Vāmacāra Viṣṇu in Hindu Iconography A Problem in Sociological Values

Author(s): Raju Kalidos

Reviewed work(s):
Source: East and West, Vol. 44, No. 2/4 (December 1994), pp. 275-291
Published by: Istituto Italiano per l'Africa e l'Oriente (IsIAO)
Stable URL: http://www.jstor.org/stable/29757154 .
Accessed: 22/11/2012 17:23

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp

.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.

Istituto Italiano per l'Africa e l'Oriente (IsIAO) is collaborating with JSTOR to digitize, preserve and extend
access to East and West.

http://www.jstor.org

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
 V?mac?ra Visnu inHindu Iconography
A Problem in Sociological Values

by Raju Kalidos

The Hindu Trinity consists of the gods, Brahma, Visnu and Siva, who are
considered to represent the cosmic activities of Creation, Preservation, and Destruction

respectively. Normally in iconographical illustrations, they appear independently and

rarely get united in syncretistic forms like Harihara, Som?skanda and the Elephanta
type of Trim?rti. In most of these forms Visnu is usually assigned the v?mac?ra
'lefthand' place. A typicalexample of this typeof illustrationisHarihara in the
iconographical constitution of which Visnu has only a v?mac?ra place from very early
times (e.g., Kus?na and Gupta) down to the late Vijayanagara period. Literary accounts,
both Saiva and Vaisnava, are almost unanimous in assigning the v?mac?ra status to
Visnu which has a significant connotation of its own in the sense that the 'left-hand
place' in Hindu tradition is supposed to be characteristic of woman, the Feminine
Principle. A very good iconographical example of this concept is Siva-Ardhan?risvara.
When Visnu, an importantgod of theTrinity, is assigned such a positionwhich is
subordinate social circles, it has to be viewed with rapt attention.
in Hindu If the
Saivas could treat him on a par with a feminine concept, its raison d'etre could be
obviously explained as a sectarian treatment, but why should the Vaisnavas themselves
accept this? This question acquires freshmagnitude when we come across certain motifs
wherein Visnu has a right-handplace in a composite image like theTrim?rti. For
example some of the small caves in the Gmeiz-lena group and cave XVI on river beds
at Ellora house a number of Trim?rti images wherein the place of Rudra-Siva is shifted
to the v?mac?ra place. Now the query is: why the deviation and what contributed
towards this secondary development over the iconographical norms set by the Elephanta
masterpiece? The development of this form has a significance and relevance of its
own, and hence this paper. So the aim of the present study is to trace the history
of Visnu's v?mac?ra place in composite forms through an analysis of early mediaeval
Tamil bhakti literature which contains a number of references to this aspect and its
impact over the iconographical formulations of the age, with due credence to the
socio-cultural setting of the theme.

(*) Paper in the Sixth International Seminar on 'Asian Studies', Asian Research Service,
presented
Hotel Furama,Hong Kong, July 1988. Thanks to the Indian Council ofHistorical Research who funded
the travel to Hong Kong.

[1] 275

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
 Literary Gleanings

As warranted by the problem that has been stated above we may first of all take
into account the literary gleanings from both Tamil and Sanskrit. The earliest clear
presentationof theproblem inTamil is traceablein thehymnsofAlv?r andN?yanm?r,
custodians of the early mediaeval bhakti movement in Tamil Nadu. The Alv?r were
Vaisnava savants whose contribution to literature is collectively called N?l?yirativiya
pirapantam. The N?yanm?r were Saivas whose writings are codified under the Tirumurai
series {l)which includes the contributionsof theearly trio (2),called Ter?vam. There
are reference to Visnu's v?mac?ra status in these works.
The Ter?vam has a number of allusions in which the composite character of the
iconographical form called Harihara is affirmed. To cite few examples: M?lorup?kam
{Ter?vamIV 87.7) 'havingM?l (Visnu)on one side' and Tirum?l?rp?katt?n{Ter?vam
VI 238.5) 'having Tirum?l (Sri-Visnu) on one side', and so on. There is a specific
indication to itap?kam 'left-hand place' in which particular allusion it is stated
that the Lord, Siva, had N?r?yana merged with him on the left side: N?r?yananai
yitap?kattataint?r {Tev?ram VI 267.2). This gives a clear perspective to the composite
form under review. There is also an attempt to equate Visnu with Siva's consort,
the Siva-Sakti Um? or P?rvati, who in another composite iconographical form called
Ardhan?risvara occupies Siva's left side. The interesting passage is: M?torup?lu
M?lorup?lum {Tev?ram I 97.2), meaning that m?tu (Skt. n?rt) 'woman' and M?l (Visnu)
are the lefthalf of Siva. So from this account it is clearly stated thatVisnu is not
only assigned a v?mac?ra place but also placed on par with Siva-Sakti.
The ?lv?r call the united form, Aran-N?ranan = Skt. whose
( Hara-N?r?yana),
v?hanas 'vehicles' are the bull (Nandi) and kite {garuda)and weapons the trident
and disc wherein the body is united {Tiruvant?ti 1.5). There are also a number of
allusions in which the syncretic character is pointed out: malai mankaitan pankanai
pankilvaittu {PeriyaTirumoli7.10.3) (3) 'havingplaced onewho possess thehill-woman
(Ardhan?ri)on his own part'; Icar kicaintuutampil?rk?rut?nkotutt?n {ibid. 9.10.4)
'hewillinglygave a place to Isvara (Siva) on his own person'; p?kam nalki {ibid. 10.9.4)
'having given the half, meni yonru {Tiruvant?ti 1.5) 'the body is one', Cataiydnai

(*) These are twelve in number, called Pannirutirumurai,and supposed to be the work of the
N?yanm?r who lived during the period fromA.D. 500 toA.D. 1300 (at the latest).
(2) The early trio consists of Tirun?nacampantar, Appar alias Tirun?vukkaracar and Cuntarar,
authors of the first seven Tirumurais: 1-3, 4-6 and 7 respectively. The first seven Tirumurais are collectively
named Tev?ram and the eighthTiruv?cakam and Tirukk?vaiy?r.
compareswith Tev?ram (I 97.2) cited above. Mankai tanpankan 'having
(3) This passage strikingly
the woman on one side' would denote Siva's Ardhan?ri form. Siva in such a form is said to occupy
Visnu's half, i.e., the rightside, inwhich case Visnu's v?mac?raposition is not only accepted but the
equation of Visnu with Siva-Sakti also indirectlyacknowledged.

276 [2]

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
p?kattuvaitt?n(Tiruv?ymoli10.4.6) 'havingthematted-locked one (Siva) on his own
person', and so on.
In all the above cited references the core idea is that Visnu allowed Siva to occupy
one half of his body. But which part? it is not clearly stated. But there is a specific
reference regarding this question in which Visnu to possess
is said 'one who has
the matted locks with a Crescent adorning them (i.e., Siva, called Cataiyan) on his

right side': pirai tanku Cataiy?nai valatte vaittu (Periya Tirumoli 3.4.9). In all these
allusions we get the other dimension of the problem wherein the Vaisnavas themselves
acknowledgethey?mac?raplace ofVisnu incompositeimageslikeHarihara by declaring
thatVisnu had Siva on his right side.
left-hand place of Visnu inHarihara
The images is also attested by some Sanskrit
texts. It may also be stated here that references on the subject under the present
head are scanty. The Mah?bh?rata (III 40.57), supposed to give the earliest account
of the form (subject to authenticity of the relevant passage), calls the Lord Hari-Rudra.
TheMatsya Pur?na (260.22-27) and Devat?m?rtiprakarana (VI 35-41, 56-57) are almost
identical in iconographical descriptions and bear witness to Visnu's v?mac?ra status
(Desai 1973: 51-52, 60). The Matsya Pur?na uses the phrase v?m?rdha M?dhavam,
meaning therebythatM?dhava (Visnu) occupies the lefthalf {v?ma 'left'and ardha
'half'). The Devat?m?rtiprakarana handles the name Harihara while describing the
formand states that Siva is always on the right side and Visnu on the lefthalf.
The concerned passage is: Daksin?rdhe Sad?sivam v?man?rdhe ca Hrsikesa. Apart from
them, there are a number of silpas?stras in South India which speak of pratim?laksana
and enlist the Harihara form. The S?rasvatiyacitrakarmas?stra (chap. XXVII) calls
theLord Kesav?rdham?rti 'theLord with theKesava half and says that the left is
occupied by Kesava and that this half exhibits the characteristic attributes of Visnu
as follows: with all kinds of ornaments including the krita makuta, hands in ?ru hasta
(lower) and bearing the sahkha (upper), wearing pit?mbara, green in complexion and
so on. Similar accounts are also found in the K?syapasilpas?stra (patalam 76) and
Silparatna (chap. 22. vv. 129-37).
Lingodbhava is an important form, often described by the N?yanm?r {Tev?ram
1 13.9, 36.7, 37.9, 39.9, 86.9, 17.9, III 344.9), which attests the secondary and tertiary
place accorded to Brahm? and Visnu in thisbasically Saiva formwhich projects the
glory ofMah?linga whose head and foot the other two wanted to scale and ultimately
failed miserably. But in the hymns of N?yanm?r nothing regarding Visnu's v?mac?ra
position is specifically stated (4). Being the product of sectarian animosity on the part
of the Saivas which purports to project the place of Siva asMah?linga, the Alv?r never
allude to this form. But in their hymns there are a number of references to a form
calledM?var?yamutalvanwhich has earlierbeen identifiedwith theElephanta type
of triunebust (Kalidos 1986c: 319-21; 1988: 416-37).

(4) It may be observed that in iconographical specimens of the age, Visnu retains his v?mac?ra
status. See below for details.

[3] 277

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
 We may now take into account the data forthcoming from the N?l?yirativiyapira
pantamwhich refersto the typeof triunebusts found in theElephanta and Ellora
caves. These images are supposed to represent Siva in his Sad?siva form by experts
in the field (Kramrisch1946: 4; 1984: 65-85; Sharma 1976: 364). But an analysis
of the data found in the Tamil bhakti literature has provided scope for a reinterpretation
to the effect that these triunes represent a synthesis of Visnu (in centre), Rudra (to
the right)and Brahma (to the left) (5). The nomenclatureof the formisnoted several
times which supports the present thesis. The following for example may be noted:
m?var?kiya oruvan 'the one who became the three' (Periya Tirumoli 3.1.10), m?varum?y
mutal?ya m?rtti represented the three and became first among them'
'the Lord who
(ibid. 4.1.2), 'one
m?vuruv?navane who became three in form' (ibid. 6.1.6), m?varil
mun mutalvan 'first and foremost among the three' (ibid. 9.9.1), m?varil mutalvan 'first
among the three' (Tiruv?ciriyam, v. 3), M?var?kiya m?rtti 'the Lord who acquired a
triune form' (Tiruv?ymoli 3.6.1), m?vuruv?m mutalvan 'first in a triune form' (ibid.
7.9.2), and m?var mutalvan 'first among three' (ibid. 9.8.9). The core thought behind
these cryptic passages is: (1) that Visnu represented the Trim?rtis, (2) that he was
the first and foremost among them, and (3) that these three either appeared in a triune
polycephalous form (as in the Elephanta type of image) or represented three distinct
physical forms in a panel (as in the Tiruccir?ppalli lower cave or Lankesvara and Milk
Maids caves at Ellora). The polycephalous aspect of this typeof Trim?rti is clearly
pointed out in a particularhymn (PeriyaTirumoli 2.5.3) wherein theLord is said to
be 'three but in form one and yet the persons (m?rti) different':

Utampuruvil m?nronr?y m?rttiver?y [...]

In another significant reference the Lord is said to 'stand as the great one among all
other gods and amidst them in triune form, being the primeval one'. The relevant
passage is:

Tevar?y nirkumat tevum attevaril

M?var?y nirkum mutu punarppum
(N?nmukan Tiruvant?ti, v. 54).

(5) This aspect has been elaborated inmy paper, presented in theEUora National Seminar (Kalidos
1988) and the forthcoming monograph on Vaisnava iconography(Kalidos 1986c). The argumentscentre
on the following ideas: (i) the facial expressions of triunebusts represent the sattva,r?jasa and t?masa
gunaSycharacteristicofVisnu, Brahma and Siva respectively;(ii) in theElephanta triunethe central face,
representingVisnu, is to the left side of Rudra, the v?mac?raposition under investigation;and (iii) the
referenceswhich appear profusely in the hyms of Tamil bhakti savants justifythe existence of such a
form which however is missing in the early mediaeval and pre-early mediaeval Sanskritic texts, and so
it is a product of the formvisualized by theTamil writerswhose ideasmust have spread far and wide
in South India.

278 [4]

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
Though thereare a number of such referencesin the hymnsof the?W?rs nothing
definitely is stated regarding the v?mac?ra status of Visnu. But with due reference
to the iconographical specimens of the age (infra)we may assertively state that Visnu' s

place was mostly the left half.

Iconographical Representations

artistic tradition from early times, at least the Kus?na,

In Indian attempts have
been made to syncretize the personalities of Visnu, Siva and Brahma into a physical

entity. Such forms are Harihara and Trim?rti. There are also some representations
in which the Trinity appear within a frame but as separate beings. Examples are

Lihgodbhava, Kaly?nasundaram?rti, Som?skanda and Tripur?ntaka. There are also

rare examples of Trim?rtis, appearing independently on a status of equality as primus
interpares ones, but within a bigger frame. Examples are found in the Tiruccir?ppalli
lower cave, Lankesvara in the Kail?san?tha temple (cave XVI) at Ellora and the
Milk Maids cave (no. XXVII) at Ellora. There are rare pieces of Yoga-Nrsirhha (or
Kevala-Nrsirhha in seated aspect) and Vaikuntham?rti images in a cave temple at
N?makkal in which Siva and Brahma inducted as a sort of corollary
find themselves
to Lingodbhava images. During period a form called Ekap?dam?rti
the later mediaeval
'the Lord with one leg' became very popular in the Tamil country which appears to
be an elaboration over the Elephanta type of triune, not in bust form but a full-fledged
image.

Harihara

The earliest surviving images of the theme have been traced back to the Kus?na
andGupta periods (Sahai 1975: 136-37;Desai 1973: 52-53) which set thepioneering
model for later types, appearing in a redundant quantity during the early and later
mediaeval periods. The pattern set was that a line drawn through the centre of the
image from top to bottom represented the characteristic aspects of Siva and Visnu
on the right and left sides respectively. Mostly the images were samap?da-sth?naka
which means standing in an upright position (Fig. 1). Rarely we come across seated
images of which we have a specimen in one of the left kosthas on the mukha-mandapa
part of theVir?p?ksa temple at Pattadakkal (Fig. 2). Rare instancesof tribhahga
images have also been reported (Kalidos 1986c: 510). Mostly the images are caturbhuja
and very rarely astabhuja. A number of specimens, subscribing to the usual typology
(i.e., samap?da-sth?naka and caturbhuja) are traceable in the early mediaeval cave and
structuraltemples at Bad?mi (caves I and III), Pattadakkal (Vir?p?ksaT.) Aihole
(R?vulaphadi cave and Durga T.), M?mallapuram (Mah?var?ha-Visnu-gr?? cave),
Kannakkuti caves and N?mmakkal (Cave I). Iconographically the features characteristic

[5] 279

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
 - - Harihara
Fig. 1 Harihara, Durg? temple,?ihole. Fig. 2 (seated), Vir?p?ksa temple,
Pattadakkal.

of Siva (e.g., jat?makuta, tanka or mrga, a short loin-cloth and the vehicle nandi)
and Visnu {Kirttamakuta, cakra or sankha, pit?tnbara and the vehicle, garuda) are
represented in the images. Visnu in all these types uniformly occupies the v?mac?ra
place (Figs. 1-2).

Trim?rtis

Literally meaning 'triune', these iconographical representations purport to

syncretizetheHindu Trinitywhich fallunder four typologicalvariables: (1) triune

busts, (2) independentTrim?rtis in a panel, (3) Ekap?dam?rti and (4) Som?skanda.
The types are mostly concentrated
triune in the Ellora caves, especially in the
garbhagrha of Lankesvar and the small caves (Ganesa-fe?tf and Cave XVI groups) on
the riverbeds of Ellora and Bhairavakonda. Normally theyappear on theback wall
of the cell in frontof which usually a Lihga is found. However the classical piece
found in the Elephanta cave is housed in a kostha 'niche'. Most of these works are

earlymediaeval and pass into oblivion, perhaps due to the unpopularityof images

280 [6]

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
conceived in bust forminHindu
tradition. The Elephanta master?

piece and its replica found in the

Ellora caves consist of poly
cephalous representations, having
Visnu in centre, Brahma to the left
and Rudra to the right(Figs. 3-4).
The facial expression in respect of
Rudra is terrific,usually the jv?la
'flame' issuing forth from the
mouth. Here the placement of
Visnu is significant
because though
in centre, he is to the left of Rudra.
Thus there is a stress on his
v?mac?ra status. However, in the
riverbed group at Ellora we find
a number of Trim?rti (totally13)
busts inwhich theplace of Rudra
is shiftedto the leftside (totally9)
and thus assigned a v?mac?ra status
(Fig. 5). This is very strikingand
significant in the context of our
study. If Sad?siva the heads
cannot be shifted on any account.
The secondvariable is theTrim?rtis. Specimens are found in theTiruccir?ppalli
lowercave (Fig. 6), Lankesvara and theMilk Maids cave (no.XXVII) inEllora. In
Bad?mi caves II and III and pillar sectionsof theVir?p?ksa temple at Pattadakkal
there are miniature reliefs. In all these representations members of the Trinity appear
standing in samap?da attitude. They are caturbhuja and hold their characteristic emblems
in para-hastasy viz., tanka and mrga or trisula for Siva, cakra and sankha for Visnu, and
kamandal? and aksam?l? for Brahma. The p?rva-hastas are uniformly in abhaya-mudr?
(right)and ?ru-hasta(left).What is importantin all these illustrationsis theplacement
ofVisnu's image. In theTiruccir?ppalli cave Brahm? is in centre,Siva to his right
side and Visnu to the left; naturally Visnu's place vis-a-vis Siva is the left and so v?mac?ra
(Fig. 6). In Lankesvara Siva is in the centre,Visnu to the leftand Brahm? to the
right. In marked contrast to these two illustrations, especially the Lankesvara where
Siva is the centre of the cosmic core, theMilk Maids cave panel represents Visnu in
centre, Brahm? to the right and Siva to the left. Thus Siva gets the v?mac?ra status
here. It isvery strikingand fallsin linewith the triuneimagesof theEllora river-bed
group wherein Siva is accorded a similar status. The miniatures of Bad?mi and
Pattadakkal are not conspicuous from the iconographical point of view and suffice
it to add that they represent Visnu in the usual v?mac?ra place.

[7] 281

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
 Fig.
5
-Trimfirti,
Ellora
(note
the
face
shifting
of
Rudra).

R4.

F
/K~.--
~
44%
*
Fig.
-Trimflrti,
4
Ellora.

a-~-~
ft K--
Yt.tvX~Q
V.2
*

00
00 00

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
 -
Fig. 6 Trim?rtis, Tiruccir?ppalli lower cave.

Som?skanda syncretic form which is mostly confined to the early

is another
mediaeval Pallava temples (both cave and structural) in the Tondaimandalam area.
These images appear mostly in the temples meant for Siva. Som?skanda appears on
behind theLihga, as we findTrim?rti images in the
theback wall of thegarbhagrha,
Ellora caves. The illustration uniformly represents Siva and P?rvati along with the
baby Skanda (T. Murukan) seated on a pedestal. In the background, behind Siva,
Visnu and Brahm? appear. Visnu's placement in all these cases is uniformly v?mac?ra.
This typeof representationfellintodisuse afterthePallava period but theCola artists
borrowed the theme for a number of bronze castings in which Brahm? and Visnu fail
to appear. Till contemporary times such images are used for processional purposes
in almost all templesof Siva inTamil Nadu.
Ekap?dam?rti is another ingeniousproduct of theHindu artists inwhich the
Trinity merge into an entity (Grossato 1987). The central m?rti is ekap?da 'one legged'
and has Brahm? and Visnu emanating fromthehip of the central core (Fig. 7). So
the Trinity are present, all caturbhuja and holding their characteristic emblems {supra).
Visnu's placement in this type is also uniformly v?mac?ra. This theme was a product

[9] 283

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
 -
Fig. 7 Ekap?dam?rti, froma temple car, Thanj?v?r Dt.

-
Fig. 8 Kaly?nasundaram?rti, Dum?rlena, Ellora.

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
of theVijayanagara period and foundmostly in latermediaeval temples (Sastri 1916:
figs. 59-60). Inthe temple cars of Tamil Nadu, wood carved chariots used for
purposes, Ekap?dam?rti appears in a prolific measure. As many as 49
processional
specimenshave been tracedfromtheTamil car temples(Kalidos 1989: 81-82, 300-302).
Mostly Siva appears as the central figure but in two rare cases Visnu replaces the former.
Such imagesare found in theS?lihgapuram (NorthArcot district)andMadurai (K?tal
Alakar) temple cars. In these carvings Siva receives the v?mac?ra treatment.

Lingodbhava

A very popular Saivite iconographical form since the early mediaeval period, it
purportstoproclaimthegreatnessof Siva-Mah?deva inhis phallicaspectas Lingodbhava
which theme receives an elaborate treatment inmah?pur?nas such as the Linga Pur?na
(Rao 1971: II, I, 105-107). Earliest iconographical specimens are found in the
Kail?san?tha temples at K?nchipuram and Ellora, Vir?paksa temple at Pattadakkal
and cave temple at Tirumeyyam (Latha 1992). In most Cola temples since the 9th
century A.D. onwards Lingodbhava appears in a stereotyped pattern on the back kostha
of themain garbhagrha,best examples being theR?jar?jesvarams at Tanj?v?r and
D?r?suram. Iconographically the Lord, Siva, appears as Candrasekhara at the lower
half of an inaccessibleLinga whose head and footBrahm? andVisnu attempt to find
out. So Brahm? appears in the zoomorphic form of a swan and Visnu as var?ha 'wild
boar' at the righttop and leftbottom of theLinga respectively.Besides Brahm? and
Visnu in caturbhuja aspect, holding their characteristic emblems in para-hastas, appear
in human form also on either side of Candrasekhara, the p?rva-hastas being in anjali
bandha. Doing so, they pay homage to Siva and thereby acknowledge his greatness.
The classicalpiece, representingthe typology,is found in the leftkostha (on themain
doorway to the temple) of theKail?sa temple at Ellora. The specimenhoused in
theback kostha of theK?nchipuram temple (Fig. 9) is also remarkableand early in
point of time (6). Visnu and var?ha in all these representations receive the v?mac?ra
status. The Vaisnavas have no antidote to this form, excepting the two rare pieces
of Vaikuntham?rti and seatedNrsirhha of theN?makkal cave {infra).

Other Forms

Kaly?nasundaram?rti,theLord (Siva)who becomeswedded toUm?, is a popular

idiom in South Indian art since the earlymediaeval period of which thereare two

(6)However the earliest specimenof a human being (Candrasekhara?) appearingwithin theLinga

comes fromGudimallam (Basham 1971: pi. 17b).

[11] 285

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
colossal representations in the Elephanta cave and the Dum?rlena at Ellora. It was
a popular idiomwith theC?k bronze makers and appears inmost Siva templesof
the C?la, Vijayanagara and N?yaka periods in Tamil Nadu. The usual typology
represents Siva and Um? as the central core, the former taking the hand of the latter
andVisnu pouringthe sanctifying holywater (thisis accordingto theSoutherntradition)
on to theunitedhands of thedivine couple. The mythological sanction is thatVisnu
giveshis sisterinmarriage to Siva. What he does by pouring thewater iskannik?d?na
'gift of girl'. Brahma too appears in the panel as a priest, seated before the homa
kunda 'fire pit' as the ritual head, giving the S?straic sanction for the marriage. It
is interestingtonote thatUm? appears on the rightside of Siva inboth theElephanta
and Ellora pieces. Visnu is to the right side (and thus Siva receives a v?mac?ra treatment)
inElephanta (Berksonet al. 1983: pis. 51-52) and the left in theEllora cave (Fig.
8; Kalidos 1986c: 32) (7).
Tripur?ntaka is another form akin toKaly?nasundara in iconographical composition.
In this form, Siva appears as a warrior to topple the tripuras, the triple demons who
disguised themselves as mighty forts. According to mythology, Siva was assisted by
a host of the gods in his war against the demons. Thus the earth became a chariot,
Brahm? the charioteer, Visnu the arrow and Mount Meru (Rao 1971: II, I,
the bow
164-70). We have a classical representation of the theme in the Kail?sa at Ellora,

justbehind the facade on the left side of themain gateway. There is a dilapidated
image in the Shore templeatM?mallapuram which is thematicallyunique and finds
no parallel in other parts of India either in point of time or composition. Normally
Tripur?ntaka is a sthdnaka image but at M?mallapuram Siva is seated and holds the
What is strikingin thecarving
arrowand bow inhis rightand lefthands respectively.
is that Visnu appears figuratively in his usual caturbhuja aspect, representing the arrow?
head. Here Visnu gets the right-hand place, making Siva's position v?mac?ra. In other
illustrations Visnu is never represented figuratively. So the M?mallapuram piece is
simply unique in the context of our study.
In the N?makkal cave (no. II) there are two representations of Vaikuntham?rti
and seated Nrsirhha, both attended by Siva and Brahm? to the right and left
respectively. All are caturbhuja and hold their characteristic emblems in para-hastas.
Siva bears the mrga and n?ga. N?makkal was a stronghold of the Vaisnava sect and
it is interesting to find Siva appear on the right side of Visnu, thus making the latter's
position v?mac?ra (Fig. 10; Dehejia 1977; Kalidos 1986c: 470).

(7) Carmel Berkson (1983) and most other scholars consider the figurebearing thep?ma-kumbha
'waterpitcher' as Himavan but I am inclined to treathim as Visnu because in all the latermediaeval
Visnu as sankha-cakra-dh?ri pours the water. Others may point out that in the Elephanta
representaions
piece Visnu appears separatelyas seated upon thegaruda to the leftside of Siva. This kind of duplication,
i.e., one divine person appearing in various forms on a particular occasion, is possible in the Hindu tradition.
For example see Visnu as both Vaikuntham?rti and Nrsirhha appearingwithin a panel at N?makkal
(Fig. 10).

286 [12]

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
 - -
Fig. 9 Lingodbhava, Kail?san?tha temple, Fig. 10 Vaikuntham?rti, N?makkal Cave temple.
K?nchipuram.

From the above study what is to be generalized is that a number of iconographical

^^^^^^
forms, purporting to either syncretizing (e.g. Elephanta type of triunes or Ekap?dam?rti)
or bringing togethertheHindu Trinity into an entity (e.g. Tiruccir?ppalli typeof
Trim?rtis, Som?skanda and Kaly?nasundara) were being created since the early
mediaeval period. In formslikeHarihara, Visnu and Siva alone get fused.What is
important in these iconographical illustrations is the v?mac?ra place assigned to either
Visnu (these are numerically superior) or Siva. As in the case of Ardhan?risvara or
Harihara inwhich Um? and Visnu uniformly receive the v?mac?ra treatment, the trend
is not uniform in other themes. This is an important phenomenon in theHindu social
usage wherein v?mac?ra was treated as something low, mean and not worthy of being
accorded a commendable status of equality. So it is imperative at this juncture to
explain this phenomenon in order to bring out the social values of the iconographical
motifs under review.

Sociological Context

In theHindu sociological context v?mac?ra is synonymous with weakness, frailty,

meanness, baseness and degradation. It is generally believed that the power of a

[13] 287

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
man is on the righthalf and that of a woman on the lefthalf. That is why in
Ardhan?risvara the female is to the left (Kalidos 1993). The followersof Sakti as
the primordial and omnipotent power, called S?ktas, who took to the pancamakaras (8),
are called v?mac?ris. V?mac?ra is also known as kaulam?rga which has a derogatory
sense also in as far as the word kauleyaka means 'a bastard', kaulinam 'a scandalous
act', kaulika 'aweaver' (a low caste) (9),kauleyaka 'a dog' (Apte 1976: 166). All these
will standproof of the fact that any one who occupies the left side is equated with
the feminine principle or something low. It is also worth mentioning that in the
R?jar?jesvaram at Tanj?v?r the image of Vamadeva, the feminine part of Sad?siva,
appears on the leftkostha of the pradaksina-p?tha within the inner cloister of the
garbhagrha(10). Now, why inHindu social usage the left side is treated low and
mean? Of the twohands, it is customaryin India to use the rightone forall good
or auspicious doings; say forexample eating, giving a giftand so on. The lefthand
is used for lower usage; say cleaning the secret parts of the body after a call of nature
or for doing any activity that is supposed to incur impurity or pollution. Even a left?
hander in theHindu social circles or even immature children are not expected to perform
any auspicious activity, using the left hand.
Against this background, certain ideas got established in Indian courtly life and
social organization like caste. In mediaeval Tamil Nadu, for example, the armed flanks
were divided into valankai (right-handdivision) and itankaimah?cenai (left-hand
division). Those of the valankai group were superior in status. Caste organization
was also in the valankai-itankai pattern. Needless to add that the valankai castes enjoyed
a superior social status. For more details on this subject, which I need not elaborate
here, see Taylor (1857-62), Raman (1959), Rajayyan (1966), Mahalingam (1967),
Hunumanthan (1981), and Kalidos (1989).
If one views the thesis elaborated in the present paper against the sociological

(8) Followers of theTantric cult believe in the consumptionofm?msa 'meat',matsya 'fish',madya

'wine', mudr? (symbolic gestures), and maithuna 'sexual intercouse'. See Kalidos 1986a: 78; Mookerjee
1977: 185.
(9) Apte (1976) quotes a passage to the followingeffect:Kauliko Visnur?pena r?jakany?mnisevate
'the weaver who represents Visnu in form renders his services to spinsters of the royal family'. The

metaphorical equivalence of kaulika with Visnu is to treat him as a woman, maybe even a eunuch. This
is tomake him behave like a woman in the company of royalwomen folk so thatno untoward incident,
resulting in any sexual misbehaviour, occurs while performing his services. It may also be noted here
that theK?syapa N?nak?nda (pp. 283-85) calls theweaver by the caste name ayogava, born of a vais'ya
fatherand ksatriyamother which is a pratiloma typeof birth. The VisnusarhhitaandManusamhita also
make a note of ayogava, both of thepratiloma type inwhich case the father is a s?dra and themother
a vaisya, the prescribed avocation being artistic works and carpentry respectively. to the Hindu
According
dharmas?stras those born in the pratiloma were very mean creatures, treated as pancamas (the fifth
lineage
caste which was untouchable). For a detailed studyof the points elaborated above, see Kalidos 1984:
374-87.

(10) For a detailed study see Kalidos 1986b.

288 [14]

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
context, itwill be evident that there was an effort to treat Visnu as a v?mac?ra principle
whichmust have takenplace at the instanceof Saivaswho had a domineeringfollowing
in South India through the ages down to Vijayanagara. However, the effort in certain
circles was resisted by the Vaisnavas who tried to project Siva as a v?mac?ra principle.
This effortseems to have been feeble ifwe look into the numericalvolume of the
various analysed in the second part of our study. All these were due to
typologies
sectarian animosity and rivalry, the Saivas and the Vaisnavas making persistent attempts
to proclaim the superiority of their own creeds through iconographical illustrations.
However, a paradoxical proposition emerges when we take into account the
Vaisnava Tamil bhakti literature wherein Visnu's v?mac?ra status is acknowledged.
How did this happen? It appears that the Vaisnavas had some reluctance in treating
the left as a woman half. In support of this argument we may bring two points for
elaboration inwhich Laksmi, the feminine counterpart of Visnu, receives a right-hand
treatment: (1) the srivatsa symbol, a mole representing Laksmi or Sri, is carved only
on the right chest of the images of Visnu (n), and (2) the sanctum sanctorum, meant
forLaksmi of Sriwithin a Visnu temple is to be foundonly on the rightside of the
main sanctum sanctorum, meant for the Lord (12). So the Vaisnavas generally did not
treat very seriously the v?mac?ra status assigned to Visnu in iconographical forms of
syncretic nature. But the fewer examples, in which Siva is assigned a v?mac?ra place
would allow us to generalize that in certain circles there was a reaction which was
sporadic in nature and failed to produce any pan-Indian effect due mainly to the basic
Vaisnava ideology which approved of the right-side-woman concept. If viewed in this

angle, the righthalf in formslikeHarihara, Lingodbhava and Som?skanda is feminine

according to the Vaisnavas. In any case Visnu is supreme in his own way, no matter
what the Saivite treatment was. So also Siva was supreme in his own way, no matter
what the Vaisnava treatment was.

(n) Sridevi, theMother Goddess Laksmi, having her residence on the chest of Visnu, is alluded
to in a numberof hymnsof the?hr?rs. In such a formtheLord is called Ciritaran (Skt. Sridhara) (Periya
Tirumoli2.3.9, 4.6.2; N?cciy?rTirumoli2.1, 5.9), Tirum?rvan (Tiru 'Sri',m?rpu 'chest') (PeriyaTirumoli
3A.2, 3.5.2, Periya Tirumatal 1.148), and Mankaim?rpan (mankai 'woman') {Periya Tirumoli 9.5.3).
Periy?lv?r in his Tiruppall?ntu (v. 2) gives a specific indication to the right-sidehalf of the Feminine
Principle. The beautiful expression, noted for its rhythmictone, is vativ?ynin valam?rpinilv?lkinra
mankai, meaning thereby that the gracious damsel resides in the rightchest of the Lord. In another
place (PeriyaTirumoli 3.5.9) it is stated that theLord unites (kalantu,this termalso denotes sexual union)
with Sridevi inhis chest. I am of the opinion that there is an erotic content in thisparticular reference
which would mean that the Tantric approach in Vaisnava tradition is not v?mac?ra.

(u) In Siva temples the programme is the opposite, i.e., Devi shrine to the leftof theLord (e.g.
the T?r?curam temple); the Minaksi-Sundaresvara temple, Madurai, is an exception. A good example
is theNatar?ja temple,Citamparam, where shrinesfor the goddesses appear to the rightand leftof the
houses, meant forVisnu Sesas?yi and Natar?ja respectively (Younger 1986: fig. on p. 223).

[15] 289

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
 Conclusion

It may be added here that at times religious traditions and social usages come
into clash with each other which happened in case of the paradoxical treatment that
Visnu had to receive in respect of the v?mac?ra status in iconographical illustrations.
V?mac?ra as a derogatory factor was more a social phenomenon which percolated into
the sphere of religion and art. The Visnu-Laksmi (left-right) relation was a religious
tradition because the Feminine Principle is a vital force which activates the Male
Principle in all his doings according to the Sri Vaisnava ideology. So the Vaisnavas
since the time of the Alv?rs did not view with serious concern the v?mac?ra aspect
of Visnu in iconographical motifs and they even accepted it since itwas in tune with
their religious heritage and philosophical orientation. The fewer repulsive elements,
traceable in forms like Trim?rti or Trim?rtis and Tripur?ntaka are likely to have taken

shape when religious traditions and customary social usages came into conflict with
each other, thus providing a ground to argue that Siva is also v?mac?ra at times if
Visnu should be treated as something low on this basis.

REFERENCES

Dictionary. Delhi.
Apte, V.S. (1976) The StudentsSanskrit-English
Basham, A.L. (1971) The Wonder That Was India. Calcutta.
Berkson, C. et al. (1983) Elephanta, theCave of Shiva. Delhi.
Dehejia, Vidya (1977) Namakkal Caves. Bombay.
Desai, K.S. (1973) Iconographyof Visnu. New Delhi.
Devat?m?rtiprakarana, ed. V.M. Sankhya Tirtha, Calcutta 1936.

Dutt, M.N. (1978-79) The Dharamsastra. 6 vols. New Delhi.

Grossato, A. (1987) 'ShiningLegs'. The One-footed Type inHindu Myth and Iconography. EW, 37,
pp. 247-82

Hanumantha, K.R. (1981) Untouchahility in Tamil Nadu. Madurai.

Kalidos, R. (1984)Varna-sarhkarainAncient India. The QuarterlyJournalof the

Mythic Society,LXXV, 4.
Kalidos, R. (1986a) Tantric Buddhist Interactionwith theDevelopment ofHindu Iconography inTamil
Nadu. Tamil Civilization, IV, 1-2.

Kalidos, R. (1986b) The Hindu Temple as a Symbol ofPancakrtya. Proceedingsof the45thAnnual Session
of the IndianHistory Congress. Delhi.
Kalidos, R. (1986c) Iconographyof EarlyMediaeval South India, vol. I: Vaisnava. Thanjavur. MS.
Kalidos, R. (1988) Tamil Interactionwith Ellora. Ellora Caves: Architectureand Sculpture,ed. Ratan
Parimoo et al. New Delhi.

Kalidos, R. (1989) Temple Cars ofMediaeval Tamilaham. Madurai.

Kalidos, R. (1993) The Twin-Face of Ardhan?ri. ActaO, LIV, pp. 68-106.

290 [16]

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions
K?syapas Book ofWisdom (K?syapan?nak?nda),ed. & transl.T. Gaurian, The Hague 1965.
K?syapasilpas?stra, ed. & transl. K.S. Subrahmanya Sastri, 2 vols., Thanjavur 1960-1968.

Kramrisch, S. (1946) The Image ofMah?deva in theCave Temple on Elephanta Island. Ancient India, 2.
Kramrisch, S. (1984) The Great Cave Temple of Siva in Elephanta. Discourses on Siva, ed. M.W. Meister.

Bombay.

Latha, V. (1992)Art ofTirumeyyam.M. Phil, thesis,Dept. of Sculpture& Art History, Tamil University,
Thanjavur.

Mahalingam, T.V. (1967) South Indian Polity. Madras.

Mah?bh?rata, ed. Bhandarkar Oriental Institute, Poona 1927-64.

Manusamhita, see Dutt 1978-79.

Matsya Pur?na, Anandasrama Sanskrit Series, Poona 1907.

Mookerjee, A. & Adhu Khanna (1977) The TantricWay: Art, Science, Ritual. Boston.
N?l?yirativiyapirapantam (N?l?yiram), ed. V.N. Devanandan, Madras 1971.

N?nmukan Tiruvant?ti, see N?l?yiram.

Periya Tirumoli, see N?l?yiram.

Rajayyan, K. (1961) Administrationand Social Life in theCamatic. Tirupati.

Raman, K.V. (1959) The Early History of theMadras Region. Madras.
Rao, T.A.G. (1971) Elements ofHindu Iconography.Varanasi (repr.).
Sahai, Bhagawant (1975) Iconographyof theMinor Hindu and BuddhistDeities. New Delhi.
S?rasvatiyacitrakarmas?stra, ed. & transl. K.S. Subrahmanya Sastri, Thanjavur 1960.

Sastri,H. Krishna (1916) Iconographyof Sad?siva. New Delhi.

Silparatna, ed. & transl. Devanathacharya, Thanjavur 1961.

Taylor,W. (1857-62) Catalogue Raisonnee ofOrientalManuscripts in theLibrary, (Late)College, FortSaint

George. Madras.

Tev?ram, ed. P.A. Sadasiva Chettiyar, 2 vols., Madras 1973.

Tiruv?ciriyam, see N?l?yiram.

Tiruvant?ti, see N?l?yiram.

Tiruv?ymoli, see N?l?yiram.

Visnusamhita, see Dutt 1978-79.

Younger, P. (1986) The Citamparam Temple Complex and Its Evolution. EW, 36, pp. 205-26

[17] 291

This content downloaded by the authorized user from 192.168.72.229 on Thu, 22 Nov 2012 17:23:46 PM
All use subject to JSTOR Terms and Conditions