Pallava Art

Pallava Art

Michael Lockwood
with
A. Vishnu Bhat
Gift Siromoney
P. Dayanandan

Tambaram Research Associates

 Tambaram Research Associates
MCC, Tambaram, Madras 600 059

Pallava Art, copyright 2001 by Michael Lockwood

Mahabalipuram Studies
First edition, copyright 1974 by Michael Lockwood
Mmallapuram and the Pallavas
First edition, copyright 1982 by Michael Lockwood

Typeset by T.R.A. on a Macintosh PB G3

Printed at Sudarsan Graphics, Madras 600 017
 Preface
Dr. Gift Siromoney introduced me to Mmallapuram (and, thus,
to the art of the Pallava dynasty) in the late 1960s. Over the years, he,
Dr. Vishnu Bhat, Dr. P. Dayanandan, and I visited the site many times.
All four of us were teaching at Madras Christian College, Tambaram.
Dr. Siromoney was in the department of mathematics, and later became
the chairman of the department of statistics. Sadly, he died, prematurely,
in 1988. Dr. Bhat has been teaching in the English department of M.C.C.,
and is presently its chairman. Dr. Dayanandan, who has been teaching
in the botany department, is now its chairman. My own teaching was in
the philosophy department. The four of us thus came to the study of
Pallava art from different disciplines.
By the word Art, I have intended to include not only the graphic
and plastic arts (painting and sculpture), but also literature, music, and
certain aspects of temple architecture, as well as the art of epigraphy.
This book comprises revised editions of two earlier works: Maha-
balipuram Studies (1974) and Mmallapuram and the Pallavas (1982).
These two books, which have been out of print for some years, con-
tained collections of studies originally written between 1970 and 1982.
In the decades following, we have carried out significant revisions and
corrections. I have, therefore, reorganized these studies, and have brought
them together, here, in one volume, adding twelve additional essays
some of which have never been published before.
Though I have been the principal author throughout, the follow-
ing ten chapters were co-authored by my colleague, Prof. Bhat:
3. Pallava Gagdhara
5. iva as Ligin in a Pallava Smskanda
9. The Philosophy of Mahndras Tiruchi Poem
10. iva-Gagdhara/Pallava-Kvrdhara
12. Dhvani in Epigraph and Stone
13. Queen Ragapatks Inscription
18. The Birudas of Mahndravarm
21. Kuumiymalai and Mmar Inscriptions
23. The Brhm Script and Phonetics
25. The Shore Temple Capital Inscription
Two chapters were co-authored by both of my colleagues, Profs.
Siromoney and Dayanandan:
1. Pallava Dvraplas and the Mahishamardin Cave-Temple
2. Pallava Smskanda
One chapter was co-authored by Prof. Siromoney:
6. Authorship of Mmallapuram Monuments
 Twelve chapters (4, 7, 8, 11, 14, 15, 16, 17, 19, 20, 22, and 26)
were authored by me, alone.
The 23rd essay in this book, The Brhm Script and Phonetics:
An Isometric Analysis of Vowels, may appear to be rather abstruse and
to have little to do with the art of the Pallavas. But the script of their
writing, engraved on their monuments and on their royal copper plate
grants, has evolved from the Brhm script and has created some of the
most beautiful calligraphy in the world. Some knowledge of the evolu-
tion of such beautiful writing may deepen our appreciation of it.
The 24th essay has been written and illustrated by Ms. Carmel
Berkson, and was first published by the Lalit Kal Akademi in its jour-
nal, Lalit Kal, Number 23 (1988). I consider her article, which com-
pares an Amazonmachy sarcophagus with the Mahishsuramardin panel
at Mmallapuram, to be a rare, groundbreaking investigation of Greco-
Roman influence on the art of the Pallavas.
The 25th essay, The Shore Temple Capital Inscription, deals
with a recently unearthed capstone of a presumed victory pillar. The
capstone has, engraved around its rim, four royal titles of the Pallava
king, Rjasiha. A study of these titles throws interesting light on scribal
practices of the early eighth century, A.D.
Chapter 26 presents additional facsimiles of Pallava inscriptions.
Based on estampages published by the Archaeological Survey of India
and on some of my own photographs of the original engravings, I have
created delineations of the inscriptions which, I hope, reveal their artis-
tic form more clearly than most of the estampages could.

Michael Lockwood
Milton, Mass., 2001
 CONTENTS
Preface .......................................................................................... v
Introduction .................................................................................. 1
1. Pallava Dvraplas and the Mahishamardin Cave-Temple .. 7
2. Pallava Smskanda .............................................................. 21
3. Pallava Gagdhara ............................................................... 47
4. God/King Images and Cult Worship ..................................... 53
5. iva as Ligin in a Pallava Smskanda ............................... 67
6. Authorship of Mmallapuram Monuments ........................... 73
7. Mmallapuram Chronology: Part I The Cave-Temples ...... 91
8. Mmallapuram Chronology: Part II The Rathas ................ 123
9. The Philosophy of Mahndras Tiruchi Poem ....................... 129
10. iva-Gagdhara/Pallava-Kvrdhara ................................. 137
11. A Mystery Dog in Sculpture .................................................. 143
12. Dhvani in Epigraph and Stone ............................................... 145
13. Queen Ragapatks Inscription .......................................... 149
14. Notes on Mmallapuram ....................................................... 155
15. Notes on Pallava Art .............................................................. 167
16. Royal Titles of Rjasiha and Mahmalla ............................ 173
17. Mahndras Paradoxical Birudas ........................................... 189
18. The Birudas of Mahndravarm ............................................ 193
19. Newly Discovered Monuments at Mmallapuram ................ 223
20. Vdhara Ardhanrvara ..................................................... 235
21. Kuumiymalai and Mmar Inscriptions ......................... 239
22. The Rape of the Liga ........................................................... 254
23. The Brhm Script and Phonetics .......................................... 255
24. Comparison of an Amazonmachy with the
Mahishamardin Panel by Carmel Berkson ................... 258
25. The Shore Temple Capital Inscription ................................... 263
26. Additional Facsimiles of Pallava Inscriptions ....................... 266
Bibliography .................................................................................. 293
 Key to Transliteration and
Pronunciation

Vowels
A a [ i ] u e A ?
(mica) (fill) (full) (prey) (go) (merrily)
Aa [- } eo ai AaO au
(father) (police) (rude) (aisle) (owl)

Anusvra = = nasal m or n
Visarga : = = voiceless aspiration

Consonants
Voiceless Voiced
Unaspirated Aspirated Unaspirated Aspirated Nasal
k\ k K\ kh ga\ g Ga\ gh =a\
ca\ c C\ ch ja\ j Ja\ jh Ha\
T\ z\ h D\ Z\ h Na\
%a\ t qa\ th d\ d Qa\ dh na\ n
p\ p f\ ph ba\ b Ba\ bh ma\ m

Semi-vowels Sibilants & Voiced h:

ya\ y Ya\
r\ r Xa\
L\ sa\ s
la\ l h\ h
va\ v
 Introduction

On the coast, almost sixty kilometers south of the city of

Madras, at a place called Mmallapuram, there are some of the most
famous ancient monuments in India. They are appealing to the casual
visitor. And to the student of South Indian art and architecture they
are of fundamental importance.
In the seventh and eighth centuries A.D. and, perhaps, even
earlier, kings of the Pallava dynasty created cave-temples, monolithic
shrines, structural stone temples, and expansive relief panels carved
on the open rock-face of the hillsides. These monuments are impor-
tant because they were among the first major artistic monuments to
be fashioned out of hard rock in South India. More than one thousand
two hundred years have come and gone, and yet these works are still
fresh before our eyes. The structures of all other temples of those
days and earlier times have long ago vanished because they were
made of relatively perishable material.
Although these monuments and their figures are all carved out
of stone, every inch would have been covered by the artisans with a
thin layer of fine, white plaster and then painted so as to simulate the
materials and color of ordinary temples. All of the human and animal
figures would have been painted so as to impart a startling realism to
them. The paint, of course, has disappeared except for traces.1
Mmallapuram has more than 14 cave-temples, 9 monolithic
shrines, 3 structural stone temples, and 4 relief-sculptured rock panels,
all of which were created by the Pallavas in those early centuries.
The structural temples imitate, in the hard medium of stone
blocks, the traditional temples which were built with brick, mortar,
and wood. Each of the monolithic shrines is a whole temple carved
out of a single mass of rock. They are sculptured replicas in the
round, so to say.
In their cave-temples, the Pallavas have reproduced the interior
aspect of shrines along with their porch-like pillared maapas by
scooping and carving into the solid rock of the hillsides. Since the
frontal maapa with its pillars is visually the most prominent feature
of the cave-temple, these temples are often called simply maapas.
The most unusual and impressive sight at Mmallapuram must
surely be the so-called Penance panel. Popularly, it is believed to be
an artistic representation of Arjunas penance. However, certain
scholars have persuasively shown it to represent Bhagrathas penance
and the descent of the river Gag.2 In this huge open air relief-
carving with its multitude of figures (animal,3 human and divine), the
Pallava artists have used for their canvas the sheer rock which rises
perpendicularly on one side of the hill.
 2 The story of Bhagrathas penance is given, among other
Pallava Art places, in the epic of the Rmyaa. Bhagratha wished to sanctify
the ashes of his ancestors with the holy water of the Gag. This
divine river was at that time confined to the heavenly realm. In order
to bring her down to earth, Bhagratha practiced severe penance.
Brahm finally agreed to grant his request, but warned Bhagratha
that in the mighty rush of her descent, the Gag would devastate the
earth. Therefore, Bhagratha continued his penance in order to win
ivas protection against her terrible onslaught. For a whole year
Bhagratha remained standing on one foot with his arms upraised, his
body becoming emaciated. iva, pleased by Bhagrathas austerities,
appeared and granted him his boon. It is this moment which is
portrayed in the Mmallapuram panel, to the upper left of the central
cleft in the rock which divides it into two sections.
The water which the Pallava engineers planned to have cascade
down the cleft into a pool below would represent the Gag reaching
the earth.4 All the figures, human and divine, are thus shown gravitat-
ing towards the central cleft to behold this glorious miracle of the
Gags descent.
Between the point in the Rmyaas account where iva
grants the boon to Bhagratha and the part in which the Gag reaches
the earth, there is the scene of ivas carrying out Bhagrathas request
an act which is not depicted in this panel, though it was a favorite of
the Pallavas and appears twice elsewhere in Mmallapuram. It is the
Gagdhara theme in which iva controls the fury of the descending
Gag by holding her captive in the locks of his hair until she flows
gently to earth. The oldest Pallava representation of the Gagdhara
theme (even pre-dating by one generation the Mmallapuram Penance
Panel) is the Gagdhara panel in a cave-temple in Tiruchirapalli.
The significance of the Tiruchi panel in relation to the art of the
Pallavas at Mmallapuram will be discussed in the third, ninth, and
tenth studies in this book.
The Mahishamardin Cave is one of the most remarkable of the
cave-temples at Mmallapuram. It takes its name from the Mahisha-
mardin panel carved on the right wall of its maapa. On the wall
opposite there is a panel cut in deep relief, depicting Vishu in trance-
like sleep, reclining on the great serpent, Ssha. These panels
represent two scenes described in the Dv-Mhtmya, an episode in
the Mrkaya Pura. Particularly effective is the striking contrast
achieved by the artists between the calm potency of the Reclining
Vishu panel and the vigorous action in the other panel which depicts
Durg waging her victorious battle against the buffalo demon,
Mahisha.
There are three cells or sanctums cut into the rear wall of the
maapa of this cave-temple. At the back of the central sanctum there
is a large carved panel representing iva together with his consort
Um, and their little son Skanda. All three are shown seated together 3
on a royal throne. This image is called Smskanda. The very Introduction
earliest Smskanda panel was a creation of the Pallava king,
Paramvara-I, in the latter half of the seventh century. In the first
study of this book, we discuss the Smskanda panel of the Mahisha-
mardin Cave and try to show that it is an addition which was ex-
ecuted at a date distinctly later than that of the Vishu and Mahisha-
mardin panels of this same cave-temple.
The Shore Temple is the most important structural temple at
Mmallapuram. Built by the Pallava king, Rjasiha, in the early
eighth century, it is picturesquely situated on the edge of a promon-
tory jutting into the ocean. There are actually three separate shrines
which form the Shore Temple complex. The eastern and western
shrines which have high towers are dedicated to the god iva. In
between them is one dedicated to Vishu. On the back, inner walls
of the two aivite shrines there are Smskanda panels.
Our second study is devoted to an analysis of the stylistic
development of the Smskanda panel during the successive reigns
of several Pallava kings. In their extant art, it is by far the most often
repeated image. More than 40 Pallava Smskanda panels remain to
this day, providing thus an important key to the problems of the
chronology of Pallava monuments. The fourth and fifth studies also
investigate various aspects of the Pallava Smskanda.
The finest examples of monolithic shrines at Mmallapuram
are found in the group popularly called the Five Rathas. The word
ratha, which means chariot or vehicle, has been imaginatively
applied to these temples. Of these five, the so-called Draupad Ratha
is actually a small shrine for the goddess Durg. Her image is carved
in relief on the back wall of the sanctum. Two devotees are shown
kneeling at her feet. One of them is in the act of making the supreme
sacrifice of cutting off his own head! That this practice actually
existed in Tamilnadu is revealed elsewhere both by inscription and
in literature.
The great importance of the five shrines to the study of the
development of temple architecture in South India lies in the fact that
each one of them has a different form. The Draupad Ratha is the
simplest. The Dharmarja Ratha is the largest and most elaborate. It
is pyramid-like in form, with three stories. On the top level there is a
small sanctum scooped out of the solid rock. On the back wall of this
cell is carved the oldest extant Smskanda panel. There is an
inscription outside claiming that this is the vara (iva) shrine of the
Pallava king called Atyantakma. There are many other inscriptions
on the walls of this temple.
The sixth study in this book considers the evidence provided
by these inscriptions, plus evidence from several other sources, in an
attempt to throw light on the problem of the authorship of the monu-
ments of Mmallapuram.
 4 In dealing with various problems of the history and art of the
Pallava Art Pallavas, our studies have generally emphasized the importance of
stylistic analyses of the dress and ornaments depicted in the sculpted
figures. There are, however, several studies in the book which deal
with the inscriptions of the Pallavas: the Tiruchi poem of King
Mahndra (9th & 10th), Queen Ragapatks Inscription (13th), the
inscribed royal titles of Rjasiha and Mahmalla (16th) and of
Mahndra (17th & 18th). Among the remaining assorted subjects, the
recently discovered monuments at Mmallapuram are covered in two
studies of the book (19th & 20th).
____________________
1In a letter to the editor of The Hindu which appeared in the
issue dated January 18, 1970, Gift Siromoney, P. Dayanandan, and I
made the following observations about the painting of Mmallapuram
(only a part of the letter is quoted here):
A group of small school children found it most amusing that we
three adults should be craning our necks and peering so intently at
the upper reaches of the Rathas. And we were quite ready to
smile back at them because, on the basis of a little detective work,
we were enjoying in our minds eye a view of the monuments of
Mmallapuram which they did not see. Imagine the Rathas
completely covered outside and in with bright colors of paint.
Imagine the many graceful figures which people the niches of
these temples rendered in life-like color, their bright jewels and
gold ornaments glittering, the stone pillars which they lean against
(pillars imitating structurally the earlier style of wood) painted in
an imitating maroon. Imagine further the great panel of Arjunas
Penance alive with color! I say imagine because, as any visitor to
Mmallapuram knows, we see everywhere only the uniform grey-
brown hue of the carved granite rock. Everywhere that is, unless
you look as intently as we three were doing to perceive the unmis-
takable traces of plaster and paint which have survived perhaps
more than a thousand years of weathering. . . . On the Arjuna
Panel, traces of plaster and paint can be seen easily (especially
with binoculars) under the upraised and joined hands of the ascetic
practicing austerities. And there are many other places on the
Panel where plaster and paint are quite evident.
2First advanced by V. Goloubew in 1914, this view has been
ably supported later by G. Jouveau-Dubreuil and the archological
evidence noted by A.H. Longhurst. The point which is absolutely
fatal to the Arjunas Penance interpretation is the fact that some of
the heavenly beings depicted in the panel actually have their backs to
iva as he grants the boon to the ascetic who is supposedly Arjuna.
The problem vanishes if it is the descent of the Gag which is the
center of attention (the boon granted to Bhagratha).
 3Some 150 animals representing 16 different species. 5
Introduction
4Longhurst describes the discovery of the stone-lined pool at
the foot of the Penance Panel (Pallava Architecture, Part II). This pool
was very likely a royal bath at the time of the Pallavas. There is also
archological evidence of a storage tank for water on top of the hill
just above the central cleft. Thus, at special times of celebration, water
could be let out of this tank by the Pallavas so as to produce an
artificial waterfall down the central cleft and thus simulate the Gag
descending to earth. It would have been quite a spectacle even by our
modern standards!
 ONE

Pallava Dvraplas and the

Mahishamardin Cave-Temple1

The Tamil word for temple (kyil) can also mean palace.
Usually, the temples of the gods are shown with guardians posted at
the entrance to the sanctum. This only imitates the practice of the king
in his palace, with guards protecting the royal chamber.
Our study would establish for the first time the fact that the
carved guardians or door-keepers in many Pallava temples are really
anthropomorphic representations of weapons or emblems peculiar to
the god enshrined within. In Sanskrit such weapon-men are called
yudhapurushas. In Pallava temples, the particular weapon or other
emblem which a guardian represents is usually shown on his head-
dress.
The second part of this study is devoted to a comparative
analysis of carved panels. The purpose of such an analysis is to help
discover the chronological development in Pallava art. It is concerned
with details of the dress and ornaments depicted on figures of people
and gods. Fortunately for our study, Mmallapuram has an impres-
sive population of stone figures. Gods and goddesses are represented
in idealized human form. Also shown are many of the lesser divinities.
Most of these are also depicted in human form, though some are part
animal. Of great interest are the numerous ordinary humans who have
been sculpted. Common people are seen tending cattle and carrying
children. Hunters are shown in their forest habitat. There are ascetics
and holy men with beards. Even kings and queens have been por-
trayed.
It is important to note that in the sculpture of this period, very
little difference is seen between the dress and ornaments of divine
beings and those of humans. The one really distinguishing feature of
the gods is the addition of extra arms, with their identifying emblems.
The lesser divinities have only two arms, but they can be distinguished
easily when they are shown flying through the upper regions a feat
not possible for ordinary mortals! Other semi-divine beings are half
human and half animal. The upper half is usually human. The lower
half may be of a bird, or snake, or some such creature.
 8 Our analysis of stylistic development together with the signi-
Pallava Art ficance of the discovery that Pallava temple guardians are yudha-
purushas has been used by us in this study to show that the Mahisha-
mardin Cave has had an erratic history of development (it still
remains unfinished) and that there are reasonable grounds to
suppose that what was originally planned as a Vishu sanctum was
transformed into a aivite one during the reign of Paramvara-I, a
Pallava ruler in the latter part of the seventh century.
Scholars have long been aware of the fact that there was a
period in Mmallapurams history (some would say, the 13th century)
when Vaishavite sectarians took possession of aivite temples there.
This take-over has been signified by their engraving the emblems of
Vishu (the discus and the conch shell) on the walls of these appropri-
ated temples. Our study, however, would for the first time show that
much earlier there was an appropriation of a Vishu sanctum by
aivites, undoubtedly on the direct order of King Paramvara,
himself.
The curious horns on some of the dvraplas (door guardians) in
early Pallava temples gave us the clue to the surprising conclusion that
in the Mahishamardin Cave-Temple at Mmallapuram, the main sanc-
tum was originally planned for Vishu, not for the Smskanda panel
which we see today.
In regard to the horns on the dvraplas, there are several
conflicting views among scholars as to their significance. One view
would have it that they are a kind of mutation of the early Buddhist
motif of Ngarja as dvrapla. That is, the multi-headed snake-hood
of Ngarja develops into two horns.2 A second theory is that the
horns of the dvraplas can be explained with reference to the practice
of wearing horns by such tribes as Ngas and the Gonds. Still a third
explanation is that the horned dvraplas represent a humanized form
of the bull, Nandi.3
While photographing a dvrapla in the upper cave-temple at
Vallam (two miles east of Chingleput town), we were struck by the
similarity between the horns of this dvrapla and the outer prongs of
the trident or trila as represented in Pallava sculpture elsewhere.
These horns and the outer prongs of the trila have the same peculiar
compound curve at their base. Further, the so-called horns in the
6.2 Vallam example are not shown attached to the head or head-dress in a
very realistic manner. We concluded, therefore, that the horns of the
dvrapla along with his elongated makua (as the central prong) did,
in fact, represent the trila, an emblem of iva.
At Vallam, only the dvrapla on the proper right of the
entrance has horns. However, we soon discovered that, although the
dvrapla on the left did not have horns, he did have an axe-blade
projecting edge-forward from the front of his head-dress. The axe is
another aivite emblem. The trila horns and the axe-blade, then,
Trident can be clearly recognized as aivite symbols which, along with certain
 9
13

19

Horned Dvrapla, Vallam Cave-Temple

 10 other characteristics such as the snake-entwined club, go to indicate
Pallava Art quite unambiguously that these dvraplas are guarding a aivite
shrine.
Other examples of dvraplas with horns and axe-blades on
their head-dress are to be found in the Kailsantha temple at Kanchi-
puram, at the Atiraachavara cave-temple at Saluvankuppam, and
at various shrines at Mmallapuram. In most of these cases, a knowl-
edge of the significance of the trila horns or the presence of the
axe-blade is not necessary for an identification of the shrines as aivite
because within the shrines there is a liga. However, consider the
shrine on the western side of the second level of the Dharmarja Ratha
at Mmallapuram. This sanctum is empty and unfinished, and there is
nothing inside it now that would indicate which god it was fashioned
4.5 for. Therefore, it is the horned guardian to the proper right of this
shrine which reveals it was intended as aivite.
The practice of showing the emblems of the deity on his guard-
ians head-dress is applied by the Pallavas to Vaishavite shrines as
well as aivite. A clear example of this is found in the Varha-II cave-
temple at Mmallapuram: the dvrapla immediately to the right
6.2 (proper) of the sanctums entrance has a discus represented edge-
forward at the very top of his head-dress. The dvrapla to the left has
a conch placed at the top of his head-dress. The discus and conch are
Vishus insignia. That this Varha cave-temple is a Vaishavite
temple is undisputed, and we find here the Varha, Trivikrama, and
Gajalakshm panels which are all Vaishavite themes. But the discus
and conch emblems on the head-dress of the dvraplas give additional
confirmation that the (now empty) sanctum was for Vishu.
Conch Shell Personified Another important example of Vaishavite emblems on the
Varha-II Cave-Temple head-dress of dvraplas is to be found in the divarha cave-temple at
Mmallapuram. Here the Varha figure in the central shrine is under
worship. The modern walls which enclose the front of this shrine hide
4.5 parts of the dvraplas. However, one is still able to see the discus at
the top of the head-dress of the right dvrapla and the conch similarly
placed on the left dvrapla.
We must also mention that the guardians of King Mahndras
Vishu cave-temple at Mahendravadi also have the discus and conch
on their head-dress.
6.2 In the case of the goddess Durg, the dvrapliks (female
guards) in her shrines at Mmallapuram are shown with a sword in
hand (guard to the proper right) and with a bow (left guard). There are
two Durg shrines at Mmallapuram: the Draupad Ratha and Kikal
Maapa. The two young fighting women accompanying the goddess
in the Durg panel of the divarha cave-temple are similarly armed
and provide an analogous example, though, strictly speaking, they are
not guarding a door here.
Dvrapliks Our main conclusion so far, then, is that dvraplas are often
Draupad Ratha
shown with emblems or weapons which are characteristic of the deity
 11
13

19

Dvrapla with axe-blade on head-dress, Vallam Cave-Temple

 12 they guard. They are, in effect, yudhapurushas. In the case of many
Pallava Art aivite shrines, one dvrapla has horns and the other an axe-head
shown on the head-dress, and both may have clubs with snakes encir-
cling them. In the case of Vaishavite shrines, we find the following
arrangement: one dvrapla has a discus represented on his head-dress,
and the other, a conch.
With these facts in mind, let us turn to the famous Mahisha-
mardin Cave-Temple at Mmallapuram. There are three sanctums
in this cave-temple, and one naturally thinks of the many Pallava cave-
temples created for the Hindu Trinity. The central sanctum of this cave
is given special prominence by having before it a raised porch with two
lion pillars in front. But considering first the right (southern) sanctum,
one finds that the dvrapla to its proper right has horns. The
dvrapla to the left has a single axe-blade projecting edge-forward
above his forehead. The right dvrapla has a club with a snake around
it. We conclude from these facts that the right sanctum is clearly for
iva.
Considering next the left (northern) sanctum, one does not find
any of the above aivite emblems. Further, both the dvraplas wear
the long dress and the uttarya (upper cloth) which are uncharacteristic
of aivite dvraplas. We conclude that the left sanctum of the
Mahishamardin Cave-Temple is distinctly non-aivite.
With a clearly aivite sanctum to the right, with a distinctly non-
aivite sanctum (undoubtedly for Brahm) to the left, and, further, with
a large panel on the porchs right wall depicting Vishu reclining, one
would naturally expect the main, central sanctum to be for Vishu. But
surprisingly, one finds instead a large Smskanda panel on the back
wall of this main sanctum.
This led us to examine with care the dvraplas of the central
sanctum. At first glance, both dvraplas seem to be aivite: they both
have clubs the club of the proper right dvrapla being encircled by a
three-headed snake. The dvrapla to the right has horns (in light
relief), and the dvrapla to the left has a triple-bladed axe-head repre-
sented on the head-dress above his forehead.
But there are several puzzling aspects about the way in which
these two dvraplas have been sculpted. In fact, it looks as though
these niches may have been originally intended for dvraplas without
clubs the kind of dvraplas one would expect to be guarding a shrine
for Vishu. The reason we say this is that the clubs seem like an
afterthought. The clubs are carved where the pilasters should be, and
completely break the orderly boundary of the rectangular niches. It
would be interesting to know whether there is a single other example
in Pallava sculpture of such an extreme disregard of the rectangular
boundaries of the niche.
It is possible that work had begun on these niches at a time
when the main sanctum was intended for Vishu. At that time, the
 13

13

19

Mahishamardin Cave-Temple
14

14.5

Porch within a porch Mahishamardin Cave-Temple

boundaries of the niches and the general pose of the dvraplas were 15
established. For one reason or another, the work was not completed. Pallava Dvraplas
At a later date, when aivism was in the ascendancy, the details of the
dvraplas were finished as aivite, including the horns in very
shallow relief on one guard and an axe-head on the others head-dress.
The clubs had to be added in a most unusual place: where the pilasters
normally would come. To accomplish this addition of the clubs, the
rock area for the pilasters and all the rest of the architectural ornament-
ation of the main sanctums faade had to be removed. This refacing
of the rock has left only a plain surface around the niches for us to see
today.4
This evidence of re-working led us to note, first, the obvious
fact that the Smskanda panel of the central sanctum is different
stylistically from the other two panels (of Vishu and Durg) in this
temple; and, secondly, that there is a striking similarity between this
Smskanda panel and like panels found in the eighth century Kailsa-
ntha temple at Kanchipuram.
We, therefore, began to feel certain that the Smskanda panel
in this cave-temple was a later addition, transforming what was origin-
ally planned as a Vaishavite main shrine into a aivite shrine.
Speaking generally of Mmallapuram, one can observe a
marked difference in style, as shown in the dress and ornaments of the
sculptured figures. Just as fashion changes today, so it must have
changed in the time of the Pallavas. This change is reflected in their
sculptural art and thus provides us with a means of dating the monu-
ments.
As we have noted, even in one and the same cave-temple one
finds distinctly different styles. To help us date the panels of the
Mahishamardin cave-temple, we examine them in detail with regard
to the style of dress and ornaments of the figures portrayed. As a basis
for our argument, we mention certain general observations we have
made about the dress and ornaments of Pallava sculpted figures.5
(i) Early Pallava Characteristics
In early Pallava sculpture (roughly, around the period of the
Great Penance Panel and the Five Rathas, which are usually ascribed
to King Narasiha-I in the seventh century A.D.), men do not wear
any leg ornaments and are shown with only one diagonal band (sacred
thread, etc.6) across the body. In the early period, women do not wear
any diagonal band and have only single anklets on each leg.
(ii) Later Pallava Characteristics
In the later Pallava sculpture (eighth century, around the time
the Kailsantha and Shore temples were built), we notice that men
now sometimes have leg ornaments and often have more than one
diagonal band. Leg ornaments as a common feature for men appear
to have been introduced gradually for the first time in Indian art by the
 16 Pallava sculptors of the early 8th century. In the whole sweep of art
Pallava Art history from Bharhut in the centuries B.C., through Amaravati and the
earlier phases of Ajanta up till the end of the 7th century A.D., men do
not wear leg ornaments. The very few exceptions to this claim will
certainly prove the general rule.
In the later period of Pallava sculpture, women are seen wearing
the diagonal band; they frequently have multiple ornaments on each
leg; shoulder straps for the breast-band are introduced; and the head-
dress which looks like a turban around the base of a crown develops
two distinctive characteristics: the turban-like portion is slightly
pinched (indented) in the front, and the crown-like portion is unusually
tall. (This is actually only an arrangement of tying up the hair and is
neither a turban nor a solid crown.)
It is on the basis of these general observations that we have
analyzed the panels of the Mahishamardin cave-temple and have
concluded that the Smskanda panel was done at a distinctly later
time than the other two panels in this cave.
(iii) The Smskanda Panel
To establish that the Smskanda panel of the cave-temple has
the characteristics of the later (8th century, Kailsantha) period, we
mention some of the close similarities between the figures of the
Smskanda panel of this cave-temple and the figures of like panels
in the Kailsantha temple in particular, the Smskandas of the two
sub-shrines centrally located on the northern and southern sides of the
main sanctum of the Kailsantha temple. In both the Mahishamardin
cave-temple Smskanda and the Kailsantha examples, one finds
these characteristics of the later period: Um has a diagonal band,
multiple anklets, and the characteristic late-period head-dress. iva has
multiple diagonal bands.
Next, to show that the Smskanda panel of the Mahisha-
mardin cave-temple is quite different stylistically from the early
Smskanda panel of the Dharmarja Ratha, it should be noted that
the following characteristics of the later period, all of which are found
in the cave-temple panel, are absent in the Ratha panel: Ums
characteristic late-period head-dress, her diagonal band and multiple
anklets, and ivas multiple diagonal bands. In addition, Ums profile
pose in the Ratha panel is absolutely unique; whereas, in the cave-
temple panel, she strikes the oft-repeated pose found at the Kailsa-
ntha, Shore temples, etc. Further, in regard to the small Vishu figure
appearing in the cave-temples Smskanda panel (above and behind
ivas throne), Vishus discus and conch are depicted with flames
(generally accepted as a later characteristic); whereas the discus and
conch have no flames in the Rathas depiction of Vishu in an adjoin-
ing side panel to the Smskanda proper.
 17
This photograph, courtesy of the Archaeological Survey of India:

7 6

7 7

Smskanda, Mahishamardin Cave-Temple Smskanda, Kailsantha Temple

14.5

Smskanda, Dharmarja Ratha

18 10.5

Reclining Vishu, Mahishamardin Cave-Temple

7 7

5 5

Reclining Vishu, Kailsantha Temple Mahishamardin Panel, Sluvakuppam

Mahishamardin Panel, Mahishamardin Cave-Temple

 Thus, the Smskanda panel of the Mahishamardin cave-temple 19
has much in common, stylistically, with Smskanda panels of the later, Pallava Dvraplas
Kailsantha period; and it is significantly different from the earlier
Smskanda panel of the Dharmarja Ratha. It would seem, therefore,
that the Smskanda panel of the cave-temple was executed much
closer to the period in which the Kailsantha temple was built than
were the other panels.
Finally, it must be shown that the other two panels of the
Mahishamardin cave-temple (the Reclining Vishu and the Mahisha-
mardin panels) were done during an earlier period in the mid-7th
century.
(iv) The Reclining Vishu Panel
Considering first the Reclining Vishu panel in this cave-
temple, one finds these early characteristics: no man wears more than
one diagonal band, and none has any leg ornament; the women have no
diagonal bands, only single anklets, no characteristic late-period head-
dress, and the breast-band is depicted without shoulder straps.
On the other hand, the Reclining Vishu panel of this cave-
temple (as an early example) contrasts with the little-known, and much
smaller Reclining Vishu panel of the Kailsantha temple (as a later
example). This latter panel is found directly above the entrance to the
Smskanda sub-shrine centrally located on the northern side of the
main sanctum. The patchy coating of plaster on this panel makes any
job of detailed study risky guesswork. However, mention may be made
of the following later characteristics of it which are free of plaster
covering: the woman (Bh-dv) kneeling at Vishus feet wears
shoulder straps on her breast band and she has the characteristic late-
period head-dress; and the five heads of the great serpent on which
Vishu reclines are ornately carved as horned-yi-type heads (which
contrasts with the more naturalistic treatment of these heads in the
cave-temple panel).
(v) The Mahishamardin Panel
Considering, finally, the Mahishamardin panel of the cave-
temple, it contrasts (as an early work) with the Saluvankuppam and
Kailsantha Mahishamardin panels (as later works): in the cave-
temple panel, there are these early characteristics: Durg has no
distinctive late-period head-dress, no diagonal band, no shoulder straps
on her breast-band, and only single anklets; whereas, in the Kailsa-
ntha and Saluvankuppam panels, one finds the later characteristics.
Again, in the cave-temple panel, the buffalo demon has only one
diagonal band and no leg ornaments, whereas in the Saluvankuppam
panel he wears two diagonal bands and has prominent anklets.
(vi) Summary
Let us summarize our stylistic analysis. The Smskanda
panel of the Mahishamardin cave-temple is a relatively later Pallava
 20 work as it compares with similar panels of the 8th century Kailsantha
Pallava Art period, and contrasts with the 7th century Smskanda panel of the
Dharmarja Ratha. The other two panels of the cave-temple are earlier,
7th century works as they have the early characteristics, and contrast
with panels of the same themes created in the Kailsantha period.
The conclusion that the Smskanda panel of the Mahisha-
mardin cave-temple is a decidedly later work than the other two panels
of the same cave strengthens the claim we have made earlier (on the
basis of an examination of the cave-temples dvraplas) that there are
reasonable grounds to suppose that the main, central shrine was origin-
ally planned for Vishu.
____________________

1This first study is based on Pallava Dvraplakas and the

Mahishsuramardin Cave at Mahbalipuram, by Michael Lockwood
and Gift Siromoney, a paper read at a meeting of the Archological
Society of South India, April 4, 1970, and on its modified version
which appeared in The Sunday Standard, Madras, in two parts: Guard-
ians of Pallava cave temples (February 14, 1971) and Changing
fashions in Pallava art (February 28, 1971).
2P.R. Srinivasan, Beginnings of the Traditions of South Indian
Temple Architecture, Bulletin of the Madras Government Museum,
New Series General Section, Vol. VII, No. 4, 1959, p. 34.
3K.R. Srinivasan, Cave-Temples of the Pallavas, Architectural
Survey of Temples Series, No. 1 (New Delhi: Archological Survey of
India, 1964), p. 36.
4Surprisingly, the faades of the other two sanctums seem to
have been re-faced in a similar way. In doing this job of recessing the
walls, the feet of the dvraplas of the left sanctum have been sheared
off. In the case of the right dvrapla of the right sanctum, his right
foot remains projecting out beyond the walls surface in a most unusual
manner. While re-facing the wall, a portion of the rock was left under-
neath this foot to give some sort of support to it.
We must mention, in passing, two other puzzling aspects:
(1) the dvraplas of the main, central sanctum are noticeably smaller
than the dvraplas of the other two sanctums; (2) the entrances of the
two side shrines are in poor alignment with the stairways provided for
them.
5Some of these observations have been discussed in Maha-
balipuram: Costumes and Jewellery, by Gift Siromoney, M.C.C.
Magazine, 1970.
6As there is much confusion in the application to early sculp-
ture of the term sacred thread, we have deliberately coined the more
general term diagonal band which we intend to include the sacred
thread as well as other similarly worn items.
 TWO

Pallava Smskanda1
The Smskanda images of the Pallavas are carved stone
panels which portray iva and his consort Um, seated together on a
royal throne with their little son, Skanda, between them. Of all the
Pallava images which have survived to the present, the Smskanda
panels are by far the most numerous. There are more than forty of
them. They offer an extremely important key to the solution of several
thorny problems in the history of the development of Pallava art.
The Smskanda image was most probably the creation of the
Pallava king Paramvara-I. However, there are only four extant
Smskanda panels (plus one which has been effaced) which can be
attributed to his reign. Fortunately, almost forty Smskanda panels
survive from the period of his son, King Rjasiha.
The Smskanda image continued to be popular with later
Pallava kings. For instance, there is a fine example at Kanchipuram in
the sanctum of the Muktvara temple which was built around the 28th
regnal year of the Pallava king Nandivarm-II (during the latter half
of the eighth century). The Smskanda was also very common in the
Chola period, especially in the medium of bronze casting. Its popular-
ity with South Indian artists continued into the modern period.
We give below a list of the Pallava temples which have the
Smskanda panel on the inner back wall of their sanctum:

Pre-Rjasiha Style
Mmallapuram:
1. Dharmarja Ratha (3rd level shrine)
2. Rmnuja Maapa (main shrine)

Rjasiha Style
Mmallapuram:
3. Kshatriyasihvara
4. Rjasihvara (3 and 4 belong to the Shore Temple)
5. Mahishamardin Cave-Temple (main sanctum)
6. Mukundanayanr

Saluvankuppam:
7. Atiraachavara (main + 2)

Tirukkalukkunram:
8. Vdagirvara (main + 1)
 22 Kanchipuram:
Pallava Art 9. Mahndravarmvara Kailsantha (1 + 28)
10. Piravtanvara
11. Iravtanvara
12. Amarvara (also called Tripurntakvara)
13. Airvatvara
14. Muktvara
15. Mtagvara
Panamalai:
16. Tlagirvara

The Smskanda theme originated in a period when the Pallava

kings of the seventh and eighth centuries A.D. made a distinct effort to
integrate the worship of iva with the Dv cult and the Murugan cult.
In the Smskanda panels carved in relief on stone, and in later Sm-
skanda bronzes, these three deities are shown as a family group. iva
and Um are portrayed sitting on a throne with their son, Murugan, in
the form of the young child, Skanda, between them. The term Sm-
skanda (sa-Um-Skanda), translated into English, literally means,
with Um and Skanda.
Ordinarily, in aivite temples, where the main object of
worship is the liga, no anthropomorphic form of the deity, either
in painting or in carving, appears in the sanctum. However, in the
Pallava period the custom was different. The carved Smskanda
panel is commonly found on the back inner wall of the sanctums of
their aivite temples. This practice was not continued by later dynas-
ties. So, as a rule of thumb, we can say that if a aivite temple has a
sculptured panel in its sanctum, almost certainly it is a temple of the
Pallava period.
In our first study, we have, on the basis of an analysis of the
dress and ornaments of sculpted figures, established two distinct
styles for the Pallava Smskanda panels. The earlier style we call
pre-Rjasiha and the later style, Rjasiha (after the eighth
century Pallava ruler whose identified temples have a total of around
forty Smskanda panels in them).
We know of only two examples of the pre-Rjasiha style
Smskanda. One of them is found in the third-level sanctum of the
Dharmarja Ratha at Mmallapuram. The other, which is on the back
wall of the central cell of the Rmnuja Maapa cave-temple of the
same place, has been destroyed. Only a rough outline of the figures
remains.
Dharmarja Ratha, Mmallapuram
The Smskanda panel in the third-level shrine of the
Dharmarja Ratha, therefore, is unique in that it is the only well-
preserved Smskanda which is of a distinctly pre-Rjasiha style.
It is, thus, the earliest extant Smskanda.
 There are some interesting details of the Rathas Smskanda. 23
In this panel, iva, as indicated by his attitude, is imparting words of Pallava Smskanda
wisdom, and Um is bending the tip of her right ear with her fore-
finger so as to catch every word. There is a figure of a bird which is
carved in light relief immediately above ivas upper left hand. This is
most probably the cock standard of Skanda, but the details are indis-
tinct.
In our first study, we have noted those characteristics of the dress and ornaments which distinguish
the Rjasiha-style from the pre-Rjasiha style in Pallava works of art. Such an analysis of dress and
ornaments, we argued there, shows that the Rathas Smskanda belongs to the pre-Rjasiha period.
With regard to our present comparison between the pre-Rjasiha style Smskanda (Dharmarja Ratha)
and any of the numerous Rjasiha-style Smskandas, we note here the following points of contrast:

Pre-Rjasiha Style Rjasiha Style

Smskanda Panel (Shore Temple and
(Dharmarja Ratha) 40+ other examples)
1. Um is seated in profile. 1. Her torso is always turned front.
2. Ums back abuts niches edge. 2. Because of her frontal posture, her
back never abuts niches edge.
3. Ums left hand is in front clasping 3. Her left arm is always on her left
Skandas waist. side supporting her body.
4. ivas lower left hand rests 4. His lower left hand always rests on
clenched on his left knee. his right ankle in dhyna mudr.
5. ivas right leg only is down. 5. Always only his left leg is down.
6. ivas lower right forearm is held 6. His lower right forearm is held
vertically close to his chest horizontally away from his body.
(with hand in chin mudr).
7. Two gaas with fly-whisks hover 7. Never any hovering gaas above.
above iva and Um in corners They are replaced by Brahm and
of the panel. Brahm & Vishu Vishu standing directly behind
stand on either side in adjoining the throne, just above ivas
niches. upper arms.

Pre-Rjasiha Style Smskanda, Rjasiha Style Smskanda,

Dharmarja Ratha Shore Temple
 24 Rmnuja Maapa, Mmallapuram
Pallava Art
As we have said, the Rathas panel is the earliest preserved
Smskanda. We would maintain, however, that the smashed
Smskanda panel of the Rmnuja Maapa, Mmallapuram, is also
pre-Rjasiha style. Another table of characteristics will show why
we take the Rmnuja panel to be pre-Rjasiha:

Pre-Rjasiha Style Rjasiha Style

Dharmarja Ratha Rmnuja Maapa Shore Temple & 40+ others
1. Um in profile. 1. Also in profile. 1. Never in profile.
2. Ums back abuts niche. 2. Also abuts niche. 2. Never abuts niche.
3. ivas lower left hand rest on 3. His lower left hand also on thigh 3. His lower left hand always in
his left thigh. (certainly not dhyna mudr). dhyna mudr.
4. Two gaas hover above iva 4. Also two hovering gaas and no 4. Never any hovering gaas; instead,
and Um; no Brahm and Brahm and Vishu. Brahm and Vishu are behind
Vishu in the panel. ivas throne.

The Rmnuja Maapas Smskanda relief has been chiseled

and leveled off. However, the outline of figures remains, and the
outline is enough to allow one to deduce the characteristics which are
listed above.
It should be added that the details which are discernible in the
smashed Durg panel of the Rmnuja Maapa are similar to those
of the Durg panel in the divarha cave-temple of Mmallapuram.
These observations, taken together with an acknowledgment of the
early architectural characteristics of this cave-temple, all go to support
a pre-Rjasiha date.

Five more Temples, Mmallapuram

Mmallapuram has five more temples whose Smskanda
panels are in the Rjasiha-style. They are:
1. the Kshatriyasihvara,
2. Rjasihvara,
3. Mahishamardin cave-temple,
4. Mukundanayanr stone-structure temple, and the
5. Atiraachavara cave-temple.
We include the Atiraachavara cave-temple in the list since it is
only a short distance away from the town of Mmallapuram.
The Shore Temple actually has two aivite temples, each of
which has a Smskanda in its sanctum. The Atiraachavara cave-
temple has, in addition to the Smskanda in its sanctum, two other
Smskandas carved on the rear wall of its maapa.
The Mahishamardin Cave and the Mukundanayanr structural
temple each has a Smskanda in its sanctum sanctorum.
 25
14.5

Smskanda, Dharmarja Ratha

10

10

Smskanda (outline), Rmnuja Maapa

26

15

19

Rjasiha-Style Smskanda
In the west-facing shrine of the Shore Temple
All of these Smskanda panels are of the Rjasiha-style, 27
as a summary of their characteristics will indicate. The following Pallava Smskanda
characteristics are common to all of these Smskanda panels.
Indeed, these characteristics are common to practically all of the
Rjasiha-style Smskandas. We, therefore, call it the:

Standard Table of Characteristics of

the Rjasiha-Style Smskandas
iva:
1. left leg only down.
2. four arms:
upper right: holding snakes tail.
lower right: chin mudr.
upper left: jna mudr.
lower left: ardha-dhyna mudr.
3. lower right forearm held horizontally away from his body.
4. ear ornaments are both makara kualas.
Um:
1. left leg only down.
2. torso turned to the front (non-profile).
3. two arms.
4. leaning on her left arm.
5. peculiar head-dress: a turban-like portion which is pinched in the
middle and a tall crown-like portion.
6. ear ornaments are both patra kualas.
Skanda:
1. has the same peculiar type of head-dress that Um has.
General:
1. no gaas in upper part of panel.
2. Brahm and Vishu in panel immediately above ivas upper hands
(Brahm always to proper right, Vishu to proper left).
3. umbrella above Um.
4. sana is a royal throne.

Vdagirvara Temple, Tirukkalukkunram

In addition to the Dharmarja Ratha and Rmnuja Maapa,
the only other temple we could think of which might boast of a pre-
Rjasiha style Smskanda was the famous Vdagirvara structural
shrine on top of the hill at Tirukkalukkunram. It was with great
interest, therefore, that we visited it some time ago and had a look at
the three carved stone slabs which form a major part of the inner back
and side walls of the sanctum sanctorum.
The inner structure of the sanctum probably dates from the time
of the Pallava king, Paramvara-I, the father of Rjasiha. It is not
generally appreciated that this ancient Pallava shrine is completely
encased within a later Chola vimna. It is a temple within a temple.
From the outside, only the Chola structure can be seen. The
inner shrine belonging to Paramvaras reign, therefore, is the oldest
extant structural temple under worship in South India. There is another
 28 temple of Paramvaras time at Kuram, but only the basement
Pallava Art of the original structure remains, and no regular worship is con-
ducted there.
It must be said right away that the various descriptions of these
relief carvings inside the sanctum, beginning with those of the Annual
Report on South Indian Epigraphy of 1909 (pp. 76-77), were based on
mere hearsay. That information, unfortunately, was over-imaginative.
The Report claims, for example, that:
(1) Mrkaya appears in the Smskanda panel [he doesnt!];
(2) two ishis appear in the northern panel [they dont!]; and
(3) Nandikvara and Chaikvara appear in the southern panel
[a puzzling way of describing iva-Ardhanr seated on the
bull, Nandi].
On the Reports authority, these misleading descriptions were re-
peated.2
Our own report follows: On the back inner wall of the sanctum
which faces east is a typical Rjasiha-style stone Smskanda panel
of impressive dimensions. The pilasters framing the panel and the
portion of the wall above it are clearly brick, not stone. On the inner
side wall, facing north, is an equally large relief of iva-Ardhanr.
Ardhanr, holding a v and other insignia, is seated on ivas
mount, the bull, Nandi. To the upper right (proper) of Ardhanr, in
this panel, is a small bust of Brahm with three of his faces showing.
To the upper left (proper) is a small bust of Vishu, wearing kira
makua.
On the inner side wall facing south is a panel showing a four-
armed figure seated by itself on a royal throne, in almost the same pose
and regalia which iva has in the Smskanda panel. We shall call this
figure Rjamrti. In this panel of Rjamrti, above and behind his
throne, on either side of him, are two ladies of royal appearance, with
their hands held in ajali mudr. There are no other figures.
On the outer sides of the sanctum walls, in deeply recessed
niches, there are similar but smaller and very badly worn panels.3
The unusual depth of the niches is due to the fact that the Pallava
shrine with its panels is encased within the later Chola structure.
These outer panels duplicate the inner ones. That is, on the back wall
of the sanctum, outside, facing west, is a second Smskanda panel;
on the southern wall, facing south, is a second Ardhanr; and on the
northern wall, facing north, is a second Rjamrti flanked behind by
two ladies, with their hands in ajali mudr.
The two Smskanda panels of this temple agree completely
with all of the characteristics listed in the Standard Table (Rjasiha-
Style) given earlier in this study.
 We give further details of the two Smskandas below: 29
Pallava Smskanda
Inner Smskanda Outer Smskanda
iva:
1. leg ornaments: none none
2. diagonal bands: two (at least) one (visible)
over right arm? no no
Um:
1. leg ornaments: 4+1 indeterminable (worn)
2. diagonal band: one (strands of pearls?) one
between breasts? yes no: down her left side
General:
1. Vishus emblems:
(a) flames? no indeterminable (worn)
(b) valampuri? no indeterminable (worn)
2. moon yes: disc raised and no
crescent raised further
3. Nandi below no no
4. attendants below one (as in Mahish. Cave none
Smskanda panel)
5. vessel below yes: water pot type (spout) yes: wide-mouth bowl
6. throne legs: non-animal non-animal

We also give a detailed analysis of the Ardhanr and Rjamrti panels found in the same sanctum
of the Vdagirvara temple:
Inner Outer Inner Outer
Ardhanr Ardhanr Rjamrti Rjamrti
The Deity:
1. leg ornaments: iva-half: none -half: indet. none none
Um-half: none U-half: silambu
2. diagonal bands: two indeter. two 1 visible
over right arm? no no no no
3. ear ornaments: -half: makara -half: indet. both makara both makara
U-half: patra U-half: indet.
4. leg position: left down left down left down left down
5. four arms:
upper right: trila shaft indeter. snake tail indeter.
lower right: snake staff indeter. abhaya abhaya
upper left: v neck indeter. chin mudr indeter.
lower left: v neck indeter. ardha-dhyna ardha-dhyna
General:
1. figures above: Brahm & Vishu nobody two ladies two ladies
2. sana: Nandi Nandi throne throne
(a) throne legs: non-animal non-animal
(b) ends of back rests: makara head above makara head above
rampant lion rampant lion
3. figures below: none none none none
4. yga paa on right knee indeter. no no

We add a few comments on the inner panel facing south with the figure we have called
Rjamrti. The Annual Report on South Indian Epigraphy of 1909 describes this panel as representing
Yga-Dakshimrti and two ishis. R. Nagaswamy has said that it represents Mdha-Dakshimrti and
 30 two female chauri-bearers.4 It is difficult to see how a kingly figure
Pallava Art seated on a royal throne, flanked by two ladies with their hands held
in ajali mudr (they do not have chauris) can be Dakshimrti.
There are no sages, no tree, nor any of the other characteristics which
usually go along with the Dakshimrti theme. The figure is cer-
tainly not seated out in the forest, and, as mentioned before, he has
almost the same pose and regalia which iva has in the Smskanda
panel.
It is interesting to note that, in the courtyard of the Shore
Temple, Mmallapuram, there is a stone block which has panels
carved in relief on its four sides, two of which are similar Ardhanr
and Rjamrti panels. [1997 note: this block is now in the ASI
museum, Mmallapuram.] These panels, much more modest in size
and execution, and with variations of emblems and sanas, neverthe-
less reflect the style we see in the earlier and bigger panels of the
Vdagirvara temple.

Kailsantha Temple, Kanchipuram

The visits to the Vdagirvara temple had aroused our curios-
ity concerning the Smskanda panels in the Kailsantha temple,
Kanchi that fountainhead, as it were, of Rjasihas art. We soon
found an opportunity to go there. Examining first the Smskanda in
the sanctum of the smaller temple, the Mahndravarmvara, we
found a panel which in every respect was typically Rjasiha in style.
It agrees in every detail with the characteristics listed in the Standard
Table (Rjasiha-Style).
We were stunned, therefore, when we saw next the Sm-
skanda in the main sanctum of the Rjasihvara: a diminutive panel
which in no way can be considered the work of Rjasihas period.
It is certainly a later addition.
Anyone who has first seen the large and imposing Sm-
skanda panel of the Vdagirvara temple (whose sanctum is of
modest dimensions: 187 cm. length by 170 cm. breadth, approx.),
would naturally expect an even more imposing panel in the Kailsa-
ntha temple (whose main sanctum is 265 cm. in length and 273 cm.
in breadth, approx.). But this is not the case. The Vdagirvara panel
is roughly 160 cm. high and 122 cm. broad (a vertical format).
Whereas the Rjasihvara panel is only 94 cm. high and 115 cm.
broad (a horizontal format).
But it is not just the small size of this panel which is unex-
pected. The details of the figures themselves are completely at
variance with the usual Rjasiha-style Smskanda (of which there
are 29 such examples in this temple alone). In particular, the main
sanctums aberrant panel has:
1. ivas right leg down.
2. iva has an axe in his upper right hand and a deer in his
upper left.
 3. ivas lower right forearm is not held horizontally away, with the 31
chin mudr (his lower left arm, unfortunately, seems to be Pallava Smskanda
broken off).
4. Brahm and Vishu are not included in the panel.
5. There are no gaas above, either.
6. Nor any umbrella.
7. Um is seated with both legs drawn up on the sana.
8. The sana has lost any resemblance to a royal throne.
9. Skanda is standing on the sana (between Um and iva).
Further analysis of details in dress and ornaments is impossible
because the panel has a thick coating of plaster on it.
One more anomaly is that whereas the panel in the Mahndra-
varmvara sanctum shows iva and his family seated on a throne in
the faade of a shrine or pavilion which is carved in relief with side
pillars and kapta, there is no indication of such a faade in the Rja-
sihvara sanctum. However, such a faade is found framing the
Smskanda panels of Rjasiha-style in all of the structural temples
in the sanctums sanctorum. The only possible exception is the Vda-
girvara sanctum where the side pillars framing the panel are distinct-
ive in both form and material (brick) and where there is no kapta.
Where, then, is the original Smskanda? Hidden behind an
added wall and this later panel? Removed as a war trophy, many
centuries ago, by the Chlukyas? It is difficult to say.
There are fragments of painted (not carved) Smskanda
panels which have been uncovered in a couple of the enclosure shrines
of the Kailsantha temple.5 Although little remains of the complete
scene, there are some interesting details which add to our knowledge
of the carved Smskanda panels. For instance, in shrine No. 41, the
three separate loops of ivas diagonal band are clearly shown in the
painting. The large (and thick) diagonal band is made of many strands
of pearls. The other two narrow diagonal bands seem to be strips of
cloth: the shorter loop passing around his chest rather high on his right
side; the longer loop falling almost vertically downward and disap-
pearing beneath his belt and waist garments.
Two side loops of the waistbands are each weighed down by a
heavy ring (with ornamental knob and tassel) through which they
pass. The waistband, itself, is a long strip of folded or pleated cloth
which is striped with transverse bars of color.
The glimpse one gets of Ums bust, in the painting of shrine
No. 23, is a perfect illustration of one of the ways in which women
used to paint their breasts in the early period. In the Kailsantha
painting the red color of her breasts contrasts with the normal flesh
color of her stomach. Some art historians have long been attempting
by argument to clothe the heavenly maidens of the famous Sigiriya
frescoes in Sri Lanka with diaphanous blouses. But it is quite clear in
this Kailsantha painting that the colorful, but otherwise invisible,
blouses are merely applications of sandal paste.
 32 Questions have been raised about the age of the fragments of
Pallava Art painting found in the Kailsantha temple. It is true that one can find
several layers of plaster and paint one on top of the other. We have
observed up to three layers of plaster and paint. But it is natural, in the
absence of any contrary evidence, to take the layer of plaster and paint
nearest the stones surface to be the original. And when the paintings
themselves (for example, in shrines No. 41 and No. 23) not only paral-
lel the details of the sculptured panels, but actually make clear certain
points which are otherwise obscure, then we are inclined to believe that
the lowest layer of paint in these cases is coeval with the original
construction of the temple.
The eastern and western enclosure shrines contain sculpted
Smskanda panels in typical Rjasiha-style. In several cases,
Brahm and Vishu have been completely hidden by plaster during
renovation.
On the wall between the shrines appear panels representing the
King and a Queen they very closely resemble iva and Um in the
Smskandas. The King, of course, has only two arms. At the back,
stand two female chauri bearers.
It is interesting to note that the ilparatinam prescribes that, in a
Smskanda, iva must be like Rjarja. Other works prescribe
Rjaguam for iva. Thus, the tradition of iva being represented as
the king continues even into the post-Rjasiha period.

Six Pallava Temples, Kanchipuram

Having seen the thirty Smskanda panels of the Kailsantha
temple, we next turned our attention to six minor Pallava shrines all
of them also in Kanchipuram. In 1971, we visited all six of them and
made a detailed comparison. All six of the Smskanda panels in their
sanctums exhibit the characteristics listed in the Standard Table
(Rjasiha-Style) with the following exceptions.
iva, in the Iravatanvara, Amarvara, Muktvara, and
Mtagvara, has his lower right hand in abhaya mudr. And in the
Muktvara, ivas upper hands hold an axe (right hand) and deer
(left). It must be emphasized here that the Amarvara, Airvatvara,
and Muktvara temples all have Smskanda panels which are
heavily plastered. So heavily so that even the details of emblems are
conjectural. For instance, we find in the Muktvara Smskanda that
iva has the axe and deer emblems in his upper hands, which are post-
Rjasiha characteristics. But it is anybodys guess whether these
stucco emblems truly represent the stone carving beneath.
 33

6.5 6.5

8.5 8.5

Iavttavara Mtagvara

6.5 6.5

8.5 8.5

Airvatvara Muktvara
34 Additional details are these:

Piravtanvara Iravatanvara Amarvara Airvatvara Muktvara Mtagvara

General:
1. gaas
below: 3 3 none 1 none 1
2. vessels
below: none none wide-mouth ? none none
3. throne
legs: lion plain plain ? lion lion
4. sanctum
sides: carved plain plain carved carved carved

Tlagirvara Temple, Panamalai

There remained one major temple for us to see, and in August,

1971, we visited it: the Tlagirvara structural stone temple at Pana-
malai. The Smskanda in the sanctum is of the expected Rjasiha-
style. Unfortunately, it is rather thickly plastered over. What is
unusual is that the Smskanda panel is framed by a complete shrine
(carved in relief) which rises high above to a second level which is
topped by a barrel vaulted roof with ku arches and two stps.
The Smskanda panel of this temple conforms to every one of
the characteristics listed in the Standard Table (Rjasiha-Style).
Additional details are these:
General:
1. The throne has a lateral back rest which ends in makara heads
with rampant lions directly below them; the throne legs are
non-animal.
2. No gaas or attendants below.
3. But two vessels: one pot with spout; and one wide-mouth
bowl.
4. The sanctums side (inner) walls are plain.
5. There is a torch on a standard to the proper left of Ums head
as is also found in a Shore Temple Smskanda panel (in
the Rjasihvara).
iva has no leg ornaments. Um has silambu and a diagonal
band which passes between her breasts. Nothing can be said of ivas
diagonal bands, as there is a thick covering of plaster on his chest.
Post-Rjasiha Style Smskandas
The Smskanda theme continued to be popular in the bronzes
of Tamilnad for several hundred years. The later Smskandas are
distinctly different from the Rjasiha-style.
 Instead of dealing with individual Smskandas of the later 35
period, we shall contrast some of the characteristics of Smskandas Pallava Smskanda
as laid down by the ilparatinam with those of the Rjasiha-style
Smskanda:

Rjasiha-Style ilparatinams
Smskanda Smskanda
iva:
1. left leg only down 1. right leg only down
2. four arms: 2. four arms:
UR: holding snakes tail UR: axe
LR: chin mudr LR: abhaya mudr
UL: jna mudr UL: spotted deer
LL: ardha-dhyna mudr LL: kaaka or sihakara mudr
3. ear ornaments both makara kualas 3. right ear: makara or siha kuala
left ear: patra kuala, or
both ears: patra kualas
Um:
1. peculiar head-dress: turban-like portion 1. kira-makua
pinched in the middle; tall crown-like
portion.
Skanda:
1. always sitting. 1. standing, sitting, or dancing.

14

Smskanda, Post-Rjasiha Style, Trilam Temple, Pallavaram

 36 Conclusion
Pallava Art In conclusion, we would like to say that there are enough
uniformities in all of the panels which we have examined to establish
a Rjasiha-style for most of them. We have listed these common
characteristics in the Standard Table (Rjasiha-Style). We have
called it Rjasiha because his authorship of several temples which
contain the majority of Smskandas of this type is clearly established
by inscriptions. In our fourth study we shall argue, however, that it was
his father, Paramvaravarm-I, who actually initiated the Rjasiha
style and who was the author of some of these monuments.
In two cases, the Muktvara and Mtagvara temples,
inscriptions indicate that they were built after Rjasihas reign, even
though their Smskanda panels continue in the Rjasiha-style.
In the Appendix, we have applied the techniques of numerical
taxonomy to an analysis of stylistic differences in various Smskanda
panels.
_______________

1This second study is based on Pallava Somaskandas, by

Michael Lockwood, P. Dayanandan, and Gift Siromoney, a paper read
at a meeting of the Archological Society of South India, September 9,
1971, and on its modified version which appeared in two parts in The
Sunday Standard, Madras, on the 19th and 26th of November, 1972.
2For instance, see Longhursts work, Pallava Architecture
(Archological Survey of India, Memoir No. 17, 1928), Pt. 1, p. 21.
It is not until 1966 that one gets anything like an accurate description
of the Vdagirvara carvings. This description comes in the form of a
note written by R. Nagaswamy which is appended to Chapter Eleven of
S.R. Balasubrahmanyams Early Chola Art: Part I (pp. 251-52).
3The outer sides of the main sanctum are nevertheless
protected within the enclosing verandah walls and are roofed over.
Therefore, in the darkness, a light of some sort is necessary to see the
panels in the outer niches.
4See Nagaswamys note, pp. 251-52, Early Chola Art: Part I.
5In the enclosure shrine No. 44, a carved panel of iva and Um
has been inserted some nine inches in front of the back wall which
may still have the original painting intact.
6Tanjore, 1961, chp. 22.
 APPENDIX A

A Numerical Taxonomic Analysis

of Various Smskandas
Taxonomy is the study of the principles of classification. With
the advent of computers, there has been a considerable development
in the field of numerical taxonomy. We have applied the methods of
numerical taxonomy to our study of the Smskanda panels. The
results more or less confirm our main findings presented in the body
of our second study.
Numerical taxonomists recommend a large number of charac-
ters (say from 40 to 100) to be selected for study. We have chosen 40
characters as given in Table I. When a particular character is present,
it is coded with a plus (+); when it is absent, a minus (), and when it
is not possible to determine the presence or absence of the character, a
zero (0). For example, we may use the presence of a leg ornament as a
character. If a leg ornament is present, we mark + against the charac-
ter; if the leg ornament is absent, we use . In some cases, the leg
may be covered with a thick coating of plaster so that it is impossible
to determine the presence or absence of the leg ornament. Then the
corresponding code given is 0.
We have listed characters for 15 panels. However, only the first
10 panels have been used by us in our numerical taxonomic analysis.
They are the Smskanda panels of the Dharmarja Ratha, Mahisha-
mardin cave-temple; the Vdagirvara, Tlagirvara, Rjasihvara
(Shore), Kailsantha (faade panel of Shrine No. 51), Mukunda-
nayanr, and Mtagvara temples; the east gpura of the Naarja
temple of Chidambaram; and a bronze from Nidur (see Fig. 189 in
P.R. Srinivasans book on Bronzes of South India). We have not
included the panel from the main sanctum of the Kailsantha temple
because many of the characters cannot be determined due to the thick
coating of plaster on it.
We compare these ten panels two at a time, and calculate a
similarity coefficient (S) for each pair. If two panels were to have 30
characters in common out of a total of 40 characters, then the similarity
coefficient would be 75. If all characters agree, then S is 100. And if
no characters agree, then S is 0. If the number of characters which the
panels have in common is 18, and 4 out of the 40 characters are
indeterminable (allowing, then, 36 pairs of character comparison), then
S is 50.
Since we have taken 10 panels for study, we have had to make
45 different comparisons. A similarity table for the 10 panels is given
in Table II.
 38 Each value in the similarity matrix (table) is represented by a
Pallava Art square, in Fig. 1 each square being shaded, the depth of shading vary-
ing in proportion to the similarity index. Figure 1 also represents the
stage of cluster analysis, where the similarity matrix is shown rear-
ranged so as to bring together into clusters those panels which have the
greatest mutual similarity.
In conclusion, Fig. 1 shows clearly that the Dharmarja Ratha
panel (A) stands by itself. The two late Smskandas, one from Chid-
ambaram (I) and the other, the Nidur bronze (J), stand together, but at
the same time differ from the rest of the panels.
The remaining panels, with the exception of the Mtag-
varas, stand together in a group and share high levels of similarity (89
and above). They are all panels which exhibit what we have called the
Rjasiha-style: those of the Mahishamardin cave-temple (B), the
Vdagirvara (C), Tlagirvara (D), Rjasihvara (Shore) (E), and
Mukundanayanr (F) temples, & shrine No. 51 of the Kailsantha (G).
The Mtagvara Smskanda when compared with the panels
of this group yields values of similarity ranging from 76 to 86. Thus,
even though this Smskanda (H) is close to the panels of the BCDEFG
group, yet it stands significantly apart from them.
We hope that this experiment in the application of numerical
taxonomy to iconography may lead the way to wider and more inten-
sive studies using this method.

Key to the Panels Listed in TABLE I

(A) Dharmarja Ratha (Mmallapuram)
(B) Mahishamardin cave-temple (Mmallapuram)
(C) Vdagirvara (Tirukkalukkunram)
(D) Tlagirvara (Panamalai)
(E) Rjasihvara (Shore Temple, Mmallapuram)
(F) Mukundanayanr (Mmallapuram)
(G) Shrine No. 51 (Kailsantha, Kanchipuram)
(H) Mtagvara (Kanchipuram)
(I) Naarja Temple (Chidambaram)
(J) Nidur Bronze (P.R.S.s book, Fig. 189)
(K) Periyavenmani (Chingleput Dist.)
(L) Trisulam (Pallavaram)
(M) Tirupparankkunram (Madurai)
(N) Takkolam (Chingleput Dist.)
(O) Tirupanjili (Tiruchi Dist.)
 TABLE I: Coded Data 39
Panels: A B C D E F G H I J K L M N O
iva:
1. right leg down + + + + + + +
2. left leg down + + + + + + + +
3. leg ornament present + + + + 0 +
4. waist band loops down + + + + + + + 0 +
5. two+ diagonal bands + + 0 + + 0
6. LR: forearm horiz. + + + + + +
7. UR: snake tail + + + + + + + +
8. UR: axe + + + + +
9. LR: chin mudr + + + + + + + +
10. LR: abhaya + + + + + +
11. UL: jna mudr + + + + + + + 0
12. UL: deer + + + +
13. LL: dhyna mudr + + + + + + + +
14. LL: fist on thigh + +
15. LL: chin mudr + +
16. left ear: makara kuala + + + + + + + + + + + + +
17. right ear: makara k. + + + + + + + + + + + + +
18. headdress short (1.5x) + + + + + + + + +
19. udarabandha present 0 + + + + 0 + +
Um:
20. left leg down + + + + + + + + + + + + + + +
21. waist band sash down + + + + + + + 0 0
22. long diagonal band + + 0 + + 0 0 0 0 + +
23. torso profile +
24. leaning on left arm + + + + + + + + + + + +
25. right hand touching ear +
26. right ear: patra kuala + + + + + + + + + + 0
27. left ear: patra kuala + + + + + + + 0 + + 0 +
28. large patra kuala + 0 +
29. pinch in middle (hair) + + + + + + + + 0
30. headdress short (1.5x) + + + + + + + + + + +
31. headdress conical + + + + + + +
Skanda:
32. seated + + + + + + + + + + + +
General:
33. gaas above (in panel) +
34. Brahm & V. behind . + + + + + + +
35. umbrella above + + + + + + + +
36. royal throne 0 + + + + + 0 + + +
37. makara traa + +
38. attendants below + + 0 + +
39. vessel(s) below + + + 0
40. iva & Um close + + + + + + + + + + + +
40

TABLE II: Similarity Matrix

Dharmarja Ratha (A) 100

(Mmallapuram)
Mahisha. Cave-Temple (B) 44 100
(Mmallapuram)
Vdagirvara (C) 41 98 100
(Tirukkalukkunram)
Tlagirvara (D) 50 89 92 100
(Panamalai)
Rjasihvara (E) 44 95 98 95 100
(Shore Temple)
Mukundanayanr (F) 46 93 90 92 93 100
(Mmallapuram)
Shrine 51 - Kailsantha (G) 50 94 94 97 94 94 100
(Kanchipuram)
Mtagvara (H) 54 79 76 86 79 82 86 100
(Kanchipuram)
Naarja Temple (I) 47 23 21 31 23 31 29 46 100
(Chidambaram)
Nidur Bronze (J) 54 23 20 30 23 30 28 45 90 100
__________________________________________________
A B C D E F G H I J
 41

Fig. 1

Matrix Rearranged According to Cluster Analysis

42

15

17.5

1. Smskanda, Mahishamardin Cave-Temple

 APPENDIX B

Five Smskanda Panels

The five Smskanda panels illustrated in this Appendix are from
the following temples:
1. Mahiamardin cave-temple (Pallava), Mmallapuram;
2. Yamadharma temple, Tirupanjili;
3. Katriyasihvara (the east-facing shrine of the Shore Temple)
(Pallava), Mmallapuram;
4. Tiruvural-Mahdva temple (Pallava), Takkolam;
5. Kyrhaa temple (Chola), Kanchi.
All of the photographs of these Smskanda panels, excepting the
fifth, are courtesy of the Archological Survey of India, Temple Survey
Project (Southern Region). Copyright belongs to the Archological
Survey of India. The fifth photograph is reproduced courtesy of Mr. V.
Narayanaswamy, who came across this Smskanda in one of his many
investigative expeditions.

2. Smskanda, Yamadharma Temple, Tirupajili

44

15

19

3. Smskanda, Shore Temple, east-facing shrine

 45

7.5

4. Smskanda, Tiruvural Mahdva Temple, Takkolam

12.5

10

5. Smskanda, Kyrhaa Temple, Kanchi

46

Gagdhara, lateral niche, Kailsantha Temple, Kanchi

 THREE

Pallava Gagdhara1
In the Introduction to our studies, we have briefly noted the
story of Bhagratha and the descent of the river Gag, which is
narrated in the Rmyaa. The point we would like to emphasize here
is that the goddess Gag was enraged when iva commanded her to
descend to earth:
He calls me, in her wrath she cried,
And all my flood shall sweep
And whirl him in oerwhelming tide
To hells profoundest deep.
(After Griffiths Rmyaa, i, 190.)

But in the ensuing trial of strength, iva proved his superiority

by capturing the descending Gag in the locks of his hair! There she
stayed until her temper cooled down, when at last she flowed into the
Vindu lake, the source of the seven sacred rivers of India.
It may be of interest to note here that the terrific aspect of
ivas tussle with the goddess is clearly emphasized in the major
Gagdhara panel of the Kailsantha temple, Kanchipuram, built in
the early eighth century by the Pallava king, Rjasiha. This panel
which forms the inner back wall of the central western sub-shrine of
the main tower shows iva with a fearsome expression. His mouth is
slightly open, his teeth are bared, with two elongated fangs curving
downward. These are details on the original sandstone carving.
In this same panel, Prvat stands on ivas left. As a matter of
fact, Prvat appears for the very first time in any Pallava Gagdhara
when she appears in the Gagdhara panels of this temple.
The Gagdhara theme is repeated on the faades of two of the
enclosure shrines of the Kailsantha temple. And what is extremely
significant for the debate over the Penance Panel at Mmallapuram
(the question whether it is Bhagrathas or Arjunas penance) is the
fact that one of the sub-shrine panels (sub-shrine No. 50) actually
shows Bhagratha standing next to iva in the same tortuous stance as
is found in the Mmallapuram Penance Panel. There is no parallel
example in the whole range of Pallava art which thus portrays iva
and the penitent Arjuna.
The Gagdhara theme can be considered both as a terrific
form as well as a grace-bestowing form of iva. It is terrific in its
aspect of portraying his contest with Gag. It is grace-bestowing in
its showing the god as fulfilling the fervent prayer of Bhagratha. This
 48 double aspect is emphasized in the Kailsantha temple by the Gag-
Pallava Art dhara theme appearing both in the southern row of enclosure shrines
(which portray terrific forms of iva) as well as the northern row of
enclosure shrines (which portray grace-bestowing forms of the god).
It is the much earlier Gagdhara panel of King Mahndras
in his cave-temple at Tiruchirapalli which is the main subject of the
following study. This particular panel would seem to emphasize the
grace-bestowing aspect of the theme.
The new contribution which this study seeks to make to Indian
art history is the realization that an Indian king had an image of a god
carved, which image was at the same time a portrait or representation
of the king himself. That king was Mahndravarm-I, and the image is
the Tiruchi Gagdhara. Historians know that the practice of making
God-king images was common in the east Asian colonies of India.
But in the following study, we would not only establish that this
practice existed in India, we would also suggest that it most probably
originated here.

12

12

iva-Gagdhara Panel, King Mahndras Cave-Temple, Tiruchi

Near the summit of the Rock Fort Hill at Tiruchirapalli, there is a 49
cave-temple created in the seventh century A.D. by the Pallava king, Pallava Gagdhara
Mahndra-I. His craftsmen carved a large panel representing iva-
Gagdhara on the living rock which forms the western wall of the
cave-temple. In the art history of the Tamil country, this carving
marks the very earliest extant, large stone-sculptured panel represent-
ing a deity.
On the two pillars actually, pilasters which frame this
imposing carving, there is a famous inscription of King Mahndras.
This inscription was translated as far back as 1890 by Dr. E. Hultzsch
in the first volume of South-Indian Inscriptions. His interpretation has,
more or less, been followed by scholars up to the present day. How-
ever, we wish to present a fresh translation of this inscription which is
radically different at three key points.
First, Hultzsch, in his translation, says that King Mahndra
placed an image of iva in the cave-temple. The English word
placed is misleading here, and Hultzsch and others have concluded
that a separate piece of sculpture was brought from somewhere and
placed in the cave-temple. But, in fact, the image referred to in the
inscription is the obvious one: the figure of Gagdhara in the relief
panel itself which was carved in situ. The Sanskrit word nidhya
may be translated, poetically, as established.
Secondly, when King Mahndra had the figure of iva-
Gagdhara carved in anthropomorphic form, it was given the human
form of the king himself. That is, when we look at the Gagdhara
panel, we are actually seeing a figure of iva which is at the same time
a portrait of King Mahndra. This is the significance of the passage in
the inscription which claims that in the making of the image of iva
the king became himself sthu (fixed, immortal) together with [iva]
on earth. We probably see in this figure of Gagdhara not only the
bodily and facial likeness of the king, but also his royal dress and
ornaments. If this appears vainglorious on the part of the king, one
ought to remember that in aivism, as in other faiths, the human
person, itself, has been taken as a true temple or house of God. This is
certainly the idea conveyed in the inscription when it speaks of God
being immanent in the king.
Thirdly, in the inscription, the title Daughter of the Mountain
was taken by Hultzsch, and by everyone else since his day, to refer to
Prvat. But we wish to submit that in the context of the Gagdhara
theme, the Daughter of the Mountain is none other than Gag.
Gag, as well as Prvat, is referred to in literature as the Daughter of
the Mountain. And it is extremely significant that in the story of
Bhagratha, in the Rmyaa, where the theme of iva-Gagdhara
occurs (the very same theme of the carved panel), Gag is referred to
as the elder daughter of Himavn, the king of the Himlaya
mountains.
 50 Mahndras Inscription
Pallava Art The inscription begins on the northern pilaster:
(Verse 1) When King Guabhara [Mahndra] established a
stone figure [the relief image of iva-Gagdhara] in the wonderful
stone abode on top of the King of Mountains [the Rock-Fort Hill],
this ruler, (entitled) Vidhi [the Creator], made Sthu [iva] true
to His name [sthu: stationary / firmly fixed] and became
himself sthu [fixed, immortal] together with Him, on earth.
(V. 2) The lord of wealth, atrumalla [Mahndra], made on
this mountain an abode for the husband [iva] of the Daughter of
the King of Mountains [Gag], so that the meaning of His
[ivas] title Giria (i.e., Mountain Dweller) would be made
literally manifest.
(V. 3) Having affectionately been asked by Hara [iva], How
can I, while remaining in an earthly abode, see the abundant wealth
of the Chas and the river Kvr?, this Supreme Ruler, Gua-
bhara, the fame of whose empire rivals that of Manu, ordered for
Him [iva] this sky-scraper [cloud-licking] mountain-abode.
(V. 4) By first raising iva, the God within (his) heart, to his
head, an incomparable stone figure of Hara [iva] was then, with
pleasure, raised to the top of the mountain by this Puruttama
[Mahndra]. And by thus himself first bearing, and then by making
the mountain bear, God immanent, on top, the Exaltedness of the
Immovable One [acalasya] was made a concrete reality by him.
The inscription continues on the southern pilaster:
(V. 5) Suspecting that the God [iva], who is fond of rivers,
on seeing the Kvr, whose waters please the eye, who wears a
garland of gardens, and who possesses attractive qualities, might
fall in love with her also, the Daughter of the Mountain [Gag]
has left her fathers family to reside, I reckon, permanently here on
this mountain, calling this river [Kvr] the beloved wife of the
Pallava (king).
(V. 6) As the king called Guabhara has become embodied in
this image [ligini = Kvrdhara / Gagdhara], let the Faith,
which has been brought back from the encircling opposition, be
forever spread by this same image [ligna] throughout the world!
(V. 7) This mountain is like the diadem of his [Mahndras]
Cha province, this abode of Hara his (diadems) chief jewel, and
the splendor of akara [Gagdhara] is, as it were, his [Mahn-
dras / Kvrdharas crest-jewels] splendor.
(V. 8) This bodily image [of Satyasandha (God/king)] was
created out of the stone inscription [ilkara] of Satyasandha
[the poet-king]. By the same imperishable character, an embodi-
ment of His/his fame was made imperishable.
(Coda) The firm, surpassing devotion within Guabhara [king/
Mountain King] was (thus) scooped out and made manifest! . . .
 As we have mentioned above, the Gagdhara panel is framed 51
by two pilasters, and it is on these two pilasters that the inscription is Pallava Gagdhara
engraved.
If one were to find an inscription on the pedestal of a statue, it
would be most natural to expect some intimate relation between the
statue and the inscription. The same thing should be expected here in
the case of the Gagdhara panel. The inscription refers to the panel
itself, and to the figures therein, and not to some supposed separate
pieces of carving which would have been placed at the opposite end
of the cave-temple (far away from the inscription).
It has long been known that from a very early period in Greater
India, there existed the practice of creating images of gods which were
at the same time portraits of royal persons. In R.C. Majumdars work
on Champa, there is an ancient inscription which explicitly and
unequivocally mentions this custom. It is the Hoa-Que stel
inscription of Bhadravarman-III. The relevant passage is translated
thus:
[Ugradvs] brothers, being of one mind and with the
permission of their mother, have erected in the middle of their
native place, in the aka year denoted by gagana-dvi-magala
(820), an image of r Mahrudradva, out of devotion to and in
imitation of the features of their father, named Aja Sarthavaha,
brother of the chief queen of king r Indravarman. . . .
To the north of this they erected, in their native place, in the
aka year denoted by kha-vahni-tanu (830), an image of
Bhagavat, out of devotion to and in imitation of the features of
their mother named Pu Pov ku Rudrapura, . . . who had issued
from a family, pure from time immemorial, and who had herself
established in the aka year denoted by chandra-agni-tanu
(831), the images of Dv, Gaa and Kumra. . . .4
Since the brothers made an image of a goddess in the likeness
of their mothers features in the aka year 830, and we learn from the
inscription that the mother was herself alive and active in the following
year (aka 831), we have a record of the practice of making an image
of a god in the likeness of a living person.5
Ananda K. Coomaraswamy, in his book, History of Indian
and Indonesian Art, speaking of the cult of deifying royal ancestors,
says that the custom existed in Java, and he mentions in particular the
portrait image of King Erlanga as Vishu. Coomaraswamy further
adds, however, that in
India, royal images were indeed often set up in temples, but so far
as we know always in human form.6
Mahndras Tiruchirapalli cave-temple inscription reveals,
then, that, contrary to Coomaraswamys supposition, the custom of
making an anthropomorphic image of a god, which was at the same
time a portrait of a person, was practiced in the Mother Land, and has
 52 been documented in the early seventh century A.D. We may
Pallava Art reasonably assume from this that Greater India was only following a
custom which had developed at some earlier period in India itself.
_______________

1This third study is based on Pallava Gangadhara, by

Michael Lockwood and A. Vishnu Bhat, a paper read at a meeting of
the Archological Society of South India, March 20, 1973, and subse-
quently published under the same title in the Journal of the Ganga-
natha Jha Kendriya Sanskrit Vidyapeetha, Vol. XXVIII, Parts 3-4
(July-October, 1972), pp. 159-166. A modified version of this paper
appeared in The Sunday Standard, Madras, on April 22, 1973.
2The Rmyaa, Blakaa, Chapter 42, lka 23 (Sanskrit
edition published by Jalana Motilal, Gorakhpur, p. 82).
3This figure is simply the Gagdhara image in its aspect of
being also a portrait of King Mahndra.
4R.C. Majumdar, Champa, Vol. I, Book II of Ancient Indian
Colonies in the Far East (Lahore: The Punjab Sanskrit Book Depot,
1927), p. 120.
5It must be admitted, however, that the grammatical structure
of the passage throws doubt on the correctness of the reading of the
dates.
6Ananda K. Coomaraswamy, History of Indian and Indonesian
Art (New York: Dover Publications, Inc., 1965 first published by
Karl W. Hiersemann in 1927), p. 185.
 FOUR

God/King Images and Cult Worship1

There has been a difference of opinion among scholars over
the question of a liga cult in early Pallava aivite temples. Liga
worship was a common practice in many parts of India during the
seventh and eighth centuries A.D., and it was perhaps only natural to
suppose that the Pallavas followed the same practice in their aivite
temples. But some scholars have tried to argue that the liga was not
the object of worship in any Pallava temple until a date later than the
construction of Rjasihas temples in the early part of the eighth
century.
Our fourth study goes against this view and supports the
opinion that the consecrated object of worship in Rjasihas temples
was indeed the liga, and that the same was true of Mahndras cave-
temple at Tiruchi.
We are not claiming that every aivite temple of the early
Pallavas originally had a liga in its sanctum. The central sanctum of
the Trimrti cave-temple at Mmallapuram, for instance, did not. The
object of worship there was the relief image of iva in anthropomor-
phic form carved on the back wall of the sanctum.
However, we are arguing for an original liga cult specifically
with regard to Mahndras Tiruchi cave-temple and all of Rjasihas
structural temples.
The subject matter of our third study, Pallava Gagdhara,
especially the famous inscription of the Tiruchi cave-temple, provides
supporting evidence for an early Pallava liga cult. And the practice
of making God-king images, which is introduced there, is developed
further in the present study.

Two major problems are dealt with in this paper. One of them is
the question of the God-king relationship expressed in the art of the
Pallavas. The other problem is the question of whether there was liga
worship in the early Pallava aivite temples.
These two problems are indirectly related, and we have tried to
draw upon the evidence in one field for enlightenment in the other.
1. Pallava Liga Worship
K.R. Srinivasan in the Sankara Parvati Endowment Lectures,
1959-60, advanced the following thesis:
 54 . . . the sanctums in the early Pallava cave-temples dating upto
Pallava Art 730 A.D. in Tondaimandalam and dedicated to Siva were devoid
of a liga of Pallava origin. Even in the structural temples of
Rjasiha with the Smskanda relief on the hind wall of the
sanctum, forming the primary object of worship, the installation of
the liga was an afterthought, as the in situ evidences would
indicate.2
The evidence put forward in the above lectures was developed
and augmented by K.V. Soundara Rajan in his 1964 paper, Cult in
the Pallava Temples.3 In this paper he points out that during the
Mahndra, Mahmalla, and Paramvara reigns:
There was no provision for any liga to be fixed in the centre of
the shrine chamber. . . .4
And a little later in the same paper:
Although ligas are found in most of the temples of Rjasiha,
as we see them today, there are strong grounds in favour of their
being later insertions.5
Some of these arguments are based on the observation that the
arrangements for abhika in early Pallava temples follow no rational
plan and betray a make-shift workmanship and crude improvisation a
crudeness which is not in keeping with the care and precision shown in
the plan and the construction of the temples themselves.
The abhika arrangements which appear crude are as follows.
First, the channel on the floor for removing the abhika water is often
crudely cut, and the spout on the outside appears improvised and in
some cases was not even provided. Secondly, some of the liga phas
are oversize for the sanctum and have therefore required assembly in
parts. Thirdly, in the aivite cave-temples of the early period which
now have ligas, these ligas are sometimes not truly centered in the
cells. The Cult article concludes that the use of regular praa
[spout, with properly oriented channel] came into ritual use by about
the end of the eighth century A.D.6
Now, let us grant the contention of these two scholars that
some of the present ligas and phas are later additions (on the basis
of their sound observations). One can, nevertheless, still maintain the
thesis that an earlier form of liga was the central object of worship in
many of these very same early aivite temples, especially those of
Rjasiha.
For instance, the abhika ritual, itself, might have been only
of a token nature, and therefore would not have required any channel
or spout. If these original ligas were anything like the one pictured
in the bas-relief panel of the Airvatvara (Pallava) temple, Kanchi-
puram, this could have been the case. Such a liga has a square base
with miniature rampant lion pilasters at its corners and an elaborately
carved padmabandha on the ligas shaft. This liga has no apparent
arrangement whatsoever for the abhika ritual as practiced today. 55
Further, such a form of the liga (especially if it were carved out of a God/King Images
single block of stone) might not have required any special provision for and Cult Worship
being fixed in the center of the shrines chamber.
Or again, another possibility (in case there was abhika
water flowing off these earlier type ligas) is that the abhika water
was collected in a container placed in the cella, itself, and therefore
the channel and spout outside were not originally required.
But there is still another argument which has been used to back
up the thesis that ligas represent a later development in the ritual of
these temples. According to K.V. Soundara Rajan, some of the
foundation inscriptions of these temples actually state explicitly that
iva in the Smskanda group was the main object of worship in many
early Pallava aivite temples. In his book, Indian Temple Styles, he
says:
For Rajasimhas explicit reference to Somaskanda as the
consecrated God in his temples, we must refer to the inscrip-
tions found in the cave-temples of his at Saluvankuppam near
Mahabalipuram.7
The reference here is to the fifth lka of the Atiraa-
chavara inscription. This lka may be translated as follows:
(King) Atiraachaa, the lord of the rulers of the earth, is the
cause of making this temple (called) Atiraachavara. May
Paupati (iva), together with the Daughter of the Mountain
(Prvat / Um), Guha (Skanda), and his retinue of gaas, always
be happy here.
At face value, this passage would seem to support the claim that
iva-Smskanda was indeed the consecrated object of worship in this
cave-temple. And there is, in fact, a Smskanda panel carved in bas-
relief on the rear wall of its sanctum.
But the famous inscription of the Pallava king, Mahndra-I, in
his cave-temple, Tiruchi, provides evidence for an alternate interpreta-
tion.
There is a much disputed passage in this inscription which has
crucial significance for our study. It reads as follows:
Guabhara-nmani rjany-anna ligna ligini jnam |
Prathat-cirya lk vipaka-vtt parvttam ||
Dr. E. Hultzsch, in the first volume (p. 29) of South-Indian
Inscriptions (1890), translated this passage as follows:
While the king called Guabhara is a worshipper of the liga,
let the knowledge which has turned back from hostile (vipaksha)
conduct, be spread for a long time in the world by this liga!
 56 If this translation were to be accepted as a correct reading of
Pallava Art the Sanskrit, it would naturally provide almost conclusive evidence that
the liga was an object of worship in Mahndras kingdom and most
probably in this Tiruchi cave-temple, itself.
But there are more ways than one of interpreting the above
passage, and our two scholars have taken exception to Hultzschs
interpretation. In the Cave-Temples of the Pallavas, K.R. Srinivasan
says of this cave-temple of Mahndras:
The temple is called il-bhavana (the wonderful stone house)
and the installed object is referred to as ailtanu (stone body or
form), which seems to suggest a stone image or sakala liga and
not perhaps a symbol or nishkala liga.8
Thus, on his interpretation, the consecrated object of worship
which was placed in the shrines chamber would have been an anthro-
pomorphic image of iva and not a symbolic liga. The author then
explains his understanding of the expression ligna ligini in the
inscription:
In the context of the preceding verses liga would denote only
the entire work (excavation of the cave-temple and the installation
therein) of the ligin viz. king Guabhara.9
In the article, Cult in the Pallava Temples, K.V. Soundara
Rajan gives the following comment on the Sanskrit passage under
discussion:
To begin with, liga as well as ligin used by the royal author
of the epigraph should at once put us wise about the ghrtha
rather than the vyakta character of the nomenclature. If Mahndra
meant a physical liga the object of worship he would have
certainly been more explicit and less pedantic. That he did not
imply the material liga is also borne out by the rest of the sentence,
which also indulges in denominational jargon of vipaka vtti
etc.10
Now, I fully agree with these two scholars when they claim that
in the Tiruchi inscription the primary meaning of liga refers to the
anthropomorphic form of iva. But I must disagree with them in their
assumption that the anthropomorphic image of iva referred to in the
inscription was an image installed in the sanctum of the cave-temple,
which image is now missing.
On our interpretation, the entire inscription (which is found on
the two pilasters framing the Gagdhara panel) refers in its primary
meaning to the contents and figures of this panel.
The significance of our interpretation is this: the words anna
ligna ligini do refer in their primary meaning to the anthropo-
morphic form of iva specifically to iva in the Gagdhara panel.
But the expression ligna ligini is an unusual one, to say the least,
and the poet must surely be punning here. Thus, the secondary
meaning of ligna ligini should be understood in the sense in which 57
Hultzsch has translated it: that King Guabhara (Mahndra) was a wor- God/King Images
shipper of the liga (the aniconic form of iva). and Cult Worship
That the poet is punning here is quite in keeping with the
general style of this inscription. For instance, in the very first lka
he puns repeatedly on the word sthu.11
The outcome of this line of reasoning is the conclusion that
Mahndras inscription definitely refers (though in a secondary
meaning) to liga worship.
Our translation12 of the disputed passage, giving its primary
meaning, is as follows:
As the king called Guabhara has become embodied in this image
[ligini = Gagdhara/Kvrdhara], let the Faith which has been
brought back from the encircling opposition be forever spread by
this same image [ligna] throughout the world!
The same passage, giving a secondary meaning, would be:
As the king called Guabhara is a worshipper of the liga, let the
Faith which has been brought back from the encircling opposition
be forever spread by this liga throughout the world!
Now let us return to the claim in the book, Indian Temple
Styles (p. 105), that a Pallava king made explicit reference to iva-
Smskanda (and not to the iva-liga) as the consecrated God in his
Saluvankuppam cave-temple. It seems to me that the Tiruchi inscrip-
tion of Mahndras provides grounds for an alternate interpretation
which could challenge the above claim.
The Tiruchi inscription repeatedly declares that King Mahndra
made the cave-temple there for iva. And throughout the inscription
the explicit reference to the God is only to his anthropomorphic form!
Take, for example, the following lka:
Having affectionately been asked by Hara [iva], How can I,
while remaining in an earthly abode, see the abundant wealth of
the Chas and the river Kvr?, this Supreme Ruler, Guabhara,
the fame of whose empire rivals that of Manu, ordered for Him
[iva] this sky-scraper [cloud-licking] mountain-abode.
And yet we have seen that the anthropomorphic form of iva referred
to by the inscription was not any consecrated image installed in the
sanctum, but rather, it was the figure of iva-Gagdhara in the panel
carved on the wall opposite the shrines chamber (sanctum sanctorum).
Further, we have seen that the secondary meaning of the words
ligna ligini is that King Mahndra worshipped the liga, and thus
the liga should have actually been the consecrated form of iva
worshipped in the sanctum of this particular cave-temple.
We may conclude, on this interpretation, that God was One for
the poet whether in the anthropomorphic form of iva-Gagdhara,
or the form of the consecrated liga, or the Spirit indwelling in
 58 the kings consciousness God immanent. That the poet chose to
Pallava Art speak explicitly of iva in the anthropomorphic form rather than in the
form of the symbolic liga, should not surprise us. The impressive
panel of Gagdhara, which was also a portrait in stone of King
Mahndra, was there for all to gaze upon and admire.

2. Image of Deity and King

The Tiruchi Gagdhara is the earliest documented example in
India where the artist has combined in one anthropomorphic figure both
a major image of a deity as well as a royal portrait. Was this artistic
synthesis of the divine with the human continued in the art of the
Pallavas? We suggest that the Smskanda image represents just such
a combination. Only, in the Smskanda panels there are three figures
which represent both divine beings as well as royal persons. A well-
known inscription of the Kailsantha temple, Kanchipuram, outlines
such a parallelism in poetic language:
Just as Guha (also called Subrahmaya or Kumra) took birth from
the supreme lord (iva), the destroyer of the war-like (demon) Pura,
thus from the supreme lord [A]gradaa [King Paramvara-I], who
was born in the race of these (viz., the Pallavas), . . . there took birth
a very pious prince (subrahmaya, kumra), the illustrious
Atyantakma [i.e., King Rjasiha], the chief of the Pallavas. . . .13
In this lka, King Rjasiha and his royal father (King
Paramvara-I) are compared to the divine Skanda and his father, Lord
Paramvara (iva). It is significant that the Smskanda panel (show-
ing iva, his consort Um, and their infant son Skanda, all seated on a
royal throne) is repeated more than 28 times in the Kailsantha temple
built by King Rjasiha.
The same comparison between kings and gods is drawn in the
Panamalai inscription of King Rjasiha:
From the lord kamalla [King Paramvara] . . . was born, like
Guha [Skanda] from the great vara [iva], he . . . who was well-
known as [King] Rjasiha. . . .14
Or, again, to return to the Kailsantha temple, Kanchi, there are
the famous, so-called Ragapatk inscriptions,15 one of which likens
the dowager queen to Um (Prvat) and compares her husband, the late
king, to Paramvara (iva):
(Her) husbands [i.e., King Paramvaras] well-merited fame
being widespread as Klakla on account of his bows power
(having been made) manifest in the destruction of cities, (thus)
like the Daughter of the Great King of Mountains, (she,) the
dearly beloved wife of Paramvara, the Bull-Bannered One, . . .
shines with surpassing splendor. . . .
 On the inner back walls of the sanctums of the two shrines 59
associated, in the past, through their inscriptions, with Ragapatk, God/King Images
are stone bas-relief Smskanda panels. We may conclude from our and Cult Worship
analysis above of the inscriptions, that the comparison between gods
and royal persons is maintained:
King Paramvara = Paramvara (iva)
his chief queen = Um (Prvat)
This comparison fits in with the other two inscriptions already
mentioned which made the following comparison:
King Paramvara = Paramvara (iva)
King Rjasiha = Guha (Skanda)
Now, besides this parallelism between poetic comparison and
sculptured panel, is there any further evidence to support our claim that
the Smskanda figures originally possessed an aspect of royal repre-
sentation? Fortunately, there is a Pallava temple which provides sub-
stantial support for our theory. It is the Vdagirvara shrine at Tiruk-
kalukkunram.
In the paper, Cult in the Pallava Temples, there is the
following statement:
. . . according to religious canons, normally only one exclusive
object of worship is to be installed.16
The import of such religious canons for the main thesis of the
Cult article is this: since one finds the Smskanda panel on the inner
back wall of most of the temple sanctums belonging to King Rjasiha,
then one ought to conclude that the Smskanda was originally the
exclusive, consecrated object of worship not the liga which, though
perhaps the chief object of veneration today, represents nevertheless a
later intrusion.
The Vdagirvara sanctum presents a serious blow to this line
of reasoning. There are no less than three equally large carved panels
of deities which fill up most of the space of the inner walls of the sanc-
tum of this temple.17 On the inner back wall is a superb Smskanda
panel. On the inner wall facing north is an equally imposing panel
depicting a four-armed iva-Ardhanr holding a v and trident and
bow, and seated on the bull, Nandi. On the inner wall facing south is a
third panel which depicts a royal-looking figure with four arms. This
kingly figure is seated on a throne which is identical with the type of
royal throne found in the Smskanda panels. Standing in back of this
king-like figure, just behind his upraised left and right hands, are two
queen-like ladies with their hands in ajali mudr.
With these three equal-size panels of deities in the sanctum, it
would be difficult to maintain that only one of them was the conse-
crated object of worship.
But what is one to make of the seated king-like figure (which, in
a previous study, we have called Rjamrti) in the panel facing
 60 south, who has two queen-like ladies in attendance? It is specifically
Pallava Art this figure which reveals to us most clearly that behind these works of
art depicting deities, there is also a positive aspect of royal representa-
tion, if not portraiture.
Others have seen in this Rjamrti panel a type of Dakshi-
mrti.18 But frankly, the only thing in common, here, would be
the aspect of meditation (indicated by the Rjamrtis ardhadhyna-
mudr of the lower left hand) and wisdom (indicated by his chin-
mudr of the upper left hand).
A more significant comparison can be made between this figure
4.5 of Rjamrti and figures in the earlier19 panels of the divarha cave-
temple and the Dharmarja Ratha, both of Mmallapuram.
The figure we wish to draw attention to in the divarha cave-
temple is the portrait of King Sihavishu. The figure of Rjamrti
and the portrait of Sihavishu have the following characteristics in
common:
6.2 (1) both have the bearing and full regalia of kingship;
(2) both are seated majestically on a royal throne;
(3) both figures have a hand in chin mudr; and
(4) both are attended by two consorts (queens) who are standing
respectfully either to the side of the throne (divarha) or
behind the throne (Vdagirvara).
Again, the portrait relief sculpture of King Sihavishu in the
King Sihavishu, divarha temple may very well be taken as the model for iva in the
divarha Cave-Temple early Smskanda panel in the third level sanctum of the Dharmarja
Ratha.20 Both figures (the portrait of Sihavishu and the Rathas
iva) are very similar in their general posture, and have the following
characteristics in common:
(1) both have right hand (ivas lower right) in chin mudr;
4.5 (2) both have left hand (ivas lower left) clenched in a fist and
placed on the left thigh;
(3) both are seated on a royal throne (ivas is unfinished, though).
The line of evolution can thus be traced as follows:
(1) first, the figure of iva-Gagdhara, Tiruchi, which is also a
portrait of King Mahndra (clearly establishing for us the
6.2
God-king synthesis in Pallava art);
(2) second, the straight portrait of King Sihavishu in the
divarha temple of Mmallas period;
(3) third, the similar looking representation of iva (God-king)
in the pre-Rjasiha-style Smskanda panel of the
Dharmarja Ratha;21 and
(4) finally, the transformation of the pre-Rjasiha-style
iva-Smskanda
Smskanda into the Rjasiha-style Smskanda,
Dharmarja Ratha and its widespread repetition in the many shrines of King
Rajasiha more than 40 such Smskanda panels have
survived.
 Even considering only the Rjasiha-style Smskanda 61
panels, there is discernible among them a definite evolutionary trend. God/King Images
We would hold that those Smskanda panels which tend to fill up the and Cult Worship
entire back wall of the sanctum are the earliest. Specific examples of
such early panels would be the huge Smskanda of the Mahisha-
mardin cave-temple at Mmallapuram and the inner Smskanda of
the Vdagirvara temple at Tirukkalukkunram.
In filling up the back wall of the sanctum, these examples only
follow the existing practice in the early temples of Mmallapuram,
such as the Draupad Ratha, the third-level shrine of the Dharmarja
Ratha, the central shrine of the Rmnuja Maapa cave-temple, and
all three cells of the Trimrti cave-temple.
Further, another aspect of the evolution of the Smskanda
panels which should be kept in mind is that the God-king equation in
them is most appropriate and flattering to the earlier king, Param-
vara, since the parallel is between himself and Lord Paramvara
(iva), the head of the divine family. The God-king relationship is
not as flattering to his son, King Rjasiha, since the parallel would
be between Rjasiha and the infant Skanda, who as an infant is out-
ranked by iva, his father, and, iconographically speaking, even by
his mother, Um.
Let us then postulate the following: the Vdagirvara Sm-
skanda and the Mahishamardin cave Smskanda are the works of
King Paramvara-I.22
Now, when we compare the Smskanda panels in the estab-
lished temple of Rjasihas with the above two panels, we note
several things. First, the relative size of the Rjasiha panels (when
compared to the dimensions of the sanctums back wall) is drastically
reduced. The panels are small. They occupy just a fraction of the
space on the back wall. Secondly, the relative size of the three main
figures (iva, Um, and Skanda) in relation to each other become more
stylized. For instance, in the Rjasiha temple panels, Um is distinct-
ly smaller in relation to iva than she is in the Mahishamardin cave-
temple Smskanda panel. The relation of size between iva and Um
in the Mahishamardin panel is far closer to what would be the case
between an actual human male and female. In other words, the
Mahishamardin Smskanda is closer to actual royal portraiture than
is any of the Smskanda panels in Rjasihas temples.
This obvious departure by Rjasihas panels from the physical
norms of relative figure size, together with the reduction of overall
panel size is quite in keeping with the process of ritual formalization
going on during Rjasihas reign and with the fact that the
parallelism between King Rjasiha and the infant Skanda is less
appropriate. Should we not, then, expect that the actual consecrated
objects of worship in the sanctums of Rjasihas temples were ligas
and not the Smskanda panels?
 62 In further support of this conjecture, we wish to point out a
Pallava Art fact which is otherwise extremely hard to understand. In two out of
the seven subordinate lateral shrines of the Kailsantha temple,
Kanchi, there are huge carved Smskanda panels which fill not only
the back wall of the shrines but spill over into the side walls also. It
seems clear that no ligas were planned for these subordinate lateral
shrines.23 Now, if one believes that the Smskanda panel was the
exclusive consecrated object of worship in the main sanctum, one has
to answer this question: How is it possible that the Smskanda panels
in the main sanctums of the Kailsantha (in both the Rjasihvara
and the Mahndravarmvara) are very much smaller than those in the
subordinate lateral shrines of the Rjasihvara? It seems to us that
the proponents of the thesis that the Smskanda panel was the exclus-
ive object of worship in the sanctum sanctorum have no adequate
answer to this paradox.
But there is no paradox when one supposes that there was a
liga as the consecrated object of worship in the sanctum sanctorum
from the very beginning, but no ligas in the subordinate lateral
shrines. In this case, the Smskanda panel in the sanctum would be
only of secondary importance, and understandably small, whereas, in
the subordinate lateral shrines, the Smskanda carving would be the
exclusive object of veneration, and thus understandably large.
We must point out one more paradox which is created by the
insistence that the Smskanda panel was the exclusive consecrated
object of worship in the sanctums of Rjasihas temples. In the
sanctum of Rjasihas Tlagirvara temple at Panamalai, we see
very clearly that the Smskanda panel is placed within the sculptured
relief of a full pavilion-like shrine. This image of a shrine is complete
with roof surmounted by two stps (all in bas-relief, of course). Now,
if the Smskanda panel were really the consecrated object of worship,
then the actual vimna of the Tlagirvara temple would be its shrine,
and not a mere bas-relief image of a shrine. The actual stp on top of
the Tlagirvara temple would be the ritually placed part consecrating
the object of worship within. What then would be the significance of
the two stps on top of the relief-sculptured shrine on the back wall?
They would be absolutely redundant!
In concluding the arguments advanced by us to show that the
Smskanda panel in Rjasihas sanctums sanctorum was not the
primary object of worship, it should be noted that these panels are
raised a significant distance above the floor level of the chamber. For
example, in the Tlagirvara temple at Panamalai, the bottom edge of
the Smskanda panel is 188 cm, above the floor level of the chamber
that is, more than 6 feet! This elevation provides ample visual
clearance above the prismatic liga which is there now. On the other
hand, this elevation of more than six feet would be hard to explain on
the view that the Smskanda panel was the exclusive object of
worship.
 The Cult article has shown us that in the Pallava art of Rja- 63
sihas period, we have an example of a sculptured panel in which both God/King Images
the liga and the anthropomorphic form of iva are shown together.24 and Cult Worship
In this panel an eight-armed deity is shown offering worship (flowers)
to an elaborately designed liga. That the anthropomorphic image of
iva in the same panel is subordinate to the liga is proved by the fact
that iva in his anthropomorphic form is on a distinctly smaller scale
than the eight-armed figure who is offering flowers to the liga in
worship. It should also be noted that the anthropomorphic form of
iva (together with Um) appears in the panel above the liga! This
example shows that the Pallavas were perfectly familiar with the
simultaneous representation of iva in his iconic and aniconic forms
and familiar with a representation in which the worship being offered
to the aniconic form is given unequivocal primacy! We may conclude
then that this panel mirrors the actual set-up inside the sanctums of
Rjasihas temples.
The article, Cult in the Pallava Temples, also mentions the
figure of Ligdbhavamrti found on the outer side of the main vimna
of the Kailsantha temple, Kanchi. It is thus admitted that this repre-
sentation of iva which combines both his iconic and aniconic forms
was propagated by Rjasiha himself. But the article has overlooked
still other examples of the Ligdbhavamrti in the Kailsantha
temple complex. For example, there is a Ligdbhavamrti panel on
the faade of the enclosure shrine No. 49. Again, it is found on no less
than three of the eight shrines in front of the main precincts of the
Kailsantha temple:
(1) in the southern niches of the second shrine to the right of
the entrance to the main precincts;
(2) in the northern niche of the fourth shrine to the right; and
(3) in the northern niche of the fifth shrine to the right.
Is not this five-fold repetition of the Ligdbhavamrti panel at
the Kailsantha temple good evidence to support the claim (based on
other grounds) that liga worship was original to this temple?
Five of the six shrines to the right of the entrance of the
Kailsantha have ligas in them now. It must be granted, however,
that these particular ligas are probably later replacements and thus
not original. Yet, it seems that scholars have failed to notice a unique
square sandstone liga pha in the fourth shrine to the right. Further,
the peculiar, indented sides of this pha are duplicated almost exactly
in the rectangular foot-rest for iva in the Smskanda panel which is
directly in back of the pha. It would thus seem that this unique pha
is an original one, whereas the circular phas in the other shrines are
admittedly later substitutions. In passing, it should be noted that the
very fact the square pha is made of friable, unpolished sandstone (and
would thus require a suitable coating of plaster over the rough surface)
provides additional evidence against an original ritual of full-fledged
abhika.
 64 Finally, in all the representations of ligas in the panels of
Pallava Art these Pallava temples, not one of them is shown faceted in the manner
so common to the ligas actually found in these temples sanctums.
What are we to make of this?
_______________

1This fourth study is based on Some Thoughts on the Early

Temples of Tondaimandalam by Michael Lockwood, a paper read
at a seminar organized by the Archological Society of South India,
October 4, 1973.
2K.R. Srinivasans Lectures, published as, Some Aspects of
Religion as Revealed by Early Monuments and Literature of the South
(Madras: University of Madras, 1960), p. 61.
3K.V. Soundara Rajan, Cult in the Pallava Temples,
Transactions of the Archological Society of South India: 1962-65
(Madras: Archological Society of South India, 1969).
4Ibid., p. 144.
5Ibid., p. 145.
6Ibid., p. 154.
7K.V. Soundara Rajan, Indian Temple Styles (New Delhi:
Munshiram Manoharlal, 1972), p. 105.
8K.R. Srinivasan, Cave-Temples of the Pallavas, Architectural
Survey of Temple Series, No. 1 (New Delhi: Archological Survey of
India, 1964), p. 87.
9Ibid., p. 88.
10Transactions: 1962-64, p. 150.
11This lka reads: When King Guabhara [Mahndra]
established a stone figure [ailn-tanu: the relief image of iva-
Gagdhara/Pallava-Kvrdhara] in the wonderful stone abode on
top of the King of Mountains [the Rock-Fort Hill], this ruler, (entitled)
Vidhi [the Creator], made Sthu [iva] true to His name [sthu:
stationary/firmly fixed] and became himself sthu [fixed, immortal]
together with Him [iva], on earth.
12I am indebted to Prof. A. Vishnu Bhat and his brother, r
Subraya Bhat, for their aid in all matters Sanskrit.
13The full inscription and translation are given by Hultzsch in
South-Indian Inscriptions, I, pp. 12-14.
14See Epigraphia Indica, XIX, pp. 113-115.
15Our full translation of the Ragapatk inscription is given
later in this book, in the article, Queen Ragapatks Inscription.
16Transactions: 1962-65, p. 156.
17These panels have been described in detail in our second
study, Pallava Smskanda.
 18The Annual Report on South Indian Epigraphy of 1909, 65
pp. 76-77; Longhurst, Pallava Architecture (Archological Survey of God/King Images
India, Memoir No. 17, 1928), Pt. I, p. 21; and a note written by R. Naga- and Cult Worship
swamy which is appended to Chapter Eleven of S.R. Balasubrahman-
yams Early Chola Art: Part I, pp. 251-52.
19The clear priority of these Mmallapuram panels has been
shown in (or would be evident from) our study, Pallava Smskanda.
20And the model also for the destroyed Smskanda panel in
the Rmnuja Maapa. See our study, Pallava Smskanda.
21And also the Smskanda of the Rmnuja Maapa.
22These panels, nevertheless, as far as style is concerned, have
been classified by us in the study, Pallava Smskanda as belonging
to the Rjasiha-style group.
23This fact is evident from the presence of a granite plinth-like
altar at the foot of the Smskanda in the north-central lateral shrine
(and in some of the other lateral shrines, also). The altar is actually a
sandwich of a sandstone slab between two granite slabs.
24This panel in the Airvatvara temple, Kanchipuram, has
been described and illustrated in the article, Cult in the Pallava
Temples.
66

iva-Smskanda, Periya Venmani

13

17

copyright: Institut franais de Pondichry / Ecole Franaise dExtrme - Orient

 FIVE

iva as Ligin in a Pallava Smskanda1

A Pallava Smskanda panel was discovered some time ago by
R. Champakalakshmi and A. Swamy in the village of Periya Venmani
in the Madurantakam Taluk of Chingleput District.2
In their description of the Periya Venmani Smskanda panel,
Champakalakshmi and Swamy did not notice the liga which is por-
trayed immediately behind ivas right shoulder.3
In the various sculptured panels found elsewhere which illus-
trate the theme of iva as Ligin, iva, in anthropomorphic form, is
shown carrying, supporting, or otherwise possessing the liga (the an-
iconic symbol of iva). Images of iva as Ligin usually portray the
god holding with one of his arms the aniconic liga on his shoulder, or
just behind his neck. In some sculptures iva supports the liga in front
of himself. There are further variations. (Even other gods and goddess-
es are sometimes portrayed carrying the liga.) In the Periya Venmani
Smskanda panel, iva does not hold the liga rather, the liga
stands directly behind ivas right shoulder.

R. Sen Gupta has written two articles on sculptures of iva as

Ligin.4 In Sen Guptas first article, Two Sculptures of iva as Ligin
from the Kailsa Temple at Ellora, the author actually discusses many
more than two ligin images. Here is a list:
1. iva as Ligin, in an Um-Mahvara panel (on a wall
flanking the gpura), illustrated in Sen Guptas first article,
Pl. I, fig. 1.
2. iva as Ligin (alone, in a panel of the garbhagha), Pl. I,
fig. 2.
3. iva and Prvat both holding a liga (a panel on the north
side of the prkra), Pl. II, fig. 4.
From Aihole, now in the Prince of Wales Museum:
4. iva as Ligin, in an Um-Mahvara panel.
A fragment in the Bhrat Kal Bhavan:
5. A liga on padmapha held with two hands atop a mans
head, Pl. II, fig. 3.
From Plikher Well II, now in the Mathura Museum (No. 882):
6. Dv (4-armed) as Ligin, the liga held atop her head with
the extra pair of hands illustrated in Sen Guptas second
article, Pl. V-A.
At the Siddhvara temple, Haveri:
7. Dv as Ligin, a piece lying loose.
 68 8. Viu as Ligin (6-armed), carrying a liga in his proper left
Pallava Art hand with the right hand held over it.
At the Pandharpur temple, the main image:
9. Vihb (Viu) as Ligin.
In Sen Guptas second article (1962), he has discussed several
more examples of ligin images. The list continues:
At the Kailsa temple, Ellora:
10. iva as Ligin, in an Um-Mahvara panel (immediately
above the principal ukans of the prkra to the north).
11. iva and Prvat both holding the same liga (sculpture found
on the adhihna of the prkra to the north), Pl. VI-B.
At Pattadakals Sagamvara temple:
12. iva as Ligin (and as Vdhara), alone, on the outer south
wall, in a miniature dvakha set in a pajara, Pl. I-A.
At Pattadakals Mallikrjuna temple:
13. iva as Ligin (on the outside, north wall of the vimna).
14. iva as Ligin (playing a game of dice with Um), Pl. II.
At Pattadakals Virpka temple:
15. iva as Ligin (2-armed), liga on right shoulder not held by
either arm; iva holds a snake in his right hand, and embraces
Prvat with his left arm; on eastern side of vimna.
16. iva as Ligin (2-armed), alone (southern side of vimna).
17. iva as Ligin (4-armed), alone (southern side, above Rvaa
panel).
18. iva as Ligin (on southern side).
19. iva as Ligin (4-armed), with Um (northern side), Pl. I-B.
20. iva as Ligin (4-armed), alone (on the lintel of the entrance to
the shrine), Pl. III-A.
At Kanchipurams Muktvara temple:
21. iva as Ligin (4-armed), with Um in Rvanugrahamrti
panel, in the temples sabhmaapa, Pl. III-B.
22. iva as Ligin (4-armed), with Um in facing panel, same
place, Pl. IV-A
The Mathura liga with figure of iva in front (2nd century A.D.).
23. iva as Ligin (4-armed), with his extra pair of hands he is
holding his jas which encircle the liga and thus support it,
Pl. IV-B.
The Kolhapur image of Mahlakm:
24. Lakm as Ligin she carries the liga on her head.
At the Bhadvara temple, Tanjore:
25. An asura as Ligin a painted 2-armed figure carrying a liga
on his head; on the outer face of the south wall of the garbha-
gha, part of a Triprntaka scene, Pl. VIII.
At the Khajuraho site museum: 69
26. Trimrti as Ligin, a 3-headed figure (in the round) surmount- iva as Ligin
ed by three more heads, with a liga at the top, Pl. VII-B.
At the Kandariy Mahdva temple:
27. Trimrti as Ligin, a similar figure, with the vhanas of the
three deities indicated.
Sen Gupta notes, in his earlier paper (p. 41), that there is epi-
graphical evidence that the Bhraiva kings used to carry a liga as a
load on their shoulder (Aabhra sannivita iva ligdvahana . . .,
J.F. Fleet, Gupta Inscriptions, pp. 236, 245). Sen Gupta also mentions
the Vraaivas and Ligyats as worshipping iva by carrying a small
liga tied around the neck. It is further suggested by him that in doing
this they only followed the Buddhists who used to carry relics, symbols,
and effigies of the Dhyn Buddhas.
Sen Gupta emphasizes the fact that the representation of iva
as Ligin has a philosophic dimension. He says that according to the
Vdntins, iva is the father or manifestation, whereas the liga
stands for the Divine Essence. On this view, the liga symbolizes the
Absolute or the Unmanifest, whereas the anthropomorphic form of iva
represents the manifest form.
Sen Gupta considers this dual aspect a parallel to the doctrine of
Kya in Mahyna Buddhism. In Buddhist iconography, a parallel to
iva as Ligin can be seen in the sixth century A.D. relief sculpture at
Kanheri: a small image of a seated Buddha is carried over the head of a
standing Buddha. In this way, Sen Gupta says, Dharmakya (the
Reality) is shown as the Buddha being held over the head of standing
Buddha, and the latter Buddha represents Rpakya (the Unreality or
the subtle form).
We add that the Bdhisattvas, in Buddhist art, have a small
image of the Buddha or stpa portrayed on the front of their headdress.
Sen Gupta then goes on to point out parallel images also in
Vaiava and Saura examples. He mentions the small image of
Ygsana Viu, seated on a flying Garua; and again, the small image
of Yga-Nryaa seen at the top of a Viu figure in the Khajuraho
Museum; and, finally, a similar small image appearing above the Bh-
Varha (432) of the Allahabad Museum. And he concludes:
Thus it will appear that the same conception of the Supreme
Spirit was entertained by the different sects: be it aiva, Vaiava,
Brahm or Saura and was represented in their images to show the
relationship with its respective manifested forms as was done in
turn by the Mahyna Buddhists to show the doctrine of Kya.5
We would like to carry this idea even further. The manifest
form of the deities or Buddhas could, and did, represent actual
contemporaneous human beings. The Kanheri Buddha (standing
image) could represent a particular monk who had achieved the highest
level of wisdom. And the many Bdhisattvas in Buddhist art quite
clearly represented kings or other rulers. Similarly, the Hindu
 70 images of the gods can be understood as representing the manifesta-
Pallava Art tion of the Supreme Reality in a particular ruler on earth.
The Tamil word for temple (K-y-il, i.e., kings abode) is
absolutely appropriate to such a localization in a shrine-house of the
embodiment (the god/king image) which is the manifestation of the
Unmanifest (the Transcendent Reality).
Even the Sanskrit names given by the Pallava kings to their own
temples exhibit this god/king dhvani. For example, consider:
1. Mahndravarmvara-Gha
2. Atyantakma-Pallavvara-Gha
3. Rjasihvara
Our modern minds usually construe these names as:
1. The Shrine of (iva,) the Lord of (King) Mahndravarm
2. The Shrine of (iva,) the Lord of the Pallava (king,) Atyanta-
kma
3. (iva,) the Lord of (King) Rjasiha
Fair enough! But this way of interpreting them gives only one level of,
perhaps, several levels of meaning. Dhvani also gives us the following
legitimate renderings:
1. The Shrine of the Lord (god/king) Mahndravarm
2. The Shrine of the Pallava Lord (god/king), Atyantakma
3. The Lord (god/king) Rjasiha
It is in the latter sense of these names of temples that the Tamil
word kyil is really appropriate.
To return to the Smskanda panels of the Pallavas, the
manifestation here takes the form of the royal family. iva is king, Um
is queen, and Skanda is the baby prince. The actual liga, which would
be standing in front of the Smskanda panel in the garbhagha of the
Pallava temples, would symbolize the Supreme Being in its unmanifest
form.
This relationship of the manifest with the Unmanifest is
represented in the Periya Venmani Smskanda panel, and the liga
appears in the panel itself, where it is portrayed behind ivas right
shoulder. This explicitly represented relationship throws further light
on the function of god/king images in Indian art.
In the context of the discussion so far, how should we inter-
pret then the passages in the Pallava inscriptions which have led such
scholars as H. Krishna Sastri to hold that these passages indicate the
practice of wearing an image of iva on the royal headdress.6
In the second half of the fourth verse of the famous Tiruchi
cave-temple inscription of King Mahndra-I,7 the religious and philo-
sophical basis of the identification of God with king is clearly implied:
Ktv iva irasi dhrayattma-sastham=uccai
irastvam=acalasya kta ktrttham ||
The gist of the above passage may be given as follows: 71
(King Mahndra) bore on his head (that is incarnate in his features iva as Ligin
and in his mind) God immanent.
As we have already maintained, the bearing of iva on ones
head, as expressed in this inscription, is a metaphor expressing Gods
immanence in ones mind, soul, and self.
Two Pallava inscriptions of Mmallapuram have also confused
scholars in this regard. Consider first the ninth verse of the inscriptions
of the Gaa Ratha and the Dharmarja Maapa:8
Abhika-jal-pr citra-ratnmbujkar |
st vil su-mukha iras-sarasi akara ||
This lka poetically describes the anointed head of King Paramvara-I
(not Rjasiha, as Krishna Sastri holds), and we translate it as follows:
In the lofty head-lake
full of the water of coronation,
A mine of multi-colored jewel-lotuses,
the handsome-faced akara is manifest.
We shall maintain that it is the idea of God being incarnate in
human form which is expressed by the poetry and not that an actual
image of iva was fixed on the headdress of the king. The portrait
sculptures of Pallava kings and queens do not have any such images on
their headdress.
Another passage referred to by Krishna Sastri is the third verse
of the inscriptions of the Gaa Ratha and the Dharmarja Maapa:
Yasyguha-bharkrnta kailsas-sa-danana |
Ptlam-agaman-mrdhn rnidhis-tam bibharty-ajam ||9
This we translate as follows:
The weight of (ivas) big toe was enough to plunge (Mount)
Kailsa together with the Ten-Faced (Rvaa) down to the
underworld, (and yet) rnidhi (the king) (manages to) bear that
Unborn (iva) on his own head!
Here, again, we would maintain that the king bears iva on his
own head in the sense that God is spiritually immanent within the mind
of the king.
Furthermore, these verses really make better sense esthetically
when the metaphors they contain are understood in the philosophical
sense of the Unmanifest and the manifest, and are not taken literally. In
fact, a literalism would ruin the whole effect of the poetry. To empha-
size this point, consider what literalism would do to the following
example taken from the poetical work, Kakarmta, by Ka
Lluka. In this lka, a gp speaks to her lord, Ka:
Urvy kpi mahdhar laghutar drbhy dht llay
Tna tva divi bhtal ca satata gvardhanddhraka |
Tv trailkyadhara vahmi kucayr-agr na tad-gayat
Ki v Kava bhana bahun puyair-ya labhyat ||10
 72 Our translation:
Pallava Art
Some small hill, with your hand,
you easily held on high,
And now, as Gvardhanddhara
youre ever praised from earth to sky.
I hold you, Bearer of the Three Worlds,
on the tips of my breasts!
But why talk so much, Kava? Who takes account?
On ones luck it merely rests.

_______________

1This study is based on an article of the same title by M.C.

Lockwood and A.V. Bhat, appearing in rnidhi: Perspectives in Indian
Archology, Art and Culture, ed. by K.V. Raman et al. (Madras: New
Era Publications, 1983), pp. 131-35.
2First reported in the Indian Express, Madras, February 4, 1972,
and later described in their article, Pallava Antiquities in Periya
Vemai, Journal of the Madras University, Vol. XLI, Nos. 1 & 2,
(pre-dated) 1969, pp. 129-37.
3Ibid., pp. 131-32.
4Two Sculptures of iva as Ligin from the Kailsa Temple
at Ellora, Journal of the Asiatic Society, Vol. I, No. 1, 1959, pp. 41-45;
and More Sculptures of iva-Ligin, J.A.S., Vol. IV, No. 2, 1962,
pp. 41-47.
5More Sculptures of iva-Ligin, p. 45.
6Epigraphia Indica, XVIII, pp. 149-50.
7South-Indian Inscriptions, Vol. I, p. 30. In our own book,
here, in several studies, we argue that the Tiruchi image of Gagdhara
represents at the same time both the god, iva, and the king, Mahndra-
varm-I.
8S.-I.I., Nos. 18 & 19.
9Ibid.
10Llukas Kakarmtam, ed. and trans., K.P.A. Menon
(Delhi: Nag Publishers, 1994), pp. 111-12.
 SIX

Authorship of Mmallapuram Monuments1

One of the outstanding problems concerning Mmallapuram
has been to determine who exactly it was that created the monuments
there. After centuries had wiped away the memory of those early days,
various answers to this question have been forthcoming. In the early
eighteenth century, one observer even suggested a Chinese influence.
Later guesses included the Siamese and Roman. However, scholarly
historical research in the nineteenth century has satisfactorily fixed the
authorship on the Pallavas. In the twentieth century, then, the chief
problem has been to determine which particular kings of this dynasty
were responsible for the monuments. The research of such scholars as
Messrs. G. Jouveau-Dubreuil, A. H. Longhurst, and K.R. Srinivasan
began to bring about a consensus of opinion that several Pallava kings
were consecutively responsible for the great monuments of Mmalla-
puram, and that one king in particular had created the majority of
them in the seventh century, that king being Narasihavarm-I,
otherwise known as Mahmalla.
However, in recent years a dissenting view would move
forward to the eighth century the building of all the monuments of
Mmallapuram. Mr. T.N. Subramaniam, in his book, The Pallavas of
Kch in South-East Asia, and Mr. R. Nagaswamy, in a research
paper, have proposed that the Pallava king, Rjasiha (Narasiha-
varm-II), was the sole author of all the Pallava monuments at
Mmallapuram.
According to this latter view, Rjasiha was the greatest
Pallava king, and his title Atyantakma indicates his ability to have
created the unlimited variety of monuments and sculpture at
Mmallapuram.
At the time this debate was developing over the authorship of
the monuments, no statistical analysis had been made of the problem
of stylistic variation. However, there are, today, scientific tools which
can be used to attack general problems of variation.
It is a plain fact that the monuments of Mmallapuram reveal
a great variety of architectural and sculptural styles. The more widely
accepted view on the chronology of the monuments takes this variety
as evidence of an evolutionary development during the reigns of
several Pallava kings. However, the opposing view would have it that
only one king was responsible for all the variety we find at Mmalla-
puram.
Now, the scientific study of variation is not new. The problem
of variation is of great importance to many scientific disciplines, such
as agriculture, animal husbandry, and psychology, to mention only
 74 three. The scientific tool which is common to them in such a study is
Pallava Art the statistical analysis of variance or variation.
Let us consider, for example, an agricultural experiment
involving two different varieties of paddy. Let each variety be grown in
10 plots of equal area. Suppose that the total yield of the first variety
works out to an average 1000 gms per plot, and that that of the second
variety, to 1500 gms per plot. Nevertheless, if the plots are considered
one by one, it will be seen that in the 10 plots of the first variety of
paddy there is bound to be a certain amount of variation from plot to
plot. Thus, one particular plot may yield 900 gms, while another yields
1100 gms. Whereas, in the 10 plots of the second variety, there may be
a variation between different plots ranging from, say, 1200 gms to
1700 gms.
When one is confronted with the variations in yield between
all 20 of the plots, it is possible, therefore, to separate out the variation
due to differences between the two varieties and the variation within
the two varieties. When the difference between varieties is significantly
higher than the difference within varieties, we say that the two varie-
ties of paddy give significantly different yields.
The same kind of statistical analysis can be applied to stylistic
variations found in art and architecture. Such an analysis was basic to
our study Pallava Smskanda. It is also fundamental to a full
understanding of several sections in the following study.

In February, 1962, at a meeting of the Archological Society of

South India, Mr. R. Nagaswamy read a paper entitled New Light on
Mamallapuram.2 This paper radically challenged the accepted
position developed by such outstanding students of the subject as
G. Jouveau-Dubreuil,3 A.H. Longhurst,4 and K.R. Srinivasan,5 who
held that several Pallava kings were consecutively responsible for the
great monuments of Mmallapuram, and that one king in particular
had created the majority of them, that king being Narasihavarm-I,
Mahmalla. As against their position, Nagaswamys thesis was that
Rjasiha (Narasihavarm-II) was the sole author of all the
Mamallapuram monuments and inscriptions.6
Many years have passed since Nagaswamys paper was pre-
sented, and there is still no general agreement on this issue. There are
many who, on reading Nagaswamys published articles, take it for
granted that his position has been indisputably established. On the
other hand, those who support the traditional view seem to continue
confident in their own position, paying slight attention to Naga-
swamys challenge. To our knowledge, no scholar has yet subjected
Nagaswamys thesis to a detailed, critical analysis. This kind of analy-
sis is what we shall attempt in what follows. We shall argue that the
traditional view is quite right in asserting multiple authorship. Our
main point, however, will be negative: Rjasiha was not the sole
author of Mmallapurams monuments. And, therefore, we shall not
attempt, in this essay, the positive, and much more difficult, task of 75
establishing a comprehensive chronology of the monuments. Authorship of Monuments
The supporters of the traditional view might ask us what value
there is in trying to disprove a theory which, from their points of view,
is so obviously untrue: that Rjasiha was the sole author of Mmalla-
purams monuments. However, we suggest there is value in marshaling
the various types of evidence so as to have an overall, systematic view
of the problem. Further, this kind of preliminary effort may serve as the
basis for that positive task of establishing a comprehensive chronology
of the monuments which will eventually earn general agreement.
Therefore, although we may take issue with T.N. Subramaniam and
Nagaswamy, we nevertheless feel that they have done good service to
scholarship on Mmallapuram by presenting a bold challenge to the
traditional view.
Summarizing the position he wants to attack, Nagaswamy says
that the supporters of the traditional view held that:
Mahendra introduced the rock cut technique to South India for the
first time and that before him, all the temples were built of brick,
mortar and other perishable materials. His caves were
characterised by simplicity in plan and in the treatment of pillars
which were square [in cross-section] at the top and bottom and
octagonal in the middle. His son Narasimha also known as
Mamalla continued the rock cut caves and for the first time
conceived the idea of cutting the huge boulders into monolithic
temples, familiarly known as rathas. He also introduced the sedant
lion at the base of the pillars and bulbous capitals with palaka at the
top. Paramesvaravarman-I who succeeded Narasimha-I, continued
the monoliths. . . . Rajasimha who succeeded Paramesvara-I was a
great builder of structural shrines as evidenced from the
Kailsantha temple of Kanchi and the Sea-shore temple at
Mamallapuram. Except the stray example of Saluvankuppam cave,
excavated by Rajasimha, there are no other caves, which could be
ascribed to him. Rajasimha for the first time introduced the
rampant lions at the base of the pillars.7
Now, according to Nagaswamy, the above hypothetical frame-
work suggested by the supporters of the traditional view runs into
several serious difficulties which would ultimately force them to give
up their position altogether. These difficulties would include:
(1) the lack of Literary evidence concerning the authorship of
Mmallapuram;
(2) confusion involving Paleographic evidence;
(3) evidence derived from a study of the evolution of temple
Architecture; and
(4) evidence from Inscriptions.
 76 1. Literature
Pallava Art
Concerning evidence from literary sources, Nagaswamy says:
no light is thrown on the subject by literature, for there are very few
references to Mamallapuram.8 Without giving any reason, he dis-
counts the references to the Vishu sea-shore temple in the Avanti-
sundarkath. But the evidence in Dains Avantisundarkath and
its abstract, Avantisundarkathsra, is extremely important and should
not be summarily dismissed without specific reasons being given.
Obviously, at the time Dain was writing, King Rjasiha had not
yet built the Shore Temple as we see it today. Only the Vishu shrine
existed with the waves brushing the feet of the image. And Dain,
who must have been writing during the reign of Paramvara-I, speaks
of the Vishu image as a work of the great ancient architects. Since
Paramvara was the father of Rjasiha, the term ancient must take
the origin of the Vishu shrine back to a time long before Rjasihas
reign.
2. Paleography
Concerning the evidence provided by a paleographic study of
the various scripts found on the monuments of Mmallapuram, Naga-
swamy holds that it will be of little value in providing any support for
the traditional position. Nagaswamy points out that, in the recording
of more than 200 royal titles of Rjasihas in the Kailsantha temple
(Kanchi), several different forms of alphabet were used. On the basis
of these differences, some scholars (Hultzsch, in particular) had sup-
posed that these inscriptions belonged to successive rulers of Kanchi,
and thus represented an evolutionary development of the script.9 The
same view was held regarding the two epigraphs of the Atiraacha-
vara cave-temple at Saluvankuppam. In this cave, several verses
praising the king, Atyantakma, have been inscribed on one wall in
one script and then the same verses on another wall in a second script.
The theory was that one inscription was a later copy of the other.
This theory, according to Nagaswamy, has been discarded:
It was only in later times that the suggestion of successive engravers
was discarded and [it was] rightly noted that since the inscriptions
on the south and north wall are identical verses, they were written
by the same king Rajasimha. In the same vein it was [rightly]
concluded that the inscriptions in four different alphabets, found at
the Kailsantha temple, which were the repetitions of the same
titles of the corresponding tiers, were all inscribed by Rajasimha
himself to exhibit varieties. Thus . . . paleography [will certainly
fail] in determining the age of the monuments of Mamallapuram.10
Our Comments 77
First, we think it should be noted in passing that the inscrip- Authorship of Monuments
tions of the Atiraachavara cave-temple are assigned to Rjasiha
by scholars on the assumption that the title Atiraachaa, in this
inscription, belongs firmly, and, perhaps, solely to Rjasiha.
Secondly, it should be noted that it was Dubreuil, in 1916, who
clearly sounded the warning about using the different forms of the
alphabet as supposedly conclusive evidence concerning chronology:
. . . we have come to the conclusion [from a paleographic study of
the inscriptions of Rjasiha] that the form of the alphabet is not
an absolute test of the age of antiquities and that inscriptions
which, by their alphabet, seem to belong to different epochs, can,
in reality, be contemporaneous.11
Yet, even this awareness of the pitfalls in drawing chrono-
logical conclusions from paleography does not in the least weaken the
evolutionary theory, a fact which can be seen from Dubreuils own
pioneering work.
In regard to the form of the script employed for the titles en-
graved on the Dharmarja Ratha, we must point out the fact that it is
closer to the Badami stone inscription of Vtpi-Koa Narasiha-
varm and some of the inscriptions of Mahndra-I than it is to any one
of the several forms of script used by Rjasiha in the temples indis-
putably assigned to him.
But there are two label inscriptions found on the third level of
the Dharmarja Ratha which are distinctly different in form of script
from the other titles on the same monument. These two label inscrip-
tions read: Atyantakma-Pallavvara-gham, and they are written in
a form of script quite similar to the one belonging to Rjasiha which
is found on the base of the main shrine of the Kailsantha, Kanchi.
Now, it has been suggested by some (who support the tradi-
tional view) that Paramvara-I continued work on the Five Rathas,
which monuments were begun by his grandfather, Mahmalla.
Nagaswamy, however, disagrees with this supposition.
The view that the monuments were consecrated by Paramvara-I
is untenable since most of the monuments at Mamallapuram are
unfinished and were never consecrated.12
We agree with Nagaswamy with regard to the work done on the
Five Rathas. It seems to us that work on them was started and came to
a premature halt in a relatively short span of time. Where we disagree
with Nagaswamy is concerning the labels: Atyantakma-Pallavvara-
Gham. These labels, we suggest, represent an appropriation by a
King Atyantakma of the unfinished Dharmarja Ratha (with special
reference to the 3rd-level cell with its Smskanda panel on the back
wall). In suggesting this we go along with the traditional view
 78 that this particular Atyantakma was certainly not Mahmalla, but
Pallava Art a later king.
3. Architecture
Turning next to the evidence for multiple authorship of
Mmallapurams monuments provided by a study of the evolution
of architectural style, Nagaswamy also rejects such evidence.
(i) Pillar Styles
Speaking of one of the key elements in the traditional argument
the evolution of pillar styles Nagaswamy says:
We all owe a great deal to Prof. Jouveau-Dubreuil for his illum-
inating study of South Indian architecture. . . . The evolution of
pillar [styles] as shown by Dubreuil was perhaps the best study
from which we were able to arrive at some tangible conclu-
sions. . . . Dubreuil suggested that beginning from the Mandaga-
pattu cave, the pillars of Mahendra are plain; Narasimha I intro-
duced the sedant lion[-based pillar] and Rajasimha introduced the
rampant lion motif [as pillar base]. But I am afraid that Dubreuil
made [a] fundamental mistake and scholars subsequent to him,
followed suit without pausing to question the suggestion. In my
opinion the evolution of architecture as suggested by Dubreuil is
of little help for our study as we shall presently notice.13
A little later in the same paper, Nagaswamy outlines the method
he will use in his attempt to discredit the architectural evolutionary
theory as applied to the monuments of Mmallapuram:
I shall now proceed to prove that the study of architecture falls
short of expectation. . . . If it is proved that during the rule of one
and the same king the architectural details exhibit great variety,
then the evolutionary theory which is based on the conviction that
with one king only one form of architecture prevailed and each
king introduced a novel theme will certainly fall short of any satis-
factory [confirmation].
We hold that it is Nagaswamy who, at this point, has funda-
mentally mistaken what is at issue. Dubreuil, in his Pallava Antiquities,
never makes such a claim: that with one king only one form of
architecture prevailed. After all, just because each king may have
introduced a novel architectural feature, this would not necessarily
mean that he gave up entirely the features established by his
predecessors.
Be that as it may, let us continue with Nagaswamys attempt at
proof:
The rock cut cave of Saluvankuppam excavated by Rajasimha has
very simple and plain pillars very much like the Mahendra pillars.
Therefore Saluvankuppam cave posed a great problem for scholars
in placing it in the evolutionary [framework]. . . . [L]et us take the
 case of [the] Vayalur inscription of the same king. The inscription 79
is engraved on a pillar itself, which is plain and is in the so-called Authorship of Monuments
Mahendra style. The huge Nandi Maapa in front of the Kailsa-
ntha temple of Kanchipuram has four pillars with sedant lions at
the base. The small cells running along the outer walls of the same
temple, also have pillars with sedant lions at the base.15
Continuing with a consideration of the Kri Maapa and the
divarha cave (unfortunately his description of these cave-temples is
mixed up), Nagaswamy concludes:
Thus the theory that with one king only [one] form of architecture
prevailed will not hold good. . . . Thus it is quite evident, the
evolution of architectural motifs fails with reference to our present
study.16
(ii) Our Comment
Let us be perfectly clear about our criticism of Nagaswamys
methodology. In attacking the claim of one king, only one form of
architecture, Nagaswamy is attacking a position which certainly
Dubreuil and K.R. Srinivasan never held!
(iii) Variety
But it is not just evidence based on the evolution of pillar styles
which is discredited in Nagaswamys eyes. He compares the three
major temples which are now unanimously assigned to Rjasiha (the
Kailsantha, Tlagirvara, and Shore temples) and finds such a
bewildering variety in their architectural details, that had
Rajasimha not left his inscriptions in these temples, certainly these
monuments would have been ascribed to various monarchs and
would have been ascribed to various centuries.17
What are these variations in architectural details which lead
Nagaswamy to the above conclusion? They are variations in:
(1) ground plans,
(2) shapes of vimna superstructures,
(3) variations as to whether the temple walls are plain or relief-
sculptured, and
(4) whether ligas are present or absent in lateral shrines (of the
Kailsantha and Tlagirvara temples).
(iv) Our Comments
Now, interesting as these variations may be, Nagaswamy has
chosen to deal with features which do not provide in themselves the
most adequate basis for either establishing or challenging a given
chronology of evolutionary development. During the earliest develop-
ment of structural stone temples in the Tamil country, it is not surpris-
ing that a great deal of experimenting was done with regard to ground
plans, shapes of the vimna, and such details as whether to have plain
walls (that is, walls with only paintings of figures on them) or to have
 80 sculptured walls (that is, walls with paintings which are enhanced by
Pallava Art the tri-dimensionality of relief carvings). In regard to this latter varia-
tion, it cannot be over-emphasized that it is the art of painting which is
absolutely fundamental. Painted sculptural reliefs are primarily paint-
ings, and only secondarily carvings! Thus, the plain walls of the
Tlagirvara temple side-shrines with their paintings (now almost
irretrievably lost due to centuries of deterioration) and the sculptured
walls of the Kailsantha temple (which have also lost their original
cover-paintings) do not represent a variation which significantly chal-
lenges or weakens the evolutionary analysis of the development of
Pallava architecture.
What then (we may be asked) are the significant features for
such an evolutionary analysis?
First, and most important, would be a minute and exhaustive
study of variations in the dress and ornaments of figures in sculpted
panels. Charles Fabri has rightly expressed the importance of such a
study:
Dress, as must be obvious to anyone interested in humanity, is a
marked characteristic of any culture. . . . [The] tastes and tenden-
cies of an age are clearly indicated by the type of clothes a period
fancies. . . . Because fashions change, a careful observation of
these changes is one of the most powerful tools in the hands of an
art historian. For it is possible to date paintings and sculpture
within a generation when no other data, such as inscriptions, are
available, by an accurate attention to the clothes worn by the
human figures depicted.18
Secondly, even details of sculpture which are not connected
with human dress and ornaments may be quite significant. To mention
one example as illustration, Nagaswamy has tried to show that any
argument supporting a given chronology which is based on a supposed
evolution of pillar styles is worthless. According to Nagaswamy, the
existence of all three types of pillars (plain Mahndra, sedant-lion-
based Mmalla, and rampant-lion-based Rjasiha) in Rjasihas
Kailsantha temple is enough to prove the worthlessness of such an
analysis. However, we suggest that a detailed and careful study of the
features and characteristics of the carved lions, themselves, which form
the base of the pillars of the Mmalla type and the Rjasiha type
would enable one to distinguish easily a sedant-lion-based pillar carved
in the time of Rjasiha from a sedant-lion-based pillar carved in an
earlier kings reign.19
Thus, if we avoid the over-simplification involved in the belief
that with one king only one form of architecture prevailed, then the
study of pillar style evolution will surely continue to be one of the most
important elements in any effort to establish a chronological develop-
ment of early Pallava temples.
Another detail of sculpture which underwent an evolutionary
development, and which was noted by Dubreuil in Pallava Antiquities,
is the tiruvchi (the term Dubreuil uses for the ornamental arch span- 81
ning the top of the niche and issuing on both sides from the mouths of Authorship of Monuments
makaras). Dubreuil had noted that in Mahndras time (for instance,
on the faade of the Dalavanur cave-temple), the tiruvchi is double-
arched. In all of the undisputed temples of Rjasiha, the tiruvchi is
single-arched. It is therefore significant that on the Draupad Ratha
and the Trimrti cave-temple at Mmallapuram, the tiruvchi is
double-arched.
Finally, K.V. Soundara Rajan has pointed out certain other
features which seem to be significant for an evolutionary analysis.
About one, he says:
An important compositional feature of the free-standing monoliths
of Mamallapuram is that almost all of the series . . . show the hra
of karakas and bhadra las in each of the talas, including the
topmost. [A] significant modification of this rule is the ending of
the last tala of the vimna with a kapta and prastara above, but
without the kudra alpa ikhara above them in their respective
places along the periphery of that tala. This [modification]
becomes the norm in all the structural temples of post-Mmalla
period which further shows a secondary variation by replacing the
hra of miniature ikharas by the nandis placed in the corner.20
Another significant architectural change, according to Soundara
Rajan,
was the dropping of the hra in almost all Rjasiha temples
around the lowest tala as well, except on the mukhamaapa
roof.21
But let us return to Nagaswamys paper. Having attempted to
disprove the evolutionists position by linking it with the untenable
claim of one king, only one architectural form, Nagaswamy turns,
finally, to the evidence available from inscriptions.
4. Inscriptions
Speaking of the various kinds of evidence examined by him so
far, and considering their failure in establishing the chronology of the
various monuments of Mmallapuram, Nagaswamy has this to say
(and we paraphrase him):
Neither literature nor paleography nor for that matter architecture
helps us in determining the age of the monuments. We find
ourselves on no better ground when we turn to the inscriptions of
Mmallapuram. The reason for this predicament is that many
kings are said to have assumed the same names and titles, and with
respect to Mmallapurams monuments, where we have only titles,
the difficulty is all the greater.22
(i) Nagaswamys Hypothesis
In the very next paragraph following the above quotation,
Nagaswamy says:
 82 When thus, all our tangible sources fail how are we to arrive at a
Pallava Art conclusion? My answer will be that the clue to our problem lies in
the very failure of all these sources. Paradoxical it may seem,
when I say that all these evidences do not fail us when we reverse
our process of enquiry by first taking inscriptions, applying it to
architecture and applying both to paleography. We arrive at a
solution which is quite convincing.23
What Nagaswamy means, of course, is that if one proceeds on
his hypothesis (that is, that Rjasiha was the sole author of all the
monuments and inscriptions at Mmallapuram), then one may, accord-
ing to him, arrive at a convincing solution by the route he outlines
(examining first inscriptions, then architecture, and finally pal-
ography). It must be noted, however, that in fact he never did go
beyond a discussion of inscriptions.
First, Nagaswamy notes that Atiraachaa is given as one
of the many titles of King Rjasiha in his Kailsantha temple
inscriptions. There is also a cave-temple at Saluvankuppam, near
Mmallapuram, which has foundation inscriptions clearly stating that
Atiraachaa made this (temple called) Atiraachavara. There-
fore, assuming that the same title refers to the same king (Rjasiha),
Nagaswamy says:
The same king Atiraacaa has excavated the Saluvankuppam
cave and left his inscriptions. . . . But strange indeed, [some of]
the same verses are found [in inscriptions] in the Dharmarja-
maapa and Ganesa ratha! Verse for verse, word for word and
syllable for syllable they are identical. . . .24
This concordance of verses is enough to convince Nagaswamy
that the author of the Dharmarja Maapa, the Rmnuja Maapa,
and the Gaa Ratha was also King Rjasiha.
Secondly, the king who caused the Atiraachavara cave-
temple to be made at Saluvankuppam also had the title Atyantakma.
The same title, Atyantakma appears on the Dharmarja Ratha, and
Nagaswamy points out that the Dharmarja Ratha also bears the label
Atyantakma-Pallavvara-gham. And thus Nagaswamy adds the
Dharmarja Ratha to his list of Rjasihas monuments.
Of course, one of the key assumptions upon which Naga-
swamys argument is based is that the titles Atiraachaa and
Atyantakma were titles not shared by other Pallava kings.
We must emphasize the fact, here, that there simply is no
sound method available to Nagaswamy or anybody else to prove that
a given title belongs exclusively to one king. Using Nagaswamys
methodology, one might as well argue that because Mahndra had the
title Avanibhjana, and that title appears on the Kailsantha temple,
therefore Mahndra built that monument! Or, vice versa, because
Rjasiha had the title Avanibhjana, and we find this same title on
several cave-temples commonly ascribed to Mahndra, we must con- 83
clude that these cave-temples were really built by Rjasiha! Authorship of Monuments
(ii) Concordance of Titles on Dharmarja Ratha and Kailsantha
Nagaswamy makes the following claim:
Of the thirty titles inscribed [on the] Dharmarjaratha, over fifteen
titles are found in Kanchi inscriptions of Rajasimha.25
The list of royal titles which he says are common to both the
Kailsantha temple and the Dharmarja Ratha are given below in the
left-hand column. We give our comments and the niche numbers of
the Kailsantha (where a given title is found) to the right:
1. Narasiha (Narasihaviu, on one of the small
shrines in front of the Kailsantha)
2. rbhara (3-3)
3. Bhvanabhjana (24-1)
4. rmgha (4-1)
5. Trailkyavarddhana [only Trailkyantha (17-3) at Kanchi]
6. Atyantakma (1-2)
7. Kmalalita (not at Kailsantha)
8. Nayanamanhara (44-4)
9. Sarvvatbhadra (15-1)
10. rnidhi (not at Kailsantha)
11. Niruttara (not at Kailsantha)
12. Parvara (essentially the same title as 14 below)
13. Raajaya (1-3)
14. Parbhara (21-1)
15. Mahmalla (23-1)
16. Apratihatasana [only Apratihata (UG-2) at Kanchi]
It can be seen from our analysis that of the 29 different titles
(not 30) inscribed on the Dharmarja Ratha, only 12 are identical (or
very similar) to titles found at the Kailsantha. That is, only 41% of
the Dharmarja Ratha titles are duplicates (even approximately) of
titles found in Kanchi.26
(iii) Our First Point
Nagaswamy thinks that this fact (that nearly half of the titles
on the Dharmarja Ratha are found in Kanchi) provides significant
support for his hypothesis that Rjasiha built the Dharmarja Ratha.
But we are of the opposite opinion that this fact actually goes against
his hypothesis. There are over 250 different titles given to Rjasiha
in his Kailsantha inscriptions. How is it, we ask, that with this
exceedingly large collection of titles available to Rjasiha, only 41%
of the Dharmarja Ratha titles are titles which are also found in the
Kailsantha inscriptions? On the other hand, this low percentage is
 84 quite understandable if the Dharmarja Ratha inscriptions are by
Pallava Art predecessors of Rjasiha.
(iv) Our Second Point
Of the 252 different royal titles which are engraved on the
shrines surrounding the Kailsantha temple, only 2 titles appear twice
(that is only 2 titles are repeated on a given level). But of the 29
different titles engraved on the Dharmarja Ratha, 7 appear twice on
this ratha, and 1 title (Vidhi) appears 3 times. Why are the titles on
the Dharmarja Ratha so repetitious?
(v) Our Third Point
Rjasiha is a title not found on the Dharmarja Ratha!
though it is found on all of Rjasihas undisputed temples: the
Kailsantha, Tlagirvara, and Shore temples.
(vi) Our Fourth Point
In fact, except at the Shore Temple, the title Rjasiha does
not appear at all at Mmallapuram!
(vii) Our Fifth Point
Rjasiha is the very first title one meets when circum-
ambulating the Kailsantha temple. Similarly, Narasiha is the first
title one meets when circumambulating, at ground level, the Dharma-
rja Ratha. And, to take an even earlier example, Mahndravikrama
is the first title given in the Pallavaram cave-temple inscription of
Mahndra-I. Isnt there some significance in these first-place titles?
It should be noted in this connection that Narasiha is not among the
252 titles engraved in the Kailsantha. Isnt there some significance in
this omission, which surely must have been deliberate? The name
Narasihaviu appears only on one of the little shrines outside of
the main precincts of the Kailsantha temple. As we have noted
above, the title Rjasiha does not appear at all on the Dharmarja
Ratha. Finally, it should be noted that in Rjasihas Vayalur inscrip-
tion, it is Rjasiha (not Narasiha) which is the title given the
king, whereas, in the same inscription, Narasihavarm is the given
name of his great-grandfather. We may therefore assume that though
Narasiha was Rjasihas coronation name, yet he preferred
Rjasiha, or other titles, so as to distinguish himself from his
illustrious great-grandfather, Vtpi-Koa Narasihavarm.
(viii) The aivite Curse
Nagaswamy notes that the last verse of the Gaa Ratha and
the Dharmarja Maapa inscriptions (a curse) is found repeated at the
Rmnuja Maapa and the divarha cave-temple. This verse has
been rendered thus:
Six times cursed be those, in whose hearts does not dwell Rudra
(iva), the deliverer from the walking on the evil path!27
 The concordance of this verse, together with a concordance of 85
other verses, leads Nagaswamy to add the divarha cave-temple to the Authorship of Monuments
Rmnuja Maapa, the Dharmarja Maapa, the Gaa Ratha, and
the Atiraachavara cave-temple, as monuments built by Rjasiha.
However, there are a few points we would like to make concerning this
aivite curse which are in opposition to Nagaswamys thesis.
(ix) Our First Point
The aivite curse does not appear on any of the monuments
which are indisputably assigned to Rjasiha (the Kailsantha,
Tlagirvara, and Shore temples).
(x) Our Second Point
The curse, by itself, appears on the floor of the divarha cave-
temple (a Vishu temple still under worship today). It is extremely
unlikely, to say the least, that the builder of this structure would have
put a aivite curse on the floor of his own temple dedicated to Vishu!
If this reasoning is sound, then the following logical deductions may be
made:
Let the author of the curse = x;
Then the author of the divarha temple is an ancestor of x;
If x is Rjasiha, then the builder of this temple was Param-
vara-I and/or previous ancestor(s);
If x is Paramvara, then the builder was Mahmalla and/or
previous ancestor(s).
(xi) A Final General Observation on Inscriptions
We should like to emphasize the fact that not one of the
following monuments at Mmallapuram has any foundation inscrip-
tion: the Five Rathas, the divarha and Varha-II cave-temples, the
Kikal, Rmnuja maapas, and the Mahishamardin cave-temple.
This is unlike Rjasihas practice in those temples which are ascribed
to him by scholars.
So much for stone inscriptions, admittedly an area in which
there seems to be no proof positive, one way or the other, on the issue
of the authorship of Mmallapurams monuments. However, we hope
that we have raised enough points to indicate the serious problems for
anyone trying to use inscriptions to confirm the hypothesis that Rja-
siha was the sole author of the monuments and inscriptions of
Mmallapuram.
5. Dress and Ornaments
Our first study, Pallava Dvraplas and the Mahishamardin
Cave, provided overwhelming evidence that in one cave-temple there
were at least two distinct stages of work. The most obvious evidence is
the fact that, stylistically speaking, the Smskanda panel on the back
wall of the central sanctum of the Mahishamardin cave-temple is quite
different, in many points of dress and ornaments of the figures depicted,
 86 when compared with the Reclining Vishu and Mahishamardin
Pallava Art panels on either side of the rock-cut maapa of the same temple.
Further, the evidence from a study of the figures of guardians
carved on the sides of the entrances to the three sanctums of this cave-
temple indicated that the main sanctum was originally intended for
Vishu, but that it was converted at a later date into a aivite sanctum
with the Smskanda panel on its rear wall.
In regard to our stylistic analysis of the three panels of the
Mahishamardin cave-temple, we demonstrated in the earlier study
the following relationships. The Smskanda panel of the cave-temple
is a relatively later Pallava work, as it compares with similar panels of
the eighth century Kailsantha period, and contrasts with the seventh
century Smskanda panel of the Dharmarja Ratha. The other two
panels of the cave-temple are earlier, seventh century works, as they
have the early characteristics, and contrast with panels of the same
theme created in the Kailsantha period.
It is therefore difficult to believe that one king, Rjasiha,
created all the monuments at Mmallapuram, when in this cave-temple
there is such a change in the style of panels, and when there is evidence
for a shift in the dedication of the main sanctum from Vishu to iva-
Smskanda!
6. Size of Ear Ornament
One of the most important characteristics in a study of the
evolution of dress and ornaments of Pallava-period sculpture is the
relative size of ear ornaments. In particular, the circular patra kuala
is easy to measure and compare. Now, the figures in Mahndras cave-
temples (mostly dvraplas) have enormous ear ornaments, extending
well below shoulder level. But in all of the temples unanimously
attributed to Rjasiha, the figures have very much smaller ear orna-
ments. The patra kualas in the Rjasiha period often do not even
touch the shoulder.
What then is the relative size of ear ornaments of figures
belonging to the Mmallapuram monuments under dispute? Well, the
ear ornaments of figures in the divarha cave-temple, the Kikal
Maapa, and the Kisha Maapa are very large approaching the
relative enormousness of the Mahndra period! And the ear ornaments
of figures on the Five Rathas, the Penance Panel, Varha-II, and
Trimrti cave-temples are of a size intermediate between the Mahndra
and Rjasiha periods. (There is no doubt, however, that they are
distinctly larger than those of the Rjasiha period!)
Now, an interesting point arises. According to Nagaswamy,
Rjasiha created all of the (Pallava) monuments at Mmallapuram.
But the Five Rathas are incomplete. So are many of the cave-temples
and both Penance Panels. Nagaswamys chronology, then, would have
Rjasiha completing all of his known structural temples, but leaving
unfinished the monuments listed above. That is, the Five Rathas, many
of the cave-temples, and both Penance panels are the very latest monu- 87
ments to have been attempted by Rjasiha, but he was unable to Authorship of Monuments
complete them. We feel that this is a very strange order of events.
And our study of the evolution of ear ornament size would provide
clear evidence against such an order.
7. Rjasiha and Variety
As mentioned earlier, Nagaswamy has tried to argue that
Rjasiha was the greatest Pallava king and quite capable of creating
all of the various styles found at Mmallapuram. Nagaswamy has
equated Rjasiha with King Atyantakma, and has interpreted this
biruda as meaning a king capable of creating unlimited variety. We,
of course, feel that this is stretching too far the meaning of Atyanta-
kma.
Now, fortunately, because the Smskanda panel was almost a
trademark of Rjasiha, we were able to make a detailed study of the
degree of variety this king was capable of in all of his known temples.
In the Kailsantha temple alone there are thirty Smskanda panels!
In the Shore Temple, there are two Smskandas. In the Tlagirvara,
one. In our second study we have shown that a detailed comparative
study of Smskanda panels will provide overwhelming evidence
against Nagaswamys contention about Rjasihas creative capacity.
The Rjasiha-style Smskanda repeats itself more than 46 times
almost monotonously, when one carries out such an overall compari-
son!
8. The Great Gap
There is a general observation which we would like to stress
at this point. If, on Nagaswamys view, all of the monuments at
Mmallapuram are to be assigned to the reign of Rjasiha, there is
then a perplexing gap of rock-cut architectural and sculptural inactivity
between the time of Mahndra-I and the time of Rjasiha. Mahndra
created more than nine cave-temples. And Mmallapuram is a show-
case of many different types of stone monuments. But if the monu-
ments of Mmallapuram are all assigned to Rjasiha, then what were
all the artisans and sculptors doing during the reigns of the great Mah-
malla, his son (Mahndra-II), and Paramvara-I? Was there really a
gap of some 70 years when no rock-cut caves or stone temples were
being created? Prima facie, this seems highly unlikely, indeed.
9. A Last Word from Inscriptions
N. Ramesan has edited two copper plate grants in a publication
of the Government of Andhra Pradesh.28 One of these grants, the
Chir copper plates of the Pallava king Nipatuga, gives us inform-
ation about a Vishu shrine (an abode built out of stones) constructed
on the sea-shore by King Narasiha. Since this information is given in
the genealogical account of King Nipatuga, it is clear that this Nara-
siha is Mahmalla (Narasihavarm-I).
 88 The relevant Sanskrit passage actually reads:
Pallava Art Ya-ayy-gham-amabhir-j-jala-nidhau cakr Mah-c-cakrina ||
This passage may be translated into English as follows. It speaks of
King Narasiha:
who built out of stones, on the ocean, an abode (for) the One who
possesses the mighty discus (Vishu) to recline in.
The reference, unquestionably, is to the Vishu shrine belonging
now to the Shore Temple complex at Mmallapuram.29
Some objections have been raised concerning the genuineness
of the Chir grant. And even if it were genuine, the fact that it is
removed some eight generations from the days of King Narasiha-I
would not allow us to accept all of its statements blindly. Nevertheless,
until some specific arguments falsify it, the statement stands as a clear
contradiction of the hypothesis that King Rjasiha built all the
monuments at Mmallapuram.
_______________

1This study is based on a paper entitled, On the Authorship

of Mahabalipurams Monuments, by Michael Lockwood and Gift
Siromoney, which was read at a meeting of the Archological Society
of South India, March 20, 1971.
2Published in the Transactions of the Archological Society
of South India: 1960-62 (Madras: The Archological Society of South
India, 1962), pp. 1-50.
3Pallava Antiquities, Vol. I (London: 1916).
4Pallava Architecture, 3 Parts, being Memoirs of the Archo-
logical Survey of India, Nos. 17, 33, and 40 (Archological Survey of
India, Simla, 1924, and Calcutta, 1928 and 1930).
5Cave-Temples of the Pallavas, Architectural Survey of Temple
Series, No. 1 (New Delhi: Archological Survey of India, 1964).
6Nagaswamy, Transactions: 1960-62, p. 34.
7Ibid., p. 2.
8Ibid., p. 5.
9See South-Indian Inscriptions, Vol. I (Madras: Archological
Survey of India, 1890), p. 10.
10Nagaswamy, pp. 6-7.
11Pallava Antiquities, Vol. I, p. 74.
12Nagaswamy, p. 25.
13Ibid., p. 7.
14Ibid., p. 9.
15Ibid., pp. 11-12.
 16Ibid., p. 12. 89
17Ibid., p. 11. Authorship of Monuments
18Charles Fabri, A History of Indian Dress (Calcutta: Orient
Longmans, 1960), p. 1.
19For instance, lions of the pre-Rjasiha style often have
the hair of their mane and head arranged in circular whorls, and their
canine teeth are only moderate in length. But in the lions of Rja-
sihas time, there are no whorls, and the canine teeth are extraordi-
narily long almost half again as long as those of the earlier lions.
20K.V. Soundara Rajan, Rjasihas Temples, Transac-
tions: 1962-65, pp. 173-74.
21Ibid., p. 176.
22A paraphrase of a passage from page 12 of Nagaswamys
article.
23Nagaswamy, p. 12.
24Ibid., p. 14.
25Ibid.
26Whereas with Rjasihas 34 titles given in the Shore
Temple inscription (No. 18A, Vol. XIX, Ep. Ind.), 65% are duplicates
of the Kanchi titles; of his nine titles given in the Vayalur inscription,
67% are duplicates; and of his 16 titles given in the Tiruporur pillar
inscription, 63% are duplicates.
27See the 11th verse of Inscription No. 18, S.-I.I., Vol. I.
28N. Ramesan, Studies in Medieval Deccan History (Late
Pallava and Telugu Chola Period) being Copper Plate Inscriptions of
the State Museum, Vol. III, Archological Series No. 29 (Hyderabad:
The Government of Andhra Pradesh, 1972).
29In his work on the Chir plates, Ramesan at first jumped to
the conclusion that Mahmalla built the Shore Temple complex as we
see it today (see his article, New Light on Shore Temple, The Sunday
Standard, Madras, November 12, 1967). But to try to maintain such a
theory in the face of all the evidence to the contrary would be futile. As
any close study of the Shore Temple complex would reveal, the present
superstructure of the Vishu shrine is obviously of the later Rjasiha
style of architecture. So are the two aivite shrines. But the base of the
Vishu shrine, which together with the image inside is carved out of
the living rock, has a plinth molding which is clearly of a pre-
Rjasiha style. The proper interpretation, then, in the light of the
Chir plates and the architectural and inscriptive evidence at hand, is
that Mahmalla built the original shrine house for the Reclining Vishu
image, and that Rjasiha, in his reign, rebuilt its superstructure and
added two new shrines dedicated to iva. Ramesan accepts this
position in the final publication of the Chir plates, in 1972.
Dharmarja Maapa (in the foreground)
 SEVEN

Mmallapuram Chronology
Part I: The Cave-Temples1
In our previous study, we tackled the problem of the author-
ship of the monuments of Mmallapuram. There, our main aim was to
challenge the claim that all the Pallava monuments at Mmallapuram
were created during the reign of only one king, Rjasiha.
The present study pursues the more positive task of establishing
the chronological order in which the cave-temples were excavated.
One of the important tools needed for this task is a clear understanding
of the various levels of meaning in the stone inscriptions associated
with several monuments at Mmallapuram. These inscriptions provide
an important, though slender, link with the past history of those monu-
ments.
The Sanskrit verses of the inscriptions contain various levels of
meaning. Woven into these verses are many titles (or birudas) which
apply at one level of meaning to the deity and at another, to the king.
The surface or obvious meaning of the poetry is often concerned with
the praise of a god. The suggested or implied meaning (dhvani) of the
very same verses, however, praises the king.
One might be tempted to interpret these passages in terms of
their surface meaning alone. The verses would then be seen as fulsome
praise of some deity. However, I would like to stress the point that it is
the suggested or implied meaning of the verses which is by far the most
important. Any learnd person of the seventh century would clearly
understand this literary convention and appreciate the fact that these
poetical verses are most fundamentally a glorification of the king.
One of the key points in my study will be the claim that the ex-
pression Paramvara in these inscriptions refers to King Param-
vara-I, and is not a title of King Rjasiha.
There are more cave-temples at Mmallapuram than any other type
of monument. But the majority of these cave-temples were never
finished. The more complete ones all have dvraplas (door guardians)
sculpted in relief on either side of the entrance to their sanctums.
Let us list, then, the eight major Mmallapuram cave-temples
which do have dvraplas:
1. Kikal Maapa
2. Dharmarja Maapa
3. divarha cave-temple
4. Varha-II cave-temple
5. Rmnuja Maapa
6. Mahishamardin cave-temple
7. Trimrti cave-temple
8. Kri Maapa
 92 Previous studies of ours on the history and art of the Pallavas
Pallava Art provide the background for my chronological analysis of these monu-
ments in this study.
In the first study of this book, we have drawn a sharp distinction
between a pre-Rjasiha style of Pallava art and Rjasiha style.
This distinction involves differences in dress and ornaments of the
figures portrayed in Pallava sculptural art.
The criteria we used in distinguishing these two styles can be
applied to all of the Pallava monuments at Mmallapuram. When this
is done, we see that the only monuments which exhibit the Rjasiha
style at Mmallapuram are the structural temples (the Olakavara
and Shore temples) and the isolated Smskanda panels, themselves,
in the Mahishamardin cave-temple and the Mukundanayanr temple.2
1. The Mahishamardin Cave-Temple
Of the eight major cave-temples which I have noted, the
Mahishamardin triple-shrined excavation stands out as being the only
one in such an unfinished state. Furthermore, there is the peculiarity
that at least three distinct stages are evident in the work on this cave-
temple.
The most obvious stage is that in which the Reclining Viu
and the Durg panels were done, as well as the details of the small
porch with lion pillars which is in front of the central sanctum.
As we have mentioned above, the Smskanda panel was
introduced into this cave-temple at a definitely later stage. However,
what I wish to suggest here is that there was also a distinct stage of
work prior to the major work on the two panels of the maapa. In a
footnote to our earlier study, we observed a puzzling fact about the
three pairs of dvraplas in this cave-temple: the dvraplas of the
main, central sanctum are noticeably smaller than the dvraplas of the
other two subordinate sanctums. Now, this extraordinary discrepancy
demands an explanation! There is no other example in the whole range
of Pallava cave-temples where, if there is more than one pair of
dvraplas in a given temple, there is a difference in size.3
The explanation I suggest for the difference in size of the
dvraplas of this cave-temple is as follows. The initial excavation,
including the two pairs of dvraplas guarding the two side sanctums,
was carried out in the first stage of work. There was then a distinct
break between this stage and the second stage. In the second stage, the
dvraplas of the central sanctum (which was originally intended for
Viu), the panels of the maapa, and the porch and pillar details
were done. Then, in the third stage, after another break, the iva-Sm-
skanda panel was cut on the back wall of the central sanctum, and the
dvraplas of the central sanctum were altered by sculpting in the clubs,
snakes, horns, and axe-blade details, thus transforming them from
Vaiavite into aivite guardians.
There is even a fourth stage which is evident. Vaiavite
sectarians, at some later date, re-appropriated this cave-temple. There
are signs that they walled up and closed off the central sanctum with its
The Olakavara Temple, above, being used as a lighthouse!
The Mahishamardin Cave-Temple, below (19th century photo)
 94 aivite Smskanda image inside, and then transformed the lion-
Pallava Art pillared porch into a new Viu sanctum. There is a hole cut in the
floor of the porch to hold the base of an image. And an area has been
crudely cut out of the front of the porch, evidently in order to facilitate
the ritual practice of an officiating priest within the new sanctum. In
this fourth stage, Vius emblems, the discus and conch, have been
engraved prominently on the walls of this cave-temple, signifying its re-
appropriation by these Vaiavite sectarians.
What I wish to emphasize here, and to offer as an explanation
of the difference in the dvraplas size, is the claim that when there
was a break at the end of the first stage of work on the Mahishamardin
cave-temple, the artisans who took up the work in the second stage
were never really interested in finishing the cave-temple along the lines
of whatever the original design might have been. Nor were they partic-
ularly concerned with achieving a really finished monument of their
own design somehow rationally superimposed on the work already
done during the first stage. For all we can say, the two side sanctums
may have been abandoned after the first stage. I suggest that the
workers of the second stage did as much as they ever intended to do on
this cave-temple. Thus, the question of harmonizing the sizes of the
dvraplas may not have even entered the minds of the artisans. In
spite of this somewhat casual attitude towards the overall design of the
temple, the artisans of the second stage produced two of the most
famous examples of South Indian art the Reclining Viu and the
Mahishamardin panels.
The workers of the third stage were even less interested in the
overall design or finish of the Mahishamardin cave-temple. They were
intent merely on transforming the main, central sanctum into a shrine
for iva and his family. They accomplished this transformation by
carving the huge Smskanda on the back wall of the central sanctum
(the other two sanctums are without any carving inside) and by carving
appropriate aivite insignia on the already existing Vaiavite-type
dvraplas guarding this main sanctum.
2. Pallava Smskanda
In an earlier study on Pallava Smskanda, we analyzed the
more than forty examples of this theme found in sculpted stone panels
of the Pallava period.5 We divided them into two categories according
to style: those of pre-Rjasiha style and others of Rjasiha style.
There is, in fact, only one known pre-Rjasiha style Smskanda
panel which has survived intact. It is found on the back wall of the
third-level sanctum of the Dharmarja Ratha, Mmallapuram. There is
one other pre-Rjasiha style Smskanda panel which, though most
of it has been destroyed and leveled off, still can be identified as such
from the remaining outline of its figures. This destroyed pre-Rjasiha
style Smskanda is found on the back wall of what was once the
central sanctum of the three-celled Rmnuja Maapa cave-temple.
Of the Rjasiha-style Smskanda panels, there are more than
forty remaining examples from the Pallava period.
 As we have explained in previous studies, we have given the 95
name Rjasiha-style to Smskanda images of a certain type be- Chronology Part I:
cause that style of Smskanda is uniformly and prolifically found in The Cave-Temples
the well-identified temples belonging to the Pallava king Rjasiha. It
may seem paradoxical, but what we have called the Rjasiha-style
appears to me to have been initiated late in the reign of King Param-
vara-I (the father of King Rjasiha).
Two earlier studies of ours provide the basis for this assertion.
The first, Pallava Gagdhara, establishes the fact that the Pallava
king Mahndravarm-I created in his cave-temple near the summit of
the Rock-Fort Hill, Tiruchi, an image of iva-Gagdhara which was
also at the same time a portrait or representation of himself, the king.
The other study, God/King Images and Cult Worship, shows
that this god-king synthesis in Pallava art was continued in the
Smskanda images.
There are, in the inscriptions of Rjasiha, the well-known
poetical comparisons between his father, King Paramvara, and iva
(Lord Paramvara), and between himself and Skanda. The Sm-
skanda image, then, at its inception was peculiarly appropriate to King
Paramvara when he was reigning, and Rjasiha, a baby prince.
A consideration of the evolution of the Smskanda image in
our study on God/King Images substantiated the view that the Sm-
skanda image originated in the reign of Paramvara-I. In several
respects, the Smskanda images in Rjasihas temples reveal an
advanced stage of formalization. For instance, (i) the small size of the
Smskanda panel in relation to the size of the back wall of the sanc-
tum on which it is placed; (ii) the abnormally exaggerated size dif-
ferences between the principal (male) figure of iva (large), on the one
hand, and the subordinate (female) figure of Um (small), on the other;
and (iii) the Smskanda panels being raised a significant distance
above the level of the sanctums floor all of these characteristics are
typical of the Smskanda panels in the sanctums of those temples
which are indisputably credited to King Rjasiha.
On the other hand, the Smskanda panels of the Mahisha-
mardin cave-temple and the Vdagirvara temple at Tirukkaluk-
kunram have the characteristics which could be interpreted as typify-
ing an earlier date: (i) the panels tend to fill the entire back wall of the
sanctum; (ii) the relative sizes of iva and Um are much closer to
those of actual human males and females, and (iii) the panels begin
near the floor level.
We suggested tentatively in the earlier study, therefore, that the
Mahishamardin and Vdagirvara Smskanda panels belong to the
reign of King Paramvara-I.
To whose reign, then, can we assign the two pre-Rjasiha
style Smskanda panels (the intact one of the Dharmarja Ratha and
the destroyed one of the Rmnuja Maapa)? On the basis of my
research in the God/King Images study, the parallelism between King
Paramvara and Lord Paramvara (iva) and between Prince Rja-
siha and the child Skanda appears so strong and so specific to these
96

15

10

Atiraachavara Cave-Temple
persons, that I am compelled to conclude that these two pre-Rjasiha 97
style Smskanda panels must also be credited to the reign of King Chronology Part I:
Paramvara-I, and not earlier. The Cave-Temples
The line of development which seems to emerge is as follows.
At the end of the reign of Mahmalla (Narasiha-I), the artisans con-
tinued the Mahmalla style (pre-Rjasiha style) over the very short
reign (of about three years) of his son, Mahndra-II, and into the begin-
ning of the reign of Paramvara-I. We know from various sources that
during the reign of Paramvara-I, the Pallava kingdom was thrown into
confusion by enemy attacks and that probably it suffered several years
of famine and utter disorder.4 We may suppose then that it was when
Paramvara managed to restore his rule from Kanchipuram that a new
group of artisans was employed and the so-called Rjasiha style was
actually initiated. This style was continued by King Rjasiha through-
out his reign. In fact, many of the characteristics of the Rjasiha style
Smskanda are found in the Smskandas belonging to the later reign
of Nandivarm-II (Pallavamalla).
On my interpretation, then, the most dramatic break in the
continuity of Pallava art style over the two centuries of its greatest glory
(the seventh and eighth) occurred sometime during the reign of
Paramvara-I.
Thus, the style of sculptural art during the early part of
Paramvaras reign would be included by me within the style of King
Mahmalla (the pre-Rjasiha style group); and the style of sculpture
during the latter part of Paramvaras reign I would include within the
Rjasiha-style group.
3. The Atiraachavara Cave-Temple
Very near Mmallapuram, at a place called Saluvankuppam,
there is a Pallava cave-temple called the temple of Atiraachavara.
There seems to be a general consensus among scholars over the years
though not complete agreement that this cave-temple was created by
King Rjasiha.5
I wish to suggest, however, that Rjasiha had little to do with
the creation of this cave-temple.
The two foundation inscriptions (virtually the same verses in
each inscription, but in two different scripts) located on the southern
and northern walls in front of the maapa must be assigned to Param-
vara-I (for reasons I shall set forth in detail later in this study).
There are three Smskanda panels found on the walls of the
Atiraachavara. These carvings were probably done late in Rja-
sihas reign or in the post-Rjasiha period. Here are my reasons.
There are twenty-nine Pallava Smskanda panels in Rjasihas
Kailsantha temple, Kanchipuram; there are two Smskanda panels
in his Shore Temple, Mmallapuram; and there is one Smskanda in
his Tlagirvara temple, Panamalai this comes to a total of thirty-
two Smskanda panels attributable to Rjasiha in the three temples
assigned to him on indisputable grounds. In Pallava Smskanda panels
of the Rjasiha-style, there is above Ums head a royal parasol.
 98 which has a garland hanging from its center, vertically downwards.
Pallava Art (This garland may be mistaken by some observers for the umbrellas
handle, but it is not.) In the many Smskanda panels of Rjasihas
three major temples, the garland always falls to the proper right of
Ums head. However, in the three Atiraachavara Smskanda
panels, the garland falls to the proper left (main sanctums panel) or is
above Ums head (the two porch panels).

10

6.5

Smskanda in the Atiraachavara sanctum

10

6.5

Smskanda, Atiraachavara porch

 The three early Rjasiha-style Smskanda panels which I 99
have tentatively attributed to Rjasihas father, Paramvara-I (that is, Chronology Part I:
the panel in the Mahishamardin cave-temple and the two in the Vda- The Cave-Temples
girvara temple), all have the umbrellas garland hanging to the proper
right of Ums head.
But what about the post-Rjasiha period? In the Muktvara
panel, the umbrellas garland is carved directly above Ums head. In
the Mtagvara panel, it is to the proper left of her head.
Tenuous as all these comparisons of garland positions may
seem, yet they are at least something positive by which one could
guess at the chronological ordering of the Smskanda panels. A more
exhaustive comparison of all the known Smskanda panels, perhaps
involving a numerical taxonomic study of the various proportions of the
figures in the panels, might either confirm or disconfirm my hypothesis.
In this study, then, I shall proceed on the assumption that the
three Smskanda panels of the Atiraachavara cave-temple were
carved in a period definitely later than the reign of Paramvara-I.
Now, the two inscriptions (which are, in the main, identical
verses, but in two different scripts) do refer rather pointedly in their
fifth lka to the Smskanda group of deities:
May Paupati (iva), together with the Daughter of the Mountain
(Prvat/Um), Guha (Skanda), and his retinue of gaas, always be
happy here (in this temple).
I suggest that King Paramvara appropriated this Mahndra-
style cave-temple (which may have been lying in an unused or un-
finished state) and brought it near to its present state of completion and
caused a painting of the Smskanda group to be executed on the back
wall of the sanctum. Then, sometime afterwards, perhaps late in Rja-
sihas reign, but more probably in the post-Rjasiha period, the
painted Smskanda was transformed into a carved bas-relief (painted)
Smskanda panel. The two porch Smskandas were also carved at
the same time.
The two dvraplas of the Atiraachavara cave-temple are
definitely Rjasiha-style, what with their torso-twisted stance. But
even accepting this fact of style, there is still the question where exactly
these dvraplas should be placed: (1) in the late Paramvara period,
(2) in Rjasihas reign, or (3) in the post-Rjasiha period?
4. Concordance of Verses and Royal Titles
Let me turn then to an analysis of the foundation inscriptions of
the Atiraachavara cave-temple. These two inscriptions must be
studied alongside several inscriptions found at Mmallapuram, which
share some of the same verses (word for word). Perhaps the clearest
way of presenting these different inscriptions and of indicating the
degree of concordance among them is through the following type of
arrangement:
 Concordance of King Paramvaras Inscriptions

Gaa Ratha & Dharmarja Maapa Inscriptions

1. May he (iva) who destroyed (Kma) the God of Desires (nevertheless) be the fulfiller of the
countless desires of mankind he (iva) who is the cause of (all) creation, preservation, and
destruction, (but is) himself uncaused.
2. May he be victorious! he who is immutable (amya), (and yet) the ground of all transient existence
(Citramya); who is without qualities (agua), (and yet) the receptacle of (all) qualities (Guabhja-
na); who is self-dependent (svastha), (and yet) without superior (Niruttara); who is without any lord
(ana), (and yet is himself) the Supreme Lord (Paramvara)!
3. The weight of (ivas) big toe was enough to plunge (Mount) Kailsa together with the Ten-faced
(Rvaa) down to the underworld, (and yet) rnidhi (the king) bears that Unborn (iva) on his
head!

4. May (he) be victorious always, that rbhara (the king) who so easily bears Bhava (iva) in his
mind which is filled with devotion, and who bears the burden of (ruling) the earth as lightly as a mere
__ ornament on his arm.

5. This temple of ambhu (iva) was caused to be made by King Atyantakma, conqueror of his
enemies territory and renowned by the title Raajaya!
6. May he be victorious! (he) who is unmoving (Sthu), (yet) aware of everything (ja); who is
fiery souled (Pvaktma), (yet) whose body is (infinite) space (viyadvapu); who is fearsome (Bhma),
(yet) auspicious (iva); who is the Destroyer of Desire (Kmasdana), (yet) the Comforter
(akara).
7. May (King) Tarukura be victorious! (he) who is Rjarja (King of Kings a title also of
the god, Kubra), (yet) not uncultured (virasa as is Kubra); who is Cakrabt (Emperor , also a
title of Viu), (yet) not Janrddana (Vius title, here punned on suggesting Torturer of Mankind);
who is Trakdhipati (a title of the moon god), (yet in his supremacy ) completely sound (svastha
unlike the moon which waxes and wanes).
8. This lord of wealth (rman, Lord of the goddess r) and of unlimited desire (Atyantakma), who
strips his enemies of their pride (Dviaddarpppahri), who is the Storehouse of Prosperity
(rnidhi), who possesses the charm of the god of love (Kmarga), worships Hara (iva) ardently
(Harrdhanasagin).
9. In the lofty head-lake (i.e., the anointed head of the king), full of the water of coronation, a mine of
__ multi-colored jewel-lotuses, the handsome-faced akara (god iva) is manifest.

10. This lofty temple of Dhrjjai (iva) was caused to be made by him (the king) who was desirous of
attaining the eight-fold treasure of akara (iva) and of (thus) providing his subjects with all their
desires.
[The following verse is also found in the Rmnuja Maapa & divarha Cave-Temple:]
11. Cursed be those, cursed be those, and again cursed be those, cursed, cursed, cursed be those in whose
hearts does not dwell Rudra (iva), the deliverer from treading the evil path.

The temple of Atyantakma Pallavvara.6

 Concordance of King Paramvaras Inscriptions
Atiraacavara Inscription (all seven verses & titles): South Wall

This framed portion (just six verses) is also found on the North Wall
1. This lord of wealth (rman, Lord of the goddess r) and of unlimited desire (Atyantakma), who
strips his enemies of their pride (Dviaddarpppahri), who is the Storehouse of Prosperity
(rnidhi), who possesses the charm of the god of love (Kmarga), worships Hara (iva) ardently
(Harrdhanasagin).

2. In the lofty head-lake (i.e., the anointed head of the king), full of the water of coronation, a mine of
multi-colored jewel-lotuses, the handsome-faced akara (god iva) is manifest.
3. For the welfare of this earth, he, who is foremost among the rulers of the world, caused to be made
this temple of ambhu (iva) which shines like the Kailsa (mountain) peak.
4. May (he) be victorious always, that rbhara (the king) who so easily bears Bhava (iva) in his
mind which is filled with devotion, and who bears the burden of (ruling) the earth as lightly as a mere
ornament on his arm.
5. Atiraacaa, the lord of the rulers of the earth (Avanibhjmpati), made this temple (called)
Atiraacavara. May Paupati (iva), together with the Daughter of the Mountain (Um), Guha
(Skanda), and his retinue of gaas, always be happy here.

6. May the eight-formed Lord of animate beings abide eternally in this (temple called) Atiraaca-
vara which was made by him who possesses along with the title of Atiraacaa a deep devotion to
na (iva), and (also) the heavy burden of (ruling) the earth, an extraordinary liberality (to the
needy), and the widely famed title of Raajaya!

(He is one) who is inclined to be gentle (Anugrala).

7. Except for Vidht (Brahm), Bharata, Hari, Nrada, and Skanda, who is there who can understand
the music of Klakla (the king)?
The Arjuna in War (Samaradhanajaya); who is brave in battle (Sagrmadhra).

_______________

*Please note that verse 4 (shaded) is the same in the Gaa Ratha, Dharmarja Maapa, and
Atiraacavara Inscriptions. And note that verses 8 and 9 (shaded) of the former two are the same as
verses 1 and 2 (shaded) of the Atiraacavara Inscription.

**The titles rman and rnidhi are also appropriate to the god Viu. And Atyantakma
and Kmarga are suggestive of the god of love. By the clever device of dhvani, the poet appears to make
Viu and Kma the ardent worshippers of iva. In this context, consider the eleventh verse of the inscrip-
tion opposite which upholds the worship of iva. Note that this imprecatory verse is also found on the floor
of the divarha cave-temple a temple dedicated to Viu in his Varha avatra. This verse bespeaks a
clear attempt to subordinate Viu to iva.
 102 The above concordance deals with inscriptions found in five
Pallava Art different monuments:
1. Atiraachavara cave-temple;
2. Gaa Ratha;
3. Dharmarja Maapa;
4. Rmnuja Maapa; and
5. divarha cave-temple.
The inscription of the Dharmarja Maapa is identical with
the inscription of the Gaa Ratha. There is not only an agreement
here, verse for verse and word for word, but also the form of the script
used in both inscriptions is identical. Now, three of the lkas in these
two inscriptions are also identical to three lkas in the Atiraacha-
vara inscriptions although their order has been altered in the case of
two of these lkas. The first and second lkas of the Atiraacha-
vara inscriptions are the eighth and ninth lkas of the other inscrip-
tions.
Now, I wish to claim that all of these inscriptions belong to
Paramvara-I.
There are others who would assign some or all of these in-
scriptions to Rjasiha. The main reason given for their doing so is
usually the fact that several of the royal titles (birudas) appearing in
these inscriptions are also titles applied to King Rjasiha in his
Kailsantha temple inscriptions. Let us look into this matter more
closely.
The collection of Rjasihas titles found in his Kailsantha
temple is perhaps the largest single collection of royal titles in India.
The full list is given in Chapter Sixteen. More than 250 different titles
of his are inscribed on the little shrines which form the enclosure of the
main temple.
Of the 13 royal titles found in the Atiraachavara inscrip-
tion, six of them are common to the Kailsantha also (that is, 46%):
1. Atyantakma (1-2) (niche & place no., Kailsa.)
2. Dviaddarpppahri
3. Srnidhi
4. Kmarga
5. Harrdhanasagin
6. rbhara (3-3)
7. Atiraacaa (3-2)
8. Avanibhujm pati
9. Raajaya (1-3)
10. Anugrala (48-4)
11. Klakla (front shrine, 3rd to right, applied to Paramvara-I)
12. Samaradhanajaya (20-1)
13. Sagrmadhra
 Next, let us consider the twenty royal titles found in the verses 103
which are common to the inscriptions of the Gaa Ratha and the Chronology Part I:
Dharmarja Maapa. Four out of their twenty titles are common with The Cave-Temples
Rjasihas titles in the Kailsantha temple (that is, 20%):
1. Citramya
2. Guabhjana
3. Niruttara
4. Paramvara (front shrine, 3rd to r., applied to Paramvara-I)
5. rnidhi
6. rbhara (3-3)
7. Atyantakma (1-2)
8. Raajaya (1-3)
9. Sthu
10. Sma
11. Pvaktma
12. Bhma
13. Kmasdana
14. Tarukura
15. Rjarja (13-4)
16. Cakrabht
17. Trakdhipati
18. Dviaddarpppahri
19. Kmarga
20. Harrdhanasagin
We have noted that 46% of the royal titles found in the
Atiraachavara inscription are also Rjasihas titles in his
Kailsantha inscriptions. And we have noted that 20% of the titles
found in the Gaa Ratha and Dharmarja Maapa inscriptions are
also Rjasihas titles in his Kailsantha inscriptions.
Now, when we make a similar comparison between inscriptions
elsewhere (definitely known to belong to Rjasiha) and the more than
250 different royal titles found in the Kailsantha temple, we get
significantly higher percentages.
Let us consider first Rjasihas Vayalur inscription. Six out of
its nine titles are also found in the Kailsantha temple (that is, 67%):
1. Rjasiha (1-1) (niche & place no., Kailsantha)
2. Katriyasiha
3. Yuddhrjjuna (15-4)
4. Narndrasiha
5. Atyantakma (1-2)
6. rmgha (5-1)
7. Mahmalla (23-2)
8. Raajaya (1-3)
9. rnidhi
 104 Next, consider Rjasihas Shore Temple inscription.
Pallava Art Twenty-four out of its thirty-four titles are also found in the
Kailsantha temple (that is, 71%):
1. Apratima (29-1)
2. Avanibhaa
3. Akalaka (modified: 9-1; 20-6)
4. Dharaicandra
5. Arimarddana (5-4)
6. Atulabala (28-1)
7. Kulatilaka (5-3)
8. Bhayarahita (?) (23-1) or Chalarahita (10-4)
9. Bahunaya (3-4)
10. Atyantakma (1-2)
11. Aparjita (2-1)
12. karja (on Rjasihvara Kailsantha)
13. Candrrdhakharaikhmai
14. Adbhuta (modified: 11-3; 29/30-3)
15. Caani (modified: 11-1)
16. Udayacandra (12-3) (niche & place no., Kailsa.)
17. Rjasiha (1-1)
18. Raajaya (1-3)
19. rbhara (3-3)
20. Citrakrmmuka (14-2)
21. kavra (on Rjasihvara Kailsantha)
22. ivacmai (on Rjasihvara Kailsantha)
23. Kmuka (modified: 14-4)
24. Klakla (front shrine, 3rd to right, applied to Paramvara-I)
25. Abhirma (1-4)
26. Raabhma
27. Gulaya (33-1)
28. r-vallabha (16-1)
29. Atimna
30. Raavra (26-3)
31. rjjita (2-4)
32. Unnatarma (7-3)
33. Yuddhrjjuna (15-4)
34. Narndrasiha
Rjasihas inscription on the Tlagirvara temple at
Panamalai (at least, the portions which are exposed) has only one title:
Rjasiha. So, it hardly affords us a comparison. However, one out
of one is 100%.
 Finally, there is the Tirupporur Pillar Inscription which is usually 105
taken as Rjasihas. There are sixteen royal titles in this inscription. Chronology Part I:
Ten out of the sixteen are ones which are also found in the Kailsantha The Cave-Temples
inscriptions (that is, 63%):
1. Kmalalita
2. Kulatilaka (5-3) (niche & place no., Kailsa.)
3. Guavinta (20-4)
4. Dharaitilaka (51-4)
5. Jnasgara
6. Tribhvanadpa
7. Aviratadna (18-3)
8. naaraa (12-2)
9. Pthivsra
10. Samaradhanajaya (20-1)
11. Atyantakma (1-2)
12. Abhayakara (5-2)
13. Avanidivkara (20-5)
14. Atiraacaa (3-2)
15. Avritavryya
16. Arikariksar
To summarize the concordance of the royal titles in this last
group of inscriptions with Rjasihas many titles found at the
Kailsantha temple:
1. The Vayalur inscription yields a concordance of 67%.
2. The Shore Temple inscription yields 71%.
3. The Tirupporur pillar, 63%.
These percentages, as I have said, are significantly higher than
what is the case with the Atiraachavara inscription (46%) and
with the Gaa Ratha and Dharmarja Maapa inscription (20%).
The evidence along this line of investigation, then, would
indicate that the inscriptions of the Atiraachavara, Gaa Ratha,
and Dharmarja Maapa are more likely to be Paramvaras than
Rjasihas. But there are further grounds for assigning these inscrip-
tions to Paramvara-I.
Early scholars dealing with the Gaa Ratha and Dharmarja
Maapa inscriptions held that the word Paramvara which appears
in the second lka has a double meaning. In its primary reference, the
term Paramvara refers to iva, who is being praised as the Supreme
Lord of the universe. In its secondary reference, however, it is a play
on the name of the king, himself, Paramvara-I.
More recently (1962), however, R. Nagaswamy, discussing the
same lka, has denied that the reference is to King Paramvara-I, and
instead he attributes the title Paramvara in this inscription to King
Rjasiha. Thus, Nagaswamy believes that the author of the Gaa
Ratha and Dharmarja Maapa inscriptions was Rjasiha. Speaking
of the first two lkas of these inscriptions, he has this to say:
 106 . . . the word Atyantakma is primarily employed to denote bound-
Pallava Art less desires, but also implies a reference to the King Atyantakma
(whose prosperity Siva may fulfil). It is in the same context the
word Paramesvara in the second verse must be taken to refer to
Siva primarily. [But it] also implies a reference to a title of the
King as Paramesvara. [That the] title Paramesvara was borne by
Rajasimha also is seen from his Kanchi inscription as
PARAMESVARA. In [the] Rangapatka inscription [of the
Kailsantha temple, Kanchi,] he is referred to as PARAMES-
VARA. . . . Thus the secondary reference to the title of
Paramesvara in the second verse of the Ganesaratha and the
Dharmarja maapa is only a reference to Rajasimha.7
I cannot agree with Nagaswamys concluding sentence. As
Nagaswamy himself points out, Rjasiha assumes (in the niche of
shrine 31 of the cells surrounding the main temple complex) the title,
Il-Paramvara. The qualification Il- is necessary precisely
because Paramvara by itself would not be appropriate to Rjasiha.
After all, Paramvara was the coronation name (abhika-nma) of
Rjasihas father. It would be very odd within the Indian context for
a royal son to assume his fathers coronation name as one of his own
titles. Therefore, Rjasiha had to add the qualification Il. In San-
skrit, one meaning of Il (or I) is the earth or the world. Thus,
Hultzsch has translated the whole expression (Il-Paramvara) as
the supreme lord of the earth.8
Thus, I do not believe that the Paramvara in the Kanchi-
puram title Il-Paramvara can be taken alone as a proper title of
Rjasihas that is, as a title of his on which could be based the kind
of punning and double entendre which we find in the second lka of
the Gaa Ratha and Dharmarja Maapa inscriptions.
Neither do I agree with Nagaswamy when he voices the gen-
erally held view that the name Paramvara in the so-called Raga-
patk inscription (Kailsantha, Kanchi) refers to King Rjasiha (in
addition to its alternate reference to the god iva). The reasons for my
disagreeing with this view are put forward in the study, Queen
Ragapatks Inscription.
It is my opinion, then, that with regard to the Gaa Ratha and
Dharmarja Maapa inscriptions, the earlier interpretation of scholars
is the correct one: the second lka of the inscriptions does refer to
King Paramvara-I (and not to Rjasiha). These inscriptions, along
with those of the Atiraachavara, therefore, can all be assigned to
Paramvara-I.
I shall mention another fact which would support the view that
these inscriptions all belong to Paramvara, and none to Rjasiha.
That fact is that in the Atiraachavara, Gaa Ratha, Dharmarja
Maapa, Rmnuja Maapa, and divarha cave-temple inscriptions
which we are considering, not one of them has a royal title using a term
meaning lion, nor is there any reference in them whatsoever to lions,
metaphorical or otherwise. The significance of this omission can
perhaps be appreciated when we note that in every inscription which 107
has been positively assigned to Rjasiha, there is always given at least Chronology Part I:
one of his titles which is based on a word meaning lion. Further, there is The Cave-Temples
often praise of the king which employs the metaphor of lion-like bravery.
This kind of thing is to be expected since Narasiha (the name of Vius
Man-Lion avatra) was the coronation name of Rjasiha. In the Kailsa-
ntha temple, Kanchi, these are his lion titles:
1. Rjasiha (the lion among kings) (1-1) (niche & place no.)
2. Puruasiha (the lion among men) (21-4)
3. havaksar (the lion in battle) (8-3)
4. Vraksar (the lion among heroes) (14-3)
5. Vikramaksar (the lion in valor) (57-3)
6. Prtthivasiha (lion among princes) (54-1) (3rd tier down)
In Rjasihas Shore Temple inscription, there are these lion titles:
1. Rjasiha (1-1)
2. Narndrasiha (the lion among rulers of men)
In his Tlagirvara inscription at Panamalai, we find one such title:
1. Rjasiha (1-1)
In his Vayalur inscription, there are three lion titles:
1. Rjasiha (1-1)
2. Narndrasiha
3. Katriyasiha (the lion among warriors)
And, finally, in the Tirupporur Pillar Inscription, we find the following
lion title:
1. Arikariksar (a lion to the elephants, his enemies)
Further, it should be noted that wherever we do have the original names
of Rjasihas temples given in the earliest inscriptions, they are all
lionized:
1. Rjasihvara was the original name of the Kailsantha
temple, Kanchipuram.
2. Rjasihvara, Katriyasihvara and Narapati-
siha-Pallava-Viu-Gham were the names given by
King Rjasiha to the three shrines of the Shore Temple
complex, Mmallapuram.
In concluding these arguments, I must also mention the fact that
the lion pillars, used everywhere in the architecture of Rjasihas
temples, provide simply another device which was ultimately intended
to emphasize the lion-like nature of the king.
Both Rjasiha and his ancestor, Mahmalla, had the same
coronation name: Narasiha. The lion-based pillars were introduced
by Mahmalla (Narasiha-I) and vigorously continued by Rjasiha
(Narasiha-II). Of course, architectural motifs such as the lion pillar
were employed by kings who had no such lionized coronation name.
For instance, Nandivarm Pallavamallas temples make liberal use of
the lion pillars. In passing, I would like to point out that the Gaa
Ratha (one of the few Mmallapuram monuments which seem to
 108 belong to Paramvara from original plan to final execution) does have
Pallava Art two lion pillars (vya pillars), but that the two vya-like pilasters on
either side of the entrance have peculiar faces with bird-like beaks (they
appear to be griffins).
However, with regard to verses praising the king, the situation
is more strict. The punning use of titles, signifying by means of meta-
phor or double entendre the lion-like nature of the king, would not be
appropriate to a king whose coronation name was, for instance,
Paramvara!
It is against this background of the traditional use of lionized
titles and metaphors by King Rjasiha, and the inappropriateness of
such titles and metaphors with regard to Paramvara, that we must see
the significance of their complete omission in the inscriptions which we
have examined in the Atiraachavara, Gaa Ratha, Dharmarja
Maapa, Rmnuja Maapa, and divarha cave-temple.
5. The aivite Curse
If we go along with the traditional view that it was Mahmalla
who introduced the lion-pillars in the architecture of Mmallapuram,
then the divarha cave-temple should be assigned to Mahmalla, as
the lions of this cave-temple and the other sculpture in it are of the early
style. The divarha is a Viu temple, and is still under worship to-
day. Yet, on the floor in front of the sanctum, engraved in large letters,
is the following aivite curse (I give here Hultzschs translation of it):
Six times cursed be those, in whose hearts does not dwell Rudra
(iva), the deliverer from the walking on the evil path.9
This curse, as we have seen, is also found in the Rmnuja
Maapa, and it forms the last verse of the Gaa Ratha and Dharma-
rja Maapa inscriptions inscriptions which pun on the royal-divine
name, Paramvara.
By no stretch of the imagination is it reasonable to suppose
that the creator of the divarha cave-temple, a Viu shrine, would
have engraved such a aivite curse in front of the very sanctum he has
dedicated to the Varha form of Viu! The author of the aivite curse
inscriptions, then, must be someone who came after Mahmalla.
Paramvara-I is well known for his zealous, even exclusive
devotion to iva. And from the evidence we have already given that it
was he who appropriated the once Vaiavite Mahishamardin cave-
temple and who transformed its main, central sanctum into a shrine for
iva-Smskanda, it would seem that Paramvara-I was probably the
author of the aivite curse. In this connection, it must be noted that
nowhere does the aivite curse appear in any of the inscriptions and
temples which are indisputably assigned to King Rjasiha. This
negative fact, therefore, provides additional confirmation that Rja-
siha was not the author of the aivite curse, nor the inscriptions
which contain it.
 6. Review of Major Cave-Temples 109
In the beginning of this study, I gave a list of eight major cave- Chronology Part I
temples at Mmallapuram. These eight are distinguished from the others
at Mmallapuram by being more complete and by having dvraplas
sculpted on either side of the entrances to their sanctums. Lets look again:
1. Kikal Maapa
2. Dharmarja Maapa
3. divarha cave-temple
4. Varha-II cave-temple
5. Rmnuja Maapa
6. Mahishamardin cave-temple
7. Trimrti cave-temple
8. Kri Maapa
The first two cave-temples in this list, the Kikal and the
Dharmarja Maapas, are distinguished from the others by belonging
to the early style so typical of Mahndras cave-temples. This early
style is characterized by a simplicity in the general plan and execution
of the temple. Pillars are massive, with plain square section (except for
the middle third of the pillar which is chamfered to an octagonal sec-
tion). There is usually very little in the way of sculpture sometimes
no figures at all. If any sculptured figures are present, they are almost
always only door guardians.
On purely architectural grounds, these two cave-temples would
be placed in the Mahndra period or even earlier. They have the same
general simplicity in plan and detail. Their pillars are massive and are
typical of the Mahndra type pillar. The only sculpture these two
temples have is a pair of door guardians.
(i) Kikal Maapa
In the case of the Kikal Maapa (a cave-temple which was
dedicated to Durg [Koi]), the two female guardians are more crudely
carved than their counterparts found guarding the Draupad Ratha.
(Rather surprisingly, the Kikal guardians have no leg ornaments an
almost unique omission for females.)
(ii) Dharmarja Maapa
In the case of the Dharmarja Maapa, which has three sanc-
tums, the two dvraplas of the central shrine have been chiseled off,
but their outline remains. There is no sign of any dvraplas for the two
side shrines of this same cave-temple.
Though this cave-temple would, on purely architectural grounds,
be placed in or before the Mahndra period, these architectural consid-
erations, for most scholars, have been completely over-ruled by the
presence of a single inscription. This inscription one we have already
dealt with very clearly states that King Atyantakma caused to be made
this temple for iva. Since most scholars consider the King Atyantakma
of this inscription to be either Paramvara-I or Rjasiha, this cave-
temple is accordingly attributed either to Paramvara or to Rjasiha.
 110 I go along with the view that the inscription in the Dharmarja
Pallava Art Maapa belongs to Paramvara-I, but I do not think that this inscrip-
tion provides conclusive proof that Paramvara was responsible for the
excavation of this cave-temple. Instead, I believe that this cave-temple
existed prior to the time of Paramvara, and that its main, central sanc-
tum was originally intended for Viu. What suggests this to me is the
character of the two dvraplas of the main shrine which have been
chiseled off. The remaining outlines of these two dvraplas show us
that they are not the usual aivite type of dvraplas. They have no
clubs. Their hair-style (judged by the outline) is moderate. And their
general pose and slender appearance is counter to what we would
expect in the case of guardians of Pallava aivite shrines. Finally, we
should note that these dvraplas were facing the observer, standing in
relatively spacious niches an early, Mahndra-period characteristic.
Now, this kind of observation about the character of the
dvraplas is merely suggestive, and I realize that it cannot, in itself,
settle the issue about the origin and development of this cave-temple.
Is there any other evidence, then, which could support my view that in
consecrating the Dharmarja Maapa to iva, Paramvara had
appropriated a cave-temple already started and probably fully estab-
lished by a predecessor of his?
Speaking generally, it can be said that the more temples we
discover at Mmallapuram which show signs of having been appropri-
ated, the more we would, perhaps, be willing to suspect such a thing
with regard to the Dharmarja Maapa. (In passing, it should be
noted that the Dharmarja Maapa has Vius emblems, the discus
and conch, engraved on its walls by Vaiavite sectarians who, thus,
signified its re-appropriation by them.)
But more specific evidence of Paramvaras having appropri-
ated the Dharmarja Maapa is to be discovered in his foundation
inscription in this same cave-temple. First, the fact that this inscription
is an exact duplicate of the foundation inscription of the Gaa Ratha
should make one stop and think. If a king had actually been completely
responsible for two such different types of monuments, why would he
repeat his foundation inscription word for word in both places even
going to the extent of giving both temples the same name: Atyanta-
kma-Pallavvara-Gham? Rendered into English, this name means
the temple of (iva) the Lord of the Pallava (king,) Atyantakma.
While I am touching on this point, I must also emphasize the
fact that the third-level sanctum of the Dharmarja Ratha also bears the
label inscription, Atyantakma-Pallavvara-Gham! And in this
particular case, we have an obvious example of an appropriation by this
king, Atyantakma, of a sanctum in a monument which most certainly
was started by his predecessor. (The Dharmarja Ratha, of course, is
still very much unfinished with regard to its overall design.) For those
of us who hold that the king, Atyantakma, of these inscriptions was
Paramvara-I, the inscribed label on the third-level sanctum of the
Dharmarja Ratha is prime evidence of his appropriative tendencies.
 Furthermore, it is quite extraordinary that three different 111
temples in the same place (and within a few hundred meters of each Chronology Part I:
other) should bear exactly the same name with respect to the same The Cave-Temples
king. Appropriation by King Atyantakma is almost certain in the case
of the Dharmarja Rathas upper shrine. Is it not probable, then, that
the unusual repetition of both the Gaa Rathas inscription and name
on the Dharmarja Maapa occurs because (like the Rathas shrine)
the Maapas shrine was also merely an appropriation?
It is true that, in the case of Rjasiha, there are two temples
built by him which were both originally named Rjasihvara. But
one of them is in Kanchipuram (the Kailsantha temple), whereas the
other is the west-facing shrine of the Shore Temple complex, Mmalla-
puram. And when it came to naming the three shrines of the latter,
Rjasiha used three different variations of his own titles:
1. Rjasihvara (the west-facing shrine)
2. Katriyasihvara (eastern shrine)
3. Narapatisiha-Pallava-Viu-Gham (central shrine)
To return to the Dharmarja Maapa inscription, there is
another peculiarity in it. It gives us no information concerning the fact
that this cave-temple has three sanctums. Thus, the same inscription
and temple name have been applied, on the one hand, to a monolithic
temple with a single sanctum (the Gaa Ratha), and, on the other
hand, to a cave-temple with three sanctums (Dharmarja Maapa).
In contrast, the Mandagappattu inscription of Mahndra-I was
perfectly clear in its reference to the three separate sanctums of that
cave-temple:
[This temple (ayatana)] was caused to be made by King Vicitra-
citta for Brahm, vara, and Viu.
There is even a third oddity of the Dharmarja Maapa inscrip-
tion. As we have said before, it is an exact duplicate of the inscription
in the Gaa Ratha. Now, the tenth verse of these inscriptions reads
(in part) in translation:
He (the king) . . . caused to be made this lofty dwelling of Drjai
(iva) in order to procure the fulfilment of their desires to his subjects.10
The term lofty may, with poetic license, be applied reason-
ably to such a monument as the Gaa Ratha. But, when we consider
the Dharmarja Maapa, the adjective lofty seems positively absurd.
It is a cave-temple and one with a none too high ceiling!
Let me summarize, then, the three peculiarities of the inscrip-
tion in the Dharmarja Maapa:
(1) it is an exact duplicate of the Gaa Ratha inscription, even
repeating the same name, Atyantakma-Pallavvara-Gham;
(2) it in no way acknowledges the fact that there are three sanc-
tums in the Dharmarja Maapa (contra the Gaa Rathas
single sanctum);
(3) it repeats the term lofty with respect to the Dharmarja
Maapa, a cave-temple which is not at all lofty!
 112 These oddities can be explained if we understand that this King
Pallava Art Atyantakma (Paramvara-I) was not responsible for the creation
of the cave-temple, but that he summarily appropriated the Dharmarja
Maapa and dedicated it anew to iva. Such an act of appropriation
did not call forth the originality and care in framing an inscription of
re-dedication as would be the case if it had been an original dedication
climaxing the entire creation of the cave-temple.
Thus, for the various reasons I have outlined above, I feel that
the Dharmarja Maapa must be dated on the basis of its architectural
features. And so I would place both the Kikal Maapa and the
Dharmarja Maapa in the Mahndra or even pre-Mahndra period.
At present, I know of no way we could positively assign them to
Mahndra instead of some earlier king. Therefore, I only conclude by
assigning these two cave-temples the earliest relative position in the
chronology of Mmallapurams monuments.
We next turn to the four cave-temples which are the classic
examples of the Mahmalla style:

Finished monuments:
divarha cave-temple
Varha-II cave-temple
Rmnuja Maapa (cave-temple)

Unfinished monument (multi-stage):

Mahishamardin cave-temple

In the Mahmalla style, we may first mention that the pillars

are slender, and have eight- or sixteen-sided shafts, with elegant orna-
mentation. And their most outstanding feature is present when the
base of the pillar is carved in the form of a seated lion or seated vya.
Secondly, the general details and decoration of the Mahmalla style
cave-temples are far more elaborate than those of the Mahndra style.
Thirdly, the walls of the maapas of these cave-temples have been
transformed into impressive sculptured panels depicting gods and
goddesses in traditional scenes or illustrations of episodes from Hindu
scriptures.
While touching on the subject of the great sculptured panels
found on the maapa walls of Mahmallas cave-temples, it should be
noted that of them not one single major panel deals with the god iva!
Instead, they all deal with Viu or the two goddesses, Lakm and
Mahiamardin (Durg [Jayalakm]). In fact, in the maapa panels,
iva appears in his own right only in the minor niche found on the
northern wall of the divarha cave-temples maapa. In this niche,
iva is portrayed as Gagdhara.

(iii) divarha Cave-Temple

One other appearance (or half-appearance) of iva is in the
Harihara figure (half iva, half Viu) found in the same cave-temple 113
(the divarha). Chronology Part I
Both the Gagdhara and Harihara figures are in narrow niches,
and cannot be considered major maapa panels when compared to the
Gajalakm and Durg panels of the divarha cave-temple, or the
large panels of the other cave-temples.
Now, the mere appearance of iva in a Viu temple is remark-
able.11 Of course, it is true that their joint portrayal was started earlier
in the famous Viu cave-temples of Badami. And it is true that
Mahndra had established several triple-celled cave-temples dedicated
to the Trimrtis (Brahm, Viu, and iva). But in the divarha cave-
temple, there is only one sanctum, and that one is dedicated to the
Varha form of Viu. The images of iva-Gagdhara and Harihara
are subordinate images, outside of the sanctum. On the southern wall,
directly opposite, and facing, the Gagdhara image, there is a figure of
Brahm. Thus, in a sense, we do have the Trimrtis in the divarha
cave-temple, but Brahm and iva are clearly subordinate, in that they
do not have sanctums of their own.
I should also add that in the Durg panel of this same temple,
there is depicted behind Durg (a little to her right) a tall standard with
the trila (trident) emblem of iva at its top.
Because of this admixture of aivite images and emblems in
the divarha cave-temple, I would consider this to be the earliest of
the Mahmalla style cave-temples. It is certainly nearest in spirit to
the earlier Chlukyan examples and to the inclusiveness of Mahndras
triple-celled cave-temples dedicated to the Trimrtis.
(iv) Varha-II Cave-Temple
Almost immediately, however, the Viu temples of the Pallavas
were to drop the practice of showing anything aivite. Thus, the fact
that in the Varha-II cave-temple at Mmallapuram, no trident or other
aivite emblem is shown in its Durg panel, is an indication to me that
this temple is later than the divarha. Though the Varha-II cave-
temple does have a small image of iva in its Trivikrama panel, yet
iva is shown in diminutive size when compared with the Gagdhara
and Harihara figures of the divarha cave-temple.
It may be of interest to note that Vius emblems, the discus
and conch, are found engraved on the faade sides of the Varha-II
cave-temple.
(v) Rmnuja Maapa
We turn next to the Rmnuja Maapa, a cave-temple which
has, unfortunately, been radically altered. Its maapa panels have
been chiseled away, leaving only outlines of the figures. The front and
separating walls of what were once three sanctums have all been
excavated away. The iva-Smskanda panel on the back wall of what
was once the central sanctum has also been chiseled off, so that only an
outline of the figures remains.
 114 From the outline of the figures on the southern wall of this
Pallava Art cave-temples maapa, we can easily identify a Durg panel which is
very similar to the one in the divarha cave-temple.
Unfortunately, the outline of the figures on the northern wall of
the maapa is not enough of a clue to identify that panel.
King Atyantakmas (Paramvaras) aivite curse is found on
the floor of this cave-temple, in front of what was once the central sanc-
tum. Thus, there is a parallel here with the manner in which the same
aivite curse is engraved on the floor of the divarha Viu cave-
temple. The paleography of the two floor imprecatory inscriptions is
practically identical. The size of the letters in both cases is large, and
the engraving deep.
From this parallelism, I would judge that the central sanctum
of this cave-temple was originally dedicated to Viu, and that Param-
vara-I appropriated it and had the Smskanda panel carved on the
back wall of the central sanctum. (It is significant, in this connection,
that there is no trace of any carvings on the back walls of the other two
sanctums.) Vius emblems, the discus and conch, are engraved on
the walls of this temple, signifying the re-appropriation of it at a later
date by Vaiavite sectarians.
From all indications (including the early type lion pillars), the
Rmnuja Maapa is a cave-temple belonging originally to Mah-
mallas time. But we have tried to show that the Smskanda image
was a creation of Paramvara-I. Hence, our conclusion is that the
Smskanda panel in this Maapa was a later addition to this cave-
temple, which transformed the central sanctum into a aivite shrine.
(vi) Mahishamardin Cave-Temple
Finally, in concluding my survey of this group of the Mah-
malla style cave-temples which have maapas, I shall consider again,
briefly, the Mahishamardin cave-temple. As I have already pointed out,
it stands apart from the other three in revealing a very erratic develop-
ment and an unfinished appearance. Our previous studies suggested
that during the Mahmalla period, the main sanctum of this cave-temple
was intended for Viu, but that later, King Paramvara-I transformed
it into a aivite shrine and had a Smskanda panel carved on its back
wall. At the same time, aivite emblems and weapons were added to
the dvraplas of the main sanctum in an obvious attempt to give them
a aivite appearance.
Our discussion so far, concerning the development of cave-
temples at Mmallapuram, would indicate that Vaiavism was domi-
nant in them throughout the period of Mahmallas reign. However, in
Paramvaras reign there was a vigorous completion or conversion
of these earlier temples into aivite shrines, and only a few new monu-
ments were created (the Gaa Ratha being the foremost example).
There is evidence that the Pallava kingdom suffered invasion by
enemy forces during the successive reigns of Mahmalla, Mahndra-II,
and Paramvara-I. The Chlukyas of Badami were long-standing
enemies of the Pallavas. In the Gadval copper-plate grant (dated A.D. 115
674) of the Chlukyan king, Vikramditya, it is stated that this king Chronology Part I:
invaded the Pallava capital and crushed the glory of Narasiha (Mah- The Cave-Temples
malla), caused the dissolution of the valor of Mahndra (Mahndra-II),
and subdued vara (Paramvara-I) with his eyes.12 It is possible that
an invasion disrupted the temple-building going on at Mmallapuram.
It is even likely that some of the temples were damaged by the enemy.
There is plenty of evidence of deliberate destruction of the shrines at
Mmallapuram. Thus, Paramvara might have had on his hands sev-
eral abandoned monuments at Mmallapuram. His completion or con-
version of these monuments could be viewed within such a context.
Nevertheless, it is important to keep in mind that there does
seem to have been a significant degree of resentment on the part of the
Vaiavite community at Mmallapuram, which was later to reassert its
claims to these monuments. For instance, the discus and conch, em-
blems of Viu, are engraved indicating such re-appropriation on all
of the following cave-temples:
1. Dharmarja Maapa (cave-temple)
2. Varha-II cave-temple
3. Rmnuja Maapa (cave-temple)
4. Mahishamardin cave-temple
5. Kri Maapa (cave-temple)
(vii) Trimrti Cave-Temple
There is one major cave-temple at Mmallapuram which appears
to belong to the Mahmalla style group, but is different from the others
in that it has no maapa and was primarily dedicated to iva. It is the
so-called Trimrti cave-temple. The name is appropriate as there are
three sanctums having in them relief figures of iva, Viu, and
Subrahmaya.
Subrahmaya, here, replaces Brahm. He is carved in a stand-
ing pose on the back wall of the northern sanctum. He has four arms.
iva, similarly, is carved on the back wall of the central sanc-
tum. This central sanctum is given additional prominence by being set
forward in front of the other two side sanctums.
In the southern sanctum is a four-armed figure of Viu.
An eight-armed figure of Durg is carved in a niche, outside, on
the southern side of the Viu sanctum.
(viii) Kri Maapa
The last of the major cave-temples of Mmallapuram (and the
eighth in our list) is the Kri Maapa. It has slender, Mahmalla
type pillars but none with lions or vyas. There are five cells (the two
which flank the central one are set back a little from the others). Most
of the cells seem to have been dedicated originally to forms of iva,
since their guardians have aivite emblems (trident horns or axe-
blades) depicted on their headdress.
 116 There are three features which indicate to me that this cave-
Pallava Art temple should be assigned to the early part of Paramvaras reign:
1. It is predominantly, if not totally, iva-oriented.
2. Though all the pillars of this cave-temple are slender, and
those of the inner row are of the elegant Mahmalla style, yet
there are no lions at the base of these pillars.
3. The dvraplas of this cave-temple have leg ornaments, a
characteristic not found on the dvraplas of monuments more
certainly belonging to Mahndras or Mahmallas reign.
7. Additional Notes
(i) On the Number of Sanctums
Of the eight major cave-temples at Mmallapuram, three are
single-celled (i.e., have only a single sanctum), four are triple-celled,
and one has five sanctums. Thus, there are more triple-celled cave-
temples in this group than any other type!
(ii) On the Introduction of Relief Images in Cave-Temples
The relief images found in the early temples are only an exten-
sion of the art of painting. Mahndras early cave-temples have very
few figures carved in them. His cave-temple at Pallavaram, for
instance, has none. However, every inch of these temples would have
been plastered and painted. And we may be sure that the walls of the
maapas would have been decorated with large painted panels dealing
with the same kind of subjects which we find in the later, carved
maapa panels. Mahndras title Citrakrapuli, which glorifies his
mastery of painting, surely refers especially to the paintings which
originally adorned the walls of his own cave-temples.
Now, the introduction of relief-carvings is only a three-dimen-
sional enhancement of the wall painting technique, itself. Thus, the
famous carved Gagdhara panel of Mahndras Tiruchi cave-temple
was fundamentally a painting, whose realism was enhanced by its
relief-carved ground. It is merely an accident of time that the plaster
and paint of this image have all but disappeared, and that we now per-
ceive this work solely in terms of the plastic art of carving.
The above comments apply equally to the maapa panels of
Mahmallas cave-temples at Mmallapuram. And, also, to the great
Penance Panel of the same place. The Penance Panel was basically
a great open-air painting. Today, we admire only the carved skeleton
of that great work.
When we turn to the question of the nature of the image wor-
shipped within the sanctum of the early Pallava temples, K.R. Srini-
vasan has this to say:
A close scrutiny of the earlier cave temples and rathas reveals
that though Mahndra and Mmalla deviated from the traditional
materials of construction, they perhaps could not do so in respect
of the principal image consecrated. In the earlier and contempor-
ary temples, the principal object of worship consecrated was a
painting on the wall or one fixed to the wall, or picked out or
 moulded in stucco and painted, or of wood, carved and appropri- 117
ately painted.13 Chronology Part I
Several supporting references are then quoted by him from
Sagam and post-Sagam works. He adds:
The Avanti-Sundar-Kath-Sra narrates how the queen of Rja-
hamsa offered worship to Guha in the cave temple and saw the
wall painting (bhitti citra) of Guha playing beside his parents
(evidently the Smskanda panel), and a son was born to her, as
a result of her wish and prayer.14
Since, on my view, a carved sanctum wall with the gods
relief image painted, would not be significantly different from a plain
sanctum wall painted with the gods figure (both are basically paint-
ings), I would say that the reference in the Avantisundarkathsra to
the cave-temples wall painting (bhitti citra) could very well be to the
kind of carved (and originally painted) Smskanda image which we
find in the Mahishamardin cave-temple, Mmallapuram.
The developments which led to carved stone images of deities
in the sanctums sanctorum must have been gradual. Perhaps the earliest
such creation in the Tamil country is that of the Reclining Viu in the
Shore Temple, Mmallapuram. This may very well have been created
during the reign of King Sihaviu (father of Mahndra-I). It would
appear that this image, in the beginning, was lying in the open air.
(Only much later did Mahmalla construct an abode out of stone for
the One with the mighty discus.)
Apart from this unique image, some of the earliest 3-D
paintings on stone of anthropomorphic figures were of the dvraplas
guarding the entrance(s) to the earliest Pallava cave-temples and their
sanctums. Then there are the 3-D paintings on stone of deities found in
the maapas, outside the sanctum sanctorum:
(1) the sizable figures of Brahm and Viu in the Orukkal
Maapa cave-temple, Tirukkalukkunram;
(2) the small Naarja and Vabhntika panels in the Pallava cave-
temple at Siyamangalam; and
(3) the large Gagdhara panel of Mahndras Tiruchi cave-temple.
Finally, judging from the evidence at hand, Mahmalla became
the first of the Pallava kings to introduce a 3-D painting on stone of a
deity in a sanctum of his own temple: it is the Durg image in the
Draupad Ratha. (Thus, I must disagree with K.R. Srinivasans first
statement above which implies that it was only during Paramvaras
reign that a carved image in stone appeared in a Pallava temples sanc-
tum.) Not long after this, the 3-D paintings on stone reliefs of Subrah-
maya, iva, and Viu were executed in the Trimrti cave-temple
sanctums, along with the adjacent Durg figure.
Once again, to judge from the evidence at hand (this evidence
being the remarkable image of Cmu at Mmallapuram), it was dur-
ing Mahmallas reign that a stone figure of a deity was created clearly
in the round. Is this stone image of Cmu to be considered as fund-
amentally a painting in the round? Or shall we finally admit that
 118 sculpture has taken precedence over painting?
Pallava Art Whatever way we answer these questions, we ought to note
that, from being simply the background material (a flat wall) on which
figures of deities were painted over a plaster base, the use of the mater-
ial stone evolved gradually till this stone became the very substance
which takes the form of the gods themselves.
_______________

1This study is a revised version of On the Chronology of

Mahabalipurams Monuments, Part I: the Cave-temples, a paper by
Lockwood read at a meeting of the Archological Society of South
India, October 22, 1974.
2I am ignoring the very small monolithic shrines found on the
beach to the south and north of the Shore Temple.
3Consider, as examples, the Mahndra caves at Kuranganil-
muttam and Mamandur (the Rudravlvara), and also the Trimrti
cave-temple and Kri Maapa at Mmallapuram. I am, of course,
excluding any comparison between dvraplas carved on either side of
the faade of a cave-temples maapa and dvraplas inside guarding
the entrances of sanctums.
4These sources are discussed in detail in Chapter VIII of C.
Minakshis book, Administration and Social Life under the Pallavas
(Madras: University of Madras, 1977).
5The following scholars have all assigned the excavation of
the Atiraachavara cave-temple to Rjasiha: G. Jouveau-
Dubreuil, in his book, Pallava Antiquities, Pt. 1 (1916), pp. 66-68;
R. Nagaswamy, in his paper, New Light on Mamallapuram, Trans-
actions of the Archological Society of South India: 1960-62, p. 11;
K.V. Soundara Rajan, in his paper, Rajasimhas Temples in Tondai-
mandalam, Transactions: 1962-65, p. 169; and K.R. Srinivasan, in his
book, Cave-Temples of the Pallavas (1964), pp. 128-29.
6In the Dharmarja Maapa inscription, this name of the
temple comes between verses 10 and 11.
7R. Nagaswamy, New Light on Mamallapuram, Transactions:
1960-62, pp. 23-24.
8South-Indian Inscriptions, Vol. I, No. 25, p. 20.
9Ibid., the 11th verse of Inscription No. 18.
10Ibid., Inscription No. 18, p. 5.
11Other notable appearances of iva in Viu temples in the
Tamil country are to be found in the Namakkal Viu cave-temples in
the Salem district.
12Epigraphia Indica, Vol. X, No. 22, p. 101.
13Some Aspects of Religion . . . (Madras: Madras University,
1960), p. 10.
14Ibid., p. 11.
 119

Gaa Ratha Inscription

10

11

12

del. Michael Lockwood

120

The Eleven Verses of the

Gaa Ratha and Dharmarja Maapa Inscriptions

1. Sambhavasthitisahrakraa vtakraa |
Bhydatyantakmya jagat kmamarddana ||
2. Amyacitramysvagu guabhjana |
Svasth niruttar jydana paramvara ||
3. Yasyguhabharkrnta kailsassadanana |
Ptlamagamanmurddhn rnidhistambibhartyajam ||
4. Bhaktiprahva manas bhava bhaallay |
D ca y bhuv bhra jytsa rbharaciram ||
5. Atyantakm npatirnnirjjitrtimaala |
Khyt raajaya ambhstnda vma kritam ||
6. Ja sthurnnikala sma pvaktm viyadvapu |
Bhma iv vijayat akara kmasdana ||
7. Rjarj na virasacakrabhnna janrddana |
Trakdhipati svasth jayatttarukura ||
8. rmattyantakmasya dviaddarpppahria |
rnidh kmargasya harrdhanasagina ||
9. Abhikajalpr citraratnmbujkar |
st vil sumukha irassarasi akara ||
10. Tnda kritantugandhrjjarmmandira ubha(m) |
Prajnmiasiddhyarttha karm bhtimicchat ||
|| Atyantakmapallavvaragham ||*
11. Dhiktn-dhiktm-punarapi dhig-dhig-dhigastu dhiktm |
Ynna vasati hday kupathagativimkak rudra ||
Atyantakmapallavvaragham ||**
_______________

*The temple name (together with the symbol for ) is found in

this position only in the Dharmarja Maapa Inscription.
**The temple name is found in this position only in the Gaa
Ratha Inscription. does not appear at all in this inscription.
 121

Atiraacavara Cave-Temple Inscription

1. rmattyantakmasya dviaddarpppahria |
rnidh kmargasya harrdhanasagina ||

2. Abhikajalpr citraratnmbujkar |
st vil sumukha irassarasi akara ||

3. Tnda krita ambhrbhavana bhtay bhuva |

Kailsamandaranibha bhbht mrdhni tihat ||

4. Bhaktiprahva manas bhava bhaa[]llay |

D ca y bhuvandhatt jytsa rbharaciram ||

5. Atiraacaa patiravanibhujmatiraacavaramidamakart |
Iha giritanayguhagaasahit niyataktaratirbhavatu paupati ||

6. Gurvvmnabhakti riyamatiayin durvvaha bhramurvvy

nissmnyaca dna samamati(ra)acakhyay y (bibhartti) |
Sthn nirmmpitsminvidi(taraa)jayakhytin tna (bha)rtt
bhtnmaamrtticiramatiraacavar ytu nihm ||

A(nugra)la ||

7. Yadi na vidht bharat yadi na harirnnrad na v skanda |

Bddhu ka iva samartthassagta klaklasya ||

Samaradhanajaya Sagrmadhra || ||
122
Atiraachavara Inscription Grantha version

del. Michael Lockwood

10

12

14

16
 EIGHT

Mmallapuram Chronology
Part II: The Rathas1
This study concentrates on the group of five monolithic
temples in Mmallapuram called the Five Rathas. King Narasihas
name appears twice on the Dharmarja Ratha, and his chief biruda,
Mahmalla, the very root of the towns name, Mmallapuram, is
engraved in large letters on the parapet wall railing of the stairway
between the 2nd and 3rd levels (eastern side) of this Ratha. A detailed
study of the architecture and sculpture of this Ratha by K.R. Srinivasan2
has clearly shown that this Mahmalla must be Narasiha-I (mid-7th
century A.D.) and not the later king, Narasiha-II. Our earlier work
has supported this position through a comparative study of the dress
and ornaments of the sculptured figures in Pallava art. Though
Narasiha-I was responsible for the major work on these Five Rathas,
there remains the question of later stages in their development.
At the outset, two important observations should be made about the
Five Rathas. First, these monolithic monuments are very much unfin-
ished. Second, they all have been systematically and thoroughly
damaged.
That these Rathas have been systematically and thoroughly
damaged is not so obvious a fact. But let me present the following
details. The upper levels of all the Rathas, excepting the Draupad
Ratha, have rows of miniature hut-like, barrel-vaulted roofed structures
called ku sls. At the ends of each of these ku sls, there are
horse-shoe shaped window arches called kus. At the top of each arch
there was a shovel-shaped finial projecting upward. And between the
two shovel-shaped finials of each ku sl, there were carved in stone
two pot-shaped pinnacles called stps. So each ku sl had two
finials and two stps carved in stone projecting upward. On the
 124 corners of each upper level, there is a hut-shaped roofed structure of
Pallava Art square section. Each of these had a single stp projecting upward,
but no shovel-shaped finial as vulnerable as those on the ku sls.
Let us now add up all these upward projecting parts:
Dharmarja Ratha:
1st level: 26 stps & 22 finials
2nd level: 20 stps & 16 finials
3rd level: 4 stps & 8 finials
Top: 1 stp (separate piece now on ground, broken)
Bhma Ratha:
1st level: 36 stps & 32 finials
Top: 18 stps & 12 finials
Arjuna Ratha:
1st level: 8 stps & 10 finials
2nd level: 4 stps & 8 finials
Top: 1 stp (separate piece now on ground, broken)
Nakula-Sahdva Ratha:
1st level,
front: 4 stps & 2 finials
side: 2 stps & 14 finials
2nd level,
Top: ? stps & 1 finial
Draupad Ratha:
Top: 1 stp (separate piece now on ground, broken)
The totals of these are 127 stps and 137 finials: 264 stone projections
in all. Why have I presented all these details? To emphasize the point
that someone (or some group) took the trouble of smashing and break-
ing off every one of these 264 projections! Consider how much work
this destruction, itself, would have taken.
Further, there is hardly an example of a Pallava nose left to
see today on the faces of the figures at the Five Rathas. The Archo-
logical Survey attempted to restore new ones made out of cement, but
with unhappy results. In the Draupad Rathas sanctum, Durgs arms
have been broken off, and there are many other victims of mutilation
the various gargoyles on the Dharmarja Ratha, for example.
When did this destruction take place? And by whom? Two
Pallava monuments a little distance away from the Five Rathas, the
Gaa Ratha built by King Paramvara-I (who ruled around the end
of the seventh century) and the two towers of the Shore Temple built by
his son, King Rjasiha (who ruled around the beginning of the eighth
century), do not reveal any such systematic and thorough damage
though the weathering of the stone in the Shore Temple has been
severe. On the Gaa Ratha, one of the two major finials, which are in
the shape of a trident with a mans head at the base, is still intact. And
eight crowning stps are still safely atop its vaulted roof. As for the
Shore Temple, there are really quite a few well preserved stps. These 125
include the crowning ones atop the two towers, cut out of black basalt Chronology Part II:
rock (reported to have been quarried in the Cuddapah region of Andhra The Rathas
Pradesh). The crowning stp on top of the bigger tower is perfectly
preserved.
Detour
The question may arise whether these beautiful black basalt
stps are original. To help settle this question, I have photographed 4.75
the topmost stps and some of the others, and one may make a visual
comparison of their shapes. The crowning stps, in black basalt, are
convincingly identical in shape to the other stps. The only difference
in treatment is that the basalt stps have sixteen facets (each facet is
cut with a slight concavity), whereas the other stps are smoothly
rounded. Now, the surviving, damaged Pallava liga recovered from
6
the sands some decades ago and restored in a somewhat haphazard
manner to the sanctum of the bigger, sea-facing shrine, is made of the
same highly polished black basalt rock as the two crowning stps.
The liga is also cut in sixteen facets. And the facets are slightly
concave, too. Thus, we have formal similarities which tie the black
basalt liga to the basalt stps, and those stps to the other stps, and
thus to the original construction of these shrines by King Rjasiha.
I make one more observation, in passing. The shaft of this Black basalt stp
Pallava liga stands implanted (head up) in the stone floor of the sanc-
tum. The liga pha is not missing, though. The pha is carved in
light relief on the surface of the stone floor, itself. A circular depres- 4.75
sion on the floor surrounds the liga and drains off to the northern
side of the sanctum. Elsewhere, however, there is evidence in other
Pallava temples that people at a later time were not satisfied with this
3.5
modest and unobtrusive form of the pha. At the Atiraachavara
cave-temple at Saluvankuppam, for instance, a massive pha has been
crudely placed over the Pallava liga. And in Rjasihas Kailsa-
ntha temple, Kanchipuram, the original Pallava liga was so large that
a later pha had to be introduced into the sanctum in three pieces as
Stps, Shore Temple
otherwise it would not have been possible to get it inside the sanctum!
Back to the Main Argument
We may infer from our earlier observations, that the systematic,
thorough destruction of all 264 stps and finials of the Five Rathas
should have been carried out prior to the creation of the Gaa Ratha
by Paramvara-I and the construction of the Shore Temple towers by
Rjasiha. We may further infer that the systematic and massive
destruction of all the stps and finials of the Five Rathas occurred
before the completion of these temples, and that in fact this destruction
is probably the very reason why the Rathas were abandoned and
forever left unfinished. (Note that these monoliths were all carved from
the top down so that all 264 upward projecting stps and finials on
top of the temples were finished, inviting the attention of the desecra-
tors, whereas the lower parts were largely unfinished.)
 126 Nevertheless, there is evidence that a small amount of work
Pallava Art was continued on the Rathas subsequent to their being massively dam-
aged. K.R. Srinivasan, in his book on the Dharmarja Ratha, has, with
painstaking detail, discussed the architectural and sculptural develop-
ment of this Ratha. And I have already suggested, in a previous study,
that the Smskanda panel carved on the back wall of the sanctum of
the third level of the Dharmarja Ratha was a creation of Paramvara-I
(during the latter part of the seventh century A.D.). I would like to add
here that the bhtamla lintel, above the third level shrines entrance,
very clearly does not belong to the original design of this Ratha. The
lintel has been cut unceremoniously through the existing architectural
details of the cornice. The bhtamla carving should be contemporan-
eous with the later Smskanda panel inside this shrine.
The form of the script of the two label inscriptions naming this
third level shrine Atyantakma-Pallavvara-Gham is very close to
those other Mmallapuram inscriptions which we have already attrib-
uted to Paramvara-I. This act of Paramvaras naming the shrine
after a biruda common to himself and Narasha-I is not unique. His
son, Rjasiha, was later to do it with the Viu shrine in the Shore
Temple complex. And K.R. Srinivasan has pointed out the fact that the
name of the Chola king, Vijaylaya, was given to the Vijaylayac-
vara in Narttamalai, though, according to its own clear foundation
inscription, it was built by an earlier ruler not by Vijaylaya.
To go back to the Dharmarja Ratha in Mmallapuram, I thus
see this panel and its two related label inscriptions as part of an approp-
riative act of Paramvaras, creating a sanctum dedicated to iva-
Smskanda in what had otherwise been an abandoned monument.
I wish now to suggest that King Narasiha-I (Mahmalla),
himself, had the eight imposing figures on the first level of the Dharma-
rja Ratha carved after the Rathas had been massively damaged. I offer
the following observations in support of this claim. The first level
carvings are equal, if not superior, to the sculpture on the other levels,
yet it would appear that the artisans were no longer interested in main-
taining architectural symmetry and order. One has merely to stand at
the northeast corner of the Dharmarja Ratha and look at the two ad-
jacent niches with the figures of Harihara and iva-Ardhanr. They
are superb carvings, but the bottom edges of these two adjacent niches
are not at all on the same level! I would maintain that this inequality
would have been architecturally unthinkable in the ordinary order of
events. If one inspects the upper levels of this same Ratha, there is no
evidence whatsoever of such a disregard of symmetry. A comparison
of the other proportions of these first level niches will strengthen my
claim that there has been an architecturally lax approach in executing
the niches proportions.
The inscriptions on this first level begin above the Harihara
figure with the name, r-Narasiha. It would seem reasonable,
therefore, to assume that these first level figures and the inscriptions
above them were also carved and inscribed during the reign of Nara-
siha-I. Thus, we would have the following sequence of events: the
Rathas were started by Narasiha-I (Mahmalla), but during his reign 127
they were massively damaged by his enemies; nevertheless, Narasiha Chronology Part II:
had the figures on the first level of the Dharmarja Ratha executed after The Rathas
this destruction, and his name and birudas added at that time above the
figures. The last stage of work on this Ratha, carried out after a gap of
some time, was the carving by King Paramvaras artisans of the
Smskanda panel in the third-level sanctum and the bhtamla lintel
above the sanctums entrance, and the engraving of the two label
inscriptions relating to this third-level shrine.
Conjecture
The Gadval copper plate grant of the Chlukya king, Vikram-
ditya-I, declares that victory was achieved by the lord rvallabha
(Vikramditya), who crushed the glory of Narasiha (Mahmalla).
(E.I., X, p. 105.)
Earlier, Narasiha-I had fought three battles with Vikram-
dityas father, Puliki-II at Pariyala, Maimagala, and ramra.
Maimagalam is a village a short distance south of Madras city, and
therefore not far away from Mmallapuram. Thus, we have evidence of
the Chlukyan army invading the Pallava territory, first, in the reign of
Puliki-II, sometime before his defeat and death at the hands of Mah-
malla, in 642 A.D., and next in the reign of Vikramditya-I, sometime
before 668 A.D., when Mahmallas rule is supposed to have ended.
Narasiha-I succeeded his father Mahndra in 630 A.D. In
his 13th regnal year (642 A.D.), Narasiha crushed Puliki and des-
troyed the Chlukyan capital, Vtpi (Badami). After this victory,
Narasiha ruled for another 26 years.
I would suggest that the victorious Mahmalla brought back
artisans from Vtpi. Mmallapuram was then developed by him and
took its name from his victorious title Mahmalla. At some time
during the period when most of these monuments (including the Five
Rathas) were being created, Vikramditya-I invaded and crushed the
glory of Narasiha (Narasiha yaas vihita). (E.I., X, p. 105.)
Many years later, in 735 A.D., Vikramditya-II (the grandson
of Vikramditya-I) invaded Kanchipuram during the reign of Nandi-
varm-II. It should be noted that, though Vikramditya-II captured the
capital city of the Pallavas, he expressly stated that he did not destroy
it. (E.I., IX, p. 206.) At Kanchi, the invading king rejoiced Brahmins
and poor and helpless people by his uninterrupted liberality, (and he)
acquired high merit by restoring heaps of gold to the stone temple of
Rjasihvara and other gods, which had been caused to be built by
Narasiha Ptavarman. (E.I., IX, p. 206.)
This account of the gracious behavior of Vikramditya-II may
be supposed to contrast pointedly with the more destructive campaigns
of Mahmalla (Vtpi) and of Vikramditya-I (Mmallapuram).
Vikramditya-II, at the end of his campaign, took back with
him to his capital some of the leading southern architects. This fact is
evidenced in the inscriptions of the Virpka and Ppantha temples,
Pattadakal. (I.E., X, pp. 165 and 171.)
 128 Postscript (1997):
Pallava Art In his review of Mmallapuram and the Pallavas in The
Indian Express, Madurai, 13 Nov. 1982, the late Mr. N.S. Ramaswami
had this to say about my Conjecture in this study:
[O]ne point must be taken up here. Asserting that the Five Rathas
are not merely unfinished but also have been systematically and
thoroughly damaged because he [Lockwood] has found that all
264 stone projections have been broken, he conjectures that
Vikramaditya I, the Badami Chalukya, who invaded the Pallava
kingdom, was responsible for it. This is hardly conceivable and
quite opposed to old Hindu practices.
I regret my response to Ramaswamis claim that such dese-
cration is inconceivable comes more than ten years after his review.
However, I do have a response. In South Indian StudiesII, in his
article, Purananuru and a Rethinking on Ganapathi Worship in Tamil-
nadu, M. Arunachalam has noted the following (p. 43):
A laudatory verse on Maravarman Sundara Pandya says that when
he conquered the Chola country in the days of Raja Raja III, every
temple and monument in the land was razed to the ground except
the sixteen pillared hall which commemorated the grant of King
Karikla Chola to the poet Rudrankanna for the song Pattinapplai.
And I would mention one further observation. In the book,
Tamil Epigraphy A Survey (Madurai: Enness Publications, 1980),
p. 11, N. Subrahmanian and R. Venkatraman write:
A stone inscription at Trivendipuram in South Arcot district
inscribed during the 16th regnal year of Rajaraja III Chola (A.D.
1231) is an example of this class [of political inscriptions]. It says:
Kpperunjinga imprisoned Rajaraja III at Sendamangalam, devas-
tated the Chola country and desecrated the temples.8
(The emphasis in the above quotation is mine.) The footnote, No. 8, is
their footnote. That footnote reads:
8E.I., Vol. VII. It is interesting to note that a Hindu chieftain
destroyed Hindu temples.
Still other examples could be given to emphasize my point that
the claim that Hindus desecrated the temples and monuments of other
Hindus is not only conceivable, but, unfortunately, is supported by
historical facts, but I shall rest my case with the above observations.
_______________
1This study by Lockwood was first published in Mmallapuram
and the Pallavas (1982).
2The Dharmarja Ratha and its Sculpture: Mahabalipuram
(New Delhi: Abhinav Publications, 1975).
 NINE

The Philosophy of
Mahndras Tiruchi Poem1
In the first volume of South-Indian Inscriptions, E. Hultzsch
edited King Mahndravarms inscription which is engraved on two
stone pilasters flanking the famous iva-Gagdhara panel in the
kings cave-temple on the Rock-Fort Hill, Tiruchirapalli. A puzzling
error in Hultzschs reading of the Tiruchi inscription has necessitated
a fresh examination of it.
A point which we shall also discuss now is the popularly held
identification of King Mahndra with the Pallava king in the Periya
Puram account who was converted from Jainism to aivism. Mahn-
dras Tiruchi inscription is often offered as historical evidence of his
conversion. We wish to insist in the following study that Mahndras
inscription does not really support such an interpretation.
Mahndras Sanskrit inscription exhibits dhvani it possesses
different levels of suggested or implied meaning in addition to the
surface or obvious meaning. This inscription refers to the adjacent
stone sculptured Gagdhara panel. We reveal how the dhvani in the
inscription is echoed by a kind of parallel dhvani in the stone sculpture
to which it refers.

The reading of ancient inscriptions of the Pallavas is beset with

many difficulties. There are the usual problems of philology. And in
many cases, these records have suffered from the passage of time and
are damaged and fragmentary. But it would seem to us that the greatest
problem standing in the way of a correct understanding of many of
these inscriptions is a proper interpretation of their underlying spirit and
philosophy. This observation is especially relevant to King Mahndra-
varms famous inscription found in Tiruchirapalli. The Pallava king,
Mahndravarm-I, excavated a cave-temple in the Rock-Fort Hill, in
the center of this town, in the early part of the seventh century A.D.
In this cave-temple there is a carved wall panel depicting iva-Gag-
dhara. And on the hard rock surface of the pilasters which frame this
panel, Mahndras inscription is engraved.
In 1890, Hultzsch edited and translated this inscription. With
all due respect to him, we have maintained in previous studies that
Hultzsch had misunderstood three things with regard to the interpre-
tation of this inscription.2
First, Hultzsch, in his translation, had given a misleading inter-
pretation of the Sanskrit word nidhya, and said that King Mahndra
placed an image of iva in the cave-temple. Because of this misin-
terpretation, Hultzsch failed to understand that the inscription was
specifically related to the Gagdhara panel which is carved in situ.
 130 Secondly, Hultzsch did not understand the inscriptions
Pallava Art import that when King Mahndra had the figure of iva-Gagdhara
carved, this figure was also fashioned as a portrait or representation of
the king, himself.
Thirdly, in the inscription, the expression Daughter of the
Mountain can refer to the goddess Gag who is depicted in the carved
panel, and not just to Prvat, as all scholars have been assuming since
Hultzschs day.
Our reinterpretation of Mahndras inscription is significant in
that it shows that the making of a major image of a god which was also
a representation of a human being was practiced in India in the early
seventh century A.D.
In this study, we would like to point out that Hultzsch mis-read
as ilkhara a word in the last verse of the Tiruchi inscription. An
examination of the original inscription reveals very clearly that the
proper readings is: ilkara. As the word ilkara is clearly
engraved in this inscription, the question naturally arises how Hultzsch
could have read il[kh]ara in its place. We can only guess that
from the interpretative framework which he had established for the
whole inscription, this particular verse would make very little sense to
him with ilkara instead of his reading: ilkhara (stone-chisel).
Thus, Hultzsch considered it a scribal error and corrected it editorially.
Hultzsch translated this verse, therefore, as follows:
By the stone-chisel a material body of Satyasandha was executed,
and by the same an eternal body of his fame was produced.
Here are the actual inscribed words:
ilkara janit satyasandhasya bhautik |
Mrtti krttimay-csya kt tnaiva vat ||
One solution we propose and we assume that there was no
scribal error is that the expression ilkara, in its most easily under-
stood meaning here, should be interpreted as imperishable stone. We
would, therefore, translate the above lka thus:
Through Satyasandhas bodily image [bhautik-mrtti the
Gagdhara image is meant by this expression] created out of
imperishable stone [ilkara], an imperishable embodiment of
his [Satyasandhas i.e., the kings/Gods] fame has been made.
Satyasandha is a well-known title of Mahndras. It is found
in the list of royal titles engraved on the faade pillars of this cave-tem-
ple, as well as in other cave-temples of his. Satyasandha is also one
of the Thousand Names of the god iva. Thus, we have an example
of dhvani in the dual reference of the title Satyasandha in this passage.
The whole verse may be read as referring to the god iva or, alternately,
it may be read as referring to King Mahndra.
The plastic form of the carved Gagdhara figure which repre-
sents Satyasandha is, in a parallel way, a kind of sculptural dhvani,
and it also has a dual reference to both God and king. (This point is
being made, we believe, for the first time in Indian epigraphy and art.)
 Another interpretation of this verse is possible. The word 131
Akara is also a name of iva. Akara has the meaning of im- Mahndras Tiruchi Poem
perishable, and as such it may stand for the immutable god-head, iva.
Thus, the expression ilkara can be read as il-iva (i.e., stone-
iva). In this context, the verse may be read as:
Through this stone-iva, a physical embodiment of Satyasandha
[King Mahndra] was created, and through this form, his fame was
made eternal.
This interpretation would again support our claim that the
Tiruchi iva-Gagdhara image is also at the same time a representa-
tion of King Mahndra.
There is one more level of interpretation which may be given,
which we consider to be the most fundamental level. King Mahndra
was a noted poet. He pioneered the writing of farcical drama in
Sanskrit with his two plays, Mattavilsa and Bhagavadajjuka. The
author of this Tiruchi inscription was very likely the king, himself.
Mahndra was also a noted artist. The royal title Citrakrapuli
(Tiger among artists), which appears in this very same cave-temple at
Tiruchi, testifies to his artistic ability. The kings creative and invent-
ive powers are praised here in another of his titles, Vicitracitta. Thus,
we may understand that both the poetry of the inscription and the
remarkable sculpture of the panel in this cave-temple owe their exist-
ence to his creative inspiration. In this context, the above verse, with
the existing word ilkara, can be rendered in English as follows:
This bodily image [of Satyasandha (God/king)] was created out of
the stone inscription [ilkara] of Satyasandha [the poet-king].
By the same imperishable character, an embodiment of His/his
fame was made imperishable.
Our view, then, is that Mahndra made the image of iva-
Gagdhara in his own image. Consider, for instance, the first verse of
the inscription. In this verse, the self-identification of King Mahndra
with iva is expressed quite emphatically. However, in literally inter-
preting the word nidhya, Hultzsch ends up with a translation at once
perplexing and erroneous:
When King Guabhara placed a stone-figure in the wonderful
stone-temple on top of the best of mountains, he made in this way
Sthu (iva) stationary and became himself stationary (i.e.,
immortal) in the worlds together with him.3
Hultzschs reading of nidhya as meaning literally placed has
led to the supposition by him and subsequent scholars that no less than
three separate statues were placed in the sanctum of the cave-temple
by King Mahndra!:
1. a stone statue (anthropomorphic) of iva;
2. a portrait statue of himself (the king); and
3. a statue of Prvat (this statue being postulated on the basis of
another verse which speaks of the Daughter of the Mountain
taking up permanent residence on this mountain).
 132 There is not a trace of any of these separate statues. Nor need
Pallava Art there be any! There never were such separate pieces. Once the proper,
poetic interpretation of nidhya in this context is understood together
with the true nature of the God/king image, the meaning of the whole
inscription with its specific reference to the figures in the Gagdhara
panel becomes obvious. Here is our translation:
When King Guabhara [Mahndra] made a stone figure [the relief
image of iva-Gagdhara] in the wonderful stone abode on top of
the King of Mountains [the Rock-Fort Hill], this ruler, (entitled)
Vidhi [the Creator], made Sthu [iva] true to His name
[sthu: stationary / firmly fixed] and became himself sthu
[fixed, immortal] together with Him, on earth.
Now let us consider the fourth verse where there is an identi-
fication of God and king. In this verse, the religious and philosophical
basis of the identification of God with king is specifically stated:
Puruttama (Mahndra) bore on his head (that is, incarnate in his
features and in his mind) God immanent.
The full verse may be translated thus:
By first raising iva, the God within (his) heart, to his head, an
incomparable stone figure of Hara [iva] was then, with pleasure,
raised to the top of the mountain by this Puruttama [Mahndra].
And by thus himself first bearing, and then by making the mountain
bear, God immanent, on top, the Exaltedness of the Immovable
One [acalasya] was made a reality by him.
The seventh verse, with its underlying metaphor comparing
the Rock-Fort Hill to the kings crowned head, may be translated thus:
This mountain is like the diadem of his [Mahndras] Cha
province, this abode of Hara his (diadems) chief jewel, and the
splendor of akara [Gagdhara] is, as it were, his [Mahndras /
Kvrdharas crest-jewels] splendor.
The metaphorical comparison is as follows:
Cha province = king
mountain = diadem of king
cave-temple = crest-jewel of diadem
ivas splendor = splendor of crest-jewel
Mahndras metaphor stands at the root of various titles assumed
by later Pallava kings:
(1) iva-cmai4
(2) Candrrdhakhara-ikhmai5
(3) Mahvara-ikhmai-dptamauli6
And the key to a proper understanding of the meaning of these titles is
found in the Tiruchi inscription in the phrase:
. . . iva irasi dhrayattma-sastham . . .
The bearing of iva on ones head is merely a metaphor to express
God immanent in ones mind, soul, and self.
 Various scholars have suggested that some of the Pallava kings 133
wore an image of iva (iconic or aniconic) on their heads. For instance, Mahndras Tiruchi Poem
H. Krishna Sastri, in his commentary on the Vayalur Pillar Inscription
(Ep. Ind., XVIII, pp. 149-50) of Rjasiha Pallava (Narasiha-II),
says:
The adjunct [Mahvaraikhmaidptamauli] which occurs in
these verses and which, literally rendered, means one whose
diadem shines with the head-jewel, viz. Mahvara (iva), is rather
perplexing. Comparing this with titles like [ivacmai] etc. and
the verse [yasyguhabhrkrnta] etc. which occur in the
South-Indian Inscriptions, Volume I, Nos. 18 and 19 (v. 3)
[abhikajalpr] etc. in ibid., Nos. 21 and 22 (v. 2) all with
reference to king Rajasimha it looks as if the king did actually
wear a figure of iva or rather his symbol, the liga, on his head.
This fact is evidently also hinted in the verse [Guabharanmani
rjanyanna ligna ligini] etc. (ibid., No. 33, v. 2), which refers
to the conversion of the Pallava king Mahndravarman I Gua-
bhara from Jainism to aivism. Again, v. 4 of No. 34 in the same
volume speaks of iva fixed in the mind, being worn on the head.7
We feel that this is to take too literally the metaphor and fail to
give due weight to the philosophy underlying the metaphor. Of course,
one might argue that in Buddhist iconography, a small image of the
Buddha or the symbol of the stpa is found as a head ornament. But in
Pallava iconography, there is no known example of any gods image
(iconic or aniconic) appearing on the heads of the various royal portraits
at Mmallapuram and Kanchipuram.
Further, the very passages which Krishna Sastri cites as sup-
porting the suggestion that a figure or symbol of iva was worn as a
royal head ornament are themselves perplexing when interpreted in this
way. Consider, first, the second verse of the second half of the Tiruchi
inscription (S.-I.I., I, No. 33):
Guabhara-nmani rjany-anna ligna ligini jna |
prathat-cirya lk vipaka-vtt parvttam ||
We have tried to show in a previous study8 that King Gua-
bhara (Mahndra) possessed the liga (or anthropomorphic form of
iva) primarily in the sense that his portrait was combined with the
image of iva-Gagdhara. We, therefore, gave the following trans-
lation of this passage to bring out this primary meaning:
As the king called Guabhara has become embodied in this image
[ligini = Kvrdhara / Gagdhara], let the Faith which has been
brought back from the encircling opposition be forever spread by
this same image [ligna] throughout the world!
Hultzsch has given an alternative reading of the same verse in
his translation of it:
While the king called Guabhara is a worshipper of the liga let
the knowledge which has turned back from hostile (vipaksha)
conduct, be spread for a long time in the world by this liga!9
 134 Following Hultzschs reading of this verse, we have seen how
Pallava Art some scholars, including Krishna Sastri,10 considered the knowledge
(jnam) which turns back from hostile conduct to be the kings
knowledge, and thus this reading would give support to the story that
King Mahndra was converted to aivism from Jainism.
However, according to our own reading, jnam, here, should
be understood as the faith of the people in general, and therefore the
kings spiritual enlightenment is expressed by the two words: ligna
and ligini. Thus, the, kings enlightenment would be the instrument
of re-converting others back to aivism from rival faiths.
Our view is strengthened from the dhvani (suggested meaning)
of this same verse a meaning rather vaguely hinted at by Hultzsch in
one of his footnotes:
This whole verse has a double entendre. It contains allusions to
the Indian logic (tarkastras), in which ligin means the subject
of a proposition, liga the predicate of a proposition and vipaksha
an instance of the opposite side.11
This suggested dhvani with reference to Indian logic has been
repeated by later scholars, but the appropriateness of the logical terms
in the present context has not been made evident by any of them.
First, we think that the proper logical basis for the dhvani is
not that ligin means the subject of a proposition and liga, the predi-
cate, but rather that ligin means the conclusion to be arrived at in an
argument or inference, and liga means a reason advanced in support
of the conclusion:
Ligin = conclusion to be arrive at (pratij)
Liga = supporting reason (htu)
The whole inference is known in logic as anumna.
In this context, then, ligin would represent the conclusion to
be established, viz. King Guabharas (Mahndras) identity with lord
iva. And liga would represent the artistic work expressing this.
(And what is true for the king is true for everyone and everything.)
And, further, in this context, the verse expresses the hope that
this artistic work (image, temple) should become the instrument by
which others were to be brought back to the fold of aivism from rival
(atheistic) faiths (such as Jainism and Buddhism).
It is significant that one of the titles of Mahndra in the Tiruchi
cave-temple inscriptions is Anumna.12 This title of his, in the above
context, should be understood as indicating that the king had given
artistic expression to his (and others) spiritual self-identity with God;
and, in still another context, that he would be able to defend this
enlightened position through disputational arguments and the satirical
plays which he wrote which especially poked fun at degenerate
Buddhists.
It would seem, then, that for hundreds of years now, people have
gazed on the Gagdhara panel in the Tiruchi cave-temple and
have not realized that they were also looking straight at a portrait of the 135
great Pallava king, Mahndravarm-I. It is the philosophic dimension Mahndras Tiruchi Poem
of dhvani which has allowed us such an insight.
_______________

1This study is based on The Philosophy of Mahndravarmans

Tiruchchirapalli Epigraph, by M.C. Lockwood and A.V. Bhat,
published in the Journal of the Epigraphical Society of India, Vol. III,
1976, pp. 91-102.
2Refer to our study, Pallava Gagdhara.
3S.-I.I., I, p. 30.
4A title applied to Rjasiha both in the Kailsantha temple
inscription and Shore Temple inscription.
5A title applied to Rjasiha in the Shore Temple inscription.
6A title applied to Rjasiha in his Vayalur Pillar inscription.
7See also T.V. Mahalingams endorsement of this interpretation
in his book, Kcpuram in Early South Indian History (Bombay: Asia
Publishing House, 1969), p. 124.
8Pallava Gagdhara.
9S.-I.I., I, p. 29.
10See also T.V. Mahalingam, op. cit., p. 76.
11S.-I.I., I, p. 29.
12This title appears in the list of royal titles engraved on the
pillars of this cave-temple. The same title, Anumna, is also applied
to King Mahndra in his Pallavaram cave-temple inscription.
 136

16

16

iva-Gagdhara panel, upper cave-temple, Rock-Fort Hill, Tiruchi

 TEN

iva-Gagdhara/Pallava-Kvrdhara1

Work of the epigraphist includes discovering, reading, and

interpreting and translating inscriptions. After some introductory
remarks, we give a detailed word-for-word translation of King Mahn-
dravarms long inscription (eight verses) found in his cave-temple
near the top of the Rock-Fort Hill.
First, we note that the first four verses of this inscription are on
the northern pilaster, and the last four verses are on the southern one.
The number 2 is actually engraved at the end of the second verse on
the northern pilaster. If the inscription had begun on the southern
pilaster, this verse would have been number 6.
Our previous studies of this inscription have shown how the
iva-Gagdhara image is also a portrait of King Mahndra. We would
make one additional comment here. In verse 5, the poet reckons that
the Daughter of the Mountain (Gag) has left her fathers family to
stay permanently on this mountain (the Rock-Fort Hill), calling the
river Kvr the beloved wife of the Pallava king. Though the central
figure of the panel is to be viewed at the primary level as iva receiving
the descending river Gag on the locks of his hair, at another level
this same figure can be viewed as King Mahndra slowing the descent
of the river Kvr. May we speculate that Pallava engineers had been
involved in some way with the damming of the Kvr?
138

King Mahndras Tiruchi

Gagdhara Poem Inscription

Beginning on the Northern Pilaster:

1 ailndra-mrdhani il- bhavan vicitr

mountain=king top-of-on stone abode-in wonderful-in
2 ailn-tanu Guabhar npatir-n-nidhya [|*]
stone-body Guabhara king established-having
3 Sthu vyadhatta Vidhir= a yathrtha saja
Sthu made Creator this meaning-true-to name
4 sthu svaya- ca saha tna- [j]agatsu jta [|| 1 ||*]
fixed himself also together Him-with worlds-in become-has
5 Gham=akta atrumall gir[]ndra-kany-
abode made atrumalla mountain=king daughter-of
6 patr-g- girv= asmi [|*] Giriasya giria-
husbands mountain-on this Girias Mountain-Dweller
7 sajm=anvartth- kartum= artthapati || 2 ||
name meaningful to-make wealth-lord
8 Vibh[]ti-Cn katham=aham=avk
wealth Chas-of how I see-will
9 ya vipulm nad v Kvrm=avani-bhavanva-
which abundant river and Kvr earth-abode=remain-
10 sthita iti [|*] Harkta prty vibhur= adia-
ing thus Hara-by=asked-having-been affection-with the-king ordered
11 d=abhra-liham=idam-Manu-prakhy [rjy*] giri- bhavana-
cloud- licking this Manu-famous country-in mountain- abode
12 m=asmai Guabhara ||[3 ||*] Nirmmpit[m]=it[i] mud
Him-for Guabhara made-was thus pleasure-with
13 Puruttamna ail Harasya tanum=aprati-
Puruttama-by stone Haras body incompara-
14 mm=anna [|*] Ktv iva irasi dhrayattma-
ble him-by made-having iva head-on holder-by=heart
15 sastham= uccai irastvam= acalasya kta kt-
firmly-fixed-in lofty- mindedness mountains made-was real-
16 rttham ||[4 ||*]
ity
2

10

12

14

16

Delineation and photograph of the inscription on the northern pilaster

140

Continuing on the Southern Pilaster:

1 Kvr-n-nayanbhirma-salilm= r-
Kvr eye=pleasing water-possessor gard-
2 ma-ml- dharm dv vkya nad-priya
en- garland-bearer the-god on-seeing river-lover
3 priya- gum= apy=a rajyd= iti [|*] Sa-
pleasing qualities-possessor also this desiring thus with=suspi-
4 k giri- kanyak pit- kula hitvha many gi-
cion mountain-daughter-of fathers family having-left=here I-guess moun-
5 rau nityan- tihati Pallavasya dayitm=t bru-
tain-on forever stations (herself) Pallavas wife this call-
6 v nadm ||[5 ||*] Guabhara-nmani rjany=anna li-
ing river Guabhara- named king this-by im-
7 gna ligini jnam [|*] Prathat- cirya lk vi-
age-by image-having-become-embodied-in Faith renowned-be for-long the-world-in en-
8 paka-vtt parvttam ||[6 ||*] Ca- viayasya ail
emy circle-from brought-back Cha province-of mountain
9 maulir= ivya mah-mair= ivsya [|*] Hara-gham=ta-
diadem like=this great jewel like=his Hara-abode this (his Cha province)
10 j-jtis- tadyam= iva kara jyti ||[7 ||*] ilkar-
splendor his (crest jewels) like akaras splendor stone=inscrip-
11 a janit Satyasandhasya bhautik [|*] Mrtti krttima-
tion-out-of created-has-been Satyasandhas bodily image fame-full-
12 y- csya kt tnaiva vat ||[8 ||*] Niky[]cal-sa-
of and=his made-has-been it-(stone)-by eternal scooped-out=firm well
13 m-adhyi Guabhar bhakti [par] . . .
made-manifest Guabhara-in devotion surpassing

_______________
1Based on part of the paper, Trichy Pallava Kvr-dhara,
by M.C. Lockwood and A. Vishnu Bhat, published in the Journal of the
Epigraphical Society of India, Vol. XX, 1994, pp. 4-9.
 2

10

12

Delineation and photograph of the inscription on the southern pilaster

142

iva-Gagdhara, Mtagvara Temple, Kanchi

 ELEVEN

A Mystery Dog in Sculpture1

This article and the following study continue the examination
of the various levels of meaning of a given Pallava sculpture.

When the Gag was called down to earth from her heavenly
abode by the great tapas of King Bhagratha, she would have destroyed
the earth in a cataclysmic deluge had it not been for the intervention of
iva who received her mighty force on the locks of his hair and held her
there as easily as a single droplet until she was prepared to flow gently
down. Thus the Rmyaa recounts the event which, represented in
Indian art, is called Gagdhara.
This theme was very popular in Pallava art. In fact, the very
earliest extant major sculptured panel in the Tamil country is the
Gagdhara panel carved in Mahndravarms cave-temple in Tiruchi.
This earliest of major panels, which belongs to the seventh century
A.D., was followed by many other Pallava renditions of Gagdhara.
There are two Gagdhara panels at Mmallapuram, and four at the
Kailsantha temple, Kanchipuram. In the same city of Kanchi, there
are at least three smaller Pallava iva shrines which have them.
What is surprising is that in many of these Gagdhara panels
a dog appears in one of the upper corners. To put it mildly, the dog is
considered a lowly creature in Indian tradition. It is therefore difficult
to guess why the Pallava artists should have introduced a dog into the
Gagdhara theme a theme which represents such an auspicious event
for the whole world.
Mayilai Seeni. Venkatasamy, in a learned journal, noted that a
passage in a Tamil stone inscription at Tiruvannamalai (North Arcot)
which was engraved during the reign of Kpperujiga, who claimed
Pallava descent, can be interpreted as saying that iva, the Primeval
Being, at the time of receiving the Gag on his head, created the
illusion of a dog. Unfortunately, the inscription provides no other
information about the significance of this incident. Further, the crucial
passage in the Tamil inscription is open to other interpretations which
would eliminate any reference to a dog.
To the best of our knowledge, the puras are silent about any
dog in relation to the Gagdhara story. Apart from Venkatasamys
suggestion, we have not yet met a single person who could enlighten us
from other sources about the mystery dog.
But still the plain fact remains that a dog does appear in many
Pallava Gagdhara panels, and even in a few Gagdhara panels found
in other regions. The accompanying photograph is of the dog in the
Gagdhara panel of the Mtagvara temple at Kanchi. Fortunately,
 144 this carved dog is well preserved and is not covered with that thick
Pallava Art plaster which obscures so many of the great works of the ancient past.
The hitherto enigmatic animal carved in the Tiruchi Gag-
dhara panels of Mahndras, which in its present damaged state has
mystified generations of scholars, is now known to be a dog. The crea-
ture in the upper left (proper right) corner of the Gagdhara panel of
the west-central lateral shrine of the Kailsantha temple at Kanchi can
now confidently be accepted as a dog, if any doubt may have existed
earlier. Similarly, we can be sure that it is a dog appearing in Gag-
dhara panels of the Muktvara and Iravtanvara Pallava temples at
Kanchi.
But the basic mystery remains over the question why the dog
appears in any Gagdhara panel. Somewhere there should be a ver-
sion of the Gagdhara story which would account for this unusual
appearance of a dog.
_______________

1Based on A mystery dog in sculpture, an article by M.C.

Lockwood, published in The Indian Express, Madras, March 6, 1976.
 TWELVE

Dhvani in Epigraph and Stone1

This study is devoted to the further investigation of dhvani
in epigraph and stone sculpture. It will become evident that dhvani in
Pallava art is not merely one level of implied or suggested meaning,
but rather a rich spectrum of different levels of suggested meaning.
The various inscriptions of the Pallavas which are in poetic form
are excellent examples of the use of dhvani. We have already exam-
ined King Mahndras poem inscribed in his Tiruchi cave-temple.
Mahndras Tiruchi epigraph refers specifically to the adjacent
carved Gagdhara panel. We pointed out the fact that the dhvani in the
poem is paralleled by a type of dhvani in the sculpture itself (a God-
king image).
Mahndras inscription, however, does not give us any clue to
the significance of the two prince-like figures with ja-makuas who
are kneeling on either side of iva-Gagdhara. Nor does it give any
clue to the recumbent creature carved to the upper proper left of iva.
The head of this animal has unfortunately been damaged, making its
identification difficult.
Mayilai Seeni. Venkatasamy was the first to point out the
presence of a dog in a Pallava Gagdhara panel found in Kanchi-
puram.2 In his article, Venkatasamy noted that there is a passage in a
Tamil stone inscription at Tiruvannamalai (North Arcot District) which
provides an explanation of the dogs presence in the Gagdhara panel.
The inscription was engraved during the reign (in the 13th century
A.D.) of the ruler Kpperujiga, who claimed Pallava descent. The
relevant passage in this inscription describes iva receiving the Gag
on his hair. And Venkatasamy finds in it the clue to the dogs presence:
Kautar-perum=tintany va kou pypuar
Kakaiy=yira-mukako=rtteum=a=u-k-koa
tiruntiya piai muiy=aruntava-c-catarar. . . .3
However, Venkatasamys interpretation of this passage has
been questioned.4 Whatever be the proper interpretation of the Tiru-
vannamalai inscription, the fact remains that a dog does appear in
Pallava Gagdhara panels and in Gagdhara panels elsewhere also.5
Following Venkatasamys paper, we wrote an article published
in The Indian Express about several other Pallava Gagdhara panels
which have a dog portrayed in them.6 A photograph accompanying the
article clearly showed a dog seated on its haunches in the upper corner
of the panel opposite the half-anthropomorphic image of Gag.
 146 But the basic mystery remained over the question of why a
Pallava Art dog should appear in any Gagdhara panel. In the Express article, we
appealed for any further information which might solve this problem.
One reader, in a letter to the editor, suggested that the dog should be
one of the two dogs guarding Yamas gate.7 Another reader felt that it
should be Yama himself, in the form of a dog, as told in a story in the
Mahbhrata.8
The most thought-provoking idea offered as a solution to our
problem came in an article which proposed that the mythology and
iconography of the Gagdhara theme involved the constellations in
the heaven.9 iva-Gagdhara, on this interpretation, is imaged in the
constellation called Orion by the Greeks. The Gag is the heavenly
Milky Way. The dog would then be the constellation Canis Minor,
the Little Dog.
But let us return to the Tiruchi Gagdhara panel and to the
realm of epigraphy. Perusing the early volumes of The Indian Anti-
quary, we came across a surprising fact which would seem to have
relevance to an interpretation of the dog portrayed in the Gagdhara
panels of the Pallavas. More than one of the Kadamba copper-plate
grants have a dog engraved as the emblem of the royal seal of the
grants.10 Now, it is well known that the western Gags and the Kad-
ambas were important feudatories of the Pallavas. We therefore
suggest that at one level, at least, the implied meaning of the sculptural
dhvani of the Pallava Gagdhara panel in Tiruchi is as follows: the
image of Gag, with her hands held in ajali mudr, may be taken as
an emblem of the Gag feudatories of the Pallavas; and the image of
the dog may be taken as an emblem of the Kadamba feudatories. We
have already argued in detail earlier that the iva image is also a repre-
sentation of King Mahndra, the Pallava King of kings. The two
prince-like figures which are carved in the Tiruchi panel, kneeling on
either side of the iva/Mahndra image, would then represent the
respective kings of those two subordinate dynasties. The kneeling
figure on the proper right, under the Gag figure, would portray the
Gag king. And the figure on the other side, beneath the dog, would
portray the Kadamba king.
One level of the implied meaning of the dog in these panels
would then be apparent. But the mythological significance of the dog
in relation to the Gagdhara theme would seem to require further
investigation.
_______________
1Based on Dhvani in Epigraphy and Stone, a paper by Lock-
wood and Bhat read at the Fifth Annual Congress of the Epigraphical
Society of India, Bangalore, Feb. 3-5, 1979.
2M.S. Venkatasamy, Kaktara mrttiyi ariyatoru cipa
vaivam (in Tamil), Journal of Tamil Studies, Vol. V, Sep. 1974,
pp. 70-74.
3South-Indian Inscriptions, Vol. VIII, No. 69, p. 40. (A.R. 480
of 1902.) On the west wall of the first prkra of the Arunchal-
vara temple. A record of the Pallava king, Kpperujigadva, the 147
protector of Mallai (Mmallapuram). Records gift of ornaments by the Dhvani in Epigraph
king and erection of buildings by his son. In interpreting the relevant and Stone
passage of this inscription, Venkatasamy actually suggests that iva,
the Primeval Being, took the form of a dog: tinta ny va kou.
The precise reading of the beginning of this passage, however, is
tintany . . ., not as Venkatasamy reads it. This difference has left
room for alternate interpretations.
4See, for instance, Ta. Mu. Subrahmanyams rebutting article,
Ny vakoa nampa (in Tamil), Koku, Vol. V, May 1975.
5See, for instance, plate XXVI, fig. 1, the Gagdhara panel of
the Kailsa temple, Ellora, in J. Burgess, Elura Cave Temple, Vol. V,
Archological Survey of Western India (reprinted in 1970 by Sagar
Publications, New Delhi); and fig. 27, the Gagdhara panel of the
Garua temple, Alampur, in C. Sivaramamurti, Nataraja in Art,
Thought and Literature (New Delhi: National Museum, 1974), p. 186.
6Lockwood, A mystery dog in sculpture, The Indian
Express, Madras, March 6, 1976.
7M.E. Adiceam, The Indian Express, March 17, 1976.
Adiceam refers to an article of hers (in French) published in Ars
Asiatica, Vol. 32, 1976. (The scriptural reference is to the g Vda,
X.14 & 15.)
8G. Basker, The Indian Express, Madras, March 27, 1976.
9R. Venkatram, A Mystery Dog in Sculpture, Journal of
Tamil Studies, Vol. 8, Dec. 1975, pp. 12-17. (It should be noted that
this journal is pre-dating its publication! Venkatrams article was ac-
tually written after Lockwoods Express article of March 6, 1976, to
which Venkatram refers and whose title he borrows.)
10See the seal of the Kadamba copper-plate grant of
Kkusthavarm (The Indian Antiquary, Vol. VI, 1877, No. 20) for
a clear example of the dog. It is interesting and, at the same time,
puzzling that the emblem on the seal of the Uruvupai grant of
Yuvamahrja Viugpa, issued during the reign of the Pallava king,
Sihavarm, is also a dog (The Indian Antiquary, Vol. V, plate oppo-
site p. 50).
Faade of the Fifth Shrine (note the inscription!)
 THIRTEEN

Queen Ragapatks Inscription1

In the first volume of South-Indian Inscriptions, the pioneering, German epigra-
phist, E. Hultzsch, erred in fixing the location and sequence of some of the inscriptions on
the small shrines in front of the Kailsantha temple complex at Kanchipuram. One group
of these seemingly related epigraphs includes the delightful foundation inscription of the
Pallava queen, Ragapatk.
While correcting the sequential order of these inscriptions, Prof. Bhat and I have
also given a fresh translation of them. One of the key points made in our new translation is
that the expression Paramvara of these inscriptions refers to King Paramvara-I, and
is not a title of King Rjasiha. When this fact is clearly established, it demolishes the last
support of the mistaken claim that King Rjasiha assumed his fathers abhika-nma
(coronation name), Paramvara, as his own biruda (royal title). This is an important
issue because this mistaken claim is the mainstay of the misplaced attempt to credit
Rjasiha with the creation of all the Pallava monuments at Mmallapuram.

The Kailsantha temple at Kanchipuram is rich with inscriptions of its builder, the
Pallava king, Narasihavarm-II (Rjasiha) as well as of his son, Mahndravarm-III. In
front of the main temple complex, just outside its enclosing wall, are several small shrines
which belong to the same general period. On three of these small shrines are some inscrip-
tions which relate to their foundation by other members of the royal family.
Dr. E. Hultzsch, who edited and translated the Kailsantha temple inscriptions
in Volume I (1890) of South-Indian Inscriptions, included in that volume the inscriptions
found on these small shrines in front.2 The most notable of these inscriptions are three
verses in Sanskrit poetry ascribing the creation of one of the shrines to Queen Raga-
patk. There is an error in Hultzschs location of Ragapatks inscription. Hultzsch
located the verse which contains the name Ragapatk on the faade of the third shrine
to the right of the front entrance to the main temple complex. But this is not its correct
position. This verse is actually found on the faade of the fifth shrine to the right of the
front entrance.
This error in location is serious because the verse which contains the name,
Ragapatk, does not stand alone. Hultzsch read it in conjunction with two other verses
which actually are to be found on the third shrine. But now we shall have to read the
Ragapatk verse in conjunction with the two different verses found on the fifth shrine!
Because of this mistaken juxtaposition of verses, Hultzsch and all scholars since
his day have unquestioningly thought Ragapatk to be the queen of Narasiha-II. For a
clearer understanding of why they did so, we give below, in the order in which Hultzsch
presented them in Volume I of South-Indian Inscriptions, the several verses inscribed on the
third and fifth shrines.
Here follows Hultzschs translation along with his location of the various verses:
150

On the Third Shrine3

to the Right of the Front Entrance
On the faade, first line:
Adoration to iva!
(Verse 1.) She, who was the dearly beloved mistress of her hus-
band, the supreme lord, who was famed by the name of Kla-
kla, whose sign was the bull, and the strength of whose bow
had become manifest at the destruction of cities, just as the
daughter of the king of mountains (Prvat) is the dearly be-
loved mistress of her husband, the supreme lord (iva), whose
sign is the bull, and the strength of whose bow has become
manifest at the destruction of (the demon) Pura;

On the back:
(Verse 2.) She, who is resplendent, as she has attained the mighty
position of favourite with king Narasihavishu, who has split
the hearts of his foes, and who has devoted himself to the
protection of the circle of the world, and as thus she seems to
have subdued the pride of Pushkaradevat (i.e., Lakshm, the
wife of the god Narasiha-Vishu);

On the faade, second line:

(Verse 3.) That Ragapatk, who was, as it were, the banner
(patk) of women, caused to be built this lovely dwelling of
(iva,) whose crest-jewel is the moon.

Faade inscription on the Fifth Shrine

 151

On the Fifth Shrine4

to the Right of the Front Entrance

On the faade:
Prosperity!
(Verse 1.) She, who, full of loveliness, softness, grace and clean-
liness, seemed to be the master-piece of the first creator, whose
skill had attained perfection at last, after he had created thou-
sands of good-looking women;

On the back:
(Verse 2.) She, who was charming through genuine sweetness, who
was adorned with grace, coquetry and feeling, who, like the art
of attraction, . . .

* * * * * * *

It should be immediately obvious, though, that if Verse 3,

which mentions the name, Ragapatk, is not located (as stated
above) in the inscription of the third shrine, but rather is Verse 3 of the
inscription of the fifth shrine, then the heretofore unquestioned identity
of Ragapatk as the favourite queen of King Narasiha must be
examined afresh and established solely by some new evidence!
But this is not the only major reassessment called for with
respect to the inscriptions on these shrines. In re-translating the two
verses which properly belong to the third shrine, we would like to em-
phasize the point that there are four royal persons (not two) who are
involved in the poetical comparison with four divine beings:
1. King Paramvara-I (who is evidently deceased);
2. his wife (the surviving Queen Mother);
3. King Narasiha-II (son of Paramvara-I);
4. King Narasihas wife.
The similes are as follows:
King Paramvara-I = the god, Paramvara (iva)
Paramvaras queen = the goddess, Prvat
(Daughter of the King of Mountains)
King Narasiha-II = the god, Narasihaviu
Narasihas queen = the goddess, Lakm (Pukaradvat)
Here, then, is our own translation of the verses, with the
correction of the location of the Ragapatk verse:
152
On the Third Shrine
to the Right of the Front Entrance

On the faade:
Salutations to iva!
(Verse 1) (Her) husbands [i.e., King Paramvaras] well-merited
fame being widespread as Klakla on account of his bows
power (having been made) manifest in the destruction of cities,
(thus) like the Daughter of the Great King of Mountains,
(she,) the dearly beloved wife of Paramvara, the Bull-
bannered One,

On the back:
(Verse 2) attaining supremacy [as Queen Mother], shines with
surpassing splendor, subduing, as it were, the pride of Pukara-
dvat, while god-like Narasihaviu, true to his sacred vow,
is protecting the encircling world, tearing out the hearts of his
enemies.

On the Fifth Shrine

to the Right of the Front Entrance

On the faade, first line:

Prosperity!
(Verse 1) (She,) who, full of loveliness, gentleness, grace, and
purity, seemed to be the masterpiece of the primeval creator,
Brahm, whose craftsmanship had attained perfection at last,
after he had created thousands of good-looking women,

On the back:
(Verse 2) (she,) who was so appealing because of (her) genuine
sweetness, adorned with sentiments (both) charming (and)
fascinating, (who,) like the art of attraction, . . .

On the faade, second line:

(Verse 3) that Ragapatk, who was, as it were, the banner of
women, caused to be built this lovely dwelling of (iva),
whose crest-jewel is the moon.

* * * * * * * *

Finally, we give the Sanskrit texts of the inscriptions of the

third and fifth shrines in their correct order:
 153
On the Third Shrine
to the Right of the Entrance

On the faade:
Namaivya [||*]
(Verse 1) Bharttu purnmathana-da-dhanurbbalasya
aildhirja-tanayva vadhvajasya [|*]
Y Klakla iti viruta-puya-krtt
Knt nitnta-dayit Paramvarasya ||

On the back:
(Verse 2) Dv jagad-valaya-rakaa-baddha-dk
Nirbbhinna-atru-hday Narasihaviau [|*]
Vllabhyam-rjjitam-avpya virjat y
Nirjjitya-garvvam-iva Pukaradvaty ||

On the Fifth Shrine

to the Right of the Front Entrance
On the faade, first line:
r [||*]
(Verse 1) kra-sundara-vilsavat-sahasra-
sarggaprabandha-cira-[saskta-kau]alasya [|*]
Lvaya-mrddava-vilsa-mj samagr nirmma-
siddhir-iva y prathamasya dhtu ||

On the back:
(Verse 2) Aklia-mdhuryya-vilbhany vibhit
vibhrama-hva-bhvai [|*]
kara-vidym-iva l . . .
. . . . . . . . . . . . . . . . . . . [||*]

On the faade, second line:

(Verse 3) Nirmmpitam-idan-dhma tay Candra-[ikh]man [|*]
Pat[kayva] nr ramya Ragapatka[y ||*]
_______________

1This study is based on Pallava Queen Ragapatks

Inscription, by M.C. Lockwood and A. Vishnu Bhat, a paper pub-
lished in the Journal of the Epigraphical Society of India, Vol. IV,
1977, pp. 67-69.
2South-Indian Inscriptions, Vol. I, Nos. 28-30.
3Ibid., No. 29.
4Ibid., No. 30.
154

Chmu
 FOURTEEN

Notes on Mmallapuram1
There is much to be seen at Mmallapuram. And many interest-
ing things will be missed on a first visit. The following observations
were made on return visits to the site.
A Unique Image of Chmu
Dr. Gift Siromoney and I first noted the significance of this
image in a newspaper article (1972).2 Most visitors to Mmallapuram
never get around to seeing the stone-carved Saptamtks (Seven
Mothers) which are placed in a row on a raised platform near the local
Branch Library. Actually, there are eight separate figures in the group.
The few guide books which mention these Saptamtks consider them
all to be Pallava creations. But in our opinion, seven of them are
definitely late-Pallava or post-Pallava.
The remarkable exception is the central figure of Chmu
(Cmu) which is considerably larger than the others. What is so
rare about this piece? First, it is an image in the round datable to the
mid-seventh century A.D. Such a free-standing figure of the Pallava
period would be rare enough, but here is one which is contemporaneous
with the Penance Panel and the Five Rathas. There is no other Pallava
example of such an early free-standing image of god or goddess known
to us.
Second, even taking the relief images into account, it would
seem that this Cmu is the only example which we have in the south
of a Saptamtk of the seventh century the earliest period here of
such surviving stone sculpture. We have to turn to the eighth century
to find the Saptamtks carved in relief on the enclosure wall of the
Kailsantha temple in Kanchipuram.
Since Cmu represents the fearful destroying power of
Time, the Mmallapuram image of her has a skull on the center of her
headdress. She has a diabolical grimace with bulging eyes, pointed elf-
ears, and two fangs protruding downwards from her mouth. The orna-
ment hanging through her pierced right ear-lobe is a corpse (prta-
kuala). (The goddesss size must then be envisioned as gigantic.)
She wears, diagonally across her body, a garland of severed human
heads strung together. There is a thin band tied around her torso above
her slightly drooping breasts.
The Mmallapuram image of Cmu also has several unusual
features which indicate an early experimentation unfettered by the ster-
eotype of later tradition. She holds a dagger in her lower right hand and
grips a second corpse in her lower left hand. Originally, the figure
 156 had four arms, but, unfortunately, the upper left one has broken off
Pallava Art entirely. Her upper right hand seems to be holding a bell.
What leads us to claim that this is a seventh century Pallava
figure? The more obvious characteristics of this early period which we
can list in summary form3 are: the very large circular ornament in her
left ear (such a large size goes out of fashion by the time of King
Rjasiha Pallava, in the early eighth century); the bikini-like lower
garment that she is wearing, which has no waist ornaments whatsoever
(this extremely simple womens dress, it should be noted, is found only
in the earliest period of Mmallapuram art); the plain, single anklets;
and a general slenderness in the treatment of the torso that we connect
with the early Pallava style.
Overlooked Heterodox Image
Another overlooked piece is a badly damaged, but surprising
stone image found by us originally on top of the hill, opposite the so-
called Dharmarja Lion Throne (this area is generally thought to be the
site of palace buildings in the Pallava period). Dr. P. Dayanandan and I
noted this piece in a newspaper article (1970).4 Since then, the Arch-
ological Survey of India have removed it from the hill top and have
kept it in their museum, adjacent to the A.S.I. office, nearby.
The image is that of a seated figure in a yogic posture. The
upper half of the image was missing when we originally photographed
it. The pedestal of the image measures 31 inches across and has no
design on it.
It was difficult to identify this broken fragment because there
was no other image like it in Mmallapuram. Our immediate reaction
was to consider it to be part of an image of the Buddha or a Jain saint.
As the upper portion of the figure was later found and restored, it can
now be identified confidently as a Jaina image. Quite a surprising find,
this lone heterodox figure, among all the Hindu art at Mmallapuram!
 157

18

Jain image found on top of Mmallapuram hill

 158 The Reclining Viu Image in the Shore Temple
Pallava Art
The reclining Viu in the central shrine of the Shore Temple
complex has a really unusual feature which has somehow escaped
notice: Viu is portrayed with ja-makua. The Ja style of hairdo,
of course, is a well-known characteristic of iva and aivite images.
But it is unknown on images of the Reclining Viu. The Shore
Temple Viu image is thus unique in this respect.
A second aspect of this same image which I would like to dis-
cuss is the claim by many scholars (which claim has hitherto remained
unchallenged) that there is no serpent, Ananta, portrayed with this
image of Viu. Now, it is true that there is no elaborate and massive
carving of Ananta here as there is in the Mahishamardin cave-temple
panel of the same theme. However, as the Reclining Viu image is
carved out of the living rock at this very spot in the Shore Temple, the
original rock formation may have limited the sculptors. In any case,
there are two wavy, engraved lines running somewhat parallel immed-
iately in front of Viu. It has always seemed obvious to me that these
lines represent the body of Ananta gradually tapering to the right.
If anyone were to object that Anantas multiple heads are
nowhere to be seen, I would only answer that this shrine (which existed
from before Mahmallas time, well before the aivite shrines of the
Shore Temple complex were raised by Rjasiha) was, from the Mah-
malla period a composite structure: a built-up stone superstructure on a
rock-cut base formed from the living rock in situ. Outside, on the back,
at the base of this Viu shrine, one can still see the bottom portion of
figures in niches carved in the living rock. These figures must have
been continued in the stone superstructure which King Mahmalla
built, sheltering the pre-existing image of the Reclining Viu. (The
present superstructure is a later rebuilt one dating only from the days
of King Rjasiha. Rjasihas workmen never bothered to recreate
again the upper portions of these outside figures. And the stones in the
reconstructed wall are placed in a hodge-podge manner.) I suggest that
a similar fate befell the upper portion of the serpent Ananta, so that a
lack of heads is no proof that Ananta never existed in this Viu shrine!
(The repaired heads may have been made of stucco.) In conclusion, I
note that the mass of rock (the original mother rock) under the head of
Viu is ribbed horizontally in representation of the layered coils of
Ananta.
Worlds Oldest Childrens Slide 159
Notes on Mmallapuram
My brother, Dr. Merrick Lockwood, pointed out to me what is
plainly before everyones eyes: a childrens slide cut into the living
rock. What makes this particular slide so unusual is that it was created
over 1,200 years ago, and is located to the immediate right of the
famous Penance Panel of Mmallapuram.
The children of Mmallapuram also know a slide when they see
one and make proper use of it. One often sees children sliding down
it. In a newspaper article (dated April 16, 1972)5 which first reported
this slide, I also raised some questions concerning it. Was the slide
created only for children or for grown-ups too? Since we see only the
upper two meters of the slide (and the steps leading up to it), how far
down below the present ground level does the slide extend? Did the
slide, perhaps, end in a watery splash in the same pool which received
the cascading Gag?
Thanks to a little dig which the Archological Survey of India
carried out, I was able to report the following in a newspaper article
dated October 1, 1972.6 From the excavation, it was found that the
slide continued for approximately another one meter below the present
ground level. The total length of the slide, therefore, is about three
meters. Five more steps were uncovered by the digging, making a total
of 12 steps leading up to the top of the slide. The bottom of the steps is
at the same level as the bottom of the slide.

13

8.5
 160 The last question, whether the slide ended in the pool, could
Pallava Art now be answered negatively the pool level (assuming it to have been
under the elephants feet) would have been far below the end of the
slide. The earlier question, whether the slide was created for children
only or for grown-ups too, would (in view of its three-meter length)
probably be decided in favor of children alone.
The Penance Panel Its Interpretation7
The great open air bas relief at Mmallapuram continues to be
the subject of scholarly controversy. Does it portray Arjunas penance,
or King Bhagrathas? The debate is still very much alive.
Some time ago, I entertained the idea that possibly both sides
in this debate could be right! The figure standing on one leg doing
penance could represent both King Bhagratha and Arjuna at the same
time. This suggestion is not as preposterous as it might at first seem.
Several studies in this book, including Dhvani in Epigraph and Stone,
should explain the mechanism by which this double meaning is possible
for a given sculptured figure. It was around the period when Mmal-
lapurams monuments were being created that the great poet Dain is
credited with having written a type of poem, a dvisadhna-kvya.
This work of his could be read either as an account of the Rmyaa or,
alternately, of the Mahbhrata. One particular manner of arbitrarily
dividing the compound Sanskrit would result in the story of the
Rmyaa. But if the compound expressions were divided differently,
it was instead the story of the Mahbhrata. Which epic did Dains
dvisadhnakvya really relate? The answer is: both.
In a parallel way, couldnt the Mmallapuram Penance Panel
portray both Arjunas penance and King Bhagrathas? The single
penitent figure could then be both Arjuna and Bhagratha at the same
time. Theoretically, there is no reason why this figure might not have
represented both. However, when all the available evidence is weighed,
I feel that the great panel does not satisfactorily allow for the Arjuna
interpretation. Therefore, my loyalty remains undividedly with King
Bhagratha. In what follows, I argue for the Bhagratha interpretation
and against the Arjuna interpretation.
From [Aokavarm] descended the powerful, spotless race of the
Pallavas . . . which resembled the descent of the Gag (on earth),
as it purified the whole world.8
This passage is from the Kakkui Copper Plate Grant of
Nandivarm Pallavamalla (8th century A.D.). The comparison made
between the advent of the Pallava race and the descent of the Gag had
already been given a graphic and concrete form a century earlier in the
Great Penance Panel of Mmallapuram.
C. Minakshi pointed out to scholars, many years ago, another
graphic representation of this same idea in the series of sculptured stone
panels in the Vaikuhaperum temple, Kanchipuram, which illustrate
the history of the Pallava race. Describing the fourth panel in the upper
row to the left of the entrance, she wrote:
 The . . . idea that the Pallava race resembled the descent of the 161
Ganges is expressed by the artists by depicting a man, obviously Notes on Mmallapuram
Bhagratha, performing penance just as in the Gagvataraa scene
on the rock at Mmallapuram. Resting on one foot, . . . his ja and
beard and his uplifted arms mark him out as one in severe penance.9
A third Pallava representation of Bhagratha (chronologically
midway between the Penance Panel and the Vaikuhaperum panel) is
found in the faade sandstone carving of the enclosure shrine No. 50 of
the Kailsantha temple, Kanchipuram. There can be no doubt that this
figure, standing on one foot, with upraised hands and ja hairstyle, is
Bhagratha, as the main figure of the same panel is iva-Gagdhara.
Is it possible to find an unequivocal, Pallava representation of
Arjuna in penance which will similarly parallel the debated figure in the
Mmallapuram Penance Panel? The answer is a clear-cut No. There
is only one unquestionable appearance of Arjuna in the whole range of
extant Pallava art, and that is in the faade sandstone carving of the
enclosure shrine No. 16 of the Kailsantha temple, Kanchi. In this
panel, Arjuna is portrayed fighting with iva, who is disguised as a
huntsman (Kirta). The boar, which is at issue in this fight, is shown
prominently at the bottom of the panel.
In our book, Mahabalipuram Studies (1974), we mentioned in
passing, in the Introduction (in a footnote), that the Penance Panel of
Mmallapuram was first interpreted as the Descent of the Gag by
V. Goloubew in 1914, and that the
point which is absolutely fatal to the Arjunas Penance interpreta-
tion is the fact that some of the heavenly beings actually have their
backs to Siva as he grants the boon to the ascetic who is supposed-
ly Arjuna. The problem vanishes if it is the descent of the Ganga
which is the centre of attention (the boon granted to Bhagiratha).10
Having pronounced on this matter in a somewhat off-hand
manner, we were censured by a reviewer of the book. We had, the
critic said, endorsed the claim
that the great open air bas relief represents Bhagirathas penance.
It might have been thought that the identification with Arjunas
penance is final and complete after Mr. T.N. Ramachandrans
study of Bharavis Kiratarjuniyam. It is disheartening that
scholars should continue to argue about it.11
Disheartening or not, the debate continues, and there are many
who disagree with Ramachandran and such like-minded scholars as C.
Sivaramamurti. These two scholars would interpret the Penance Panel
as a gigantic and detailed illustration of Bhravis Kirtrjunyam (the
famous Sanskrit poetic composition dealing with Arjunas penance and
his ensuing combat with iva, who took the form of a hunter [kirta]).
There is no doubt that these two scholars are backed by a deep
knowledge of Sanskrit sources, but the vessel of their argument, con-
structed as it is out of speculative comparisons, is destined, in my
opinion, to be shipwrecked on that fatal rock of objective fact which
we have footnoted in Mahabalipuram Studies.
 162 Let us take a closer look at this question. Ramachandran puts
Pallava Art it thus:
A rocky fissure has been turned into a natural causeway such as
would suggest a river course and the right half of the relief is filled
up with beholders, participants and applauders of the grand event,
the event in the present case being Arjunas penance, victory and
reward. This event was witnessed by the whole creation of the
Lord of the three worlds.12 [Italics added.]
And Sivaramamurti, in the official guide book on Mmalla-
puram published by the Archological Survey of India, writes:
Arjunas Penance: This magnificent carving is unique in the
range of Indian art. Two large boulders with a narrow fissure in
between have been chosen to represent a series of rows of gods
and goddesses like Chandra, Srya, pairs of Kinnaras and Siddhas,
Gandharvas, Apsaras, etc., rushing towards a central point near
the cleft where a sage stands on his left foot deeply engaged in
penance. . . .13 [Italics added.]
Now, both of these learned gentlemen are contradicted by the
fact that just at the foot of the man doing penance are two heavenly
couples flying by with their backs to what Ramachandran calls the
grand event supposedly Arjunas penance.
This contradiction is stunningly clear in the Minor Penance
Panel (near the light house), where iva and the penitent figure are
isolated in the upper left-hand corner of the relief, and all the creatures
of the three worlds (animals, humans, and demigods), which are
portrayed under them and to their left, have their backs to iva and
instead have their attention focussed on, and are moving toward, the
cleft to the right, which represents the path of the descending Gag!
These observations may have been made by others before us;
but they need to be repeated. And the proponents of the Arjunas
Penance interpretation must be specifically challenged to explain the
above-mentioned anomaly in their interpretative framework. To my
knowledge, Ramachandran and Sivaramamurti never gave such an
explanation, in spite of their elaborate theorizing.
Let me next take up an objection put forward by Ramachandran
to the Gagvataraa interpretation. He says that iva (in the Great
Panel) is by no means iva as Gagdhara:
Gangadhara must be Sivas form if we accept the theory of
Bhagirathas penance. As Gangadhara he should stand with his
right leg planted vertically on the earth and the left slightly bent.
His upper right arm should be raised to support a braid of his locks
on which river Ganga descends or settles (cf. Trichinopoly cave
temple and Adivaraha Cave).14
Ramachandran, evidently, was not familiar with Pallava
Gagdhara images, for his prescription is inaccurate on every point
with relation to the majority of their Gagdhara panels. As a matter of
fact, the Tiruchi Gagdhara image, which he himself refers to, has
iva with his left foot planted solidly, and his right leg bent; and nine
 163
14

20

The Great Penance Panel (central cleft area), Mmallapuram

 164 out of eleven Pallava Gagdhara panels have ivas left hand raised
Pallava Art to hold his locks. But all these details are neither here nor there.
Why should the Pallavas have to portray iva-Gagdhara in order to
satisfy the Gagvataraa theme? It would only be an anachronistic
imposition of the later rigidity in art traditions on the creative freedom
of the Pallava artists. In fact, in this particular case, such a requirement
would have resulted in the ludicrous juxtaposition of an anthropo-
morphic form of iva, a few feet tall, with the actual torrents of a real
(but artificially created) waterfall (which the Pallava engineers had
provided) dropping fifty feet from top to bottom of the central cleft.
No, the Pallava artists chose to represent iva at the moment he appears
before Bhagratha to assure him of the boon. This event precedes the
episode in which iva takes the form of Gagdhara. In the Penance
Panel, the Gagdhara form is skipped over, and the grand, climactic
event of the Gag reaching the earth (with a real waterfall) is shown.
There is no difficulty in the Indian art tradition of thus showing chrono-
logically distinct episodes in one and the same panel.
How very popular the Gagvataraa theme was with the
Pallavas, may be indicated by the following list of Gagdhara panels
which have survived from the Pallava period.
1. In Mahndras Tiruchi cave-temple; this is the first major
stone sculpted panel of the Pallavas and of Tamil Nadu!
2. In the divarha cave-temple, Mmallapuram.
3. In the central niche, north side, second level, of the
Dharmarja Ratha, Mmallapuram.
4. In the central, west-facing lateral shrine of the Kailsantha
temple, Kanchi.
5. In the same temple, on the outer wall of the main sanctum.
6. In the same temple, the faade panel of the enclosure shrine
No. 24.
7. In the same temple, the faade panel of the enclosure shrine
No. 50.
8. In the mukha-maapa of the Mtagvara temple, Kanchi.
9. On the north side of the vimna (outside) of the same temple.
10. In the mukha-maapa of the Muktvara temple, Kanchi.
11. On the north side of the vimna (outside) of the Iravtanvara
temple, Kanchi.
As against these eleven panels, many of which are of impres-
sive size and in important locations, there is the sole instance of Arjuna
fighting with iva portrayed in an enclosure shrines faade panel in the
Kailsantha temple. This panel cannot compare in importance, for
instance, with the Gagdhara panel in the lateral shrine of the same
temple.
If the significance of all this is not lost, and we recognize the
Penance Panel of Mmallapuram for what it is Bhagrathas penance
and reward then we can appreciate the impress which this spectacle
made down the ages.
 The Cha emperor, Rjndra-I, proclaimed, in his Tiruvlagu 165
Copper Plate Grant, that he, Notes on Mmallapuram
the light of the solar race, mocking Bhagratha who by the force of
his austerities caused the descent of the Ganga, set out to sanctify
his own land with the waters of that stream brought by the strength
of his arm.15
In bringing back water from the Gag in golden vessels
carried on the heads of the rulers defeated during his victorious march
to the North, and then in ceremoniously pouring it into the great man-
made lake at his capital city, Gangaikondacholapuram, Rjndra meant
not only to mock Bhagratha, but, for us, more significantly, to mock
the Pallavas and their Mmallapuram make-believe Gag flowing
down into the small pool below.
Empires have come and gone. Fortunately for us, Bhagrathas
Penance Panel at Mmallapuram has survived.
______________
1These notes, except for the last one, are based on a paper,
Mamallapuram Assorted Observations, by M.C. Lockwood, read at
the Symposium on Mahabalipuram held in Washington, D.C., Jan. 31 to
Feb. 3, 1979, organized by the American Committee for South Asian
Art. This paper was subsequently published in The Madras Christian
College Magazine, Vol. XLVIII, 1979, pp. 41-44.
2The Sunday Standard, Madras, October 1, 1972.
3For the detailed analysis which forms the basis of our present
observations, see the earlier studies in this book.
4The Indian Express, Madras, February 28, 1970.
5The Sunday Standard, Madras, April 16, 1972.
6The Sunday Standard, Madras, October 1, 1972.
7This last Note is based on a paper of the same title submitted
by M.C. Lockwood to the Symposium on Mahabalipuram (Jan. 31 to
Feb. 3, 1979, Washington, D.C.), published subsequently in the
crya-Vandan D.R. Bhandarkar Birth Centenary Volume, ed. by
S. Bandyopadhyay (Calcutta: University of Calcutta, 1982), pp. 272-
276, and which then formed the second study in the book,
Mmallapuram and the Pallavas (1982).
8South-Indian Inscriptions, Vol. II, Part III, p. 355.
9C. Minakshi, The Historical Sculptures of the Vaikuha-
perum Temple, Kch, being Memoirs of the Archological Survey
of India, No. 63 (Delhi: Archological Survey of India, 1941), p. 9.
10Lockwood, Siromoney, and Dayanandan, p. 6.
11The Indian Express, Madras, December 28, 1974.
12T.N. Ramachandran, Mmallapuram, Marg, Vol. XXIII,
No. 3 (June 1970), p. 36.
13C. Sivaramamurti, Mahabalipuram, third edition (New Delhi:
Archological Survey of India, 1972), p. 21. Sivaramamurti has
 166 indicated his preference for the Arjuna Penance interpretation else-
Pallava Art where in more scholarly publications. See, for instance, his Early
Eastern Chalukya Sculpture being the Bulletin of the Madras Govern-
ment Museum: New Series General Section, Vol. VII, No. 2 (Madras:
Madras Government Museum, 1962), pp. 42-46.
14Ramachandran, p. 50.

15S.-I.I., Vol. III, p. 109.

 FIFTEEN

Notes on Pallava Art1

1. Smskanda
Since the 1974 publication of our study on Pallava Sm-
skanda, friends of ours have discovered three more important exam-
ples of the Pallava Smskanda.
i. Kanchipuram, in the aknthan Talvara temple
Mr. A. Ekambaranathan directed us to a small shrine, the
aknthan Talvara, in Kanchipuram, which is situated to the north-
west of the kmbarantha temple. The aknthan Talvara is
locally called the vara. The shrine itself is a modern structure,
but placed on the inner back wall is an ancient carved Smskanda
panel which certainly belongs to the classical Rjasiha style and
Rjasiha period.
ii. Periya Venmani, loose lying panel [photograph, p. 66]
Ms. R. Champakalakshmi and Mr. A. Swami noticed a carved
stone Smskanda panel lying near two old brick temples in the village
of Periya Venmani, Madurantakam Taluk, Chingleput District.2 This
panel is approximately 3 ft. 6 in. in height and 3 feet in breadth. iva is
four-armed. His upper right hand holds the shaft of a trident; his upper
left, the shaft of an axe. In both these hands the shafts are grasped by
the tips of the index fingers and thumbs, the other fingers being folded
downwards, except for the little fingers which again point upwards.
ivas lower right hand rests in a clenched fist on his right
thigh. His lower left hand holds a flower. The positioning of ivas
two lower hands and his legs are almost an exact mirror image of the
positioning of the same limbs of iva in the pre-Rjasiha Smskanda
of the Dharmarja Ratha, Mmallapuram, which in turn reflects the
posture of King Sihaviu in the portrait sculpture of that king found
in the divarha cave-temple of the same place. In the Periya Venmani
Smskanda panel, Um has her left leg down at almost the same angle
as in the Dharmarja Ratha panel. These are thus similarities which tie
in with a pre-Rjasiha style.
The axe (usually held by ivas upper right hand) is common in
post-Rjasiha style Smskandas. But the trident is unique. iva
wears a stomach band (udarabandha) and both his ears have makara
type ear ornaments. He has no leg ornaments. He wears the vi (long
lower garment), which reaches down to the ankle of his left leg.
Ums torso is twisted toward the viewer, whom she faces. This
attitude is in keeping with the Rjasiha style Smskandas. Her left
hand is on her left hip; her right hand supports the infant Skanda,
 168 who is seated on her knees (a characteristic which is shared with the
Pallava Art pre-Rjasiha Smskanda of the Dharmarja Ratha). Um wears a
patra-kuala in each ear. Ums wearing two patra-kualas is a
standardization reached in the Rjasiha style Smskandas. Further-
more, the diameter of these earrings is relatively small, a characteristic
also in keeping with a Rjasiha period date. Ums hair is done up in
the karaa-makua style. This hairstyle for Um in a Smskanda is a
post-Rjasiha characteristic. But the lower band of hair on Ums
head is pinched in the middle just above her forehead and this peculi-
arity is characteristic of the Rjasiha style. Because of this single
characteristic, I would not place the Periya Venmani panel in the pre-
Rjasiha period.
Thus, we see that the various characteristics of the Periya
Venmani Smskanda oscillate between pre-Rjasiha and Rjasiha
characteristics, with one or two post-Rjasiha characteristics thrown
in to confuse the issue. I would tentatively date this Smskanda in the
late Rjasiha period, and see it as carrying over some of the character-
istics of the pre-Rjasiha style.
Behind the throne, mid-way between the heads of iva and
Um, appears one of the two animals connected with the goddess,
especially in her Durg form. It is a deer, with an attendant bearded
sage having a ja hair-style. That this animal is a deer is clearly shown
by a comparison of it with the deer in the contemporaneous panel of
Mahiamardin, also from Periya Venmani. The heads of the two
animals are practically identical. The fact that Durg is to be identified
with ivas consort, Um, in Pallava art, is thus established quite
conclusively.
A unique aspect of the Periya Venmani Smskanda panel is
the appearance of the liga (as a smooth cylinder with hemispherical
top) just behind ivas right shoulder. We have discussed the signifi-
cance of this aspect elsewhere in this book in our study of iva as
Ligin in a Pallava Smskanda.
This appearance of the liga recalls to my mind the various
small sculptured panels (e.g., 30 x 20 cm.) found in the Toaima-
alam (Pallava) area, at Kanchipuram, Munnur, Manimangalam, Teneri,
Madurantakam, Uttaramerur, Ukkal, and Brahmadesam.3 Typically,
these small panels show seated in a row on a common throne (which
here appears as a long bench-like sana) the following deities: Brahm,
the aniconic liga, Um, Subrahmaya (Skanda grown up!), and Viu
in the form of Narasiha. These small panels, some of which are Sati
stones, are thus transmuted iva-Smskanda panels, iva being repre-
sented only in the liga form, and Skanda being shown full-grown, in-
stead of as an infant. I would therefore suggest that these panels be
dated sometime after the early Smskandas belonging to the Param-
vara and Rjasiha reigns at the end of the seventh and beginning of
the eighth centuries A.D. The Periya Venmani Smskanda panel can
be viewed as an important link between the standard type of Pallava
Smskanda and later variations or mutations of it.
iii. Vallam (near Sriperumbudur), Sadaya Temple 169
Mr. V. Narayanaswamy discovered another Pallava Sm- Notes on Pallava Art
skanda panel in a small shrine in the village of Vallam, situated about
six miles to the south-east of Sriperumbudur, on the road to Chingleput
town.4 The shrine is called Sadaya. The Smskanda panel is found
inside the sanctum, on the wall behind the liga. It is of the classical
Rjasiha style. We note that the umbrellas garland is directly above
Ums head; and a crescent moon is carved to the immediate proper
right of the top of ivas ja-makua.
2. Lion-face Buckle
In the sixth study of our book, Mahabalipuram Studies, we
mentioned, and illustrated, one of the earliest examples of the lion-
face belt buckle, on a Viu figure of the Vaikuha-Perum temple,
Kanchipuram, built by King Nandivarm-II, Pallavamalla, in the
eighth century A.D. However, K.R. Srinivasan, in 1964, had pointed
out an example of the siha-mukha (lion-face) clasp or buckle on the
dvrapla carved on the eastern end of the faade of King Mahndras
Mandagapattu cave-temple.5
Since the Mandagapattu cave-temple is usually considered
Mahndras earliest, we thus have an example of the lion-face buckle
in a Pallava monument excavated around the beginning of the seventh
century A.D.
3. Pallava Paintings
In King Rjasihas temple at Panamalai, there are fragment-
ary remains of paintings on the inner walls of one of the lateral shrines.
On the inner back wall of this shrine, in the central and most important
position, only an outline remains of the major painting of iva dancing.
From this very fragmentary outline, one can make out iva in the
dancing pose called lha. The stance called lha, in Sanskrit, is the
position taken by an archer when he kneels on one knee and keeps the
other leg advanced with that foot squarely on the ground. (It is the half-
kneeling stance taken by a person being knighted.) iva is said to strike
the lha stance in his victory dance after having destroyed the
Tripras with his mighty bow and flaming arrow.
On the inner flanking wall (to the proper left of iva) is a less
fragmentary painting of Um standing, watching iva dancing.
A carved sandstone panel in a niche in the Kailsantha temple,
Kanchipuram, shows clearly the same dance pose of iva, and in a side
niche (to the proper left), Um is seen standing and watching iva
dancing.
Some of the small shrines which surround the courtyard of the
main shrine have patches of paintings on the inner walls of their cells.
The French scholar, G. Jouveau-Dubreuil, discovered these remnants in
the early part of the twentieth century.
One may ask whether these fragments belong to the Pallava
period. What would be the methods of dating such early paintings?
Since the paintings are found on shrines built by King Rjasiha,
 170 obviously the paintings cannot be older than the buildings. But how
Pallava Art could one establish that they are not much later?
One method would be to find out whether there existed other
layers of paintings underneath the visible painting. That layer nearest
the stone surface underneath would be the earliest. However, there is
no certain way of knowing whether the original layer had not peeled off
and someone had painted the stone surface subsequently. As a matter
of fact, there is a Chola inscription engraved on the inner stone wall of
one shrine which was subsequently covered by thick plaster till the
plaster fell off very recently.
Another method would be to date a painting on general stylistic
grounds. This method would require several samples of well-dated
works of art from other places for a satisfactory comparison. Unfortun-
ately, in the Pallava period, the only other known examples of paintings
are found in Rjasihas Panamalai temple. And the same questions
could arise with regard to the age of the Panamalai paintings. There-
fore, Gift Siromoney and I suggested, in an article in 1975, a compari-
son of the paintings with Rjasihas sculptural works with respect to
the dress and ornaments of the figures.6
In one of the enclosure shrines of the Kailsantha temple, there
is a notable fragment of a painting portraying the Smskanda theme.
We proceeded to show that this painting does indeed belong to the
Rjasiha period. There are several significant parallels between this
painting and sculpture of the Rjasiha period in terms of character-
istics peculiar to this period. For instance, iva is shown seated on a
throne with only his left leg extending down. This is a standard charac-
teristic of the Rjasiha style sculptured Smskandas, and is found in
more than twenty examples of the Smskanda panel in this very same
temple. In the post-Pallava period that is, throughout the Chola and
Vijayanagar periods, iva is always portrayed with only his right leg
down in Smskandas.
Again, in the Kailsantha painting, the maid at the feet of
Um wears a breast-band with shoulder straps of the Rjasiha style.
(Um is not shown with a breast-band, but her torso is represented as
colored with sandal paste.) And again, Um is shown wearing a sr
from the waist down, just covering her knees. This is exactly as
depicted in the Smskanda sculptures of the Rjasiha period.
The parallels in dress and ornaments between the paintings of
Smskanda, on the one hand, and sculptures belonging to Rjasiha,
on the other, were very close except for two peculiar ornaments por-
trayed in the painting. Both are found on Ums arms, above the
elbow. One is an unusual upper arm-band with spaced rosettes. The
other is a simple single band worn just above the elbow. This elbowlet
is similar to those so commonly found on figures of the Chola period.
From the time we first noted the elbowlet in the Kailsantha
Um (detail of painting) painting, we began to wonder whether this painting was indeed contem-
Kailsantha, Kanchi poraneous with the temples construction.
However, while examining a niche of the main shrine, we
came across both the unusual ornaments on a sculpture of Um. This 171
figure is found in a niche next to that of iva dancing. It is reached Notes on Pallava Art
through the front hall of the main shrine. The sculpture in this niche is
part of the original temple. Thus, the parallel between painting and
sculpture is complete. The unusual rosetted upper arm-band found on
both painting and sculpture is striking confirmation that the painting
belongs to the Rjasiha period.
Another important outcome of this discovery is that the
elbowlet, so common on Chola figures, clearly appears in sculpture
(and painting) of the Rjasiha Pallava period, around 700 A.D. This
appearance is some two hundred years before the coming to promin-
ence of Chola art.
4. Earliest Sculpture of Kaappa7
Tirukkalukkunram is a town situated between Chingleput and
Mmallapuram. The town lies at the foot of a low range of four hills
which are said to represent the four Vdas.
Um (detail of sculpture)
On the highest of the four, the sacred Vdagiri, there are two
Kailsantha, Kanchi
Pallava temples, one of which is the picturesque hilltop shrine, the
Vdagirvara, with the nearby noon-time feeding of the sacred birds.
There is also an important temple complex at the foot of the hill,
the Bhaktavatsala. An inscription on the prkra wall of this temple
states that in the 9th year of Javarm Sundara Pya, around 1260
A.D., the present main shrine of the Bhaktavatsala was built.
The Bhaktavatsala has for its strong room, however, a struc-
ture which is far older than its main shrine. This room is, in fact, the
garbhagha of a ninth century apsidal temple belonging to the late
Pallava period. The upper storey of this temple is now missing, and its
garbhagha has evidently been repaired and given a flat roof in more
modern times. Just under its cornice, however, remains an original
feature: an interesting row of small, sculptured figures, impish and pot-
bellied, called bhtagaas. This group of figures is especially import-
ant because it contains the earliest sculptural representation of the story
of Saint Kaappa so far discovered.
It was on a recent trip that we discovered amidst all the frolick-
ing gaas a clear portrayal of Saint Kaappa, who is here also one
of them. This ardent devotee of iva is shown kneeling next to a liga,
ready to gouge out his right eye with an arrow held in his right hand.
Among the earliest references to the Kaappa theme are
brief passages in the seventh century hymns of Appar and Sambandhar.
The basic point being made in the Kaappa story is that the intense
devotion of the rough hunter, Kaappa, was as pleasing to iva as
the more refined worship of the orthodox priests. Brief references in
akaras ivnandalahari and Sundaras Tiruttoattogai, in the
eighth and ninth centuries, led up to the more detailed twelfth century
account in Skkirs Periya Puram.
It will be evident, then, that the ninth century sculptured repre-
sentation of Kaappa which we have noticed at Tirukkalukkunram
takes one back more than a thousand years, and is very close to the
 172 period of the earliest literary references to Kaappa. Heretofore,
Pallava Art the earliest known sculptural representations of Kaappa have been
bronze images assigned to the late tenth and the eleventh centuries.

15.5

The relevant figures in the Tirukkalukkunram panel, from

left to right, are as follows. First, comes a hunter (who is Kaappa)
shouldering a pole with two pigs suspended by the hind legs, one at
each end of the pole. Next, is a hunting dog, with its head turned back
towards the pigs. Then, we see Kaappa kneeling next to the iva-
liga, ready to gouge out his right eye with the tip of an arrow (the bow
is portrayed just below the arrow, and above the dogs head). A hand
can be seen projecting out of the liga, indicating to Kaappa that he
should desist from his extreme act of self-sacrifice. On the other side
of the liga are two figures (also gaa-like) representing the orthodox
priesthood, shouldering baskets of flowers for worship, and holding
lotuses in their left hands.
_______________
1Notes on Pallava Art formed the tenth study in the book,
Mmallapuram and the Pallavas (1982).
2Indian Express, Madras, February 4, 1972. This Smskanda
panel was discussed in greater detail later in 1972 (though the journal is
pre-dated, Jan.-Jul., 1969) in an article by the same authors published in
the Journal of the Madras University, Vol. XLI, Nos. 1 & 2, pp. 129-
137, and fig. 3.
3Damilica, I, Dec. 1970, pp. 1-2, figs. 6a & 6b.
4Reported in an article, A new Pallava Somaskanda, The
Sunday Standard, Madras, April 8, 1979.
5Cave-Temples of the Pallavas (New Delhi: Archological
Survey of India, 1964), p. 50; Pl. III-A shows this dvrapla, but the
details of the buckle are not distinguishable in the photograph.
6Pallava paintings of Kanchipuram, Indian Express, Madras,
September 20, 1975. This part of the study is based on that article.
7This last note is based on an article of the same title, by M.C.
Lockwood and Gift Siromoney, first published in The Indian Express,
Madras, March 3, 1977.
 SIXTEEN

Royal Titles of
Rjasiha and Mahmalla
Two sets of inscribed royal titles (or birudas) are given below.
The first set, 252 titles of King Narasiha-II (Rjasiha), is from the
Kailsantha temple, Kanchipuram. The second set is from the
Dharmarja Ratha, Mmallapuram, and belongs to King Narasiha-I
(Mahmalla).
Each set of titles is first given in the order in which the inscrip-
tions are located on the temple structures. Next, the same titles are
given in alphabetical order along with a translation into English.

Rjasihas titles are engraved in four tiers on the faades of the

many little shrines which form an enclosure around the main structures
of the Kailsantha temple. The first (and uppermost) tier is formed by
granitic stone slabs, and because of this hard medium, the inscribed
titles on this level have been well preserved. (It should be noted that
the script used on this level is a southern variety of Ngar not the
usual Pallava Grantha.) The lower three tiers are of soft sandstone.
The inscriptions on these have, in many places, been badly weathered,
and therefore the titles are often fragmentary or missing altogether.
From the fragments, however, it appears that the lower three tiers
usually but not always repeat titles which are found on the first tier.
The script of the second tier is simple, plain Pallava Grantha. In the
third tier, a florid Pallava Grantha. And in the fourth and lowest tier,
an extremely florid Ngar more decorative than readable!
In the alphabetical list of King Rjasihas titles, the three
numerals in brackets after each title indicate, first, the tier, then the
shrine number, and, finally, the serial order of the given title. For the
shrine numbers, I have used the numbers which have been engraved on
the shrines by the Archological Survey of India. There are five titles
which are engraved on the small gpura which is situated between
shrines Nos. 29 and 30. In the alphabetical list, I have used the initials
UG (Unnumbered Gpura) to indicate the location of these five titles.
There are eleven titles of Rjasihas ending in short i (plus )
which the first editor of these inscriptions, E. Hultzsch, noted and said
should be corrected to the long . I would like to point out that this

_______________
*This study is based on Appendix A of Lockwoods
Mmallapuram and the Pallavas.
 174 shortening is no simple scribal error. Instead, it represents the strong
Pallava Art influence, in the Tamil country, that the Tamil language and its scribal
conventions had on the writing of Sanskrit. The eleven titles are:
gamnusri (I: 39.3)
vijayi (III: 30.2)
havaksari (I: 8.3)
Khinnnukampi (I: 10.1)
Gandhahasti (I: 50.1)
Daridrnukampi (I: 18.2)
Dradari (I: 44.1)
Dharmmavijayi (I: 42.4)
Naynusri (I: 44.3)
Vikramaksari (I: 57.3)
Vraksari (I: 14.3)
One other title in this series which was influenced by Tamil
is Lkaikmai (I: 53-4), where correct Sanskrit would have the
aspirate kh in ikhmai.
In the Dharmarja list, the following titles have been influ-
enced by Tamil:
Softened Other
Correct Sanskrit Actually Inscribed Consonant Changes
Pthvsra Pridhivisra th to dh to ri
to i
Ankpaya Ankbhaya p to bh
Parpara Parbhara p to bh
Parpara Parvara p to v
Bhuvanabhjana Bhuvanabhchana j to ch
* * * * * * * *
Postscript (1997):
K.G. Krishnan, begins his article, Convertibility of Surds and
Sonants Historical Evidence (Indo-Iranian Journal, Vol. XIV, No.
3/4 [1972], pp. 241-46) with these statements:
Professor Kuiper has summarised clearly the results of the attempts
of scholars made so far both in favour of and against the theory of
the convertibility of Surds and Sonants originally propounded by
Caldwell. He has come to the conclusion that the modern opposi-
tion between a tense voiceless articulation of the plosives in initial
position, and a lax (more or less voiced) articulation with weakened
occlusion intervocally seems essentially to have existed already
about the beginning of our era.
Krishnan goes on in his article to cite examples from (1) Tamil
written in Kannada script, (2) Tamil in Grantha script, and (3) Tamil in
Ngar script, in support of Caldwells theory. All the examples he
gives date from around 1000 A.D. or later. The examples which I have
given above (in the softened consonants category), which date from
around 650 A.D., give further evidence supporting Caldwells theory.
Samples of Rjasihas Titles Inscribed in the Kailsantha Temple, Kanchi
All four tiers of a single enclosure shrine (No. 8) giving royal titles in four different scripts:

1. First (Uppermost) Tier, titles in early, plain Ngar script:

2. Second Tier, titles in plain Pallava Grantha script:

3. Third Tier, titles in florid Pallava Grantha script:

4. Fourth (Lowest) Tier, titles in extremely florid Ngar script:

176
Birudas of King Narasihavarm-II
(around the inside of the enclosure of the Kailsantha temple, Kanchipuram)
First Tier Titles
South side of east enclosure:
Shrine No.
1 r Rjasiha r Atyantakma r Raajaya r Abhirma
2 r Aparjita r Amitramalla r Akutbhaya r rjjita
3 r Jayapara r Atiraacaa r-bhara r Bahunaya r Udayabhskara
South enclosure:
5 r-mgha r Abhayakara r Kulatilaka r Arimarddana
6 r Uditaprabhva r Uditakrtti r abhadarppa r abhalchana r kac . . .
7 r Ugravryya r Uditdita r Unnatarma r Ugrapratpa
8 r Atyadra* r Anunayasddhya r havaksari*
9 r Kalakavarjjita r Kcmahmai r Kharavikrama r Cakravartt*
10 r Khinnnukampi* r Cpadvitya r Chinnasaaya r Chalarahita
11 r Amitrani r Apratimalla r Adbhutacarita r Ibhavidydhara
12 r Icchpra r naaraa r Udayacandra r Parjanyarpa
13 r Paracakramarddana r Narndraclmai r Nityavara r Rjarja
14 r Vdyavidydhara r Citrakrmmuka r Vraksari* r-kmuka
15 r Sarvvatbhadra r Katraclmai r-vilsa r Yuddhrjjuna
16 r-vallabha r Sagrmarma r Srvvabhauma r Katravidrvaa
17 r havabhma r Amitaprabhva r Trailkyantha r Dnavara
18 r Tpraa r Daridrnukampi* r Aviratadna r Dptapaurua
19 r Dnara r Dharmmanitya r Dhavalaya r Dharmmakavaca
20 r Samaradhanajaya r Bhaacpa r Ajjaya r Guavinta r Avanidivkara
r Kalakarahita r Kalsamudra r havadhra r Dadamana r Pallavditya
21 r Parpara r Parahita r Nitytsha r Puruasiha
22 r Puyalka r Prttavikrama r Bhmaknta r Bahudakia
23 r Bhayarahita r Mahmalla r Mattapramatta r Mattavikra
24 r Bhuvanibhjana* r Mahndraparkrama r Mahprabhva r Manucarita
West enclosure:
26 r Mycra r-pativallabha r Raavra r Yugntditya
27 r Raadhra r Rakmai r Raacaa r Raavikrama
28 r Atulabala r Ahitntaka r Apravikrama r Avapriya
29 r Apratima r Akhaasana r Akhaani r Amghavikrama
UG r natamaala r Apratihata r Adbhutaakti r jrasa r caryyavrya
30 r ptadurddhara r viyi* r havddhura r Ibhavatsarja
31 r Iddhasana r Ilparamvara r Ugradaa r Unnatamna
32 r Ucchritavryya r Udayatuga r Uttarttara r Ugrasana
33 r Gulaya r Udayavasanta r kasundara r Mahnubhva
 177

North Enclosure:
35 r Upndravikrama r pra r Kuladhvaja r Gunnata
36 r Unnatccha r Utkhtakaaka r kadhanurddhara r Udrakrtti
37 r crapara r rttyana r rtavatsala* r titana
38 r tdhyatumburu r gamapramna r jlakta r Itihsapriya
39 r Atishasa r Anavagraha r gamnusri* r Utthnala r Udaynnata
r Udvttadamana r karja r Klavikrama r Jayanidhi r Klavasana
r Garvvitadamana
40 r Jtigambhra r Cracaku r Jnkua r Taptaaraa
41 r Damitavyla r Dnavara r Dvadvabhakta r Durvvravga
42 r Cruvilsa r Tugavikrama r Tvrakpa r Dharmmavijayi*
43 r Dvgni r Davarddhana r Dradurita r Dharmmastu
44 r Dradari* r Dptasana r Naynusri* r Nayanamanhara
45 r Anindyacarita r Agdhagmbhryya r Anabhravi r Atanupratpa
46 r Adharmmabhru r Arina r Avanibhjana r Aprativryya
47 r Avandhyakpa r Amitrntaka r Avihataakti r Anavagta
48 r Artikla r Anavagraha r Atishasa r Anugrala
49 r Abhayari r hatalakaa r Utshanitya r Upyanipua
50 r Gandhahasti* r Kmavilsa r Kviprabdha* r Kraakpa
51 r Caadaa r Asahyakpa r Chyvka r Dharaitilaka
52 r Varuapa r Dhairyyasgara r Pravttacakra r Ngapriya
53 r Niramitra r Nirarggala r Parantapa r Lkaikmai*
54 r Prttivasi* r Balaprama* r Bhridna r Pratibhaya
North side of east enclosure:
56 r Bhmavikrama r Rjakujara r Lalitavilsa r stradi
57 r Vraabhagadatta r Viktavilsa r Vikramaksari* r Vinrada*
58 r akarabhakta r rgragaya r Tatvavd r varabhakta

Fourth Tier Titles

Shrine No. Shrine No.
2 r Atyantakma r Amitramalla 13 r Kharavikrama r Khinnnukamp
3 r Guavinta r Aparjita 14 r Cakravartt r (Cpa)dvitya
5 r Avanidivkara r rjita 15 r Amghaba r Asahyamrggaa
6 r Uditaprabhva r Uditakrtti 16 r Ugrasyaka r Uddhataviikha
7 r Kalakarahita r Kalsamudra 17 r Bhmakrmmuka r Bhaacpa
8 r Ugra(v)rya r Uditdita 18 r Avismita r Amitrani
9 r Atyudra r Anunayas(ddhya) 19 r Iavara r Indralla
10 r Unnatarma r Ugraprat(pa) 20 r Amitra(marddana*) r jimarddana
11 r havadhra r havaksar 21 r Duadamana r Durutsaha . . .
12 r . . . r Klakpa
 More Examples of Rjasihas Titles Inscribed in the Kailsantha Temple

On enclosure shrine No. 9:

On enclosure shrine No. 10:

 179

Alphabetical List
(of Royal Titles in the Kailsantha Temple, Kanchipuram)

A
Akutbhaya (I:2.3) Always the Fearless [Also the title of a commentary by the Buddhist, Ngrjuna!]
Akhaasana (I:29.2) Unswerving Rule
Akhaani (I:29.3) Mighty Thunderbolt (literally: an unbranched, single, solid bolt [cf. Vilviugu
Agdhagmbhryya (I:45.2) Unfathomable Profundity and Pakppiuku])
Ajayya (I:20.3) The Invincible
Atanupratpa (I:45.4) No Small Prowess
Atiraacaa (I:3.2) The Exceedingly Fierce in Battle
Atishasa (I:39.1 & 48.3) The Exceedingly Daring
Atulabala (I:28.1) Matchless Strength
Atyadra* (I:8.1) [A misspelling; see: Atyudra, below]
Atyantakma (I:1.2 & IV:2.1) Boundless Desires (not limited, in the sense of being selfish)
Atyudra (IV:9.1 & misspelled Atyadra in I:8.1) The Exceedingly Noble
Adbhutacarita (I:11.3) (He of) Astonishing Deeds
Adbhutaakti (I:UG.3) (He of) Astonishing Strength
Adharmmabhru (I:46.1) Fearing (only) Injustice
Anabhravi (I:45.3) Cloudless Showerer (of benefits)
Anavagta (I:47.4) The Irreproachable [See line 44 of Mattavilsa]
Anavagraha (I:39.2 & 48.2) The Unrestricted
Anindyacarita (I:45.1) (He of) Blameless Behavior
Anugrala (I:48.4) (He of ) Gentle Character
Anunayasddhya (I:8.2 & IV:9.2) (He who is) Won by Gentle Means (only)
Aparjita (I:2.1 & IV:3.2) The Invincible
Apravikrama (I:28.3) (He of) Boundless Valor
Apratimalla (I:11.2) The Unchallenged Wrestler
Apratima (I:29.1) The Incomparable
Aprativryya (I:46.4) The Irresistible
Apratihata (I:UG.2) The Imperishable [One of the names of Viu]
Abhayakara (I:5.2) The Creator of Safety
Abhayari (I:49.1) The Ocean of Safety
Abhirma (I:1.4) The Charming
Amitaprabhva (I:17.2) (He of) Unlimited Power
Amitramarddana (IV:20.1) The Foe-Crusher
Amitramalla (I:2.2 & IV:2.2) The Foe-Mauling Wrestler [Cf. Mahndras title, atrumalla]
Amitrntaka (I:47.2) The Foe-Destroyer
Amitrani (I:11.1 & IV:18.2) A Thunderbolt (to his) Foes
Amghaba (IV:15.1) (One whose) Arrows (do) not (fly) in Vain
Amghavikrama (I:29.4) (He of) Not Unavailing Valor
Artikla (I:48.1) Death (to his) Enemies
Arina (I:46.2) Destroyer (of his ) Enemies
Arimarddana (I:5.4) Crusher (of his) Enemies
Avanidivkara (I:20.5 & IV:5.1) Sun of the Earth
Avanibhjana (I:46.3) Receptacle of the World [Mahndras title also; see Mattavilsa]
 180

Avandhyakpa (I:47.1) (He whose) Anger (is) not Fruitless

Aviratadna (I:18.3) (He of) Unceasing Donations
Avismitah (IV:18.1) The Never Perplexed
Avihataakti (I:47.3) The Unquellable Power
Avapriya (I:28.4) A Lover of Horses
Asahyakpa (I:51.2) (He of) Unbearable Anger
Asahyamrgga (IV:15.2) (Follower of the) Arduous Path
Ahitntaka (I:28.2) Destroyer (of his) Enemies

gamapramna (I:38.2) (One whose) Authority (is) the gamas

gamnusri* (I:39.3) [Correct: sr] Follower (of the) gamas
crapara (I:37.1) (One who is) Devoted to Tradition
jimarddana (IV:20.2) The Crusher in Battle
jrasa (I:UG.4) (He who) Relishes (issuing) Orders [Adding one more rasa to the list of nine?]
jlakta (I:38.3) (He who is) Adorned (with the power of issuing) Orders
tdyatumburu (I:38.1) Tumburu with Musical Instruments [Mcchakaikam, Act V, l. 11]
natamaala (I:UG.1) (He to whom) the Provinces Bow
ptadurddhara (I:30.1) The Irresistible (in) Attacking
rttyana (I:37.2) The Refuge of the Distressed [One of the functions of a katriya]
pra (I:35.2) The Fulfiller (of) Hopes
vijayi* (III:30.2) [Correct: vijay] Successful (in attaining his) Desires [Incorrect: viyi, I:30.2]
caryyavryya (I:UG.5) (He of) Amazing Prowess
rtavatsala* (I:37.3) [Correct: rita] (He to whom) Refugees (are as his) Children [Cf. Trimrtis]
hatalakaa (I:49.2) (He of) Numerous Good Qualities [Cf. Bhagavadjjuka, l. 3]
havaksari* (I:8.3 & IV:11.2) [Correct: ksar] The Lion in Battle
havadhra (I:20.8 & IV:11.1) The Steady in Battle
havabhma (I:17.1) The Fierce in Battle [Bhma, a title of iva]
havddhura (I:30.3) The Unrestrained in Battle

I
Icchpra (I:12.1) The Wish-Fulfiller
Itihsapriya (I:38.4) (One who is) Fond of History
Iddhasana (I:31.1) (He of) Fiery Command
Indralla (IV:19.2) (He who in his) Exploits (resembles) Indra
Ibhavatsarja (I:30.4) (He who is like) the King of Vatsa [Udayana] (in dealing with) Elephants
Ibhavidydhara (I:11.4) Musician expert in capturing Elephants [as Udayana did]
Ilparamvara (I:31.2) Supreme Lord of the Earth [I in III:31]
Iavara (IV:19.1) Showerer of Desires [I.e., Fulfiller of Desires]

titana (I:37.4) The Protector from Epidemics

naaraa (I:12.2) (As) Protector (he is like) na (the Supreme Lord = iva)
varabhakta (I:58.4) The Devotee of vara (iva)
 181

U
Ugradaa (I:31.3) (He whose) Punishment is Terrible [Cf. Klidsas Raghu.: to maintain order]
Ugrapratpa (I:7.4 & IV:10.2) (He of) Terrifying Bravery
Ugravryya (I:7.1 & IV:8.1) (He of) Terrifying Prowess
Ugrasana (I:32.4) (He whose) Commands are Terrifying
Ugrasyaka (IV:16.1) (He whose) Arrows are Terrifying
Ucchritavryya (I:32.1) (He of) Heroic Prowess
Utkhtakaaka (I:36.2) (One who) Uproots Evildoers
Uttarttara (I:32.3) The Ever-Progressive
Utthnala (I:39.4) By Nature, Active
Utshanityah (I:49.3) The Ever-Active
Udayacandra (I:12.3) The Rising Moon (in beauty?)
Udayatuga (I:32.2) The Highly Exalted
Udayabhskara (I:3.5) The Rising Sun (in majesty?)
Udayavasanta (I:33.2) The Rising Spring (of happiness?)
Udaynnata (I:39.5) The Extremely Exalted
Udrakrtti (I:36.4) (He of) Exalted Fame (for his munificence) [Cf. Kirtrjunyam: Canto I:18]
Uditakrtti (I:6.2 & IV:6.2) (He of) Ever-Fresh Fame
Uditaprabhva (I:6.1 & IV:6.1) (He of) Ever-Freshly (manifested) Power
Uditdita (I:7.2 & IV:8.2) (One who is) Ever Progressive
Uddhataviikha (IV:16.2) (He whose) Arrows (i.e., defenses) are Ever-Raised (ever-ready)
Udvttadamana (I:39.6) The Subduer of Rebels
Unnatamna (I:31.4 & IV:10.1) (He of) Exalted Honor
Unnatarma (I:7.3) The Extremely Gracious
Unnatccha (I:36.1) (He of) Lofty Desires
Upyanipua (I:49.4) (He of) Adroit Diplomacy
Upndravikrama (I:35.1) (He who has) The Valor of Upndra (Viu)

rjjita (I:2.4 & IV:5.2) The Mighty

abhadarppa (I:6.3) (He who takes) Pride in the Bull (as his emblem)
abhalchana (I:6.4) (He whose) Emblem is the Bull

kadhanurddhara (I:36.3) Foremost among Archers (lit., . . . among Those who Hold the Bow)
karja (I:39.7) Foremost among Kings
kasundara (I:33.3) Foremost among the Handsome
kac . . . (III:6.5) [We suggest that this title is: kachatra = Foremost among Kings; cf. Klidsa]
 182

K
Kalakarahitah (I:20.6 & IV:7.1) The Spotless [Superior to the moon which has a blemish]
Kalakavarjjita (I:9.1) The Spotless [Superior to the moon]
Kalsamudra (I:20.7 & IV:7.2) The Ocean of Arts
Kcmahmai (I:9.2) The Great Jewel of Kanchi
Kmavilsa (I:50.2) (He who embodies the) Sport of Kma (Cupid)
Kraakpa (I:50.4) (One who gets) Angry (only for good) Reason
Klakpa (III:20.9 & IV:12.2) (He who has) The Anger of Death (Yama)
Klavasana (I:39.10) (He who has) The Appearance of Death (to his enemies)
Klavikrama (I:39.8) (He who has) The Valor of Death; or: (whose) Valor is Death (to enemies)
Kviprabhda* [Correct: Kavi] (I:50.3) (He of) Poetic Insight
Kulatilaka (I:5.3) The Ornament of his Clan
Kuladhvaja (I:35.3) The Banner of his Clan
Katraclmai (I:15.2) [cmai in III:15.2] The Crest-Jewel of Warriors
Katravidrvana (I:16.4) The Router of Warriors

Kh
Kharavikrama (I:9.3 & IV:13.1) (He of) Fierce Valor
Khinnnukampi* (10.1 & IV:13.2) [Correct: kamp] (He who is) Compassionate to the Distressed

G
Gandhahasti* (I:50.1) [Correct: hast] The Musth Elephant (i.e., one who is unrestrainable)
Garvvitadamana (I:39-11) The Subduer of the Haughty
Guavinta (I:20.4 & IV:3.1) (He who has) the Virtue of Modesty [!]
Gulaya (I:33.1) The Abode of Virtues
Gunnata (I:35.4) (He of) Exalted Virtues

C
Cakravartt* (I:9.4 & IV: 14.1) [Correct: vartt] Emperor
Caadaa (I:51.1) (He whose) Punishments are Fierce
Cpadvitya (I:10.2 & IV:14.2) (He whose) Constant Companion is the Bow
Cracaku (I:40.2) (He whose) Eyes are Spies [I.e., the spies (agents) who serve him are his eyes]
Cruvilsa (I:42.1) The Graceful Sport
Citrakrmmuka (I:14.2) [krmuka (III:14)] The Wonderful Archer

Ch
Chalarahita (I:10.4) (He who is) Not Rash
Chyvka (I:51.3) The Shade-Giving Tree
Chinnasaaya (I:10.3) (He of) Dispelled Doubts

J
Jayanidhi (I:39.9) The Treasure of Victory
Jayapara (I:3.1) (He who is) Determined on Conquest
Jtigambhra (I:40.1) (He whose) Profundity is Inborn
Jnkua (I:40.3) (He who uses) Wisdom (as his) Goad
 183

T
Tatvavd* (I:58.3) [Correct: vd] Philosopher
Taptaaraa (I:40.4) The Refuge of the Distressed
Tvrakpa (I:42.3) (He whose) Anger is Fierce
Tugavikrama (I:42.2) The Eminently Valorous
Tpraa (I:18.1) The Fulfiller of Desires
Trailkyantha (I:17.3) The Lord of the Three Worlds

D
Damitavyla (I:41.1) (One in whom there is) No Trace of Villainy
Daridrnukampi* (I:18.2) [Correct: kamp] (He who is) Compassionate to the Poor
Dnavara (I:17.4 & 41.2) The Showerer of Gifts
Dnara (I:19.1) (He who is) Keen on Giving Gifts
Dvgni (I:43.1) The Forest-Fire (to his enemies) [Davgni was a title of Mahndras]
Dptapaurua (I:18.4) (He of) Brilliant Manliness
Durutsaha (I:21.2) (Of) Unthwartable Resolution
Durvvravga (I:41.4) The Unthwartable Force
Duadamana (I:20.9 & IV:21.1) The Subduer of the Wicked
Dradari* (I:44.1) [Correct: dar] The Far-Seeing
Dradurita (I:43.3) (He who is) Far from Sin
Dptasana (I:44.2) (He who) Firmly Commands
Dvadvabhakta (I:41.3) The Devotee of the God of Gods (iva)
Davarddhana (I:43.2) (One who causes) The Prosperity of (his) Country

Dh
Dharaitilaka (I:51.4) Earths Beauty-Mark
Dharmmakavaca (I:19.4) Protector of Dharma
Dharmmanitya (I:19.2) (One who) Ever Abides by Dharma
Dharmmavijayi* (I:42.4) [Correct: vijay] (He whose) Victories (are always) Dharmic
Dharmmastu (I:43.3) The Bridge (for others to reach) Dharma
Dhavalaya (I:19.3) [aya, in II & III:19.3] The Pure in Heart
Dhairyyasgara (I:52.3) The Ocean of Fortitude

N
Nayanamanhara (I:44.4) (He who is) A Delight to the Eye
Naynusri* (I:44.3) [Correct: sr] The Follower of Polity
Ngapriya (I:52.4) The Lover of Elephants [Nga also = snakes; thus, Ngapriya = iva]
Narndraclmai (I:13.2) [cmai, in III:13.2] The Crest-Jewel of Princes
Nityavara (I:13.3) (He who is like) Continual Rains
Nitytsha (I:21.3) The Ever-Resolute
Niramitra (I:53.1) (He who has) No Enemies (left)
Nirarggala (I:53.2) The Unimpeded
 184

P
Paracakramarddana (I:13.1) The Destroyer of Hostile Empires
Parantapa (I:53.3) The Mortifier of (his) Enemies
Parahita (I:21.2) (He who) benefits others
Parpara (I:21.1) The Supreme Lord [Brahm]
Parjjanyarpa (I:12.4) (He who) Appears Cloud-Like (in raining prosperity)
Pallavditya (I:20.10) The Sun of the Pallavas
Prtthivasiha (III:54.1) The Lion among Kings [Incorrectly, Prtthivasi, in I:54.1]
Prtthavikrama (I:22.2) (He who has) The Valor of Arjuna [Prtha]
Puyalka (I:22.1) (He whose) Fame is Pure
Puruasiha (I:21.4) The Lion among Men
Pratibhaya (I:54.4) The Formidable Antagonist
Pravttacakra (I:52.3) The Emperor
B
Balapramathana (III:54.2) Subduer of the Mighty [ivas Yajur-Vdic title] [Balaprama*, in I:54.2]
Bahudakia (I:22.4) The Munificent
Bahunaya (I:3.4) The Great Statesman

Bh
Bhayarahita (I:23.1) The Fearless
Bhmakarmmuka (IV:17.1) (He whose) Bow is Terrifying
Bhmaknta (I:22.3) (He of) Awe-inspiring Loveliness
Bhmavikrama (I:56.1) (He of) Awe-inspiring Valor
Bhaacpa (I:20.2 & IV:17.2) (He whose) Bow is Terrifying
Bhuvanibhjana* (I:24.1) Receptacle of the World [Correct: Bhuvana] [See, also: Avanibhjana]
Bhridna (I:54.3) The Bounteous

M
Mattapramatta (I:23.3) The Intensely Passionate
Mattavikra (I:23.4) The Intensely Active
Manucarita (I:24.4) (Like) Manu in Demeanor
Mahnubhva (I:33.4) (His) Majesty
Mahprabhva (I:24.3) The Immensely Powerful
Mahmalla (I:23.2) The Mighty Wrestler [This was the No. 1 title of Narasimha-I]
Mahndraparkrama (I:24.2) (God) Mahndra in Heroism [Mahndra also = his g.g. grandfather]
Mycra (I:26.1) The Diplomat

Y
Yugntditya (I:26.4) The Sun at the End of the World; or, The Sun till the End of the Eon
Yuddhrjjuna (I:15.4) The Arjuna in Battle
 185

R
Rakmai (I:27.2) The Jewel of Protection
Raacaa (I:27.3) The Fierce in Battle
Raajaya (I:1.3) The Victorious in Battle
Raadhra (I:27.1) The Steady in Battle
Raavikrama (I:27.4) The Valorous in Battle
Raavra (I:26.3) The Hero in Battle
Rjakujara (I:56.2) The Elephant among Kings
Rjarja (I:13.4) King of Kings
Rjasiha (I:1.1) The Lion among Kings

L
Lalitavilsa (I:56.3) [Laita, in III:56.3] (He of a) Graceful Smile; or, The Graceful Sport
Lkaikmai* (I:53.4) [Correct: ikhmai] The Crest-Jewel of the World

V
Varuapa (I:52.1) The Noose of Varua
Vdyavidydhara (I:14.1) A Vidydhara (with) Musical Instruments
Vraabhagadatta (I:57.1) (He who resembles) Bhagadatta (in the knowledge of) Elephants
Viktavilsa (I:57.2) (He of an) Ironic Smile
Vikramaksari* (I:57.3) [Correct: ksar] A Lion in Valor
Vinrada* (I:57.4) [Correct: V] Nrada (in playing the) V [Mcchakaikam, Act V, l. 11]
Vraksari* (I:14.3) [Correct: ksar] The Lion among Heroes

akarabhakta (I:58.1) The Devotee of akara (iva)

stradi (I:56.4) (He whose) Eyes are the stras
rgragaya (I:58.2) Foremost among Heroes
rkmuka (I:14.4) (He who) Desires Prosperity
rpativallabha (I:26.2) The Favorite of rpati (i.e., of Viu)
rbhara (I:3.3) The Bearer of Prosperity
rmgha (I:5.1) The Cloud (which showers) Prosperity
rvallabha (I:16.1) Favorite of the Goddess of Prosperity (i.e., of rdv)
rvilsa (I:15.3) The Smile of Fortune; or (He who) Sports with r (the goddess of prosperity)

S
Sagrmarma (I:16.2) (He who is like) Rma in War
Samaradhanajaya (I:20.1) The Arjuna in Battle [Implied: The Conqueror of Wealth in Battle]
Sarvvatbhadra (I:15.1) (He who is) Auspicious in All Respects
Srvvabhauma (I:16.3) (Possessor of) The Whole Earth [I.e., The Emperor of the Whole World]
186 Birudas on the Dharmarja Ratha

Bhuvanabhchana (A1) Pridhivisra rbhara (B1)

Ardhanr Subrahmaya
r Narasiha Harihara iva Atyantakma-
(A1) Ankbhya (C)

First Level
Dharmarja Ratha

North

Brahm king [r]mgha-

Bhairava iva Trailkyavarddhana-
Vidhi (D)

Note: Engraved on the parapet of the stairway connecting the 2nd and 3rd levels, eastern side, is the title: Mahmalla.

Anupama Vma
Naykura[] (A2) Parbhara (B1)

Sthirabhakti- "W" "P" "W" "P" (bell)

Madanbhirma iva & iva Vidhi
Vidhi[] Nandi Vibhrnta (B2)
(A1)
iva & Kla iva-Andhakavadha

r Narasiha- rnidhi
Bhuvanabhjana- Viu & Garua Vdhara (iva) Niruttara (B1)
rmgha
Apratihatasana
(A3) Gagdhara Harihara

Kmalaita-
Amyamya iva & Caa iva & Nandikvara Nayanamanhara
Sakalakalya Sarvvatbhadra (E)
(F1)
iva & Tau Ka & Kliya

(F2)
Nayanamanhara- Vdhara 2nd Level iva Laita (E)
Vma dvraplas
Atimna
4-armed figure female
Satyaparkrama
Parvara
(F1)

I claim that all the above titles are birudas of Narasiha-I (Mahmalla). But on the 3rd level, east side, just
above the central figure of Srya, there is a label inscription, in two lines, by Paramvara-I, naming the sanctum:
r Atyantakma-Pallavvara-Gha[m ||]
Raajaya
The title, Raajaya, in the second line, therefore belongs to Paramvara-I. The name of the 3rd level
shrine minus the r and the title Raajaya is also found engraved by Paramvara on the west side of the
3rd level, just above the entrance to the sanctum in which he had carved the iva-Smskanda panel.
 Alphabetical List 187
(of the Royal Titles engraved on the Dharmarja Ratha, Mmallapuram)
In the following alphabetical list of the titles engraved on the Dharmarja Ratha, the numbers in
the brackets after each title indicate the level (1 = ground level; 2 = second level; and 3 = third level), and
the letters N, S, E, and W, indicate north, south, east, and west sides of the Ratha.
Atimna (2-N) The Most Honorable [Also in Rjasihas Shore Temple inscription]
Atyantakma (1-S) (He of) Boundless Desires [Also, in Kailsantha temple, I:1.2]
Ankbhya* (1-S) [Correct: Ankpya] The Highly Enterprising
Anupama (2-E) The Incomparable
Apratihatasana (1-E) (He whose) Commands are Unopposed [Apratihata in I:UG.2]
Amyamya (2-N) (He whose) Power is Unfathomable
Kmalaita (2-N) (He who has the) Charm of Kma (Cupid) [Also, in Rjasihas Tiruporur inscrip.]
Trailkyavarddhana (1-S) (He who) Prospers the Three Worlds [Also: Tiruporur pillar inscription]
Nayanamanhara (2-N & 2-S) (He who is) A Delight to the Eye (I:44.4)
Naykura (2-E) The (Flower-)Bud of Wisdom in Polity
r Narasiha (1-N & 2-N) The Illustrious Narasiha [the kings abhika nma]
Niruttara (2-S) The Unexcelled
Parbhara* (2-E); Parvara* (2-W) The Omnipotent [I:21.1 Parpara, which is correct Skt.]
Pridhivisra* (1-E) [Correct: Pthiv] The Essence of the World
Bhuvanabhchana* (1-E & 2-N) [Correct: bhjana] The Receptacle of the World (I:24.1)
Madanbhirma (2-N) The Charming God of Love
Mahmalla (2/3-E: on the outer side of parapet railing) The Mighty Wrestler (I:23.2)
Raajaya (3-E) The Conqueror in Battle [This title alone, here, belongs to Paramvara-I] (I:1.3)
Laita (2-S) The Charming
Vma (2-N & 2-E) The Handsome
Vidhi (1-S & 2-N & 2-S) The Ruler [Also a title of King Mahndra in his Tiruchi poem inscription]
Vibhrnta (2-S) The Extremely Passionate [Mahndra has the simpler title, Bhrnta]
rnidhi (2-S) The Treasure-House of Prosperity [Also, in Rjasihas Vayalur pillar inscription]
rbhara (1-E) The Bearer of Prosperity (I:3.3)
rmgha (1-S & 2-N) The Cloud (which showers) Prosperity (I:5.1)
Sakalakalya (2-N) (He who causes) Prosperity (to) All [Cf. Sarvvakalya, Vsanta plates, l. 9.]
Satyaparkrama (2-W) The Truly Heroic
Sarvvatbhadra (2-S) (He who is) Auspicious in All Respects (I:15.1)
Sthirabhakti (2-N) (He who is of) Firm Devotion [Cf. Mahndras similar title, Dhabhakti]
_______________
The titles of the king are found engraved above a number of the figures carved in niches of the
Dharmarja Ratha. In the diagram, on the facing page, are given the titles engraved on the first and
second floor plans of the Ratha (that is, at the ground level and middle storey).
All of the titles are in the nominative case and end with the letter (visarga).
In the diagram, I have identified the location of the sculptured figures by labeling them in the area
inside the floor plans. Thus, for instance, on the lowest level of the Ratha, the name of the king, Nara-
siha, is engraved above the image of Harihara on the north-east corner.
The capital letters W and P stand for worshipper and priest.
The royal titles appear as though written by different scribes. My guess is that six different scribes
(A to F) were responsible for the 35 titles (27 different ones) inscribed on the first and second levels.
(The numbers appearing with the letters indicate slight sub-variations which I perceived.) The two label
inscriptions, the ones which give the name of the third-level shrine as: Atyantakma-Pallavvara-
Gham, are in a very different style of writing, and belong to the later reign of King Paramvara-I.
 SEVENTEEN

Mahndras Paradoxical Birudas1

Subtle dimensions of the various levels of possible meanings of
the titles glorifying gods and kings are to be found in the paradoxical
birudas used in Pallava inscriptions. The following study investigates
these dimensions.
The Pallava king, Mahndravarm-I, is widely acknowledged as
one of the outstanding rulers of South India. His reign, before and
during the early part of the seventh century, A.D., was a time of great
revival for the aivite and Vaiavite faiths, even as Buddhism and
Jainism were declining.
The religious and philosophical outlook of King Mahndra can
be gathered from a study of the two plays which he authored and which
have come down to us: the Mattavilsa Prahasanam and the Bhagavad-
ajjukam.2 Another source of his views is found in the various stone
inscriptions engraved on his monuments. In his cave-temple, near
the summit of the Rock-Fort Hill, Tiruchirapalli, there is, besides the
famous poem inscribed there, a long list of his birudas. Ordinarily, a
biruda is a title glorifying a king in an unambiguous way. And the
majority of Mahndras birudas do conform to this norm. Take, for
instance, his title Dharmmaplaka, which in English may be ren-
dered as the Protector of dharma. Or, again, Satyasandha, which
means (He who) sticks to the truth. Mahndras best known biruda,
Guabhara, is of this type, and means The bearer of virtues.
But Mahndra had an unusual sense of humor, and delighted in
the paradoxical by way of adopting many strange birudas which have
baffled scholars.
If historians had had no knowledge of Mahndras authorship
of the Sanskrit farcical drama entitled Mattavilsa-Prahasanam, what
would they have made of his biruda, Mattavilsa, which may be
translated as Drunken sport? It is not exactly the most complimentary
biruda for a king, at face value. Or, again, the biruda, Virasa, which
at face value translates to Tasteless (or Obscene). Another biruda
of his is Akarua, The merciless. And a final example here is
Sakrajti, which, again at face value, means Of mixed caste.
I suggest that a linguistic and philosophical analysis will help in
understanding the paradoxical nature of these titles. And the key to that
understanding is the Gaa Ratha Inscription of Mmallapuram, which
is attributable to King Paramvara-I (Mahndras great grand-son). In
the Gaa Ratha Inscription, which is Sanskrit poetry, iva (or,
alternately, the king, himself) is described in seemingly paradoxical
pairs of epithets. For instance, consider this passage:
 190 May he (iva) be victorious! he who is both Amya, yet
Pallava Art Citramya; who is Agua, and yet Guabhjana; who is
Svastha, and yet Niruttara, who is Ana, and yet Param-
vara!
Lets consider the first apparent paradox, where iva is claimed
to be both Amya and yet Citramya. If the mundane world of ord-
inary sense experience is taken to be the effect of Gods creative power,
mya, then, to say that iva is Amya would, in one sense, mean that
he does not have such creative power. But the next half of the figure of
speech asserts that iva is Citramya. That is, he not only possesses
the power of mya, but it is amazing (citra).
But this apparent paradox is resolved once the term Amya
is rightly perceived to mean that the primeval God (iva) is immutable
and cannot be equated with the transient world which is produced
through his creative power.
Thus, we may translate the first portion of the passage as
follows:
May he (iva) be victorious! he who is both immutable (amya)
and yet the ground of all transient existence (citramya); . . .
There is nothing paradoxical about the passage when interpret-
ed in this way.
The other apparent paradoxes may be resolved in a similar way.
We give below a table to illustrate this:
Seeming Paradox Resolution
Negative & Positive & Negative & Positive &
undesirable desirable desirable desirable
Amya vs. Citramya Amya & Citramya
(powerless) (amazing power) (immutable) (ground of all
transient being)
Agua vs. Guabhjana Agua & Guabhjana
(no virtues) (vessel of all virtues) (quality-less) (vessel of all qualities)
Svastha vs. Niruttara4 Svastha & Niruttara
(alone) (superior to all) (self-sufficient) (superior to all)
Ana vs. Paramvara Ana & Paramvara
(leaderless) (Supreme Lord) (no Lord (Supreme Lord)
[above Him])
The extreme left column lists the negative term of each seem-
ingly paradoxical pair. It is the undesirable connotation latent in each
of the negative terms which gives rise to the paradoxical effect.
However, each negative term also has a desirable connotation.
When each of the negative terms is understood in its desirable sense,
then the paradox disappears:
May he (iva) be victorious! he who is immutable (amya), (and
yet) the ground of all transient existence (citramya); who is
without qualities (agua), (and yet) the vessel of all qualities
(guabhjana); who is completely self-dependent (svastha), (and
 yet) without superior (niruttara); who is without a master (ana), 191
(and yet is) the Supreme Lord! Paradoxical Birudas
Mahndras linguistic exercise with regard to some of his own
titles is even more compressed than Paramvaras. This is because
each title or biruda is a unit in itself. It is this group of paradoxical
titles of Mahndra which we shall now analyze. Each separate biruda
in this group generates two or more distinct levels of meaning to create
its effect of paradox.
The first level involves the most obvious connotation of the
word which is used as a title which connotation is negative and
undesirable. Thus, Virasa is Tasteless (Obscene); Akarua is
Merciless; and Sakrajti is Of mixed caste.
The second level is generated from the possible desirable
connotations of the terms, still negatively conceived. Thus, Virasa
becomes One who has attenuated his sensual experience as, for
instance, a sannysin would do; Akarua becomes (One who is)
merciless (to evil-doers); and Sakrajti becomes a name, some
scholars would say, of a type of mixed rga (melodic scale) in music
and thus would personify Mahndra, its inventor. Be that as it may, we
shall suggest that there are still further levels of meaning of this biruda
which can co-exist with the above.
This brings us to the third level of meaning generated by a few
of these title terms. At this third level, Virasa sheds its negative
aspect altogether and is transformed into the wholly positive meaning
of (He of) extraordinary sensitivity (in matters artistic or philosophi-
cal). This transformation is possible because the prefix vi-, in San-
skrit, can have a positive as well as a negative effect.
Sakrajti may turn out at the third level to have a rather
philosophical import. This title appears in Mahndras Tiruchi cave-
temple which also has the famous poetical composition inscribed near
the Gagdhara image carved on one of the rock walls of its maapa.
And the clue to the higher meaning of Sakrajti is found in the
fourth line of the poem:
. . . sthu svaya-ca saha tna-jagatsu jta |
The context of this passage is the claim that King Mahndra had
an image of iva created in his cave-temple (Tiruchi), which image was
at the same time a portrait of the king, himself. Thus, the poem informs
us, by the creation of this fixed image-cum-portrait, ivas title,
Sthu, became literally meaningful (fixed, stationary), and the king,
himself, became sthu (fixed, immortal) together with iva in the
world.
The religious and philosophical significance of this verbal
and visual pun or dhvani, to use the more appropriate Sanskrit term
would be that the king had attained awareness of his spiritual union
with God. Sakrajti would signify precisely this spiritual birth
(-jti) into the knowledge of this union (sakra-) with God. We
should also be aware of the presumption that the union is an eternal fact
 192 with regard to all creatures, but it is the birth of the realization of this
Pallava Art union to which the king alludes.
_______________
1Based on Mahndravarmans Paradoxical Birudas, by
M.C. Lockwood, a paper first published in the Bulletin of the Institute
of Traditional Cultures: Madras, Jan. to Jun., 1976, pp. 11-16.
2Both plays are edited and translated into English by Lockwood
and Bhat in Metatheater and Sanskrit Drama (Madras: Tambaram
Research Associates, 1994).
3The full passage reads:

Sambhavasthitisahrakraa vtakraa |
Bhydatyantakmya jagat kmamarddana ||
Amyacitramysvagu guabhjana |
Svasth niruttar jydana paramvara ||
4In this pair, Niruttara is actually a negative term having a
desirable meaning and Svastha is a positive term having an undesir-
able connotation (alone and helpless). The paradox and its resolution,
however, are otherwise similar to the other pairs.
 EIGHTEEN

The Birudas of Mahndravarm1

Introduction
From the stone inscriptions of the Pallava king, Mahndra-
varm-I (who flourished around 600 A.D.), an astonishing number
of royal titles (or birudas) can be collected nearly 130 of them!
Most of the available titles have been published in text form
(Ngar script). And most (though not all) of these titles have been
illustrated in the accompanying facsimiles of the original inscriptions.
However, the published sources are scattered; the facsimiles in certain
parts are difficult to read; and the editors have left many of the titles
uninterpreted as to their meaning.
To date, the most vigorous effort to interpret Mahndras titles
is found in T.N. Ramachandrans article (published in the 1930s),
The Royal Artist Mahendravarman I.2 Even so, this was a general
article on Mahndras artistic accomplishments, and Ramachandrans
reading is doubtful in several places, and there still remain many titles
whose meanings he was not able to suggest.
In 1967, T.N. Subramaniam proposed that some of the titles of
Mahndra appeared mysterious because they were in a foreign language
of the Indochina region colonized by the Pallavas.3 T.V. Mahalingam
repeated this view in 1969 in his book on the Pallavas:
Mahendravarman assumed a bewildering variety of birudas after
some of which he named his architectural excavations. . . . He
bore also a considerable number of Telugu titles, the import of
many of which is mysterious, while a few like Bpu or Vlasu,
Mlayu, Kaumtayu, Kaumtarambu etc. look more like foreign
titles, probably having Khmer origins.4
It is against this background of lingering mystery that I
present here the full range of Mahndras titles (in transliteration and
facsimile). It is my hope that a renewed interest in these titles by
scholars will lead to a fuller understanding of most of them.
First, in Part One, all of the titles which are engraved in list
form are given in the order in which they appear in situ in the Pallava-
ram and Tiruchirapalli cave-temples, and on one of the four Kanchi
Pillars which are now placed in the front entrance of the main building
of the Madras Museum.
More than one hundred of Mahndras titles in list form are
engraved in his cave-temple at Pallavaram, near Madras.5 About eighty
titles in list form are legible in his cave-temple at Tiruchi.6 Of these
eighty, 60 or so are titles also found at Pallavaram. The Kanchi Pillar
has fourteen titles engraved on it, and every one of these titles is also
 194 found either at Pallavaram or at Tiruchi (or at both places).7 These
Pallava Art three sources give the kings titles in list form, one biruda after another.
In Part Two, a concordance is given in alphabetical order of
all these titles together with facsimiles arranged in 3 columns according
to their three sources: Tiruchi, Pallavaram, and Kanchi. A comparative
study of these titles will be useful to epigraphy because the majority of
Mahndras titles appear in the different locations written in various
styles of early Grantha script, and a few are even written in the early
Pallava Tamil script. Such a comparative study will also give us a good
idea of the actual variations in spelling which occur when titles taken
from different languages (Sanskrit, the Prakrits, Telugu, Tamil, and
Kannada) are written in Grantha.
In Part Three, a separate concordance is given for the few
remaining titles which are gleaned from the dedicatory and poetical
inscriptions at Mahndras cave-temples at Mandagappattu,8 Vallam,9
Mahendravadi,10 Mamandur,11 Dalavanur,12 Siyamangalam,13 and
again also from the same cave-temple at Tiruchi,14 and from Mahn-
dras Chezarla Slab Inscription.15 From these sources, six new (differ-
ent) titles can be added to our list.
Finally, in Part Four, a fresh attempt is made in this study to
interpret as many of the birudas as is possible. For many titles, this
task is difficult, and necessarily tentative. Further, any given title of
Mahndras most probably had multiple intended meanings. It would
be a serious mistake to necessarily limit each title to a single meaning!
Therefore, I have often given more than one meaning which may be
appropriate to an intended context of a title.
That some of the royal titles actually had an obvious (prima
facie) derogatory connotation may seem shocking and even absurd.
Our previous study, however, has attempted to provide a basis for
interpreting such titles.
_______________
Please Note: Most of Mahndras titles have been written in what is
loosely called the Pallava Grantha script. However, a few of his titles
are written in the Pallava Tamil script a script for Tamil developed
from Pallava Grantha by scribes of the royal court. (The modern
Tamil script has directly descended from this Pallava Tamil script.)
In the following pages of Parts I and II, the transliterated forms of those
titles of Mahndras written in Pallava Tamil are printed in bold type!
 195

PART I
Mahndravarms Pallavaram Cave-Temple Inscription (Titles)

1
r Mahndravikrama Mattavilsa Ceakri Vicitracitta Ciundu Ceruuccemprru
luptakma Vimya Da(rppa)via Lkavaya (Dav)gni Kalahapriya
Laitkura (Mayamakku) Citrakrappuli Nivambu Nilvulneyyambu Vampu Vuk
Vaka()mpu Ku17 Kaumpu18 Kaukyu Vil(y)a Pala(pi) Vntulavittu Alavala19
sii Akkap(s)umbu20 Daikalla Vyivei Yamuku Sakrajti ()u21

2
Iukai Pisugu Piuvi22 Perindhi23 Prakraa() Pravtta-mtra()
Pukpiuku (Pa)sarambu (Ananya ?) Pavi(u)24 (Pa)kai Bku Bujjanakanthu Bhrnta ()u25
tha26 Ututi27 Upamna Ummkuu28 Kmrjjava Kaukrka29 Kauterambu Guabhara
Teppu T()vi Taudaa30 Tanumpu(n)mi Tuknu T()u31 Tda Ta32
(r Dhabhakti) (Dunuvryya) (Uduksii) . . . (Dharmmaplaka) . . . .

3
Satyasandha Lakita Iaduabhraacarita Naihikmutrika Nayambu Nauku33
Mahmgha() Ma(npr)vu Maum[]a34 (M)ilcuo35 (Mrkhavijja) (Moggara) (Cacui)
(Curmbu)
(Pau[si]dhdha) (Caisa)ppuru()u (Vstha)36 Vu(ntha)37 (Vambara) (Vvei)38 . . . (Vakiu)
(Vpu)39 (Virasa)40
Vyavasthita Vyavasya Anityarga Avanibhjana Anumna Abhimukha Akari41
hryyabuddhi (yati)42
196

Mahndravarms Tiruchi Cave-Temple Inscriptions (Titles)

On outer row of pillars:

Proper right43 pilaster (bracket): Vajavalava

First pillar: Sarvvana(ya)44

Taudaa Nityavinta Nivambu
Tanumpunmi Nirapka Nayambu
Tuknu Nilvulneyambu Nauku
Tpuk Naihikmutrika Narpaa(ka)45
U(l) . . . ku

Second pillar: Sakrajti

Virasa Anityarga Vambu
Vyavasthita Anumna Vuk
Vyavasya Avanibhjana Vpu
Cittira(kk)rappu(li)

Third pillar: Satyasandha

Kaukyu Abhimukha Vstha
Kauntarambu Akarua Vakiu
Kaumpu Alavala Vaka
Piapiakku

Fourth pillar: (Lai)tkura

Maku Calambu Emuku
Kaa Kilambu thi
Kuambu Mlyu Kuhaka
Vambara Vvei Vuntha
Kucaa

Proper left pilaster (bracket): Lakita()

 197

On the west-facing side of the proper left (eastern) pilaster (outer row of pillars):
1 Svasti r Mahndravikrama
2 Mattavilsa Mayamayakku
3 Mauma Mahmgha
4 Manprvu Milcuo
5 Mrkhavijja Moggara
6 Ma . . i Ceakri
7 (Ciundhu) (Ceruuce)mbru
8 Cui . . . (S)aa
9 . . . (Caisappuruu)
10 (Vi . . .) (Vicitrac)itta
11 . . . . . . . . . . . . . . . .

On the inner row of pillars:

Proper right pilaster (bracket): Guabhara

First pillar: (bracket): Akkapsunu

(central facet): T . .

Second pillar: (bracket): sei ya(t)i

(CF): Tea

Third pillar: (bracket): luptakma

(CF): Te(a)

Fourth pillar: (bracket): hryyabuddhi

(CF): Kaa

Proper left pilaster: (bracket): damaged

On the west-facing side of this same left (eastern) pilaster:

1 (Iaduabhraa)carita
2 damaged
3 damaged
4 Prak(raa) the rest damaged
5 This and following lines are unreadable
198

Mahndravarms Kanchipuram Pillar Inscription (Titles)

First face: thi

Kucaa
Mahmgha
Dhabhakti

Second face: Abhimukha

Citrakrapuli
Kuambu

Third face: Curmbu

Vakabu
Vntha

Fourth face: Pisugu

Vambara
Bhrnta Akari
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214

Mandagappattu

Siyamangalam
Mahendravadi

Tiruchirapalli

Pallavaram
Mamandur
PART THREE

Dalavanur

Chezarla
Vallam
Titles also in non-list form

Avanibhjana + +

Guabhara + +
(Kuapara) +

Narndra +
(Danatanarndra) +

Nityavinta +

Pukpiuku
(Pakppiuku) +

Puruttama +

Bhbhuja +

Mahndra +
(Mahndravikrama) +
(Mayntirappttarcaru) +

Lakita +

Laitkura + +
(Laitkura) +

Vicitracitta +

Vidhi +

atrumalla + + +
(Catturummalla) +

Satyasandha + +

Sthu +

+ = titles found in dedicatory and poetic inscriptions

= titles found in list form inscriptions (Parts I & II)
 PART FOUR 215

Akkapsunu (T); Akkapsumbu (P). Tel., In battle (akam), the one who (wields) the noose,
i.e., The God of Death in battle (Yama). See Epigraphia Indica, Vol. VI, p. 178, where the king,
Kisha-III, has the title, Akatrintra: a very Trintra (iva) in battle (the Atakur inscription of
A.D. 947-50). See my comments on the -nu and -mbu endings under the entry, Kuambu.
Akari (P)(K). Akkari fr. Tamil Akkarai, foreign/er; or fr. Skt. Akari, Maker of inscriptions or
Immortal; or kari (anger, rage; poison; fearful) and a- (the negation of those qualities).
Akarua (T). Skt., Merciless the satirist is merciless to his victims (especially to his Buddhist
victims, to whom karua is such an important virtue!): see Mahndras plays, Mattavilsa and
Bhagavadajjuka, for a humorous treatment of the emotion of karua. Comedy demands a suspen-
sion of mercy (Lee Siegel, p. 19).
Anityarga (T) (P), Skt., Not permanently or continually drawn to pleasure. Anitya may also mean
unusual or unstable hence, (One of) unusual or unstable desires.
Anumna (T) (P). Skt., Reasoner, logician, disputant. Philosophical: the second of three stages,
where one attains the knowledge of God through reason. One of the means of obtaining knowledge
(pramna) according to the Skhya system (or Nyya).
Abhimukha (T) (P) (K). Skt., Forward looking; daring; straight-forward.
Alavala (T) (P). Skt., vala (darkness; dryness); alati fr. root ala (prevent, ward off): Vala alati iti
Alavala. There was a large irrigation lake at Mahendravadi named after Mahndra, and we have
suggested elsewhere that the king may have involved his engineers with damming the Kvr.
Avanibhjana (T) (P) (Siyamangalam) (Chezarla). Skt., The receptacle of the earth. In the invocation
(Nnd lka) at the opening of King Mahndras play, Mattavilsa, there occurs the expression
vyptvanibhjanam which incorporates his title, Avanibhjana, and indicates that we may
interpret it philosophically as, He in whom the world is manifested. This title is also found at the
Kailsantha temple, Kanchipuram (3rd title inscribed on niche 46 [46.3] ).
yati (T) (P). Prob. Skt., Majesty; dignity; height (tall); restraint of mind; extension, length, prolixity.
Kan., majesty; length, restraint of mind; an ascetic (in the present context, a Rjari).
luptakma (T) (P). Skt., (He who has) completely reduced to nothing (his) desires ( = all-round,
as far as; -lupta = destroy, suppress; -kma = desire, passion). See also: Nirapka of this list, i.e.,
One who has attained cessation (or perfection) of his desires.
sei (T); sii (P). Kan. (Kit.), The creator, fr. sii (a tadbhava), creation, etc. Or: Prkt. fr. Skt.,
ryas, most excellent, superior; auspicious. = in all respects. Syn., Sarvvatbhadra.
hryyabuddhi (T) (P). Skt., Incarnate wisdom; (One) knowledgeable about costumes, dress;
Awareness of indwelling Self.
Iukai (P). Tam., iu = obstacles; kai = remove: Remover of obstacles. Or, iu = give; kai =
excessive, abundant: The munificent. Or, iu = put, place; erect; set up; kai = stick, staff: (He
who) wields the stick. Also, cf. the expression, Iu ka kai, in the Tirukkual.
Iaduabhraacarita (T) (P). Skt., ia = friends; that which one holds dear; dua = rogues, that
which is undesirable; bhraa = degenerates, those who have lost caste or status; carita = behavior
or character. (He who) act-ivates the good, bad, and degenerate. We see these types of characters
acting in Mahndras two farces and the prakaraa, Crudattam, and its adaptation, Mcchakaikam.
(Uduksii) (P). I cannot make out this word from estampage or site. As given in South-Indian Inscrip-
tions, XII, No. 13, it might be interpreted as derived from Skt., uduka = star; -siddhi = com-
pletely accomplished (i.e., Master of astrology or astronomy). In the prologue of Mahndras play,
Bhagavadajjuka, he refers to astrology.
Ututi (P). From Skt., uddhathi = (His) eminence.
Upamna (P). Skt., Standard of comparison (logic); resemblance, analogy; recognition of likeness.
Awareness of indwelling Self? Third of four pramas (means of attaining correct knowledge).
Ummkuu (P). Tel., He who has Um seated on his thigh (Umm = Um; aka = thigh; uu
= one who has). I.e., iva.
U[l] . . . ku. (?)
Emuku (T); Yamuku (P). Kan., amuku = press down, crush. Tam., amukku = press down
(Tiruvcakam). The oppressor (of the wicked).
thi (or i) (T) (K). Tam., to shine brilliantly; or i = to reach high estate. Tel., eriya = king;
Eikal = a title of a Renati Cha king. Tam., u = bull. I.e., He who has the bull (iva and
King Mahndra, because the Pallava kings had the bull as their emblem, too). ttu = extol, praise.
()u (P). Tam., to raise, teach; am = excellence, superiority; fame. a = Lord of the Bull.
a (or tha) (P). Skt., tha = extending in all directions: All-encompassing. Tam., tam = ocean.
(O)u (P). Tam., u = frets or plectrum. Ottu = keep time to music. Tel., ottu = drone instrument.
Kaukyu (T); Kau[]kyu (P). Tel., fr. Tam., kaun = strong; fierce; kayam = compulsion;
force, power; constraint; certainty: (He of) fierce power.
Kaukrka (P). From Skt., kau = biting; krkam = saw. The saw which cuts down (enemies).
Kauntarambu (T); Kauntrambu (P). Tel., kaun = fierce, mighty, trambu = chariot, accd. to
I. Mahadevan. Compare Mahrathi, the name of an early, Andhra dynasty. Trambu = Tam. tr.
Kaumpu (T) (P). Tel., Tormentor. Or, kaumpu = mighty. Tam., k[u]umpu = Mischief-maker
(similar to Kalahapriya).
Ka[]a (T) (P). Tam., kaar = The learnd.
Kalahapriya (P). Skt., (One who) enjoys a fight. An epithet of the mischievous Nrada. A fight is
also acted out in Mahndras play, Mattavilsa.
Kaa (T). Fr. Skt., Calamity. In the Mattavilsa, the Kplika cries out, Bh kaam! (line 59).
Ku (P). Tam., wind. The wind (storm).
Kmrjjava[] (P). Skt., Subduer of desires. Skt., (One of) honest or straightforward desires.
Kilambu (T) Skt., kila = sport, amorous play. Kan., kila = laughter, playing. Tam., exaltation,
emergence. Ambu = pallava = akura. A title equivalent to Laitkura (see below).
Kucaa (T); Kucaa (K). The Tamil script is used in both cases. And as the second consonant c,
in the Tamil script, could be pronounced , the first half of the title could derive from Skt. kua
and mean, sharp (as the tip of a blade of kua grass); and a would derive from Skt. jna and
mean intellect. Also, consider: Kua, a son of Rma, who was an actor or kua, an actor, in
general: the whole title would then mean, One who is knowledgeable about acting. Or, one more
possibility: Tam. kuca = a wicked, depraved, mad, or inebriated person. Thus, A rogues cun-
ning or (One who is) knowledgeable about rogues. This last reading would be a Tamil equivalent
of Mrkhavijja, a Prakrit title also found below in this list of Mahndras birudas.
Kuambu (T) (K). Tel. form of Tam., kuam. The -mbu ending, along with -nu (this latter which, if
written in the Tamil script would be - or -u), indicates the gender of the word in archaic
Telugu (6th/7th centuries, A.D.). I. Mahadevan noted that the dual appearance of the very first title
of King Mahndra in my present list illustrates this: Akkapsunu (m.); Akkapsumbu (n.).
Kuhaka (T). From the Skt. root, kuha = to surprise, cause wonder. Abhinavagupta . . . commenting
on Bharatas list of determinants of the comic sentiment, understands kuhaka, the term used for
trickery, to refer specifically to tickling, to touching childrens necks, armpits, and the like, in
order to surprise them (Problems). This quotation is from Siegels book, Laughing Matters, p. 16.
Guabhara (T) (Mahendravadi) (Vallam). Skt., Bearer of virtues. It can also be taken as a synonym
of Stradhra. Vallams version in the Tamil script is: Kuapara. This title, Guabhara, is
played upon in the second line of the farcical comedy, Mattavilsa.
Calambu (T). Tel., The charming; or, Persevering antagonist. Kan., Calla = great mirth, fun, jest,
laughter. Cf. Smantha, Padhita-radhya Charita Basavapuram, wherein he mentions methods of
manipulating the strings of a v including challaamu and malapu; also cf. Bharatas
experiment on the Dhruva and Cala vs.
Caisappuruu (T); Caisappuruu (P). Kan., caisu = to be in motion, shake, tremble (v strings?);
be confused; go astray; sport about. Purau from Skt. paryastha = thrown or cast about; surround-
ed, encompassed, ensnared; overturned, upset, inverted; struck, killed; eyes rolling, etc. (RSA)
Cittira[kk]rappu[li] (T); Citrakrappuli (P); Citrakrapuli (K). The Tiruchi title is in the Tamil
script. Tam. & Tel., Tiger among artists. Citra may be taken as the fine arts, in general.
(Ciundhu) (T); Ciundu (P). Tam., cilampam = fencing, sword play; cilaiya = bow-man. Hence,
ciundu = The one who wields the bow (or staff) i.e., iva or Kma.
(Curmbu) (T); Curmbu (K). Tel.& Skt., cur = theft; thief (ref. Sajjalaka, in the play, Chrudattam)
(Cui) (T); Cacui (P). Tam., cui = whirlwind; whirlpool; cunning; afflicted in the mind. Cacui =
Whirlwind to the Chas.
Ceakri (T) (P). Tam., ceam = anger, fury, rage (ceampukkra = malicious person). Ceu =
to engrave (ceakkri = Engraver [of inscriptions ?]). Or, cththa- from Skt., ca = creative;
and kri = person. From the last century, however, Cththa- has been interpreted by scholars as
Caithya- and translated as temple; and hence the full title has been understood as Builder of
temples.
Ceruucembru (T); Ceruuccemprru (P). Tel., ceruu fr. Skt. cl = to move; cm = fierce; and
br = (eye) brow. See Epigraphia Indica, X, pp. 101ff., the Gadval Plates of the Chlukya king,
Vikramditya-I, for an illustration of this figure of speech. Another possibility: Skt., cru =
respectfully, worshippingly; cebrolu = a town/city in Andhra. I.e., One who is worshipped in
Cebrolu.
Tanumpunmi (T) (P). Skt., The purified one (tanum = body; punmi = purified).
Tvi (P). Tam., t = Enterprising; energy, exertion, perseverance; t kual = low tone in music.
Again, t = ivas foot. See the Nnd lka of Mahndras farce, Bhagavadajjuka.
Taudaa (T) (P). Tam., tau = fierce; restraining; daa = scepter. Thus, (He who wields) a
fierce (restraining or punishing) scepter.
Tuknu (T) (P). From Skt., dukha = distress, suffering; and anu = look after. I.e., One who looks
after the welfare of those in distress. Tam., tukkam = distress; anu = benefit, goodness. Hence,
the same meaning as in Skt.
Teppu (P). From Tam., teppam = raft; i.e., One who helps people cross over the ocean of life (with
all its difficulties and suffering). Tel., teppu = relief. Ma., Tppu = rubbing, polishing; scratching.
Te(a) (T). Kan., the state of being clear. Equivalent to the name Prasanna.
Tea (T) (P). Tam., team = assurance, relief from doubt; comfort; certainty; perseverance. Tu =
clearness; certainty. Again, equivalent to Prasanna.
To . . . (T); To(da) (P). Tam., toakkam = beginning, origin, commencement.
To()u (P). Tam., tam = appearance (equivalent to Skt. bhsa); strength, power; fame; Source
(cf. Tiruvcakam, 5:70 and 7:20); tu = create, generate, invent.
Topuk (T). Tam., tou = worship; serve; puk = undivided. I.e., One whose devotion (to God) is
undivided, steady. This would be the Tamil equivalent of his title, Dhabhakti. Also, consider:
One of steady friendship (ta = friend).
Daikkalla (P). Kan., Mace bearer. Pli, kalla = clever, able; dai = scepter. I.e., One who wields
a mighty royal scepter. Also consider: Tel., dai = neck of a v (Krishnadevaraya, in his work,
muktamlyada).
Darppavia (P). Skt., Poison to (enemies) pride; or He whose pride is poison (to his enemies). Refer
to the expression, Dviaddarpppahri (He who deprives [his enemies] of their pride) found in
the Gaa Ratha Inscription, Mmallapuram.
Davgni (P). Skt., The forest-fire, or Wildfire (to his enemies). This is also one of Rjasihas
titles (Dvgni, Kailsantha temple, Shrine No. 43, first title). The Dirghasi inscription of
Vanapati (E.I., IV, 45, pp. 317-18) elaborates the forest-fire metaphor; and contains also the title,
Calamartiganda.
Dunuvryya (P). Skt., The preventer of hardships (dunu = hardships; vryya = preventer). See
also: Aprativryya (Kailsa, Shrine No. 46, fourth title), The Irresistible.
D[habha]kti (P); Dhabhakti (K). Skt., (One whose) devotion (to God) is firm. Mahndra has
a Tamil title, Topuk, which has a similar meaning. Ref. also to the title, Sthirabhakti, which is
found on the Dharmarja Ratha, Mmallapuram.
Dha[r]mm . . . ka (P). In S.-I.I., XII, 13, this title is given as Dharmmaplaka, i.e., The protector of
dharma.
Nayambu (T) (P). Tel. ruler, diplomat; pleasant. Ref. also to Naykura, Sprout of polity, found on
the Dharmarja Ratha; Bahunaya, Great ruler, or Great statesman, Kailsantha, Shrine No. 3,
fourth title; and, Naynusri, Follower of polity, Kailsantha, Shrine No. 44, third title.
Narpaa[ka] (T). Skt., (One who is) Fearless of men.
Narndra & Danatanarndra (Dalavanur). Skt., King of men; and King of men (before whose)
scepter (others) bow.
Nauku (T) (P). Tam., & Tel., & Kan., cut, chop to pieces, crush. Also: ngasvaram.
Nityavinta (T) (Mamandur). Skt., Ever-modest.
Nirapka (T). Skt., Desirelessness. As a Buddhist term the word [Nirapka] indicates a virtue, a
lack of concern with mundane things. Through the satirical character to whom it is applied, however,
the virtue becomes a vice, an expression of expedient heartlessness rather than spiritual accomplish-
ment, Lee Siegel, Laughing Matters, pp. 215-16. See also: luptakma of this same list, with
similar meaning. Nirapka is the name of a character in the bhna, Pdatitaka.
Nilvulneyambu (T); Nilvulneyyambu (P). Tel., invincible; invincible arrow; invincible friendship
(i.e., an ever-constant friend) similar to Tpuk (Tamil).
Nivambu (T) (P). Tel., The exalted. Tam., nivappu = elevation, height.
Naihikmutrika (T) (P). Skt., Neither this (world) nor other(-world); i.e., as expressed in Pua-
u 134: One who is righteous and liberal not because of any thought of reward either in this
world (immai) or in the next (maumai), but because that is just the way of a noble person.
(Pa)kai (P). Kan., a corolla or row of petals. Tam., kai = jollity, hilarity; intoxication; bewilderment
(Mayakkam); frenzy (Bhrnta).
Pakppiuku (Vallam). Tamil script. See Pukpiuku in this same list. Same title spelled differently.
Pau[si]dhdha (P) Kan., Pau = sharp, clever, cunning; eloquent; cruel; sidhdha = accomplished.
Skt., pau = proficient; persevering.
Palapi (P). Tam., pala = much, many; pi = sung, singer i.e., The singer (of) many (songs).
Pavi[u] (P). Prkt. fr. Skt., pavitha = The purified.
Piuvi (P). Tam., piukal = grip; viai = bewilderment; frenzy. Tamil equivalent of Mahndras
Skt. title, Bhrnta, which is listed below.
Piapiakku (T). Tamil script. Tam., A devil to devils. Tel., piakku = champion (in single combat)
thus, Champion of champions (in single combat).
Pisugu (P) (K). Tel. & Kan., to squeeze, knead; stingy. Tam., picaku & picukku = failure, mistake
(syn., tavau); blunder (syn., kuam); disagreement; alteration; deviation; picaku = miss a note
or beat (aaital).
Pukpiuku (P). In the Vallam inscription of Mahndras reign, this title of Mahndra clearly appears in
the Tamil script as: Pakppiuku. Thus, puk = pak = undivided, indivisible (e.g., Tam.,
Pakpporu = The Indivisible The Supreme Being); and Tel., piuku (piugu) = thunder-
bolt. Full title: Undivided thunderbolt, i.e., a single, solid bolt of lightning (not branching). An-
other Pallava Tam./Tel. title equivalent to this is Vilviuku (Vilviugu). The Skt. equivalent is:
Akhaai (King Rjasihas title found in his Kailsantha temple, Kanchi [29.3]). The later
Pallavas also assume such titles as Perumpiuku and Mrppiuku.
Puruttama (T). Skt., Best among men; Noblest of men.
Perindhi (P). Kan., per = great; indhana = kindling, fuel. Thus, Great fire. Indha, from the root,
indh, applied to Indra as God of lightning. Thus, per = mah; and indhi = Indra. And we
finally have the equation: Perindhi = Mahndra (King Mahndra, identified with the King of the
gods).
Prak[raa] (T); Prakraa[] (P). Skt., The creator (author) of a prakaraa. [Crudattam?]
Pravtta-mtra[] (P). Skt., pravtta = worldly affairs; mtra = solely. I.e., (One who is) devoted
(to the) welfare (of his people). Or: Ever-enthusiastic; Ever-active. Or: Ever-creative, inven-
tive.
Bku (P). Tam., pku = beauty. Tel., bgu = beauty, charm. Kan., bku = dagger.
Bujjanakanthu (P). Prkt., Downfall (kanthu) to the Buddhists (Bujjana).
Bhbhuja (Mamandur). Skt., (He who supports) the world (by the strength of his) arm.
Bhrnta (P) (K). Skt., frenzied; varied, versatile. Ref. also to: Vibhrnta, The highly passionate, a
title on the Dharmarja Ratha, Mmallapuram. We find these ideas given flesh humorously in the
Madman, in the Mattavilsa.
Maku (T). Kan., maku = dullness, stupidity. Skt., maku = to go, to move, to act (for the sake of
others); also, shaking, vacillating; and adorned, decorated. Kan., magu = masculine, manly.
Mattavilsa (T) (P). Skt., (One who) sports (with) revelry. Some similar Pallava titles are: Matta-
pramatta, The madly excited, Kailsantha (23.3); and Mattavikra, The madly passionate,
Kailsantha (23.4). Madanavilsa, Manifestation of the god of love, was a title of the Atiya
king, Guala, in Namakkal. Consider, also, Mudamudit, One who enjoys revelry, 8th century
Tel. title. Most importantly, Mattavilsa was the title of King Mahndras farce dealing with
drunkenness and madness.
Manprvu (T) (P). Skt., man = desires, wishes; prvu, fr. Skt., prav = satisfy. Thus, Manprvu
(Tel. form) may be translated as: (One who) satisfies desires.
Mayamayakku (T); Mayamakku (P). Tam., mayakku = enchantment, spell, swoon, etc.; the doubling
may indicate the causative form or simply an enhancement of the root word. Maya was the name
of an asura, described sometimes as the artificer of the daityas, versed in magic, astronomy, and
military science; and therefore this title of the king could indicate that the king has the abilities of
Maya which astonish and dumbfound.
Mauma (T); Maum[]a (P). Tam., maumam = renaissance, re-conversion (the king is
the agent of a renaissance). We may understand that this refers to a renaissance of Hinduism, a
reconversion of the people back to their old faith, away from the heterodox religions of Buddhism
and Jainism. Also, the overthrow of King Plaka by ryaka in the Mcchakaikam.
Mahmgha (T) (P) (K). Skt., The great cloud; or, Thundercloud. See also: rmgha, The
cloud (which showers) prosperity a title found on the Dharmarja Ratha, Mmallapuram.
Mahndra (Mahendravadi); Mahndravikrama (T) (P) (Chezarla); Mayntirapptarecaru (Vallam).
Skt., Mahndravikrama = The valorous Mahndra. Tam./Tel., Mayntirapptarecaru is in the
Tamil script and is a Dravidianized form of the Sanksrit, Mahndra-Pta-Rja (pta = Pallava).
Milcuo (T) (P). Kan., mie = to love, unite, understand; seize, hurt; cuo = thief. Thus, (One
who) understands thieves. Similar to his title, Mrkhavijja.
Mrkhavijja (T); [Mrkhavijja] (P). Prkt. fr. Skt., mrkha = rogues; vijja, from vidy = knowl-
edge. The title may be translated as, (One who is) wise to rogues. Or: A rogues cunning.
Moggara (T); Mogga[ra] (P). Guj., Mogra= croc god. Skt., mudgara = mace, club; bud (pallava)?
Mlyu (T). From Skt., mlyin = growing thin or emaciated (through tapas?). Wither(er) (of enemies).
Or: derived from malyu (an area in Sumatra?). Mlyu is another reading.
Yamuku (P). Refer to Emuku, above, in this list.
Lakita (T) (P) (Mandagappattu). Skt., Distinguished; auspicious, having auspicious marks;
authoritative!
[Lai]tkura (T) (Siyamangalam); Laitkura (P); Laitkura (Vallam). Skt., The charming
rake (or scion = pallava). The title at Vallam is written in the Tamil script and in Tamilized form.
Related titles, Laita and Kmalaita, are found on the Dharmarja Ratha; Naykura,
Dharmarja Ratha; and Tarukura, Gaa Ratha and Dharmarja Maapa.
Lkavaya (P). Skt., Controller of the world (by attraction, positively, rather than by force).
Vaka (T); Vakampu (P); Vakabu (K). In the Tamil epic, Maimkalai, King Cekuttuva
crosses the Gag river in a boat called vakam. Consider Mahndras title, Teppu. Tel., vaka
= stream, current of water, torrent. Pli, vaka = crooked, deceitful. The idea of crookedness here
may have some reference to musical scales differing in descent from ascent one meaning, perhaps,
of Mahndras title, Sakrajti.
Vakiu (T); Vakiu (P). Tel., upper arm band. Or: crooked musical scale?
Vajavalava (T). Tam., vacam = cruelty; violence; wickedness; revenge; valava = conqueror;
emperor; strong man. Thus, the title may be translated, Subduer of the wicked (or rebels). Also:
Tam., vaca, from Skt. vaa = lineage; thus, (Of the) family of emperors. Vaja-valava =
Vaa-rja = Vatsa-rja (ref. to the hero of one of the so-called Bhsa plays). Finally, consider
the equation: Vajavalava = Vntulavitta (Tamil), another title in this list.
Vambara (T) (P) (K). From Skt., Remover of burdens (lit., Vomiter of burdens). Tel., vembara = a
fool; a wicked person, a madman; Vempara = plague; trouble; annoyance.
Vambu (T); Vampu (P). Tam., ribaldry, obscenity; breast-band; quarrelsomeness; newness, novelty
(all of this is representative of Mahndras two farces, Mattavilsa and Bhagavadajjuka).
Vyivei (P). Old Tel., vyi = mouth; vei = silver; pure. Silver tongued (orator, singer). He of
pure speech (connected with his title, Satyasandha?).
Vvei (T). As above?
[Vicitra]citta (T); Vicitracitta (P) Skt., Inventive mind; a mind of many-sided talents.
Vi[mya] (T); Vimya (P). Tel., vi = unsplittable; unstoppable; mya = power. Thus, (He
of) unstoppable power. Ref. to Amyamya, (He whose) power is unfathomable, a title on the
Dharmarja Ratha. The title, Myacra, Diplomat, is found in the Kailsantha list (26.1).
Vidhi[] (T). Skt., Ruler; Fate.
Virasa (T) (P). Skt., Tasteless (or: Exalted taste!). Also: Vulgar, Obscene the Tamil form is
viracam (see the Dictionary of Contemporary Tamil), a synonym of which is pcam = porno-
graphic; salacious. Mahndras two farces may appear to portray such tastelessness in certain
passages.
Vil[y]a (P). Tam., Vilaiyar = seller, dealer; (public) servant?
Vuk (T) (P). From Tam., puk = Indivisible; Impregnable. Skt., buka = hsya, laughter (also
written vuka); the long may indicate the causative form: to make laugh.
Vuntha (T); [Vuntha] (P); Vntha (K). From Skt., vintha = having no lord or master; and, there-
fore, unprotected. Note the use of the terms, Ana/Paramvara, in the Gaa Ratha Inscription.
Vntulavittu (P). Tam., vntu = king; royal; vittu = race, lineage; i.e., of royal lineage. Vntulavittu =
Vajavalava.
Vstha (T); [Vstha] (P). Pli, Vesrajja from Skt., Vaia-radhya = perfect enlightenment.
Vyavasya (T) (P). Skt., Resolution (personified); energy; enterprise; perseverance; painstaking;
industry; diligence. The fourth sandhi in dramaturgy is concerned with frustration one element of
which is vyavasya (perseverance, determination).
Vyavasthita (T) (P). Skt., Ordered rule; perseverance, determination.
Vpu (T) (P). [F]rom old Tamil vi and the grammatical suffix -pu. The words va or vi in Tamil
mean to pervade.46 One/which pervades (all)
atrumalla (Mamandur) (Dalavanur) (T); Catturummalla (Vallam). Skt., The foe-mauling wrestler.
This title of Mahndra was used by him in the Bharatavkyam (Benediction) of his play, Matta-
vilsa.
Sakrajti (T); Sakrajti (P). Skt., mixed caste; mixed (musical) rga; awareness of the
indwelling Self [sthu svaya-ca saha tna-jagatsu jta |, i.e., awareness that the king was
united with iva (Sthu) in (this) world the Tiruchi poem of Mahndras]. In the drama,
Crudattam, the Stradhra is born again (in another jti).
Satyasandha (T) (P) (Mamandur). Skt., (One who) sticks to the truth. One of the 108 names of iva.
Sarvvana[ya] (T). Skt., All-round master of polity. Note that in Mahndras play, Mattavilsa, this
title is used by him with reference to his father, King Sihaviu:
Pallava-kula-dharai-maala-kula-parvatasya sarvanaya-vijita-samasta-smanta-maalasya
khaala-sama-parkrama-riya r-mahimnurpa-dana-vibhti-paribhta-rjarjasya r-
Sihaviuvarmaa. . . .
Sthu (T). Skt., firm, steady, etc. Also a name of iva.
[s]aa (T). (?)
gu (P). (?) This title has been read as Ananya (S.-I.I., XII, 13) which is certainly a misreading
of the word. The conjunct letter is definitely not -nya. I read it -gu. Other possible consonants
which could be found with this particular form of the u are bh, t, and .
_______________
1This study by Lockwood and Bhat is a revised version of an earlier one which, on March 25th,
1977, was read at the Third Annual Congress of the Epigraphical Society of India, meeting at Udupi.
2Journal of Oriental Research, Madras, Vol. VII (1937), pp. 219-246 and 303-330.
3The Pallavas of Kch in South-East Asia (Madras: The Swadesamitran Press, 1967), pp. 76ff.
4Kcpuram in Early South Indian History (Bombay: Asia Publishing House, 1969), pp. 70 & 71.
5PALLAVARAM: the text and facsimile are given in South-Indian Inscriptions, XII, No. 13. See also:
Archological Report on South Indian Epigraphy, 1908, No. 369. The temple is described in the Mem-
oirs of the Archological Survey of India, No. 17, p. 16.
6TIRUCHIRAPALLI: S.-I.I., XII, Nos. 8 & 9. A.R.S.I.E., 1908, Nos. 411 & 411-A.
7KANCHIPURAM PILLAR: S.-I.I., XII, No. 14.
8MANDAGAPPATTU: E. I., XVII, pp. 14-17; S.-I.I., XII, No. 12; A.R.S.I.E., 1905, No. 56.
9VALLAM: S.-I.I., II, No. 72 (with estampage, Plate X).
 10MAHENDRAVADI: Ep. Ind., II, pp. 152-53; G. Jouveau-Dubreuils Pallava Antiquities has an
estampage.
11MAMANDUR: S.-I.I., IV, No. 136 (text and estampage); A.R.S.I.E., 1888, No. 58.
12DALAVANUR: Ep. Ind., XII, pp. 225-26; noted in S.-I.I., XII, No. 10; A.R.S.I.E,, 1905, No. 51.
13SIYAMANGALAM: Ep. Ind., VI, pp. 319-320 (text & estampage).
14TIRUCHI: S.-I.I., I, Nos. 33 & 34 (text of poem); the estampage for No. 34 only is found in S.-I.I.,
II, Plate X; A.R.S.I.E., 1888, Nos. 63 & 64; S.-I.I., XII, No. 9 (text and estampage of dedicatory inscrip-
tion); A.R.S.I.E., 1904, Nos. 411 & 411-A.
15CHEZARLA: S.-I.I., VI, No. 595 (text only); A.R.S.I.E., 1899, No. 155-A.
16See the chapter above, in this book: Mahndras Paradoxical Birudas.
17S.-I.I., VI, No. 13: Kaththu. 18Ibid.: Kathumpu.
19Ibid.: Alarvale. 20Ibid.: Arkkap(s)umpu.
21Ibid.: ()ththu. 22Ibid.: Piuvr. 23Ibid.: Perinthi.
24Ibid.: Pavi(thuh). 25Ibid.: ()ththu. 26Othu?
27S.-I.I., VI, No. 13: (Udhdha)ti(). 28Ibid.: Ummkuu.
29Ibid.: Kaukraka. 30Ibid.: Tathudaa.
31Ibid.: T(thth)u. 32Ibid.: Tththa.
33Ibid.: Nathuku. 34Ibid.: Mathumtha.
35Ibid.: Vithlcuth. 36Ibid.: Vsi. 37Ibid.: Vu(stha).
38Ibid.: (Vvti). 39Ibid.: V(su).
40Ibid.: Vrasa. 41Ibid.: kara. 42Ibid.: yatha.
43Proper right pilaster, high up on the bracket, facing south.
44The y can be deduced with reasonable certainty from the remaining fragment of that letter; the
a are postulated.
45A tiny fragment of a letter (ka?) remains; this and the other letters are postulated.
46Iravatham Mahadevan suggested this interpretation in an e-mail sent to us on the 19th of June,
2004. The relevant passage in full in his e-mail is the following:
According to me, the biruda [Vpu] is made up of the stem v (< from old Tamil va or vi) and
the grammatical suffix -pu. The words va or vi in Tamil mean to pervade or occupy space,
thus being the equivalent of Skt. vyp. It is interesting that both citations of this word in the Tamil
Lexicon are from [the] Divya Prabandham referring to the legend of Trivikrama (TL, vol. 6, p. 3728).
On this evidence I would translate Vpu as One/which pervades (all).
This title of Mahndras may, therefore, strengthen the possibility of the royal poet being the author of the
short Sanskrit play, the Traivikramam.
 NINETEEN

Newly Discovered Monuments

at Mmallapuram1
The little shrine and examples of sculpture discovered (1990) by
the Archological Survey of India beneath the sands, immediately to
the north of the main complex of the Shore Temple, Mmallapuram, are
important new additions to the known works of the Pallava king, Nara-
sihavarm-II, more commonly referred to by his title, Rjasiha.
The center of interest in the area uncovered is a small cylin-drical
shrine which stands in the focal point of the protective stone wall
which curves around it on the southern side. Positioned against this
wall on the southern side is a massive image of the Varha (Boar)
Avatra of Lord Viu, shown here completely in animal form, with
its snout rooting downward into the earth. Water is indicated by the
lotuses under its body. This image is carved in the round from the
bedrock.
Inscriptions
On the front face of the pedestal of the Varha image are en-
graved three of Narasihavarms titles: r Rjasiha, r
Raajaya, and rbhara. On the west flank of the pedestal is in-
scribed: r Citrakrmmuka. These four titles, along with the honor-
ific r, may be translated as: The illustrious Lion among Kings,
The illustrious Victor in Battle, The Upholder of Pros-perity, and
The illustrious Wonderful Archer.
These four titles appear in other inscriptions of Rjasihas
elsewhere in the Shore Temple, itself, and in Kanchi and Panamalai.
In Kanchi
The four titles are among the birudas inscribed on the faades of
three of the little shrines surrounding the main Kailsantha temple.
(Ref. to Chp. 16, above, for further details.) r Rjasiha is the
first title engraved on Shrine No. 1. r Raajaya is the third title on
the same shrine. rbhara is the third title on the third shrine; and
r Citrakrmmuka, the second on shrine 14. (The shrine numbers
which I am using are the ones engraved on them by the A.S.I.)
Again, these same four titles appear in the final verse (V. 12) of
King Narasihas inscription around the outside of the main shrine (the
main vimna) of what is today called the Kailsantha temple, Kanchi.2
This verse reads:
Rjasih Raajaya[] rbhara-Citrakrmuka [|*]
kavra-cira ptu [i]va-cmair-m-mahm [||12||*]
May Rjasiha, the Victor in Battle, the Upholder of Prosperity, the Wonderful Archer,
First among Heroes (kavra), (who has) iva for his crest-jewel, for a long time protect the earth!
 224 In Panamalai
Pallava Art In a natural cavern near the hill at Panamalai, there is a slab
inscription which repeats just this one verse (the twelfth verse of the
Kailsantha inscription).
In Mmallapuram
This same verse is part of Rjasihas inscription on the
baliphas (the stone altars for offerings) for the three shrines of the
main complex of the Shore Temple.
Fragments of a second newly discovered inscription are found
on the inner faces of three upper rim stone slabs on the south western
curve of the retaining wall which surrounds the cylindrical shrine.
They read, in the following order, from left to right:
(1) . . . Rjasih ya katrasiha iti viruta-puya-krtti jy-
(2) t-[Ma]hvara-ikhmai-dpta-mauli-
(3) [ryyu]ddhrjjun nikhila-lka-nar . . .
These lines are identical with the major portion of the following
passage found in Rjasihas Vyalr Pillar Inscription:
r-Pallavnvaya-kulcala-Rjasih
ya katrasiha iti viruta-puya-krtti |
Jyt-Mahvara-ikhmai-dpta-maulir-
y-yuddhrjjun nikhila-lka-Narndrasiha ||3
Which may be translated:
The illustrious Rjasiha, of that mountain range which constitutes the Pallava dynasty,
Whose well-merited fame is widespread as the Lion among Warriors (Katrasiha),
(For whom) Mahvara is his crowns gleaming crest-jewel, (and) who is the Arjuna-of-War,
Long may he live, that Lion among Kings (Narndrasiha), Ruler of the entire world!
This verse proclaims King Rjasihas devotion to Mahvara
(iva). It is, therefore, interesting to find such a striking image of the
Varha Avatra of Viu among these monuments. Unfortunately, this
sculpture has been deliberately broken into pieces at some time in the
distant past. More than one series of wedge holes were chiseled into
the back and flank of the Boar, and then the stone was split apart. The
pieces of the image have been reassembled by the A.S.I. so that it is
now close to its original form.
Cylindrical Shrine
The little cylinder-shaped shrine is very unusual. Its adhi-
hna the part from the molded base up to the floor of the miniature
sanctum has been carved out of the bedrock, a fine quality, beige
gneiss. The rest of the shrine is made up of three carved blocks of
another type of stone placed one on top of the other on this fixed base.
These three blocks were lying scattered under the sand until the A.S.I.
cleared the area, found them and refitted them, forming, once again, a
complete shrine.
The first block, which forms the main walls of the sanctum
(the pda), with four pilasters having rampant yis with riders carved
in front of each, rises from the base of the sanctum to the top of the 225
pilasters. New Discoveries
The second block forms the cover to the sanctum, with over-
hanging cornice (kapta) and the narrow neck (grv). The cornice is
decorated with four horse-shoe shaped windows, unfinished in detail,
which are positioned, one each, over a pilaster. On the upper surface of
the cornice is a ring of four little, horned yi busts alternating with four
little gaas (goblins). The grv has four conch-blowing gaas carved
directly above the yis on the cornice.
The third block forms the crowning cupola (ikhara), with
four horse-shoe windows (nsiks or kus) carved on the east, south, 5.5
west, and north sides. A miniature bas-relief image of Gaapati is
found within each of these windows.
The topmost finial (stp) is missing.
Every part of this little shrine is round on the outside except
for two sub-plinth moldings (upnas) at its very base. The lowest of
these moldings forms a regular octagon; the one just above it has six-
teen sides.
It is an interesting fact that another example of a slender,
cylindrical shrine is found carved in high relief under the great arches
of the Gaa Ratha, Mmallapuram. This rendition of a cylindrical
Cylindrical Shrine,
shrine is severely simple, lacking ornamentation such as pilasters, yis, Shore Temple
and gaas.
On the eastern side of the Shore Temples cylindrical shrine is
the opening to the cubical sanctum. On the back wall of this tiny cell is
carved an image of Ardhanrvara (the hermaphrodite, composite
image of iva and his consort, Um). The right half of the figure is 5.5
iva; the left half, Um. This figure is seated on the bull, Nandi.
Ardhanr holds a v (lute) diagonally across the chest.
I must add that these details are not clear in the little image of
Ardhanr in this shrine. One must be acquainted with this form of the
god/dess as portrayed by Pallava artists elsewhere. One of the finest,
clearest, and most accessible examples of Vdhara Ardhanr is found
carved on the outer, western side of the Kailsantha temple (main
vimna), Kanchi. The largest and most impressive panel showing just
this form of Ardhanr, seated on Nandi, holding a v, is found on the
inner left wall of the sanctum sanctorum of the Vdagirvara temple,
Tirukkalukkunram. However, a visitor to this temple might not even
notice this masterwork of Pallava art because there is such dim light Cylindrical Shrine in
within the sanctum.4 relief, Gaa Ratha
Historically, there was a close connection from the time of
Rjasiha between the iva temples at Tirukkalukkunram and the
Shore Temple (iva shrines), Mmallapuram. For hundreds of years,
the deity was brought from Tirukkalukkunram to Mmallapuram in a
ritual annual ceremony. In recent years, however, this custom has
been abandoned due to friction between dominant groups in each place
(aivites in Tirukkalukkunram; Vaiavites in Mmallapuram).
 226 The extraordinarily slender, cylindrical shape of the newly
Pallava Art discovered shrine suggests that it is a novel expression of the Ligd-
bhava theme. That the Ardhanr form of the Almighty is at the heart
of the Divines creative power, symbolized by the liga form, would be
most appropriate.5 The suggestiveness of this interpretation would
incorporate into the Ligdbhava myth the image of the Boar nearby
Vius taking this form to root downward to try to find the lower limit
of the pillar (liga). Will an image of the sacred Goose (Hasa the
corresponding form of Brahm) also be found in the sands nearby?
Cistern
To the north of the little shrine, at ground level, there is a small
circular cistern carved out of stone. Recessed into its eastern side is a
small bas-relief carving of a royal-looking lady seated at ease on a
throne, with two attendant females standing behind.
Retaining Wall
Some observers have proposed that the wall around the little
shrine was the wall of a spacious temple, apsidal in form, and that the
little cylindrical shrine was the central object of worship within this
much larger temple.
This view seems to me to be mistaken for the following two
reasons. First, the wall is stepped a characteristic more appropriate to
a retaining wall (to keep sand out) than to a wall of a temples sanctum
(vimna). Second, the inscriptions on the Varha images pedestal and
on the inner face of the wall would indicate that there was sufficient
light to read the inscriptions, and that people moved freely within the
confines of the wall. This freedom would not have been there if this
were the sacred area within the sanctum of an apsidal temple.
Chronology
We learn from the Cir Grant (copper plates) of the Pallava
king, Npatugavarm, that one of his predecessors, King Narasiha-
varm (the First), built out of stone, on the ocean, an abode for the
One who possesses the mighty discus [i.e., Viu] to recline in:
Siha r-Narasiha ity-ari-kula-sthambramm-abhd-ya-
ay-gham-amabhir-j-jalanidhau cakr Mah-cakria ||7||6
The great poet Dain, in the latter half of the 7th century,
A.D., tells of his visit to the Reclining Viu image on the sea-shore
at Mmallapuram. It is evident from his account that the two iva
sanctums had not at that time been built by King Rjasiha. Dain
speaks of the image of Viu as having been made by the ancients.
Therefore, I would suggest that the Viu image was carved in a period
even earlier than the reign of Narasiha-I (Mahmalla) probably
during the reign of Sihaviu, Mahmallas grandfather and that it
originally was in the open air. Then, in the reign of Narasiha-I, the
king had a superstructure built of stone blocks. Five or so decades later,
Rjasiha, first, created the little cylindrical shrine for the Ligd-
bhava Ardhanrvara, establishing the presence of iva and Um in
this holy place, along with an image of Varha, and, then, at some later
date, he radically transformed the site by constructing two iva shrines,
 227
12.5

10.5

Cistern

12.5

8.6

Stepped wall around cylindrical shrine

 228 one (the Katriyasihvara) in front of, and the other (the Rjasi-
Pallava Art hvara) in back of the Reclining Viu. At the same time that Rja-
siha built these two iva temple towers, he rebuilt the superstructure
of the Viu shrine and named it Narapatisiha-Pallava-Viu-
Gham, incorporating this shrine architecturally and visually into the
prkra wall of the larger, east-facing iva shrine, the Katriyasih-
vara. The assortment of shrines on the shore of Mmallapuram was,
thus, finally, brought to the number which we see today at this place.
Further Discoveries at the Shore Temple
The Eastern frontage of the prkra wall of the Shore Temple,
with its little proto-gpura doorway, was built on a large, massive rock
which forms the very foundation of the whole temple complex.
Until a few decades ago, the waves of the Bay of Bengal broke
against this rock formation which curved downward into the sand and
water. Then, in the mid-century, a groyne wall or breakwater was
constructed several meters in front of the temple to protect it from the
sea. The rock formation just in front was covered with sand.
In 1991, the Archological Survey of India began a little dig
and exposed once again the rock in front of the temple. What is very
interesting, indeed, is the fact that there is a slide carved in this rock.
Did the children of that time, more than 1200 years ago, slide with a
splash into the waves?
The rock surface to the left (south) of the slide has six sockets
cut into it, presumably to hold six pillars of a small shrine. This is the
second six-sided shrine we have noticed at Mmallapuram. Skanda
(Subrahmaya/Muruga), the six-faced god, was born on the banks of
the river Gag. We have, here, a re-creation of river and bank!
A groove in the rock runs straight across the floor of this
shrine, from back to front. The groove, in part, appears to have been
formed by nature. Was it venerated as a svayambh (self-born) yn,
the feminine counterpart of the liga, enshrined in the Shore Temple?

10

6.7
 229

15

18

The rock formation on which the Shore Temple stands is exposed once again.
The sea water used to reach this rock until a groyne wall was built in front during
the early part of this century. Note the (childrens?) slide cut in the stone.
 230 An Ancient Breakwater
Pallava Art In 1992, yet another ancient feature of the Shore Temple
complex was uncovered beneath the sands. An excavation, to the
south-west of the temple, has exposed a stepped structure. What at
first appeared to be the stepped side of a temple tank, has turned out to
be the elaborate stonework foundation of a breakwater. This ancient
groyne wall is presently a considerable distance inland from the
shoreline. In the seventh and eighth centuries, however, the sea must
have reached around in back of the complex, almost creating a little
island on which the monuments were located.
Blocks of reddish laterite stone form the basic material of this
wall. But a stepped system of keyed granite slabs was designed to keep
the wall from yielding to the force of the waves which dashed against

15

11

Keyed structure Breakwater foundation

 231

15

11

Breakwater wall intact section

 232 it. A short length of this wall is [was] still relatively intact from top to
Pallava Art bottom. It can be seen that rough lime plaster had been used to fill the
gaps between the stone blocks.
_______________

1Based on M.C. Lockwoods essay of the same title published

in Indological Essays: Commemorative Volume II for Gift Siromoney
(1992), pp. 44-56.
2First published by E. Hultzsch in South-Indian Inscriptions
(1890), Vol. 1, pp. 12-13.
3Epigraphia Indica, Vol. XVIII, pp. 145-152.
4We have described this panel in detail in the second study of
this book.
5In May, 1994, I visited an exhibition of South Asian Art at
the Metropolitan Museum of Art, New York. One of the exhibits
displayed one half each of two portable, stone ligas, only a few inches
tall. (Only the right half of the larger liga has survived; and only the
left half of the smaller liga.) The divide in each liga is from top to
bottom and front to back. When closed, the two halves of each liga
would have formed an ka-mukha-liga (a liga having ivas anthro-
pomorphic face on the front). When opened up, in the manner of a
diptych, there would have been revealed carvings on the inside of both
halves. On the inside of the extant, larger liga-half, there are four
relief images: iva, Um, tiny Skanda holding the vl (spear), and
Nandi (all of them, thus, forming a miniature Smskanda!). The
exhibitors had dated the larger liga in the seventh century A.D. The
smaller one, from the sixth to the seventh century. These miniature
ligas could have been used in household worship, and, thus, may be
considered portable, private versions of the type of cylindrical shrine
built by King Rjasiha, which was fixed, royal and public. (I am
speaking, here, of general practice, and am not suggesting any direct
copying, one way or the other!) The monthly magazine, Span, pub-
lished by the U.S.I.S., Delhi, had a photograph of the larger of these
two portable ligas on the cover of its June, 1992, issue. The following
information about it was given: One of the Indian art objects from the
Samuel Eilenberg Collection on display at the Metropolitan Museum
of Art in New York City Section of a Portable Linga with Shiva and
Parvati, Kashmir, 7th century, chlorite, height 7.6 cm. [3 in.].
6See Copper Plate Inscriptions of the State Museum, Vol. III,
edited by N. Ramesan (Hyderabad: Govt. of Andhra Pradesh, 1972),
pp. 170-171.
 233

Half of the larger ka-mukha-liga,

Samuel Eilenberg Collection

The Metropolitan Museum of Art, gift of

Samuel Eilenberg, 1987. (1987.142.66)
Photograph by Otto Nelson.
All rights reserved, The Metropolitan Museum of Art.

Inner figures of the ligas half

The Metropolitan Museum of Art, gift of

Samuel Eilenberg, 1987. (1987.142.66)
Photograph by Otto Nelson
All rights reserved, The Metropolitan Museum of Art.
234

6.5 6.5

8.5 8.5

Fig. 1. In the foreground: Mini Well, Fig. 3. Ardhanr image inside

Cylindrical Shrine, and Varha. the Cylindrical Shrine.

6.5 6.5

8.5 8.5

Fig. 4. Relief image of a cylindrical Fig. 5. Mini Well, with carving of a

shrine, Gaa Ratha. seated, regal-looking lady.
 TWENTY

Vdhara Ardhanrvara
A popularized and somewhat (editorially) transmogrified
account of my previous essay on newly discovered monuments at
Mmallapuram, together with a number of photographs, was published
in the July 30th, 1993, issue of the magazine, Frontline. It generated an
interesting exchange of letters to the editor, which appeared in the
August 27th issue of that magazine. I reproduce, below, with permis-
sion, the two letters and the three photographs which accompanied
my letter:

Mamallapuram
This has reference to Dr. Michael Lockwoods article, Stone
stories (July 30), on the recent discoveries at Mamallapuram. The
writer neither was associated with the excavation nor seems to have
consulted any archaeologist who made the discoveries. He has wrongly
identified some important sculptures, and I am writing because this
excavation has added a new chapter to Pallava history and architecture
at Mamallapuram.
Fig. 1: The author identifies the engraved figure in the sanc-
tum as Ardhanarisvara. In fact, it is Vrishabhantika-Siva (Siva leaning
on his vehicle, Vrishabha or bull). His description of the Varahas
snout pointing downwards to burrow through to the lower extremity of
the Sivalinga is purely imaginary. No Sivalinga was installed in the
Garbhagriha or sanctum sanctorum during Rajasimhas time, whereas
the famous Somaskanda panel can be seen in all temples of his period.
Fig. 3: The engraved figure is identified as Ardhanarisvara as
well as Veenadhara-Ardhanarisvara. As far as I know there is no
Veenadhara-Ardhanarisvara, especially in the Vrishabhantika pose,
occurring during the Rajasimha period. Here, the figure in the mini-
ature temple is that of Vrishabhantika-Siva.
Fig. 4: The author compares the miniature temples vimana
with the Ganesa rathas kanta-bhithi relief. This is not acceptable
because the vimana has kudus, ganavali and kirti mukha while the
Ganesa rathas relief is simple and plain.
Fig. 5: The seated figure is identified as a royal lady whereas it
is the river goddess attended by her servants. No royal lady would sit
on the rim of a well.
K.T. Narasimhan
Archaeological Survey of India
Madras Circle
Madras
 236 Dr. Michael Lockwood writes: With regard to K.T. Narasimhans
Pallava Art first paragraph, with all due respect to the Archaeological Survey of
India and its officers, they are not the sole, nor necessarily the final
authority in identifying and interpreting ancient sculptures.
Regarding the problem of identifying the bas relief figure in the
sanctum of the cylindrical shrine (Figures 1 & 3) I said in the article
that a familiarity with similar portrayals by Pallava artists elsewhere
would help. There is a very large carved panel, representing this very
same figure (Veenadhara-Ardhanarisvara seated on the bull, Nandi) on
the inner proper [right] (north facing) wall of the sanctum sanctorum of
the Vedagirisvara temple at Tirukkazhukkunram. Almost 20 years ago,
my colleagues, Dr. P. Dayanandan and the late Dr. Gift Siromoney,
and I published a detailed description of this Rajasimha-style Veena-
dhara-Ardhanarisvara in our book Mahabalipuram Studies (1974).
As this detailed and well-preserved carved figure is inside the sanctum,
photography is not permitted. However, anyone who would take the
time and trouble to make a careful visual study of this Vedagirisvara
carving would understand the logic of my identification of the Shore
temple figure.
To support my argument further, however, I have photographs of
three smaller sculptures of Veenadhara-Ardhanarisvara belonging to
the Rajasimha period.

Photograph A is of a sandstone image which [was] found in the

courtyard of the Kailasanatha temple, Kanchipuram, and is remarkably
similar to the one in the much larger Tirukkazhukkunram panel. The
figure in this photograph, like those of Tirukkazhukkunram and
Mamallapuram, is also seated on Nandi.
 Photograph B is of a figure of Veena-
7
dhara-Ardhanarisvara seated on a plain
throne not on Nandi. This panel, carved on
one face of a four-sided block of granite, was,
at the time the picture was taken, in 1969,
located in the forecourt of the Shore temple.
The figure in this panel is almost identical in
attributes and pose to the Tirukkazhukkun-
ram, Mamallapuram and Kanchipuram
images. Yet, as there is no bull in this panel,
obviously, this figure cannot be called
Vrishabhantika-Siva.
Photograph C is of a figure of Veena- 10.7
dhara-Ardhanarisvara, also seated on a plain
throne. It is carved on the west side of the
outer wall of the vimana of the Kailasanatha
temple, Kanchi.
In response to Narasimhans claim that
during Rajasimhas time no Sivalinga was
installed in the garbhagriha of his temples,
I need only point out that this thesis has been
a matter of scholarly debate for some years
now. My own position is that the Sivalinga
in the Shore temple is original, and I have Photograph B
argued this point in my book, Mamallapuram
and the Pallavas (1982).
In all Pallava art, be it poetry, drama, or
sculpture, there is dhvani (suggestiveness). 7
Thus, I have suggested that the image of the
Boar Incarnation of Vishnu (Fig. 1) can be
viewed as illustrating the Lingodbhava
legend. This theme is found well illustrated
in the sculpture of Rajasimha.
In Fig. 4, I illustrated another cylindrical
shrine carved in high relief under the south-
ern arch of the Ganesa ratha. That there is
some similarity, as I have stated, between
this relief carving of a shrine and the newly- 9
discovered cylindrical shrine, I leave to the
readers to judge for themselves.
Finally, there is the question of the
identity of the seated figure, carved in relief
on the rim of the well (Fig. 5). Is it a royal
lady or a river goddess? Let me answer this
question by saying that in the time of the
Pallavas as in our own the line between
queen and goddess was exceedingly thin,
and sometimes non-existent.
Photograph C
 238 Postscript 1997:
Pallava Art Photograph A was taken by me in the late 60s. This
Vdhara Ardhanrvara carving has, at some later time, been
removed from Kanchipuram and is presently being exhibited,
along with the carved block (Photograph B), in the A.S.I.s site
museum at Mmallapuram!
 TWENTY ONE
The Kuumiymalai and Mmar Inscriptions
of King Mahndravikramavarman: A Review1
The Kuumiymalai and Mmar Inscriptions of the great Pallava king, Mahndravikramavarman, are
extremely tantalizing each of them in its own way.
The Kuumiymalai Inscription, though excellently preserved, presents a puzzling record of musical
exercises involving various notes, to be played on the Parivdin, an ancient type of v, having seven strings.
Among the few scholars who have attempted to solve the enigma of this inscription, there has been no consensus
about its ultimate interpretation.
The Mmar Inscription, on the other hand, has suffered grievously from the ravages of time. The rock
surface on which it was engraved has crumbled away in many places leaving only patches of readable words and
passages, which now constitute, perhaps, a little more than half the original inscription. It has been extremely
difficult to get a coherent idea of the over-all flow of the text.
To assist in the re-translation of the Mmar Inscription, we have devised a transliterated version of the
text, together with a word-for-word translation, directly underneath, which reproduces (approximately) the spacing
of the words in the original inscription. This version is given on the next page. Our attempt at a running translation
of the main body of the text, starting with line 6, now follows:
Mmar Inscription (lines 6 to 15)
[6] . . . Naka (one of the ten types of drama) . . . Vysas equals (work) . . . entitled Bhagavadajjuka (i.e., the
work of him who is Vysas equal, which is entitled Bhagavadajjuka), (which manifests the essence of) Hsya
(Rasa) ( i.e., which manifests the Laughable); Mattavilsa (which is both the title of one of the kings plays and
a word signifying the highest degree of uncontrollable laughter), the quintessence (uttama) of the Prahasana
(one of the ten types of drama), (which represents life from its) beginning (di) (to its end!).
[7] . . . Prkt . . . exciting . . . (these) four (plays) . . . She, who having (taken) delight in (her) victorious
husband, King atrumallas (singing),
[8] and gaining (through concentrated practice) a voice (resembling) the sound of honey-bees,
Who established (herself as) possessing the enlightening poetic intellectual insight, which
equalled that of her husbands,
[9 end] . . . Who, earlier, attained to the discipleship (of atrumalla) in the hallowed stras
[9 beginning] . . . (on account of) her (singing) lofty-pitched syllables (conveying) poetic speech, full of meaning.
[10] . . . Who became the veritable goddess of music and art, in the company of her creator-husband . . .
. . . (He, the king,) together with his senior wife,
[11] . . . analyzing the rules (culled) from tradition, distinguished (the three ways in which the instrument may
relate itself to the voice:) Vtti, Daki, and Citr.2 . . . Having carried (this) out (i.e., accomplished it)
according to rules,
[12] (by) arranging (musical) syllables into groups of four, he established instrumental music (at a level) not
achieved previously (by any of his predecessors), . . . (following her singing, in which) she (had
distinguished herself) as (being) fully accomplished . . .
[13] . . . (on account of her) extraordinary wealth of vocal musical quality, which, with elation, was made
manifest by him (on the instrument) . . .
[14] . . . With her (i.e., in the company of her), (one) who, needless to say, possessing the enthusiasm of
Lakm, as well as extraordinary character, . . .
[14 & 15] . . . whose inner brilliance was manifest in her teeth,
[15] resembling the crescent moon (Candralkha) . . .

* * * * * * * * *
avinitaka-samddha 1 vttamasya-vija[l] gandharvvastramakhila khilamukhdgata
arrogant arise conduct-of music science whole mouth-sprung
2 prjpatya nd-bhvangama[na] pu gandhn impclanirmmita
Brahm-of imagination-coming fragrance Pcla-in made
khya ndh 3 tt [a]iktuna manbhirma ma vlmkivar[]ita
aiktu-by mind-pleasing by-Vlmki described
bharatnyaka sabh 4 vastrpahra [ri]pvajrasyaka[]
actor-leading (hero) assembly clothes-stripping (grabbing) enemy thunderbolt-arrow
kma[kta] ndhata cra 5 mudrrtthamurvvasarvvabhan
crying Urva all- shining
n[]rca nakam || 6 vysakalpasya bhagavadajjuk[khyah]syammattavilsdipadamprahasanttama
woman & Naka Vysas equals Bhagavadajjuka titled the laughable Mattavilsa starting word Prahasana quintessence
[r]m-prgta 7 samttjita [c]tuaya jayasyapatyuatrumallasyabhbhuja[]
prkt exciting four victorious-husbands atrumallas kings
d-bhramarvptasammad 8 y kavnm praka vatt [sa]mpattissamabuddhirivasthit
bee like=gained delight who poets-to enlightening containing wealth equal-know-how remained
gurusvaravaray 9 purtasy kavigi[rajasyrtha]vat []strapuntuiyatvamyua
pitch note syllables-by-her in the past her poetic speech born meaning-full-of stras- hallowed discipleship attained
prajpati samanvitm 10 pr[a] []tta jyyay-svay[]
Creators with consort (Sarasvat) elder wife-with his own
kalpt pravibhajya 11 [v]tti-dakia-citrkhya[] ra [kra]yitvyathvidhi
(from rules established by) tradition analyzing Vtti- Daki- Citr=called having done accd.-to-rule
[ya]cavividhaiktvvaracaturtthaya 12 aprptaprvvannivuvdya ravaa ya ta ktavatva
various having-made syllables fours-groups-of un-attained previously establish-to instr. sound-heard (she) fully-accomplished=as-if
[kaha]ruti-gud- asdhra[a]sampad 13 dvvtkara[]na yin kva[sdhakasa]
vocal note- quality-by extraordinary wealth-by having-made-it-manifest elation-with him-by
yat visavda lakmy[mantsu]katay 14 yasylavi[a] snyata ydantam-antara
because needless-to-say Lakm-with mind-excitement due-to whom-by character extraordinary-with ablution whose teeth inner
[knti]yathrtthna candralkva ygat 15 gtrandhaha ta stru nityavihitabudha
brilliance truly-like moon-crescent-like whos-attained body-wealth (all) stras-in ever well-versed scholar
yu parprtissapanna syavapu 16 nityavintna satyasa[ndhna] sya bhaktyvarjjitamaulin
in-whom highest affection wealths abodes Nityavinta-by Satyasandha-by devotion-filled mind

[sa]hasrajharasapramghaymasyagarjjita 17 [madama] pati vatay

thousand torrents-full cloud dark roar jointly-with
 241
Line 6 of the Mmar Inscription mentions the titles of two plays which are close to our
hearts. We have made a concentrated study of these plays over the last twenty years, publishing the text
and translation of Bhagavadajjuka, first, in 1978, and of Mattavilsa in 1981. Our fourth edition of the
plays (both revised for the third time) was published in 2005.3
There has been a long-standing debate about the authorship of the play, Bhagavadajjuka. An
anonymous, sixteenth century commentary on this comedy was found in Kerala, which declares that
Bdhyaa was the author of Bhagavadajjuka. But, who was this Bdhyaa? No one has come
forward with an answer. Therefore, the Mmar Inscriptions mentioning the titles Bhagavadajjuka
and Mattavilsa in the same sentence, one after the other, led early scholars, such as C. Minakshi and
V. Raghavan, to take it for granted that the author of Bhagavadajjuka was King Mahndra since
Mattavilsa is acknowledged by everyone to be his work. But other scholars, especially those from
Kerala, have denied King Mahndra his due, and have maintained that Bhagavadajjukas author was
Bdhyaa, who was supposed to have lived several centuries earlier than the Pallava king. In our
editions of the two plays, we have marshalled detailed arguments in support of Mahndras authorship
of Bhagavadajjuka. In this paper, we shall just take for granted his authorship of it.
Let us, then, examine Line 6. The beginning of the line is obliterated. The first three readable
words are n[]r-ca nakam. We can only say that a reference is being made, here, to a Naka (one
of the ten types of drama), in which a woman (nr) plays an important part.
The lka, immediately following in the same line, begins with the expression Vysa-kalpasya.
The poet king, Mahndra, is being compared, here, to the great Vysa. But it would be a mistake to
interpret this comparison literally. Mahndra enjoyed poking fun at the whole world, including himself.
Vysa, as author of the Brahma-Stra, was famous as a Strakra. In the comedy, Bhagavadajjuka, the
stra-spouting of the main character, the Parivrjaka, is a parody of the outpouring of the great Strakra,
Vysa. The comparison made between Vysa and King Mahndra, in this line of the inscription, is,
therefore, itself, only a pleasant joke.
The author of Bhagavadajjuka did not include the term Prahasana in its title. Its title is,
simply, Bhagavadajjukam or Bhagavadajjukyam. Whereas, the other plays title is Mattavilsa-
Prahasanam. In Line 6, the play Bhagavadajjuka is linked (according to our interpolation) to the techni-
cal term Hsya (The Laughable), whereas Mattavilsa is specifically associated with the highest
degree of uncontrollable laughter (mattavilsa), and is called a Prahasana. The word di, immediately
following Mattavilsa, should not be translated simply as etc.! We get a clue to the primary meaning
of di, here, from the Prastvan of Bhagavadajjuka. The playwright (Mahndra), through the mouth of
the Stradhra, declares that, of all the different types of Rasas, the Hsya Rasa is the primary, most
important Rasa. And he is not talking only about dramaturgy. He is talking about life in general from
its beginning, through its middle, to its end! It is just this expansive idea which has been encapsulated in
the little word di. It is in this sense that we understand the last part of Line 6 to be an assertion that the
play, Mattavilsa, conveys the quintessence (uttama, i.e., it is the very best embodiment) of the Highest
Degree of the Laughable (that is: mattavilsa), which is the predominant flavor in the Prahasana and in
all of life, from birth to death!
Line 7 is also only readable after an obliterated initial stretch. The first word clearly decipherable
is Prgt (for Prkt). From this one word we can deduce that King Mahndras scribes mother
tongue was Tamil. Just as we see today, in Tamil Nadu, the name Rnuk inscribed in roman letters as
Renuga, the same process was occurring over 1300 years ago at Mmar in the Pallava script. The
next expression, samttjita translates as exciting. Then, after an obliterated gap, there is the word
ctuaya, which sums up four of something. Since the preceding passage has touched on a Naka, a
Comedy, a Prahasana, and another work where Prkt was important, may we not assume that the four
somethings referred to by the term ctuaya are four plays which the king wrote which excelled in
different genres?
 242
Then, after another gap (but still in Line 7), the inscription begins to describe the senior queen
of King Mahndra (the queen he is holding by the wrist in the portrait sculpture in the divarha Cave-
Temple at Mmallapuram). The inscription, from this point up to but not including the last readable
fragment in Line 10, gives grammatical priority to the queen. This queen, whom we shall call
Chandralkha (on the basis of dhvani in Line 15), was inspired by her husbands singing. It should be
noted, here, that Mahndra assumed the Tamil biruda Palapi (One [who sings] many songs i.e.,
One who has a great repertoire of vocal music).
Line 8: There is an obliterated portion in the beginning. Then we learn that Chandralkha
developed a voice which had the quality of the sound of honey bees. The end portion states that her
intellectual grasp of the arts was equal to that of her husbands.
Line 9: Again, the beginning of the line is unreadable. The first fragment of this line seems to
indicate that her voice was excellent in the high registers, and was very expressive of poetic meaning
(in the lyrics). The readable portion at the end of this line states that she had become a iy of the king,
well educated in the sacred stras.
Line 10: Another break. Then a fragment indicating that she was like the goddess of music and
art (Sarasvat) in the company of her creator-husband (Brahm).
Towards the end of this line (10), the inscription shifts to the king as the primary grammatical
subject, though his queen still shines supreme. The king, together with her, [Line 11:] studied the musical
traditions of the past and distinguished what are called Vtti, Daki, and Citr (the three ways in which
the instrument may relate itself to the voice, as described in the Nya-stra). Then, creating an ordered
system of rules, [Line 12:] by arranging musical syllables in groups of four (as we find in the musical
inscription at Kuumiymalai), he established instrumental music (for the v) at a level which had not
been attained before. And the king did this by being able to follow (on the instrument) the music she was
so expertly singing.
T.N. Ramachandran, one of the earliest scholars to have commented on these passages, gave
(in 1931) an interpretation quite opposite to what our translation, above, indicates concerning the relation
of instrumental music to singing:
The king wanted to achieve what was not achieved before in the realm of music. Seeing the various
intonations of sounds that stringed musical instruments like the v alone could produce and actu-
ated by a zeal and determination to produce the same results in vocal music, he designed, by dint of
uncommon and superior resources probably a successful notation of musical sounds to be produced
in vocal music. . . .4
Ramachandrans statement seems to fly in the face of the almost universal belief that the human
voice is the supreme instrument. Now, there are several accounts, coming down to us from the Pallava
period, which go against Ramachandrans view. Minakshi mentions the incident, recorded in the Periya-
Puram, involving the singing saint-child, Tirujnasambandhar, and the renowned v player,
Tirunlakahappar. After the two met, Tirunlakahappar
resolved to spend the rest of his life in the company of [Sambandhar,] playing on his matchless y
[v] every song that his young master produced.5
Unfortunately for him, there was one song which Sambandhar sang which he was not able to
follow on his instrument. He then determined to master the difficult passage or never play again. How-
ever, in this attempt, he broke his instrument. (Though it was probably his pride which was broken, and
not the instrument, since the child-saint, in a following verse, requests him to continue playing it.)
Then, there is the lka, attributed to di-akara (who lived during the Pallava period), which
seems to stress the same point:
 243
Vipacy gyant vividham-apadna paupats-
Tvayrabdh vaktu calita-iras sdhu-vacan |
Tadyair-mdhuryair-apalapita-tantr-kala-rav
Nij v v niculayati clna nibhtam ||6
Once, again, the superiority of the voice (Prvats) over the instrument is demonstrated.

Kuumiymalai Inscription Sketch of Publishing History:

1. The Kuumiymalai musical inscription is first discovered in 1904.
2. First reported in the Annual Report on Epigraphy Southern Circle, Madras, 1905.
3. First edited by P.R. Bhandarkar in Epigraphia Indica, Vol. XII, pp. 226-237, this scholar lays the foundation for
further research.
4. T.A. Gopinatha Rao comments, in the Indian Antiquary, Vol. LII, on the Tirumayyam Cave inscriptions (com-
panion inscriptions of those at Kuumiymalai though, unfortunately, the main Tirumayyam inscription has,
in the past, been almost entirely erased).
5. T.N. Ramachandran, in a paper presented during the inaugural session of the Indian Historical Congress, 1931,
brings to the notice of the scholarly world the import of the Kuumiymalai Inscription. His paper, The
Royal Artist, Mahendravarman I, is then published in two parts in the Journal of Oriental Research, Vol.
VII: part iii, pp. 219-246, and part iv, pp. 303-330.
6. V. Raghavans note, Dakia Citra, in The Journal of the Indian Society of Oriental Arts, Vol. VI (1938),
corrects Ramachandrans interpretation of the terms, Vtti, Dakia, and Citra (as pertaining to the art of
painting) in the 11th line of King Mahndras Mmar Inscription. Raghavan correctly points out that
these three are musical terms, which are discussed in the Nya-stra. (The Mmar and Kuumiymalai
inscriptions should be read together.)
7. C. Minakshi, in her book, Administration and Social Life under the Pallavas, first published in 1938, presents the
most detailed study, as yet, of the Kuumiymalai inscription. This book, along with two others, represents
the results of her research for the Ph.D. degree (Madras University), which she earned in 1936.
8. K.R. Srinivasan edits a transliterated version of the text of the Kuumiymalai Inscription in the 1941 publica-
tion, Inscriptions in the Pudukkottai State, Translated into English, Part I, Early Pallava and Chola Inscrip-
tions (Pudukkottai: Sri Bhadamba State Press), pp. 3-10.
9. V. Premalatha produces a Ph.D. thesis (Madras University) in 1964, Sources for the Construction of a
Detailed History of Indian Music. She is guided, in her research, by her supervisor, P. Sambamoorthy,
and T.N. Ramachandran. After C. Minakshis pioneering analysis, hers is the most thorough study of the
Kuumiymalai Inscription.
10. V. Premalatha and S. Ramanathan present separate, short papers on the Kuumiymalai Inscription at the 1966
Seminar on Inscriptions, Madras. These two papers, Kudumiyaamalai Inscription on Music, pp. 29-31 (by
V.P.), and Music from Inscription, pp. 32-35 (by S.R.), are among the speeches and papers published in the
proceedings, Seminar on Inscriptions 1966, ed. by R. Nagaswamy.
11. C. Minakshis book, Administration and Social Life under the Pallavas (listed as no. 7 above) is reworked by
K.K. Pillay and transmogrified into a revised edition, which is published in 1977 by Madras University.
Inserted as an Appendix to her XVIth chapter on Music is a short piece written by P. Sambamurthy7
(supervisor of V. Premalathas Ph.D. research). Sambamurthys Appendix is a paraphrasing and sometimes
verbatim reproduction of passages from his own candidates (1964) thesis, which, surprisingly, he does not
acknowledge.
12. V. Premalathas Ph.D. thesis is finally published in 1985, with a new title, Music through the Ages (Delhi:
Sundeep Prakashan).
 244

13. Richard Widdesss book, Ragas of Early Indian Music: Modes, Melodies and Musical Notations from the Gupta
Period to c. 1250, is published by Oxford University Press in 1995.
What we present in the following eight pages are the text of the seven Svargamas (pp. 245-47);
Minakshis Table III8 (page 248), showing the varjya svaras or rutis in each of the seven Kuumiy-
malai Groups (we include, in this table, a few interpolated corrections); Minakshis Table, divided
further by us to show the same for each of the 38 lines of the seven Svargamas (pp. 249-50); the same
38 lines, but shown with the actual number of occurrences of a given ruti in each line (pp. 251-52).
_______________
1This study is based on a paper read by M.C. Lockwood and A.V. Bhat, on April 26, 1997, at a
meeting of the XXIII Annual Congress of the Epigraphical Society of India, Thanjavur, Tamil Nadu.
2The three ways in which the instrument may relate itself to the voice, according to
Abhinavaguptas commentary on the relevant Nya-stra passage, are these: the instrument may
dominate, and be played in a rapid, showy manner (Citr); the instrument may be played in a subdued
manner, allowing the voice to dominate (Daki); or the instrument may be balanced with the voice
(Vtti). See the Appendix to this paper for a detailed account of these three.
3In Metatheater and Sanskrit Drama: Second, Revised and Enlarged Edition (Madras:
Tambaram Research Associates, 2005). Distributed by EastWest Books, Madras (Chennai).
4Journal of Oriental Research, Vol. VII, Pt. iii, p. 237.
5Administration and Social Life under the Pallavas, 2nd ed., revised (Madras: University of
Madras, 1977), p. 262.
6Saundarya-Lahar. Translation:

When you, with a movement of your head, say Bravo, Bravo

in appreciation of Sarasvat, playing on her v (songs celebrating)
Many a noble deed of Paupatis, (Sarasvat, finding) the v string badly
articulating the low tone, quietly covers the v with its sheath.
7A variant of the earlier spelling of his name.
8Administration and Social Life . . . , p. 297.
 Siddham | Nama ivya ||
I. Madhyamagrm Catuprahra Svargam
1 sa-ne-pu-sa gi-ne-gi-sa ne-dhu-ne-sa mu-pu-ne-sa mi-ra-gi-sa ru-ge-nu-su sa-gi-ne-sa ne-mu-pe-su
mi-ga-ne-sa pe-mu-ne-sa ra-mi-ga-se dhu-ne-gi-sa ne-pu-ne-sa pi-ma-pi-se ga-dhu-ne-sa mu-ne-pu-? [||]
2 na-pe-ru-ge mi-ga-re-ga ne-sa-ra-gi dhu-ne-ra-gi sa-gi-ne-gu pe-mu-ra-gi mu-pe-ru-ge gi-sa-ra-gi
sa-ne-ra-gi ru-ge-nu-ge pi-ga-re-ga ne-pu-ra-gi su-ge-ra-ge ga-re-mi-ga pi-ne-ra-gi se-ra-mi-ga [||]
3 pu-sa-mu-pe gi-sa-ne-pu ne-sa-ne-pu ma-ni-ma-pi dhu-ne-mu-pe sa-mu-ne-pu ni-ma-ni-pa re-ga-mu-pe
dhi-su-(ne)-pu mi-dhu-ne-pu sa-dhu-ne-pu ne-sa-mu-pe gu-pe-mu-pe sa-gi-ne-pu ne-dhu-ne-pu gi-sa-mu-pe: ||
4 ne-pu-dhu-ne mu-sa-dhu-ne ra-gi-dhu-ne gi-sa-dhu-ne ni-ma-pa-ni ne-sa-dhu-ne mu-ne-dhu-ne sa-mi-dhu-ne
mi-ga-se-nu se-ga-se-nu ga-se-mu-ne pu-sa-gu-ne sa-ne-dhu-ne mu-gi-dhu-ne ni-ma-dhe-na pe-su-gi-ne ||
5 mu-sa-pe-mu gi-ne-sa-mu ne-mi-sa-mi sa-dhu-ne-mu ne-gi-sa-mi mu-pe-sa-mi ra-gi-pe-mu gi-sa-pe-mu
dhe-sa-ne-mu ga-se-pi-ma su-ge-su-ma mi-dhu-ne-mu ra-gi-se-mu ne-sa-ne-mu ni-se-pi-ma ru-me-gu-me ||
Sampt (Sva)(rgam*)

II. ajagrm Catuprahra Svargam

6 sa-gi-dhe-sa su-ru-ge-su mi-ga-dhe-sa dhe-se-pe-su gi-ne-dhu-se pu-sa-dhe-sa pe-na-pe-su sa-gi-pe-su
sa-mu-gi-sa pu-dhu-ne-sa ne-ra-gi-sa dhi-ne-pu-sa ru-ge-dhu-se dhi-ma-ge-su sa-pu-dhe-su ne-pu-gi-sa ||
7 su-ru-ge-ru ge-dhu-su-re gi-dhe-gi-ra ga-se-gi-ra pu-ne-gi-ra dhe-sa-ne-ra dhi-ma-ge-ru ra-dhe-gi-ra
dhe-sa-gi-ra pi-se-gi-ra pu-gi-dhe-ru na-(pe)-gi-ra dhu-ne-sa-ri ne-pu-se-ra dhe-ru-gi-ra gi-ra-dhe-ru ||
8 gu-pe-ru-ge gi-dhe-ra-gi sa-dhe-ra-gi dhe-mu-ra-gi dhu-se-ra-gi sa-pe-ru-ge mi-ra-mi-ga dhe-gi-sa-gi
ru-ge-su-ge pi-se-ra-gi sa-mi-ra-gi (dhe)-gi-ra-gi dhu-se-mi-ga ne-pu-se-ga dhi-pa-su-ge ga-pe-su-ge ||
9 ne-pu-ra-pi pu-dhu-ne-pu gi-dhu-(ne-pu) na-pe-gu-pe mu-dhe-na-pe ra-gi-ne-pu gi-ra-ne-pu ne-sa-ra-pi
pu-sa-ne-pu dhe-na-gu-pe (sa-ra)-gi-pe pi-ra-gu-pe pa-se-ga-pi dhu-se-ga-pi dhe-sa-gu-pe gi-sa-gu-pe ||
10 dhu-ne-pu-dh(e) ga-pi-ga-dhe gi-sa-ne-dhu (ne-pu)-gi-dhe gu-pe-na-dhe pi-se-ga-dhu se-ga-ne-dhu sa-mi-ga-dhe
dhu-su-ge-dhu pe-su-na-(dhe) ra-gi-pu-dhe pi-ra-gi-dhe su-ru-ge-dhu ge-su-ge-dhu sa-gi-pu-dhe mi-ra-(gi-dhe) ||
11 ne-gi-dhu-ne dhu-se-dhu-ne gi-pu-dhu-ne sa-pu-dhu-ne pu-gi-dhe-na dhe-gi-dhe-na gi-dhu-sa-ne ra-(dhe)-sa-ne
pu-dhu-se-nu re-ga-se-nu gi-sa-gi-ne ga-se-pu-ne pu-(nu)-dhu-ne ga-pi-dhe-na pe-sa-gi-ne pi-ga-dhu-ne ||
12 gu-na-dhe-mu sa-gi-(dhe)-mu gi-dhe-sa-mi se-ra-ga-mi ru-me-su-me gi-sa-(dhe)-mu ne-sa-dhe-mu ma-re-(ga)-mi
ge-ru-gu-me ru-gu-dhe-mu ga-dhu-ne-mu mi-ga-dhe-mu pi-re-ga-mi mu-gi-dhe-mu re-ga-dhe-mu sa-pu-dhe-mu ||
Sam(pt)( Svargam*)

245
 III. ab Catuprahra Svar()gam
13 sa-dhu-ne-sa mu-dhu-ne-sa ne-dhu-ra-se mi-ne-dhu-sa ru-u-me-(su) me-su-pa-sa a-mi-pe-su ri-sa-dhu-se
mu-ne-dhu-se ra-dhu-ne-sa dhe-mu-ri-sa ri-dhe-ri-sa dhe-na-ri-se mu-dhe-ra-se su-re-dhu-se mi-ra-dhu-se ||
14 ru-dhe-sa-ri dhe-na-dhe-ru mi-ra-dhe-ru sa-ri-dhe-ru a-mi-dhe-ra dhu-su-me-ru me-su-me-ru ri-sa-dhe-ru
ne-dhu-se-ra mu-dhe-sa-ri sa-mu-(dhe)-ru dhi-e-ma-re dhu-ne-se-ri dhe-ru-mi-ra mi-ra-me-ru ra-mi-se-ra ||
15 me-su-re-dhu su-ru-na-dhe sa-dhu-ne-dhu ri-sa-mu-dhe sa-ne-mu-dhe (ru)-me-na-dhe mi-ra-ne-dhu se-mu-ne-dhu
se-(ra)-mu-dhe na-dhe-mu-dhe u-me-na-dhe mi-(ra)-mu-dhe mu-dhe-na-dhe ri-sa-ne-dhu ne-mu-ne-dhu sa-dhe-mu-dhe ||
16 ne-dhu-ne-mu ri-dhe-na-mu ma-se-a-mi dhu-se-a-mi dhu-ne-u-me (a)-se-a-mi me-ru-u-me dhi-ma-se-mu
pa-dhi-e-ma r(e)-su-e-ma dhu-ne-u-me dhi-su-e-ma pu-dhe-na-mu na-dhe-u-me mi-dhu-se-mu dhe-ru-(u-me) ||
Sam(pt Svargam*)
IV. Sdhrit Catuprahra Svargam
17 sa-pu-ke-sa mu-dhe-pu-sa dhe-pu-ke-sa ri-sa-pe-(su) ka-si-pe-sa su-ru-pe-sa dhe-ru-pe-sa ri-dhe-pu-sa
pu-dhu-ke-sa me-ru-me-su mi-se-ra-se ra-pu-ke-sa me-ra-pe-su mi-ra-ku-se a-mi-ra-se pu-dhe-ra-[se] ||
18 se-pu-dhe-ru sa-me-su-re mi-se-dhu-re pu-dhe-sa-ri u-me-su-re dha-si-dhe-ra dhe-ra-dhe-ru sa-dhe-sa-ri
pu-dhe-mi-ra dhe-pu-dhe-ru mi-pu-dhe-ru sa-dhe-mi-ra mi-ra-su-re su-ru-me-ru dhi-(ma)-dhe-ra dhe-pu-sa-ri ||
19 dhe-sa-pu-dhe pu-ke-sa-dhe sa-pu-sa-dhe ri-sa-pu-dhe su-ru-sa-dhe ri-ke-sa-dhe mi-ra-se-dhu ke-sa-pu-dhe
ra-pu-se-dhu ke-ri-sa-dhe mu-sa-pu-dhe dhi-re-pa-dhi se-ra-pu-dhe ru-dhe-pu-dhe sa-dhe-pu-dhe mi-ra-pu-dhe ||
20 pu-sa-dhe-[p]u dhe-sa-dhe-pu ma-e-dhi-pa ri-sa-dhe-pu sa-dhe-ru-pe ru-pe-ra-pi dhe-ru-dhe-pu ma-e-ma-p(i )
s(a)-pu-dhe-pu dhe-sa-ru-pe u-mu-dhe-pu dhi-pa-re-pu mi-r(e)-dhe-pu ke-sa-dhe-pu a-mi-ru-pe mu-sa-dhe-[pu] [||]
21 pu-sa-dhe-mu mi-ra-se-mu se-mu-a-mi a-mi-se-mu sa-dhe-u-me ri-sa-dhe-mu pu-dhe-(sa)-mi dhe-pu-?-?
?-(dhi)-e-ma su-re-dhi-ma re-pu-dhe-mu mi-(su)-dhe-mu sa-dhu-se-ma pu-ke-se-mi me-su-dhe-mu dhi-?-?-(mu) [||]
V. Pacam Catuprahra Svargam
22 pu-ne-dhu-se ra-pi-ma-se pi-mu-pe-su ne-u-me-su sa-mi-ra-se dhu-ne-mi-sa ne-sa-pe-su ri-sa-me-su
sa-dhe-ri-sa ne-pu-ri-sa pu-se-ra-se dhi-a-mi-se na-ru-pe-su ne-ra-mi-se dhu-ra-pi-se mi-a-me-(su) ||
23 mu-pe-sa-ri ne-mu-pe-ru sa-dhu-ne-ra pi-ma-se-ra ru-u-me-ru ma-dhe-na-ri ne-dhu-ne-ra pi-ra-se-ra
dhe-na-pe-ru ri-sa-me-ru ma-pi-ma-re pu-dhu-ne-ra ra-na-pe-ru dhe-na-(me)-ru pi-a-mi-re sa-u-me-ru (||)
24 pe-mu-ra-mi ru-me-u-me ma-pi-a-mi ne-ra-se-mu pi-ra-pe-mu pi-ma-se-mu na-pe-ru-me ru-dha-u-me
ru-pe-su-me na-pe-u-me mi-se-ra-mi ne-pu-ne-mu se-ra-a-mi ra-se-ra-mi dhu-se-pi-ma pi-ra-se-mu ||
25 dhu-ne-sa-dhe na-dhe-pu-dhe pi-ra-mi-dhu ne-pu-ne-dhu ra-dhu-ne-dhu ne-ri-sa-dhe mu-pe-mu-dhe ri-sa-ri-dhe
a-mi-ra-dhe mu-pu-ne-dha se-ra-mu-dhe pi-sa-ne-dhu se-mu-pu-dhe ru-me-sa-dhe ra-mi-pu-dhe dhi-ne-mu-dhe ||
26 ne-ru-dhe-na u-mu-dhe-na ri-sa-dhu-ne ra-ne-dhu-ne pu-ra-dhu-ne su-ri-dhe-na pe-mu-dhe-na mi-ra-dhu-ne
r(i)-mi-dhe-na ri-pu-dhe-na dhe-ru-dhe-na ru-u-me-na mu-pu-dhe-na dhe-ru-sa-ne dhi-pu-dhu-ne sa-ri-dhe-na ||
27 r(u)-e-ma-pi me-u-mu-pe ri-sa-ne-pu ra-dhu-ne-pu ma-pi-re-pu ne-u-mu-pe ru-pe-u-pe dhi-e-ma-pi
ra-mi-u-pe ru-na-mu-pe mi-a-ma-pi dhe-na-ri-pe ri-pu-ne-pu dhe-ru-na-pe su-pa-ma-pi ra-(ne)-mu-(pe) [||]

246
 VI. Kaiikamadhyam Catuprahra Svargam
28 sa-mu-ke-sa mu-dhe-ka-si dha-me-ke-sa mi-ra-mi-se ri-(sa)-ke-sa su-me-dha-si dhe-mu-ke-sa a-mi-ke-sa
dhu-se-ri-sa dhe-ri-ke-sa ri-sa-re-su sa-(dha)-ke-sa (su)-ru-ke-sa mi-ra-ke-sa ra-mi-ke-sa me-(si)-?-? [||]
29 dhi-ma-se-ra dhe-ri-sa-ri ru-dhe-sa-ri se-ri-su-re sa-ke-sa-ri mu-dhe-mi-ra ke-sa-mi-ra mi-ke-sa-ri
dhu-ke-sa-ri ke-sa-dhe-ru ri-ke-sa-ri dhe-ru-sa-ri sa-e-ma-re ra-mi-dhe-ru ri-dha-me-ru sa-ke-?-? [||]
30 dhu-ke-sa-dhe mi-(sa)-mu-dhe ma-dhe-se-dhu se-ku-se-dhu ke-sa-mu-dhe u-mu-sa-dhe ri-dhe-sa-dhe sa-u-mu-dhe
sa-mu-sa-dhe dhi-e-ma-dhi dhe-sa-mu-dhe su-dhe-sa-dhe dhi-re-ma-dhi dhe-ru-sa-dhe mu-ke-sa-dhe me-dhi-?-? [||]
31 sa-me-ru-me ri-sa-dhe-mu ke-sa-u-me dhu-re-su-me sa-mu-dhe-mu ke-sa-dhe-mu dhi-ma-e-ma dhu-se-dhu-ma
ru-me-dha-me sa-dhe-ru-me ri-ke-sa-mu dhe-ka-si-mu ma-dhi-ru-me dhu-me-ru-me mi-ke-sa-mu dhe-(sa)-?-? [||]

VII. Kaiik Catuprahra Svargam

32 sa-u-me-su ri-ke-ri-sa dhe-mu-dhe-sa mu-ke-ri-sa ri-dhu-ke-sa dhu-ke-ri-sa ke-su-dhu-se a-mi-ke-sa
ru-me-dha-si me-dha-me-su si-dhu-me-su sa-ke-ri-sa ke-sa-ri-sa u-mu-dhe-sa dhi-su-me-su ka-(si)-?-? [||]
33 mu-dhe-mi-ra pi-se-mi-ra me-u-me-ru a-mi-ma-re mi-ra-me-ru sa-ke-mi-ra su-pa-me-ru dhe-pu-sa-ri
u-mu-ke-ra mi-a-dhi-re a-pi-se-ra pe-ru-dhe-ra ke-ra-se-ra mi-se-dhi-ra su-pe-su-[re] ma-pi-?-? [||]
34 (su)-ru-me-u si-dha-me-u me-dha-me-u mi-a-ke-u ma-pi-ma-e dhu-se-mi-(a) dhi-(ma-ke)-u dhe-ru-me-u
ra-se-mi-a ku-se-mi-a mi-a-me-u me-ru-me-u ra-pa-ma-e ru-u-me-u ma-sa-?-? ?-?-?-? [||]
35 se-ku-se-mu ra-pe-u-me (su-re-su)-me pi-e-pi-ma se-ma-pe-(mu) ra-pe-ru-me pi-re-(a)-mi ru-dha-su-me
u-me-ru-me sa-ri-pe-mu dhi-pa-e-ma e-ma-pi-ma u-mu-(dhe-mu) ri-dhe-ru-me a-pi-e-[ma] ?-?-?-? [||]
36 sa-mi-ra-dhe ru-ke-sa-dhe mi-a-mi-(dhi) (pe)-mu-ra-dhe ru-(me)-ra-dhe mi-a-(pa)-dhu (ra-ma)-ri-dhe ra-mi-ra-dhe
ru-u-me-dha si-dhe-su-dhe a-pi-ma-dhi dhe-mu-sa-[dhe] (pu)-e-ma-dhi dhe-pu-ma-dhi ?-?-?-? ?-?-?-? [||]
37 ke-ra-se-(ku) sa-mu-sa-ke su-ru-sa-ke ri-dhu-sa-ke su-ru-(ma)-ku ra-mi-sa-ke ri-ka-sa-ke e-ma-se-ku
ma-dhi-se-ku dhi-ma-se-ku (se)-a-mi-ke dhu-ra-se-ku sa-dhu-(sa-ke) mi-ra-se-[ku] ?-?-?-? ?-?-?-? [||]
38 dhu-(ke)-sa-pu (ke)-sa-mu-pe sa-ri-mu-pe ri-sa-ru-pe mu-(ke)-sa-pu (ke)-sa-u-pe sa-ma-(e)-pu ri-ke-sa-[pu]
sa-dhe-sa-pu mi-dhu-re-pu (ke)-sa-ru-pe mu-pe-ra-pi se-ku-se-pu mi-se-ma-pi ?-?-?-? ?-?-?-? [||]

247
 Minakshis Table III, showing the varjya svaras or rutis in each of the Kuumiymalai Groups
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28.
Madhyama-grma: (+ indicates the presence of svaras of the ruti and 0 indicates the absence of svaras of the ruti)
1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4
sa si su se ra ri ru re ga gi gu ge ma mi mu me pa pi pu pe dha dhi dhu dhe na ni nu ne
+ 0 + + + 0 + + + + + + + + + + 0 + + + 0 0 + 0 + + + +
21 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
[where we differ: + + .]
aja-grma:
sa si su se ra ri ru re ga gi gu ge ma mi mu me pa pi pu pe dha dhi dhu dhe na ni nu ne
+ 0 + + + 0 + + + + + + 0 + + + 0 + + + 0 + + + + 0 + +
21 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
[where we differ: + + + .]
ava:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe na ni nu ne
+ 0 + + + + + + + 0 + + + + + + + 0 + + 0 + + + + 0 0 +
21 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20.
Sdhrita:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe ka ki ku ke
+ + + + + + + + + 0 + + + + + + 0 + + + 0 + + + 0 0 0 +
20 21 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
[where we differ: + + + + .]
Pacama:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe na ni nu ne
+ 0 + + + + + + + 0 + + + + + + 0 + + + 0 + + + + 0 0 +
21 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
[where we differ: + + .]
Kaiika-madhyama:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe ka ki ku ke
+ + + + + + + + + 0 + + + + + + 0 0 0 0 + + + + + 0 + +
20 21 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19.
Kaiika:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe ka ki ku ke
+ + + + + + + + + 0 + + + + + + 0 + + + + + + + + 0 + +
[where we differ: 248 + ]
 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Madhyama-grma: Line Totals
sa si su se ra ri ru re ga gi gu ge ma mi mu me pa pi pu pe dha dhi dhu dhe na ni nu ne
1 + 0 + + + 0 + 0 + + 0 + + + + 0 0 + + + 0 0 + 0 0 0 + + 17
2 + 0 + + + 0 + + + + + + 0 + + 0 0 + + + 0 0 + 0 + 0 + + 19
3 + 0 + 0 0 0 0 + + + + 0 + + + 0 + + + + 0 0 + 0 0 + 0 + 16
4 + 0 + + + 0 0 0 + + + 0 + + + 0 + 0 + + 0 0 + + + + + + 19
5 + 0 + + + 0 + 0 + + + + + + + + 0 + 0 + 0 0 + + 0 + 0 + 19
aja-grma:
sa si su se ra ri ru re ga gi gu ge ma mi mu me pa pi pu pe dha dhi dhu dhe na ni nu ne
6 + 0 + + + 0 + 0 + + 0 + + + + 0 0 0 + + 0 + + + + 0 0 + 18
7 + 0 + + + + + + + + 0 + + 0 0 0 0 + + (+) 0 + + + + 0 0 + 19
8 + 0 + + + 0 + 0 + + + + 0 + + 0 + + + + 0 + + + 0 0 0 + 19
9 + 0 0 + + 0 0 0 + + + 0 0 0 + 0 + + + + 0 0 + + + 0 0 + 15
10 + 0 + + + 0 + 0 + + + + 0 + 0 0 0 + + + 0 0 + + + 0 0 + 17
11 + 0 0 + + 0 0 + + + 0 0 0 0 0 0 0 + + + 0 0 + + + 0 + + 14
12 + 0 + + + 0 + + + + + + + + + + 0 + + 0 0 0 + + + 0 0 + 20
ava:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe na ni nu ne
13 + 0 + + + + + + + 0 + 0 0 + + + + 0 0 + 0 0 + + + 0 0 + 18
14 + 0 + + + + + + + 0 0 + + + + + 0 0 0 0 0 + + + + 0 0 + 18
15 + 0 + + + + + + 0 0 0 0 0 + + + 0 0 0 0 0 0 + + + 0 0 + 14
16 0 0 + + 0 + + + + 0 + + + + + + + 0 + 0 0 + + + + 0 0 + 19
Sdhrita:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe ka ki ku ke
17 + + + + + + + 0 0 0 0 0 0 + + + 0 0 + + 0 0 + + + 0 + + 17
18 + + + + + + + + 0 0 + 0 + + 0 + 0 0 + 0 + + + + 0 0 0 0 17
19 + 0 + + + + + + 0 0 0 0 0 + + 0 + 0 + 0 0 + + + 0 0 0 + 15
20 + 0 0 0 + + + + 0 0 0 + + + + 0 + + + + 0 + 0 + 0 0 0 + 16
21 + 0 + + + + 0 + + 0 + + + + + + 0 + + 0 0 + + + 0 0 0 + 19

249
 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Pacama: Line Totals
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe na ni nu ne
22 + 0 + + + + + 0 + 0 0 0 + + + + 0 + + + 0 + + + + 0 0 + 19
23 + 0 0 + + + + + + 0 + 0 + + + + 0 + + + 0 0 + + + 0 0 + 19
24 0 0 + + + 0 + 0 + 0 + 0 + + + + 0 + + + + 0 + 0 + 0 0 + 17
25 + 0 0 + + + + 0 0 0 0 0 0 + + + 0 + + + + + + + + 0 0 + 17
26 + 0 + 0 + + + 0 0 0 + 0 0 + + + 0 0 + + 0 + + + + 0 0 + 16
27 + 0 + 0 + + + + + 0 + + + + + + + + + + 0 + + + + 0 0 + 22
Kaiika-madhyama:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe ka ki ku ke
28 + + + + + + + + + 0 0 0 0 + + + 0 0 0 0 + 0 + + + 0 0 + 17
29 + 0 + + + + + + 0 0 0 + + + + + 0 0 0 0 + + + + 0 0 0 + 17
30 + 0 + + 0 + + + 0 0 + + + + + + 0 0 0 0 0 + + + 0 0 + + 17
31 + + + + 0 + + + 0 0 + 0 + + + + 0 0 0 0 + + + + + 0 0 + 18
Kaiika:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe ka ki ku ke
32 + + + + 0 + + 0 + 0 + 0 0 + + + 0 0 0 0 + + + + + 0 0 + 17
33 + + + + + + + + + 0 + 0 + + + + + + + + 0 + 0 + 0 0 0 + 21
34 + 0 + + + 0 + 0 + 0 + + + + 0 + + + 0 0 + + + + 0 0 + + 19
35 + 0 + + + + + + 0 0 + + + + + + + + 0 + + + 0 + 0 0 + 0 20
36 + + + 0 + + + 0 + 0 + + + + + + + + + + + + + + 0 0 0 + 22
37 + 0 + + + + + 0 + 0 0 + + + + 0 0 0 0 0 0 + + 0 + 0 + + 16
38 + 0 0 + + + + + 0 0 + 0 + + + 0 + + + + 0 0 + + 0 0 + + 18

250
 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Madhyama-grma:
sa si su se ra ri ru re ga gi gu ge ma mi mu me pa pi pu pe dha dhi dhu dhe na ni nu ne
13 0 2 2 2 0 1 0 3 5 0 1 1 3 4 0 0 2 4 2 0 0 3 0 0 0 1 14
4 0 1 1 10 0 3 3 7 9 1 6 0 3 2 0 0 2 1 3 0 0 1 0 1 0 1 6
8 0 1 0 0 0 0 1 1 3 1 0 3 1 7 0 1 1 9 7 0 0 4 0 0 3 0 12
6 0 1 4 1 0 0 0 3 4 1 0 2 2 4 0 1 0 2 1 0 0 9 1 1 3 2 16
9 0 2 2 2 0 1 0 1 5 1 1 3 5 11 2 0 2 0 4 0 0 2 1 0 1 0 8
aja-grma:
sa si su se ra ri ru re ga gi gu ge ma mi mu me pa pi pu pe dha dhi dhu dhe na ni nu ne
13 0 7 3 1 0 2 0 1 6 0 3 1 1 1 0 0 0 5 4 0 2 3 4 1 0 0 5
3 0 2 3 12 1 6 1 1 11 0 3 1 0 0 0 0 1 3 (1) 0 1 2 7 1 0 0 4
4 0 3 4 8 0 3 0 4 11 1 6 0 4 1 0 1 1 1 3 0 1 2 5 0 0 0 1
5 0 0 2 6 0 0 0 2 5 5 0 0 0 1 0 2 5 7 8 0 0 3 3 3 0 0 7
3 0 4 2 3 0 1 0 5 6 1 4 0 2 0 0 0 3 4 2 0 0 8 10 2 0 0 4
5 0 0 4 1 0 0 1 4 8 0 0 0 0 0 0 0 2 6 1 0 0 9 5 3 0 3 12
5 0 1 1 1 0 3 3 6 4 3 1 1 5 11 3 0 1 1 0 0 0 1 10 1 0 0 2
ava:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe na ni nu ne
9 0 4 7 4 5 1 1 1 0 1 0 0 3 4 2 1 0 0 1 0 0 9 4 1 0 0 6
5 0 2 3 7 5 10 1 1 0 0 1 1 5 2 4 0 0 0 0 0 1 3 10 1 0 0 2
5 0 2 2 3 2 2 1 0 0 0 0 0 2 9 3 0 0 0 0 0 0 7 13 5 0 0 7
0 0 2 5 0 1 2 1 3 0 5 3 5 4 5 6 1 0 1 0 0 3 5 4 3 0 0 4
Sdhrita:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe ka ki ku ke
11 1 4 5 6 2 3 0 0 0 0 0 0 3 1 3 0 0 7 5 0 0 1 5 1 0 1 4
6 1 4 2 6 3 6 4 0 0 1 0 1 5 0 3 0 0 6 0 1 1 1 12 0 0 0 0
10 0 1 3 4 3 2 1 0 0 0 0 0 2 1 0 2 0 11 0 0 1 2 17 0 0 0 4
8 0 0 0 1 1 5 2 0 0 0 2 3 2 3 0 2 2 11 4 0 2 0 13 0 0 0 1
5 0 3 5 1 1 0 2 1 0 1 1 3 6 9 2 0 1 5 0 0 2 1 8 0 0 0 1

251
 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Pacama:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe na ni nu ne
7 0 6 8 5 3 1 0 2 0 0 0 1 5 1 3 0 3 3 3 0 1 3 1 2 0 0 5
4 0 0 3 7 3 8 2 1 0 2 0 5 1 1 4 0 4 1 3 0 0 3 3 4 0 0 5
0 0 1 7 8 0 4 0 2 0 3 0 3 6 6 6 0 5 1 5 1 0 1 0 2 0 0 3
5 0 0 2 5 3 1 0 0 0 0 0 0 3 6 1 0 2 5 1 1 1 6 12 1 0 0 8
3 0 1 0 4 4 4 0 0 0 1 0 0 2 3 1 0 0 4 1 0 1 5 11 10 0 0 8
1 0 1 0 3 3 4 1 1 0 4 2 5 2 4 1 1 5 5 9 0 1 1 2 3 0 0 5
Kaiika-madhyama:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe ka ki ku ke
15 3 3 2 3 4 1 1 1 0 0 0 0 5 3 3 0 0 0 0 3 0 1 3 1 0 0 10
12 0 1 2 4 11 5 2 0 0 0 1 2 4 1 1 0 0 0 0 1 1 1 6 0 0 0 7
12 0 1 3 0 1 1 1 0 0 2 2 3 1 7 1 0 0 0 0 0 5 3 16 0 0 1 3
9 1 1 1 0 2 5 1 0 0 1 0 4 1 7 10 0 0 0 0 1 2 4 6 1 0 0 4
Kaiika:
sa si su se ra ri ru re a i u e ma mi mu me pa pi pu pe dha dhi dhu dhe ka ki ku ke
12 3 6 1 0 7 1 0 1 0 1 0 0 1 3 6 0 0 0 0 2 1 4 3 1 0 0 8
2 1 3 4 11 1 4 2 1 0 1 0 2 7 2 4 1 3 1 2 0 2 0 3 0 0 0 3
1 0 1 3 2 0 4 0 4 0 10 2 5 5 0 9 1 1 0 0 2 1 1 1 0 0 1 2
1 0 3 3 2 2 4 2 0 0 1 3 5 1 5 7 1 5 0 4 1 1 0 2 0 0 1 0
3 1 1 0 6 1 3 0 2 0 1 1 4 5 2 2 1 1 2 1 2 4 1 10 0 0 0 1
8 0 2 7 4 2 2 0 1 0 0 1 4 3 1 0 0 0 0 0 0 2 3 0 2 0 6 8
11 0 0 3 1 3 2 1 0 0 1 0 2 2 4 0 1 2 6 6 0 0 2 1 0 0 1 6

252
 Appendix
The Terms Vtti, Daki, and Citr in the Nya-stra
Tisras-tu vttaya-citr-daki-vtti-sajit |
three indeed [are] vttis Citr Daki Vtti designated
Vdya- gtbhaya-gu nirdis-t yathkramam ||71||*
instrument voice=both lead shown [are] they in-order

Ukt vtti nirpayati | Tisras-tu vttaya iti | Vttir-gua-pradhna-bhvtm

the-said vtti defines-he three indeed vttis thus vtti role- foremost consists-of
vyavahra iti smnya-lakaam | Tata- vdya-prdhny gta-guatti
performing thus general definition stringed-instrument playing predominant-when voice subdued=thus
citr vtti | Gta- mukhpkat-virahita hi vdya yath-vidhi vaicitrya-carcita kriyat |
Citr Vtti voice turning-to=regard devoid-of indeed instrument accd.to improvis. variations repetition played-is

tad- viparyayas-tu daki vtti | Gta hi viia[] sthnya pradhna tad-anukla-

this-(of) converse indeed Daki Vtti voice indeed special place important that accompanying

cya vyavahra | Yatra samam-va dvayr- vyavahras-tatra smyna vartand-vttir-nma |

and=this performing where equal indeed tween-the-two performing there balance-with being Vtti- called

Vtti-sajita-abdasyyam-aya | Vttir-iti smny vi ca prayukttra | . . .

Vtti- designated words=this intended-meaning Vtti thus general-sense-(in its) special-sense & employed=here

Gua- abdtrtkara-vc na tv- apradhna- vc tna vdyasya gtasya ca gua

gua the-word=here-lead meaning not-indeed secondariness meaning in-this-way instruments voices & gua

utkar yatra prdhnyam-iti yvat | . . .

lead where importance thus
*********
There are three Vttis, called Citr, Daki, and Vtti |
They are listed, in order, as (possessing) Vdya-, Gta-, and both [Vdya-Gta-] Guas ||71||
He [Bharata] defines the (above) mentioned (general) concept Vtti, as well as the three (particular)
Vttis. Thus, the general definition of the term Vtti is the performing (vyavahra) of the foremost
role (gua-pradhna) [by either the instrument or voice].
When the stringed instrument predominates and the voice is subdued, then Citr-Vtti (is manifested),
and the instrument, turning away from (accompanying) the voice, plays a series of improvised variations
[on themes of the Rga].
The converse of this (Citr-Vtti) is Daki-Vtti, where the voice, occupying the dominant position,
becomes important, and the instrument accompanies it.
When there is a balance between the two (instrument and voice) in their performing, then that is called
Vtti-Vtti. This is the meaning of Vtti as given in its technical definition.
In this way, Vtti, (both) in its general usage and in its special (technical) usage, is employed here
[by Bharata]. . . .
The word, Gua, here, means Lead Role (and) does not mean Subordinate-ness or Secondariness;
and, in this way, it [Gua] (refers to:) (i) the taking of the lead role by the instrument, (ii) by the voice, [and
(iii) where both are equal].
_______________
*Nyastra of Bharata Muni, with the commentary, Abhinavabhrat, by Abhinavaguptchrya
(Delhi: Parimal Publications, 1984), Vol. IV, Chp. 29, p. 98.

253
 TWENTY TWO
A Note on the Rape of the Liga by Lord Hobart
The monolithic shrine which is today called the Gaa Ratha was actually called the Arjuna
Ratha prior to the nineteeth century. This temple was originally dedicated to iva, as evidenced by the
long Sanskrit inscription on its maapa wall. And there was once a liga in the sanctum sanctorum.
J. Goldingham, in his article, Some account of the Sculptures at Mahbalipuram, published in the
Asiatic Researches, Vol. V, 1798, observed that the shrine had a liga within:
[Ones] attention . . . is first arrested by a Hindu pagoda, covered with sculpture, and hewn from a
single mass of rock; being about twenty-six feet in height, nearly as long, and about half as broad.
Within is the lingam, and a long inscription on the wall, in characters unknown.1
Five years later, in 1803, Kvali Lakshmayya, the Brahmin scholar-assistant of Colonel Macken-
zie, wrote:
On the South side of [the Butterball] is Arjunas Ratha [todays Gaa Ratha], cut out of a
single stone. The stone has been cut into the shape of a Mantapam with two pillars and a
Garbhagriha, in which was placed Siva [liga]. When that Linga was carried off by Bu ** [sic],
the people of this place took an image of Vinyaka [Gaa] which was near and put it in the
Garbhagriha.2
In the book, The Seven Pagodas on the Coromandel Coast (1869), edited by Captain M.W.
Carr, B.J. Babington, in his article, An account of the Sculptures and Inscriptions at Mahmalaipr,
referring to the monolithic temple near the Great Penance Panel, speaks of this small monolithic
pagoda, now dedicated to Ganesa, and situated on the north side of the hill. Babingtons footnote ()
here reads as follows:
When Mr. Goldingham wrote his account, this pagoda contained a lingam, so that it has passed
from the Saivas into the possession of the Vaishnavas since that period. . . .3
In an editorial footnote on the above footnote, Captain Carr, in 1869, disputes the details of
the transition: it was not the Vaishava Brahmins who introduced the image of Gaa, but the
villagers!
An old Vaishnava Brahman tells a different tale: he states that the Lingam was taken away, (with
an image of Hanumn,) by Lord Hobart (?) and sent to England, Lady H. giving 20 pagodas to the
villagers as a consideration! Lord Hobart was Governor of Madras from Sept. 1794 to February
1798. Mr. Goldinghams account was published in the latter year.4
Here we learn that it was Lord Hobart who removed the ligam from this Ratha shortly
before his departure in 1798. In any case, he must have removed it sometime during his governership
(Sep. 1794 to Feb. 1798).
_______________
1Reprinted in The Seven Pagodas on the Coromandel Coast (Madras, 1869), which, in turn,
was reprinted by Asian Educational Services (New Delhi, 1984), p. 30. Of course, we do not know
how long it was before the original publication of this article in 1798 that Goldingham visited
Mahabalipuram and made his observations.
2Description of the Pagodas, &c., at Mvalivaram, written in the Telugu language by Kvali
Lakshmayya in 1803 [with a translation], The Seven Pagodas . . ., pp. 200-201. [The system of translit-
eration applied to Indian terms and names in this 19th century publication is responsible for what other-
wise might appear as an accidental confusion of typographical fonts!]
3The Seven Pagodas . . ., p. 56; Babingtons article was first published, in 1830, in the Transac-
tions of the Royal Asiatic Society (Vol. II).
4The Seven Pagodas . . ., pp. 56-57.
 TWENTY THREE

The Brhm Script and Phonetics: An Isometric Analysis of Vowels*

More than one scholar has suggested that the Brhm alphabet was devised by a grammarian.1
We propose to detail briefly the isomorphism which exists between the shapes of vowels of the Brhm
script, on the one hand, and the phonetic analysis (Pinis) of these same vowels, on the other.
In the Gua sandhi () and (Dx) are used as substitutes for i (V) and u () respectively.
In all guating processes, a (D) remains unchanged [or, as it is sometimes said, a (D) is its own Gua].
Both a (D) and (D) remain unchanged in the Gua and Vddhi sandhis. Thus, it can safely be said that
and are the corresponding Gua vowels to i and u, ai (x) and au (D}) being the corresponding
Vddhi vowels. In the Prtikhys, and are called Sandhyakars. But still the authors of the
Prtikhys give rules respecting their pronunciation in a manner implying them virtually to be unitary
sounds. From these, the heavier ai and au were distinguished by the length (indicating growth/increment
= Vddhi) with which they are invested. By the time of Pini, and and ai and au came to be treated
as purely unitary sounds and, accordingly, Pini treats them, on the basis of tradition, as Gua and
Vddhi Sajs and uses them as substitutes for i and u and and in the Gua and Vddhi sandhis.
The inventors of the Brhm script must have had in mind both the Prtikhys and Pinis
dhyy, especially, in their treatment of and and ai and au because they seem to take into consid-
eration the Gua of i () and the Gua of u (), and not the Gua of a () because and share the
characteristics both of i and a, and of u and a, respectively; whereas the Gua of a (that is, a) does not.
While interpreting Pinis stra d-gua, the commentator Bhaji Dikita says, Avarad-aci par

Basic Gua Vddhi

i ai

u au
2
prva-parayr-k guda syt. And he means by the expression prva-parayr-k guada:
one gua is substituted for the final vowel of the preceding word and the initial vowel of the following
word, and that one gua substitute must necessarily share the characteristics of both the preceding vowel
and the following vowel. And, similarly, for any further increase of and , but also take into considera-
tion Pinis idea of sajs (three strokes have been used to denote Gua, and the additional stroke
found in ai and au to denote Vddhi):
(1) In the formation of vowels, the inventors seem to have taken the short a, i, and u as the basic
characters, whereas in the formation of the and and the ai and au, they seem to have followed
Pinis definition of the Gua and Vddhi concepts as well as Pinis rules regarding the Gua and
Vddhi sandhis. Pinis rules, in this context, are purely phonological in character, not grammatical.
_______________
*Based on a paper read at the XV Annual Congress of the Epigraphical Society of India, Nov.
18th to 20th, 1988, Anantapur, Andhra Pradesh, and published in the Journal of the Epigraphical Society
of India, Vol. 17 (1991), pp. 117-118.
 256

(2) According to Pini, a, , and are called Gua letters, whereas , ai, and au are called
Vddhi letters. The Gua and Vddhi letters are used as substitutes whenever Gua and Vddhi sandhis
take place.
(3) The fourteen Mhvara stras given at the beginning of Pinis work establish the inde-
pendent origin of the Gua as well as the Vddhi letters. Pini never says that the Gua letters are
developed out of a, i and that the Vddhi letters, out of , , and . The terms Gua and Vddhi are
used as Sajs (the proper, as well as technical, names of and , and of ai and au, respectively). In
the rules regarding sandhis, Pini uses only these technical names and not the specific letters (e.g., d-
gua, 6:1:87, and Vddhi ci, 1:1:1, where gua means and , and vddhi means ai and au).
(4) Did the devisors of the Brhm script follow Pinis rules? Certainly they did. Fundamental
to the whole sandhi concept, there is a stra in Pini: Sthnntaratama, 1:1:50. While discussing
sandhis, Pini uses only common terms like ik (i, u, , ), ya (y, v, r, l), Gua (a, , ), Vddhi (, ai,
au), etc. Thus, Pini simply says, d-gua, which means: If the vowel is followed by a simple
vowel, the Gua (a, , or ) is the single substitute for the final a or of the preceding word and the
simple vowel of the succeeding one. In this context, it is absolutely necessary to keep in mind the
fundamental stra, Sthnntaratama. Even if there is the chance of all the letters included in the
technical term being comprehended on substitution, still the likeliest of its significates to that in the place
of which it comes, should be accepted as the actual substitute. Thus, in the case of the Gua sandhi,
when we propose sandhi, for example, between Upa and Indra (Upndra), we first of all get a
Gua letters. The sandhi in this particular case has to take place between a and i. Here, according to
Pini, we must have regard to the proximity of the organ of utterance. A is guttural and i is palatal.
As is both guttural and palatal, that letter is the actual substitute here. Similarly, in the place of a and u,
is the actual substitute. In the case of Vddhi sandhi, in the place of a and , ai should be the actual
substitute, and in the place of a and , au should be the actual substitute.
It should be borne in mind that Pini uses the word substitute (da) and never says that a
and i, a and u, a and , a and become respectively , , ai, and au. Hence, any attempt to see a
combination of the three basic vowels in the Gua and Vddhi letters will be mistaken. We note that,
following Pinis analysis, the devisors of the Brhm script used special symbols to denote Gua and
Vddhi vowels. For Gua, they have used three strokes, and for Vddhi, they have used four.
It is interesting to note that bears closer similarity to i, to u, ai to , and au to . This is,
perhaps, due to the fact that in many cases, in Pinis work, the substitutes are used in the place of the
following vowels, like i, u, , , without taking into consideration the preceding a or .

i u

or:

ai au au
258

Fig. 1. Mahishsuramardin panel, c. 7th cent. A.D., Mmallapuram

Fig. 2. The battle of the Amazons, c. A.D. 180, Amazonmachy sarcophagus

at Tel Mevorakh, near Caesaria in Israel.
 TWENTY FOUR

Comparison of an Amazonmachy with the

Mahishamardin Panel*
The recent discovery of an Amazonmachy sarcophagus (Fig. 2) at Tel
Mevorakh, near Caesaria in Israel,1 is important, as it contributes substantially
to the body of knowledge regarding the flow of currents of culture between
Greece and India in ancient times. While interchange between Greco-Roman
and North Indian and Deccan art forms has been studied in detail, less attention
has been directed towards the fecund and critical exchanges resulting as a con-
sequence of the direct Arabian Sea route. When observed in relation to the
Durg Mahishssuramardin panel (Fig. 1) at Mmallapuram, the sarcophagus,
carved in c. A.D. 180, silently speaks volumes about the interconnections.
This combined evidence of intercourse between the Indian and the
western world2 permits us to formulate a theory that the Pallava sculptor in the
seventh century had at his very fingertips an intimate familiarity with western
art forms. Thus the relationship of the two works of art becomes entirely plau-
sible. It is suggested here that, either by direct or indirect contact, the Durg
panel relied partially on Greek art for elements of the configuration and the
postures of individual figures and for some of the themes and motifs, even
while the particular style characteristics cannot but reflect the unique qualities
of each of the two great civilizations.
The Tel Mevorakh sarcophagus and the Durg panel can be compared
in regard to:
(1) the similarities of the configurations and certain postures of individual
figures;
(2) the analogous, underlying geometrical diagrams;
(3) the subject matter: the battle scene;
(4) the implicit symbolic references to the heroic element in each of the
cultures.
(1) The most salient of the influences is exhibited in the analogous configura-
tions. In both, a female warrior, in symbiotic connection with her mount, sits
straight, wields a long spear and assumes an assertive attitude. She is shown to
be in battle with a male opponent of approximate weight who stands in the
_______________
*Reprinted with permission. This article, by Carmel Berkson, was
first published in Lalit Kal, Number 23 (1988). Its full title, there, was Com-
parison of the Recently Discovered Amazonmachy with the Mahishsuramardin
Panel at Mmallapuram. The two photographs are by Berkson. Her study
remains the groundbreaking investigation of the influence of Greco-Roman art
on the Pallavas. In the much more detailed form of a monograph, her thesis
appeared under the title, The Amazon and the Goddess: Cognates of Artistic
Form (Bombay: Somaiya Publications Pvt. Ltd., 1987), and is to be re-issued,
soon, in a revised edition. (ML)
 260 archaic position. On the ground, between the two, a warrior falls or
Pallava Art has fallen. Space separates the two opponents. A very long Greek
tradition anteceded this unique combination of figures which form an
original triad. With infinite variations, the individual elements and the
triple configuration appear on vase paintings, friezes and sarcophagi
first in Assyrian art and then in Greek. On the other hand, in India,
Durg and the buffalo had, in the previous centuries, been depicted in
an entirely different formulation, as goddess and buffalo were united in
a single volume, with the buffalo, in animal form, either spread across
the front of Durgs body or standing in front of her on the ground.
The separation into two distinct and isolated entities, divided by ample
space and more or less equally weighted in the Durg panel appears in
India for the first time here. It is safe to conclude that because of the
complexity of these relationships, the composition could not have
arisen spontaneously or indigenously in the mind of a single artist or
group of artists in India, since in Greece, a millenium of experience had
preceded the fixed relationship of the Amazon on her vehicle, the male
opponent and the fallen warrior. It is a momentous achievement in the
history of art, and most likely the artists at Mmallapuram borrowed the
triad from the west.
Other motifs are the largely identical, accompanying fallen
warriors (they appear earlier in Indian art, but rarely) and the position-
ing of a truncated figure the horse in the sarcophagus and the warrior
on the ground in the Durg panel. An illusion of depth is sought by
positioning each in a perpendicular relationship to the background.
The warrior who lies with his head downward and serves as the center
is also a Greek theme. Indian artists may have gained experience from
working somewhere in the west, or the Yvana artists, settled in India,
may well have contributed their experience to their Mmallapuram
counterparts. Since the icon is not sacred until duly consecrated,
foreigners might have been permitted to work even directly on the
panel.
(2) While styles evolve each out of the particular tradition of the past,
underlying both panels are identical geometrical diagrams which the
artist engraves upon the surface of the stone prior to initiation of the
work. Both the sarcophagus and the Durg panel carry within their
internal structures inherent diagonal and circular arrangements. How-
ever, in this regard, it is not possible to determine how these methods
came to be employed by the two cultures.
(3) The goddess riding on her mount as warrior in an aggressive stance,
in the midst of battle, is portrayed at Mmallapuram for the first time in
India, although the theme of goddess on lion had earlier been trans-
ported and adapted for the Brahmanical icon on coin and statue in the
northwest. Prior to this in Indian sculpure references to battle were
limited to processional scenes or to an occasional, single, or two male
foot soldiers or to several cavalry members. Even though Purnic
depictions teem with graphic descriptions of savage battles, the field
itself, where the clash of battle is literally portrayed, had not been
conceived as a theme for carved panels in India. This implies that the
the Pallavas were seeking an entirely changed frame of reference, and 261
most likely, they turned to the west and adopted the triad as a reductive Amazonmachy
symbol for the whole field of battle as an apt form by means of which
to express a growing concentration on the implications of the king, with
his sacred power, as hero, and Durgs role in regard to this.
(4) It has been observed that the Tamil king embodied in his authority
the sacred power which found ultimate expression on the field of battle.
Conquest in war is followed by prosperity, while defeat of the king is
disastrous for the entire community.3 Also, Durg or Koavai, as
goddess of the battlefield, rides at the head of her troops, and her
powers determine the outcome of the struggle. It follows that to carve a
monumental bas-relief, concentrating on the most dangerous instant in
the battle, is an attempt to externalize fear, to maintain control and to
influence the results; the execution of the panel is in itself deemed to be
apotropaic. It would thus appear that the Pallava artists were attracted
to the Greek Amazonmachies because the motif and the configuration
best symbolized their most potent beliefs. The panel in the shrine thus
serves a dual purpose. The goddess requires sacrifices and propitiation
as the sacred deity. On the field of battle she also will assure the health
of the kingdom.
To summarize: In order to best portray their ideological focus
on the battlefield, as scene of heroic exploits effecting creation and the
well-being of the community, the Pallavas turned to the long-term
western experience which had developed out of the Greek heroic
tradition. Primary influences seem to have travelled with the trade
across the Arabian Sea.
_______________
1
It is now in the Rockefeller Museum in Jerusalem.
2
Surrounded on east, west and south by oceans, the subcontinent of
India is favorably suited for outward expansion of an active commercial life
which radiates in many directions. We learn from Roman, Christian and
Chinese treatises, from the Tamil Sangam literature and from archaeological
evidence that, in the first centuries A.D., both India and r Lak were very
much part of the known world of geographers, traders, ambassadors, Christian
missionaries and piratical adventurers, and that Yvana (Greek) craftsmen
settled in South India and were employed by local rulers. Images for protection
and painted vases for storage and sale came along with the trade, and Yvana
settlers built in their own styles. Since until the seventh century in South India
images were constructed in perishable materials, it is not possible to know
precisely when the intermingling of forms occurred; nevertheless, the Chinese
Buddhist Fa-Hien and St. Jerome in the fifth century and Procopius and Cosmas
Indicopleustes in the sixth, graphically described the ongoing mercantile
expeditions, as the passion for Indian spices, silks and other goods did not
diminish in these later centuries. Alaric, for example, in 408 A.D. postponed
conquest of Rome for ten years in exchange for 3,000 pounds of Indian pepper.
After the destruction of Rome, Byzantium developed a vast commercial
network, and trade with India played a crucial role in the empire.
3
George Hart, III, The Poems of Ancient Tamil, Berkely, 1975.
 TWENTY FIVE
The Shore Temple Capital Inscription1
Excavations carried out by the Archological Survey of India, in the sands immediately south of the Shore
Temple, Mmallapuram, have unearthed remaining parts of what seems to have been a rather large pillar. (The shaft,
itself, unfortunately, is missing.) Spokespersons of the A.S.I. have suggested that it should have been a jaya stambha,
victory pillar. The presumed capital of this massive pillar is lying on the ground at the edge of the pond which has
recently formed in the cavity of the excavations. Encircling this capital are four inscribed titles, three of which are
well-known birudas of the Pallava king, Narasihavarm-II, and the fourth, a title of his, not previously found in his
inscriptions:
r Rjasiha || r Katrasiha || r Narndrasiha || r Mahvara cmai ||
These four titles may be translated as The illustrious Lion among Kings, The illustrious Lion among Warriors,
The illustrious Lion among Rulers of Men, and The illustrious (One whose) Crest-jewel (is) Mahvara.
Of the three well-known titles of King Narasihavarm-II, Rjasiha was so widely used that he is
actually better known today by this biruda than by his abhika-nma. The title Katrasiha appears in his
Vyalr inscription and is also incorporated in the inscribed name of his east-facing shrine of the Shore Temple.
The title Narndrasiha is found in his major Shore Temple inscription as well as in his Vyalr inscription.
There is, however, no previously known appearance of Mahvara-cmai among the lists of his
many birudas, though the king has several titles incorporating the element, clmai or cmai or synonym:
Katraclmai (I:15-2)2 [cmai (III:15-2), Kailsantha Temple, Kanchipuram]
Narndraclmai (I:13-2) [cmai (III:13-2), ditto]
ivacmai or cmai (Shore Temple; Kailsas vimna, Kanchi; Vyalr inscriptions)
Mahvara-ikhmai-dptamauli (Vyalr inscription)
In the recently discovered Capital Inscription, there is a gap between the Mahvara and the cmai
which is a puzzle. However, it doesnt make sense to read the separated parts as two different birudas since there is
no visarga and no daa after mahvara, and, in the title lists of King Rjasiha, clmai/cmai always
appears together with some other element. As there is space enough in the gap for two syllables, we suggest some
such missing letters as -candra- should have been intended: Mahvara-candra-cmai quite appropriate for a
king belonging to the lunar dynasty.
r precedes each of the four titles engraved on the capital, just as it precedes the more than 250 different
titles of his engraved in the Kailsantha Temple, Kanchi. But there is an interesting twist in this matter which must
be noted here.
The earliest editor of the Kailsantha titles (in 1890), E. Hultzsch, observed that there were four tiers of
titles. He conjectured that the titles on the third tier were the first to be inscribed and that they were engraved at the
time of the building of the temple by Rjasiha. The first and second tiers, he thought, were later copies which
were executed by some descendants of Rjasiha.3 He goes on to say that the fourth tier is written in a peculiar
ornamental alphabet, which is based on an alphabet of the same type, as that of the first tier; . . . perhaps the first and
fourth tiers were contemporaneous.4 So, according to the first editor of these inscriptions, the four tiers of titles at
the Kailsantha Temple were inscribed in the reigns of three successive Pallava kings: the third tier, first, in
Rjasihas reign; then the first and fourth tiers, in the reign of a descendent of his; and, finally, the second tier,
during the reign of some even later descendent.
In the decades since Hultzschs pioneering work, scholars have realized that the titles engraved on the first
and fourth tiers are written in the contemporary, northern, Ngar script of the early eighth century, A.D.: those of the
first tier, in a plain style; and those of the fourth, in a pronounced floriated style, which even incorporates the long,
graceful neck and head of birds to represent the medial i and . It was observed, further, that the titles of the
second and third tiers are written in the southern, Pallava Grantha script: those of the second tier, in a plain style; and
those of the third, in a slightly more ornamental script. And, over the years, scholars have come around to the view
that all four tiers of titles were inscribed during the reign of King Rjasiha, himself. In this paper, we would like to
go one step further and suggest that all four tiers were the handiwork of one outstanding royal scribe and that it is
the Capital Inscription at the Shore Temple, Mmallapuram, which holds the key to this insight.
 264
The surprising thing about Rjasihas Capital Inscription is that while its four titles are written in the
southern, ornamental Pallava Grantha script, the r which precedes each of these four titles is written in the northern,
floriated Ngar script! This would seem to us the playful mixing of styles by a scribe adept at both. The implication
of all this is that the hundreds of different titles in the Kailsantha Temple, inscribed in four different forms at four
different levels, may represent an artistic tour de force of the highest order, designed by one single scribe! These
titles deserve more attention from the epigraphical clan than they have heretofore received. Have these titles ever
been honored by having their facsimiles published?
One lesson which we can learn from the hand of this extraordinary scribe is that, in the chief cultural center
of eighth century South India, there was not a trace of the linguistic chauvinism which plagues us today. Southern
and northern scripts were both treated with respect.
A concluding guess: since we have argued (in chapters four and seven) that Rjasihas father, King
Paramvara, was responsible for the Atriraacavara Cave-Temple inscription at Saluvankuppam, near
Mmallapuram, which appears in two versions, one in the southern, Pallava Grantha script, and the other in the
northern, Ngar script,5 this scriptal cosmopolitanism could be due to King Paramvaras having taken as his chief
queen a princess from the northern Ngar Nu! Such an event would then pleasantly account for her sons having
honored her, the dowager Queen Mother, when he, as the reigning sovereign, built his great temple in Kanchipuram
and had his numerous royal titles engraved first and foremost in the Ngar script on the first tier of the dozens of
small shrines surrounding the main vimna.6
_______________
1Paperby Michael Lockwood and A. Vishnu Bhat read at a meeting of the XXIV Annual Congress of the
Epigraphical Society of India, Trichur, Kerala, May 16, 1998.
2The three numbers in parentheses indicate: 1) on which tier (roman numeral), 2) on which shrine (using
the A.S.I.s engraved number), and 3) in which ordinal position on the front of each shrine the given title is to be
found.
3South-Indian Inscriptions, Vol. I, p. 10.

4Ibid.

5The Pattadakal pillar inscription of the Chlukyan king, Krtivarm-II (c. 757-57 A.D.) copies this practice

initiated by Paramvara-I (see Epigraphia Indica, III, pp. 4-6).

6Rjasihas sense of filial devotion to his mother is beautifully evidenced in the poetic inscription on the

third shrine to the right of the entrance to the Kailsantha Temple complex an inscription erroneously mixed up by
Hultzsch and all later scholars with the so-called Queen Ragapatk inscription, which actually is to be found on
the fifth shrine to the right! The inscription on the third shrine (we quote from chapter thirteen) which essentially
praises Rjasihas mother, reads:
Namaivya [||*]
(Verse 1) Bharttu purnmathana-da-dhanurbbalasya aildhirja-tanayva va-dhvajasya [|*]
Y klakla iti viruta-puya-krtt knt nitnta-dayit paramvarasya ||
(Verse 2) Dv jagad-valaya-rakaa-baddha-dk nirbbhinna-atru-hday narasihaviau [|*]
Vllabhyam-rjjitam-avpya virjat y nirjjitya-garvvam-iva pukaradvaty ||
Which we have translated:
Salutations to iva!
(Verse 1) (Her) husbands [i.e., King Paramvaras] well-merited fame being widespread as Klakla
on account of his bows power (having been made) manifest in the destruction of cities, (thus) like
the Daughter of the Great King of Mountains, (she,) the dearly beloved wife of Paramvara, the
Bull-bannered One,
(Verse 2) attaining supremacy [as Queen Mother], shines with surpassing splendor, subduing, as it were,
the pride of Pukaradvat [Lakm, here, also, Rjasihas chief queen], while god-like
Narasihaviu [i.e., King Narasiha (Rjasiha)], true to his sacred vow, is protecting the encircling
world, tearing out the hearts of his enemies.
 265

The Shore Temple Victory Pillar Capital

r Rjasiha || r Katrasiha ||

r Narndrasiha ||

r Mahvara . . . . . . cmai ||
 Photos of a few of Mahndras titles on pillars of his Tiruchi cave-temple:

Avanibhjana

Kaunt[]rambu
Kaumpu

Abhimukha
Akarua

Emuku
266 thi
 TWENTY SIX

Additional Facsimiles of Pallava Inscriptions

Some of the delineations of facsimiles of the Pallava inscriptions reproduced in the following
pages are based on estampages printed in publications of the Archaeologica Survey of India. The rest
are based on photographs by the author. All of the following renditions are delineations by the author.
*********
The Vallam inscriptions of Skantasa, son of King Mahndras feuditory, Rja Vasantapriya:

Northern pillar faade: 1 cTMmLl bpr

catturu-m-malla kuapara
2 mEyNrPEpaTtErc ya
mayntira-p-pttarcaru aiy
3 vyNtP Erc mk kNtEc
vayantappiri arcaru maka kantaca-
4 ecTt EtvlM
ceyivitta tva-kulam

Southern pillar faade: pkaP ltar

pakppiuku laitkura
(delineations by Lockwood from his photos)
268

King Mahndras Cave-Temples Foundation Inscription at yamagalam

1 Lalitkura rjva- by King Lalitkura

2 nibhjana-pallavvaran-nma [|*] this (temple) named Avanibhjana-Pallavvara
3 Kritam-etat-svdh1-karaa- was made at his (wish)
4 m-iva puya-ratnnm [||*] like a reliquary for jewels of merit
_______________
1
Hultzsch, considering this word to be a scribal error, corrected it to svccha
(his own wish Epigraphia Indica, VI, pp. 319-322, with plate).
 269

King Mahndras Cave-Temples Foundation Inscription at Mahndravi

1 Mahitatama satm=upa-mahndra-takam=idam1 Next to the splendid, expansive Upa-Mahndra Reservoir,

2 sthiram=uru krita guabhara vidryya ilam1 [|*] (King) Guabhara caused to be excavated out of hard rock
3 Jana-nayanbhir[]ma-gua-dhma mahndra-pur An eye-pleasing, fine temple for the people of Mahndrapuram,
4 mahati mahndra-viu-gha-nma murri-gha[m ||*] an abode for Murri, this temple called Mahndra-Viu-Gham.
_______________
1
The final m in each of the first two lines, because of space limitation, has been dropped down
(by the scribe) just above the final letter of the next line. See E. Hultschs article, Mahndravi
Inscription of Guabhara, Epigraphia Indica, IV, pp. 152-153.
270

King Mahndras Cave-Temples Foundation Inscription at Dalavr

1 Danata narndra by scepter (order) bowed to by princes of men,

2 narndraia krita [|*] by this king of men was made,
3 atrumallna ailsmin by atrumalla, on this hill,
4 atrumallvarlaya [||*] the abode of atrumallas Lord (iva)
_______________

First edited by E. Hultzsch in Two Cave-Inscriptions, Epigraphia Indica, XII
(1913-1914), pp. 225-226, with plate.
 271

King Mahndras Cave-Temples Foundation Inscription at Maagapau

1 tad=aniakam=adrumam=al- Without brick, timber, metal, or

2 ham=asudha Vicitracittna [|*] mortar, by King Vicitracitta was
3 Nirmmpitan-npa brahm- this distinguished temple made
4 vara-viu-lakityatanam [||*] for Brahm, vara (iva), and Viu.
_______________

First edited by T.A. Gopinatha Rao, Mandagapattu Inscription of Vichitra-
chitta, E.I., XVII (1923-24), pp. 14-17. See also S.-I.I., XII, No. 12.
272

Tirukkaukkuam Tami Inscription1

of King Narasihavarman-I

1 rr [||*] Tirukkauk[ku*]attu perum- rr || KkKT epma-

2 aigaukku [|*] Kaattr-kat- kK. kTREkaT-
3 [tu . . . Tirukkaukkua-]2 [ . . . KkK-]
4 ttu r mal[ai]ml3 T r mwlEmL
5 mlatattu perum- ltaT epma-
6 aigaukku vaipuppuam- kK vpaPma-
7 ga Vtpikoa-Naracigap- k vataekaBdnrkP-
8 ptta[yara]4 vaytatu [||*] EpaTt[yr] vYt.
_______________
1
Annual Report for 1932-33 on South Indian Epigraphy, p. 55.
2
The third line has proved beyond my power to decipher. The reading which I have given here
in brackets is that suggested by previous editors. I have left the estampage copy untouched in the third
line so that the difficulty may be appreciated! (ML)
3
My delineation of the medial vowel ai in malaiml of the fourth line is purely conjectural!
4
A reversed mix-up of the syllables rya?
 273

King Candrdityas Cave-Temples Foundation Inscription at Mlaccheri

1 Kritam-idan-npatin caused to be made by king

2 candrdityna sarvvan[tha*]n [|*] Candrditya, emperor, was
3 r-ikhari-pallavvaram-iti r-ikhari-Pallavvaram,
4 aivan-dhma simhapur1 [||*]2 this iva temple, in Simhapuram
_______________
1
The engraver forgot to engrave on the stone surface the right half of the m of
this word! I have re-instated it in my delineation. (ML)
2
Edited in South-Indian Inscriptions, XII, No. 115, with plate V.
274

Badami Rock Inscription of King Narasihavarman-I1

Text and word-for-word translation:
1 . . . [sa*]vatsar tman rjya-var- ca varddhamn tra[yda*] . . .
year-in-(the) own regnal year-in and running thirteenth
2 . . . Narasihna Mahmallna vidvim Vtpir-atimn . . .
Narasiha-by Mahmalla-by enemys Vtpi haughtiness
3 . . . [Bhradv*]ja . . . iur-atula gtra guair-tman . . .
Bhradvja -iu matchless gtra deeds (his)-own
4 . . . t . . . taya kiti bhujm-agrsara Palla[va*] . . .
earth rulers foremost Pallava
5 . . . [s]ihaviur-api- [ya] stambha-jaya . . .
Sihaviu veritable who pillar (of) victory
6 . . . [*]r-b-bahu-mat prakhyta . . tth . . .
many opinion esteemed

Running translation:
. . . in the 13th year running of his reign, Narasiha Mahmalla wiped out the haughtiness of (his) enemy
Vtpi and brought renown to the matchless Bhradvja gtra. . . . That veritable Sihaviu who (is)
the foremost of the Pallavas and the kings of the Earth . . . (captured the) pillar of victory. . . .2
_______________
1
The delineation, opposite, is based on facsimiles in Indian Antiquary, IX (1880), and South-
Indian Inscriptions, XI, Part I.
2
In the Vlrppaiyam Plates of Nandivarman-III, it is claimed that Narasihavarman, after
defeating the host of his enemies, took from them the pillar of victory standing in the centre of Vtpi
(S.-I.I., Vol. 2, pp. 508 & 511):
Tad-tma-jd-virabhm-Mahndrd- Upndra-krtti-n-Narasihavarmm [|*]
his son-from appeared Mahndra-from Upndra fame Narasihavarman
Vtpi-mady vijitrivargga sthita- jaya- stambham-alambhayad-ya [||11*||]
Vtpi middle-in conq. enemy-host standing victory pillar removed who
 275

6
276

1 Sambhavasthitisa[mh]rakraa vtakraa [|] Bhydatyantakmya jagat

2 kmamarddana || Amyacitramysvagu guabhjana [|] Svasth

3 niruttar jydana paramvara || Yasyguhabharkrnta kailsa-s-sada

4 nana [|] Ptlamagamanmrddhn rnidhistambibhartyajam || Bhaktiprahva manas bhava bh-

5 aallay [|] Do ca y bhuv bhra jytsa rbharaciram || Atyanta-

6 km npatirnnirjjitrtimaala [|] Khyt raajaya ambhstnda vma

7 kritam [||] Ja sthurnnikala sma pvaktm viyadvapu [|] Bhma iv vijaya-

8 t akara kmasdana || Rjarj na virasacakrabhnna janrddana [|] Trakdhipati svasth

9 jayatttarukura || rmattyantakmasya dviaddarpppahria [|] rnidh k-

10 margasya harrdhanasagina || Abhikajalpr citraratnmbujkar [|] -

11 st vil sumukha irassarasi akara || Tnda kritantugandhrjjarmmandira

12 ubha[] [|] Prajnmiasiddhyarttha kar bhtimicchat || * || Atyantakmapallavvaragham ||

13 Dhiktndhiktmpunarapi dhigdhigdhigastu dhiktm [|] Ynna vasati

14 hday kupathagativimkak rudra ||

 277

10

11

12

13

14

King Paramvaras Dharmarja Maapa Inscription

 278

Inscription of King Paramvara, the First, on the floor of the divarha Cave-Temple, Mmallapuram,
in front of the sanctum sanctorum:

1 Dhiktn-dhikt punar-api dhig-dhig-dhigastu dhikt [|*] Yn-na vasati hday

2 ku-patha-gati-vimkak rudra ||
 King Paramvaras inscription on the floor, at the entrance of the Rmnuja Maapa, Mmallapuram:

1 [r |* Dhi]kt[]-dhikt[] punar-api dhig-dhig-dhigastu dhikt

2 [|*] Yn-na vasati hday ku-patha-gati-
3 vimkak rudra ||

279
280
Atiraacavara Cave-Temple Inscription (Grantha Script)
(Delineations based on plates No. 1 & 2 in Epigraphia Indica, X, No. 12)

1 rmattyantakmasya dviadda-

2 rpppahria [|*] rnidh kma-

3 rgasya harrdhanasagina ||

4 Abhikajalpr citraratnmbujkar [|*]

5 st vil sumukha irassarasi akara ||

6 Tnda krita ambhrbhavana bhtay bhuva [|*] Kai-

7 lsamandaranibha bhbht mrdhni tihat || Bhaktiprahv-

8 a manas bhava bhaa[]llay [|*] D ca y bhuvandhatt

9 jytsa rbharaciram || Atiraacaa patiravanibhu-

10 jmatiraacavaramidamakart [|*] Iha giritana-

11 yguhagaasahit niyataktaratirbhavatu paupa-

12 ti || Gurvvmnabhakti riyamatiayin durvvaha bhramurvvy-

13 nissmnyaca dna samamati(ra)acakhyay [y bibhartti] [|*]

14 Sthn nirmmpitsminvi[ditaraaja]yakhytin tna [bhar]tt bhtn-

15 maamrtticiramatiraaca[va]r ytu nihm [||*] A(nugra)la ||

16 Yadi na vidht bharat yadi na harirnnrad na v skanda | Bddhu ka iva

17 samartthassagta klaklasya |||| Samaradhanajaya Sagrmadhra ||||

 281

10

12

14

16
282
Atiraacavara Cave-Temple Inscription (Ngar Script)

1 rmattyantakmasya

2 dviaddarpppahria [|*] rni-

3 dh kmargasya harrdhanasagi-

4 na || Abhikajalpr citraratn-

5 bujkar [|*] st vil sumukha irassara-

6 si akara || Tnda krita ambhrbhava-

7 na bhtay bhuva [|*] Kailsamandaranibha bhbht

8 mrdhni tihat || Bhakti[prahv]a manas bhava[] bhaa

9 llay [|*] D ca y bhuvandhatt jytsa rbharaci-

10 ram || Atiraacaa patiravanibhu[jma]tira-

11 acavaramidamakart [|*] Iha giritanaygu-

12 hagaasahit niyataktaratirbhavatu paupati ||||

13 Gurvvmnabhakti riyamatiayin durvvaha bhramurvvy niss-

14 mnyaca dna sama[ma]tiraa[cakhya]y [y bi]bhartti [|*] Sthn

15 nirmmpitsmin[v]iditara[ajayakhytin t]na bhartt bhtn-

16 maamrtticira[matiraaca]var ytu nihm |||| Svasti ||

 283

10

12

14

16
 The Shore Temple Sacrificial Altar (Bali-Pha) Inscriptions of King Rjasiha:
(delineations based on facsimiles in Epigraphia Indica, XIX, No. 18-A, pp. 105-9 and Plate)
Altar A
West side:

South side:

East side:

North side:
 The Shore Temple Sacrificial Altar (Bali-Pha) Inscriptions of King Rjasiha:

Altar A
West side:
r | Apratimam=Avani-bhaam=Akalakan-Dharani-candram=avanndr [| *] Arimarddanam=Atula-
fortune matchless earth- ornament umblemished earth- moon earth-rulers enemy-crusher unequaled-
South side:
(ba)lam Kula-tilaka y namanti t . . . [|| 1*] . . . (Bhaya)rahit Bahunaya
prowess race-forehead-mark who salute those fearless great-statesman
East side:
Atyanta- kmam=Aparjitam=karja-Candrrddha-khara-ikhmaim=Adbhuta yam [|*] Caa-
boundless-desires invincible #1-king moon-half- headed- crest-jewel marvelous-(one) whom fierce-
North side:
ni kitibhtm- mahatm=asahya samprpya kmam=iva nandati jva- lka || [2*] r- Udaya-candra ||
thunderbolt earth-bearers-to great unbearable attaining well thus enjoys living-beings-world illustrious-rising-moon

Verse 1: Good Fortune! Those rulers of the earth who salute the Matchless One, the Ornament of the Earth, the Unblemished Moon (of this) World,
the Crusher (of his) Enemies, (He of) Unequaled Prowess, the Forehead Beauty-Mark (of his) Race.
Verse 2: The Fearless, the Great Statesman, (He who has passed) beyond limiting desires, the Invincible, the Foremost among Kings, (He whose)
Crest Jewel is the One (i.e., iva) whose head is ornamented by the crescent moon, the Marvelous, the Fierce Thunderbolt, (He who is)
Unbearable to the great Bearers of the Earth (i.e., to other kings), through whom the world of the living attains full enjoyment.
The illustrious Rising Moon!

285
 The Shore Temple Sacrificial Altar (Bali-Pha) Inscriptions of King Rjasiha:

Altar B
West side:

South side:

East side:

North side:
 The Shore Temple Sacrificial Altar (Bali-Pha) Inscriptions of King Rjasiha:

Altar B

West side:
r Rjasih Raa- jaya() r-bhara- Citra- krmmuka [|*] ka-vra-ciram-ptu iva-c- mair-m-mahm [|| 3*] r- kmuka
illust. Rjasiha battle-victor prosp.-possessor astonish.-archer #1 - hero ever protect iva-crest-jewel- earth prosperity-desirer
South side:
Kla- kla Kl . . . . . . . . . [|*] (A)bhirm vijayat Raa- bhm Gulaya || [4*] r- vallabham-Ati-
deaths-death handsome prospers battle- terror virtue-abode fortunes favorite most-
East side:
mna Raa-vra Kula-(dhvaja?) . . . . . . [|*] rjjitam-Unnata- rmam praamata Yuddrjju(nam) . . . . . . [|| 5*]
honored battle-hero race- (banner) mighty extremely-graceful salute-lets battle-Arjuna
North side:
. . . (y)am-arttha-d (ru)taam-ma . . . ma . . . [|*] Tri-nayana- bhakta- mitran- Narndrasihan-namanti np || [6*]
whom wealth seen 3-eyed (one) devotee friend-of king- lion salute kings

Verse 3: May the illustrious Rjasiha, the Victorious in Battle, the Possessor of Prosperity, the Amazing Archer, the Foremost Hero, (He whose)
crest-jewel is iva, forever protect the earth!
Verse 4: The Lover of Prosperity, the Death to Death, Kl . . . . . . The Handsome, the Terror in Battle, the Abode of Virtue, prosper!
Verse 5: Let us salute the Favorite of Fortune, the Most Honored, the Battle-Hero, the Banner (of his) Race, . . . the Mighty, the Most Pleasant, an
Arjuna-in-Battle, . . . . . .
Verse 6: the Wealthy, . . . the devotee-friend of the Three-Eyed One (i.e., iva), the Lion among Kings, on seeing whom, the rulers of men salute.

287
288

Early Pallava inscriptions in Kchpuram

Madras, July 19: A historic Pallava inscription has been found on the steps of the
Mangalatheertham in front of the Kanchi Sri Sankaracharya Mutt. It may be recalled
that this tank, not in use, was in ruins and at the instance of H.H. Paramacharya of
Kanchi, the HR and CE department renovated the tank, at a cost of Rs. 6 lakh. During
the Paramacharyas visit to this tank, His Holiness noticed a few inscriptions on the
steps of the tank. At his suggestion Dr. R. Nagaswamy, Director of Archaeology,
Government of Tamil Nadu, who visited the site and studied the inscriptions, found
them to be Pallava Grantha inscriptions assignable to A.D. 700 to 725.
It is the beginning of the inscription of the Pallava ruler Rajasimha who ruled
between 690-728 A.D. and exactly the same words are inscribed in the Ganesa ratha,
monolithic temple and the Dharmaraja mantapa at Mamallapuram.* Rajasimha built
the historic temple of Kailasanatha in Kanchi.
The inscription also furnishes clues closely connecting the Mamallapuram monu-
ments with the ruler of Kanchi.

[Printed in the Indian Express, Madras, Wednesday, 20 July 1988, with facsimile.]
_______________
*Since I believe that the inscriptions of the Gaa Ratha and Darmarja
Maapa belong to King Paramvara (King Rjasihas father), I would attribute
this beginning fragment of those longer (and complete) inscriptions to Paramvara.
(ML)
289
 The divarha Cave-Temples Tamil1 Inscription of Nandivarman-II

Front of Slab
1 Svasti r Palla[va*]-vasasya [||*] Na[ti-]
2 bdhuvarmmaku rjyvibhid[dhya-*]2 2
3 =celniatu upattu
4 antvatu3 [|*] Mmallap[u-*] 4
5 rattu nakarattr Iaivaa[-*]
6 c Kaa kau-m-[kaa-*] 6
7 niyka koa nilam [|*] [-*]
8 mr-nu Kuattr-i[ai*] 8
9 vum Kuattr-ki[-*]
10 r Ia-Nauvur maka . . 10
11 Kaan-i[ai] kau-m-ka[ani-*]
12 yka viukoa nilat[tu-*] 12
13 [k*]k=ellai [||*] Maiai Nall[u-*]
14 r ta[t*]ti[] mlkum [te-*] 14

Back of Slab
15 pl-ellai kn[ri*]
16 [i] vaakkum mlpl-[e-*] 16
17 llai Mntai talaiva[*]
18 []rikkum peru-vai i-[ki-*] 18
19 [ak*]kum vaapl=e[l*]lai Pa-
20 nappi-e[l*]lai i-[te-*]
20
21 [ku*]m ik=ellai [a-*]
22 ka[p*]paa nilamum Kaan
22
23 [i*]ai viukou po-
24 kau-m kaaniya- 24
25 [kku*] peyit [||*] Itu a-
26 [i*]ymai ktt=ai e- 26
27 [*]allai mlatu [||*]

_______________
1
The first two lines are corrupt Sanskrit written in the Pallava Grantha script. The scribe has
omitted the final va of Pallava. In the 25 lines which follow, the Tami spelling is variable, and the
Pallava Tami script, crude. Restorations and interpolations by me have been marked with asterisks in
the transliteration. I have interpolated some of the puis in my facsimile delineation. (ML)
2
Read: pta-varmmaku rjybhivddhya- for this second line.
3
Read: aintvatu WNtav.
 291

1 Svasti r Palla[va*]-vasasya [||*] Na[ti-]

2 bdhuvarmmaku rjyvibhid[dhya-*]

3 ecla ApT
4 Ntav. mamLl-
5 rT nkrTtaR iwdvz-
6 ca kBd kEm-kz-
7 yak ekaBd lM. A-
8 R naD TR iwd
9 vaM TR za-
10 R i-nbaR mk . .
11 kBdwd kEm kz-
12 yak ekaBd lT-
13 KekLwl. mw nL-
14 zaR EtaDdT EmLM et

15 paLLwl EkaEn
16 i vdKM EmLpaL -
17 Lwl maNwt twlv
18 KM epv i -
19 zKM vdpaelLwl p-
20 nPpa Lwl i et-
21 M iaekLwl -
22 kPpDd lM kBdN
23 iwd ekaB epa-
24 kEm kzy-
25 K epta. i -
26 yawm kaTta -
27 Lwl Eml.
292
 Bibliography
BOOKS
Balasubrahmanyam, S.R. Early Chola Art: Part I. Bombay: Asia
Publishing House, 1966.
Berkson, Carmel. The Amazon and the Goddess: Cognates of Artistic
Form. Bombay: Somaiya Publications Pvt. Ltd., 1987.
Burgess, J. Elura Cave Temples. Reprinted in 1970 by Sagar Publica-
tions, New Delhi.
Coomaraswamy, Ananda K. History of Indian Indonesian Art. New
York: Dover Publications, Inc., 1965.
Dain. Avanti-Sundar-Kath-Sra. Edited by G. Harihara Sastri.
Madras: Kuppuswami Sastri Research Institute, 1957.
Fabri, Charles. A History of Indian Dress. Calcutta: Orient Longmans,
1960.
Jouveau-Dubreuil, G. Pallava Antiquities. Two volumes. Pondicherry:
1916 & 1918.
Lerner, Martin, and Steven Kossak. The Lotus Transcendent: Indian and
Southeast Asian Art from the Samuel Eilenberg Collection. New
York: The Metropolitan Museum of Art, 1991.
Lockwood, Michael. Mmallapuram and the Pallavas. Madras: The
C.L.S., 1982.
. Mmallapuram: A Guide to the Monuments. Madras:
Tambaram Research Associates, 1993.
Lockwood, Michael, Gift Siromoney, and P. Dayanandan. Mahabali-
puram Studies. Madras: The C.L.S., 1974.
Lockwood, Michael, & A. Vishnu Bhat. Metatheater and Sanskrit Drama:
2nd, Rev. & Enlgd. Ed. Madras: Tambaram Research Associates, 2005.
Longhurst, A.H. Pallava Architecture, being Memoirs of the Archo-
logical Survey of India, in three parts: Nos. 17, 33, and 40. The
Archological Survey of India, Simla, 1924, and Calcutta, 1928
and 1930.
Mahalingam, T.V. Kcpuram in Early South Indian History.
Bombay: Asia Publishing House, 1969.
Minakshi, C. Administration and Social Life under the Pallavas.
Revised edition. Madras: University of Madras, 1977.
. The Historical Sculptures of the Vaikuhaperum Temple,
Kch, being Memoirs of the Archological Survey of India,
No. 63. Delhi: Archological Survey of India, 1941.
Nagaswamy, R. New Light on Mmallapuram, Transactions of the
Archological Society of South India: 1960-62. Madras: 1962.
Raman, K.V., et al. (eds.). rnidhi: Perspectives in Indian Arch-
ology, Art and Culture. Madras: New Era Publications, 1983.
Ramesan, N. Studies in Medieval Deccan History (Late Pallava and
Telugu Chola Period), being Copper Plate Inscriptions of the
State Museum, Vol. III, Archological Series No. 29. Hyderabad:
The Government of Andhra Pradesh, 1972.
 294 Siegel, Lee. Laughing Matters: Comic Tradition in India. New Delhi:
Pallava Art Motilal Banarsidass, 1989 (1987).
Sivaramamurti, C. Early Eastern Chalukya Sculpture, being Bulletin of
the Madras Government Museum: New Series General Section,
Vol. VII, No. 2. Madras: Madras Government Museum, 1962.
. Mahabalipuram. Third edition. New Delhi: Archo-
logical Survey of India, 1972.
Soundara Rajan, K.V. Cult in the Pallava Temples, Transactions of
the Archological Society of South India: 1962-65. Madras: 1969.
. Indian Temple Styles. New Delhi: Munshiram Manoharlal,
1972.
. Rjasihas Temples, Transactions of the Archological
Society of South India: 1962-65. Madras: 1969.
Srinivasan, K.R. Cave-Temples of the Pallavas. Architectural Survey
of Temple Series, No. 1. New Delhi: Archological Survey of
India, 1964.
. Some Aspects of Religion as Revealed by Early Monuments
and Literature of the South. Madras: University of Madras, 1960.
. The Dharmarja Ratha and Its Sculptures: Mahbali-
puram. New Delhi: Abhinav Publications, 1975.
Srinivasan, P.R. Beginnings of the Traditions of South Indian Temple
Architecture, being Bulletin of the Madras Government Museum,
New Series General Section, Vol. VII, No. 4. Madras: Madras
Government Museum, 1959.
Subramaniam, T.N. The Pallavas of Kchi in South-East Asia.
Madras: The Swadesamitran Press, Ltd., 1967.
The Rmyaa. Sanskrit edition published by Jalana Motilal,
Gorakhpur, undated.
Widdess, Richard. Ragas of Early Indian Music: Modes, Melodies and
Musical Notations from the Gupta Period to c. 1250, OUP, 1995.
JOURNALS
Bulletin of the Institute of Traditional Cultures, Madras, Jan. to Jun.,76.
Damilica, Vol. I.
Epigraphia Indica, Vols. X, XVIII., & XIX.
The Indian Antiquary, Vols. V & VI.
Journal of the Asiatic Society, Vols. I & IV.
Journal of the Epigraphical Society of India, Vols. III, IV, & XX.
Journal of the Madras University, Vol. XLI, Nos. 1 & 2.
Journal of Tamil Studies, Vols. V & VIII.
Koku (in Tamil), Vol. V.
Lalit Kal, Number 23 (1988).
The Madras Christian College Magazine, Vol. XLVIII.
Marg, Vol. XXIII, No. 3.
South-Indian Inscriptions, Vols. I; II, Part 3; III; VIII; XII.
Transactions of the Archological Society of South India: 1960-62, and
T.A.S.S.I: 1962-65.