anyathaitat
Xvili. 46.
It will be clear from the above that Panini does not always
choose the shortest term for his samjfias, but sometimes follows
in the foot-steps of his predecessors.* In the present case,
though virama contains a lesser number of syllables than
avasana, and though virama must have been more easily
understood in his day than avasdna, since he explains avasana
by virama, still he uses avasdna throughout his work and
not virama. Cf, his use of vibhasa for na va.
Of the later grammarians some use virama which is generally
defined as ‘paravarnabhava’ and save one syllabie, others
go a step farther and use anta, thereby effecting a saving: of
two syllables. K. which does not usually care much for
matralaghava in the matter of its technical terms uses virama
as the better known term; e.g., va virame ii. 3. 62 correspond-
ing to P&nini’s vdvasane. viii. 4. 56. Candra also uses
virdma as being non-technical. Thus for Panini’s Khar-avasa-
nayor visarjaniyah viii. 3.15 Candra has virame visarjaniyah
vi. 4. 70 and khari vi. 4.21, and for Panini’s vavasane Candra

entreaties usecase
*Some times the words of our everyday speech appear to be too
hackneyed and undignified for the needs of science and so Jearned
words, even though they are longer and less understood than their
synonyms, are used for purposes of science. It is for this reason
that Adhikarana is used as the name. of the Loc. in preference to
Adhara, Avasana is preferred to Virama and Vibhas4 is used for na va.
 AVASANA 299

has va virame vi. 4. 149.

H¢c., as usual, follows K. J. uses
anta and Mu. follows J. Vopadeva also coins the term pha for
both virama and vyafijana (haso’ ntah phah 86). Pha might
have been suggested to Vopodeva by repha which, being a con-
sonant, stands for all consonants, and which, being represented
in writing by a stroke above the next letter, may easily mean
pause also. The terms “pha” and “bhi” are peculiar to Mu.
Many grammatical operations take place in pausa or before
consonants (virama. vyafjandadisu—K.., (j)hali padante—P.).
So a technical terms covering these two would come in very
handy. One expected Vopadeva to use va for this purpose,
since the semi-vowel v occurs at the beginning of both ‘virama’
and ‘vyafijana’ but that is not possible as va is required to
designate “‘avyayibhava.”” So Vopadeva combined the p of
‘padante’ with the ha of ‘hali’ and formed the technical term
Pha, It is interesting to note in this connexion that the term
is used for the first time the rule “sror vih phe” 103 which
has to do with ‘repha’ and pha is the second syllable of ‘repha’.
Sar., though generally following K., uses avasdna in its rule
vavasane vil. 66. Sam. uses avasana in vavasane hal dvi§ ca i.
189, where Goyicandra says: “viramo ’vasinam iti purvaca-
ryasya samjna, athava, prasiddha evayam avasana-Sabdah
samaptim abhidhatte” on which the Tippani-kara says: etad-
vyakarane’ vasana-samjiiabhavad aha—athaveti. In the kreche-
savyaya-pada, however, we find mo binduh padavirame
783.
Under i. 189 Goyicandra explains avasane as pada-viradme.
RT. uses ma for virama in its rule unnice me 54 which is
thus explained in the commentary: ucci-bhavati tat svarit
am
nice va pratyaye virame va.
 300 TECHNICAL TERMS OF SANSKRIT GRAMMAR

TINT | Aare aarfa: wg a faun afa waar |

afeatacog | ze fefaar faca:—anramfag:
auqarzegqea | da amamafagisgamirag arar-
Freaaarg waarat a aafe, zaceafaaaae: |
(tT 12 )
alacasaTeararc: | facet f& fefaa:— aefacrat aet-
faced, cat ‘aera arma’ zenat asta ances
Road, ‘aay gard’ cara) werfaua fesdnat acai
fag irararcey [21318] Fata area alad gaa 1 (<1z12)

KARYIN, KARYA AND NIMITTA.

Grammarians generally divide rules into three parts. That

which undergoes a grammatical operation is known as karyin,
that which is the cause or condition of the operation is known
as nimitta, and the operation itself is known as kdarya or
vidheya. Now since we always proceed from the known to
the unknown and since the karyin and nimitta are more or less
known to us and since it is the karya or vidheya that supplies
us with the new information about the karyin and nimitta,
generally, the karyin, being best known to us, comes first, then
the nimitta and last of all the karya. This is the reason why
in an Arthantaranyasa the statement bearing on the matter in
hand comes first and the statement supporting it comes after-
wards and stanzas like the following are regarded as open to
objection :
ctaReag¢e fe frat fanecaafa agarraar |
wrarras fedora ofr: acageraty ||
 KARYIN, KARYA AND NIMITTA 301

Durgadasa quotes a karika in his commentary on Mu. 21 :

arti nia’ fafa fafa: qaqeean |

nafaq aaa afadq atafafaaa: |)
aey fatgaaa wrt’ @ areal wfeat ga: |
frat Fa Aa BAASATTATTA |
wareWt Tt akdatafad feat aa |
ARH J AITAGTT Te AIT I
‘A sutra is composed of three parts—kdaryin, and karya and
nimitta. Sometimes it consists of the kdryin and karya
and sometimes of the karya and nimitta. That for which some
grammatical operation is prescribed is known as karyin and
‘the thing prescribed is Known as karya which may be a sub-
stitude (ddesa), suffix (pratyaya) or augment (agama).
Nimittas again are divided into two classes. When an operation
takes place after a nimitta it is known as parva nimitta and
when an operation takes place when a nimitta follows it is
known as para nimitta.’
In the first karika ‘nimittam’ is put after ‘karyam’ for the
sake of metre, for the kdrya being the vidheya ought to come
last and does generally come last, as in ‘‘akah savarne dirghah”’
vi 1. 101, “namo hrasvad aci namun nityam’’ viii, 3. 32,
‘‘samanah savarne dirghibhavati paras ca lopam” K. i. 2. 1. “ad
igecor nu-vri’” Mu. 23 and so on. The commentators of K.
say distinctly under “ner yah” ii. 1. 24: karyi nimittam karyam
ityesa nirdegakramah.”’ Acc. to HN. the order is as follows:
Pirvanimitta, karyin, karya, paranimitta, but generally the
order mentioned by the commentator of K, is followed as being
 302 TECHNICAL TERMS OF SANSKRIT GRAMMAR

strictly logical, though sometimes for the sake of vaicitrya

(variety) or for other causes the order is changed. Thus
Panini puts the vidheya ‘vrddhih’ first in the very first rule of
the Astadhyayi: vrddhir ad aic. In the next rule he follows
the normal order and says ‘‘ad en gunah”. Commentators point
out the word vrddhi denoting prosperity has been put first in
violation of rules of logic and rhetoric, for the simple reason
that it denotes increase, prosperity and is a fit and proper word
to begin a work with. In VP. also “vrddham vrddhih” occurs
at the end of each chapter. But P&nini again places the
vidheya first in the rule “‘aprkta ekal pratyayah” i. 1 9 where
no reason is. discernible Similarly in K. we find ‘“‘smai
sarvanamnah”’ ii. 1. 25, after a pronommal a-stem the dative
singular ending ne is changed into smai. Here the normal
order has been reversed and the vidheya ‘smai’ put first merely
to draw pointed attention to it. In the previous rule “ner yah”
the author says that ya is the substitute for ne in the case of
a-stems. This he follows up by ‘“‘smai sarvanamnah’’. In
Mu. “it krte’ 4 the vidheya has been placed first for no
apparent reason*. In Panini’s “iko yan aci’’ vi. 1. 77, the
nimitta has been placed after the karya, so also in K.’s ‘‘ivarno
yam asavame na ca paro lopyah” i. 2. 8 the vidheya is placed
before the nimitta.
Patafijali says in the Mahabhasya (i. 1. 1) :
aaa qarantta: deft, wrafrar dar |
Ha Tie] 7
*It is just possible that the reversal of the normal order here is
due to a desire to avoid hiatus and bandhaSaithilya.
 KARYIN, KARYA AND NIMITTA 303

aan fe arto: arto afyacaq | ag an—racenfy

aa arafaceer 2aza eft dat fraz |
wu afgcatafa
CAthAAeT ASS AeA) wsfs mart
Hea: welae ager afeargmfza: sged |
agerdifa fe anerft sud, adteqerarftr warenaad-
geri a, ataear sigan ama, aaafe
aratt gaiatita: dal, qtrnftar de, ate ae:
afar reat |
A commentator of KP. follows in the footsteps of Patajijali
and says in justification of the use of the word “‘vrddhih”’ first
in the rule “‘vrddhir adeSasya”’ i. 1 :

“Area Ad aH sfaa: a8 ae fatre-

aalarfafa fataaserfinad careaia ar fe HAA |
aed ageedd fastae fafaafasieaar
cenceaeyeieten sfe: ware |
Now in ASS. and most of the Pratisikhyas the karyin is
generally put in the nomitative and the karya in the accusative.
Thus we find in ASS. :
eaaty SAATMA | Ata at atagarferany
TAMATTGEAMTT | e1RILE-Le
RP. lays down:

Rareg has aH AAATT | Rag

'When a word in the nominative and one in the accu
sative
are used together, it is to be understood that the
former is
 304 TECHNICAL TERMS OF SANSKRIT GRAMMAR

changed into or becomes the latter, and the change takes place
acc. to Gntarya or closeness of relationship of the two letters’.
TP. follows RP. and says:
Aart anra-faarhe-ataarq) tk
a fanicet) gre
‘A word in the nominative stands for an agama, or for an
element that undergoes alteration or elision...... The product
of alteration is put in the accusative.’
The Vaidikabharana explains:
aaa ofcafsanmt afaarni amartfzarfaat aeqrat
faafafazint fara | anit zaatfaasadiaat ag-
franaiacnfaga mata fae: | a aq aRS-
ACNAIeN awragqzaizfe wafa ; wate sant faafs-
BUSA |
aaa faaataeg at am: aaasfioe:
ATTA A Awa xa aa: ||
THfaweq We VA AEANTT WA |
aa WNFaAt aa: a faniciz ATaAA;: |
cafaeeg 7: eq, dfearai a feaa |
qaqa AT WeAISea: B SIGt aaTrgA: |
reaarraziat saat faafeaeat vafa; amaza
qa —aginagqa: waRaT: [ey], TaCraAGcrafasay
[ eared | afa, faarfcor TA — AAT STHIT: [je],
ar ware [ie] af) xeafer—todi were:
[eR ] xfer
-_
————— ee

su-ti-sy-aprktam hal. vi. 1. 68.

 KARYIN, KARYA AND NIMITTA 305
afte frcter Rrafemreeat—
ATR ATTA ATA WAL VIAT t
fadiamn g FFX ala TaTat AA: |
TAN — AAA STITT: [Ie], TaTaraAaraeata [ogo]!
aa fedtat fai: searachrnreaa
faved, ag aT-
fan; 8a qdeatisgarfes: [ ey], waarai waz:
[ 2e1¢°0 ] cenfeaofates
fee sora |
Similarly VP. has:
afafa farmt: | £1238
Uvata explains :
afreazaaattrnr fedter frafestea | fxeteat at
fafzqaa
a fant: setae: | TT, agra
Thea AHTT
Le lefa) aamdsaent frcaqaa; arin:
a: ¢
aaragqa [ aie
| tft. aac: wart fara] |
Similarly RT. has:
ord at ce, O aa ea (eae Tears |— ahr:) |
In Panini’s system followed by most later grammarians the
karyin is as a rule, put in the genitive and the kdrya in the
nominative. Thus Panini lays down:
THT AAATT Less
Curiously enough the identical rule is found in VP. (i. 136),
though: it is difficult to understand what purpose is served by
the rule in this particular PratiSakhya which uses the nominative
with the kdrya and the ‘accusative with the karyin. Most pro-
bably the rule has been framed for cases of elision. Contrast
TP. i. 23, v. 11 ff. and Panini vi. 1. 68 which is evidently a
borrowed rule.
O.P. 129—20
 306 TECHNICAL TERMS OF SANSKRIT GRAMMAR

The Pratisakhyas of the Atharva Veda follow this method.

Thus we find in CA.:

aT FANTAT | Vee, TT Gaga THIET xy,

ARICT MAR AHIT: | AZo
Similarly APr. has: eee ne

THEE tH: 209 faasitiaer WHIT: 203, walgtal

wat eT: Veo |
The following stanzas are current among the adherents of
the Mugdhabodha school:

samt wai fedtrat a fastest |

ddia aeam a aqel a fafads |
GaalA aguet THY Aa aa: Vet |
aaa a at area wea atog? afad |
‘When the nominative is used the verb bhavati is to be
understood after it, the accusative has the sense of the adverb,
the instrumental is used in connexion with or in the sense of
saha (cf. vrddho yina i. 2. 65), the dativeis the final dative
(cf. it krte Mu. 4), the ablative shows that some operation is
to take place after it (cf. tasmad ity uttarasya i. 1. 67), the
genitive has the sense of ‘in the place of? (cf. sasthi sthaneyoga
i. 1. 49), the locative means ‘when such and such an element
follows’ (cf. tasminniti nirdiste piirvasya i. 1. 66), ‘in the sense
of? (cf. vrtti-sarga-tayanesu kramah i, 3. 38) and ‘upapada’
cf. tatropapadam saptamistham iij. 1. 92, karmanyan iii. 2.
L« ‘éte.):.
Sometimes in the same word in the same rule, the same
 KARYIN, KARYA AND NIMITTA 307

vibhakti is used in different senses. Thus the Kasikai explains

the rule afar TUCTTAR aq lu. 1. 51 thus:

atgara fast sac? a ata; care? arfeaa fia.

avg aarnfsaia an aE aA |
Similarly the rule qmenng
aft gy: K. ii. 2. 11 is thus
explained by commentators:

42: TS BUA WW ageafeqer quawa gery

aafa) af. marie qyrasegn qacfepnaeeron
WRasaT | |ZTHT |
In the vrtti on ii. 156 HN. says :
ctetefird aa rat art’ achat
TA KAN IHet TA: TIT AI |
HAI WAT TH AWAT Baa ar |
afay acfafadaa ene faaraat
BATS Tat MNT HINA F aha |
ala J AIA seat aaat rea gq?|)
faarna fatara’ fediar afafeead |
aaa Aa Bearssgits ta F |
We find in the Vrtti on Sam. v. 207 :

TH Ga Ad: era Gaal a aguz|

Gaal A A AKA We ala? afaq ||
zal anfa* [eienos] gH: ena aha Gz genes wafer |
arTRAMTTCAA [21383] | ava wafer sy ay arenas
vata. atfa vat aa [aleve] | ae weaMA qT
a
Faas
—AA Ul, MARAT TIMA—|q BRA, Ad TA gH
— GA arow
308 TECHNICAL TERMS OF SANSKRIT GRAMMAR

aft aa sat us wafa| (ad TH mreeraEyast

qsaratraraaiaeste afafcfa faarrara orf
araaitan qian fare: | aa wa “gat anfa” “ara-
aurea’ “afta aa ad” zfa ofidtraaatereca-
erarin gana gaat | afafafa ada aie
darmea meme fanfafanad aq aad tfa aa at wet
ar ae ena afacat | afag-ceord azaratat ay ea-
masa Taga, wat “eat wart mcraftaeqy1”
qaint cag are | afacsemra Bengtal a aT
aan afeaa at, afena area, aftaa nea, aftararae|
afaenamg afracardl a) ariaeg: | adareag fad
faufaarigeen Tasaag facae—agifa | tad cat
maaigna—aaraeste aa frakrratarag) [ arrarhes
cfa] aa afienta afiacdivarc:, wearaqseg
aria, waandomiefa), waa ad afae. “manant-
uraa”’ zfa qe fafencraiaia, mannaififa afifa-
ai aaa wana, werafea annvaqey scareaear-
Jad: | faaqrsgqe |
Similarly Padmanabha says under Su. i. 1, 35:
aa Weal aa: Caled TATA aa Tae |
aarti Wt afta meat ata afaa |
gi weal aa weaaeag aA aq ar wats
[21312823 ]1 eaTerat |) «aarai aeat aa
wart :
aa
Sat) gard—[ zeiee] | wat At maT At ar) aura
wert afer gtgre BA | THT GOAT (gg eR] 1 TeTA
 i a é
KARYIN, KARYA AND NIMITTA 309

Tt | ome eT [aie | Se ae: [ereee)y

SQytTe: |
K. generally follows (i) the method of the older Pratisakhyas,
sometimes it follows (ii) the newer method of Panini and the
later Pratisakhyas. In several cases (iii) it puts both the karyin
and karya in the nominative.

(i) We Te SATS: RRR, (ii) TRAPS BIRIBC,

(iii) fares fac: | see szPa: |) aRIyo-ug
C., J., Sak., He. and Su. follow Panini. Sam. and Mu.
generally follow Panini but sometimes follow K. also. Thus
Sam. has jardci jaras va vi. 113 and Mu. jaras jardci tu 116
corresponding to jardya jaras anyatarasyam vii. 2. 101 of Panini.
Sar, HN. and PR. follow K.
Violations of the principle that the genitive is to be used
with kdryin are aften met with in the: Astadhyayi, the best
known instance being sadir aprateh viii. 3. 66. ‘Commentators
sometimes quote this rule as sader aprateh, but the original
reading is undoubtedly sadir aprateh. Here Panini is either
borrowing from a predecessor who follows the principle laid
down in RP. or uses the first case-ending with the karyin for
the sake of variety.
Rama Tarkavagisa is very explicit on the point under
Mu. 6:

aa saed afstet gard aq ora fe az |

Sard Fe AST a aT, Arete
at | aT|
Te ETA WAT BIA TA ed BATTAT |
310 TECHNICAL TERMS OF SANSKRIT GRAMMAR

caed TH Tt asa afenrquae fe act

afag saeaqaal araaa Titoisasaray i
(st—saar, gi—fadiar, di—adtat, ti—qeant, T1—
qs! ai—aaat ) |
These vyatyayas would have been much easier to explain
as due to the exigencies of metre, if we could show that
Panini’s AstadhyAyi was originally composed in verse, like RP.
among the Pratisakhyas, and portions of K. and the whole of
PR. among grammars, and portions of the Jaimini sutras among
philosophical sitras, but this is hardly possible. No doubt the
thé first two rules of Panini form the second or fourth quarter
of an anustup stanza (vrddhir ddaijaden gunah). Similarly
‘“cvam a-jiati-dhanakhyayam” i. 35, “kumarah Sramang-
dibhih” ii. 1. 70, “‘bhafija-bhasa-mido ghurac” iii, 2. 161,
“sanghe cdnauttaradharye” iii. 3. 42 and the first portions of
‘“vigesanam visesyena bahulam” ii. 1. 57, “‘harater drtinathayoh
pasau”’ iii. 2. 25 also form quarters of anustup stanza, still this.
verse-form would appear to be purely accidental, even as the
following prose passage in Whewell’s Elementary Treatise on
Mechanics (1819) is an instance of accidental metre and
rhyme: :
And so no force, however great,
Can stretch a cord, however fine,
In to a horizontal line
That shall be absolutely straight.
Poets have been busy making stanzas out of Panini's rules,
the most well-known examples of such stanzas being :
 KARYIN, KARYA AND NIMiTTA 311

Chane efa* ofa a fasta |

Taredt Sree A TMS Waa BASS |]
‘He kills birds and fishes and beasts, is tefractory, has no
fixed profession but makes his living by bodily labour and is
not now obedient to us as before’. This was said by a parent
in reply to an enquiry about how his son was getting on.
The rule “sgaktau hasti-kavatayoh” iii. 2. 54 is also the
second or fourth quarter of an Anustubh stanza as the follow-
ing samasya-pirana Sloka testifies:

Ts sae & aE arfeyrareal |

aunty qaz ata efeanarzat: il
‘Though your arms, O King, long to embrace the beloved,
still for shattering in battle they have strength in the case of
clephant-kavatas (i.e., they are strong enough to shatter the
temples of elephants in battle).,
The very first line of the Mahabhasya “atha éabdanu-
Sasanam.” which
is in all probability the first rule of the
Astadhyayi and other ancient Sanskrit grammars, is also the

tiv 4. 35 iv, 4 36 Ft
ttvy. 3. 17 §vi LL 20 §§ 1.3, 62,
eal mafia est waAE fete s fig: |
me: wa ara: Uialaaa’ y Aeaq |
Caat TAT URRY, Ba! Va’ YR )
far are fateiae’ a1 Ben aaa yea wale |
fra: he eae: a" TeE RET |
( fattened 1 sper)
312 TECHNICAL TERMS OF SANSKRIT GRAMMAR

second or fourth quarter of an Anustubh stanza and a poet has

composed the following samasya-piirana Sloka with it :
wafeae fear qa’ aera |
ecfa aeqaafaca MAM aAAay ||
‘In the presence of the elders the bashful young bride
instructs her husband in love by means of gesture, then (after
the elders have moved away) she conveys her instructions in
words.’
Similarly the following stanza has been composed with
several rules of the Dhatupatha :
yaaa farsa aaraet Taste & |
ma a aafsent az waasa Hr HAT II
‘When the existence of Janded property has come to an end,
when wealth and corn are gone, when the breadth of the body
is vanished, tell me how can you speak of tranquility ?’
Similarly the rule: ‘‘svaro hrasvo napumsake” K. ii. 4. 52
has been made use of in the following stanza :
aa: ag wef guard anfa alagng
fafae Haste qar tad geal AWA |
The first two words have also inspired the following stanza :
CaUT BEA! AlawIHA WlaHeaN ALT AAA |
area arf Paar arf aatfor ara )
Similarly “Se se se va va para-rupam”’ K. i. 5. 6, one of
the longest rules in K., has not failed to attract the attention
of poetasters, and we find the following stanza :
a eft: Ta al a fase at agers: fH caviiag|
AT Haea Ben AA WT SBaT aT Taq |
 KARYIN, KARYA AND NIMITTA 313

In the earliest portion of K., i.e., in the Sandhi-section na

and na va are used to denote vikalpa. Commentators point out
that there are two varieties of va. One is referred: to as
samuccayarthaka, i.e, in it there is the samuccaya of the karyas,
the other is referred to as vikalparthaka, i.e., in it there is the
samuccaya of the nimittas. Thus under “‘va virame’’ ii. 3. 62,
the commentator Durga says : Va-Sabda iha samuccdyartho no
vikalparthah. Kaviraja explains: va-Sabdah karyam eva
smuccinoti, na nimittam.
It would appear from a consideration of the rules in the
Sandhi-section that :
i. na va is used in the sense of samuccaya, i.e., the vikal pa
of karyas and not of nimittas;
ll. na va is used when only one side is stated, when both
the alternatives are mentioned in so many words or when one
alternative is mentioned and the other obtained by anuvrtti,
va alone is used;
lili ma vd is used only in the case of aprapta-vibhdsa where
the operation is prescribed for the first time and that optionally
All the above follows from the fact that ‘tna va” was
originally used as a separate sentence. Cf. ASS. vii 10. 7, viii.
1. 14.
We shall now consider the rules one by one:
1. In “ayadinam ya-va-lopah padante na va” i. 2. 16.
the kdrya_ elision is prescribed for the first time and that
optionally. So it is a case of aprapta-vibhasa and only one
alternative is stated here.
2. In “paficame paficamams trtiyan na va” i. 4.2, na is
 314 TECHNICAL TERMS OF SANSKRIT GRAMMAR

either an interpolation or inserted for the sake of- thé metre.

Since both alternatives are directly mentioned here, va alone
was expected.
3. In the next rule “varga-prathamebhyah Sakarah svara-
ya-va-paras chakaram na va”, na va is quite regular as this is
also a case of aprapta-vibhadsa of karyas and only one alter-
native is mentioned here.
4. In the next rule “tebhya eva hakarah purva-caturtham
na va’’ also, na va is regular for the same reasons.
5. In “‘Sificau va’ i. 4. 13 va@ is quite regular, since one:
alternative is stated and the other obtained by anuvrtti.
6. In “varge tadvarga-paficamam va”’ i. 4. 16 va is regular
for the same reasons.
7. In “ka-khayor jihvamuliyam na va” 1. 5. 4. na vad has
been used because it is a case of aprapta-vibhasa of the karyas
and only one alternative has been stated.
8. In the next rule “‘pa-phayor upadhmaniyam na va’’ also
the same reasons apply.
9. In the next rule “Se se se va va para-rupam’ va alone
is used quite correctly with the nimittas, but the use of the
second va is open to objection, since it is used of the karyas
and only one alternative is stated. Here va has evidently been
used for the sake of the metre, and it is significant that hence-
forth na va is discarded altogether.
10. In “a-paro lopyo ’nya-savare yam va’ i. 5. 9. va is
quite regular, since both sides are stated.
It may be mentioned in this connexion that Hc. also uses
va and na va, but for different reasons. Under “sau navetau”’
i. 2. 38, the Laghunyasa says :
 KARYIN, KARYA AND NIMITTA 315

aah fans fag weafanicy Haq, aa aaa

qa adfa aatfante: | awa afa aa arta: |
Similarly the Haimapraka§a says:

‘a ar eeqarwazad ‘ar wae, Wt Aa at-qegaaNT-

anna ga fared: cada ; a4 TaL-aequaad aEg
aay fancdisgada eras faa: |
The first rule of the Mimamsa Sitra “athato dharmajijiiasa”’
as also the first rule of the Brahma Sitra ‘“‘athdto Brahma-

In the Kavyalamkara-Sutra of Vamana the very first rule

is kKavyam grahyam alamkarat which’ cannot fail to strike the
reader as the first or third quarter of an Anustubh. Again, the
second and third rules of the next section, viz., pirve Sisya
vivekitvat, netare tad-viparyayat forms a hemistich of Anustubh.

Sometimes it is difficult to determine whether a passage was

composed originally in prose or in verse. Thus in the Vyakti-
viveka we find:
a aaqa ae: AMAA, Boge setae,
af g fetaaar sparta agsaiia,) aaa ain og
wan, a Pate: | agate fe qa: wal seaa |
a ainaarad ,wate firafparana |
In the Sahitya Darpana we read :

aaa af atmerzerhrae as: araeT: ea aaa

aa resarfatard oh Tet, Ta ae, feateqaay waata
TRS RAIA AT, MII | WET—
 316 TECHNICAL TERMS OF SANSKRIT GRAMMAR

aeaistamear fe qa: Har aatiaa |

a agisaara g RAvataaraa:
afer |
No doubt, several MSS. read das kartari niner vidhanat
which is supported by the commentator MaheSvara, still the
Sahitya Darpana would appear to have preserved the original
reading in this particular instance.
Personally I have not the slightest doubt that the whole of
K. was originally composed in verse. It is only on_ this
hypothesis that we can explain ‘‘eva” in “‘tebhya eva hakarah
purva-caturtham ra va” i. 4. 4, “paras tu” in “‘dadha-na paras
tu, nakaram”’ i. 4. 14 comming after “ta-thayoh sakaram, ta-
thaoh sakaram, le lam, ja-jha-na-Sakaresu nakaram”’ etc.,
‘“‘vibhasyete”” instead of “vibhasa’’ or “va” in “‘vibhasyete
purvadeh” ii. 1. 28, “‘sarvatah” in “sur Ami sarvatah” ii. 1. 29
and so on. No doubt the K&atantra, as it is extant in Bengal,
is for the most part in prose, but this is only due to the fact
that the versified Karikas were divided into short prose rules,
and in course of time, due-to certain additions and- alterations
they appeared to have changed beyond all recognition. The
first few rules* of K. may thus be reconstructed in verses-form :
fag ataaiaa aia Aqea MT: |
aa GAIAAT Bt GraealeaET Aa ||
gat gta: TUT Ste: AT SATA. ATT |
arcaatanicite arate. areparter & |
sfaQ awaHtara: | aa Mgemet qu) oem BAA) asl A waMae
aaut 1 gatve:) Ga Ae) @isauasf at) ward saqqutia
alafa aaafa, a aati qa ga ae |
 ADESA.
The word “‘adesa”’ (from:a-dis ‘to point out, direct, teach’)
is used for the first time in the Brihmanas, Upanisads and
Srauta-sutras in the sense of ‘advice, instruction’. In grammar
it means ‘that whichis ordered in the place of another’, hence
‘a substitute’. The followers of Apigali show the distinction
between adesa”’ and: “‘Agama’’ etc. thus :
ATASTTAMAT (ARICA TATA |
AAR TAG A BIT: AAR |
‘An augment comes in without injury (to the original), a
vikara comes in by smashing i.e. ousting the original, a subs-
titute comes in the place of something else and elision takes
place by dragging away i.e. destroying the whole.’
Others put the same thing more vividly thus:
CUT TATATSNT ATS FTSTTTTAT: |
qeavaaAg SIT: ARIA TAT: qe 1
‘An Adega or substitute is like an enemy taking possession
of the domain of a vanquished opponent, an agama or augment
is like the mark of ashes on the forehead, lopa or elision is
like the falling off of the teeth and pratyayas or suffixes are
appendages following one like umbrellas.’
When one letter is substituted for another, it is known as
vikara. This is the view of the followers of Apiéali :
carne fant | anata cenfesetad
Adz |
We find the same thing in the Kaviraja on K. ii. 3. 33.
Bhattoji also says :
faanret ara qorters onda: | seealena:, g: age
 318 TECHNICAL TERMS OF SANSKRIT GRAMMAR

The Hemacandra Laghunyasa (i. 4. 90) says:

mR at vata a ore fant Tt
treme yfta cafe: aaiqazias: |
Panini uses adeéa in the rules “‘eca ig-ghrasvadese”’ i. 1. 48,
“‘sthani-vad-adeso’nal-vidhau”’ i. 1. 56 etc. Agama does not occur
in the Astadhyayl.
Other grammarians use these terms when necessary HN.,
however, uses Virinci’’ for “‘adesa’’ and “‘Visnu’’ for ‘‘agama’’.

Zz. ange faftfa 1 eae

faftfgaen aaa aeqTararad Ufa, aur at fare:
cada a aaa fafigersaa, star
sanrat fry: | x20
yen aera: see Treat wate, ae at fafa:
cada @ aaa fryers |
The older name of ‘‘agama”’ is “‘upajana”’ which is found in
the Nirukta. Patanjali says: upajana a4gamah. ....vikara
adesah (ed. Kielhorn, Vol. I, p. 31, 11. 17-18).
RP uses both Agama and Upajana. Other Pratisakhyas use
Agama only.
The position of the agama is indicated in may systems of
grammar by the indicatory letters t, k and m. Hence in a stanza
quoted in J. (1. 1. 3 Laghuvrtti), a4agamas are referred to as
ka-ta-metah—

aaa: waa a HzRar f= weafer |

Aaa TTA Hered 2aafefx:
Acc. to Panini indicatory ¢ and & show that the agama will
come in at the beginning and at the end respectively. When
 ADESA 319

there is an
indicatory m the agama is placed after the last
vowel of the word, adyantau ta-kitau, mid-aco’ntyat parah i. 1.
46-47. J. follows Panini, splitting up the rule “adyantau
takitau” into two for the sake of consistency:

frarfe: | fara: | qdsat fit eeier-ea

Durgadasa says under Mugdhabodha 21 :
Tad: TIT Araes at waafy |
AATATAaTMUTA: SW TANT: |
SNS STATA) A: THA: TATAET AT |)
K. generally uses “‘vikara” for ‘‘adeéa’’ which occurs once
in “‘sphayer vadeSah”’ iv. 2. 25.
The earlier names of vikara were : vydapatti and Gpatti.
Panini generally uses the ablative and the locative respectively
with the preceding and following sounds between which the
agama is inserted, the 4gama itself being put in the nominative.
Commentators often take great pains to show that in these cases
the ablative is used for the genitive. Other grammarians
generally follow Panini.
R.P. uses the locative with the preceding sound as also with
the following sound which is often construed with the word
‘““pare’’, and the word “antara is used in such cases. Cf. RP.
pe

iv. 16-18 and iv. 84-89.

VP. distinctly says that an agama is indicated by the ins-
trumental :

aeanra: | as (date ae) |adtrar at

frigate ana: sata) ae, “eal areata”
[ aie |, “saat tararfifrar ee” [ eee ] eft) )
320 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Similarly we find AC.:. garzratfaeaa | 2]200

It is clear that in such cases thee instrumental is construed
with the word “‘vyavadhana”’ understood.
TP., as already pointed out uses the nominative with the
agama and “‘narah” with the word after which the agama 1s
prescribed. Thus we find :

agfrrga: THTIIT: | OITA APATT: | ATTA AH:

HOTT: | ARI AITIT | ATS THI TANT: |
‘a|'3-¢
RT. uses the word ‘‘vyavadhana”’
as “‘adhikara’” and the
nominative with the intrusive sound, evidently taking “‘vyava-
dhana”’ not in the sense of ‘intervention’ but in that of the
‘intervening sound’:

ATI? | aT TAtSaTe) «ara fa’

964-229
Panini was probably thinking of the word “Takka’ when
framing the rule “‘adyantau takitau”. He owes his idea of the
anubandha m probably to the fact that the most prominent
Agama which is inserted after the final vowel is the nasal n.
This particular nasal cannot be utilised for the purpose, since
it will involve complications in accentuation (jinityadir nityam
vi. 1. 197), fi and n will involve lengthening (aco finiti vii 2. 115)
and n is required for adeSas. So he uses the remaining nasal
m which happens also to be next to n.
K. uses the anubandha yu for this purpose (agama udanu-
bandhah svarad antyat parah ii. 1. 6). It must have got the
idea from the 4gama num of Panini and others, Sarvavarman
 STHANIN 321

probably thought that since the addition of a vowel was necessary

for ease of utterance, it would be conducive to brevity to use
the particular vowel as an anubandha. No doubt in the case
of vocalic 4gamas after the final vowels the anubandha u cannot
be used, still such 4gamas are rare.
The nasal 7 is used to indicate antadeéa, because it itself
stands at the end of a varga, and because it is generally used
by itself at the end of words.

STHANIN

“Sthanin” is derived from “‘sthana” ‘place’ with the help

of the possessive suffix ‘ini’ and means literally ‘having a place’.
It is used in Aévalayana'’s Srauta Sitra in the sense of ‘being
in the proper place, appropriate’. In grammar it is used in two
senses. As an adjective it means that which should be in the
place, but is not there, that which is to be supplied as in
“‘kriyarthopapadasya ca karmani sthaninah” ii. 3. 14. As a
noun it is used in opposition to adeSa ‘substitute’ and means
the original form or primitive element of a word in place of
which something else is substituted. Thus when bhi is subs-
tituted for as, as is ‘“‘sthanin” and bha “‘ddeéa.”’ Sthana is used
in the sense of place, stead in AB. and the Srauta and Grhya
sitras. This is but a slight development of meaning from the
above sense. Hence Sanskrit grammarians generally explain
“‘sthana”’ as ‘prasanga, contingency, case, event.’ Under “‘sthani-
vad AdeSo’nal vidhau” i. 1. 55 the Mahdbhdsya says:
eurrat fe ara at year w aeafe, enganr fe ara aise
wate|
O.P, 129—21
 322 TECHNICAL TERMS OF SANSKRIT GRAMMAR

The Balamanorama explains “‘sthanin’’ and adesa thus :

eat TAF, | we eatseage fstiat ad ear) aa
ftiaartarad saa Fada @ anes: |
Regarding the first sense Bhattoji says in his §abda-
kaustubha :
Vae-MeA AASAAIATAT FATHCOTTS TTT: |
Later grammarians became chary of the use of ‘‘sthanin”
which soon became obsolete, though ‘“‘sthani-vad adesgah”’ is
used as an atidega-rule in most systems. K. has ‘‘etvam
asthanini” ii. 3. 17 where “‘asthdnini’’ means that for which no
AdeSa or substitute has been prescribed. Su. uses “‘sthanin’’ in
“sthani-tumo dhator icchayam”’ iii. 1. 3. Sam. uses ‘“‘gamyamana”
in “gamyamane’ pi namni prathamo yusmad-asmados ca
madhyamottamau” ii. 17 corresponding to Panini’s ‘‘yus-
madhyupapade samanadhikarane sthaninyapi madhyamah.”
The Nyasa says:

waageakatafa aay | arensifa nat) waza

aafa—reae: catrat Fy eTitsfet @ ensieasaz |
<a fasta geaisifa fe-aezeq-
sfafrafraarar arraae ge: |
ag afr arate: efra) a-
fafa fafeageranaac: wifestSTT, T
ATT FQie9
Can it be that this cuckoo being asked at the break
of day
by the crow that had its speech restricted to the (nom.
dual)
forin of the interrogative pronoun kim, ‘“‘say what are the
 STHANIN 323

two original endings for which tata is substituted in the

system of Panini” (lit. in the Mahabhasya) made his reply in
the form of the sound tuhi ?’ |
At the first blush of the moming the crow saws and the
cuckoo cuckoos. The poet imagines the crow uttering the
word “‘kau,”’ the nom. dual of the interrogative pronoun kin
. with a view to asking the cuckoo what are the two endings for
which the substitute tatan is prescribed in the system of Panini,
and the cuckoo utters “‘tuhi’” in reply meaning that the original
endings are ‘tu’ and ‘hi. The reference is to P&anini’s rule
““‘tuhyos tatan aSisyanyatarasyam”’ vii. 1. 35.

efApargeet fears ater Fatehy aTas: TAI

fe enfragrear
ard Zo ASH HACAHT: Ta: Se i
ATT Oda
‘Ah! making himself substituted by Naisadha and also be-
coming the man Nala for the sake of the end in view (even in
the case of a non-anal-vidhi),
why did Indra, who gave an ex-
planation like that (who had composed such a grammar), retain
his originall evil nature (accept sthanivadbhava which is not
valid in this case) . |
In a previous stanza :

finda mich afoot cat dionad qercnesarat |

aaa Hrasaaigaeral vfs safe Aerarfy 4
Indra had compared Damayanti with the Apsarases Gauri,
Harini, Vinavati, Hema and Menaka. Thereupon Nala ‘had
suspected his identity and looked at him with eyes full of
suspicion. He dispelled those suspicions by explaining that
324 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Gauri, Harini, Vinavati, Hema and Menaka were not proper

names, but meant respectively ‘white’, ‘gazelle’, ‘an adept at the
lyre’, ‘gold’ and ‘no one mine’. This explanation is referred to
by the word Vyakarana in this stanza. Wyakarana also means
‘grammar’. It is wellknown that Indra is the traditional author
of the first complete Sanskrit grammar ever composed.* And
grammar teaches “‘sthanivadadeso’ nalvidhau”’ (P. i. 1. 56) that
a substitute is to be regarded as the original except in the matter
of rules concerning grammatical operation with respect to
single letters (analvidhi). Here Indra is impersonating Nala
and so has made himself the substitute of Nala and he is not
anala also for the winning of Damayanti, consequently sthanivad-
bhava which is possible only in the case of ‘analvidhi’ is not
applicable here for the simple reason that he is not anal. He
should therefore take on the nature of the Adega Nala and
become as pure and truthful as that king. But by a curious
irony of fate, the author of the first and finest work on Sanskrit
grammar violates his own rule and exhibits falsehood, prevari-
cation and his notorious foundness for other people’s wives.

at at | we aay aa a aE Vem afal ame waa ge RWa! alfa

aaa saad | aafed Maa aqati & a elels.o
sais wed WAaNaitcughy: quemanda afeary awauequifar
Sl KLIMT Maw Hal as aT awa TH HsTETE, aa AGaamMaE GOs A
wafafai ae Marafaaiaiiea gaenfas fafa are ci wom! gata
NAAM: | AMAR 1
BUAIT MACY HVAT qeugnitarad sew |
AHMAD Td Geet amatee qufad fe qifwar |
TAIRA |
 UPAJANA, UPAPADA AND UPABANDHA 325

In Murari’s Anargha Raghava (iv. 11. 3-4) we find :

aqifucea araraneatsfa carfaagiaa saan a wef |

UPAJANA, UPAPADA AND UPABANDHA.

‘“Upajana”’ is the earlier name of ‘Agama’. Upa-jan is used

in Vedic literature in the sense of ‘to be produced or to originate
in addition, to be added’. Thus RV. i. 25 8 speaks of the
intercalary month thus:

Az AAT Taal gIzM AATAA: |

dat T STATA |
‘He, the upholder of the law, knows the twelve months with
their offspring ; he knows the month that is born besides.’
In R.P. iv. 84 we find :

qeqvahaqay WHT TIATT |

SA A TATAKS AFM TLSAATT |
Similarly we find in the Mahabhasya :
ag aM gwMTaaakada sama, a fediar feast
vafa | meee adr) eee
Hence “‘upajana” means ‘adition of a letter or syllable in the
formation of a word, letters or syllables added’ and so
corresponds exactly to later “agama”. In this sence the word
is used several times in the Nirukta :
SRAIAAT | «813 ( TIHAGQ—enfsqag)
‘Upa is used in the sense of addition.’
ATH Saaat Aafa—aAda seaa gaa 1 vps
326 TECHNICAL TERMS OF SANSKRIT GRAMMAR

‘There are meaningless additions as, for instance, kartana,

hantana, yatana.’
searfa aqatqaa:—ameag git wastfa ) =e
‘Further there is addition of letters; e.g., asthat, dvarah and
bharuja’.
‘“‘Upajana”’ is used in the sense of ‘birth’ in “‘samanajatiyasya
vopajanah”’ Nir. iv. 20 and “‘Svaropajanas cadrstah padesu
samhitayam ca’? CA. iv. 109.
The past participial form ‘‘upa-jata” is used in the sense of
‘appended member’ in CA. and its commentary:

STATA? WT} vIlKo

‘When a compound is further compounded with an appended
member, the latter constituent is separated.’
fran fe gaeaist PFfag a aA |
TIMI MAAMT FANTATE: STA: wise, ZVBT:
‘The word su here is compounded with the masculine bhisaj
to which (the compound) tama is further appended, hence the
former, i.e., subhisak is to be separated.” Whitney suggests
‘“‘parena”’ for “‘purvena”’ and translates “‘separation by avagraha
made of the latter’.

qara sfafarae sratd act a7 |

ATAAAE: HA THAT res
“When both members are severally separable, the newly
added and the ancient, the latter is to be separated by avagraha
from the former : rksGmabhydm is an instance in point.”
‘“‘Upapada”’ literally means ‘near a word, a word standing
near (another word), a word preceding or following (upocca-
 UPAJANA, UPAPADA AND UPABANDHA 327

ritam padam upapadam. Nyasa). In this sense it is used by

Panini in ‘‘mithyopapadat krfio” bhydse” i. 3. 71, “‘vibhaso-
papadena partiyamane” i. 3. 77 etc. In ‘“‘upapadavibhaktt”
as opposed to “‘karaka vibhakti’’, “upapada”? means individual
‘word’ and the expression “‘upapada vibhakti” means case-
ending in connexion with individual words’. As the Nyasa
says under He. ii. 2. 100: sa hyupapada-vibhaktir yatra karaka-
gandho’pi nasti, yatha Saktartha-vasadadibhir yoge caturthiti: In
its technical sense it refers to a word in the locative in Panini’s
rules under the general heading “‘dhatoh” iii. 1. 91: tatropa-
padam saptamistham iii. 1. 92. Thus since in the rule
“karmanyan”’ iii. 2. 1 the word ‘karmani’’ is used in the locative
it is to be regarded as an “upapada” for the purposes of the
rules ‘“‘upapadam atin”ii. 2. 19, “gati-karakopapadat krt”’ vi
1. 139 etc. |
‘“Upapada” occurs in the first sense in V.P. “yadvrtto-
papadac ca” vi. 14 (corresponding to P&nini’s “‘yadvrttan
nityam’”’ viii. 1. 66). Uvata says in his Bhasya :
aa ate AGNUTeMAaA BE araral vafa aga
fafaaa, a a afafamaie |
A AUTAAITN FIST Taq AT |
SeaT ALAA MAA
ATaAA THTTOT
In another rule in the same section “ypapadaprayoge'’pi
ca’’ vi.-23, it has practically the same meaning, referring, as it
does, to the particles ca va ha aha eva which are used in
proximity with the verb.
‘The commentary on CA. while explaining the rule vrksa
 328 TECHNICAL TERMS OF SANSKRIT GRAMMAR

vananiti vakare 11. 28 uses “upapada” in the sense of ‘the

contiguous word’ :

AIT TTA | LEAT| |TTA arat

FMT TAT. |
In APr. i. 2 “karmopado vrddhiman akdara-pratyayah”’,
“‘upapada”’ would appear to occur to in its usual technical sense.
Later grammarians generally avoid this word of four
syllables, replacing it by yoga, yukta, asyuka, nasyukta etc.,
though sometimes it is used in its ordinary sense. In the
other schools generally the ablative is used with the upapada
instead of the locative of the Panini school, and so they use
“nasyukta” or “‘asyukta” for “‘upapada’’. Thus Sak. says
svakrtasyuktam ii. 1. 22. He. uses “nhasyuktam” in nasyuktam
krta iii. 1. 49 which is explained thus in the Vrtti : krt-pratyaya-
vidhayake siitre nasyanta-namnoktam krdantena namni nitya-
Samasas tatpurusah syat. The Paribhasa ‘“‘gati-karakopapa-
dindm krdbhih saha samasavacanam prak sub-utputteh”’ is read
by the followers of this school as “gati-karaka-nasyuktanam
krdantair vibhaktyutpatteh prag eva samasah.”’
J. uses vak for upapada, probably because vd in this system
denotes “‘dvitiya” and the upapada generally takes the accusa-
tive. Thus for “upapadam atin.” of Panini J. has “vag amin.”
i. 3 85. Cf. also “ipa vak” ii. 1, 98.
Sam. uses “upapada” in the rule “tannimittopapadasya ca”
vii. 53.
Su. uses “upapada”’ in the rule ‘“‘krta tadarthopapadam”’
iv. 3. 42.
 UPAJANA, UPAPADA AND UPABANDHA 329

Rama Tarkavagisa uses “upapada” in his commentary on

Mu. 366: a-syadyantenapi krtopapadasya.
PR. defines “upapada”’ thus:

cere: fafraqeraataassae |
Cc
GATIMAITE ITT | ENR
“‘Upabandha”’ (from upa ‘near’ and bandh ‘to bind’) literally
means ‘that which is attached to’ and is used in the Nirukta in
the sense of ‘suffix’ :
ea: | AAT areata eases: | gic
‘Or the suffix yu is added in the sense of ‘adhiyana’ (he
who studies).’
The sense is not quite clear. It is possible that according
to Yaska ‘adhvaryu’ is’ derived from the root ; precede
d by
adhi with the help of the primary suffix yu. More probabl
y,
however, Yaska intends to derive ‘adhvaryu’ from the substan
-
tive ‘adhvara’ with the help of the taddhita suffix ‘yu’. The
latter is the view of both Durga and Skandasvamin and appear
s
to be the correct view.

afr ar dadagaingquaesaesia qatantay—

ata: aire: eftarat watata: | eps
‘Or else the word ‘siman’ may have taken the
suffix (tas)
in the sense of the ablative and without any other
sense.’
afran:...aft at aafitagadagraramedia e126
‘Or else the word ‘agra’ might have taken ona pleo
nastic
suffix.’
The above three passages set up a strong
presumption in
favour of taking ‘upabandha’ in the sense of ‘seco
ndary suffix’
330 TECHNICAL TERMS OF SANSKRIT GRAMMAR

As has already been pointed out (p. 71) the earlier name for
a primary suffix was ‘namakarana.’

LOPA, LUK AND LUP

“T_opa” is derived from the root Jup ‘to break, hurt, injure’
and occurs for the first time in the Kausitaki-Brahmana. In
grammatical literature it is first found in the Nirukta in the sense
of loss or elision :

aan wafa— wag: safefa | aarts aut

ama waft—acr maf, anf fane:—da
tft, 212
Here ‘lopa’ is used in opposition to ‘agama’ or. ‘upajana’ and
appears to mean ‘destruction’ or ‘loss’. In V.P. we find
“varnasyadarsanam lopah” i. 147, but in TP. occurs ““vinaso
lopah” i. 58 which is the earlier sense and which is preserved
in kurral of the Tolkapiya. RP. generally uses Jumpati (active)
or lupyate (neuter) and so does not deem it necessary to define
the term. ‘Lopa’ occurs in the expression ‘lopa-rephosma-
bhava’ twice (iv. 80, xv. 12) and ‘luptavat’ and ‘luptantam’
occur once each in x. 3. TP. uses ‘lupta’, ‘lupyate’, ‘lupyete’,
‘lopa’ and ‘lopin’ and distinguishes ‘vikara’ from ‘lopa’ in
“varnasya vikaralopau’”’ i. 57. It construes the adhikara rule
“lopah....”’ v. 11 with the nominatives in the following rules
according to its paribhasa ‘“‘ahkara Agama-vikari-lopinam” 1.
23. VP. uses ‘lopa’ and construes it with the genitive acc. to
its paribhasa ‘‘sasthi sthane-yoga’’. ‘“‘Luk’’ occurs in VP. in
‘edodbhyam akaro dug abhinihitah”’ i. 114. in the sense of ‘that
 LOPA, LUK AND LUP 331

has undergone elision’. CA. uses the nominal form ‘“lopa”’

throughout, construing it with the genitive. Panini says
“adarsanam lopah” i. 1. 60 and the Varttikakara suggests
“prasaktadarganam lopah’. The Vydsa-Siksi follows P4nini
and says: lopah syad apyadarsanam 10. RT. generally
prefers monosyllables and so uses ‘lup’ (derived evidently from
the root /up with the suffix kvip) for “‘lopa” in the rules ‘‘lub
anyah”’ 83 (lupyate ’nyo’ karah—com.), and “‘lub ud esa sa sya
vyanijane 156 (lupyata udatta esa sa sya vyafijane pratyaye ite.
pare—com.). It is to be noticed that in both these instances
Lup is construed not with the genitive but with the nominative.
It must, therefore, be explained as ‘lupyata iti lup’. In the
Slokavarttikas “‘sarva-sider dvigog ca Jah’* (Mahabhasya
iv.2. 60), “pramane lo dvigor nityam” (Ib. y. 2. 37) “uvarnal
la ilasya ca” (v. 3. 83, uvarnad ilasya ca lopo vaktavyah—
Bhasya) “‘la” is used for lopa or luk. The Nyasa says under
v. 2. 37: JukaS cayam _ pirvacdrya-vihita samjna. The
Padamafijari says under v. 2. 37: luka esa purvacarya-samjnia”’
and under v. 3. 83 : lopasya la iti piirvacdrya-samjiia.
Panini also uses “‘luk’’, “glu”? and “lup” in the sense of
elision of suffixes (including Vikaranas). Of these élu is
confined to the elision of the Vikarana and the reduplication
of the root in the case of juhotyadi roots and the ¢ has evidently
been inserted for sarvadhatuka-samjfia acc. to “tin-Sit sarvadha-

sual: safe'aeqame Sat SG () wactad:) wre) wale: we

feats eae a waft qa wailed: ) weaq: emee agorq ade:
vaya: | ‘fete’ smata ‘fediq aag’d’ [ siice ] mata aa afta:|
Tease|
 332 TECHNICAL TERMS OF SANSKRIT GRAMMAR

tukam’” iii. 4. 113), though what useful purpose is served by

this samjna is very difficult to determine.
Both “luk” and “lopa”’ occur in the Unadi-sutras :

STONY SIT 220, TaforgR T 2Rz

Lup is prescribed only in the Taddhita section and differs
from “luk” in that in the case of elision of suffixes by “lup”
the word retains its own gender and number acc. to “‘lupl
yukta-vad vyakti-vacane” i. 2. 51, ‘in the case of the elision
of a Taddhita suffix by the word Jup, the gender and number
of the word formed with the suffix are the same as those of the
original word’. The views of early grammarians are embodied
in this rule which is apparently rejected by Panini in the follow-
ing rules, still the great grammarian followed their terminology.
What is far more probable, however, is that the rules “tad
asisyam samjiia-pramanatvat” 3. 2, 53 to ‘“kalopasarjane ca
tulyam”’ i. 2. 57 are really Varttikas interpolated into the sutras.
‘The distinction between Luk and Lup will be clear from the
following rules of the Taddhita section :

gearfeeaiso |KS FR) AATaT | ST a) afragar-

fever | B1R12&2-2e9
Here Lup is prescribed in the case of the words ‘jambw’
etc. because we get the fem. sing. ‘jambth’ as the name of
the fruit. But Luk is prescribed in the case of the other words,
so that from the feminine stem ‘Amalaki’ we get the neuter
‘amalaka’ in the sense of "4Aamalakyah phalam’ after the elision
of the suffix mayat. Hence the Kasika says: yukta-vad-bhave
visesah. The Siddhanta-Kaumudi also says: lupi yukta-vat.
 LOPA, LUK AND LUP 333

Jambvah phalam jambth. The

Balamanorama explains :
lukaiva siddhe lub-vidheh phalam dha.
The distinction between Lopa and the three suffixes Luk,
Slu and Lup is that in the case of the latter the Paribhasa
‘‘pratyaya-lope pratyaya-laksanam” i. 1. 62 does not apply
according to the next rule “na lumatangasya” i. 1. 63. Thus
in the case of ‘agni-cit’ where the suffix kvip has been added
to ‘agni-ci, a ¢ comes in according to “harsvasya piti krti tuk”
vi 1. 71, even though the suffix is elided acc. to ‘ver aprktasya
’’,
vi. 1. 67, because the elision here is prescribed by Lopa. In‘the
case of ‘mrstah’, however, the Vikarana Sap is elided
by ‘“‘adi-
prabhrtibhyah gapah”’ i. 4. 72 where Luk comes by
anuvrtti from
li. 4. 58 and so the rule “mrjer vrddhih”’ vii.
2. 114 does not
apply. It may be mentioned in this connexion
that Lopa is
possible in the case of both Prakrti and Pratyaya,
but Luk, S§lu
and Lup are prescribed in the case of Pratyaya alone*.
Luk may easily be derived from the root lufic
to tear off’
with the suffix kvip, as has been actually
done by Durga in his
commentary on K, ili. 4. 92:

Gad ZH Bw ATA | ereaarixearsy wna fieq |

But in that case the stem will be
luc. and forms like
““taddhita-luki” i. 2. 49 will be hard
to explain. This derivation,
therefore, is no a par with the deri
vation ‘of “dhut” from the
root dhuks with the suffix kvip. Henc
e Goyicandra says under
‘luk cihnarthasya” (Sam. vi. 1):
gfteaqara aa, ta gak:
fkacce
ee er aI, TAT
eng BAF IT Cee
334 TECHNICAL TERMS OF SANSKRIT GRAMMAR -

afe “ofs otfafkaer a” [ afgacfcfire gues] fa

fagargrthaSaT7 |
The commentators of Mu., however, get over the difficulty
by including Luk in the nyankvadi group (P. vil. 3. 53).
It would appear that Panini wanted to have another technical
term like Lup, so he seized upon the first syllable and added
to it the first of the consonants, viz. k, acc. to the maxim:

sear eaaea Thar Hrcararare |

§lu has generally been avoided by later grammarians who
directly prescribe reduplication in the case of juhotyadi roots.
K. uses in the first rule of its second section viz.
“lopa”
samanah savarne dirghibhavati paras ca lopam, but finds it
unnecessary to define Lopa as it is sufficiently well-known (cf.
“lokopacarad grahana-siddhih” i. 1. 23). K. does not use Luk
in its earlier portions, hence it has to add “‘na ca tad uktam”’
in the rule “‘napumsakat syamor Jopo na ca taduktam”’ ii. 2. 6.
At very nearly the end of the book we find the rule “‘lug-lope
na pratyaya-krtam” corresponding to P&nini’s ‘tna lumatan-
gasya” i. 1.63, but that is evidently a later addition by the
scholar who revised the work and used Luk in several rules of
chaps. III and IV. No doubt the word Luk occurs twice in
chapter II: ‘‘kateS ca jas-Sasor luk” ii. 1. 76. and “‘anyasmal
luk” ii 4. 3. but since the author does not mention anything
about Luk at the beginning of chapter II where the other
technical terms used in that chapter have been mentioned, but
explains it at the end of ‘chapter IV where the technical terms
-used exclusively in chapters ITT and IV are explained, it is quite
 LOPA, LUK AND LUP 335

clear that Lup was not used in the original portion of chapter
If. As regards the two rules mentioned above, the second
(anyasmal luk ii. 4. 3) is undoubtedly an interpolation like the
two rules preceding it and the four following it, In the first
place they are out of place in the Karaka-section. Secondly,
Sarvavarman does not deal with these details in his grammar.
Thirdly, the author who gives a wide berth to krt, taddhita
and samasa is not likely to bother about them in connexion
with other topics.
The first rule (kates ca jas-Sasor luk ii, 1. 76) is also open
. to grave suspicion. It is difficult to understand why Sarvavar-
man should use the monosyllable “luk” in ii. 1. 76 and the
disyllable ‘“‘lopah” along with the reservation “na ca taduktam”’
only after six short rules in “napumsakat syamor lopo na ca
taduktam” ii. 2. 6. He might have easily written ‘“napumsakat
syamor luk’’ which would have sounded exactly like ‘‘kates
ca Jas-Sasor luk.”” It may be objected that in that case the rule
“virama-vyafijanadav uktam napumsakat syamor lope’ pi’’ ii.
3. 64. would not apply to words like sas and consequently the
nominative singular of sas would be Sas and not sat
To this
our reply is that “napumsakat” can easily be dropped from
the
rule which itself is of doubtful authenticity, if only for the
masculine singular form “‘virama-vyafijanadau”, the grammati-
cally regular form “‘virama-vyafijanadisu” occurring in ii. 3. 44.
. So Sarvavarman probably wrote:
Raa HAMANN FT ATA |
ADAH SANS TT ATA |:
Frrrqeaqargtaa: Saas hy
336 TECH NICAL TERMS OF SANSKRIT GRAMMAR

HAA AA-TATATT: |...

AT AR] CATS: |
without bothering about “‘pratyaya-lope pratyaya-laksanam”’.
Candra uses “‘lopa” without defining it. Thus he has “ino’c!
lopah”’ v. 4. 41 corresponding to Panini’s ““bhasya ter lopah”’
vii. 1. 88. He uses “luk” for Panini’s “luk”. All cases of
Panini’s ‘‘lup’’ are, however, explained away in the Vrtti by
resorting to abhedopacara. Thus the Vrtti on “‘ilaj dese” iv.
2. 109 corresponding to Panini’s “‘deSe lubilacau ca” v. 2. 105
says: Sarkara deéa ity abhedopacarat. yatha kuntan pravesaya,
yastih pravesayeti. Similarly we find in the Vrtti on ‘‘naksatrair
indu-juktaih kalah’’ iii. 1. 5 corresponding to Panini’s
‘“naksatrena yuktah kalah” iv. 2. 3:

‘cfrarg faaratg warg wciig a” efa aatta

AAHIS A ATAAATTTATLT|
J. uses “kha” for “‘lopa’’, because kha means Siinya which
amounts to the same thing as lopa. For Luk, Slu and Lup,
J. uses Up, Uc and Us respectively (ub-uj-us 1. 1. 71). slightly
varying the order of Panini.
64k. uses Sluc for Luk and Slup for Slu. Corresponding
to Panini’s “‘na lumatangasya” Sak. has the rule “slucigen at”
i. 1. 52. Lopa and Lup are represented here by Luc. Ct.
i, 2. 91-95.
He. uses Luk, Lup and Lopa. In the case of Luk there is
Sthanivadbhava, in the case of Lup there is no Sthanivadbhava
and Lopa is a general term for elision. Thus the author of the
Haimaprakasa says:
 LOPA, LUK AND LUP 337

saad facia: TATATA: TATRA |

aa GH GL aASa aANTAaITHT: |
aq iff) Gh BI Breasat soensaad aafra |
quaqad fais, gf enfragal wafa, of@gaa
aafa qaraa ant | 2 ae) aTfaT ar ar eee? J
fa fagie, “of a enflaxrnseeaiad art
carfafa faat az Tt, FIN) TH saat Ga veee ]
af fagft eartagrararare moara:, 2 aft” ef) ae
Wea J SH STA gaacgead cia AArKaaral STAT: |
qat adver aia: [Tee] efa sq seanfe gaa|
(CF: ee )
Sam. uses Lopa and Luk indiscriminately, though it shows
a decided preference for the latter, because it is monosyllabic.
Sometims Lopa in-the Sutra is paraphrased by Luk in the Vrtti*.
Sometimes Luk in one rule is referred to by Lopa in another.
Thus in ii. 708 we find “‘sanlopa” referring to ‘‘Sanlug inadeh”’
li. 781. P&nini’s Lup is unnecessary in this system, because.it
has the words “prakrtival lingam ca’’ in Taddhita Parisista
1402 ff.
Mu. uses Lup mostly in the Sandhi section and lays down
‘‘Jupi na sandhyddyavidhi” 15 7.e., when an element is elided
by the word “lup’’, the sounds preceding and following that
element, do not enter into Sandhi combination, nor does any
grammatincal operation prescribed with ‘reference to the

*ANQSTHISA 1 IRR
W0; aafyasasaitE aT wale |
O.P, 129—22
338 TECHNICAL ‘TERMS OF SANSKRIT GRAMMAR

preceding sound take place. Corresponding to Panini’s Luk

Mu. has Luk also and corresponding to Panini’s “na lumatan-
gasya” Mu. has “‘luki na tatra’ 94. For Panini’s Lopa it
generally uses Lopa.
Sar. distinguishes between Lopa and Lopas, the latter is a
blend of ‘lopa’ and ‘nasa’ and corresponds to Mu.’s Lup.
For Panini’s Luk, Sar. also has Luk.

amiga aa: | aotfactat Stas |

mae aor amafa, saenachs sfaaenfa a aorfacra: |
Su. follows Panini.
PR. uses Luk for Panini’s Luk, and Lopa and Lopana for
Lopa.
HN. uses Hara for Lopa and Maha-hara for Luk:

aU am amequafa aur ar fafa: saat a aia

araeaa | aa act feat aatq, |oaaiersranraegee:|
arafaneqeqaenc: | after |
For lup it uses smarahara T. 363. ff.

AKRTIGANA.
In the Pratisakhyas we find lists of words where certain
grammatical operations, e.g., lengthening, take place. In
grammar we generally find the most important word of the
group followed by adi or prabhrti in the rule and the full list
given in the Ganapatha*. The list’in the Ganapatha is generally
exhaustive and a vrt at the end of the gana often shows this.

* Sometimes the word is merely used in the

plural,
 AKRTIGANA 339

In some cases, however, the list is not exhaustive but merely

illustrative. Such lists are known as Aktrtiganas. Monier
Williams explains Akrtigana thus : a list of specimens, collection
of words belonging to a particular grammatical rule (not
exhibiting every word belonging to that rule, but only specimens,
whereas a simple gana exhibits every word). The aréa-adi list
(v. 2. 127) is a well-known instance of Akrtigana. Many ganas
which are regarded as exhaustive by Panini and read with a vrt
at the end are regarded as Akrtiganas by his followers because
of the advent of new forms in the language since the days of
Panini.*
The Balamanorama explains Akrtigana thus :
MEA Tagawa Priaatss wor gee) sR.
TATMTGZSATH:
M Wat Afra wT Fanta cfa
aa |
Later critics have often been furious against grammarians
for their failure to make the list exhaustive and having recourse
to Akrtigana. Thus we find in the Nyayamaiijari :

aa safafag acequraAtTatarefanorgesd gar G2

q? ageaad a Gacmahafeagaraareyr sath |
These Akrtiganas come in particularly handy in the case
of Arsa-Prayogas. As Bhaskara Raya says in his commentary
on the Nitya-sodasikarnava (p. 269) :

*Thus though “vrihyadi” is not regarded as an Akrtigana in the

Kasika, it is so regarded by Vamana in his Kavyalamkirasitra:
dhanviti vrihyadipathat vi 2. 57,
340 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Ua gue ASME TTT eAT-

averages Susans fH: |
safe aafratat anfienaaamaer aat:
fag: afaastsftaa aarncoramaatfa: |
‘In the words of the Siddhas in this manvantara there is the
approval of grammarians, because of roots expressing various
senses, because of the mention of ‘bahula’ (in the rules),
because of the prsodaradi group, because of the mention of
akrtiganas, because of our freedom to postulate Unadi suffixes
as it suits our convenience and because of the optional applica-
tion of all rules in regard to the Vedas.’
Murari makes use of this technical terms of grammar in the
following stanza of his Anargha Raghava :
ea: Meant afs watsa AGAMA:
Cay TAIN AMT
SAAT NATAIN ARITA |
wart afrat a anrrad wat fserarco
fre abtigaetoncaafeunadart He: |) eee
‘If these human
sprouts in the bamboo-clamp of Manu
indulge in discussions on Scripture, if the category of Brahmana
be an open one, and if you are to be included in that category,
then fie on this hand of mine on which there are the prominent
scars and callosities due to the drawing of the bow-string and
kuSa-grass and which wields (the axe which is) the most potent
cause of the chopping of emperors as well as of sacrifical fuel.’
The earliest use of Gana in grammatical literature probably
goes back to the solitary
instance in RP. “aditya deva
varunasureti cetyadisu” iv. 91, ‘the words Gditya deva
varung
 UTSARGA AND -APAVADA 34]
asura undergo shortening (of. finals).- before yd etc.’ where
“yetyadisu”’ is used for convenience, the full list
being given in
the first half. of the. next stanza (RRP. iv, 93}. ‘VP:
uses gage
omy -once“in: the rule “uttambgantaligy adj-samsayat”
v. 38,
‘the :words uttambhana and the like. are nor analy
sed im the
Pada-text ‘because of doubts. as to the initial.
of the second
member’ where “adi” is used: in the sense
of ‘prakara’ and
“uttambhanddini’’ merely means ‘words like uttambhana’ (cf,
lakarasya rephah
padam angulim ityevamadinam CA. i. 66
and....vasityeva-madinam ib ii, 29).
These three may be
regarded as very early instances of akfti-
gana and may be
compared with Panini’s “prsodarAdini yathopadis
tam” vi. 3.
109. With the exception of the above cases no
gana is found:in
RP. and VP., and TP. avoids it most rigoro
usly. A fairly large.
number of ganas, ‘however, is found in CA. K. uses them
sparingly, but in several cases Where Panini who
uses gana
extensively mentions all the words, K. uses
ganas. Thus for
P&nini’s “ap-trn-tre-svast-naptr-nestr-tvastr-hotr-ksattr-potr-pra-
Sastrnam”’ vi. 4. 11, K. has “svasradinam ca”
ii. 1 69. Of the later
grammarians Bhoja mentions each partic
ular word in his Svaras-
vatikanihabharana and thus tries to
avoid the uses of ganas.

UTSARGA AND APAVADA

Utsarga (from ut-srj) means ‘pouring out’, ‘setting free’,
thence utsargatah means generally, with
out any limitation. So
in grammatical literature utsdrga means ‘a general rule’, as
Opposed to aepavada ‘a special rule’,
Apavad. (from apa ‘away
from’ and vad ‘to speak’). meatié
ta deny, :to contradict, hence
342 TECHNICAL TERMS OF SANSKRIT GRAMMAR

apavada means ‘a rule in super-session of another rule i.e., a

special rule. In RP. Nyaya is used in the sense of Utsarga:
SIGE ICRC ICC qalaiq* i. 53 which means the same thing
as the later Paribhisi aqarzfaad oftecnrqamisfafatane |
Commentators of the later systems of grammar like the
Mugdhabodha generally use “‘simanya vidhi’” and visesa vidhi”’
for utsarga and apavdda respectively. An Apavada is generally
defined thus: Qa arma at fafaueeaa a aw Aras:
(=aqarz: wate|
‘“‘Prasanga’’ (‘contingency, case’) is a still earlier term for
Utsarga. Cf. ASS. i. 1. 22: Feagreqeizt asians |
In the introductory section of the Mahabhasya we find:
anys: wzat faecal sfavaris:; wa aaa
seq: sfagdem: = fq mareafadisagert sa
HaUett ra AEA Hea: Tet ofaa cl) fH Gata 7
sqanigaral | waardra: qaaqaed: wae: Hugrety-
Rigas: 9 | ATAAa Aa: HaeT: | AT AAT HAV
Tarte ]t aver fata: | ag zat—arsarait
wm [ 31213 J 1
Under “‘nyayair misran apavadan pratiyat” RP. i. 53. Uvata
Says :
aral Saat aerfasvar faaa:; agqarat acafaar
faz: |
Durgadasa says under Mugdhabodha 21 :
aan amarernaen faavafafarat |
azal far wea a arareafafagaq i
aay: erg frat aen a faaafataaa: |
 UTSARGA AND APAVADA 343

Vasudeva Diksita in his Balamanorama -explains Apavada

thus :

ATA ALIAIAAATNT: | TYSR: HUT IT) Az

arama oat fafaameat @ wenger reece |
amreafa
wa wm) Bafa df adiar) «at anatase
aaa | aqnanaane wat at fafraeaat a
aroamian fafreacrasaraenqaral aan xfer are: |
aay vavafag: | sasrzat
fe fea entar: adstare
oe |
ag amare frame feareataaa afasa-
sti cantatas fated aaited anaa, faae-
fafeaara facaaears | faatvare’ fe fates wef
sqaa, feraarat Karta | ATATANTE’ | Araresq-
gaa feaeg sada ofa ae ag wanafe: |) edt fae
aa way) say Weathra—
aan aad faa’ sfatraassta |
Tamavat fas erread way [ fa,
a, v2 J
The terms ‘‘Utsarga” and ““Apavada’”’ are so well-known in
our country that Kalidasa, the master of figures of similitude,
has no hesitation in using them as Upamanas in his Kumara-
Sambhava and Raghuvamsa :
santas sat ad fH aeaaT: |
aya Cara: RATT: qu: i
‘Is it that, you, who attain your dignified position before all
other beings, have now been displaced by more powerful force,
as general ordinances are displaced by special rules? Kumar.
Hs. 27.
 344 TECHNICAL TERMS OF SANSKRIT GRAMMAR

a: Baa Tan fe aA: TTRAT: |

aqag cae aaafaqyyaz: |
‘As a special rule can undo a general one, so any one
belonging to the race of Raghu could destroy his adversary.’
Raghu. xv. 7.
Under Kumara ii. 27 Mallinatha says :
SAAT: aaerafe “art fea” staat | agr-
qea afaftaqatan: | aratasaa xarza: |
Narayana says under the same stanza:
aaa.@: faatafata; saani anneafaa zt) aT
farcfafaty: anneafara: cenafaceasa, aur qaaft
quieaifzemataral wa): fatafatiat qearacm J
qrantatanaaiacaaran fafsadiar” eft arate |
aq mao arafrasat cara: ararafaay | taza
anwar watazaa garzat faaiufeay: | aa ara
aarad: sad gagdste oma watafaazad qed
faactafsn faced | aa afs faaatafaeaeaizty faaaa
ate a facaata qq eral a a faafacanmd aa
aetizay | aarafatseg aanreaty aan wae |
aa fafa teraceay|

ANTARANGA AND BAHIRANGA

Antaranga as a substantive means ‘any interior part of the

body’, as an adjective it means ‘interior’ ‘internal’. Similarly
Bahiranga as a substantive means ‘any exterior part of the body’,
 ANTARANGA AND BAHIRANGA 345

as an adjective it means ‘exterior’, ‘external... In Grammar it

rule is said to be Antaranga with reference to another which is
regarded as Bahiranga when the former is more closely related
to the interior part or base of the word than the latter. Thus
the connexion of an Upasarga with a Dhatu is Antaranga
( MTT ATA: RAAHATSH ), because we find a root changing
from intransitive to transitive, from the Parasmaipada to the
Atmanepada and vice versa in connexion with different Upasar-
gas. Again in the case of a compound within a compound,
the first compound is regarded as Antaranga in reference to
the second and so on.
Dandanatha explainsthe expressions very clearly in his
commentary on Sarasvati Kanthabharana i. 2. 84 -

AS fafaaq arcaeq aqrataq, afeaey az afersrq |

Teed, qt gare) woot afesg) ape
arate, Tard afeteq) seq” SALSA ,
qaqa wea |
Durgadasa says under Mugdhabodha 21 :
afecafafars: eiacate fais |
reararra-aiag eur II
cH eageace hats waa |
THe: WS areratwat aa ||
Similarly we find in the Haima-laghuprakriya
THAT Aa LT AE AT Ya erafeerarry |
aeq areriaiamit weary’ agsae |
rererifrd aa earg afeal azaftaar |
agfe at fafiaife afece’ agead |
346 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Jinendra-buddhi says in his Nyasa on i. 4. 2:

wate -aerqaivary asatefa | saa Ta TVA |
arat fe sreenaicacgii qanensarife 9ararfi
aoa, wag afecgrin usagenaasifa Haatca |
This is based on the Mahabhasya (1. 1. 57):
Ql Aca alaa: fast | sagadl sal weal az
TMT getisd orden ase sfaadtt ara’ arf aad
HUifa, ad: Gea aa: Arafewanry|
Antaranga and Bahiranga are used exactly in the same
senses in Patafijali’s Yogasitra where we find ATHCATH
qaeyq: and azf afer faqineq iii. 7-8. Similarly
Suresvara says: °

arate fe faaret caeaRaaaTe |

afety’ J HA Mag TRAPATT: II
In Sankaracdrya’s Sarvavedantasiddhantasara-samgraha we
find :
afece’ afa: cre seats are |
nafeazaaaageasy fagaat: |)
aracg fe wag afergrg zataa: |
carfraza faareantcaget weaATATA I
Poets have not been slow in making use of these terms in
their effusions :

aaATASA aeat favs GoRaA: |

aed agra 1 afexyy face II
aafaricadaaen aaa fag ferafa oraz |
ft a Safe afeeg harap fafata eet |
 SAMASA

‘Samasa” is derived from as ‘to throw’ preceded by sant

‘together’ with the suffix ghaf. Sam-as occurs in AV., SB.
etc. in the sense of ‘to throw or put together, add, combine,’
¢.2g., Suraqsat aneany AV. vi. 89. 3. ‘let the two ends
be united’, cx @aat aueafa combines the string in séven
ways’ etc. In AB. x. 3 we find aq q2 faacfa-:--- AAAI AL
qd ‘he separates the first two padas,........ he unites the Jast
two padas’, a fafaz: a2 aaetq ‘he should not unit
the two sentences of the Nivid’. The participle ‘“‘samasta’’
is used in SB. in the sense of ‘combined, united. The
noun Samasa is used for the first time in the Brahmanas.
Thus in AB. we find: @@eat q caafagian aArat: qadar
SAra aca: AMaAtaa i. i. ‘Prajapati is seventeenfold, the
months are twelve, the seasons five through the union of
winter and the cool season.” In grammatical literature the
sense of “‘samasa’’ became naturally restricted to ‘the union
of words’. In the Nirukta also ‘‘samasena’”’ is used in the
the sense in which it is used
in A.B. ‘qfgsy
= 9attfi
@ marhr
t
aaqenarar fa a sract amaa’ iv. 27 (i. c. taking
days-and nights together). Samasa is used in its grammatical
sense in Nir, ii. 2: ay afgaamaaaqig wane a
qa qaaquyqt ofearsa fort ara ‘now with regard to
words formed with secondary suffixes and compounds
whether of one or more than one member, one should
explain their component parts in their respective order,
348 TECHNICAL TERMS OF SANSKRIT GRAMMAR

having first divided the words into them’. Samiasa occurs in R.P.
several times in the sense of ‘compound’ AATARG qaqa
wwaqd X. 16, ‘samasas one should separate by an avagraha in
their second mention.’ According to Uvata, Samasa- is
used here in the more general sense of ‘separable words’
(aeaazerrfor qatfa), See also xi. 25, 31. Samasaéngam occurs in
qT Barats AaTA 1. 94, where “‘uttaram samasingam’’ means
‘the second member of a compound’. It is not found in TP
but VP. and the Pratisikhyas of AV use it. Thus in VP.
i. 27 we find “fram fraagqerenran:” where acc. to comm
tators “catustaya-samasih’’ means ‘compounds comprising
four classes’ but ‘catustaya’ may also have been used in a
technical sense. In the Katantra school catustaya is explain-
ed thus:
Tea AIA TA HICH a fore: |
aATaedgat FA A AGE Seas ||
This is not applicable here, as ‘taddhita’ and ‘samasa’
are separately mentioned.
CA. iv has:
VarTaraneeHerW aMNeTaaKahe UTHzTTA: |
cen saga aqua’ Te arareararrent fararareay |i
According to Katre this is a portion of .the text (New
Indian Antiquary, 1. p. 391).
The opposite of samasa is vydsa or vigraha.
Samasa in the sense of a compound occurs in the above
passage as also inqyTegq CA. iv. 43 ‘separation by ava-
graha takes place also between two compounds’ etc. In Head
 SAMASA 349

ava al Atarqerataray iv. 27 the word samisa is used in the

sense of ‘combination’. In the sense of ‘compound’ it occurs
in Apr. iii 3. 5 RT. uses both “samasa” (99) and ‘“‘masa”
(103, 125).. BD. not only uses ‘Samasa’ but mentions its six
varieties also (ii. 105-106) :

faaleraisaaahaa: Sareeca = |

aftrrste falar atari zoee cert|

‘In compounds as well as in a secondary derivation explana-
tion should proceed from analysis: one should explain after
separating (the parts) ; thus ‘punishable’ (dand-ya) as deserving
punishment (dandarha). We have seen that RT. uses ‘masa”’
for “‘samdsa.”’ In the Ganapatha of Panini ‘sa’ is sometimes
found for ‘samasa’, e.g., gTsTa iv. 1. 105, aenyqTa iv. 1. 86.
J. naturally pounces upon the abbreviation and Mu. follows
J. §. follows Panini and K, :and uses the full form ‘samisa’
as do Sam., Su., PR. and even HN. C. uses “ekartham”
abbreviated from “‘ekarthibhava” for samasa.
He, however,
uses the term. ‘“‘samdsa”’ in ‘aTepearenaenfrer:” iv. 1. 149
and in ‘etpredt: SATA at” v. 1. 97, “getsaend ewat-
waTa” v. 1. 132, “asieermre” v. 4. 26, “Tar ama”
vi. 2. 51, “AATASTATET” vi. 4. 39. “SATS: ag”
vi. 4. 66. For Panini’s' firgfraarfaxat gee” ii, 4. 68
Candra has ““fenftraarfeeeporienegq” =k. 15.
Again for Panin’s “frameqmeiad axrrahsaas:”
v3, 88 Candra has “Faegraterirgrer es” and
350 TECHNICAL TERMS OF SANSKRIT GRAMMAR

‘arataeara ara” 3. 3. 56-57. Here Vrtti says: aTst-

AAT: |
Panini does not define Samasa but “ax qafafa:” ii. 1. 1
is generally taken to be the definition of the term. Samksi-
ptasira also does not define Samasa. In the Vrtti we find
CATASARTZTARISAAT” |
aTfinfa: | aT RETA MATA: | UR
RAAT | AAT SATA BHT | VULES
TE | BLAH) Uke
HU | Al BLATT eaie-z
MHz: | BT BIT AAaal age 2212
AAA RSAC | AT BIT aenrilad aaa: |
SIRI
Sqaee: | AIT aITaRTeT aaa AEST alee
aT at: AATa: | (2
aifqaeart: | acqEgt
array | Fea Asa, are
AITAAA | AATAAIraA ATA! Az
BQIAA | AAMAt AAA: | 2W2IBz
TIITaAMS | BAteEaHTaHfasgseagqeat | tH
way a nea) afzearg:—acheqzace: aaa ra} 622
efraramgaa | Sewer fare arfa eemaaifae 1771 |
rita ANT axa Baraqafases i
aagatfefsaaran geaisfer ages:
TH FST HA ITNT WHAAEAANATG: 1
aaa age | wahraqadsetaaaaa aaa ANTE: |
sees aareaarae FAA, 1 RIE Rt
 SAMASA 351

Coming now to the different varieties of compounds we

find Panini begins with Avyayibhava (ii. 1 .5) and’ passes
successively on to Tatpurusa (ii. 1. 22), Karmadharaya (ii. 1.
49), Dvigu (ii. 1. 52), Bahuvrthi (ii. 2. 23) and Dvandva
(il. 2, 29). Of.these Avyayibhiva is self-explanatory, since in
this class of compounds the resulting word becomes an
indeclinable (Avyaya). This suffix cvi shows “abhita-
tadbhava”—that which was not an Avyaya now becomes an
Avyaya. The name is so apt that most systems of grammar
(including Sak.) have accepted it. Candra, true to his promise,
cannot use it. He uses Coppa” instead. It may be be
mentioned in this connexion that C. uses Asamkhya for Panini’s
Avyaya. Thus for Panini’s aeqepaqa aIcaNwatasay: v. 4. 107,
Chandra has greatfaettsd wart iv. 4.90. (For
asaya
aTHTe vi 3. 81, Candra has spt eaqpey_ vv. 2. 100).
J. uses
“ha” for “‘avyayi-bha’’, probably because “‘ha’’ is
the last letter of the alphabet and ‘‘avyaya’’ marks the last
stage. Mu. selects the “v” and final “a” of “avyaya”’
and
uses 46
“va” for “avyayl-bhava’’. HN. sees no reason to reject
the generally accepted name.

Glo | AeqaATa 1) AL's Mo ghee

Ho) qa’ are way Wer Assay ees | RIROR
Hol Ss) Vale
Mo, sera) mat Ty) ages, fit qemi
FECT A ASG ZESeTAATT: | VIE
go aatara fhuata sedife aato gecadtare:|
SIQIR&
352 TECHNICAL ‘TERMS OF SANSKRIT GRAMMAR

qo | Maas aeaqeasqT asa: | OK

Hol a: h Aaiereaewameagnats |
PA TAAAATEN TATA AHA |
AM SAANT AMAT AHA |
Glo | Yarsersrara: ) S123
Gol wera: |) 333
Te] aise aa atarfearfeetHaqayt | cee
Hol AsTAATa:| Lae
Tatpurusa is taken as the type of the class of compounds
known by that name. It is itself a sasthi-tatpurusa, the ana-
lysis being “‘tasya purusah”. The Prakriya-sarvasva says:

CA ATIGSIEKTST THAATAA: |
aa atarfasn: Vi Haq ATSA Aa: 1
‘TET ger: becomes TST: by qstaata:, so all words
of this type are regarded as Tatpurusa, even as all suffixes
of the type of Krt are regarded as Krt.’
J. calls it ‘sa’ after the final syllable (i. 3. 20), Mu. follows
J. HN. calls it Krsna-Purusa—hardly a happy designation. C,
generallly expresses Tatpurusa by Ananyasamkhya i.e., except
in Bahuvrihi and Avyayibhava. Cf. ‘‘asamkhyaccanguler-
ananyasamkhyarthe” iv. 4. 74 for Panini’s “‘tatpurusa-syanguleh
samkhyavyadeh”’ v. 4. 86.
The word “Tatpurusa” occurs first in T.A. x. 1.5 ‘aaqeary
faae uzitara “tafe in the sense of ‘the Supreme Being’
and is a Karmadharaya. As a Tatpurusa it occurs for the
first time in Katyayana Srauta Sitra vii. i. 8: aac aTaaecatar-
 SAMASA ms 353

gear Tat TATATA ‘or the first four, the others are their re-
tainers according to the respective Vedas’.

Gio} AeGasy 1] RII

wlo | acgerraat| [ sat fagantarcatJ
faa faatarat are actaea Z|
auerd Aria fe AataeqEsq: T Ty) 2V"s-2VeeE |
Sto} TL wer wafa st aarefsaret sa: | th:
Qo | faACAMMATEMrataATeT | xI1g120
fram (fa=afa) ga gq afe me = aaa
aaa garata aa fad anera aa anrensat
aemiantzenm teceageraat wate; anata: |
Go| Age! ArRw2e
Go, Agel
a Aw: Aaa: | 917
fo | wWhameaereqey: | 317182
yo| &: HeMarista: HoUrisra: |

Alo | HATE ATST: Ae

Teo | ATTAAA TATA BAT AT EIT: STAT: | Boos
fo | Tava Wh ARTS: BUNGE: eI1-
Next comes Karmadharaya. The word must have been
self-explanatory at one time, but the exact sense is now
shroudéd in obscurity. The Prakriya-sarvasva explains:

gate wratageaihas 27 |
fraracafet Ta MTA Mra: HAP ANTTA: ||
‘Where the entire word becomes connected with the verb
by means of the same action of both the ‘words, the Samasa

O.P, 129—23
 354 TECHNICAL TERMS OF SANSKRIT GRAMMAR

is Karmadharaya.’ Abhirama Vidydlamkara says in his

Samksiptasara-tippani: Karma kriyanvayam dharayatiti
Karmadharayah. Wackernagel mentions two views in_ his
Alt-Indische Grammatik : Karmadhadraya means the subject-
holder, as opposed to Bahuvrihi in which the second member
pertains to another subject, or it may mean (ii) owing an
object i.e. having no object to show forth or exhibit.
It is just possible that in ancient systems of grammar,
Karmadharaya was distinguished not from ‘Tatpurusa, but
from Bahuvrihi. Thus instead of the rule aeqyay: arava fit-
HcUy: BU ATT: there might have been such a rule as aeaite-
SUCG sy Alt: Hy ATT. In that case Karmadharaya would
mean that which holds i.e. is connected with the action, as
distinguished from Bah. where the action does not affect the
members of the compound but something else. Indian com-
mentators generally explain Karmadharaya in this way.
Another view may be suggested here. Karma is a masc.
substantive ending in an, accented on the final syllable,
meaning ‘doer’ and ‘dharaya’ is another substantive formed
with Sa meaning ‘holder’. The word Karmadharaya is a
Karmadharaya compound of ‘karman’ masc. and ‘dharaya’:
karma casau dharayasceti ‘maker and maintainer’.
Or karman may mean ‘sense’. In the Nirukta we find
qaimtaaa having the function of the ablative, SeaHaiery
“Taq: ‘Toots. denoting to make sound”. qe GTsUey AATS:,
acagiqencia afr aaraarcr: eq, ata GAra: |
HPaTH: |
In the system of P4nini it is possible to take ‘‘karman”
 SAMASA 355

in the sense of “samanadhikaranya” ‘agreement or apposi-

tion of the members’.

aarafray cay afiaxc aed asi qari ahr

aartifiacontas wart | arta ATA TET AT EIT
aq BAsRcTgzeAqgE: | ata aqgearatal qari
warifsacnagrnin aged afrag amarfanza-
wedare efa aarafa) at qaalenraqerter AT RIET
Ft aarnifsrancy arate aalrararer TENAIGTS 7
araree eff ager: dara BcTA
ve fe eraarerara darned otaent dart adearai
ara eft wedi dat gdaseitsaa afise Redz:
daratieata: | ta Moqertd<? ademcaq eft
qereT aarnfsacmer gett sre afiqdiaer get
faarat | Fate: |
an feat areas qed sqqerqadihr HAA: |
arfeaufosreaaeaaa T: | THPTTATST Epo HATA
faniead, aTcrdiceanm: afr) QAMMAWe aeqy
anaaTaqzrafafemqna sarcafF eqaeagafa) = aaq-
THIAAT |
J. and Mu. call it ya from the last syllable. H.N. calls
it Syama-Rama.
Glo | aeqea: AARC: HATTA: | g1RIR
Bro} Ga geathaacat fasta: saree: |. 21283
Ho | qanenainngumgtad aaa eras
ato | faataot afraid aiartaar | 2yeiac
356 TECHNICAL TERMS OF SANSKRIT GRAMMAR

aes farang; da fer | aa aera a araag-

eqfaa= apart war anrnfsncdt gad fare
afaat garda ae al wae, a a aATAageiad:
anaTadaa vafa | waranafa:|
Zo, faaigat fastetorara’ HATTA 31218
ae, faarmer faa: Hae: | GIce
qo| faunal A-2-4-T-77-a: |
le | RAITTGCAT | eI F0
go) fatuuaaaa aera aezA) elec
To, ward qearhrncoraara: HATTA: | WSs
go, faguet qearfncdia , Harari oH Aaar:
AMGOIAA: | ATT FATATIAGA | e19-%
Now we corne to Dvigu which means ‘bought for two
cows’ etc. and is itself an example of “‘taddhitartha dvigu’. It
is taken as the type of all the three varieties of Dvigu com-
pounds and 1s a generic term for those Karmadharayas in
which the first member is a numeral. The _ Prakriyasar-
vasva explains the term thus: geapaaniatfafe fea aaaa
fe3T: | J. fixes on the ‘ra’ of ‘samahara’ and calls Dvigu
ra’, Mu, probably thinks it rather far-fetched and names it
‘ga after the final syllable. HN. substitute a samahira
dvigu for a taddhitartha dvigu and uses “‘trirami”” (Rama,
Parasurama, Balarama) for all dvigu compounds.
Ge | [|aeqges, eararfisnzo: rarer} ze he |
eaearqat fag: 1 xe
ete | [ot genfirncct fasia: waar: |] dearqa’
faafefa sta: mee
 SAMASA 357

sto | [|fanfea qean eixiea [ fer: aura TNA |

Riglse
feq: emer qaedaracwwga ae st: a: [ aqqey-
aara:] aafa, aa [ sasrcag ] amet art age a
[ @=aar, ga=afga, a= saa] cesar? |
Go| Awl aA a feyarareaqay | write
ao} afeal anrer fey: dena) sare a}
CAR Gian he
fo) Beat aarEett
a fear) 31e183
Wo, asa fava ara aqeariftnanfzar|
MPR: GATE aspaA gq ||
Ato | @aaret
wa ta fem) aie
gol fag: denafgartaarmne?: | ga? ay ia
O'2-\9FG
qo| wearge: a fem: aT ansee
Zo, dearqatsal factda: 1) <yo
Bahuvrihi means literally ‘having much rice’ and is
itself an instance of a Bahuvrihi compound. The Prakriya-
Sarvasva says : qzay areaiseafa Ga SNA Bawm=ag | J. calls
it ‘“*ba’’ after the first syll, and Mu. “ha” after the last. Candra
uses “anyartha” for “bahuvrihi”. HN. substitutes Pitambarah,
an epithet of Krsna, for Bahuvrihi. Pitambara is chosen
because it is itself an instance of a Bahuvrihi compound.
Qlo| AARAAWagy! 2X21
Hlo | eral afe ws g af ar eqeeertt|
ATURE TREAT TEAR: | AIRES
3458 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Ro, Waramessas aq) w31¢e araafaae [saat

aia] aeaa qeenasfaasam gard aes: a:
[ aura: | wafa |
ao} ater agefifa:) aiaie, aearrereyf 311g, aeta
AGIT | BIBS
Zo) Gears dea Aeaa Gera agaife: | B1e1ee%
o, waar aealfe: | fRatarraeaqaey Fearaya |
SIRRV-ZRZ
qo. fraraareafzderreardtat ATTA: 34%
Go, adaacaqaet agaife: | elanse
ate | agaifereary) ele
Jo] Garang? afece aque |
fret aq a agate: eis,
To | AAHAATI TaeaT: | Z1Zo%
Dvandva is the reduplication of dva ( gy gq )and means
two by two. two and two, hence pair. It was naturally chosen
as the name of those compounds in which the majority of
instances denoted pairs. The word Dvandva occurs first in
the Samhitas of the Yajus school. Thus we find in TS. i.
6.9.3 ggg aran daeaat aeatfz zargeraraat: “The
year bas twelve months, there are twelve pairs of new
eS
E

and fullmoon sacrifices.’ Of the works belonging to the R.V.

it occurs first in AB. ix. 3: TIT a fazacat uM dial Jara
MAT MTT GHArATat fAatTar gaa, awa MNT ssH |
‘(The cups) for two deities are the breaths; they are’ drawn
in one vessel: therefore the breaths have one name. They
 SAMASA | 359
are offered in two vessels, therefore the breaths are in pairs.”
Keith. Even J. and Sak. have retained the name Dvandva,
but Mu. is not inclined to sacrific consistency and so gives
the monosyllabic name ‘‘ca” to this class of compounds in
pursuance of P&anini’s definition ‘carthe dvandvah’. HN.
names it Ramakrsna and frames the rule : TATATAAMNTAT
TlH: vi. 117. The singular in Rama-Krsna_ is jarring
to the ear and as Rama-Krsna is neither an instance of
itaretara dvandva (being singular) nor one of samahara
dvandva (being masculine), it is not quite apt here.
Dvandva as the name of a variety of compounds occurs
in VP. aamaeaife ariaftaarta ii. 48, gc" araraqg
qafmagg ii. 55, qa gefeeargg iii. 127, gegrhy fzaqa-
arafa marcragaqathr v. 28 and CA. @qatgre
a1 iv. 49.
Gro} ~ardtseas i ( Cl. creigs)
lo | ea: AAMT Aad gat ahs at vag | AIRES
Sio | areas, waies
Slo | gS: GERRI zLICo
wey are graghiancnrararat ag: | aiere,
fo} art ars: Gets! zULIEe9
eo] galseatrTaa? sere
Sle | AMT ges) ele
Ho | KAaSTt AAA a
?T& aAfFaT | gx0
No | ATT gra) VCR
Ko] LALATAATEITAT TASTY) eer
360 TECHNICAL TERMS OF SANSKRIT GRAMMAR

The order of the different compounds in the different

systems of grammar is as follows :
P. Avyayibhava, Tatpurusa, Karmadharaya, Dvigu,
Bahuvrihi, Dvandva.
BD. Dvigu, Dvandva, Avyayibhava, Karmadharaya, Bahu-
vrihi, Tatpurusa.
K. Karmadhdraya, Dvigu, Tatpurusa, Bahuvrihi, Dvandva,
Avyayibhava.
C. Avyayibhava (2), Tatpurusa (11), Karmadharaya (18),
Bahuvrihi (46), Dvandva (48). Candra does not make any
provision for Dvigu. See Vrtti on ii 2. 18.
Jai. Avyayibhava, Tatpurusa, Karmadharaya, Dvigu,
Bahuvrihi, Dvandva.
Sak. Bahuvrihi, Avyayibhava, Tatpurusa, Karmadharaya,
Dvigu, Dvandva.
S. K. Avyayibhava, Tatpurusa, Bahuvrihi, Dvandva. S. K.
has no special name for Karmadharaya and Dvigu, the
generic name Tatpurusa being used for these also.
HC. Bahuvrthi, Avyayibhava, Tatpurusa, Kurmadharaya,
Dvigu, Dvandva.
Sar. Dvigu, Avyayibhava, Tatpurusa Dvandva, Bahuvrihi,
Karmadharaya.
Sam. Tatpurusa, Avyayibhava, -Karmadharaya. Dvigu,
Dvandva, Bahuvrihi.
Mu. Dvandva, Bahuvrihi, Karmadharaya, Tatpurusa, Dvigu,
Avyayibhava.
Su. Avyayibhava, Tatpurusa, Karmadharaya, Dvigu,
Bahuvrihi, Dvandva.
 SAMASA 361

PR. Karmadharaya, Dvigu, Tatpurusa, Bahuvriht, Dvandva,

Avyayibhava.
Karmadharaya Aadyah syad Dvigus Tatpuruso’ parah.
Bahuvrihiratha Dvandvo’ vyayibhavah sadiritah 1. 725.
HN. Karmadharaya, Dvigu, Tatpurusa Bahuvrihi, Dvandva,
Avyayibhava.
Yaska (ii, 3) distinguishes between “eka-parvan” and
‘“anekaparvan’” compounds. By ‘‘eka-parvan”’ he means ‘com-
pounds of two members’ and by “aneka-parvan’”? compounds
of more than two members’ (Cf. CA. iv. 77). The examples
given by him are raja-purusah, viScakadrakarsah* and kalya-
Na-varna-rupahf. On “rajapurusa”. was founded the name
Tatpurusa by the saving of a syllable, but. rajapurusah has
remained the stock example of Tatpurusa compounds from
Yaska’s time down to the present day.
Acc. to some scholars Panini deliberately used Tatpurusa
with a view to including Karmadharayas also: Wear:
qaiamdarn agate gst ata gew: a areal qeaqata
aqaaziecoazaa a vata) eTTErfcort 312133
It will be noticed that Panini starts with Avyayibhava
in which the sense of the first member is prominent, then
passes on to Tatpurusa in which the sense of the last member
is prominent, then treats of Karmadharaya which is a sub-
class of Tatpurusa and Dvigu which is a subclass of Karma-
dharaya. Then he takes up Bahuvrihi in which the sense

*‘Attraction of a dog’. “ViScakadra” appears to be the earlier

See
| form of “vi§fvakadru”.
+ One whose complexion and form are auspicious.
 362 TECHNICAL TERMS OF SANSKRIT GRAMMAR‘

of none of the: members is prominent and concludes with

Dvandva in which the sense of each of the members is promi-
nent. Bahuvrihi is closely connected with Karmadharaya
and Dvigu in that in each of these classes of compounds, the
members are in apposition with each other, hence it is placed
immediately after Dvigu.
Padmanabha, Candra (who does not make provision for
Dvigu) and Bhojaraja (who ignores both Karmadharaya and
Dvigu) follow Panini.
Since in the vast majority of compounds the sense of the
final member is prominent, the Katantra starts with Karma-
dharaya where the members are in apposition and proceeds
to Dvigu which is merely a special case of Karmadharaya.
Then it passes on to Tatpurusa where the members are not
in apposition with each other. The mention of Karma-
dharaya naturally leads to Bahuvrihi in which also gene-
rally the two members are in apposition with each other.
Of the remaining two, Dvanda and Avyayibhava, Dvandva
being a savyaya (declinable) is placed before Avyayibhava
which being an indeclinable is dealt with last of all.
K. is followed by PR. and HN.
Sak. and Hc. start with the anyapadartha-pradhana com-
pounds and proceeds on to piirvapadartha-pradhana, ultara-
padartha-pradhana and ubhayapadartha-pradhana compounds.
Mu. follows the Bhagavad-gita where the Lord SayS: geg
alatfaner a ‘of the compounds, I am Dvandva’ and accords
the place of honour to Dvandva and gives the name of the
two deities generally worshipped as an example : MHariharau.
 SAMASA 363

‘ATA AAA Bute: FACAHAT EL: |

eae
aq waar facar [ fafas amity] tazer aren
feraeea:’ || qarfrarate: |
This a samasyapirana stanza on Katantra’s definition of
Samasa but the reading is corrupt and the sense obscure.
‘The collection of the well-known thousand names of Hari
along with (a knowledge of) their meanings destroys all sins
of created beings. The distinctions between these different
names must be obliterated.’

gear fafa are nants frearncaeftara: |

ATCTST HA WT Bare eat-agaifE: 1
‘We are a couple and I have two cows. In my home there
is perpetual absence of expenditure. Act in such a way,
sir, that I become possessed of a good deal of rice.”’
This is the form in which this stanza is current in Bengal.
The Subhasitavali reads the first line thus : fact Wsraise
We AR Baansazsrara: | —«wPeterson translates: Though I
have but two cows I have a wife to support, and in my
house the outgoing of money is unknown (because there
is none to go out): therefore, O man take some steps to
make mé rich in rice.

acre fare wifa aaraariad

a27q |
Tala CATCH TA TATAfae Te |
aagattafagatata yeatsfer aes: |
AST HAAN WHAATALATT: 4)
1 shall speak of compound words and their dissolution,
 264 TECHNICAL TERMS OF SANSKRIT GRAMMAR

to remind (the reader) as it were (of the fact) that all things
shine in epitome in the body of Krisna.’
‘I have got a mate and Iam greedy merely of two cows
along with a vast quantity of grain. O lord, act in sucha
way that my devotion may not be fruitless.”

qglaqed cat agaife’ wea: |

TALIS TAT HAA TAAAANT II
After the image of Ramesvara had been set up at Setu-
bandha, Narada asked Rama, Siva and Brahma about the
name of the compound. Rama said: Sasthi-Tatpurusa (1.e.
lord of Rama), Siva said: Bahuvrihi (he whose lord is
Rama), Brahma said : Karmadharaya (Rama and at the same
time Siva).

a Aad a User Haarafsy7 |

aga faces Ua Widget Aart|
ay TTA: |
‘Both I and you, O great king, are lokanathas. I ama
Bahuvrihi (he who has the world for his master), while you
are a Sasthitatatpurusa (he who is master of the world).

are afaat: net sfraaisaifass earfaa-

mai aq waged aad oplaararhraq|
usr a ag fafa at tag Tat
ararrararqare fracas aafea ara: af: |
SEATAVTT: Qe?
‘Women return unpleasant replies to their husbands in the
Kali age. The epithet pati-devata when applied to them is to
 SAMASA 365

be dissolved as a apfraqarer (the deities of their husbands).

In them there is neither shame nor virtue nor fear. Acting
according to their own sweet will they turn their husbands
into (domesticated) bulls led by the nose. Blessed is Kali!

Sua Bat: Berar

gtRraneaanr: fat afer |
wMeaaadtazt: AATAAT-
fand Tara: eT A |) Frew: eviae
‘The officiating priests conversant with the rules of
grammar ascertained
by means of accents the correct
dissolution of compounds which though far different in their
fruits had assumed the same form.’

QUTTMASISTHETA BIITAT-
aaaKe at aes yay aay |
aqarag aearee: egefa afsataraafzar
anata fact a sa faazt Risty Jey: |

Panditaraja Jagannatha himself says:

aa aAqaravenncame | aa fe “aq } af}

adafa fa me” ceafzat seqenal sfacazaenat
afafcfa cag faaftaq, azarrastensisdtacadzarty
<aftar | saztear a afcon—
anata: at TIAA |
ataarta aaa at deat areatt fire: 1
aft | arreaatiarshraaranared qian |
 366 TECHNICAL TERMS OF SANSKRIT GRAMMAR

‘Glory supreme to the Indescribable Being—He who shed-

ding all selfishness out of his solicitude for the interest of
others bears identity without distinction of number to those
who are subordinate, in whose heart greatness noble and charm-
ing shines forth spontaneously and who is ever able.’ The
dependent clauses support Samasa also. Acc. to one view
compounds are jahat-svartha i.e. they give up the meaning of
their constituent members and take on a new meaning. In
compounds the sense of the members is subordinate to the
sense of the whole which is one and not different from the
sense of the members. And, as a rule, the final syllables of
compounds are accented, i.e., have the acute (udatta) accent.
And a compound is always formed of words which are capable
of being compounded (samartha).

ara
feanrestrgrs
Prorat reRras iat |
¢ ¢ e

SULIT saTRTOT RAAT II

aaifiarata: 3¢3
‘She is overtaken by loss of qualities and of growth and
of complexion and by the advent of antitheses like joy and
sorrow and by misfortune. She finds it difficult to walk,
to utter words and speak coherently. She is thus like the
Operations of grammar where we find the phenomena of Guna
and Vrddhi, elision of letters, Dvandva compounds, Nipatas
and Upasargasand where diflicult words and sentences and
senses are dealt with.’

AMAIA AAS | |-BITITT, £33

‘Without a mate like words in their uncompounded state,’
 KARAKA
The word “Karaka” occurs for the first time in Panini.
It is not found in the Nirukta nor in the Pratisakhyas.
It
is formed from the root kr ‘to do’ with the agent suffix
nvyul
and means literally ‘doer’ or ‘agent’, but as the good old
word
‘Karta’ was used in this sense, the sense of ‘Karaka
’ was
generalized and it was used for ‘case’.
Panini used Karake as an adhikira rule (i. 4. 23) withou
t
defining it. The author of the Varttikas discusses the meaning
of Karaka under ‘Karake’ i. 1. 23 and shows that
other cases
also may be used as the agent, consequently
the etymology
“Karotiti karakam” applies to them also Thus in the sen-
ence “Devadattah pacati”, Devadattd is the
agent, since he
places the pot on the oven, pours water into
it, puts tice into
it and so on. When we say “‘sthali pacati’’
‘the pot cooks’,
the pot which should normally be in the Locati
ve is the: agent
and the sense is—the pot cooks a drona of
rice or an adhaka
of rice. When we say ‘the fuel will cook’ “edhah
paksyanti”
the fuel which is normally the instrument
is used as the agent
in the sense that it will burn till the
rice is cooked (softened)
and so on.
Later grammarian generally use the word
‘“‘karaka”. J.
uses the, initial syllable “ka” for “kdraka”
just as it uses ‘‘sa”’
for “Samasa” and Vopadeva follows J.

So, finarfirfid area cities

firarfreeRrercot waif srenee aah | afta: |
TTA: |
368 TECHNICAL TERMS OF SANSKRIT GRAMMAR

HICH | ‘lke
Zo | frraraeataana

fo. fHared: HITT) AIR

Ho| Aa HH) wees TAAT St KAM: ||
In his pratyahdra or rather samahara sitra Vopadeva has
dha gha bha ja da da ga va. He begins with the
jha dha
fourth letter of the varga because Karaka occupies the fourth
chapter in his grammar and goes on till the sixth letter is
reached. He does not like jha which is omitted.
Coming now to the various cases, their names are practi-

cally the same in all the systems of grammar except Candra

and Mu. The nominative is Karta which means exactly the
same thing as Karaka but is the older word going back to
the Rgveda. In its technical sense it appears to occur first
in Panini. |

Mu. wants to denote the cases by means of the sonant

aspirates. Since Mu. treats of the karakas acc. to their case-
endings and since acc. to Sanskrit grammarians the Nomina-
tive and the Instrumental take the third case-ending while
the Accusative takes the second case-ending, Mu. places the
Acc. before the Nom. Sar, follows Mu. And _ since of the
Nom. and the Inst. the latter has already been mentioned in
connection with the object of the root div in a previous rule
the Inst. precedes the Nom. The Samahdra rule reads:
(jha) dha dha gha, hence dha stands for “Karman”, “dha”
for “Karana” and “gha”’ for ‘‘Kartr.”’

Acc. to commentators “‘tantra’’ is used in the sense of the

agent (Rat) in the Unadi rule erafrrgea fieg fares firarz rgeal
 KARAKA 369

aaqeq 1. 159. Here ‘tantra’ is evidently an abbreviation of

‘svatantra’ used in Panini’s rule waza: Har Orwara is Hat
because he is tqqura, Adea is amRal, because he is 31s7=q197.

io | ara: Ha | AMAT TTT] Carasak

| a aif a aay) aes a a eg |
RIBIYB-V
[ao matt qatar, 2121e2]
Ho | Aaeal Har | ATAIAHT TAT Rg
e- ee
Go| Adea: HA) AT qalaHisaTHeT ar Faqs
AATTATA HAN FeAISHAHRAT CTHAT | LIVIRB-3e
Zo | Mara: HA! Az
Fo, o Rar TT Ve
eo, fraqeaaaial har} le
Jo, araaqarsaay RAT | AU
so | fenfaal a: eaaea:
@ Hai) RAAT Fa:
mat Sega 1 eaR- RRR
Fo, Wat AINA HT 1 vlez
The next case is Karman, the Accusative, which in its
Original sense of ‘act’ ‘action’, goes back to RV. It is used
in the Nirukta as also in some rules of Panini in its earlier
sense. From meaning action Karman came to mean the
object towards which the action is directed and in this sense
it appears first in Panini. The Katantra also uses Karman.
J. uses the same technical term defining it as ‘‘kartrapyam
karma” i. 2. 145. The word 4pya in the definition appears to
have caught the fancy of Candra who uses Apya or vyadpya
O.P. 129—24
 370 TECHNICAL TERMS OF SANSKRIT GRAMMAR

for karman, sapya for sakarmaka (i. 4. 106) and anapya (i. 2.
97, i. 4. 91 etc.) for akarmaka. Sak. also uses “‘sapya” and
‘“‘anapya’”’ for ‘“‘sakarmaka”’ and “akarmaka” respectively. For
Panini’s bhaivakarmanoh* (1. 3. 13), Candra generally uses
bhavapyayoh (i. 1. 78, v. 3. 73). In the later portion of the
Candra Vyadkarana, however, ‘‘bhava-karmanoh’’ makes its
appearance only once in the rule ‘“‘abhava-karmanorano ye” v. 3.
168 corresponing to Panini’s ‘‘ye cabhava-karmanoh”’ vi. 4. 168.
Just as Katyayana and Patafijali sometimes use Samprasarana
and sometimes Prasdrana, so Candra sometimes uses “apya”’ en
e

and sometimes “vyapya”. Thus in two consecutive rules

in Chapter I we find “veh Sabdapyat’’ i. 4. 80 and “avyapyat”
1. 4. 81 and in the next rule but one we find “‘apyat’’ once
more: kartrsthamiurtapyat i. 4. 83, follows the principle
paryaya-Sabdanam Jaghava-gaurava-carca nadriyate. Hc. uses
“apya’’, “‘vyapya’”” and ‘‘karman’’. His rule is ‘“‘kartur
vyapyam karma” ii. 2. 3 which he thus explains in his Brhad-
vrtti: kartra kriyaya yad vyaptum isyate tad vyapyam, tat
karakam karma-samjiiam -bhavati. §Prasiddhasyanuvadena’-
prasiddhasya vidhanam lJaksanarthah, tena yat kartra kriyate
tad vyapya-samijfiarn bhavatityapi sitrarthah. For P&anini’s

* Some grammatical operations take place exclusively in the ra

and aeqrays, so later grdmmarians bent on securing brevity at any
cost looked for a convenient technical term for these two taken
together. The Jainendra system hits upon fe ( wana fe: i. 1. 31)
which was probably suggested by the «& of ae used in the
Paninian system for Atmanepada. Similarly Jainendra uses fy for
waaay root. This is merely a variation of |y which is used for
roots.
 KARAKA 371

karmanyadhikarane ca”’ ili. 3. 93. Hc. has “vyapyad adhare”’

v. 2. 88. For ‘“‘samana-karmaka”’ He. uses “‘ekapya”’ v. 4. 74
and for ‘“‘akarmaka’’ he uses “‘avyapya’’.
Vopadeva for some reason does not like the aspirate jh
even as Englishmen do not like gh and so passes on to dha
which is the name of karman in the Mu. For ‘“‘sakarmaka”’ and
‘“‘akarmaka” he uses “‘sa-dha”’ and “‘a-dha”’ respectively.
Sarasvata uses karya for “‘karman”. Thus for P&nini’s
‘“‘kartrkarmanoh krti’? Sarasvata has ‘“‘kartrkaryayoraktadau
krti sasthi’’ viii. 7. Cf. also: Sesah karye kartrsadhanayor
danapatre vislesavadhau sambandhadharadheyabhavayoh viii. 5,
smrtau ca karye vill. 8.
Qio| maviftaaad SH) ag |aretiftear
1818e-",0
slo, aa feat aa HA] 2123
[ao , fag fedtat) iziea]
So) Wale) UR) Rat ward aq Hes
Hada wate| |
[aol fiaatafa wt) elaicos aeraa-afe:|
Zo, Masa HT] WRksZ
qo) aqugfes eal «Rl aa wat araa firar-
wiftaat, waral aeneatfitaat ifs" aida’ wafer|
afta: | |
go.) firrmcard Tl 2212
fo | wa nd: Sean cared aq wa ofratfaams 21420
eo | fmataqarftnt agar) yes
 372 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Karana, the name for the next case viz. Instrumental,

is used in the sense of ‘act’ ‘deed’ in RV. In the Pratisa-
khyas it is used in the sense of ‘instrument of production
of any action’. The word is used in every system of gram-
mar except Mu. which uses the monosyllable “‘dha”’ for it.
In “bhitakarana’’ (found in CA.) the word karana has practi-
cally the sense of instrument. The European name Instrumental
exactly agrees in sense.

Glo | BIraqrmay RITA] £13182

Alo | dated aq RNA] 312122
[Ae] BMT] wees
So] APMPHAA HT: | F12Q1he¢
q1zaqe | fRarar: cfefacafaas meaty |
faqeas Az aa HUNT AST AAT Il
agama a fe seq aateag |
Enea Waa zeaat faqear seid AA!
oy AMTHAA RTT] tye 4h
eo | ATTHAA BING], 22122
Jo, wamraadafadmaaes aq) ce
aie) fRafaarad Hun | 41
Fo | AIIHAR HINA] = 1919,
fo| RCT AMTRAAA | 71205 ( ATH! )
Hol RAPTHMA GED wert RT) see,
The next case Sampradana (Dative) also occurs for the
first time in the Astadhyayi as a technical term. In VP. ii
is used in the sense of ‘handing down by tradition’ and in the
 KARAKA 373

Kausitaki Upanisad it means ‘gift’; in the next stage it meant

the person to whom the gift is made and is the name of the
Dative in all systems of grammar except Mu. where it is replaced
by the monosyllable “‘bha’’.

qio | mao anfasfa

a asa] LI8122
aaa, fmarrerate wazaq|
alo} wea faqar Tad urnawad at aq armaran |
ah]eo
[wo] aaa agai; 22193|
Glo | RANA: AMAMAT) VAULRe
qo} sao fier at anfsifa @ amaaa | eee
Bo] RaAfAsa: GPIATAA! ARR
Go| WAST AtAGMAA | Lays
saaarearfrasm ara wal faarmot at wad @ AAAS
arate |
Zo) waTararaqarst AAA | {LIAR
To) ara aa aga weet sifaaieg a: |
Wes UTsTAT at waasariqaege: 1
fara afafad amare afxeara || vigeoc
Go| ocaaHarPeTalay SIT Ics
The next case is Apadana, a word which is derived from
apa-d-da, ‘to take off or away’ and which appears to have
been coined by grammarians for the purpose of their
science and is used rarelly, if at all, in literature. It is found
for the first time in Panini. It occurs in all the other systems
of grammar except Mu. which uses “‘ja’, resorting to the
374 TECHNICAL TERMS OF SANSKRIT GRAMMAR,

unaspirate sonants after the aspirates have been exhausted.

As the first unaspirate sonant in the Samahdra-sitra coming
immediately after bha is ja, it is used here.

Gio | AAATASTTZIAA | eles ,

Ho | warsafa waaed at aAgATATAA! «aR
[Mo] Aaa: TayAT x1RI1¢2 |
Sto | MATTED TSATTATAAL) |_UIRIRRY
Go| AAAWASTTAAA | RIS
Zo) aaTasafacararay] |BIRIR®
Go) ASAIRMITIWMAAA) 4WRe
aoa: ore arafrenay anges’ wate 1 Tit t
qo. safacararfzeaqqraray| 212 1R°
Fo] FALSTAATATATATCUA ART: |
WAAR: ALIA: SAT ATATATAHICHA | VIGeY.
ao) wararfecsafacazra) else
Adhikarana from adhi-kr means literally ‘the act of
placing at the head, then receptacle, thing, sense. In its
technical sense it occurs for the first time in the Astadhyayi.
Because adhara consists of a lesser number of syllables
some systems sometimes use adhara for adhikarana. Vopa-
deva uses da which comes immediately after ja to denote
this case.
GTe | AMarcrsfracoa epee"
hle | @aracehiacny | syste
[We , AAATTIZ| XkeIle¢c
Ste 1 ararersferacm: | eeiewe
 375
KARAKA

areuye | sa Racaafearrarargrcad ferare|

sapaa frarfagl measfract VTA Il
qe) adnatakafearnistncm | egies
2.) fimaerreqrararsfancna ) -RIRe
gio) aafrarafrncoa | 43's aTfanr, eNqaarrny, |
afar afiacudea waft) ata
Fol PRTswaeit
SAI ave
Bo} BALITCSHTHCOAL | ARI
Fo] PATA ISHAM | LIRR
Zo] RT RAMTCaTAshACT | Blec
Now a word as to the order of the cases. In the Karaka
that
sections of Panini, Katantra and most other grammars
of the
distinguish between Karaka and Vibhakti the order
Locative,
cases is as follows: Ablative, Dative, Instrumental,
according
Accusative and Nominative. This order is fixed
to Panini’s paribhasa : vipratisedhe param karyam i. 1. 42, 1e.,
in case of conflict between two rules the later 1s to prevail
over the earlier. Thus the order of the cases in Panini shows
the relative importance of the cases according to the view-
point of Indian grammarians. The Katantra also says:
tesarh param ubhayapraptau ii. 4. 16.* The Ablative probably
comes first because it is the local case par excellence. ‘Then

comes the Dative and then the Instrumental because of

similarity of forms. The Locative is intimately connected
ee

ee

& GUSA-SHSA-ACUIT-MRH
TUF |
mq Walaa SAS NAGA | SfaNEIt-aia: eRe
 376 TECHNICAL TERMS OF SANSKRIT GRAMMAR

with the Karta (Nominative) and the Karman (Accusative),

so it comes just before them and since the subject is more
important and of more frequent occurrence than the object,
the Karla comes last of all. As Karaka is the relationship
of the substantive with the verb, the Genitive which is generally
concerned with the relationship between substantives is not
reckoned among the cases.
Sanskrit grammarians do not regard the Vocative as a
separate case for the same reason. And as in the dual and
the plural and in some cases in the singular also the Voca-
tive is identical in form with the Nominative it is not regarded
as a separate Vibhakti even. Saunaka appears to stand
alone in regarding the Vocative as a Vibhakti in the well-known
line in BD: astau yatra prayujyante nanarthesu vibhaktayah.
The name for Karaka is ‘case’ in Latin. It comes from
casus meaning ‘a falling’. The Nominative case was regarded
as the original case and the other cases as a falling away from
this original form. Thus if the Nominative be represented by
an upright straight line, the other cases will be represented by
sloping lines according to the degrec of falling off, and are
therefore called oblique cases. In Latin grammars the order
is as follows: Nominative, Vocative, Accusative Genitive,
Dative, Ablative.
The nominative case was called onome (Skt. nda, name)
by Aristotle and orthe: ptdsis (Lat. casus rectus. straight or
direct case) or eutheia (fem. of euthus direct, pto:sis case),
The Latin name nominativus (sc. casus) means ‘of or belong-
ing toa name’. The oblique cases were called by the Stoics
 KARAKA 377

pto: sets plagiai (pl. of pto:sis plagia, Lat. casus obligi).

The genitive was called genike : lit. belonging to or connected
with the genos (Skt. janas) class or kind. The Latin word
genitivus means belonging to birth, (case) expressing origin*.
The name of the Dative was dotike : (fem. of dotikvs inclined
to give). The Latin name dativus means ‘pertaining to the act
of giving’. The Accusative was called by the Stoics aitiatike :
(fem. of aitiatikos lit. the thing caused by i.c., the effect of the
action; to aitiaton means the effect). It should have been
rendered into Latin by effectivus or causativus, but Latin
grammiarians thought that the word aitiatike : was derived from
aitiaomai accuse (aitia guilt, blame) and translated it by accusa-
tivus (casus, the case that denotes the object of an accusation).

The apdddna is not preserved in Greek. It is found in

Latin and was Arst called sextus the sixth case (the other
five being Nom. Voc. Acc. Gen. Dat.) or Latinus casus arid
then ablativus (formed from ablatus—-ab away from and
latus carried, taken) that which takes away and thus corres-
ponds exactly in meaning to apaddna. The name ablativus
was coined by Julius Caesar. Locative for adhikarana and

* “Aristotle laid the foundation of the division of words into ‘parts

of speech” and introduced the notion of case (pto:sis). His work in
this connexion was continued by the Stoics, manyof whose grammatical
distinctions and terms are still in use, the latter in their Latin dress,
which embodies some curious mistakes, as when genike: “the case of
kind or species’, was rendered genitivus, as if it meant
the case of
Origin. or worse still, when aifiatike : “the case of object’, was rendered
accusativus, as if from aitiaomi ‘I accuse.” JESPERSEN, Language, p. 20.
 TECHNICAL TERMS OF SANSKRIT GRAMMAR
378

instrumental for karana are of recent origin. Locative 1s

and
{rom lucat-(um) the past participle of locare ‘to place’
the
means pertaining to a place where. It was formed after
analogy of ‘vocative’ etc. Localis, however, would he a better
name for the case, meaning as it does, belonging to a place.

The Sambodhana was called kle:tike: fem. of Ale:ttkos

invocatory, vocative.

ava fafa gaftfa cata) arennegart oe

sfagiseqerat ferferr ata: | af camara: tard, ART
a Mea SAA: AT | wa a aa Fea BITH-
qeafaar xafetfa saeeariat Aa) ATTA
eoaifa arcahrae | a fafiseateat gad areafy,
afgaaufea saci ereafe, eravafafafi gazaaty,
gaa wt vadhaa wae frart geanaa fafacdat Raitit
satsta, fara at azretera at: ATHIT waa fastsfa
areat wadia | aa oa aaleaa arcarvafirataed | say
avamaad sa 2aedt feqea: wat wafa) ate ga:
cafkat saat aerate gerd faaeaa aa are” tafe
quat qafs aiza: Vadiarita vata) at wet AAT aw
Peatereaaarimamarrearaicage fata, 9aa
RaaRUNaaara | aaTaATAA—
Reqfrars Ha Aaaarhed HITT |
sriqicaganal wemearireasara: y xf | Tras: |
Henao | aITH afa aed dan fad | Aa a ANA
eata: | FA Tad) sweay fe damama)
aaati aa
 KARAKA 379

AEM: MAA BW Tad Wistay—saeadder ze

faga ) atidifa arcafafa|
sequin azadf aeseqrgrafe: |*
srafafe Sena HEMT TIGA | Brot HTH |
afanco arcafafa |
fag g sfaarca Pearse wartat acorftiacnat:
me ATE: |X
fag: wcorfancoat: Haws: | Ha: 2 Sfaancat ra
Rar Tardtany? oowargiat fe sfenrzat fee fad |
fafre ofaercafafa) aice src ofa sfaarcanq)
eISal cfanrcat fraraae: Tardranrq,2
afanquignmaaaaugemqaaswHanfea: oNATT
ad: UTA: 1x
afgsrqaienreaaraigarara
tsTRshe a: Hata
qa: Gadiageaa | da aar afadds) ce cena:
TH: | Uae Tag: HATA |
ait qaenes Tadifa anmaateat ncunfirat arf-
HLT ATH: |*
a qauiea: Tadifa araaafmat arcofaat a Haat
est Tadigead | at sar wfasda| ansfaane
ae: | cagfancred He aA |
mat, qaren fafedsdfeeantfa saeafnar acer
qT: |*
aa qe fafgicafeurdia sazafzal safe
380 TECHNICAL TERMS OF SANSKRIT GRAMMAR

anita tardhasaa |) aaa afeaea | TT HUTT

qrh:, TAY HUNT HA aA |
sqaafaqaatfa aa fesfatar |*
sgaafaqiaata gaa tazufasadiaead | da dat
fgfrada | aT ranges) Taq waAHT: Ra tay |
Aaa Gia ad Warssay |*
adq aaa saaa fanaa a agar feat a ITA, a7
qmeszagq | aa aad fad aq |
racal afaauaresrastaata: ear |*
at fe apaa sanafimaatsae aafa—foadita,
afaqnaseza a ar fags: eaty | azfaar faa aoarty
a faa a |
amqgztarzial casferte: |*
aqiaadia wa aefate: | aan fe waat acon
eum: nad factaq, a aumamdia saa
fazed —
-

aq al Gaeautacacag aa: Waa Fad FAAP A

aa |*
aaqer: | fae? waraqcaraand | aaa
qiaeay Ticaedy & faaheran | aa: gaan aaa | AAT:
acai: Tian aaa ufacafa | aaapat a
dat afacrfa | az wat) aarenie fama) wares
faraa | acrent facitaa cfs fe aa =aa—aargingiat
cacfafefcfa | ad afe a am:—aqaerdtat wgeet
cfafafcfa | atc sonfiacra: sad faafea-
 KARAKA 381

ATA ag afaeaT | Tata Ge ger: Tea

fazama, fRafer <a asfafe: | amar aasas
miaeay faa witaedy a, aa Wear Ha aaa maThs|
AAG Ta MarqtaeaMaA aa: Tatau aad qaqa
aI Ga caq |

aa qafte waar eat waded farted arraatrai

ara fat a gaat ene waaafa wart qeaeat ara|
aq aa sated aftadad ary) var caady enequataa
cane aftaad a afcormifa, fF af€? araaafteai
anofrat a afierdifa) aa ara) aaear) aati
qtaear |) ad afe enced ad aera eet eaqacay
ALE Aa HeTATA Wldeal | AT Aa Heeishy Fag
Rama EN) araaaat areata wow) aare)
Vaca | HB ATAT ITAaeaT NTA afe seNaa aaa eats
QtaeAl Aaa arat | ACTA! Waraai cust az
aaa qitaaa, wart eae) fi ga: vara o
eT | Bat gana ear warafafa, aq aig araag
afafaag wnat sadfsat aafai ag adr aaraafy3S
Aaa tateqi aaaratsfancat aiwafafa | oa fe ae.
fraaafacoraqar, apace at area) gal
ATT T ATHY AAT | |HAA 2 marTaAafaeara | ag
mat) omer Bara zfa) a fe wnat aeaa Baeuca-
At aT fata aT ada) sal aredteafesia aga,
THe aanarfacaa | wa afe aratraagat frat qaF azar
fate RTT saTaT MAT and taranfafa aaa
 382 TECHNICAL TERMS OF SANSKRIT GRAMMAR

nice oft | cd a gear fata savaraafa| zacanrfe

aTcafata ATT tl VIBIRR
qi RAl HA HL AWIIAAAA: |
ATAAITRLT ATarAcA ATE ||
afraransacatHncny|
‘He who is the doer, the deed, the instrument, the object
of gifts, the object from which something is separated and
the receptacle of all things, may He be associated with
everything here.” In the case of grammar, ‘that which is
the nominative, the accusative, the instrumental, the dative,

the ablative and the locative—that is to be fully dealt with

here.’

aval: tftaifee aan afraiinaaafrsat |

afafaeaaar a cat: azaaeaaTM AZT |
wo 220
‘The lovely girl about the existence of whose waist one
feels grave doubts was brought within the range of my
vision, as I had made more than one city the object (lit.
objective case) of my wanderings, with a view to contact
pools of water.’

aay THA FASTA |

faaifiasd aaa wean ae afanz |
aafa anata & aeftafia mica |
qaqa eaqafa AN AeTa: |
FTAA: ApVs2
 KARAKA 383

‘It is not proper for me to speak now, after (my elder

brother) Balarama has spoken. When a thing has been
fixed up by means of a letter, what is the use of a verbal
message ?’ Still the fact that you entertain feelings of respect
for me also as your guru becomes the operating agent of the
act of speaking on my part i.e., prompts me to speak.’

ear fyarqare
taey AT TAT A |at aaa |
TSseAHG AATAT CGSAKAA: |
amiaaftar 20/282
‘Bearing joyfully on his breast her who had been placed
on himself, he plainly indicated the fact of his being the agent
o* holding her up, i.e., of marrying her.’

ae aafa ad ath:
qaaifae + tg Afar |
waa: watfa-ae-mMeara:
catfacg frrctrrarce: 1 fitzo 2ui3a
‘In the matter of the Lord the roots srj (to create),
sam-hr (to destroy) and Sas (to govern) take on suffixes
denoting the agent only and never those denoting the
object. The root stu (to praise), here, takes (the suffixes
denoting) the opposite case.” In other words, God creates
and destroys and governs, but, is not created or destroyed or
governed; on the other hand, He is praised, but He does
not praise.
facm
farce at
os Bradag
ag e'eaeraa |
RASTA WT TF AT Hat VAae Hot TH |
PPNITAA eigoy
384 TECHNICAL TERMS OF SANSKRIT, GRAMMAR

the soldiers of Virata, unable to bind the king

‘While
it was
of the Trigartas safely were (helplessly) looking on,
both
guna (string, Bhima) united with the bow that became
the agent and the instrument in the matter, i.e., Bhima bound
him with the string of his bow.’

afzeqivata aAaqea AT fafacar-

quifaear aafa afin: vefaat |
ay Hi ATIAM ATARLA ACT HAZ
fag: agit garafrafeasafas: |
‘That the sea follows the fall and: rise of the moon—the
reason thereof is not far to seek. All powerful is the source
of the subject of the act of being born. I. €., because they are
sprung from the same source (cf. afar: gata: P. i. 4. 30).
What relationship is there that the waterlily follows its
example? Surely the pure are the spontaneous friends of
the pure.’

arainiad faa
qanmamMangertaate aRe-
Reales aga feaag | -BITTATT <%
Amo, amas, I love a lass,
As a cedar, tall and slender,
Sweet cowslip’s grace
Is her nom’native case,
And she’s of the feminine gender.
Rorum, corum, sunt Divorum!
Harum, scarum, Divo !
Tag, rag, merry derry, periwig and hatband !
Hic hoc horum Genitivo!
 VIBHAKTI

The word Vibhakti* is derived from vi-bhaj ‘to divide,

distribute’ and means ‘separation, partition, division, modi-
fication’. faufmare: gat aran: | faarraaat at) anca-
afafaaia: | fsasra ufarfearaisatfa wear), Nyasa
1.2.44. In TS. i. 5.22, discriminations from the usual form in the
Adhana are termed Vibhakti. In the Kausitaki Brahmana i.4
also Vibhakti is used in the sense of discrimination according
to Keith. Eggeling and Hillebrandt, however, hold that
Vibhakti is used here in the sense of case. Whatever be the
true meaning of Vibhakti in this passage, the fact remains
that it is concerned with the different cases of the word Agni.
wa arate tag RV. vi. 16.10 ‘Agni, come for refreshment.’

* Acc. to Kasika on Panini ii.3.32 the names of the seven case-

endings, prathama to saptami, have been taken over from Purvacaryas,
and indeed we find them with the same meaning and in the same order
in. Yaska. The origin of the name for case-ending itself, viz. vibhakti,
goes still farther back to an ancient custom in connexion with the
Punaradheya or Punaradhana, the repeated employment of the three
sacrifical fires, if the first employment did not show the required result.
In it a cake has to be offered to Agni alone and for the six stanzas
to be employed mutatis mutandis for this, such. stanzas must be chosen
in which the name of the deity always appears in a different form
because repetitions weaken the ceremony—yafaaai afa = gale
az oe Of ee Oma RATEAINTA We do not learn about
this custom from AB. which does not treat of Agnyadheya but from TS.
and other Brahmanas. LIEBICH, Zur Einfuehrung in die indische
einheimisehe Sprachwissenschaft, 11.15.

O. P, 129—25
386 TECHNICAL TERMS OF SANSKRIT GRAMMAR

afa gat qoiime 1.12.1. We choose Agni as our messenger.’

afsratia: afirera i.12.6 ‘Agni is kindled with Agni.’
afaaartor agatq vi. 16.34. ‘May Agni smite the foes.”
AQ: ta Aataz v. 13.2 ‘We meditate the praise of Agni.’
Fat at wat sa; vi. 14.1 ‘To Agni that mortal who doth
service,’
The first word in each of the above verses is known as
Vibhakti. Form this it was an easy transition to normal end-
ings (the Nirukta speaks of Namavibhakti) and from thence to
verbal endings also.
In the Tandya-Maha-Brahmana we find:

aq cfa sunt waafafaam:, afafafa fedia-

mifaata saiaen, afafcfa ager) eolsie
(fafafrasa: afieeeea oat faunal are ae aft
aaeid waaenat eq, wa amnfe aaa za fe
atawefa zaaa|, afafafa fadrn faufetedteerat
eqq| afa sd aetind cfa aa aeaa: | afaafa adtar
TaAraeT TST eqq | ada wa aaa afaarfa: afaraa zfa
zeqa| afatcfa qada agqeerat SIH Fal TT
aafae arafacta: az efa aa ZAI A | ata: | |
gear aaa aafafaafearnegr amg aazai-
feeqeatfa | ATCSTATTATAT 20/9)"
[on safatrate: aarn afafaafeaa aaa at
mag area aeqanetaheata great fase
maeaasg waagal aaanR afsen aared: aa
aqfa aa ame areata | ara: |
 VIBHAKTI 387

eeofa aemenal zofaeataae: | seafafa fede |

genta adiaer| patfa aged) warfafa qa |
wefa weet) aaa waa gata agegqafeal aera
Gata Adieasfaakarneaqr ACAI Aacalwaqyy: Tyee |
AMSAASTATAITT Q ol ci8
The NatyaSastra (xv. 30) explains the significance of
Vibhakti thus:

TH AEA at aatfoqea TATAT AT |

faqs wenig fernaeta at: TAT II
‘Since they discriminate the meaning of one or many words,
of dhatu and pratipadika, they are known as “‘vibhaktis’’.’
The Nirukta exemplifies the Nama-vibhaktis thus:
gear fra aoe ER ofaeat: | Wo ao gojealgo
Indra over Heaven rules, Indra over the earth.

geatag Waal TEI | Wo Ho gypsy

Indra alone the chanters praise highly.
ErgNa deaat Sfqarart: | Wo Mo wlecjs&
With Indra the Trtsus being active.

CATT ATA WIAA | Wo Ao <ieciy

To Indra chant the Saman.
Ararea Tad am faspat | Ae So eae
Without Indra no place whatsoever is pure.
tract J erairor 1 ara | Wo Mo 2320
Of Indra, now will I proclaim the herioc deeds.

{ex RAT AAA |

388 TECHNICAL TERMS OF SANSKRIT GRAMMAR

On Indra our desires rest.

It will be seen that Yaska does not regard the Vocative
as a Vibhakti, Panini and other grammarians agree with
Yaska on this point. Saunaka, in his Brhad-devata, how-
ever, distinctly speaks of eight Vibhaktis eqs} Td sgysard
aTATAg faa: (i. 43) and is perfectly justified in doing SO.
No doubt the dual and plural of the Vocative are identical
With those of the Nominative, still the singulars differ in the
case of the a-declension and several other declensions. The
dual and plural of the Ablative are identical with those of
the Dative, still the Dative and the Ablative are reckoned
as separate Vibhaktis.
It is clear from the following passage of the Mahabhasy4
(ii. 2. 24) that in ancient works on grammar the Vocative
was placed after the Locative.

frat wer mafisagfagft | fear aeq mrafrad

Tat | fast get marae say, faataer matagy
ma fe | fag aeq Trakagmaaa | fat ae mrefinger
wal faqr wer mafiaaat afk) fear ver m7 é
Fagat aft |---crat a Sad usagnea fasfa, cat 4
SATU TSAR TET | Te aTRA cagarat FIA!
Uist aT Gaet Tage Se cat ar gare AGAMA
AMAT | Ua atGard aagard: a) cat
at FA
TIMHATAT Feat y | TaN At HAT F crstrHmnfe xfer |
On cannot help thinking that originally the Vocative
terminations, were Sj, au, jas and these were placed jmme-
diately after the Nom, terminations,
which were su, aM, 14°
 380
VIBHAKTI

Grammari
en _
ar the Vocative
regarded i singular as a_ special
or brevity
mie Nominative for the sake of laghava
.

ra,
.
oO

for the Nominative

si is preserved as the ending
sin gular in
: K&atantra,J.,
- Sak., Hc., Mu. etc.

are generally known as prathama,

seven WVibhaktis
. .
The

ami. The
hi , pa fi ca ml , sasthi sapt
, caturt raka par
“sta trtiya co me s fi rs t as being th e Ka
ra ll y
sinless natu 64) and as being of most
karta i. 4.
Rents (svatantrah ave a senten
ce with-
since we can
tive,
Out the a
e
se nt en ce a ia a Nomina
cases, but a Accusative
expressed a ly possible. ou
8 hard
Cities ee d.
f th e ne ut er th e Nominative an
_— l e
Mesititaitos 5 al l co me s also because in th
ease of re idence in tween strong
h distingush be
whic
declensions r
and weak ose
mi na ti ve sh ar es the strons characte
With the naa the No NYS In ae
and dual of the ACCURA ely
Words, it a e th e Ac cu sa tive immediat
plac
carrie: to anamasthana-
after the he iv e fo r th e pu rpose © f the Sarv
canning. ] omin at ve comes after
of fr eq ue nc y also a Accu sati
the Nomins — im ar ily affect ed by
the action
becaus e it is pr
Of the a ative, alkest, the Instru-
me Accusati ve plural is we mes
Mental rs a t, $0 th e Instrumental co
is also weak es is
Vmiatind” on ti ve . Th e Instrumental dual
ter the Accusa the Dative p
as
identical : y af Dative
e du al , so
as
form with th after the Dative
after the . The Abla ti ve co me s
its dy
nstrumental. with. those of|
the Dative. Th
e
ar e id en ti ca l
al and plural ti ve singular exce
pt
< ; th th e Ab la
Niti ve cal wi
ular is identi e Genitive.
This is
in the © sing d so it co me s af te r th
a-declesion an
390 TECHNICAL TERMS OF SANSKRIT GRAMMAR

how the names Prathama, Dvitiya etc. came to be given to the

different case-endings.
It is just possible that the lakdras (moods and tenses) were
at one time called Prathama, Dvitiya etc., two of which names
Panicami (Imperative) and Saptami (Potential) are still presery-
ed in the works of the Katantra and its followers, Hemacandra
and others.

In “ne prathamayor am” P. vii. 1. 28 Prathama in the

dual stands for prathama and dvitiya. The practice of the
Pratisakhyas is to use the second word in the dual.

The word vibhakti occurs in CA. # farhaercifataary

CA. ii. 51, Rrareartagraateanracar CA. iii. 78.

J. takes the dual vibhakti as standing for nominal and

verbal endings and resolving the word into its constituent
letters v, i, bh, a, k, t,i adds a to the consonants, because
prathama ends in @ and p to the vowles because sUP- the
generic term for the case-endings, ends in p. Thus the seve?
Vibhaktis, according to J., are: va, ip, bha, ap, ka, t4, and
ip. Mu. appears to have improved upon the terminology of
J. by adding i to the initial consonant or consonants of the
eee sees In the case of the Locative, howevel, since
st sounds very much like si, 7 is added to the consonants of
the second syllable. The reason why i is added and not 4
is that out of the seven names of Vibhaktis four end in i. The
Names according to Mu. are; pri, dvi, tri, ci, pi, si and pti.

Sak, defines vibhakti in 1. 3. 181 and prathama ete. i

L. 3.3 182 ff., so in
. the previous rules .
he has to use svaujas eles
 VIBHAKTt 391

for “‘prathama,” as in seqarel caine 1. 3. 99, RTRIR:

FNS TA 1. 3. 100 ete.

uses the pratyahara sup and the Vrtti says under
Sam.
‘@iiserara” vi. 2: “qaarfraahersiat TT TETTETT
and in
TTT
TTT
EEE the text the word vibhakti is used in SCALA Aha E
Raftrmaerda vi. 32. etc.
c ~,

also to denote
Panini and Katantra used th e term Vibhakti
, adhuna, danim,
the Taddhita suffixes tas, tra, ha, dha, rhi
ically serve the purpose
tha, tham, astat, atas, etc., which prat
of Vibhaktis.
the word Vibhakti wit h the
Every system of grammar uses HN.
syllable “kti" and
uses the last
exception of Mu. which and converts Vibhakti int
o
Which use s the name of Visnu
Visnubhakti.
qivanarad-
ate) freatfor cifen qyAaarday: |
at
au | Ot glaleee-a | my fr

Rrafen: 1tale
ara) RR
Slo) aaa ac f | ol aiearnmtgal erat
rns:
aeey agare fe
e: | THT UR
rra
a fr
fa Rtasfa gat qt
Frafinear frat aearase- O QIRRR
EAT
e m T A G A T T E M
AM’: ad a aa aA
ferrHat wat |
aaah a
So}
Fararfenficfgeeret| uel TAAISRC%
g1
sto) aiqu er fared ESF | | ataleee-2e?
ep er yfa ra tf en : | TO RT S?
Rito 1 egafy
392 TECHNICAL TERMS OF SANSKRIT GRAMMAR‘

aol faegn feats: aat atfm aif saaaetr

war | gt swaarfsemedmaquiqadiastaa: |
PURIAS-Re
fo | ea aeaANE: - ail aat sanifz: | sarfe fete |
LIRIRS-AS
fraseat fear: saat agmaigatsat Aaa |
Fras at) @aeare:|
Ho | feearfe: fe) 22, fra: ehisttefigisieet: | 9%
Gol afaar fqatatagtinaaaaaragaaaaTa
ferried | 21213"
saa
wea aat frasrafa t : | WHT |
frre
go | favtal aqafast wal | 212098, orate AAT
aaa: aafahafaha: | ere feaa carat 212808|
eo) efecto faapreaa: } 21782
fae aq arafrafrahara
UAT AT TATA |
wT alsa aafia’ fren.
WUT VAT AAS TE LIT |) Ao AFCAT 2X2
‘If one thinks of distinguishing goo
d men, that individual
na to be mentioned first who by
the display of his prowes
is able to bring to subjection man
y refractory states, (wien
one thinks of a good Vibhakti, then the first case
-ending h4°
to be taken into consideration, which by the
manifestation of
SU, au, jas
tac 3
is able to convert a large number of stems into
inflected words,’
 VIBHAKTI 393

sah sar, aa asafefa aaa |

aqait aatafa wore: orferacte
ATTA CAT, 918°
of dalliance”
“Both the sexes should tread the primrose path
Is the view of Panini who says: Salvation is for the third
Sex (eunch). Of course the real meaning of the rule is that
g time
the third case-endings is to be used with words denotin
completion of the action.
and place to denote the
second
q a are use d in peculiar senses in the
Wet and
of the following two stanzas :
|
cacafied aft gertea farfaat 827 1
aafcerte
ae] TS: HAVA AT FATT
r facie
arte cxareaarenrfaat eate
araeaarf
rnefar
adarda ee |
e ar
aT ABT qaagHET TAT: TAA a Aa
citeenazai saa 2 aotlawatof King Vallala Sen,
in-
daughter-
It is said that the husband wrote
pan gs of sep aration from her
Pining. from the
© the wall of her boudoir : joy.
Peacocks are dancing in
It is raining incessantly.
Ej
ther my beloved or Yama will put an end to my sufferings
to-day,

of the king who forth-

This stanza happened to catch the eye
Wit
h Wrote to his son :
auehter (GBT =
This peerless ( equaitafaet=serefem ? ° fafazet ) with
FFI) of the lion of kings ( FATGAAHAATA
394 TECHNICAL TERMS OF SANSKRIT GRAMMAR

(heaving) breasts like the temples of elephants ( maMleait-

eamti=nfepriafaegfagarazct) and (arched) eye-brows -
bows (Tqay a ATTA ) tormented by the first _ O
youth (qa eae faaaARTAIReT TIATIA Aa ATTAAT
aaa aaa aT ) and having fainted like a crab or a fish
( qasiaaa ) on a place devoid of water is suffering the
Stings of the scorpion ( aentaanashrnataaay
O thou with the intellect of a bull (fediaafanaaarAysAd);
O ram (@ SaA=AG), join her quickly ( qur gar: WA=AAT
faertrrs) |
Amantrita and Sambuddhi
The vocative is termed Amantrita by Panini and as the
singular has often special forms differing from the Nom.
sing., the Voc. sing. is designated Sambuddhi. Anant
occurs in the Taittiriya Brahmana in the sense of ‘to address -
Amantrana occurs in the sense of ‘address’ in the Satapatha
Brahmana :arate q QTATAT aay vi.6.2.5. “«Amantrita”’
is the past passive participle of dmantra and denotes the person
addressed. With the suffix -kta in the bhava-vacya ( aga
ata wh: iii, 3.114), it means the same thing as amantrana Of
sambodhana. = :
Amantrita .
in the sense of the S
voc. occul
first in the Pratisakhyas and in Panini,
SHIT aatedtast
TWa: RP. 1.68, TTA aarararasatatat vP ii. 17>
arated a ii. 24% aaearaRaaat: ili, 139 ageraatetla
mafeaaa vi. 1, omaha Sanaa CA. i. 81. It Mt
TATA for the voc. sing, : Calataad Ofzaaaedea it.
47. Apr. uses “amantrita”
several times: aqagqraTfZ ae :
 395
VIBHAKTI
ntaararerararareara : 7
8. stiaf
i.| 18.
area aiafaad Ta! ,
E 72 Lehi.

anafaarargzerara, Taraleat TA
a gat ll 286
SMITA ATA: qararfaac but
“a mantrita’”
Sarv avarman does not define OF explain

.takes es itjt for granted.

; ,
Sambodhana and Sambuddhi mean. the
l
use of by Panini in 4 specia
© . 7
,

e thing, Sambuddhi is made ;

a
m ‘

arman,
. 4 followed by Sarvav
3s
sense
and in this instance he techni-
of cases Sarvavarman’'s
though gh jin the vast majority
Sambod hane su
ry. Candra uses
-_ ae self-explanato He uses = mantrita in vi.3.24 and
diacetate o n,es vi. 2. 44. into
or tens Sambodhana
49. Jainendra sh nsonant
hi in vi. 3. th the first co
Bodha it has “Ei? wi
for Sambud dh i of the
Of the a dh i. Ja in endra’s use
sambud
and the i of that “aa to
this
first alphabet one to suppose
lead
a. would §akatayana
tive.
a.
Vo c. pr ec ed ed the Nomina has mm be
as
Uses =
rya” (lit. one who
yllabic “amant “Sam-
addres @ tris “S am bo dh an a”” and for
the polysyllabic AM Alaaa-
lesan d) for ana”’.
i” he uses ‘“ekamantr Hemacandra uss
A follows Panini.
i.2.121. Bhoja sing.-
ee
adds “si” [0 denote the
and
"ya for Sambodhana es 3 both
1.4.40 Kramadisy ara us
qase fararneey as
—
amantrita and adds hy
for gaqral
hana and Sambuddhi 3 ara | 163,
denote @
6 © the sing.: rule TRaat-
the Subanta section there is 4
vi.5. In been
wT appears t0 have
rs a
)

ei s ie arey 379 where emqataCommentators however, are

Used
for Panini’s Sambuddhi.
 396 TECHNICAL TERMS OF SANSKRIT GRAMMAR

of opinion that @tagyeaisais correct reading here. Thus

Goyicandra says: mRAaAarayafefa sat TET qedag
zaaa, fag aq qagfacailg adeneargeatafircatia:
aafafa werd | wa wHaaaaragraia carAleay a
vardifa wfafradadacafactag gait A zEq
Vamsivadana remarks in his Tippani ‘ ATRIA aaa

sant waar: |
Curiously enough Vopadeva who uses monosyllabic technical
terms throughout contents himself with changing the poly-
syllabic Sambodhana into the trisyllabic Sambuddhi. For
the voc. sing. he uses ‘‘dhi’ which is a vast improvement
on Devanandin’s “ki”? being the last syllable of “‘sambuddhi’’.
Sar. follows Mu. The Supadma, like the Samksiptasara uses
Sambuddhi for Sambodhana and ‘‘su”’ for sing. The Prayo-
garatnamala follows Panini. The MHarinamamrta uses the
popular word Sambodhana for “‘amantrita’”’ and Buddha, the
name of the avatara which is nearest in sound to Sambuddhi,
for that term.

Kaccayana (Pali Grammar, ii. 1.4), has ‘“‘alapana” for

“amantrita and ‘‘ga’’ for Sambuddhi.

qio| arataaq| | at=araiaa caati] THaad

rats: | 21RNeC-¥e
Hlo | aATafead fa: arafg: 4 221%
sto, arr drag, gf: f1 eeiee-go
Go| BaaAlTiaardamady | aAtKrad arate:|
UIRNSEASE
 VIBHAKTI 397

Ho, araat fafa) <z

araraa fafa: fafaeda: era) ahr: |
ato | waraant fafa) <3
Go| araranaad arafe:) aie, amafaad
CATATTA: ABIL
go, aiafad fa: arafs:) aaaiaer afana-
ATAUAITAAIAATNA | IRco
fo] Brat gages: | ara, ez
dataraga Harearhtfs sitenaria
qargaia geranta aga arfefe av an: |
ad amfan wefta weg 8 fret ahaa
maaan fad oe aaah |
afteatiasrata fiche aqareest |
‘O daughter of the dear husband of Samjiia, O sister of
Yama, O constant beloved of Srikrsna, O annihilator of sins,
O receptacle of virtue, O Yamuna, O KéAlindi, salutation to
thee. Those who having taken a vow read these names
while performing ablutions, sins of as many lives as there are
vocatives in this stanza forthwith leave them.’

a rerfesucar feaLae TATA ATT ST: 7

H Tae agad gaiarncnafst Ha aca: |
seas: Y a aay farina:
fa aa rife fanafs a agenfefininarsty |
Tg aerate: |
Where are the combined vocative singulats of the ‘star
( at), Visnu (29), the sheep (aq), lustre (at), bird (3)
398 TECHNICAL TERMS OF SANSKRIT GRAMMAR

and of her who is pleasing to the mind ( aq)? where are

three roots belonging to the ad-class (AT Alay, atafaaeraat:
zm Tat )? Where is the knowledge of the essence of
things? Where are four words formed with primany suffixes)
with the successive elision of one letter arfa:=waqet ATA,
ofaaaarer ager, faaaer ager, TaATT ATA, )?
What is the rule of Panini ( yp iii.3.18) which does not
shine even in the resplendent sunlight even when uttered 4
thousand times (which shines even in bhasvare without is
Ss, v and r)?

frarercrearcat fret arta Suz: |

fofagraiia wagtagzy|
qeqaT RAT
eatrat gfta: freq: eeafaragtce fea: |
Ufscent faaren deaf ga: ga: A
AACATONS:

UPASARGA
| Upasarga from upa’ ‘near’ and stj ‘to discharge’ means
literally ‘discharged near’, then ‘addition’, then ‘addition t0 the
verb’ i.e. preposition. It occurs first in the Aitareya Br al
mana (xvi. 4): Mahanamninaim upasargan upasrijati, he adds
the additions to the Mahanamnis, Mahanamninam pafica
ksatan
upasargan upasrjatyekadasaksaresu pidesu, he adds five’
Syllable additions of the Mahanamni
to ‘Pidas of eleve®
 UPASARGA 399

syllables. In these passages “‘upasarga’’ is used in the sense

of ‘addition’. Curiously enough four of the five stanzas to be
added contain prepositions with their verbs. BLA TAAT-
Taafaaafar ata qaeaeyTaM: | Durga under Nirukta
13. sae ataremaaicaa faad asegqrraardiagy-
@qami:; Skanda. In medical literature upasarga means ‘a disease
Superinduced upon another: ayer erga Ta: TAT: |
Susruta. In the Yogasiitras ‘‘upasarga” means obstacle. a
aaragTea eens fags: Ul. 39.
The Nirukta (i.3-5) mentions the conflicting views of
Sakatayna and Gargya regarding upasargas: na nirbaddha
upasarga arthan nirahur iti Sakatayanah. namakhyatayos tu
karmopasarhyogadyotaka bhavanti. uccavacah padartha bha-
vantiti Gargyah. tad ya esu padarthah prahur ime tam nama-
khyatayor artha -vikaranam. ‘The upasargas’, says Sakatayana,
‘dissocated (from nouns and verbs) cannot fully express any
sense ;they, however, indicate the modified sense of the noun
‘They have diverse senses’, Says Gargya. ‘So
and the verb.’
Whatever meaning is in them they express that meaning in
the modification of the sense of the noun and the verb.’
Thus according to Sakataéyana Upasargas modify the senses
Of nouns and verbs, but are themselves practically without
any meaning. Sakatayana found that the upasargas were
always connected either with nouns (¢8- prapitamaha) or
With verbs (e.g. pratisthate) and in each case there was a

“Crtain modification of sense. So he regarded them as emply

the
Words, Gargya did not deny that the upasar “ed
gas modifi
than
Senses of nouns and verbs, He merely went a step farther
400 ‘TECHNICAL TERMS OF SANSKRIT GRAMMAR

Sakatayana and said that the upasargas have a sense of their

‘
own also by means of which they cause modification 1
. . - t é
.

senses Of nouns and verbs.

Yaska accepts the view of Gargya’ and mentions the iia

upasargas with their. meanings in the following order: (1) 4
‘hitherward,’ (2) pra and (3). para opposite of hitherward,
(4) abhi ‘towards’, (5) prati ‘opposite of towards’, (6) af? on
(7) su ‘honoured’, nir and (9) dur ‘opposite of honoured’,
(10) nf and (11) ava ‘downwards’ (12) ud ‘opposite e

downwards’, (13) sam ‘together’, (14) vi and (15) apa ‘opposit

of together’, (16) anu ‘similarity’ and ‘succession’, (17) 47%
‘commingling’, (18) upa ‘accession’, (19) pari ‘around’, (20)
adhi ‘above, supremacy’.
Here Yaska had evidently to depart from the accepted
; sense bet ween
order of the upasargas to show the contrast in
many of them.
RP. gives the prepositions in the following order: ”
abhi, 4 para, nir, dur, anu, vi, upa, apa, sam,
pari, prat!, Mi
ati, adhi, su, ud, ava, api.

Here the metre must have interfered with the accepted

order of the upasargas.
TP. stands alone in mentioning only ten or eleven upasar
e?? |
nits rule at ovargranerf of gfe
fate: |a
some read SUaTaTAT etc. Acc. to them the number is '™
The Vaidikabharana says : Saat: TAT
A ‘sa fares
caten | fide gq often aasaaiazaat 7
TrACEIGSIEA LATA: | oyracit ag fearrarr
nt
 401
UPASARGA

fafaata cater | aaeraraa fieaeqt: [to de 4I1C128]

COTA GHA A RAT | TAT Ay Area aiferaferera:|
Te TA adsad| aqgsy saeraaad [aso ao yaoJ
qeaTal qeteray at ahaa |
In VP. vi. 24 the order is as follows: para, upa, apa, ava,
dur, ud,
prati, pari, anu, api, ati, adhi, an, pra, sam, nir,
ni, vi, su and adhi. | |

It will be noticed that the disyllabic upasargas have

been placed first and then the monosyllabic ones; abhi
comes Jast because it alone 1s accented on the last syllable.
been
And the disyllabic and monosyllabic upasargas have
of sandhi
arranged in such a way as to make the maximum
possible. my Te Be
58 the list isas
In the Ganapatha on “pradayah” i. 4.
nis, dus, vi, an, ni, adhi,
follows : pra, para, apa, sam, anu, ava,
api, ati, su, ud, abhi, prati, pari, and upa.
accha, Srat and antar are also
According to Sakatayana,
ected with verbs. Thus the
Upasargas because they are conn
Brhaddevata says :

sTamteg fata: Fecerrertaret farstfat: |

Ul
Ranta Bae ararertatertitag
: T: ATH, |
rASHe HRT:
TiGaT
serexrria Rrarae8g nty because
ed as twe of
‘The upasargas are to be recognis
con nex ion wit h ver bs. Th ey introduce modifications in
their
d verbs. Acdrya Sikatdyana
the. senses of endings of nouns an
O.P. 12926
402 TECHNICAL TERMS OF SANSKRIT GRAMMAR

rearded accha, Srat and antar as upasargas—by reason of their

connexion with verbs. These three are additional upasargas.’
BD. appears to accept the view of Sakatayana and regard
upasargas as having no meaning of their own.

RP. says :“gaaqart aang” xii. 25, ie.,; the upasarga is the
specializer. Again it saysgqerm faatfacaarant: Aeqceay,
xii 22, the twenty upasargas are expressive of meaning in combi-
nation with nouns and verbs. Thus RP. also appears to follow
Sakatayana.

When RP. says “sqerait *xii, 25 it evidently means

‘of the noun and the verb.’ But since the upasarga is_primarily
connected with the verb Katyayana says in his Varttika.
Rearftaera sq: 13.1.7. Under iit.1 Patafjali says:
STAT TataARAAT aa abaq frararat qeq: TSF a aq
frarfaatrare: | Such is the nature of upasargas that where
a word denoting action (i.e. verb) is used, they express the special
character of the action. Under figarfaaqx Soa i.3.1.7,
Patafijali appears to accept the view of Sakatayana and holds
that _upasargas merely make explicit what was implicit in the
root itself.

Taetife frat aaa at st frfrafe | aafe a yaqae-

wre Th za Nese afracfs, va a wg 4
safaacta aa sana ada aft aeraa NS
wrat a afracta, wTaig ng’
cafrazfa 3
qT TaRATaeTay: ST ceife |} ag Gateqarace: re
7 att «
aqfong aay xe ate samaatad rea
a—frefa sfa8*
 | 403
UPASARGA

sft) faoafa afafnarar faafe:, sfaga sfa afataar

Waa, Rawat sraraakae Gara afafear asa
aaa) a ad arf agai
eft) sits cormarc arfraafo
aft sraqt aaedifa | agarafe: sfact esasga ant
ada) Saag avdifr) tfe: eqfeatqarirasrg ee:
acer arf ad) afaat gat afeme aAeatsga-
earadif | sofaagangata zat frviecat arti
ada) uate) malsel seaartfe wea | feraraat
aft a4, wt ge) af Fe) AeA: Fe
equate mat | cafrerfa fagfata afafarniane, fasta:
ta afafrarn faahar | |
makes use of this doctrine in the well-known
Magha
Stanza (x. 15):
Ate |
ATATTR
aeana FATATHATA
TA I
faad Mant: TATA TTAMIAAIAT TA
‘The intoxication due to dr inking brought
out in a beautt-

ful manner the amorous gest ures of the young wome n—gestures
e—because there
that had been lying dormant for 4 long tim
even as the
manifestation,
had been no occasion fo r their
t in the root.
Upasarga brings out the sense laten
under at qradiean farsa
Narayana makes this quite cleat
Hera Kumara 1. 25. MIAMAI-
AMAT
t TT! ST
E> aera

Rift Rent germ a wise avi era

a:— Age
| ATE at
Battor; gar faa nay
rary xf |
404 TECHNICAL TERMS OF SANSKRIT GRAMMAR

The NatyaSastra (xv. 26 ed. Chowkhamba) defines upasarga

thus :

mfactanaan unaquanta 2 ay: |

stam sofeoreary denreaeas
fers I
‘Those which colour with their own meanings the senses
of roots combined with the sense of the Pratipadika wee
for that reason, called upasarga in this science of the purity
of speech,’
Kautilya follows Katydyana and defines upasarga thus
in his Arthasastra : fran fratfran: (= fantfrafinan: a fg
fargivat: acc. to arfraraarfrg 1.2.37) stza STAT |
1.10 |
In the Hemacandranyisa upasarga is explained _
Say aga fray aadkerTeat: |
Sdyana says under RV. i. 164, 45: TITTasId area
TATA: |
The name upasarga has been retained in
every syst¢™
of grammar except J. Candra, Mu. and HN.
J. wanted to have a monosyllabic term and
so fixed up"
g in the last syllable of upasarga and adde
d a euphoni¢
because as many as nine upasargas have it. WVop
adeva followed
J. Candra uses pradi for upasarga. HN. must have a name
‘
Visnu and so changed upasarga into Upen
dra. ;
Sam. generally has pradi (i125, vii, 213,3
3 etc). sre
used in ii. 596,607, vii. 87.
When the upasargas govern nouns or prono
uns they
are known as karmapravacaniya. When they afe neither
 UPASARGA 405

connected with a verb nor govern cases of nouns and profiouns

they are merely nipatas. acc. to later grammarians. Thus in
pranayaka, pra is connected not with nayaka but with
the participle gata in pragata, so with reference to nayaka
pra is not an upasarga but a mere nipata, consequently
the = of nayaka is not changed into uf acc. to STaATMTAATA-
sfe oyqeareq viii. 4. 14.
The Vrttitrayavarttika contains a fine stanza regarding
Upasargas. It runs thus: |
afar mf anatorishe area: |
WHAM: AAA Aa AFTHAT FF Il
‘In certain senses even intransitive roots become transitive
in connexion with upasargas, just as idle men become full
of work through the companionship of good men.’ | |
The modifications of meaning caused by upasargas are
atha :
Specified thus at the end of the Sakatayaniya Dhatup
aradt anaa aad afererrsrae |
aia fafirreaea soettfeteret II
TIL FETA: alls ay a: |
TeaTeTSTeEA TTRferegITer TUATE |
GTA IAAT
SATA N |
TRTCRTTS ATCT TLENTA II
counter to the sense of the root, as AT-Fu,
Some run

Stsafs, en-aaya, sfa-aae, seqcafa, seca |

sense of the root, as att: n=, ai-aate,
Some follow the
root. as 4
T-erqq Some modify the sense of the
Ve-aefe |
406 TECHNICAL TERMS OF SANSKRIT GRAMMAR

- Vardhamana in his Katantra-vistara speaks of four ways in

which upasargas modify the sense of verbs :
aaiTamaga wate |
qreadt araa ahaq afraagace |
aaa fafraczreaisaaHisea: TISAI
afarq uray araat | war saafa saz | CALAIS
aagadd | aatevita wast, ana fafiraccaea: | AT
gitata fatiat | waerat rar, qazaa fadiafa | |
Vardhamana in his Katanntra-vistara speaks of four ways a
sargas used with a single verb is generally five :
AT Ura: Yasaea Gaae garsafa: |
saan: fara awed wat II
amecta | sarecfa, afaemnecfa | earfacaTecte |
qantaeatecte |
arforfer: | saerat: rer * eyes
ararat: | fearfastgn seat | |e13tehs
aaex: ) rarer fir: RIS"

"A poetaster has framed the following samasya-piirana stanza ™ ith

this rule :

SaeTt: fama qfmacie aaaq |

fats aisia aarafa: Styaa: eer Haz II
The great Panini thinks that upasarga is united to action: how
‘ted .t0
then is your enemy, though (rendered) inactive, is still united
upasarga?
There is. a pun here on the word “upasarga” which means ey
th
“preposition” and “calamity”’,
 UPASARGA 407

annzira: | orfentaea’t | aerrereaifaadateerta-

PATTI: GTR TY LIER
afa:) orf: axrpato: | a a wgutarcaaa? a
wafa | deqaeraaacas ages vate) aaa 7
Sq aa ye: ser wafa |avefa rsfee ater i
Seq wat: Braet azdarat srfzecasdat vata | ay
Taratafaadt sf, al arate eat efa, aanfernafaee
afahch artara
astfeat |
S | STAT HRTTATT | aI
TAT
ATATHU
Sacra | |ata’: gaia fararaifarata-
MATA eas: ASIA: ATL ANTI
dfixeercq) = srrerectter ofafacafadg tana gsaay-
THT:|
fanfien: fRaatgrat: afazafiaisyranaat li
reo afy |
aera CATT: | AIA
aa a
aft: | sag
t
amae ashe; a a aeafterrg aeT Sete
Maat ee ara: aaa wadifa arent my,
srretoarel: finer» atererre—fier sigerar fet| 218
BATT | gercrmede Tags A ATT AT
“ANS ft: | 20k
TAT | qrerererris qrey Se | RIgIR9
aa
efterargay) grat SUERaeT ange
TRL |lBR |
408 TECHNICAL TERMS OF SANSKRIT GRAMMAR

gemiga:) aa fra gagal areal) “fara

qaraan’” carfzfatanizegqeeadenaafafa aa aft|
sre gn” “agiciiach” genaqaanznraranifa
‘Aa Ga ETAT:|

SFT:
a fata sian aaifacieteta ante: |
etafa: | qfarantaradigaieangead | Ae
aafa vamreanieany ,gaa aageram—fafaa STAT
MATa, wagner ag: aeasaing: | eT
sad—a faagi scant auifacefefa anziaa: | ata
gfaaa) faa freee: ararenamar ozareaee™
facet: aa) Raya, TAT: | AreTTaATyAT
faaafea aeta aaiteqrani: | aaifacrgfcfa mee
fata Watsatag: | aTaa Sarat fearan fenctes
yafaciaarnfaertages: | qqT Antal qaqaaaral-
aframufaaifea, catanafr aarearafamsataae
afmaifea | a mame. ornare) mRzeaTT4|
asiferrary BR [Bee] | ARTA |
aah: | wd amrenamat aqagafraae
Tam Tataga: Maaqdy wafrqare—a7 faaqa!

zatfeat | faceftata araea:, fa: gam, ata

man? | ste aarenaanies faded qa
 UPASARGA 409

quant: | gaR saat svaa aula fade are: |

Mazar aarncorfgann wera fet VITA: |
Panini
Sakatayana was by far the greatest grammarian before
The authorship of the Unadi sttras is attributed to him. He
was the first great grammarian to postulate that all names are
derived from roots. ( ATH @ AlasArE feud cATHCT WRIST
day he was sitting by
a any | WSTATEAA | ) It is said that one
the roadside intent on grammatical speculations, when a train
of waggons rattled by. He sat there blissfully unconscious of
at
any noise. quar waft & abrennzfe adarrate
waa | ag aa saracoal WHT CAM ATA:
WRIA ated ATA |
ATAGSATATN HATAATATATT aated |
ai) are—ae aereetrdfa ? ssqq—alaleata-
Gey RATATATAH AAA | AUST ATTY: | ATAT-
Rae ao RT asa feat ahagrdass qraatea |
AAT STATS Ba sasVe | TAT
4 a AMTIRATATRAA
aafa
TAaadadt geqeq sofas fasasaaa FAT TA
TTT TETT: | :
| Fer
epee: | fate gafa? armeraniftarts
AST LIT
Wet) araenaata HAT: AURA] TE
TAK
atite: fasta: nacife;, ser TERTSATAT THANE e:
T | ea far
Waraifa | aaiqdatraraat | RATTTZTA
Reta eft ar ata: | HarTaaT aylet: Area: carfarfa
waar earn aver fa
TEfe: | aaa) #: [2113]
Rararhraca: | att BAAtie| matsatadt
410 TECHNICAL TERMS: OF SANSKRIT GRAMMAR

cfs ae: sacafaarasfaa, aka aa) a arava: |

afe? segionaarser qeteacroraea fatter arast efa |
faa Teas) Te amienadiarmana aft Tt
Gira A qd CA) apse samy ATH
THT ARTMATY aan TAAATT
afafefa aamcnresiaa TARTTAIAT: | aat a aaa
qraraieaagen fordirantragandaa GaIrearzas
sfasiad | aaatnrragerguafafcte gare: |
aia Means connexion. yyeayey—subordinate connexion,
secondary connexion modification.

SHIA: TAA weed wT: |

St | we—aseaar aaa AeA AHA

ama 2 Tsar Tear wards as: | SAT
AMAA GHAI TETHTC ELTA: | marae agraTaem
qziat wafea, fagernaf arateqrararfafa mira
arar at
aeae cf arstatta
arerate: | malsca tTA-
ATATSA
fretascatard gears: | aaa ‘a cenfraifia wet
zat ubaratiaaare reser fret RATTATEA | Fe

Taga—aafes, anarmagaq—a gare

fatag—ardar ami ef | area fe aitafirarrate
Teta | TMT aI aA: | ST I
vada aafrarrmifiatafans | agar aetaaatat
aarratacersat, aay aa age aviagae
ore Srareara eae afe, cara | afr a
TATRA TIAMAT aI | TAS carat
 | di
UPASARGA

raga: az: gay wef) aaar qecadacoanrei

aqqantinta S| TaTaTIT TEACae
ama | 4 oaaanaqafidaerd? alam waa
fagqaamnnia wa ena) afas aaa aeTquraeat
ATT LATTAA | ay Gataga IAAT STAT CT
adread afiaisfe Aania camara, wate aiet-
TS GHiarAdaraaya FeaTaySA PETIGE TLCUR GLb
maqam saadift ard: eeqrerifareqTAra fH HTT
faeatarafr cqreiiraraaaerneareaga AaTeaTa TeneAhA-
saqiy: | aa age TATATATAT stam ifs, TATAAA,
Ja
arenas sera agra wa)
aa—a fe ae at aa anal waafa eae |
aaeaTA,
aaa |aTefaars TATA TATA AATETTA-
|
wraer fra faRAwrset:, PeqTaTHTAMAMETNIT
juxtaposition of ud ca ava
The word ‘uccavaca’ is a mere
ca. Cf.
aaARAgTHASAA areata: |
I
aft qsteat faeatea fre freaccraa
[ago we eigaulne |
and
ifo rm, wit h the pas sin g days, this water mounts
“Un
falls again.
tempest-clouds give life to earth, and fires re-animate
The
the heaven,” Griffith. delight. and
offered in sacrificia 1 fires
[The oblations
Strengthen the Gods. Griffith.]
412 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Gargya is the reputed author of the Padapatha of the Sama-

veda. He is also well-known for his definition of upamé.
agay acacatafa mea: | =Nir. iii, 13.
HT | THAT aETH: areraniazenat ata
far samiga: af, @aar araaqarat geTaTael
sranigaas aanaar wees) od apa any aU
faait are fa areaga:; fe gehen? Aae—
TRAM fataval waeagifagser TaTe:,
TRURRAIENS, TATE Aaaaagy TaIAT saci
RRartiaaaggata a Raa | era: mHeaTfeaes
RASMTATINT:, A Masta AEF | afm
a qa Faarauodfarsaa, aa Faceahe walt |
ween faaat atta | Ho Mo gojgyayy |, firealanfeatcta |
aaa fe afe afaamel aera) aerial
BALA Gzait ATES [ FEIATA &|2|2¢ |, sq gata-
aaa aa fara aez: qysra aa Prafararg’, qa
Awa a warszat fratarg: fs) sat Sapiapcurl
aft arenrer aa Pre aa) gaara aanarad dd
atParfagacty | aa ad geassied agar | aa
aa Tae |. qe FT oy qqqaig araranay W
wastage waaified: setae. sefta srameatent
arerenrerticcien fieacog) geaeqat ame [ 2t2lee?J
ata waft waz, axot at) Renrcafzerct: | aa fara
afaztafasratr faaioa:, Tafa mqadifaad | afafaatr
aafafigrn: sq fradaaq—froty sfaga afa, #faas
 UPASARGA 413

efcaq—awe seraa fa, wet ga fae gente

AIGCICE IE ERICK CILIC RICEI mR Gon
aT gee |
aad) sae |
ai) saat afte s oot aferactfars
AY a); At aaa | ag aat, a qaara xfa |
aaiaf wat, waekast aquantqaaatse
FRSC AA TATSLIATATHIAATET |
ena: | orgftasi armed, faster aorfiggrara—
mm genfe | aaine: afies: | agr aadt
fee a aaa
fedtar,; said azq, wasaaetea fe afa—er
ayRererg [ Wodo wlgns ] eavay Aatarfeeafe TAA,
fora a agqaarizeafis, aarret seTeco|yHo He
alas |—
steiq|
az arfe aaa a aTElee qTaeaaAT y
aay wet gear fraa faataat atfaat TAME I
er : both there and
[‘“Depart, O Maghavan, again come hith
here thy goal is, Indra, Brother,
rest in, and where
Where thy tall chariot hath a place to
thou loosest thy loud-neighing Courser. 2
Ta: AT TT
faearfira aeE ATE — ale, qa ae
Was og agena ate) FATT sat aTTRTC
saaea ate || RANT
aH qrgraia ahaaerTay, FLAT ge te
Gataqera—an a ger a arta | Se—t
| Saar eTpEseAgia
ara: SATATET IS USA adem:
ttar | (aegaeg
we st) Raq, self swaaT AAsr fxe
414 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Baz oie agaafesiscareae (] wet: qaraahaare |

tHT aaiseaa weafsafeaa: | leat cayeA? 7a
wae ged weal fread ort anager faaraa
famaradt a argued arfaatssrer | nlexet!
Treader | arearacarfsatent qurad a araradifa ART
WAHAMa waa araara at fraaaeate
faqerea cfs aerarrse sea: | TART ATTA qeaet: |
aj means ‘hither-ward.’ ay qaata originally meant ‘from
the mountain’, then it came to signify ‘up to the mountain’.
AT-Ztq means givingto oneself, i.e., taking. en-maa mean
going to oneself, i.e., coming, gzyzfa means to let go, put aside.
arqata means to let go towards oneself, to put aside on ,
oneself i.e. to put on.

“at ara arte aast fa qa arfe az: |”

[ Ho do 70173912J
“Higher, O Wind, blow healing balm,
blow all disease away, thou Wind.”
UM Vaaarnaaaaataaa A ReseITaTs-
ward: ofa Rea fang 47d |
ad RiTeahrarer cea RTA
ATA ATaT: qangarag —— eat AGA? I
aifaafaate: |
Av. a, Lat. @ ‘from’; Lat a pueris from boyhood. Greek ©
in édé.

ALA aeT aifteatrny,

 UPASARGA 415

at) qa ginarantadarsisaer swfaairr

ated: | Ate, TUT: |
eHet: | TaTta a fo EATA TAT sifastta
farRrated: | REA TELMA— ACVFTATON A fetageat
(waQ) sqremea) rermeqer oer aretfa geeMg
SATAITA |
MqqregrercoaeTren at sa TATA |
sor ord a cette a ar aifta Farge
Ho to Qolesalk
Thou yonder
(‘Spirit of the forest, Spirit of the forest!
is it that thou
that seemest to have lost thy way! How
Has fear perchance now
dost not ask about the village?
entered thee ?’)
aAcoaitat
2aWAauTIA | 2 acoqtfa =acoaifaead
srrarerfin qari aatfcerd: | faeterat: sefifer Ba
caelataia agate |senatsefine: 18a" | TOTES |
a! at a
TeyARATE: | AT HAM a ad agor
ear vitae Raven | eRe See FTA TIpro, before in
d. Gk.
Skt. pra, Av. fra, fra forth, fore, forwar fasta
Place or time, Lat. pré », later pro befo
re, in front of,
fa remembers.
out. TAT
stands, sfir@g stands forward
i.e. sets
far empties. |
Teycta forgets. qzata fills, seycar
in
para away from , Gk. para, Lat. per
Skt. para, Av. para ,
vergelten.
Perire, Eng. for in forbear, Germ. ver in

RV.pe1. Se Le
air caarr Parettaerrmarta TTS
416 TECHNICAL TERMS OF SANSKRIT GRAMMAR

“O come ye hither, sit ye down; to Indra sing ye forth

your song.”

ama qa a afufiacam RV. 1. 164. 31.

ara qa acta ooraa. RV. 10. 17. 6.
Wandering about hither and thither, knowing the way.

f gar Tear areafwer Neet goat crcafa

RV. 10. 8. 1.
‘Again goes forth with his lofty banner; the bull is bellow-
ing to the earth and heaven.’
Pra, prati, para and pari are all derived from piparti ‘pass’.

entra ihagear |
Sr mteafgeare—efirra: |
arg: | afigearsgat) saeco sac auth
Fafaga 1eae |
aha ent ta afiatienet arate |
aqIeitaae || [Wo Ae 212813|
("To thee, O Savitar, Lord of choice things, who helpest
us continually, we come for our share.’]

GENT qsaag: afsarcarg: setae | afegram

SAAT Aras | war wai 2 aaafaa:) AA
thay ageat aratat ecoharai aataiart | 2 44
ar rea: waorforat adat Thea: ararejarert SO
ta aramae | er amye.araeaiggs: | ATI:~~

a at | WAT ACATZ AMTIE: | ARENAS "EE |

Skt. abhi, Av. aiwi, Gath aibi, to,» unto, upon, against. Gk.
 UPASARGA ; 417.

amphi (cf. amphibious), Lat, ambi (cf. ambidexterous),. Germ.

um. Cf. Panini’s rule: weayxotarfasdt ethayey ii. 1. 14,

afer car gidtad asf Ged AT | Ao eo U12A1R

To thee, for thy first draught, I pour the Soma-mingled mead.

fpr at arat aTIStl a TzTAa |

cfacqefad aa aca asfaarean | fKgmsast wee
qakgQaer qifaaieay, |
Srl nateatatera: sifesterare— safe cf
eng: | sataaenfaqer arte) szTexny
[ Ho Ho voles J—-
matey, qararay aa arar geal aftaETT
afataca ez ATPATTA, satat TEA Ta TERATTA |
[‘Bending thy shafts through sacrifices, Agni, whetting
their points with song, as if with whet-stones. :;

Pierce through the heart therewith the Yatudhanas and

break their arms uplifted to attack thee.”’]

aac | afaccan | aafcfa tal actart wat

afaeda
WI a gqda faa | ag eq | qaqa
‘Gael AATAT SATST*
Wahaaaeifieaes: | 2% CFUTANA::
wT arqenara af sgigaat = FA are
ST:
aTgararsy
TRY SAAT HATTA agence aaa
anfta: | eafadamiaa |ag ome | Save: ee
Tala: | fear: le aaa |safe
O.P. 12927
418 TECHNICAL TERMS OF SANSKRIT GRAMMAR

satorfa: | fast) spr aga | area wea aI Alas

cam fagr a sda: satatars farcianaa, WEA S"
feast: | aga ofauetie sfaarar cat ag aarara |
s.
Skt. prati (in reverse direction), to, Av. paiti, paitt, toward
against, back. Gk. proti, in the direction of, towards. Lat,
por in por.rigeére to reach out to.
prati 1s pro with fi-extension.

oftaen RV. 1. 113. 11 qfrste RV. 4. 50. 7, sfafert

RV. 10. 24. 6, ete.

ata @ rafagtaard 17411

am: ) afa oq rdtarfarfrari aqa— afane:
garam zie | r
ma: | afag afagtaaramed: | afane: anita
ah) aa afe ar aia tiaat trea asad 12494|
TUE GA aq aie: || [ We Ae opal? |
mounting heaven as “twere above
[“Soma, thou shinest
light’s triple realm,
And like the driver the Sun, thou supurrest on (thy steeds) J
ft) seaqqaenfaatanimma seq) ala ah
RTT MAT waka aera) aif fa A
areqf |degfreanmata aaafa |afamegeraata 22% |
aera arnft; = din tae via ea aren
arardtag: | trea aWMTariqaaanes facatt —
arenas sfagaara zs: | ATE: ANTH | ate
MAY BETH eR Tee freqr are rat ane
 UPASARGA 419

TT: | semaadsa wea: | seufafa arrafaaq: aa

THUAN TATE | aeanfagaa @ gu! ca
antiga | anfiewd a qlngeaq,) aa areaic enfa-
SEI | Bh Tan sigs, fage aed safes
aq: ) wzafadfagta ai a qifaa) agar ay ate:
afzaarafa age fra grata aaealg anata ata Fray
THAT: | TAT CHIT TATA Ba a Srifast-
aframsaaa, safasimaear maaan, dae q7-
faq wait fra anda gqfeqaa) gue: gen afa
Rere [wee ] saTeTEAe |
[ azar wfa aan wey 2 ae fara: |
agucta arg ara’ afacifaa | [so Meo cineier|
[Thou art a good God, O Varuna, in whose throat
(i.e. the ocean) the seven streams flow as in a reed full of
holes’.]

fatger | aad 2arsf 2 cer aqaare ae

t aq an faa: a agi aafteraat at) Ara
TWHeaama aaftaan, saat) Fer SAAN
ataeq: faagatsfe ctaaa| 4 argerctea agT=ara
aqafa ar, faq? wet ag wees Ag
ld argureaafarag | qafaargacied 2 SAA
WT agfererfiya |attrarefiee |SIATATE TUT: |_
Wretar pra) gferertee| HATH TTT _—
TI | afifcfa ga: | ag TIAA Tae |
 .
A?) TECH NICAL TERMS OF SANSKRIT GRAMMAR

under 4 (AAT P. v. 4. 69.

Cty the Varttika cafavaraa
further, besides,
Skt. ati, Av. aiti past, over, beyond, Gk eri
Lat. et besides, and.
Gk. eumenés, SKt. sumanah,
Skt. su, Av. hu, Gk. eu well.
Gk. dusmenés, Skt. durmanah.

fag fetal: fasta eR

am | fagfeaarafeett: srfacied HMA: | SAT
aUNA | Sl facrafa: guafafefa, at [ certs |
qarheea a afamoafa: Tagaisa HY AMA; |
Syed TAA aranHS THAT facad ATT |
af | ‘Sftaaq afraae ara aera [2219] ATRIA |
am left,
[‘When I resolve : I will not play or go with them, [
behind my friends as-they go away. And when the brown ones
spoken, I go forthwith to their place
have of
thrown down
assignation, like a woman who has a paramour.’]

remaer arena ad) waraferdt arftaeser: gee

nag eareadha carnfa aac: | FI, az
afaarft a gaifacarifaftfa , aaet a TR:
anni sfiaresaa: sa AF was eraqisieae: |
aftaca: afiaryaes: | A AETTEMIAIT: | ART FANT |
amet adaaeart | Fat an: Aral aye: Tey TT eter
atanma weg gaedifa az a AAT TAAA afayaara:
afeeasa aneay Thr, ete) Teas) TIAA;
armfent freed deed ntumrenrced suftoia) am
—

aufcett aft arftanarned oftasa dzaend af aiaifa |

 UPASARGA
421

Skt. air, Av. nig, nz’, out, forth, away. Skt. dus, Avy.
dus. duz’, Gk. dus.

eaadfa fafameraiay i231

gm | safe fafaoendial—trreifrvase ef |
[ fafanet aia Ranimafeatiggatagaaietna
aaqisfaamnaae fata gang | aati; fafasrereh
: |
fafazers aq fafaaerata: | fem ee STATA | J
engi | famet framaq, Ste faastdsase che I
a@ [ we Wo vizie J—
san fag san agea cafaaj arta faaae
|
at at auld afar am dara wewad a STeRY ||
[““Rise up, O Agni, spread thee out befor
e us: burn down
our foes, thou who hast sharpened arrows.

Him, blazing Agni! who hath worked us misch

ief, consume
thou like dried-up stubble.’’]

qneaeaay | Ta) sf ® ae Seaway

qaagea fara) fee aaeaisenfa: | faxrez
arratizaan ara | fier a Farhi RATT
UIA! FT Te TAT Seine siake <az |
earaaig, frase ze, farm’a afa! fie aensarey
efaug ata a ot ara—é faraka aa a: areas!
afar afraqaa, aetrania, acre franfr RTI Cay
a an nufa) aerstat ata ata: Hcat— fantiar: |
acafa ze) fafrae awada) werames HSATA |
arefira sna freetterfrare: | >
 TECHNICAL TERMS OF SANSKRIT GRAMMAR
422

aaqcaaeararecoa [ iets J—
ot at aa gener fagiegaea Baisa wat:
afar afean: srgarat fasar gaife a nee 1
away from
[(‘“‘Do thou who knowest Varuna, O Agni, put far
us the God’s displeasure.
e thou far from
Best sacrificer, brightest, one, refulgent, remov
us all those who hate us.’’]

cfr oo aiseara & at agra STA | fey? aAetig

|
THAT | THIET BqeET TAPATMTTE It GE: HVATATA
equa.
qearegate at Oe: aH ATTATT: | qatate
at arta:|
ARHATTH | |ATATTATAT: | aqadavar
atest
qaqa aaa: | wa a afae: afanaa aa
ara: | |fsa
agama afar | agar: ara Fiea
aaifn) gaifa genfir secfer at caaratta | TEE
TRAN Fy) AAT aeraisTaaeTe:
fe faerie Express the sense of antithesis, i.e., are used in
expressions.
one of two antithetical clauses or
in into, Gk. eni in, Eng. ne-ther,
Skt. ni, Av. ni down,
be-neath.

Skt. ava, AV. ava.

a agar fraat atfa afar, RV. 3. 2. 10

and he descends.”
‘Busied with sacred rites, he mounts

sean qafa vr fa qet reat A |

aaqraatia sitaa RY. {. 25.. 21.
 UPASARGA 423

Take away our upper bond, cut off the bond

between, and
loosen the bonds below, so that we may live.

stzaaar: siftateaa Fy)

gm. sfefe wate ca waa: Sfaatsaare |
Sz arate |
tHer: | Shraaw: mfasirar AANA: Wye
RATE | SITET SHA TA: SR | 8 [so Ho g1e12]—
SHI TTT aes ofzsrr afi |
aia ara vat ar agraea |)
[“Raise theyself up, O Lord of the forest, on the lofty spot of
the earth. Give splendour, fixed and measured well, to him
who brings the sacrifice:”]

fa fasntaaemne) wa seq wet wag) B TATA

eaetfafrart qq) asia gamer ofaneaaarcaata-
TAM SAFI SPT TweT: | asafs efcnreahs | afer
9
araemg Aqaz | Gfaeat afeoaae afs safe | ahead |
qatar: qaaan: | @facar waka mata saceqralea-
wan agi«largar watearfzar (2) Hizara: | aetfan qd:
ret Aaaafe | wet ame wat aged aafy
aarrata zava ofa crqaatfa agarer aamraeay |
factiargzraco[ ae He B1¢12 J—
afager AAAI, FRANZ Tal ATaTAY starz BwNey |
ar amzafad aaa saereq aed aha y
[Set up to eastward of the fire enkindled, accepting prayer
that wastes not, rich in heroes,
424 TECHNICAL TERMS OF SANSKRIT GRAMMAR

Driving far from us poverty and famine, lift thyself up to

bring us great good fortune.”
efi) afager ardaenaadtae aisaztd saan: aa-
macaa ifr: afacaa: | sae wqfasenagaaaey Faia:
aaa: ANC BATA Tcatad fet atrevar ser sateat-
aaa: aaa arafert wet ay) sata qadic-
etary) ae qeNaT | Fl ag wena: | Aa: AUNT
nfaaea a: anfa: d@ wuafa gaa arama: anaaa
saa wa’ faperd: | wed adhere guna fsa
gaa at) waeraer fiat: gaat ar emaaa aAa-
fat: |
SfaAAHVTAA
ell
aw: | afreaataraniare—aizerdtta |
arg: | afaatararaate | ale agat srt) 32
[ Ho Mo 2olzeagir |—
ft asara dared od a aaife araare |
ar ani rat TaAsa STA I
[‘“Go together, speak together, Jet your minds be all of one
accord as the Gods of old sit down unanimous to their
appointed share.’’]
fa) aaaein | ARAN Aaa AaAAawAAT |
aa a qgenfaawas Taraaagz wa: sa yeaa | Far:
Ta AA Syal AN a SF AAMT: BAT Ararat TaANTTAs
Haar sdiat car, wi =F Meas, drs a_i
qftasa dad Fett aA at aaifa aman aaaft
 UPASARGA 425

daza a: gear aaifa mad stadt FARM | Far

Sqreaanfaat AAA: |
Skt. Sant, Av. hant, together, Gk. sum (symposium a drink-
ing together, aqifa: )
Skt. w, Av. vi, vi, apart, away forth.

at oath: faqzafa zaat a a Teafe |

aa: asizfa faa: || Ae Roles
“May he who sees all living things and sees them together
at a glance, may he ferry us accross Our enemies.”’

a ae fa a Tea folescle
sqqeaaea ofa ieay Rell
at: sae ofasaaieg: | faserfe, aa-
zarfa |
eaeg: | caqaaer faa favaaraateg: | fanart
aqat facaazgar zeae: | aasfe fa maa anfee [ sto do
golgcolz |]ANRSAAT AMeT [ Wo Ho yoruxno] fa

narratfesrg [ 5120, 4128 ] area |

fam a aia: pac freer: waa ar ATA TTA: |
ae dar ofafars feed fa saa antes Han az
(‘Dwelling on the mountain like a dread beast that roams
about, hast thou come from the farthest distance! Whetting
thy bolt, thy sharp blade, cut off the head of thy foes, scatter
those who hate us.’)
 426 TECHNICAL TERMS OF SANSKRIT GRAMMAR

ama) we wadtiequarqae ane fanaa |

Wat wal aa: fest saat ar Hac: afaanal are
fasgfesa acta, fafter: waerta), aaa agig
afaqerara tfa ati waa sac firfesr eaticeeadd
faatatt a ama, waifasanfacar | a: ada acfe a
ata faacatta gat.) fafeer firfoia afina zasd
fasaifa fafest:: wea gad fafiera @ aa aim
ATR | TUAAT FTE AAT aa | TT Edis
rag | Wee FUeaeagufzad: | am a wat
ag’ dara trae | aia) vafarfasat | mead zaor-
qafe 2 aq!) fed chest aa fa aga ae) aefas-
‘al fated aqa ate 4 aeara:, 4 aaate au:
donation frgzea Sttarsqaaders: |
[ARISANT AIT FLAN A: TTT: Yuga aA |
am afte wagner eo fad afes aiacany|
(“Drive to a distance from afar the foeman with they fierce
bolt, O God invoked of many. Bestow on us wealth in corn
and cattle, O Indra, and make the prayer of the singer richly
rewarded.’)

RUaSstacay | atraase afanisrnaaqaa: |

qey aaa aaieafagT saa: | area fa qezq-
mata sR grained: | ze A zT-
tania afagt areeaeanagal vale ug:
cra iia afanqsat wat | Farad: | afaancizer
TAR AT aaey |) ATTA qT] wa arrfa aay.
 UPASARGA 427

rama aqeaoeaafasfafa grasa qatrsgt FTT-

MAPAMATHTN | ZITAT NAHVAT FIH| FTAA-
facad: | AasaT | TNSTAE MTT AT az: aay AT: |
qega @ agfega! aa fea eR aed ate aan
Mad | sad ae) sare rare cafes
dgafrad: | sual aATal asadt taadt aera: | J
afeafs ATERATATATAA 117911
qn:) afeafa areqaqaae | agar fate
AISWAA , AFM TATTATT: |
earz:| afeafa Greg TaTaa | AATaTa: TATET:,
ante | agerdafafa areran, wgresdiarca4e: |
The samahara compound shows that the two meanings are
not entirely distinct, but that the one has easily passed into
the other. Cf. the boy takes after his father. In the case of
adhi we have afaaqfiaraaeaet’ ar |
Skt. anu, Gath. 9: 2anu, Younger Avesta anu. Cf. Gk an,
from which arose ana after the analogy of Kata.

aaifa Satz eel

am: |) adifa damare |) afamstt eng) agaisti
eae |
mea) dat: aqeasfara:) af fas aft café |
fay cqfe Sard: | 32 sara ag afta arafa [ao do
gjgols] tf) satiate [21z¢] saTeaera |
428 TECHNICAL TERMS OF SANSKRIT GRAMMAR

[sa ey arst farafot acoafa dari ast afaner

: arate |
ha sant ag aa ata sasieseagagi aN |
(‘And the prize runner rushes onward before the lash of
the whip, bound by the neck, by the flanks and by the mouth.
Dadhikra, having scraped up his strength according to his will,
springs along the windings of the path.’)

qazqeaga | afte @ art ana wArWAaTa|

aftqar at ea aeafaar ar atarfaea: | feraft Sao
Sc | fedtad: wasaaAa gas | FcoAy
wpaTteuanial ofa: afaad: qroafa ware) fata
ana agi weaag | afta) adifa daar | sao
7: aae sta: atsfrra: | Raa amici aera |
aiafa ara aaa) ag RA ania aanfase
Tal apaTigney | Adaawhaq | wala: aaaal agers
az way, agaradfa ag daaafa at qarasifa|
aqueissiag | afe: alfzet) cat at a afar:
qeqaay | acamnq, wRofanfsaat; — seqsy
Ta | |
Skt. api, Av. aipi (1), Gk. epi, Lat. ab

THITHAA LEI
aT) Taeqaaag arbres are | |Tas |
HeT | TIRARTSAT TTaAATHT: | a—sq zadfe
TTT [ 2ojZo]go | fa) aa [ slxo ] saTeqRaAg Aa: |
 UPASARGA 429

[ at at at aera Bas wa ataT: BIATATAA|

zq agfe arama aeaeahiaee Gan oft aq
(‘There will surely come succeeding times when sisters will
act in an unsisterly way. Throw your arm round another bull,

seek a husband other than me, lovely girl.’)

qmManaae wal 7A AHR] al Alsaat waray |

aTHId W=estaaaa Baad | A A aarad, ASAWA: |
asa armfacaa) at at sau saat gant
gist: | seamed & ater aeifaeatea Rarit aaea
at. | aa ay admecstaaftern ama: afies:
sued afeerta aaa | afaamea sift: sada wer
agaifa afer aaaeerot RA] Ta STAT
sqaa fe) faataicomter | aeleqnawar aaRITeT aHT: |
sqagfe cHaaaTaea aA urer, srarfiefiers:|
aca at feed aad tat asfa @ aramenciaa
qfrcaerenfirered: | aTRATMNa a NYA FSA Fw

mina | gaa wf aq ad) =weadaes ofa.

eAeT: | |
sqawa means accretion, It is used in Grammatical literature
in the sense of saiatay Cf. sarsfiaa a i. 4. 87.
Skt. wpa, Av. upa (a), Gk. hupo, Lat. (s)ub, Goth. uf.

quifa AaAVATAR ||Rol4-

at: | wifs aaararame—ofeaiadifa |
 430 TECHNICAI. TERMS OF SANSKRIT GRAMMAR

wer: | ore afefasafefa | |ae—

qtrat fasyat ad aia a saa Efe: |
TAG AT aa; Aezeactt Baa araafE fir: |
HO Ho 12 01912
(‘Hence sprinkle around the pressed Soma which is the
best offering, which full of manly vigour has flown into the
waters. (The priest) has pressed out this Soma with stones.’]

aarorasy | aA a saa efaeqeend: | azz.

aanietenea a ofafateerena |) ameenemnge san:
qm: wae arated | etagt aet a: ata saaa afa-
waa earana waa a parqma at efacathy (2)
qi wat Agee whamaadeaistacfiagea
wade
feat:|) wera agi qadadeamiat wat) araaa
saTaaia Haf Haseaa:| fez, aMrafafa a aaqsy
fefiar, argaa, atfageaa gatshaga: ata afsti:
afavanrafied ufta: faga, afantis fas) ht
zartfaar aftfagae aaa: areaatern) gaafirgd
AAT |
Skt. peri, Av. pairi(1), Gk. peri.

area fearatsay at [ ate] yee

am) wirgefrarnie tay at) afafagth
afaqta: =f |
eer: afaeqafraiag—atatasta, aan |
waar fraqueafrarzaara |
qtaeat ata arate ao Ho e305
 U PASARGA 431

(‘The purifying (juices of Soma) have been effused from

heaven, from out of the grass on the summit of the earth.’]

faye: aeaTgaesy| =saafaeramaaa—maat &

gaat yea fea: ataareacfafa | [eraqerararny eens] t
a@ia AeaAT ATAAITA=Ad | TAATAT: | WATAMSisa
ainqaa:, TaanfanaT: | aa Taare TAATE | AreMNTaTT-
Aaftad agaaaqo [ aie |, aeTaa aT] «Tanta: aay
frq: Tan ZIMA | Tq reaTisaaacaieaaa ATaVya |
aa: nAafiaaad | |as: Haft agaae ofe aq |
aeaiaqa | ‘WaaTa: atawraendad agqaata-
qaqa eat) aate Ge: | Adee saeqar areragt
am cet: | He Seid) ofear ata soft arafa |
aig: agfeed vafa) aafseat waaaawears) 2S |
quar feat araatg zaafeare) =oaaR araraferet
fagaifa atarafearm | aera ofaen afeaamear sqft
aafepd efsaiaaa | ‘ean ‘atast’ cf Ae: | Beaz|
afa aazd cate: | sxfa araqzezcmne” |*
Skt. adhi, O. Pers adry.

qu: | se ataTeaaatg RAedaaant wardt-

any) Aa ae: RAGedaTaaK waedifa a dae |
sami: fart [ ies jefe sfast areata frar-

\meda adam yw sft we MAI A—wloeY oeyi Daz

argo uc | sfai
 432 TECHNICAL TERMS OF SANSKRIT GRAMMAR‘

qaqa ait a atat; scant fe fraweaaa aaa.

wHeTeaA ||
CIMA aaa WeTA SaPaTACAT: RAN
ST |o |
Chez. | Oo |
[Lalitavistara (v. 1. 52): Waa) sme) ana)
RPA! MMA! AIPA), AMT! mcg |
SMUT | ATA ITAL | «AANA |

GATI
The technical term “Gati’ is the name given to certain
particles and indeclinables for purposes of compounds and
their accentuation. In most of the later systems detailed
treatment of compounds is avoided, for they have become
much more general than in the days of Panini and when it
is found necessary to prescribe rules for compounds the
particles are merely stated by name and the generic term
“Gati” is not mentioned. The Gatis occupy an intermediate
position between nouns and upasargas.
The name Gati appears to be due to the close connexion
of some of these indeclinables with the root gam. SJeEy is
the most important of the Gati’s’ Moreover it begins with
Qi a a a

t aria midiaqqeaiaaa aaa, afafaaa agi fala syaaT,

wafand aanaaiae «fea ficaaq) wa oq amama armafy %
RAAT MDA: |
 UPASARGA 433

the first vowel g@ and occurs in the Vedas. And it is often

connected with gam. Ch. aw TAITIey, P. i. 4. 69.

RT. and ST. use “‘ti’’ for Gati though there is the v. 1.
Gati in RT. 29.
J. also uses “ti” for Gati. Sar. also appears to use “ti”
(i. 1. 26).
Hc. uses the term “‘Gati’’ in i. 1. 36 and defines it in iii.
1, 2-17.
S.K. uses the term in the rule wpaeqsq i. 1. 131.
Sam. does not use this term but mentions all the particular
cases in the krt-section in rules 447-483, Similarly Mu. does
not use the term, but mentions the special cases in the tanadi-
section in rule 767 in connexion with the root kr.
Su., like Candra, mentions the special cases of gatisamasa
in iv. 3. 41, 45-57 but does not think it necessary to use any
particular technical term in this connexion.
PR. has a sub-section entitled pradi in the ktt section (xvi)
where the gatis are mentioned, but no separate name is given
to them..
HN. has no objection to the term “‘gati’’, since Krsna
is the gati (refuge) of those who have no garfi,
The term is found in the Pratisakhyas belonging to AV:
Thus in Apr. we find:

afaqat aa ure: afaq carcieratea: |

aneaa wareaafas cf frratayti eiee
OP. 12928
 434 TECHNICAL TERMS OF SANSKRIT GRAMMAR

It is distinguished from upasarga, since in the next karika

we read :

TaN GMeMNangaTa faze |

Sait AT AAA STAM FART VIZ?
In CA also we find :
aaa THfaeaal wfacaal at ala Aq WTSATAT|
Fearn afa: gas) 81213
A PATA AGAMA: TATA TE |
almiqasafafa: AAs HAA |) VIR}
Qio) (saan: Peart) cisise ) wha) Hatfe-
Fra-sTaar | e1elgo-ee
Qe | Seeqatagacota fst were
Ws! aarare faqar ca afaqasfa Taz: |
go, garagancrfagraa aa: | alee
Go| wa Stegaqiera ward aH salsa afadan: |

KARMAPRAVACANIYA

Karmapravacaniya is probably the longest technical term in

Sanskrit grammar. It is evidently derived from “karmapra-
vacana’’ with the secondary suffix cha (-lya) and means literally
“concerned with the setting forth of an action” Acc. to Indian
grammarians it means ‘that which spoke of an action’ (karma
i.e. kriyam proktavantah. bahulakat bhute kartari aniyar).
This probably means that which one referred to actions but
 KARMAPRAVACANIYA 435

now governs substantives. This agrees fully with the views

of modern philologists.* Another sense is possible. Panini
mentions that the word pravacaniya is formed as a ho ANITA,
He also lays down that the second case-ending is, as a rule,
used with Karma-pravacaniyas. So the word may mean that
which is used with the accusative case-ending, Bhartrhari
says;

fHAlal Wahl Aas Atqateyq a Aa: |

arte FRaraqTadt Arana g ATT: I
Like many other technical terms of Panini ‘‘karma-
pravacaniya™’ 1s found in the Pratisakhyas of the Atharva Veda.
aqaHRT rs attaTraahaaarsfafaararfirg CA. iv. 3.
Baa gra za Gam fare: |
aqaern: HraaaAAT Set at fares |) Apr. i. 1. 10.
* Tt is well-known that at first every substantive marked the rela-

tions of dependence, of interiority, of instrument, etc. by modifying

‘ts formal syllables. But this method of expression was at once
complicated and insufficient, It was complicated, because substantives,
not being all constructed alike, appeared under different forms in one
and the same case (gen. domini, rosae, arboris). It was insufficient,
because the cases of the declension were too few in number to express
all the relations that the mind was capable of conceiving. This was
the reason why adverbs were placed by the side of this cases to define
them. But the habit of placing the same adverb by the side of the
same case could not fail in the long run to produce upon the mind
of men an effect of which we shall presently have other examples;
between the inflection and particle of place or time they supposed
 436 TECHNICAL TERMS OF SANSKRIT GRAM MAR’

K. does not define “‘Karmapravacaniya’’, nor does it mention

its special cases. It merely prescribes the accusative in con-
nexion with Karmapravacaniyas in the rule aringqadiay Ca.
221.
Sak. has no special term for the Karmapravacaniyas but
mentions the particular cases in i.3. 101-104, 175 etc.
S. K. (i 159 ft.) follows Panini.
Similarly He. mentions the special cases in ii. 2. 36 ff. but
avoids the term.
In the same way Vopadeva and Padmanabha do not recog-
nise the sarnjiia but frame rules for the particular cases.
PR. recognises the term though for the sake of metre it adds
a pleonastic ka to it:

ANT: Bean Belsvaraaaa: |

aaah fiaiat eae faraiisfuante an eiee's
the existence of some special, connection, some relation of cause to
effect. Instead of considering the adverb as
a mere determinant of
the case, popular intelligence saw in it the actual cause of the case;
a
well-known paralogism that philosophy designates
by the formula
‘cum hoc, ergo propter hoc. But when a paralogism
is universal, it
is undoubtedly not far from
giving the impression of truth. In the
matter of language, that which the pecple believe itself
to feel, passes
into the condition of reality. Adverbs of place
and time like apo, peri,
epi, pro, meta, para, from having been
the accompaniment of the
genitive, dative or accusative became the cause of these cases; from
having been adverbs they became Prepositions. The minds of men
endowed them with a transitive force.”
 KARMAPRAVACANIYA
437

HN. uses Krsnapravacaniya for Karmapravacaniya : Hvy-

saqaataa.a feaiat e704
The Karmapravacaniyas govern cases of substantives
generally the accusative, sometimes the locative and the
ablative.
They retain their independent accents. And they do not
cause
the mutation of@ into q.

We learn from Durga’s tika on catustaya 228 that in the

Apisaliya system of grammar samayéa, nikasa anil other
avyayas
governing the accusative were regarded as Karmapravacaniy
as,
AeA asa AAANTAat Raa zefaTy nay }
AATATAATENTAT | TANYA SraMica: | Rape.
qaratfagtasts samc sainfames Fc |
aufe fareaitanasfarhraaag¢an: sea) wre:
aaTAagaataaT ATTASTHAT: AE
faqrat | Prufaattraftaarraiasetaa: arigag_
MAT: | ATA
frat Raq wg fared agai qeparhy = |
agg ahs: wHfaserafraq |
[TFTTTT alee ] afar |
[weenzarat =afar | wenaanagearaay |
aaninareactaaragiasty MMABATT | aTeaTaT-
fagarare—aarfafa | sarareag: frereiicnaqrant-
ara swartfrre Aah anat: qa sete) frodeg
aiztiataeaneraag: wereaa ff @ ast sare.
438 TECHNICAL TERMS OF SANSKRIT GRAMMAR

fern: | aq wanaaraaziamad aut cafa asfa a,

quai asaaifa | waa ceqeremfy) Aet:
Raz ad
aqaatfe: | sarrfraaaaai aquacaifzatara: at-
saimfs:| aa ames dfeamag omaaizeral
fronnateaantisafaa a: dfeararn: area
anaes: —dfeat art as RTefaadaq:,
AMMA HATAAA AAT SAAT: | SaAHRHAZIAT |
The distinction between Upasarga, Nipata and Karmaprava-
caniya is clearly set forth in the ‘Vyaktiviveka’ (ed. Chow-
khamba, p. 37). ‘‘In-the akhyata’’, says Mahima, “the sense of
bhava or action is prominent. The upasargas, nipatas and
karmapravacaniyas are those that do not denote things, still
according to the diversity of their function and use (ie.,
position) in the sentence they are divided into three categories.
Thus those that convey the modification (lit intensification)
of the original sense of verbs are known as Upasargas.
They are: pra, para etc: Those that have their forms and
senses fixed and convey the conjunction etc. of actions and things
(verbs and nouns) are Nipatas. They are svar, etc. Thus
in pacati ca pathati ca (he cooks and reads) the particle ca is
a nipata since it conveys the samuccaya (conjunction) of the
actions of cooking and reading. In Devadattasca Yajinadattasca,
the conjunction is of two persons.. But the form is fixed and
the position after the things conjoined is also fixed. ‘Those that
define the manner in which a particular action affects the noun
or pronoun are known as Karmapravacaniyas. Thus in Saka-
 KARMAPRAVACANIYA 439

lyasya samhitam anupravarsat, anu defines the relationship

between the verb hearing (niSamya) understood, and the noun
samhitam as one of cause and effect. Here also it comes after
the noun or the pronoun.
vTIAeaafagia swamieata arad Far7|
ddim A waged: RATATAT |
RRTHEAM SRT: XR
affargzian | adataq aa, fama fae
WMATA | ATATEATAAT: aguacrpiigentelitial faatar: |
fearfaassfafaaaagieann: | aaa, fa ofr
aah | Taga acart: wafaar afa asi are ad aa
aa amnfzg ava daenfaemnafafi: adsaactat ater
way: | grant rans: | aa arama) aa daar
wadifa aemaatha | aatdetanttaqraata: Ha: FATT: |
frad arenrent arardianiaa aasaatiarafe yaaa
az) aafe gaat faaad fra xf freregern-
STAT Nad | Brae g frafeaearst-
SHIBATA
fasrfeta ead) aut wmaaarsfe aat eeak) ae
qanafaearan wa fairey, fidtfener g aal-
qaqa cafe) aa swatsyaranta fra we qq
ad, +g aivaeatarnfaf wafdtrataaacaata: |
arqaraifa wqudifa darat faga eft ad er a
erage | Usa eifearayqadiganda aa-
aareatiag| ae at aemenasaar fara watt
froranenta dai Hae: | Teasewaft qayT gy
tfa fear_|
440 TECHNICAL TERMS OF SANSKRIT GRAMMAR

It may be mentioned in this connexion that only some of

the upasargas govern cases of nouns and pronouns in Sanskrit,
several others are used as prepositions in other languages.
Thus pra is not used as a preposition in Sanskrit,.
but the
corresponding word pro is used with the genitive in Greek and
Locative in Latin. Gk. pro thurdn ‘before the door’, Lat. pro
patria ‘for one’s country’.
The Sarasvati-Kanthabharana has the rule HOTTA:
1. 1. 159 which is explained thus in the Hrdayaharini :
aH Ona: HATA: | RAEgT aT finale
qa anrdageaaraty fearaceitsafsaa |) aarfz
farkaincatiraraangaztaa: HANaaha: | |{aT
mae dfeamg sraufsfa, aa area <fear-
FaIMattaqeaRiaean: aaan facaaffacafa
ragt tqeuniern cH | HATTAge meyT-
aida Ara |
Helaraja explains the functions of a Karmapravacaniya
thus :

eae: wearer: Groea:, svarcinarfaacara aez, frar-

Waa | aa ahaa Prez: Waa, afar |
wa gad at Macey waar: | agar) aa:
eacfa, ala: wan, afd mma samt, fen fe
craned Pravda xa: dasat| aear-
carenfastasa sadasafasa at aememafigarardic
waco fa fierce cgaafr are cfr
anata | wat gd frame gat ofa: afacaraesered-
 KAR MAPRAVACANIYA 441

aferdita faotcfas: sfafraaernicearagaad as-

ara fata aasaadtraeaa | ae ‘sancqeay’
‘mera war xararaqenqeaaqearaan sfafrnfatae:,
aarti 9fufatrarafia genfe; afag
arma sfafraaihanaaarcarnn af ara |
agar ‘Wad, gen’ aq mental wcmaqaararfira
qaaranrad, feast aamaa afafaraa: | aaa
wea | add Baa aug: sardafa—acarg-
STU MAU aasaafa* | Taetai earqaaTaannng”
faagaiaa ca, afarafaannfaat frenfe a ocfiad|
amradndan frat aasaaiiean faaafefaerarmead-
araaey faaaaaiaata | agnq—
safaear far afaq araed fafrade |
yaar fares Gea sad Hag |
a aqardcaraan fafrard farenae |
HATIaMAa aa aa faarag 1 [ato qo AVA, RoR]
afal frame faaateersfaaaq, auf mere
dframg orate afa aise dframadonzatanerasean:
aren: @ frratmarmiata cagar aad | aafraracar-
frvafataraeariaarte anesrnyq dffart-
wzgareaaar aa fnraqanensaa zones
aeay |feasts a fanafafa aate farsa aes
aif | aa fragaacnd wart oa fafifreadiarer
FOAATAPTTTATT AeA TRAY | ATTRA
* The printed Bhasya reads: aqmeqswensl fafirrary 1
442 TECHNICAL TERMS OF SANSKRIT GRAMMAR

siafafa faciata ganecaataafharad: | ze a Afear-

fafs afta faaheicf wet frac: 9 feataitHatta
qUequmacaagHaaen faanad carer, agar afar
fredtfacaal | aft arqaerarfacaaarar, faataa azfa-
grief arenas cama ara: atadifa
qrixateaizaica Baeaaaseaa | ASA
fRarat Alawl A ATTAIRT A ATA: |
afta fRargztant araraea g AZ: || [Alo To 2120%]
sft) azar faaant dian geaa:| agama ara)
qraeahas «=aAaTAATTCRRTAT AT | aga Tear
arate faeafa a aa aerate: acafig’ waa afa
qzTatoaaTaciaaa AraSITaAI AaISAATAeT HA-
saadiafrara fagr| ag aazafaad arate: a af
frofasiataaean: aaerenasag areas: fafafa
aq Hea, ada Wareanaghraaata, sated |
relfasd aera TBA aac ag
qamhegaa | ads fe ara a fama
qr arian daneafraaq| ag aut
agama atarefafa a) aa cateadt—maaafant-
aenfaaneaataga if) met agunea qe
Ta Talfaaedifa A AP ITIS STISATIS ATA:
aqaa netlugqfafa nasaadafavs waa) 4
aeafa—
fafraftaa: Te araaeT aA Daa |
nuraaaiaeg a fastisqerag |) [ate To 3) FT Vek]
 KARMAPRAVACANIYA 443

afa| aa a zaagaq| Watta arararacttsan:

auaadaa araeasatsada fanfatsafiada afa|
aa weamass? fafaefmoafaaaodifa: aracafaie-
qaiaiaateear | aus afta saad wyior eft aearfa-
aaisd area caffra aa) aaa waa:
qaiera nage: «fafa aa: araffam: oafineaac-
atalfzfearmaifad Te araeal sae qaamaa| wa-
afaaegquima: sdifa sfaat arezaeanisd area af
Nad, aaa: avaferqarqaiaaaza dsecmnfsfrar-
wa xcaantad | |xeaey aracrdfeaiag orayfeerarfy
eThoMal WA aay Taaratawaagraarafa | qWe-
faarercata dfsara faaaafearanfa: | sessatafe
aasasfefa arafanteen fafietrarmaiffaanaaiaa|
Sfafsararaiza carat zeae: | waama=sasfy a far:
RMYTAA TIAA AA BA THA Teese,
TIT: fraTRaaa AAR THAT | AUT F AwWAaar
wdtted aaa set am fafrafiieasz aq
maaan tft tga) wh aad wype: ch
threaded weartrareenfaat aaset | aftrary-
wiaa: sdtaa wrafsadisgaatgacnne: sffar
vara cafe aia asa) ataaa afaeqay caret
g &: arg afactamact @ [ eieige-e4 ]genfear aa-
saaaradenfrarfcany cared rcteaa Tae aA TIS!
tt:oe
aasreadiad frarna Feitaz|
aeartzfatraad aenatat fe WAIL [Ato To Q]R08]
444 TECHNICAL TERMS OF SANSKRIT GRAMMAR

afa| almag wan fama odd fearfaamnactrancar-

sqamrae eqeaifzcearaaaita atzarfe: |
Bhoja says in his Srngarapraka§a :

cizq uq faafagdanfaaaraenaseiag: nA-

qaqa: | azg@e—
fafanIiseawral APIS ITWAat THT |
RAIMA, tata AAT: |
sfa) fe gacdinnramng? aAsaadiadaifaa
Agmida: Weal THAN wanizatsar arfdafafeta|
aM—aAgean ) sda, AR) sasha ay, aq
aaa || AS AGAaAR || sAkanqaeamandtarg
nfaqaag: | afacara ofa: sfafafaafaziaa: | afer.
qa aaa @ gam afacfeat a) aft:
qaieraariarameagaayg |) afatiat | fears
af | [ao gaice-2.¢ ]aq ag sa ag oft ene ofa
afa afa a afe ata setae ae|er: |) 2awRAIING, Gereh:,
diaat, abrsay, AKAD, ATTA, SANT, Aaya
eaTag, a0T:, fear, afafafs:, sfaaraq, aria, gM,
afamATy, Tat:, ARNTAT, WATT, Wel, BAST,
aizaq, afanre: zfa gifanfacai: | aarg: art:
afaqtr aeal, aft: qsqren, wat sail, sata gsi,
weqagq cHrat xfa |
 NIPATA

Nipata is explained thus in the Nirukta: ay faqrat:

varaacaqay facafa, scyqa, aff adtcdae
sifg gage: | 1. 4-1-2.
‘These are known as nipatas as they are used_in diverse senses,
as well in the sense of conjunction as in the sense of comparison,
as also as expletives.” BD also says: ganqayq ari faqrar:
Ayre: ii. 89. And this sense is generally accepted by later
grammarians also. The Cidasthimala says: faqaa qraafa
eqrela-afena_, But the word nipata is often used in the sense of
accidental appearance in the Nirukta, ASvalayana Srauta Sitra,
Brhaddevata etc. Rk Pratisakhya says distinctly: nipatah
padapuranah xii. 25. From this it is clear that Nipadtas were so
called because of their subsidiary character, because they were
not regarded as essential parts of the sentence nor as having
an independent existence. In the Nirukta we find : Qwasat
afaaiat worrser afersr, | araParataars: vii. 13.5-7. Again,
mg Gh unt aa afaftedtsaka atsfastaden,
faqraaaa sat wifadt waa aiadaa wR) Nir. vii.
20.4 In BD. also we find:
fanaa efaaer aa a ana a aT |
aa aa vata area farsa at ear spse
“That deity to whom the oblation is offered, and to whom
the hymn belongs, will there be the chief object (of stateey,
not (the deity) who is praised incidentally’. TIATEATAT faqrarz
446 TECHNICAL TERMS OF SANSKRIT GRAMMAR >

qT wae saroreg fa: Vill; OO s essa3Brahmanaspati is praised

either as the chief object or incidentally".
The nature of the nipata is best described in BD. :

waaay arty faqiar: AAA: |

HATA a afaMITTHITATT ||
Fat FTA al Tarataget alAaq |
fraraty aeatg GooNTaeaaT
A: I
zac efa dead faqrarai a faae |
aqui wacataa® faqrearea Te TS 4 ii. 89-90, 93
“Particles are enumerated in various senses—both for the
purpose of connecting actions and occasionally for the sake of
conparison. Others again (are) occasionally (used) for the

purpose of fiilling up defective verses (pada). Those (particles)

which in metrical books have the purpose of filling up (the
verse) are meaningless....There does not exist an enumeration
of the particles (stating explicitly) ‘there are so many’. These
particles are used at every turn on account of the subject-
matter.”” Macdonell.
With this may be compared the following from RP.

gat faqrats | eRtke

faqta: TTITM: | LRIRY,
faqrararadamifaqiaaradarataat a ata: |
Raza zeae aeae aga faatat arerfaaiazt a 2 |
RRIRE

* VaHAAMS sla GISIMATA |

 NIPATA 447

Kautilya says in his Arthasastra : fratfaaitean: (v. 1.

fara ) cea scaTi:, aaa faatat: | czle
The Natyasastra (xv) of Bharata defines nipata thus :

qa
eae
ars ASGHAR AT
qeaifaqatea 92 ataretient fararaeg i
K. uses nipata Without defining it; e.g.. ajzrar sates AT
faqrar cat WHIT | 1. 3. 1. Durga, the commentator,
explains it thus: fafga: wa astfafa faq| asaifi-
maa afrarnaanataratseaa: fag: | aan ame—
armeararmarehrar facaxdifa faaran: |
J. uses the first syllable ni for nipata. Mu. follows J. Sak.
uses cadi for nipata. He. follows Sak. Sam. uses avyaya for
nipata also. Su. follows Panini. Following in the foot-steps of
K, PR. uses nipata without defining it in qYarat faqrat az gs
a Raed i. 79. The commentary Prabha-prakasika says :
facraca aureafga ara fara cfa wafer aq, 2
ATATSIAT: |
faqrararania aaazafa & az: |
aaah: wat: aa Tseat fazalag [|
qara afa deara faqratai a faaa |
qaiaaamae farafra we Tz I
Curiously enough HN. does not disdain to the use the term
nipata.
qio, sTteactfaura: | ararsady) sea: |
RISIRR-C
448 TECHNICAL TERMS OF SANSKRIT GRAMMAR

So, fir, arferam, gfe) aienersc

Qo Fo, TAPATIPAITAT | LIRIees
seuerftolt | waeact AAMWASARG HI anal
|
SeeaaNeaer: | Fae! sa aacasa fagaedifa faatar
Psctal ara sf aaHt: RISTATHT: |
ATA TA BIST EJ: APATALT TART: Il
[ Cf. ate To 21%%2 |
da) meniuafatifiragunfeameaaaarattt-
waists TAIaqay Rroaeciacaa fait
ca aren faq| a ag faa avait ag
qm fatara fafafatarar afafafiaaraiaata oTeT
araatea |
Go| ATATSAT VWV1Ae
yo, arfatafa) ce
@te | oatfatttata: 1) 9122
fo, arizat faqradant: | 213e2
It is interesting to note that nipata is used in practically
the opposite sense of accidentally by ASvaghosa in his Buddha-
carita ii. 54:

aq a aa fafad aare afiatetord eqfaaa Fag |

Zyl au gaged gal a ad a ararfefa aaraa: Il
Similarly in Saundaranada xvii. 31 also nipata is used in
the sense of established, observed: In cases like these the
word is the past participle of the root pa to protect preceded
by ‘ni,
 AVYAYA 449

Nipata is used in its technical sense by Asvaghosa in

Saundara xii. 12:

aq Hanna Fa fasted ata: |

far aay aay faaratsfahea ea: I
This stanza has been misunderstood by many scholars. The
reference here is to sentences like Bhavabhitt's :

seqeeqasfha wa stsft AMAIA |

nla aa facafafigqat a gear i

AVYAYA

Avyaya literally means that which undergoes no change and

Was probably applied first to the supreme deity. Transferred

to the province of grammar it meant indeclinable. We find in

the Gopatha Brahmana (i. 6) :

aza fan fag aaig a fates

TAIT a aia aa safes azeaay || (AEA C1eI28}
This is expanded in the Mahabhasya (i. 1.38) into :

a adieacaafafa | # ga a aia? aigaqaantta

fa,
Mayo: crafters a | Tat Bag faa
tha fafa | Qa farfea az444 |
atam
Bhattoji says in his Sabdakaustubha on “svaradi-nip
avyayam” i, 1.37: seqeeiarat AIA ATTA qnafaaiat
afercfiy zaafa—azaa: tf cat qa cata fag

Fe THIRTY az aeaarfata ginal) 4g a, Fear

OP. 12999 |
450 TECHNICAL TERMS OF SANSKRIT GRAMMAR

| ra foxy ATA , fagfaas-

cata ararzeaatate
sfaaraa araeriarertata aa: | faaheg HITHT, AAT
analfederg | aera waft fegactHatar-
asf aaty qaarzeaaeaa |
“Vyaya” is found in the Nirukta in the sense in which it
occurs in “avyaya” .zpead g aafa ca a aea facaiaalg:
i.8. Here Yaska says that not only the form ‘tvah’ occurs, the
forms ‘tvam’ etc. containing different case-endings are also met
with. - ‘‘Drsta-vyaya” occurs again in Nirukta v. 23 in the same
sense. Similarly in VP. we find “vyayavarns cantah” ii. 26
which is explained thus by Uvata : Ara MAT fzfay:—aaar
aeqaaia | ae faaceanfefa fant a faaa arsera alt |
amae—azaq fa) oer ga fhraenfeafaar:
faa a orate |
In BD. also we find ‘‘samkhya vibhakti-vyaya-linga-yukto
bhavah’ i. 45.
The word “‘Avyaya”” is also found in CA. (ii. 48, iv. 71). APr-
(ii. 2) and Unadisttra ii 65.

Upasarga and Nipata would appear to be older terms than

Avyaya, since it is not mentioned in the traditional enumeration
of parts of speech, viz., naman, akhyata, upasarga and nipata.
Panini evidently divides all parts of speech into ‘subanta’.
and ‘tinanta’ (vyayavant’ of the older grammarians) and
‘avyay’.. The last embraces nipata, gati, upasarga, karmaprava-
caniya and other indeclinables. In more ancient times the
upasargas were probably more numerous comprising some later
 AVYAYA 45]

‘satis (cf. BD. ii. 95) and all those indeclinables that were not
regarded as upasarga were designated nipata.
Panini does not define ‘avyaya’. He simply enumerates them
In svaradi-nipatam avyayam i. 1.37 and following rules.
K., Candra, Sak., He., Su, Sam (i. 102 ff.) and PR. also use
the word.

J. uses “‘jhi’’ for ‘avyaya’. (cf. gi, ai, and ti). Probably the
author took the word ‘jhatiti’ as the type of avyas. It is
clear from his definition “asathkhyari jhi’’ i.1.74 that “asamkhya’
was the word in general use in his ‘circle and Candra actually
ses it for avyaya. Asamkhya, no doubt can be much more
easily understood than avyaya, but in such rules as sastyader
asathkhyadeh (Candra iv. 2. 54) etc. it causes ambiguity.
Candra also uses ‘“sasamkhya” (cf. early ‘vyayavant’) for
“anavyaya”’, e.g., sasarmkhyad amah kya] va i. 1. 24.
‘‘Asamkhya” is preferred to “‘alinga”’, because ‘“yusmad,””
“asmad” etc, show no difference of gender (nanta samkhya
datir yusmad asmacca syur alingakah PR.), though they are not
indeclinables. “‘Alinge yusmad-asmadi” is a well-known
Srammatical dictum. In the $akatiyana-prakriya-samgraha the
declension of ‘yusmad’ and ‘asmad’ is shown under the heading :
athalingam. yusmad-asmac-chabdav ucyete.
Jhi-did not commend itself to Vopadeva (who generally
“voids jh in his technical terms), because it bears no resemblance
‘© the word ‘avyaya’ and probably also because in the system
g (-anti). So
m Panini it stands for a well-known verbal endin
 TECHNICAL TERMS OF SANSKRIT GRAM MAR
452.

In the
he used the second syllable ‘vya’ to denote ‘avyaya’.
Jainendra system ‘vya’ stands for the krtya suffixes, being the
ya’
second syllable of ‘tavya’. In RT. ‘vya’ stands for ‘talav
being_its last syllable.
qio| ewaufefaqansaaa) 212139 |
Go| amadatanadguguia@iaqagtagiaa-
camera fir: elaice | AAT fe: | AaTaPa: Viens? |
Gio | Aa-al-S-aanaeri qrgafraguadatl
TATAA 212122 |
secant, | saofamedcardticacraifa 1 atte
diet— acaafaqamygran - xanaadta-fratasfaars-
az | ag waa afsdearitmn” eatizeacqarta |
Zo} FATITASSTTA | ATEATSATT | et ROIAR
arpa aancrafata.| fag-arca-faafr-aarad4
a ararenai sfarat 1 deaf: | a afa a arated azafa
aaqania a Deratieaatafe: | rare: |
go, eufe-fafaqat sa) eo | fasfama, a=
TAA, sA=AsTF | |
Fo | eqcite-aiz-aata-afga-Al-araggeaaa \
FIQIRY
go) eurfz-ofs-aaifecr aataifzatean: |
fra: TIAA ATA FATA: II
g1B'20 afta:
ee each ante-qarfz-afean: at ateaar GASTAA |
213%°

aR ATTA
 Appendix-I

Panini as a Poet
| Kshitish Ch. Chatterji

Indian tradition has since a fairly early period connected the

grammarian Panini with a poem entitled the /ambavativiyaya or
| Patalavijaya. There can be no doubt that these two names refer
to the same poem, for we often find quotations with sfa fasta: at
the end. Cf., e.g., Durghatavriti on III.2.162 Rina Wratata
Taqy: and Supadmamakaranda — on 1.3.79 Hyaqqadt
ahtd-faratate farsa? aa AA: Wee: !’ Now if these were two
different works, commentators would not have simply
Mentioned faa but would have stated the names in full in
Order to avoid confusion.
Several verses, some of them of exquisite beauty, are also
ascribed to Panini in the anthologies, but it 1s difficult to say
Whether they are from the Vijayakavya or not as NO manuscript
Of the poem has yet been discovered. The verses, however, do
Not appear to be the work of one author and, in all probability,
MOst of them do not occur in the Vijayakavya.
ctly
The following verse from the Saduktikarnamrta distin
refer; )
fers to the 928 as a great
son of Daksi : poet:

Gat wip: Hee Tyan A Ua

efaalahtyy
exfa siearatsty ETA
faysite:ye: wafrryT wear
| waurerautéaufa sayfa |
hu. Who ‘ —— ke =
ite have reverence for Suband e al
of Raghu:
that finds no delight in the author
icandra wins our heart.
pieased With the son of Daksi, and Har
"Ta excels in purity of diction and the words of Bharavi are
fees naeemomseseseal
9 aPofeetifa worTC |
A
 of Sanskrit Grammar
454 Technical Terms and Technique

Still Bhavabhuti affords an indescribable

sweet by nature.
pleasure to our hearts.
that the prince of
From the Mahabhdsya [2 1% Ro] we learn
et Ort: |
grammarians was the son of Daksi : ag aduereen aatqa
there 1s a
In the Haravali (as also in the Siktimiktavalz)
us a bit farthel
verse attributed to Rajasekhara which takes
inasmuch as it mentions the name of Panini’s poem.
Taher Ufa aH aed SETAE: |
eT ATH PIAA STATS
Hail to Panini who, by the grace of Rudra, first composed
a grammar and then the poem /ambavatiaya.
ekhareé
In his Kavyamimamsa (Chap. X), however, Rajas
includes Panini among the Séstrakdras and not among ve
Kavyakaras. Says Rajasekhara : —
Fad Psasaa HesHRAA —

Tradition tells us of the examination of the compose’ e

poems at Ujjaini : Here were examined Kalidasa and Menth
here Amara, Ripa, Siira and Bharavi, here in this Visala were
examined Haricandra and Candragupta.
the
Tradition also tells us of the examination of |
composers of the Sdstras at Pataliputra : Here were examine
Upavarsa and Varga, here Panini and Pingala, here vyadi, her
Vararuci and Patanjali, and thereby they all attained fame- -
Ksemendra in his Suvrttatilaka (II. 30) speaks of Pani™
excellence in the handling of the Upajati metre.

Pettit ( 2) wiforeranfate: |
FARHAN eae wfafy: 1
 Panini as a Poet 455

Panini’s skill in Upajati makes him enviable just as a

garden is lovable because of its charming jasmines.
Namisadhu in his commentary on Rudrata’s Kaévydlamkara
II. 8. says :
“aera wat aaterta panty needa
WISI Tae Tat: | dene wits: wearefasr
Herat
s
= ‘Waray Te UO Sa Yala Kay Tae: | da TeaHa: —

TASHA GRAS sia Aq WPT HITT: |

aT acatagfars desat WR THA
ROTA Uefa aa Te" She
Though the author has already spoken of Vyutpatti’ by
means of which incorrect words
should find no place in a
Composition, yet he refers to that topic again, to show that one
Should take particular care to keep one’s composition free
from incorrect words (and expressions), because even great
poets are found to have been caught in the net of apasabdas.
Thus in the Patalavijaya-kavya of Panini in the expression
Waray Tel ALOT my (having held fast the evening bride with his
hand) wy has been substituted for *a (though it is
stammatically indefensible). Then again in the verse TASHA
“tc. of the very same writer we find the word 274d where the 7
Of =i has been (wrongly) dropped. oo
In the fourteenth century Vidyadhara writes in his LRavali
I. 3-6) ;
Unto aeded: MsqeraragRe: |
se dafaen maaraa a tl
VRIN drei wyafsaqaad |
sree ead Std Aa:

TEMAS HAS ATAAT I

nn
CaPTE hel q HATA |

Rudrata defines @qafd thus :

Sloop oreo feaferrerereffearare|
Werpefacat oqafattd ar
 rit Grammar
456 Technical Terms and Technique of Sansk

The plain meaning of the Slokas is : Panini and other

learning eee
scholars who are well-versed in all branches of
three classes, VIZ.)
divided all the Sdstras in this world into
where the
yyyatad, fared and araafar. Of these the Vedas
words are of primary importance are said to be agree
gives
because any mistake, ever so slight, in reciting the text
rise to sin. History and allied sciences are known as '"
(like a friend). In them the sense is said to be of prima ry
importance as they are of the nature of agare. Sastras where ~
suggested sense is prominent are said to be PreMed ‘
one’s beloved), because there the words and their ordinary
meanings are of secondary importance, the suggested sense
being of primary importance. a
ee
Now we do not find anything like such a division
explicitly stated or suggested in the Astadhyayi of Panini. So we
are driven to the conclusion that the author of a
Jambavativijaya must have entered into such a discussion in ™
introduction to his poem.
Later writers have indeed tried to show that almost all “
defects of composition dealt with in the science of Rhee
have been hinted at by Panini in his Astadhyayi and his ae
followers — Katyayana and Patanjali. The commentator on t
Vyaktiviveka, for instance, says :
maf. Tag faerie flaferheersrTatet
RITES: A:, patdfaucraerateT4 fra
woftecar wenfe Ten: ya Ree wee
! are: ore ferret TMT
fanf aaa wa) wen‘earcarafiraf aere
...""[21888 ] ne A
fesafe waicant araq a” sf searaga epee vara: where: |S
“parE fee” [3 ¢ wo] serarquemeargerenratettarra arareaTs
ot are were” [7g§ me] Fea Pre, etal a7 fewaft TE
Gal wat ‘aquaria agate, Ba" sfa graced ee

Sarr treaty yanfeta cen gaara...” (413 694

eT d.
* I have not been able to trace the aft, but see Mahabhasy@
Kielhorn Vol. I. p. 404.11.15-16,
 Panini as a Poet foeCit“I

Tiers TaaRh SHAMIM gad wreadtaesl Aedafelauniata

Taras Seas areTasanty atfadta: we
STalFIMTa aaa eaMiadcdal aed Hefagu ariaqyey Pecahmanawa
TARA Sea A HA TATA TARTAR STATS |
[Vyaktiviveka, Trivandrum edition pp. 15-16].
As against this we have the general concesus of opinion
that a grammarian is constitutionally incapable of composing
Or appreciating poetry. As Dharmadatta says :

Ta aM TTA aT, |
area tq Tad: HIS Sua AMT: i

Where fatal: is explained as grammarians and Mimamsakas.

Similarly Ksemendra in his Kavikanthabharana remarks : —

GA
SSA ST
Ue VRRIUsRA HCMAB: |
ae ae STATA aaa: YHATTA: 8 RR
47a apes: Sifwenfasreers Grae: |
a Tesh mata faferdtsfa acheia: Gata AHA: 1k RZ

Hala Ulecaaa: SHY FAS HTT CAM |

a athe Hacwnfseh a Hage GrofaHraraay Wk RAM
He who is by nature like a block of stone, or has been
*Poiled by a grinding in grammar, or consumed by arguments
never been
About fir e and smoke or whose ears have
poets, in | him
penetrated by the composition of good
“Oquence can never be engendered even by the best training
so well-trained,
s’arted in the best way; an ass, though ever
see the sun even when
i Not sing, nor does a blind man
Pp
"Nted out. One shoult acquire learning from a literatteur for
is instructor |
© “Teation of poetry. One should never have as his 1
are
rian, both of whom
gramma
rej ctician or a mere
poetic genius.
.

WMdicial to the unfolding of

.
°
I
u

Then again we find in a subhasita —

Ta crenata frat a wR cas

|
QM Wgrata Testa qageSraraeasrea
 Grammar
458 Technical Terms and Technique of Sanskrit

marten ferrot AMSAT STAT F

HAMAS HAA ARM Ftd TIAA, Il

The muse of poetry never comes to her father the
tician, shrinks from
grammarian, nor to her brother the dialec
he were 4
and gives a wide berth to the Vedic scholar as though
d
Candala, thinks the Mimamsaka a eunuch and loses all regar
and
for him, but when she meets the person versed in Kavya
d.
rhetoric she chooses him as her husband of her own accor
Another subhasita tells us —

- : iC
In connexion with vicious arguments relating to Log!
and Grammar all poetry evaporates like mercury in fire.
_ €
It is indeed difficult to find a first-rate poet who was at rs
same time an expert grammarian. Kalidasa, no doubt,: was . weles
read in grammar as is clear from the many grammatical simile
. .

von es "é
found in his works, but he never was engrossed with the me!
.
form of words nor lavished ; and energy 0” tf1€
all his; skill
framing of grammatical rules. It has been well said :
4: TA PATA Al Chaey-
ATERAIR IRA: HTTMA Ara: |
THTaRa MAHA SACHA: Fat ;
Wael sactetal ferfrrd: Hrpey Hrsg Aad, Il
n
We must therefore have very strong grounds before we C4
as
pronounce the prince of grammarians to be the same persoP
the writer of some of the exquisite verses attributed to Pan!”
We shall now proceed to discuss the slokas one by 0n©:
arat Fir: WMAAH SVE:
URTad AAT Sea: |
aleaTST AepUReTSH,
Waleed Veargqed ARAL 1 2 I
Saduktikarnamrta v. 21.

* Subhasitavali 192,
 Panini as a Poet 459

| Yonder dove, sitting in the cool cleft of the hill and skilled
¥ & : - @ . . .

in love’s caresses, fans with his wings his dear mate wearied
with heat and coos sweet things (in her ears).
UATATaST is ATT |
yg gy hardly sounds like Sanskrit. The correct
reading would appear to be Ax f€ G34, as in Sanskrit adverbs
are used in the singular number. In the English translation I
have construed Ay with a noun understood, according to
Vamana’s dictum fasauarmyarm faseryitgd [5.1.10] ie. an
adjective alone may be used when the sense of the substantive
is Clear from the context.
dais is safes for Panini does not mention the root a
in his equ. Further, one does not expect such loose
expressions as fit: MadtHaees: from the pen of Panini
This verse, however, is a fine piece of poetry, almost every
epithet being highly suggestive. It is reminiscent of Magha’s
wm: = yeaa: (VI.61). The metre ts Sosa |
TA.
SCIRI: FakSraPTAGATT GAY
Wattd wea wreseataaya: FoI: HATTA
Senicih: wphehirrerattaratratianda
void as aay: wharves: Tata fn
Ibid., V.363.

hind legs on the

See how the troops of jackals plant their
that
Sround and, with outstretched necks, drink the thick fat
hung too high on the
Crips from the rows of boiling corpses
r dark ness and
trees that are enveloped in deeper and deepe
‘vealed to them only by the light of the flickering flames that
Sue from the cavities of their own mouths.
; Aufrecht reads 3ga4 and construes it watt ag. He
“ ‘far from all beings that are lying pe agate
inn Wesen). Bohtlingk suggests | peor
cases of such ©* I
to be the correct reading, though
ae

td Cait) aad... TAH: v1.

y|
 Grammar
460 Technical Terms and Technique of Sanskrit

in Sanskrit”. I do not understand why

aida are rare even
Aufrecht
European scholars generally read piyd for wfed.
the Malate
notices the variant wary: ... w#fyawaay: and in
Madhava v. 18 (which undoubtedly is the original) we read
Puaaeaaea:.
There is not the slightest doubt that the correct reading is
wud meaning ‘that are being boiled or scorched’. SORA:
There is a belief that when jackals howl at night their mouths
emit fire, Cf. Raghu XVI. 12 Aeqateatataaisay: ara Usa:
frarf: and Bhagavata 1.14.12 frataxin eae.
TAMA
Vafers cae aS ATA SAAT:
AAS ASA GSaA WET |
HISAPS
weed wae wactfes fafind ye sreaeere
Ward: A: Gaia Siete Aral YATE: 2M
Saduktikarnamrta V. 364.

See how the old vulture, with a fierce blow of his beak,
gulps down in a moment a great lump of half-burnt flesh from
the corpse, hot, and seeming to be itself on fire, the body
being dragged out in great eagerness from the middle of the
pyre the blaze of which is heightened as its fire is fanned by the
beating of his wings and then burning within, plunges eagerly
into water.
As the second syllable in #ieigarpeqd: is not distinct in Mss.
Peterson following Bohtlingk reads MISeTPEAM:. Aufrecht $
reading #lemapeqd: is better and appears to have been the
original reading. Of course in both cases we must have recourse
to the dictum amacdsty TAHT GAE:. SEACH does not seem to
be very appropriate here. Moreover it is Sura.
The above two verses contain a powerful description of the
burning ground and are evidently from the same pen. The metre
in both is @™a. They present several points of similarity with

* Cf., however, Magha X. 56.

argarary Pr
eratey
frayed |
ur sat fore: TOPSeY seas
 Panini as a Poet 461

Bhavabhiti’s description in the Malati-Madhava, act V and are

clearly the production ofa much later period than that of Panini.
Heresy: MRRoaanrd HAE: He
surfers aeaneataaast SAA WTA: |
sarees: teases
att a
Uae Wale Sfaaita GREATS: Ww HI
Saduktikarnamrta I, 411.

When the moon took his lady the Night in his arms (rays)
cold with the touch of white lilies and charming on account of
its acquaintance with the winter and Night’s garment of
darkness fell, methinks her friends, the East, the West, the
South and the North, looking at each other and raining down
the love begotten of old acquaintance, smiled brightly in their
€cstasy of delight.
| The metre is @R. This is a very fine verse and fully
worthy of one of the great Sanskrit poets. It is clear from the
Metre and diction that it could not have emanated from the
Pen of Panini. It must have been composed by a poet or
Thetorician as an instance of garaifp. In the first line

aifaahe: is not quite clear, as it is difficult to

Understand how their acquaintance with the winter can endow
them with charm. fafaeaftaa may also be construed with
yy. though the rhythm is against such a construction,
Ut even then the sense will not be happy. Aufrecht compares
this war
US verse ,
with SST |
oro ware ARPTAM SWIS
teadie cr
sar a ge n
gore
Cad pathy Proarparesanlesaegel
Sate wea: feraea cof cere Tareas 4
Ibid., Il, 606. Given anonymously (aeara) in Subhasitaval
2037
different.
Where the reading is slightly ing it to be a lotus, on your
settleonyour ha ndtak
they
TSN:
sarah Tee: verfired HPP STITST TT
Peta
eae: Prat Waaea eT
CY aeaty areraaenitfragaceyen gat
462 Technical Terms and Technique of Sanskrit Grammar

cheeks which they mistake for Madhutka flowers, on your eyes

deeming these to be blue lotuses, on your nether lip because
they think that a Bandhika flower, they merge in your dark
tresses to which they feel attracted because they take them to bea
swarm of their companions: truly the bees are hard to get rid of;
how many parts of your body, my girl, will you be able to guard?”
The metre is Weetasitsd. This is an excellent example of
the Figure of Speech known as Wi-atar,
This verse is ascribed to Aaa in the WseRIshd and TRAIT
and given anonymously in the quludaia and Hatxaatayeas4. It is
only in the Gefmauyd of AeeTa that it is attributed to Panini.
Acala was a great poet and it is just possible that this was from
his pen. “le<tae is tautologous and should be read Petar |
Oot wT eee eee Halereateh
Tasers: fee FPA |
We Gag AH aad Ye: Hfaq H-actt-
Waiearanrendiae: fe aa fasrst i &
Skm. 2.240; Kavindravacanasamuccaya (anon) 364.
Why, lady with the slender waist, do you spoil with tears
stained by eyesalve the lustre of your lovely cheek which 1s
resting on your reddened palm and whch is already slightly
wasted. Foolish one, the bee in fickleness may kiss the Kandali
flower, but does he, on that account, forget the fragrance of
the blooming young jasmine?
Tat GRA Sal: TareRt
PAT THAT: |
aft Feat aes STS BPTI
Hy FIT sass Fe: N19 1
Saduktikarnamrta 2.88
The Dravida damsels whose faces are fair, breasts heaving,
buttocks wide, ... ... and waists slender, how did even these
Dravida damsels go out of the heart of kings?
In the printed works the last word of the third time is

* WOH v.l,
 Panini as a Poet 463

as'amq which cannot be construed with the ablative In &a:.

Hence I have proposed the emendation Wsemmq. The second
line is open to the defect known as YOIHAA and the compounds
faraye4: and SeAMmataa: are not happy. The substitution of Jey
for ec: as suggested by Aufrecht is out of the question, and
Bohtlingk’s emendation of $4 for &: hardly gives any meaning.
It may be noted in this connection that the word #faS does
not occur in Panini.
TdaSewa waa
Tifa Gq Wate ares:
sTavadt acahha-<faral
Toa Ra CSU us
Quoted by Namisadhu in Rudrata’s Kavydlamkdra I. 8
from the Jambavativijaya of Panini; also quoted<anonymously iin
Bhoja’s Sr ngaraprakasa | where for @I@4qa: we read aan: and
for Toa we read F WaT |
The very dull rumble of the dark clouds after midnight
during the rains is but the low cry of the night unable to find
out the disc of the moon even as the cow bellows that cannot
find her calf.
uae is difficult to explain according to the rules of
Panini. We should either say W4<4or Fe. It may be noted
in this connection that Magha is very fond of the word "Ra, In
the ninth canto of the Sisupalavadha we find it used thrice (in
slokas 3, 27 and 39). In the same canto the word 44is used in
§l. 61, Wears in sl. 32, WRAST in 70 and vfererfy: in 738.
ajqzadt is incorrect on the face of it, the augment FA being
obligatory here according to the rule of Panini Werte. The
figures of sense are, however, charming.
Tard CaM War Cadastral
Waa: Bfaaht ara |
sel Ten Stace
wud ua fe emfarety ie
Subhasitavali 1994.
The last two lines as printed above and as found in the
 464 Technical Terms and Technique of Sanskrit Grammar

Subhdsitavali does not yield any sense. We may, however, read

saaaMeasel and amiayda: for the last words in c and d
respectively and translate thus —
Even the glorious sun becomes poor in lustre after
illuminating the worlds with his light. Oh, fickle is fortune
though she bestows strength and fame. No one, alas! is
immune from the vicissitudes of fortune.
The metre is ayeafad.
ASM Gl sel RR: HAA Far |
caked geycnearad |20 1
This verse is attributed to Panini in the hdl<aqtaqa4
(186). It is also found quoted anonymously in the Dasaripaka
IV.53 and Alamkaratilaka p. 14 where for a we have SMa
Taga: and for ¢ Maer, According to the Alamkaratilaka it
is a VASTT with the first hemistich.
Seeing the breasts of the slender-bodied (maidens) the
youth shakes his head as though to extract his gaze riveted
between them.
This verse is open to grave defects according to the
rhetoricians and grammarians. In the first place the word qTagnt
has apparently been used without any special significance. As
Ksemendra points out while criticising Dharmakirti’s poem it is
the poet’s passion for alliteration that has led him to use the
word which does not lend the least charm to the sense. The word
Gat would be very appropriate here, or other words denoting
exquisite beauty and loveliness. The word TI or G4! gives rise
to charming propriety of sense only when it is used in reference
to ladies pining away in separation. (3a wT ga %%
Haaser Fag 4 aifseifaersaqarainarmitasifa! Gat
gaa WTS ed se oot Pfc adm g
fercfagrertion wapreifacrnat srrata | sitfacafararcrat
i”p.117.)
It may, however, be urged on behalf of the poet that the

Arjunavarmadeva hints at the same thing when he says 3444v4

TEIAM CARTAN TATA | sat a Taft | while commenting
on Amaru’s verse:
 Panini as a Poet 465

word 4 is particularly appropriate here in as much as it lays

stress on the contrast presented by the slim bodies and heaving
breasts of the maidens.
The next word &fI also is unfortunate, though such uses
are by no means infrequent in Sanskrit. We find, for instance
in Sisupalavadha Ill. 14 FHMPTALT SUA: LAT ASV: Uagiry
and in Naisadha XVII. 12.

where dqUll means Ta Cai Hii

To justify these duals Vamana has formulated the rule -
errett fecafasr sift: WaT (Kavy@lamkadra V.117) and in his urtti
he cites the example @UaSeSsAeSX, Vamana himself is unable
to explain how a word denoting a class and not individuals can
express the idea of duality and says that he merely accepts the
views of the Jainas for the purpose of explaining grammatical
forms. 34 Yt fecafaecs
sid: ? ahs 34 4 OH, sragUCa TEM: | A Ste:
agers: |e ages St: 2S ST OP) oad og mesial
fagmaarantes: | 7A danas: aeaIMairararaeafa |
Now if this verse of Panini had been known to Vamana, he
would have regarded the use of the dual in @Al as a BVH for
such cases and would have been spared the shame of having
recourse to a queer theory for explaining forms of grammar, a
theory which he cannot bring himself to:accept.
The next word that calls for comment is #74. According
to Panini’s rule frerantsaza (1.3.87) the causals of verbs
meaning ‘to eat’ and ‘to move’ are exclusively Atmanepadi. So
the correct form according to Panini is fd and not Ha. In

heh a-caafaan farfan geida +<Aat:

Garon wear: fads raat aT Sresreanfef: |
ad: CHa VaR ATS A PETA
eared: fares Pager
Hi ASS i (aE ex)
In Kalidasa Tat Ream @fafaea vert wea, Amaru’s wenfeasa fasirofaa
AMIsta Taea:, etc., Ta is used in the sense of a beautiful girl and so the
poet, even if he has erred, is in very good company.
 of Sanskrit Grammar
466 Technical Terms and Technique

Siddhantakaumudi puyatd is
fact in both the Kasikaé and the
. In the Mugdhabodha "1 1S
given as an example under this rule
(862). It is swange that
expressly used in the sitra Parnes etc. rule
of his way to violate his own
Panini himself should go out
ld let slip this splendid
and that his commentators shou
aH or of using their
opportunity of turning this into a
ingenuity to explain away the solecism.
and other
ama is no doubt found in Bhasa, Kalidasa
early poets but it is Auf all the same.
nious but still no
The conceit is certainly bold and inge
that this is from
one in his senses would for a moment suppose
the pen of any grammarian.
facies GSH wi Ufa faaead: |
ad pen ye rena fe areat fart 1122 |
Subhasitavali, 1887; Sarangadharapaddhati (Heat) 3580.
the
The East sees the glow of the West at her union with
Sun and her face darkens. There is no such thing as 4 woman
without jealousy.
Keith construes 71 with faatad: and translates - “Dark
groweth the face of the East as she beholdeth the glow of the
sun in union with the West. What woman is free from
jealousy?” This construction is to be rejected, as a contrast
and that of the
is here intended between the face of the East
West. At her meeting with her lover the Sun, 4 solt
blush suffuses the cheeks of the East but this sight 1 toe
much for the West who is seized by a fit of jealousy and het
face grows dark. The harmonious blending of the figure?
aareite, adic, fay and wag gives us a most charming
picture.

Us UF: TUS
SCCIRICECCH CIRCE
Tareas Tacs tas
WG TAATAR 122 11
by Vamana (iv
Subhasitavahi 1815; quoted anonymously
 Panini as a Poet 467

3.27); Mahimabhatta p.75; Abhinavagupta (Dhvanyalo-

kalocana), Rajanaka Ruyyaka p. 184, Visvanatha and others.
The Autumn bearing on her pale breast the rainbow
which resembles the fresh wound with the nail and thus
propitiating (making free from clouds) the spotted moon
added to the heat (agony) of the sun.
Wsawaesy Fea
fara Ye pyets arK: |
ay BTS ACARI
WSt Ya HS WTS GH: Il Subhasitavah, 1968

The moon separated (caused to bloom) the compactly

united Kumuda flowers though they are pure in conduct.
When a person whose heart is corrupt attains greatness, he is
sure to become dull i.e. deprived of all sense of propriety. To
Whom does a crooked man ever bring profit?
TT: ale VaNATEIy
ART
Taree Peat TTT SST THT |
a Sepa sft Mae TT
feta feft ataxia Serel: Ul Q¥ Il
tafediureiian
Sabhydlamkarana Samyogasrngara Vi. 24 (b);
Subhasitaratnakara iii, 55 (anon.); Sarngadharapaddhati, 3809;
Padyaracana xi, 35; Subhdsitavali, 1765; paplninancaste
samuccyaya, 129 (anonymous); Saduktikarnamrta ll, 812

(Omkanthasya); Suktimuktivali (119 b).

Where is the sun gone now, after making short the night,
sie . io

earth and
‘obbing the rivers of their water, parching all the
s wander _— :
drying up every thicket? Methinks the cloud
o
“very direction bent on seeking out the sun with their torch
lightning.

AUSTRIA ST
:|
Te quterarycy ata

Subhasitavali, 1898.
aaefaerafqarqesy i 4 Il
.

,
 of Sanskrit Grammar
468 Technical Terms and Technique

sun) of unimpeachable
Then reached his setting (the
the man who has put far from
glory to convey, as it were, to
Everything that is born must
him the fear of death the lesson :
die as I am dying.
e, FAA
Would Panini have used AMadH: as a substantiv
with fafa and 4441 (genitive) with SIEA?
Sree Ta FATA
Tat Td SY Hd APART |
sent fe sera ard Ga
the fYaearhAgTAAG It 2 Il Ibid., 1899.

“The lake has done well in shutting up its lotus eyes now
that the sun is gone. What profit is there is eyes that see the
whole world if they see not the loved one?” (Peterson).
of Panini s rule :
The use of #al in 81 is in clear violation
aarayaa: yaaa (1.4.21).
For Panini to add % after f¥aretHqnray is like ‘guilding
refined gold’ and ‘painting the lily’.

Ae Tearatyenara: |
aRifad: We fe F aR
TSAI VT i 29 Il
in
Subhasitavali, 1943. Quoted anonymously
etc.
Kuvalayananda, Alamkarakaustubha p. 234, Prataparudratika
-_ In the Alamkarasarvasva p. 95 the reading 1s al

Observing with his lightning eyes the face of the girl ees
Git to meet her lover, the cloud thundered his loud jame™ *
saying to. himself : “Have I with my showers let fall the moo™ ; p
It is doubtful whether aftafta in its technical sense
-
use at the time of Panini.

SHSM facia sen yes ira FRM!

am aed fafttyse wen wish ame aferd a TTA 8°
|
. * .
 Panini as a Poet 469

This verse is attributed to Panini in the Sadukttkarnamrta

1.412, Subhasitavali, 1969, Sarngadharapaddhati, 3634 and
Suktimuktavali (139 b) of Jahlana. Quoted anonymously in
Subhasitaratnadkara 33a 169 11.27, Dhvanydloka p. 35,
Alamkarasarvasva p.83 He., p.274; Alamkaratilaka p.36;
Vyaktiviveka p.2, etc.
The Moon, red with love’s passion, caught the face of the
Night with all her quivering starry eyes in such a manner that
she noticed not how her whole mantle of darkness had slipped
In answering passion to her feet.

aifea Uden fete arrest

ska A AAT 18S ll

tq *

TS
Quoted as occurring in the second canto of Panini’s
Jambavativijaya in the Durghatavritz on [V.3:23.
There is, in the cave of the Himalayas, concealed by the
Waves on the sea-shore in the western sea a great city named
Dvaravati hoary with age, the renown for whose sanctity has
Spread far and wide.
Since a very early period a fierce controversy has raged
round the question as to whether 4qr is to be admitted in
Classical literature or not. On the question of usage Haradatta
: Wass a Orca asa, qarmal Tacan: Ta Ta a
Temarks
erential faareverrarrmesets sail gatgoad| [AA2T
Vedas.
‘It is well-know that 4% is confined to the ayik as,
Ancient usages, in prose and poetry, in kavyas and akhy

* OF agfatuta Ses
Ufaryer
fafat atraey: |
HoH Parisearas
Pads WAS We

TTA THAS aera aA: |

Tet Wradt Bea Bed AAA, ll
TAT |
 Sanskrit Gramma
470 Technical Terms and Technique of

writings of poets fond of grandiloquent

even in the
. _
phraseology, are not met with.”
of the ”
The Kasiké does not want to restrict the scope
Bhagavrttt Whe
to the Vedas. According to the author of the
form aiift alone is not exclusively Vedic, all other pe
forms are purely Vedic. AigitsRe avarice 7 areata “
aeanitte 1.33. % sa Fy,
aarp ecaid!
Wea AS! Mea arate AAT: | wag, Et
weaty
adaqada | a ageera seq WY SA
aivadicta ve word ary areifefa arrafa: | Bhasavrtti 1.4.74. |
The Bhdasyakara says on VI.4.87; wa afeé fas afd a
TTT APL wa Mardis, «ferret AIA WAIST?
dachiciad fe vata ATeTaTay | |
It is remarkable that neither Patanjali, nor Yamane, =
lamea
Jayaditya, nor Jinendrabuddhi nor Haradatta nor Vima
should refer to “t#difq in Panini’s poem.

Gadd Oster FeTtarary |

eau Gq Hed! AAT
fasorersqad SATA Ilo I
: rk in
Quoted from the fourth
. canto of the same
- W® or rupt
Durghatavrtti [V.3.23. The sloka has become hopelessly ©
and it is idle to attempt to translate it.
A Welisid Gal aa Tet TA |
Pau date Geereriaaa A ue il
in
Quoted from the 18th canto of the same poem
Durghatavrtti IV.3.23.
The friendship which I formed with you, the friends"
‘which has subsisted from of old, that friendship has
ge
long period been renewed in my heart. res
The printed Durghatavrtti shows Ut: is ¢ but sense —
¥i:. These three verses have been quoted in the Durghatav?,
justify the use of the word [wt by poets. Saranadeva Say$* cong?
guerre
hrary sf anfa:, Tory —" [x 13 1204] sed SI
 Panini as a Poet 47]

sea! ‘aareeprata Fora vata"!

sfa eqeqet a1 ‘areata trae
vata’ sfa arey fer gaara: pretest varagreat
sia wrvate: |
TAM, Aa: “Totter SAAT: [¢ 06 13) STA ATS Tae arpa"
[¥13 204) sft frome seraaata fag am sfa viassrvaia Sarsrary
POR waste) dag areaadiias Ultra — sted wiley gia fects
Wt, Sa aafaay sa aad, a SSI | Has q YAeuigrand Falcom,
adi ada seaarpy! =ii.e., How- is Bharavi’s line @u:
fra justifiable, for (in the case of Ww) <f and eye have
qUartyty
been superseded by the use of the word Yuin the rule Toa y,
AeA (i.e., since Panini himself has used the word WW
derived from the word YW, in the sense of7 Ya:, the rule aaa
etc. prescribing the suffixes @{ and ee in the case of
indeclinables cannot operate in the case ofW though Wis an
indeclinable, and thus the form JW is incorrect). The reply is
and @e also may be added after YM according to the UAT
“ staquaraty fora vara" i.e., “when forms which are contrary to
ordinary rules of grammar are used in rules of Panini they do not
always supersede those ordinary rules”. The author of the
Bhagavrtti says : Forms in violation of ordinary rules of grammar
used in the rules do supersede these ordinary rules — this is what
Wwe find in the Bhasya. Hence expressions like qudtya: are
Incorrect. This is not so for it is in the Bkasya on Garey BAT:
that it has been said that though the absence of the substitution
Of Mfor qin the word 314 is clear from the use of the word 313144
*:, the prohibition of the
With the dental 7 in the rule sre
h do
Substitution in the case of 7 shows that there are faa; whic
Ot supersede the ordinary rules Thus we find Panini using the
second, fourth and,
Word Wt in the /ambavativijayakavy ain the
by including it
Cighteenth cantos. Kaiyata however justifies it
“nder the yarex group. He says this on the rule Gatatft Gaara |
riti
the author of the Bhagav
The view attributed here to who says
'€Ceives confirmation from the Wf aagfa of Siradeva
“Ht px 79) —s
4 I wag: Fd ufo afy waaay a :—
te, — "Laie Re] senteqa Arete! vigil pi a
Tera af Cfnar ging] goraHtathe -
|
 mar
472 Technical Terms and Technique of Sanskrit Gram

HAA: Wasa, Tao Aa Wey: | [ Paribhasavrttt,

Waretiaadl Waid
Benares edition pp. 139-40].
It is probably Jayaditya who first laid down the dictum
aaererara foam waft in the Kasika on IV.3.105. In the
Sura: | aera Gol
printed Kasika we read : Wa 3fd Pada
Tosca vata. That this was not the original reading Is clear
from the Nydsa: QW sf Bisa wee: ? Aaa ‘ara Paty— Pe iRe]

fFoearea:i’ aaa, 3 vata Gata — 4 denic! aq HT

AOA afsaenta aah — sraererafa) srarerarafe waft GT
MEUTAaT:[8 162] FIA Mery |
This is also supported by what qeardyed says in his
a qa
afer — sarerarata fora vata Aa aferras
ayaa foreiia vast aesATT | safer gyceatererqrr
eid: — aaeefromm vasa! cl GUTS arerorneareald
ratt
roared ‘‘astoremisaragsaey’’ — sft queried eqege
ome:te
germs waft, 3a cea:quar ts
ech ether meres
Soa Wald |
Bhattoji echoes this view in his wear on gerard
bis
qamTaerg. But it is strange that neither Jayaditya nor
commentators nor Bhattoji and other followers of pani
should invoke the aid of one of these numerous gart’s bY
Panini himself to justify the usage. It is clear from the
expressions used by the “afark that in his time the
Vijayakavya was unknown.

4 oreetta Gat a feetereaiica 1 22

in Katantraparisie
Quoted from Panini’s Jamba ativijaya
74.
1.33. Manorama on Ganapradipa II.41, Bhasaortti 1.4.
What you said, viz, let there be friendship betwee? jo"
and Hari, cannot be supported by reason like the griendsh'P
between the lion and the elephant.
ee
a

Is Wimefaerd the right reading?

+ Ms. in the collection of the Varendra Research Society: Rajsa”
Bengal.
 Panini as a Poet 473

The construction, as will be evident from the translation,

Is vicious. refers to the whole sentence SRM
Fe FS 4 TI but
$4 has to be construed with F4 understood after faeferea:.
This verse has been quoted in almost all the later
commentaries on grammar but it is significant that Jayaditya
and Vamana, Jinendrabuddhi and writers anterior to them
know nothing of this verse.

ae até wit We agg GATT

SAS Tat CTS ATHISRTASATaAA: 11 23Il
This verse is quoted 1 by Qe in his wagerefagta 1.1.15
with the remark :

sfq sraadifasaare aaa storeate: |

Oh blessed am I, I bow down to myself that the girl with
the slender waist turned towards me and having dilated her
wide eyes cast looks of longing at me.
In the Subhasitavali (No. 1243) it is given anonymously in
a slightly different form :

Stel Ss AA Hal Ges AleraTaT |

SIA AT I
TAT FASS
In the Saktimuktavali it is attributed to ars.

Riceuieagt FTA, RS Il
The lion that destroys the pride of the prince of
€lephants. |
This line is quoted from Panini’s Jambavaiivyayakavya in
Bhasévrte ITI.2.162, Supadmamakaranda (4.1.136
Waa uwfift:), anonymously in the qosrare,,
commentary on the Mugdhabodha (sutra
Ramatarkavagisa’s |
t 17) ELC.
added to fag, fq and =ein
tl] According to Jayaditya PTZ 1s
er only. Jinendrabudd hi is also of the same opini | :
St . ay
“Mana, the author of the Kavy the
ilamkarasiitravritt frames
4 "7

f
O
thet Meueq: HHA A (V.2.38) and quotes severa ] instances
NE use of Fiex and fag but Hiacifost girs, 80 swel-known 1

@
 que of Sanskrit Grammal
474 Technical Terms and Techni

Is conspicuous by its absence.

jater commentators on grammar
c . narra 47
Haradatta quotes from Magha but not from
T:. urgasimne :.
: Hat WE Teaqenhaseigerang
i Vararuci’s rule fefefufefaci Gt: says
aatantiartneran facantardifa crarserfuce:, gafcaafeagt se |
the author of
If Panini the grammarian had really been
Jayaditya, Yenia
the Jambavativijaya, is it conceivable that
his authority.
Haradatta and others would not have invoked

Ta: Farah: Ser aa Ara: BA: WA: We Il

of water.
Gently blow the breezes charged with particles
-— . si¥ eran fore Rayamukt
“DR cuta’s
me
This is quoted from the /ambavativyaya
commentary on Amarkosa 2% |& 18 (yotdtaqyad sid fafues le
aera ‘0a:’ ... 3A:" sft sraadifassnrey)] and anonymously
Raghunatha Cakravartin. .

U Gatrdagaerd Weiter efomiteed: Rs Il

jth
The lion smacking the corners of his lips besmeard W
blood. aft
Quoted as Panini‘s in Rayamukuta |Amara 1.6.91 4 “tha
.. Sa:' sft aftrgaxnt a] and anonymously in Raghun
Cakravartin’s commentrary.
VaAeMSAG ayes: PUFATSACATTTE U1 VO Il
and
seated on bulls
The sky was filled with mighty Rudras
accompanied by their attendants. 4.39
The second half is quoted by Rayamukuta on Amare ‘atich
CIMT sft wraacat ofnfy:) and the entire he™
quoted anonymously by Raghunatha Cakravartin.

PATATSI AL MSA | 2S I
Embracing fast his bride the night with his arm» (rays
)
Quoted (with the exception of the last word
) 4° nk
by Namisadhu in his commentary on Rudrata’s Kavya the
1.8 and anonymously in Durgadasa’s comm
entary ort oF
Mughdhabodha (sitra 1277) as also on Jumaranandin
=
the Sanksiptasara, Krdantapada 484,
 Panini as a Poet 475

Can any one imagine for a moment that Panini the

greatest grammarian that India ever produced would go out of
his way to use such a vicious form as Ya in direct violation of his
rule GAasaayyas Wl TTT [912 Be]? Is there a single instance in
the Astadhyayi where TTY has been used by Panini where strict
grammar requires ¥al ?
Mr. F. W. Thomas wrote in 1912 in his Introduction to the
nasaaiayaa: “The question as to the relation between the
poet and the grammarian Panini and also the question as to
whether the Patalavijaya and Jambavativiyaya are one poem or
two still await a final decision”. I believe I have been able to
show conclusively that the poet and the grammarian are two
entirely different persons and that the Patalavijaya and
Jambavativijaya are one and the same poem.
It is remarkable that the Mahabhasya which gives us so
much information about the great erammarian™ and which
takes so much pains to turn and twist the rules of Panini to
make them more elastic never mentions any of Panini's poems
ed
nor makes use of any of the crammatical froms contain
therein as a WITH. |
only Patanyjali, Bhartrhari, Jayaditya, Vamana,
Not
Jinendrabuddhi, Kaiyata, Haradatta, Madhava, Bhattay,
i mention his
Nagesa none of these great followers of Panin
cases forms
Poem or make use of it in any way, though 1n some
like sTa2ifa would have been simply invaluable to them.
mentioned as
Then again, the verses that are not distinctly
belonging to the Jambavativijaya contain words and expressions
It cannot be urged that
that would make Panini stare and gasp.
ing the sutras of the
Panini has violated his own rules in fram
Astadhyayi and so there is nothing strange in his using words and
Se
* ad adver: aeitgae uarford: (1.1.211; 1.1.27).

ANPAR ast: afar: (1.4.89)

WHOM: Waa (1.2.11).
UAT Ge ute: qaes wfa: (1.3.46)
Swafeary wie: ufo (I11.2.108)
 Grammar
476 Technical Terms and Technique of Sanskrit

in the poem and

expressions not sanctioned by his grammar
Vyaktiviveka :
that he might well say with the author of the
rapfasatad: Saad arated 4 ae
arate fremaeaii SAAany TI
s be allowed to
In the first place some latitude must alway
the authors of the siitras because they aim at securing brevity at
any cost. There is a WH — aearreasas Yarqsa aad FATHOM:
save
i.e. grammarians rejoice as at the birth of a son, if they can
merely half a mdtra. It is for this reason that Patanjali says
ssion aarme al:
gat vata and this is why Panini uses the expre
These he
instead of Squad: and FAM, instead of AvaqaM,
uses deliberately with a specific end in view and consequently
him. But this
Ase: HY War faa: Wada: does not apply to
latitude has never been extended to poets in whose cas
h
~Wdepia had always been held to be a fre fault from whic
there is no fr=fa, and we can esily understand that panin!
would be the last person to stop to such forms as Yd, ues
The analogy of the physician also does not hold good in this
case, for in the matter of language grammatical purity marks
off the gentleman from the vulgar mass and so Panini cou
never dream of using words like Ja and sya.
ni easily fal] into
Then again, the slokas attributed to Pani
ivijaya often show
two groups. Those belonging to the Jambavat
been writtel
some grammatical peculiarities and might have
as
wi one who dabbled in grammar. But the other slok
of speech are of a high order and do not
illustrating figures
bear the stamp of the hard clear intellect of a grammariaD, pin"
seem to be the work of a rhetorician.*
oo, in
The fact that Panini as a poet is nowhere mentioned
* ws - rds
we
In the writings of a grammarian one naturally expects to findmatic
éand expressions which have been particularly dealt with in gram
treatises as also a certain amount of h he rhetoric?
regards harshness not as a defect ee ich of ®
but as a merit in the spec’ ae
grammarian. As Mammata says : daraconel aah yaaa tarel 4
HEF TT:, Similarly Vigsvanatha lays
down :
 Panini as a Poet 477

the Mahabhasya or in any of the later first-rate works of the

Panini school and that annotators and commentators have
racked their brains to explain away ungrammatical forms
instead of regarding the uses in the /ambavativijaya as WTF's,
that some of the verses attributed to Panini in one anthology
are attributed to other poets in others, that some of these
verses show distinct traces of borrowings of a much later
period, that none of the verses not belonging to the
Jambavativijaya bear the stamp of the grammarian on them,
that some of them seem to be composed as examples for a
work on rhetoric of a much later period, clearly indicate that
the verses cannot have been the work of the peot Panini. The
Jambavativijaya Kavya or the Patalavijaya Kavya must have been
composed by a poetaster of about the ninth centrury A.D. who
made use of many peculiar orammatical forms in it and
fathered it on Panini, the great grammarian.

Notes

10
1. qAerarsin this sloka and Tag]or Taga: in sloka
the femimine sufix8
are SuIf4, for Panini does not sanction
after a agdifé compound ending in Ay. His rule
apply here, as ag hasa
TST a TES Tea aT, [IV.1.54] does not
uded in the next
Conjuct for its penult. Nor is the word a¥ incl
Tule : “faneteagmedatiygm. It was Jayaditya who first
Cf. Haradatta
formulated the rule : agmanidssaa in the Kasika.
On Kasika 1V.1.54: aTaeate easSToAAT, |
ee ee
Saran gq wieose aie |
He S: Fd AT |
bdakaustubha
So we find Bhattoji writing at the beginning of his 3
faayi afacrd aese aisha 14
Rat atufd gated aaa I
in tthe verses
. sence
by its ab
att “MC this characteristic is conspicuou s
. *
«
B ]

e in the Jambavativyay®:
"uted to Panini (except thos
 of Sanskrit Grammar
478 Technical Terms and Technique

translated ‘enveloped
9. On p. 26 asada: yitdg has been
following Peterson and
in deeper and deeper darkness’
It appears at first sight
regarding the frin “fd as pleonastic. ed
d and translate ‘envelop
better to take *H in the sense of clou
on by clouds’ (welchen me
‘n darkness brought
gehullt ind — Aufrecht.) But it 15
wolkenerzeugtes Dunkel
ds here.
doubtful whether the poet wants to speak of clou
"Wwe: has evidently been used in the sense of FG: TW:
though we can also expound it after Panini as TTA FS: |
ranl go
3. aenefiat is explained thus in the Amarakosa : aeret
y proper 10
A BA WER at ASST: and consequently it is hardl again,
object. Then
use the word in connection with a single
p in the
though the word is found in the aRaRae grou
does not occur WW
Ganaratnamahodadhi, Sabdakaustubha etc, it
the Kasika.
The different readings noted by Aufrecht are gemfirerreoarere
be adopted 1n our text it
which is distinctly better and should
ly intended this fo}
p. 27 and wesw (the scribe probab
greet).
5. The different readings are : ¢ qal:, Hypa. ¢ page
area, (d) trated aed.
printed edition °
6. Kvs. reads Sar titraead fasta. The
the Saduktikarnamrta reads fae for fad.
7. The difficulty here arises out of our ignorance of i”
context. This verse is given in the Saduktikarnamrta UW" s
arent. If, as seems probable, the poet means how can am
ever forget Dravida damsels of forms so fair and charms °
rare, WISP will improve the verse to a certain extent —
even then we should expect fem rather than AM, If,howew
the poet means, how could ladies of such heaving breasts 2
protruding buttocks effect an entrance into the hear 2 aol
there are two courses left open to us. We may either
take fot
bull by the horns and regard @: as the acce. pl. of 4 _ jo?
a for the sake of metre or accept the slight emenday
proposed by Bohtlingk. Though in the Amarakosa We
r
Ud ea Ma SAMA TA: where € has to be regarded as news
 Panini as a Poet 479

the principle of association because the words preceding and

following it are all neuter, Vamana in his Linganusasanam
distinctly says that @{1s feminine. Cf. Karika 23:

WeleT ray MATAR:

So according to Vamana’s view &: as acc. pl. is quite
normal walt weft and aft Ae however are not very
significant in the case of the second explanation.
10, HERA Cf, tated Ge AA cls safer Serad Fafa (Tara
218) ; Sed HA (AAMT
¥ 184) |
14. The different readings are : a &7 b Warearat Bai ATTA
aaa df aaa, BAe, Axia, Ae |
16. cUeevb.
18. f£.n. aaeaz etc. The verse is attributed to Yuvaraja in
the Saduktikarnamrta 11.146.1 where the reading is TAMAR etc.
 Appendix-IT

The Anubandhas of Panini

Kshitish Ch. Chatterp

Sanskrit grammarians often add one or more letters to the stem,

root, augment, substitute etc. to show that certain grammatical
operations are to take place in connexion with them. The letter
or letters, thus added, are known as “anubandhas” and they help
greatly in securing brevity. Thus instead of saying that the suffix
31 is added to the root {to form an abstract noun, Panini say»
that the suffix 4 is to be added, thereby showing that the
penult is to get urddhi, the Wis to be changed into 7,and the first
syllable is to be accented. When, however, a letter 1s added for
no specific purpose but merely for ease of utterance, it is not
regarded as an “anubandha”. We often find in grammatical
literature SHRI Faqares:, Aa: etc.
a— (1) In the Dhatupatha an unaccented 4 indicates
Atmanepada; as V4 ggt — Tad ( 3tqenisd aera | a
IR IRR).
he
_ [Unlike later grammarians Panini has not read all t
ed
Atmanepadin roots with an indicatory & but has divid
them into two classes — those with an indicatory gand thos©
with an indicatory anuddtta vowel. This has been necessary
for the purpose of the rule aqeraaat Bale: (2 2 ge) which
prescribes 44 for roots with an indicatory qa". ]
root
(2) In the Dhdtupatha, a svarita H indicates that the ™
is used in the Atmanepada when the benefit of the action
accrues to the agent (taftafaa: safima frame | 8 IR 92).
ame
[Here the two different anubandhas for the °° |
purpose are difficult to account for. Later grammarial
generally read all such roots with an indicatory 4. Proba
the two anubandhas are due to a desire to make the can
symmetrical with the other rule ayerdtsd arcade .
 The Anubandhas of Panini 48]

(3) According to Bhattoji and others the & of the

fourteenth Sivasitra Wis for the purpose of the Ween ‘Ut’
so that in the rule RYT (8 18 KX) Tmay stand for both t
and thus giving us &{as the guna of and Aas that of
q. Subsequent commentators, however, repudiate this
view, accepting the Vart TRtika
sft AHA
(4) Roots with an indicatory unaccented 3 admit qin
the actie voice ; Wqekidsd Fee: (TW. 3 12 1kwe).
[37 is often added to consonants for ease of utterance;
Garand: or FAGAN: as FA az, WAG, etc. In such cases it
Is not regarded as an anubandhal]
ee (1) In the Dhatupatha shows that the root does not
admit the augment g¢ before the suffixes @ and ag and
admits it optionally when these suffixes are added in the
waar or in the sense of Aafemy (the beginning of an
action). sifeda | fasrat wraenHort: (191212&-9) |
(2) In the Dhatupatha an unaccented ST indicates
Atmanepada, far aot | aac |

a (1) An anudatta% denotes Atmanepada; as, eae fats

— Bd
(2) Denotes a root according to the Vartteka SHAT
MGs (3131202 12); as, Sfacafrasneat fait (1k 184), wiftrd:
WHld: (2 & 130), Aaeata (9 13 14).
(3) In the Dhatupatha indicates that the root will
admit a nasal augment after its vowel ; siedt TAL aay:
(19 12 KZ), |
[This roundabout process has been resorted to mn the
Dhatupatha instead of reading the roots oa a
Penultimate {with a view to show that in the case of these |
en
gq3
* wpssana: da werenfafs:! =arcacid gaat a
fa owt menifsacdsdt off wa CoRR —_
Waray, ; Weare Frater| fraraiqaaacd AES AM Mle T TT
AMET TATRA TAT
Rite eae, RAYA oy, TATA THA, | ‘STO TH
RATER HA
AI
al Weary
°F cect axforta, seqaubeh: wraudateone: | — TERETE |
 que of Sanskrit Grammar
482 Technical Terms and Techni
(Tard,
never elided before a weak
roots the penultimate is .
or feq) termination etc. aff ea sau: fasta (& 16 IRS)
root aq with the loss of
Thus in waar we get Mead for the read in the
but in the case of ff which 1s
the penult nasal
keeps the
uqus as fie Hq, the form is ftid which
nasal intact. }
ffs, frafa, 3M etc. it
(4) At the end of suffixes like fafa,
consonant from being
serves to prevent the preceding
| (shaq ancora) |
regarded as $q by the rule teu
on zfa it serves ©
(5) In the case of the terminati on, the Abl. and
adeclensi
distinguish it from Stas in the
Gen. singulars are not identical in form
ants for ease of
N.B. = is often added to conson
in afa in the sutra
utterance; as, sara, Taq etc. The %
Vartteka are gpa
Wiarat (612132) and in aif in the
rding to most
(S812) is for ease of utterance acco
=
grammarians, ‘SR eT: | armel «wai
of the yess
Subodhini. Jinendrabuddhi, the author Way in this
however, makes capital of the 8 in Tanta

scarmaiaata sfa ametswafaepes, Arafereareail

stata |(Nyasa IV.1.31).
Saa
ast 4 mong
There is a good deal of difference of opinion ase O
the ©
grammarians as to the anubandhas % and ein
of them would have pee!
the root amg Though either
sufficient to indicate Atmanepada, F7, (311) can only the
andha & according
obtained by means of the anub
generally ~
rule aqfada eee:. Hence grammarians
ly shows that i”
that the double anubandha mere
anudatta does ®
Atmanepada indicated by an indicatory a
hold good in all cases.
ha ows that the root does not ad
AN,
" ((1) In the Dhdtupat sh participial ouffixes @ a
e augment—_—§ before the past
( aitfedt Frerar i 12 12%). Thus at ast — oh: at fast — 9s
root 1s fimanth -
f a

(2) An anuddtta% shows that the

rn
=

eareat Gs] — ear | Wivifesarht: — qed sal

vote
 The Anubandhas of Panini 483

o—
(1) To indicate roots according to the dictum: Hat
Ug, as areifuwegiaary (9 12) where Wa means the root
WM.
(2) An anudatia 3 indicates Atmanepada; as FY Ware —
ATU.
(3) A Saha Z indicates Atmanepada when the fruit of
the action accrues to the agent; as, Gq STAC, Wy aA etc.
(4) 3 with the harsh unaspirate mutes indicates the
entire varga; thus # stands for #4, ¥Y for dat and so on.
syed Wavy AICAA: (2 Ik 1).
(5) Sin the Dhatupatha shows that the root admits the
augment 8 optionally before the gerundial suffix ¢4. staat
aT (9 12 &K&). Thus 34 — wae Weal |
(6) Zin a suffix shows that the feminine of the word
ending in that suffix is to be formed with ay. shag
(191%). Thus aq — Wad Sag — yadt) Faq— casa |
gy — stadii wa — Fei. Such suffixes are : Ag, WTS sae
Wad, HY, SITY WL AGT, AW.
(7) Aword technically termed 7 formed with a suffix
with an indicatory 3 shortens its final vowel optionally
before the suffixes ®&% 7 etc. faghau, fagaiaa; vara,
START etc. Sfiraa (& 13 4) |
[Bhattoji considers forms like fagytea incorrect.
According to him there will be two forms only, one with
at and
the short vowel and the other with Jaa. Thus fagia
the only two forms available. Trey
fag are
quotes : Wht
faguintergened after, The Bhasavrtt, however,
n 1s found in the
aun vediaat ay. The quotatio e d edition
Ravanarjuniya or Vyosakavya (XIX.19) the print
of which reads Fede, ]
augment 1 (TD
(8) shows that the s tem will add the ,
hnically
after the last vowel before the terminations tec
y to roots with
known as WaaqneaM. This does not appl
an indicatory %. sfteat adamerisad: Cok tso). Thus

4a, Far etc.

 mar
484 Technical Terms and Technique of Sanskrit Gram

tors, is for
(9) The 3 of Yaccording to some commenta
the purpose of the rule : Hamers: § Ps RW where FATS:
would mean the final of 444 is changed into { except
before terminations beginning with %,
[Like &, $ and 4% is often added merely for ease of
utterane; e.g. fT. ]
s— (1) In the Dhatupatha shows that the root admits the
augment $2 optionally before Ardhadhituka suffixes
beginning with aq, watagfagataysiadt at (9 R ws). Thus 1h,
— TTA, TTT |
(2) An anudadtta & shows that the root
Atmanepadin; as, AT Waray — ATI ay Wed — amd) A.
ome Ged A — ALI |
(3) A svarita indicates Atm. where the benefit of the
action accrues to the agent, as, Te dawn — Ted. .
m— (1) denotes roots according to the dictum safer
Ug Tea.
[According to grammar such 34’s are suffixes and !
z30¢

anubandhas. |
(2) An anuddtta®% indicates Atm.; as WY — ard, 4
aed.
is Atm. when the
(3) A svarita% indicates that the root
benefit of the action accrues to the agent; as, a4
— arafd, aad.
!t -
(4) shows that the root does not shorten its penu
, in

the reduplicated aorist. aeifrmegiedy (9 fs 12) Thus mies

HINA, AY — aaa, Gl — aeernd. ”
ed wit
(5) In a suffix shows that the feminine is form
%; Sfraa ¥ 8 &. as, AG — TART,
Such suffixes are : AqG41 [3 2 lov], J [& Ps 18], wd. :

(6) shows that the stem will admit the augment 1

after its last vowel before the terminations technice :
known as Hameed: sitesi adarmearserd: (wig to). Phu
Te, Tete.
 The Anubandhas of Panini 485

(7) A word technically known as Wet formed with a

suffix with an indicatory shrotens its final vowel
optionally before the suffixes ®, 7 etc. Thus qataa, aaa,
eraTaaT, eI.
[It is difficult to understand why in the Kasvka,
Bhasa@ortti Siddhantakaumudi etc. examples of Als from
words with indicatory % have not been given.]
@— shows that %€is to be substituted for fee in the Aorist
Parasmaipada. qwfeqmeqgicd: Wee 224. Thus TAG
ATA, Mae, ATR,
= shows that the root does not lengthen its vowel in
Aorist Parasmaipada before 4 preceded by 8
Saar eaTay-afeany 8 124. Thus, He THAN: SAHN!T,
st— shows that the root forms its past participle with 4
instead ofa. sifeda 21 4. Thus Si WF SLT.
Fi (1) In the second Sivasiitra ® s for forming the Ware's
ath, FH and Se (seh: WaT ae: & 1 Rok FHT TTTST 8 8B, SIMA
S12 1G).
at
(2) In an augment indicates that it is to be added
the end: as, 3vaait cfadt 2 ig me. [The augments with an
indicatory ® are AGH, AIR, ATR, WR PR WR, SH, fayR,

am].
that the abhydsa does not undergo
(3) shows
lengthening in 4 and 4 Gh craisfhd: 9 fs 123; as, qari
rding to
(because it has taken FRwith an indicatory # acco
the rule wasyfaartes) but WAM.
is weak and that before
(4) shows that the termination
a nor vrddhi, but
the termination there is to be neither gun R in
: to undergo
is reduction or : WAA
the radical vowel
cases, etc. fesit A iRh afacafrasnetat fafa & 18 Rs
some
aififeal we sour: feet Gx oral aftercdfa Feria 2 MeRS I:
Tenia
9121.
HT, FI, PIL
Such q suffixes are - FH HA, He, HAH,
Fors, 4 FS
ea, Berg,fee,FA, FT, HA FG, WAG,Fev,
486 Technical Terms and Technique of Sanskrit Grammar

FRG, 4, AG AF, TY, TY, BH,HT, HHT, HTT, HG, fee,THA

fry, FH, FH, HY, oH, FH,
(5) In the case of dfga suffixes indicates that the last
syllable will have the acute. fha:-& 12 1964 Thus ars
formed from 7 with the addition of the ®& has its last
syllable acute.
(6) and that the first syllable will take vrddhi; tard 4 -
912 122¢ Thus 4a + Oh — icH,
[The fsa suffixes with an indicatory Rare AK, Bh

(7) Before a suffix with an indicatory # or <4 stem

ending in a nasal lengthens its penult; aqwar4
fasta & rs 184 as, WIA, War etc.
(8) Under similar circumstances a penultimate q of a
root is dropped; afifeat ta sauna: frefa ¢ re ies. Thus ee
ead. from Aa
(9) Before an affix with an indicatory Ror {the re”
a1 df etc.
Lo

drop their nasals: AF ;

pMaHat ate teefa & rs 2. Thus a with ® gives 44, Tr with
® Md and so on.
@— (1) A word ending in a vowel as also the words 34
and ig¥q admit the augment 4 (44) if they are une
upapadas of the root to which a suffix with an indicatory
is added; Sefgyesates & 13 Io as BOATS, Fae. vel
(2) When a word of one syllable ending in any — 4
except 4 or 1 is the upapada of a word formed with at
suffix with an indicatory @ this word admits the augme
“4 which behaves exactly like the 314 of the accusalv’
ef

singular. 34 WaHrdisy yeaa e116; as, TA:

Taqar:, YAR: |

(3) Before a root with a suffix having an indicato’y .

all words shorten their final vowels except indeclinab -
as, Sfemmn, effre. This rule does not hold good Ma
case of avyayas. Tame IR IRE: e.g.,
STATE: ‘ A,

The 4 suffixes with an indicatory @ are a4, ay

GY, LAMY, GHA and S|
 The Anubandhas of Panini 487

[@ at the beginning of asa suffixes is not an

anubandha but stands for $4.].
According to later grammarians, an indicatory shows
the weak character of the suffix fasia 9 — 3a Tesh aay
fafezad! There is only one suffix with an indicatory 7 viz.
Te (Tosteyal AA 3121238). Many grammarians however
regard the suffix as Aq!
[In the afea suffix FAR (4 182%) Wis not an Asay, but
is a part of the suffix. ]
Before a $d suffix with an indicatory 4 — the palataA’s 4
and Ware changed into their corresponding gutturals Rand
7. Thus F+W] =Wh, G+ B= IM (AST: HATA: w BWR. The
Hi suffixes with an indicatory Gare ¥, Py FPLand FA
[The Gal 4 stands for the suffixes ® and 74 — aaa eT:
21212. At the beginning of a asd suffix 4 is not an
anubandha but stands for 34].
(1) In the Dhatupatha as also in root-forming suffixes
indicates Atmanepada; aTqeratsd STATE & 13 182 as WE— yd;
&
ae — Wd. The verb-forming suffixes with an indicatory
are 34e, Hag, Meand 4g
[In the case of fag the Atmanepada is due to te and
arty
not to 3tMferiet according to the afer — fasfaed we
2 IR as in the rule 3Meldisd SIAATS 2 1B 182 TFcomes after
stad This is known as weawfastaae. ]
is weak; thus
(2) shows that the suffix or termination
a fei suffix,
neither 7 nor afs of Fh can take place befor
:
but the root undergoes weakening in various ways
fereta A 2 12
& 18 18
aise freatiahirafag atareorapsral fafa FH
_
FST ATTA AAI, & 18 189
with an indicatory Gare
The suffixes and terminations
EWS IS _
ae foe wah, aE Wg AG Mg
takes place because it is not
&R
[In the case of frefg |
but & that gets gg].
s 11SSe
+ lakaraTE and Fx
qgal
(4) @in; the case of the fou
changes comm
shows certain terminational
 it Grammar
488 Technical Terms and Technique of Sanskr

them; e.g., fret fed: 3% 188 (in the case of lakaras with an
indicatory € the of the terminations for the endings of
the Ist person is always elided) Sada 3 [6 If00 and aeaearaa
MTNA: 2 6 [Q 08,
(5) In the case of the sup terminations & ofa, Sand
fs the indicatory Shas been made the common element
to show certain changes in the stem before all these
endings :
Ufeta9 13 1222
AML TET: 9 12 1822
STSTS: 9 13 1223
AATH: PISS 9 13 122
faye fecrargaransary9 13 1224
Taft BAA — 20S ie
(6) In the case of Sea (substitutes) an indicatory g
shows that it is the substitute for the final letter of the
stem (according to fe 212143) and not for the whole
stem, though the &tey may consist of more than one an
according
and should thus be the substitute for the whole
to ATH AIST 2 12 144.
Such substitutes are : 34S Hage Aas, HTS, STAG, FTX ws

[In the case of Wd since & serves cert ain othe!2

purposes viz. the prohibition of J as in ead and ek '
in t{ald, the reduction of 4, 4 etc. into , 3 etc. (TAM .
in 4a from BB etc. it does not indicate Fauew).
(7) Gin the 3rd Sivasitra is for the purp?
pratyahara VS. A
(8) In the case of the three feminine suffixes ort, ne
and @, the indicatory € helps to bring together al!"
three suffixes under a convenient designation st for
purposes of the rules : gamunifasteard iv.1.1 crear 2
ESTES Ea) el vi.l.68 Sareea ae vi.1.178 TETRIS
Vi.3.43 Sah: dase vi.3.63. A a vi3.64 FF! geal
TMTaSs vi.3.61.
 The Anubandhas of Panini 489

q (1) shows that the final vowel of the word formed with
the suffix is to take the acute accent (Sera) Fad: & 18 12&3.
(2) The 4 of J4serves to distinguish it from Jland at |
the same time enables it to be included under the general
category 7 in the rules GRBAa-| vi.4.154 Goa v.3.59 etc.
otherwise it would have been excluded according to the
paribhasa PATHS A ATH] VST, .
qin G4 serves merely to distinguish it from GX for the
purpose of the rules Gf se: vi.4.94 etc.T
i a occurs in 44 only where it merely serves to distinguish
it from A for the purposes of the rules Wa: Wi viii.1.17
TAs A: GF vi.1.103 ete.
aes (1) In the Dhdtupatha shows that the root takes
A tmanepadin terminations when the benefit of the action
accrues to the agent Takata: wafawd fearac 1.3.72. Thus
UTS] WUT Aad |
(2) In a Hd suffix shows that the suffix is strong and
causes the vrddhi of the final vowel or penultimate & of
the root saistfufa and aa sae: vii.2.115-116. Thus My
with 2 yield Mm; WW with 4 — Wa. Such suffixes are: 3%,
SHA, HA, GPL, FHA, HL, Ty Fl
(3) shows that the word formed with this suffix is
accented on the first syllable; faraericry vi.1.197.
(4) In a afed suffix shows that the first syllable of the
word to which the suffix is added is to undergo vrddhi:
atest: 0 12 220. Thus Wid + AA = Usd; SA + HL = TMG.

* “case: ‘gfetag’ ‘qeecfa’ saet aeremecnfaenand: | “Tad! Cat.

GIR RG3) geRiteaeta a vale, ‘smgena’ (7. 2B) we fs! TA
“AWA! (a1, & 122) sft Ga Fa: gery wet Fae HaGEAA! Fea TATA |
oly ceemeish afatvela veh wenife arly ‘oa’ ft
VCR.
fasting:iea
Tatterh ie | s5 araeaee afeaquent d
17 Tara
t gant que, aang Gla wea.’ siaface faafe:, GWE KISIM
RR hare eerarfiy wench ARITA: 1 CHATS GATT AFET." SetGIT
Weeritenen qh-aeEeaae: Il ain vad verre | teeta $1
aa
farrary
Raton gen ena ws aTyfela | aaa sa a, WTA! FE PTA
Sash ashacgachfar ares, SATA: eta age Tera fers ATH AA
 mar
490 Technical Terms and Technique of Sanskrit Gram

The afed suffixes with an indicatory are 24, 31, SHA, TA,
PA
Gna, Tid, 4, AG Ae, dA, SHA, SA, TS, Aq, A, Wy,
Ha, Fl, FA, A VL, AL
(5) The suffix 4% is added to a stem which has already
been formed in the sense of fam or 2444 with a suffix
with an indicatory 4. fda aqweraid iv.3.155.
(6) A qayed after a word formed with a suffix ™4 with
an indicatory Wis dropped “aataarifadt af qtoren: 1.4.58.
[In Fararad fefe III.1.40 4 serves to form the yearn
#3 according to later grammarians, but this 1s simply
absurd.] f — In the Dhdiupatha shows the root admits ®
to signify present time. Sa: w: III.2.187.
(1) shows that the augment is to be placed at the
beginning : staan efpdt 1.1.46. Such augments are : Bz, Ag
BS 84 SA, We Fe, Te We, Age, fe, Fe, ve, Wz, Te, MMe
(2) In suffixes, dhdtupdtha, ganapatha, TANARS
15
V.1.114 and tarme V.1.113 shows that the feminine
formed with 4 fegema — IV.1.15. Such suffixes are wae,
“AR, %, CH, ey,fear, Toa, am, ey, Ye, Be, MA, TH 1
Hae, ede!
se
(3) In serves to form the Wener Gz for the purpo
of the rule FStqanes 1.1.43.
[This, however, is hardly proper. Since the nom. and
acc. duals are identical in form in all Cases; the
termination ought to be the same for both.]
(4) In 4 serves to distinguish the Instr. sing. for me
purpose of the rule: fedr dreds: II.4.34: otherwis©
fede: would mean fda, sq and ¥ excluding the
Instr. sing.

(5) In te, etc. serves to distinguish them from the fet

R's for the purposes of the rule fea smeqrgertlek{11.4.79-
| (6) In the feminine suffix em merely serves -
distinguish the suffix from aqand

Though according to the paribhasa — eqn

{IAF WET, 3M without any < could have ini
 The Anubandhas of Panini 49]

distinguished from “TT and ST still STZ is required to

denote all three suffixes for the purpose of the rules: S474
gifarafeat [V1.1 Seer cre alah Set VI-1.68 3S sm:
Gs:
VIL.1.18 Baraat Fe VIL.1.54 ate as: VII.3.105 steratad
I1.4.82 etc.
the
In the Dhatupatha shows that the root will admit
suffix 344; feadtsys L1.3.89. Thus 44 Sfexut FAY:.
zat i.e. Z, d, SS
[When Z is not an arubandha it means
and U. stufeq waves GACT: 1.1.69.]
an
[At the beginning of @fsa suffixes S 1s not
Sqqnrid,
anubandha but is replaced by &* or ®. seta: and
@: VII.3.50-51).
ion of the
Indicates that before the suffix the fe port
Thus fa4_+ sit —
stem is to be dropped. (¢ ¢ — V1.4.143.)
are Sq Sq, SI, Sata,
=f The suffixes with an indicatory &

and S47.
Though Panini lays down that the fe portion of a stem
be dropped before a
technically known as % is to
& still because of the
termination with an inidcatory
is elided in those cases
presence of the anubandha %, fe
€.§., before the St of dz
also where the stem is not 4;
dhanta Kaumudi.
fecqumaeaieseata eeitg: — Sid
t the root admits the affix
In the Dhadtupatha shows tha
8. Ra
f* with the augment 7H 130: fi: IL.3.8
[V.4.20. Thus 379 We — We.
but represents wa at the
[It is not an anubandha
ing of dfz d suf fix es. aea aifaa: weaoul yeqareary,
beginn
VIT.1.2.]
or
that the final vowel
(1) In a Fa affix shows stat STATA:
take vrddhi. apqarfourfa,
penultimate short 4 is to fag, TAT,
<0"
VII.2.115-116. Such suffixes are 3,
e, wr fo
e, for
r, fo
e, re ,Fo r, 7A ,TH R,FHA,PAM, VIA MSS
or is to
the first vowel
a af zd aff ix sh ows that
(2) In 3 gives MK.
e: 9 IR IRR; qe with
undergo vrddhi, agat
492 Technical Terms and Technique of Sanskrit Grammar

Such suffixes are 3, By, 2cquy, am, sy, cay, 1, fs, %,

vad |

(3) The % of the root @™ serves to distinguish it

from the <-roots.
aq — (1) shows that the vowel to which it is suffixed will
denote itself alone and not its Fas containing a greater
or less number of mdtrds (morae). Thus 3d Sduat:
VII.2.116 means that the penultimate short 3 is to undergo
lengthening before a suffix with an indicatory 4 and "k,
but 3% =t VII.4.32 means that before the suffix 1,
‘a’short or long” is changed into %. Thus the rule applies
to Yai as well as to MM, so that we get Yaritsa and ret pa.
But if we change 34 “= into saat the rule will be
applicable to short 3 alone. TReMaHTTS 1.1.70.7
(2) shows that the suffix is waft. faq vaftaq VI.1.185.
Such suffixes are Gad, Wad, Wad, Fd, ALI
[Wis often added to a letter for ease of utterance. Thus
on the rule : atgmaeasm: fafeat fae a III.2.171 the Kasika
says : afefa dant qaqred:, 4 cad qt | ar ye arate RI BIA
RA TIS T The same appears to be the case with most
of the @’s after long vowels. ]
i a (1) shows that the word formed with the suffix 1s
accented on the first syllable. ftenfeficay VI.1.197.
(2) In the case of a @q suffix with an indicatory ",
except one with g an immediately preceding wor
technically known as "fa retains its own accent; €-8-;
Here fd is formed with the #q suffix fy which has av
indicatory 4 It is immediately preceded by ¥ which »
i
* , . .
A vowel when mentioned without an indicatory ¢ denotes itself a° 5
well as its savarnas i.e. letters hom ogeneous with it according to the rule
AUS Waa AIT: 1.1.69,
t Patanjali explains in this rule to mean both a: W FAM and a
a Thus in the rule HES TN: 1.1.2.
aq, denotes short 4 and G& ee
ong @ and 4 to the exclusion of plut
a or prolated @ and sit The same”
the case with SUEZ 1.1.9.
 The Anubandhas of Panini 493

technically known as 7fa (faa 1.4.60). Hence in wef, ¥ will

retain its accent and @fd will be unaccented.
areal a fafa Head VI.2.50. Suffixes with an indicatory {are

aq, for, Fear, fart, SHA, HA, EA, SA, SAS, PY, Tay, fey, TH,
T, dq, TH, A, TY, TH, A TY, PH, TT, TY, TF
Wer, UA, A, TA, SH, BY, Sh A
oe (1) shows that the suffix is strong i-e., before the suffix
TI or afg takes place.
vowel
(2) Before a Pq affix with an indicatory {a short
admits the augment ¢ ee fafa Hid q& VI.1.71. Thus a
fafa afufaq formed by adding T&T (0) to afar.
Such are; +,
(3) shows that the suffix is unaccented.
TM
AY, WY, HY, HET, FT, AY ITY WAY MY FY SY TTA
TY, Sorry, ST, TTL
faq #d suffixes are : F714, afr, wey,
(4) The Yo gLfforms the YER’s TL and FATT lz [Rkx.
t it closely with <TY
(5) The ¥ of WT serves to connce
and STso that all three are denoted by S14,
ns qa]
[yis not an ‘anubandha’ but mea
afad suffixes is not an
[th at the beginning of
‘anubandha’ but means 314%. ]
Ro ot s wit h an ind ica tor y {OF that are regarded as
ie (1) take vrddhi before T
having an indicatory 4 do not -_
suffixes etc.
pse "]:wi
An au gm en t wit h an indicatory floa
(2) . Su
the word. foeas<ar TG Lt.
the final vowel of
are : 7, F 3% Th TkTeh M |
augments
to say whether the
a

ficult
* In case of a fad substitute it is often very di ddition of [makes
for we often find that aaaiu
Tae is due to sale or fara,
t
a4ahie] and though there 1s the papell sfaras and
ibhasa
the Substitute
82
subtle nye ri_ aargy tee
stammarians have often drawn a
toji+: the‘ subst jtut on
i
- Th us according to Bhat ttO} "74.127
~— aprepany, wale:, TF
494 Technical Terms and Technique of Sanskrit Grammar

t~— Shows that the last vowel but one of the word formed
with the suffix is SeMd — Sord4s fia VI.1.217. The three
suffixes with an inidcatory tare Slat, SITAR SRT |
a— shows that the vowel immediately preceding the suffix
is actue. fed VI.1.193 (sera Yayera4q). Such suffixes are
aI, HY, Gea, Ia, TAA, Mei, VIL, Tea, Ae, Are, Tel, Act, At,
ae, The, [IV.2.54]. ac, fea, Ty, TA, THe, fae:, [[V.2.54} Bl,
[In ce, we etc. and Tf, Gis not an anubandha].
I — (1) An @@% with an indicatory wf becomes the
substitute for the entire word and not as usual for the
final letter : Adare Pq 1.1.55. Such substitutes are :
oI, FU, A, TA, AT, FA, Vi, VA.
(2) A suffix with an indicatory Wis regarded as SAM?
faghrg, wdergey 11.4.113 (cf. snea saewistfa VI-1.45
reqaergaai fit? VII.3.75). Such suffixes are GY, VI, WW, Mi,
WAR, WI, MAA, WA, WA, Ay, Al, A, VW.
7 (1) In the Dhdtupatha shows that the root may admit
the suffix ag Trefyeieriss 11.3.104 Tye asa — AT.
(2) A word formed with a dfed affix with an indicatory
¥ admits GY in the feminine. faentafeaa IV.1.41. Suffixes
with an indicatory Ware : ¥, 4, Wed, Ghd, BS, FA, BA, SL Bel,
th, Uhh, ME, WS, ONG, Hy.
(3) The ¥ of the suffix ay simply serves to bind the
three suffixes together under on name &. But for this %
would have meant this suffix alone to the exclussion of
and according to the paribhasa — Weqryr™
MAGLI AHA WET |
y—
Before an affix with an indicatory 4, a base is technically
termed ¥e. fafa 4.4.16. Such suffixes are Wa, Oa, 7A, TA and
ue (atti). Thus when the suffix @ is added to aad we Se
aad but when Sis added to Yad we get aaqera. Before the

4 of &, aa is not a 4 but 4 hence the rule am sai

VIII.2.39 does not apply, but since 34 has an indicatory &
before this suffix ¥aqis regarded as a ¥@ and so the rule
Sd applies and changes the dinto &.”
* T need hardly mention here th the above !ist I have
, at in compiling
derived great help from Boh tlingk’s ex Fiin his edition O fPaninl
cellent index
 The Anubandhas of Panini 495

The Vibhaktis

qa According to modern commentators of Panini the

anubandha 3 serves to distinguish it from other 4's in the
rule 3dtraaars: VI.4.127. But for this anubandha the sutra
would be read 34a@Ss!: and mean HAA {will substitute J for
its final except before endings beginning with {and thus
the form aTdda will not be correct. This, however, reminds
one of the sayings “faasf aga a wait aafe: and aan_sha 4
qurafa:, for in the case of the rule aaaraeqdanera stp:
VI1.1.168, the anubandhaZ causes confusion and we are ata
loss to determine whether the Gis the nom. sing. or loc. pl.
or both, until obliging commentators, inform us thatghere
refers to the loc. pl. It is, therefore, best to regard the J as
being inserted for ease of utterance, for though in tara.
etc. Fis followed by the vowel 3if and so there will not be the
slightest difficulty in pronunciation even if we drop the 3,
in
yet in such cases as waded: qertsararaa tie VI.1.132 and
the case of the WaeR Wz the consonant being immediately
followed by another consonant there will be considerable
difficulty in pronunciation.
that Panini should cause this
It is unfortunate
ngs by adding
confusion between the first and the last endi
Aindra_ school,
this 3 to @ The followers of the
others read fa
Sarvavarman, Hemacandra, Vopadeva and
ular of both being
for this ¥ but even then the locative sing
inevitable.
ata certain amount of confusion becomes
SLoccurs as an anubandha.
WY— Thisis the only instance where .1.17
of the rules Wa: wt VII
It is necessary for the purpose h consisting
Taree a: Ufa VI. 1.103. etc. where wand wa bot
qfa’”’ (& 12 203)
De resicorell a:

* Sq-weistereran FA:Bt’ (9 iz kV)

‘é 3 ?
.

| se ar Preg arer aree nfec a a, SatSHRTTE FSE

Serralferafaaeel E
THTREMTT stlfoe TH
AT TN ST A,
nc Ret: TETENG
Soreyeaahfecesear,aeraFrere Fere
ST ea at a er e
496 Technical Terms and Technique of Sanskrit Grammar

of a4 have to be kept separate. It may be said that the two

endings may be distinguished by adding an anubandha to
any one of them 1.e., by saying either Aq and Wor aq and
aq for nom. and acc. pls. respectively. Tattvabodhini
following in the footsteps of Jinendrabuddhi and Haradatta
answers: If you say Gail 44 then by Sandhi you will get FIA4.
and people might think that the termination is 4tqand not
ait. If, on the other hand, you drop the %LofWW then 3TH aie.
aq will give a4Sq and people might think that the
termination is 3 with an indicatory S signifying that
before the termination the fé portion of the stem is to be
dropped, so that aq + 3it would become a= and not aaa.
This explanation is most unsatisfactory and hardly requirs
any refutation. GMa srearrei VI.2.117 etc. will clearly
indicate that the termination is not feq. The fact is that it 1S
merely for the purpose of symmetry and euphony that
Panini reads both 44 and WLwith anubandha.
3itz— ~\ The anubandha Z is for the formation of the WaTeR FT
(¥ Ht HH st and 3it) for the purpose of the rule ystqaoe
I.1.43. It would have been better to make the ending for
the nom. dual and acc. du. identical as there is not 4
single instance where the actual nom. du. form differs
from the acc. du. It may be noted in this connexion that
the technical term for 4éaraeaM in Katantra, Hemacandra
etc. is YZ which appears to be coined after this 8
Samksiptasara and Supadma also use the pratyahara a
Sarvavarman, Hemacandra, Vopadeva and others who are
followers of the Aindra school read Si for both.
WH— See Fy,
ec —
found in practically all the systems with the anubandha
< though the earlier form would appear to
Panini in the rule are fam, VII.3.120. The z here serve*
be arg used i

to distinguis
h it for the purpose of the rule fcr
:
114.34 which without the z would read fidtat
eea™: al
mean fed, aq and ¥ thus excluding the Instr. Sing-
The anubandha F
is used in practically all the system
and enables us to bring together the terminat
ions
,
©
 The Anubandhas of Panini 497

&and f¥under the common name fe for the purposes of

the following rules : Sfsfal Spe: | Usa: | GAH: BTSSTAS |
fast fedrargdanary VII.3.111-115 and feta ea 1.4.6.
oa— For Ssee & The 8 of &€ serves to distinguish it from
64, as in the a-declension and in the case of pronouns, the
Abl. and Gen. singular differ from each other. Cf.
csfastarqen: VII.1.12 and Sasa: Sqr VIL.15. -
S— For Ssee &
fe— For SseeS
q— Vis for the formation of Witer’s Land Sy,
fre— In the fae fasth’s the Tin fy, fay and fAY shows that the
terminations are strong and unaccented. The dof $Z srves to
distinguish it from other 8’s for the purpose of the rule seis,
3 Ié 1808 which prescribes the substitute H for ¥ in fg. If we
simply say Winstead of ¥2lSq, would be the substitute for
the 3 of afé and 4fe also, and so the forms udafe and winfe
would become Uae and WHE respectively. Some say the qin
$2 serves no useful purpose, for even if we say Wwe get Tre
by stqgfa from the previous rule and as ¥ is the only fererext
consisting of short 8 only, the rule applies to § alone and not
to af and Afé. But this view is not correct. Since ¥ is faxrq to
fasea, $ would mean S4R according to 7 faeries -
2 12 692 and thus afe and Afe being HRM will come within the
scope of the rule. Therefore the < of %< is not at all
superfluous. The &in ufse helps to form the Waenr’s fas
(including all the frgfast’s) and te. Sakatayana has a new
yarn ‘ye’ to denote all fates {Zand fre The 4 of Wey
&.
indicates urddhi of the final vowel or penultimate short

aisq sha
:
. Vgopan amas 4 vata, secured: wafer, fed S
fagraones: | Cee” Se watery Re, oR eT TT |
ai e | nt wlafe
are
hfay | saa fae
méad wafa aH“FAST GH TATE
Peatera nycmeckay a feaaeorenr TanTTETT Sa
uf Tacae: | apr, SmI ference’ FaerTTA
 Sanskrit Grammar
498 Technical Terms and Technique of

Feminine Suffixes

Of the three feminine suffixes 29, 44 and STF only ol

remains Z, 4, § and ¥ being regarded as ‘canubandha’s. The 31 of
37is accented, whereas the 4 of e1¥ and sI¥ are unaccented.
Then again in the case of STY since © is an ‘anubandha’ the
portion of the stem is elided (2: & 1243). Thus Bray Sry, AAS
adm. The ¥ of AY simply serves to include all the three suffixes
under one designation S14, whereas the ¥ of aq and oI
indicates that the suffix is unaccented. In the case of art
though there is a conflict between the rules Tad: & 18 R&2 (1.€. a
word formed with an affix with an indicatory 4 has 1ts last
syllable accented) and aqerit aafadt 21% (ie, Case
terminations are unaccented and so are suffixes with aM
indicatory 4), still according to the panbhasa faarrae ray, (in
cases of conflict between equals it is the subsequent rule that
will prevail), fad: being the subsequent rule prevails and words
: formed with the suffix WW like sPatea, FIAT etc. accent their
final syllables. If we had read SY instead of 4, ATT in such
rules as Sargnfaafeard¥12 12 STEM Seis GfaesyeHhEL & | |e etc.
‘ would have signified 31 alone to the exclusion of eI and
for there is the Wm, tarnryHe Tet a aarEes WET
when a suffix, augment etc., are mentioned with one
“anubandha”, suffixes, augments etc. with two anubandhas are
excluded. @ has been attached as an anubandha to @ for a
same reason, for without Z, this suffix alone would be denott
by a1 to the exclusion of 49 and Sq. Ay occurs in 1V.1.74-79
aqin IV.1.4.9, Shin [V.1.13 and 24in 1.4.82, IV.1.1., VI.1.68
3.63-64, VII.1.18, 54, 3.44, 105, 113, 116, 4.15. .
Of <y, 4y and Gq only ¥ remains, = 4 Gand 4 a
and
‘anubandhas”. % shows that the suffix is unaccented
indicates that the word formed with the suffix has
. at
syllable accented (fsrrafeficay&18 1830). serves to bring at Ss
them together under a convenient designation &, for thous
even without we could have denoted all the three suffixes
 The Anubandhas of Panini 499

2 still that would have led to confasion with 3’s other than
those of the feminine suffixes. Similarly the ¥ of SY serves to
include it under the general name & for without the {or some
other letter @ would be Warmers and the pandbhasa WAMU A
gaqaarery would have excluded aq and S7 from the scope of
@ in such rules as IV.1.1 etc. BH occurs only in IV.1.73; aq in
59, 61-
1V.1.5-12, 15-16, 20-21, 24, 26, 39, Sqin IV.1.25, 40-55,
Bh and’ al in [V_1.1s V1 68, 178} 3,48, 01. 63, 64, 66.
Another feminine suffix is SF (SFI ¥IV IS). | has been
tagged on as “anubandha” to the & simply to distinguish this
ASAT: & 18 189&
feminine suffix & from other &’s. Thus the rule
suffix & like
applies only to words ending with the feminine
S's.
qela-y etc. and not to words ending with other
in the
Several Krt and Taddhita suffixes will be dealt with
next issue.
be
N.B. On page 453, line 9 “and the first syllable is to
indicates
accented” is to be omitted. Though this anubandha
ding to
that the first syllable takes the uddtta accent accor
syllable will
Panini VI.1. 197 still in this particular case the last
be accented according to Panini VI.1.159.
 oe , 4° Pats, Pits 3 tal ee
’ ’ b ) aad € Y ; . f

6's : ee mM ty:
Al: Ve ' “ va) » #,\ 4,
. ri . ne we yea:
poor MN Toe
“\
of we,pa Daa 1 ery
ef.
. 7
a4 i
Pie A’~e)*
Wn , ‘ i
a 4 } 4 NP ths, \s :
4 ‘ Qty ae b ‘
- é - fi 7 Ms ry , KY hf
5 tn a ary vPro om.
e.
a?
& 4 ‘“ a F ‘ a
: ?
5 ty : i 4 t* j5, 4) “ar
* “ a “e 4 ’ ad a 4
' » mt f i ° . .

. - 7
» -~

7
Lis *
. DAA, 7
we
ANN .
i!
PR ae ) — IB - ih
bP ’
4
4 oo - . . , a, ‘ - ‘

2 , r ais . I , fe
Phair ae, Okt
a’ 2

, - ’ 4 + r
: A - s
‘ . 4 - 7 1 4 .
, rT ar -
. f ’. , ie
’ - ‘ « of dA
‘
“/a74 re * + a ‘ ,
:
. , 7 f . dus - @ . a ma 7 .

whe or : ¢
alee,
© a ee,
oe
>* -
ae- i

“ _ | Pe as “ a’
.
4 5 A Fh
’,os . ; od Y ; { .
7
2

- a? > e \ ll

- Z 4 . £ » a
1 . r «eat .
ae 6 . . r ‘ é a
iy , - . - ‘
-
*, ' - 2 - - . - =)
* “wd é , * -
4 a may = be
4 ‘ su # ‘> -
‘4 > , 7 > F

' - + > *' :

*. - ‘ u ~
14 MS
* a it x ie i
‘ . Aw - ih
wae) (4 , f + ¥

; - rs . ~ , ° -

Mi é : t (ie .
w 7 ‘ ‘ , 4 ‘yy _

P, ¢ 3 ~y ~ i} - ng - e > e
’ 77 7 -
: - - ‘ 9
Pad : a A , 4 -) i” -

r “asFY > t % :‘ wi
gen ae “uP WS? Om, Fa ay
' mF .
Rese
yr 7 i ye ” » 7 - o/ ; Stews
f 7 s > % i @ ; ’ b a
eyx a
F -
, el
y, &-5)- on *,t % —
—
«zr
$2
: Ay f i 7 7%, “et Du wt A uF 48 .
me wi Ww j a: f” wey vv ,
maef i : 7
ed» ,
Nae
oe
ee
Ye
eg
why! , Lie? yr ’ y TA, s
Suen S o ! ay, ‘ ee yA a “y ? F 5 ;
- sap 0) Mite
, : os p ke ‘gi et, 8° pt
oie ae 4 tic Pa : 5/500 py *
4 - igitit Ae ay
‘ ld . f - rye Md
- . . - 7 wy, wins a.
: (uy ' : - ed ‘ are pmo e ‘ nr
: “ ‘ Ai : Myo . DeywSPORT
AS ae oe
i? ; re
Ss t 4 ’ : Pip > TELS +

7 ~ ae ,
e5 - a ‘ 4 , , fh
' 4 3 . 7of Ps .
by’ + >i bee AMD yi he . a“
>> 2 “egal
7 - Ww 6
° . '
t ape.as
Lawet" ew
+,” .

,
Vi 2
wa
“ ‘
=e «
- a
‘ phe As
ss A
» .
RBs
a
* ca an dd , ie . - .

- ae | ° r { Ve ,
) : : , a -- al, 4 45
. ag
pe . 25 4
_ a" - ’ 9% a _ fia
7 % , 4 SA ' i. f . -
7 “es > de? ©
j “y Ney . . - 4 i?
oe : , + “AG “- fc wl A wT,
. : ; Py. vw
{7 ‘a
-ef "aa
@ 5 ‘ ee |
' he
bh
\ a
‘ ¥« : } /
: + Le oe
ft a si of A> G. J :
d e a 7 "¢ Pk " ad ad 4
3.
i) 2
“ ie + 7 ; a 4
. . oJ Ls - J ge
‘4 ji is ‘ — Fi.
Ralgt! }a
‘sao.
wt
at
7 ; - - » The s i | % i?
° 7 - 4 a on ‘oh tf : , Ty’ D Ay .
> '’ y

’ ; ee in
bee, ‘ eS
4 ae - an
2 MER, te
oe. 45 a}
- y > Ww. wl ag
Sify ’ 1. Oe a te) 7
- ; 7 . 7 ¢22. * ab ! “i
ry ; 7 ‘ys
A . : wv “a 7 ~ >
"Y : wll 4 3) eres
‘ 7 ¢ ‘ J ¢ . a?

- 5 ¥» « +" «4 '
“a” f | ‘
rae
)
Pals’ , a vee ty
ee NY Jira 7
‘ . . “* wera Te geet :
: . ¢ - ~ ‘sh ) / r
¥ ‘ , * : dag J 4 ‘el ‘
- . , of ; - wee Vo f ,
a 7 . ¥ *ghe 7.
j x . - t
4 lf
oe *
-
4 q%
;
be
AEE
PT
. , oe jt le oe pt Ae ¥4 7
5 q » Yee Mae
,
‘ ‘ye ae , Sad - Cpr,
~J " ne, Tie ny
y 25 as? J, Y 3 - eV
7 rd
§ a. ye
; De j . 7 7 ® J ore
. of ° 2
§

Y . . 2 a: x mS ot ~ Teie’. 4 : *, ‘
ay “ P) ope te a ffi el. 2~ < -
i ox a ti Wk en, ~ ;
= on | 7
.° pi
x, ah
-myt Ys
i'sey ‘ey ‘¥ t
C+" "a 2
2 ‘NW. » 4 r; P% we "a

~ f Fad,
_ *. ;le : ey
i) ;
> j Aa
vit
: a
hy 24
ee
f ete
J “eee
‘ve a ”
id SS *\) , t'. oe em ay .
; j “f
. tel
La
o* % :
:a : rs
- ,
‘
» J - _ °

3 f rs 7 - hla . ne ron] 0 4
s ; > oe : Sn
% » a 7 \
- + 7 4 ee ‘ ’ f
; | P sy ral,
iy va y Fr ‘ ~ 4we Ted, yak ty
Berne ia) apart. , y ee
aps, d's
o $ i eevee ,. oer er a.
Py
Te les TAaS
N Degd beCana
Poe GE TI FIRSShi be
Path BataOe
7 v4 e- Wid he 3 > "$e fy - Bad? v
¢ a _ ,

tr. . i " ms ” - | i oer fa

‘he e wt fe y bys . A yh, “ - - . Te: f Jts
-
tie a=
TF » 3 ».
Os
~~ r
PL 4BRRED ; es ArNa
ots
AN¥ ata!
ler
oot
Foy
- hol te ~iy «* oh FIP
J hy
; g% ot)
: ~ 4
tous
» at.
- by By be -
eh : Fant
yee 4 Wy! ' hy Kw 4 ie s -/ “ f Be
- > * ? “hy' s 7
iy of» tf eas4
Us oo Foal
° ¥ * i.
. = oa
. x* PERS
AD hd? Ro f nN 7 yea, ’
v he P
, “s
wud r “ye .
hw ? KS
ce.
,
8 s _
tol . iy
\ ¢
on AP
Sate
i. ;
v : wsee)
; -
=. . A, : : Dy FAo2 sy * 7 » he bas Pt KsAFieSe iy
ow an

ij Ee ip ts EN rene ey
Sie ;* Aw As yaad ve if : - -
‘ : fu : ; Pull eo ae a LE SRA "
, - re 2 th MW
pg en or ae. wy
he ed
yay
aA {
N)
Nhe! ;s Pl 4%. . 4: A rk
ray 7, 4 {4
Ba bynN gry° a
tae rd
= .
ral 4 s nal
PLP: ”
ge y ¥ ae ‘ i y aes -! 1 yee | yp 45 ees
t + h,.48 ad ney
ENG A S fae PS)
YR thle it Mbcnts Se
hate ie|py
ade trae oe
- od | Aur Pee Di
qr P : 4 - ‘
pr 7 fh . ’ - Ast " id
Ls os
ab yg . ue
» ween os ’ *
ad a,4 “> f
‘ pes ‘ ‘ 4 ]’ - I -~ pd » we, ye}
= '< >} wa % ' a >?
ee * - ay, ie :*
\es
7 > y, 4 % ‘ "rue 4, ve ; :e 7
7 - si rt7 ;
ooh
ty
§ J
rh
ji rue iss eina 4An PRN
Gf gl p] 7 nery 7 Yur ’ th¢
5 -
y TS _ 4 m it DP} - “he o> : .

ydets yee AO pts et TPA Ieet.

? Prive,trae
4 ©, he
bf i , ey| TTS '. a x 6 Jy "C Orviie! & vi
’ 7 . J BY ogi olant © a yn > *
x; f ¥ aT v %, a4 aed, at. “Jf
x Ad “if ®, ah. oe oh Taps ~ ’
Bf ae: 4° Yea, ‘ F 4 ligt er nig in t . Ww
. i) yh “OR we ‘ a tes a” ~~ s i,
; i) es , vue le Lg wad §) ,
 Kshitish Chandra Chatterji
(b. 24-11-1896 d. 08-11-1961)

Technical Termsof Sanskrit Grammar

Kshitish Chandra Chatterji |
"Honour to whom honour is due. Dr. Chatterji has broken
entirely new ground in his Technical Terms and Technique of
Sanskrit Grammar and demonstrated clearly that Bengal, though
fallen on evil days, can even now produce works on Indology that
invite and bear comparison with the best work produced in any part
of the world. Who ever thought that all the technical terms admitted
of satisfactory explanation and who ever imagined that a work on
grammatical terminology could be made as delightful as a novel ? In
this volume of nearly 350 pages the technical terms of Sanskrit
grammar from the Brahmanas, Nirukta, Prati$hakhyas, Panini.
Katyayana, Patanjali, down to Prayogaratnamala, and
Harinamamrta have been considered critically and historically in
all their bearings and apparently arbitrary technical terms like gha,
bha. lat, lit etc., have been traced to their sources. We have nothing
but admiration for this masterpiece of scholarship and warinly
reccommend it to our readers."

vb OR pas ee ryiitay eee:

ST iced, ot RE) te Zathe Saies’) |
oa
| ees eee
"y dye 7
; } eat e iy etPU ox
wT(asCOS nee
yan