Karana Sculptures, Dance Practice

and the Niityasastra in Indonesia

ALESSAN DRA IYER

The Reliefs of Candi Siwa

n the village of Prambanan, aboul15 kilometres fro m Yo gyakarta, the capital

I of the Special Region of Yogyak arta, in Cent ral Java, there is a magnifice~(
temple complex of Saiva denomination . Th e co mplex is one of the largest ID
the Hindu world. Only part of the complex has been restore d, that is, the temples
in the inner court. In its heyday-it is now thou ght it belongs to the ear ly part of
the 9th century A.D.'-it must have been a most impres sive sight . It is made up of
a serie s of co ncentric courts which contained some 224 minor pariwara, sur-
round ing temples, and eight temples in the inne r court. Amo ng the latter there are
three tall structu res in the middle of the inner court. T hese are ca ndi (tem ple) Siwa
in the centre (Fig. I). to the right (of candi Siwa ) candi Brahma and to the left
candi Wi snu, so named because the main statue s at one time in worship represent
these god s. Candi Siwa is more popularly known as candi Loro Jonggrang, after
a legendary local princes s iden tified with Durga, whos e statue showing her in the
Mahisasurarnardini aspect is enshrined in the temple in the nort h-facing chamber.
All the ~hree main tem ples are rich in iconography. The re are free-standing
image s of Siwa, Durga, Ga nesa , Bhatara Guru (Agastya) , Wi sn u and Brahma-:
the first four belonging to candi Siwa alone. The inner side of the balustrade of
candi Siwa and candi Brahma are covered wit h reliefs de pic ting the .:;lory of
Rama and on the inne r balustrade of candi ~i~l)u one ca n see the Kre~l)aY3l)a
story recounting ep isodes from the life of Kresna . The outer wa lls of all the three
candi are covered with a vari ety of ornamental motifs sho wing flowers, ani m~s,
and m inor cele stials. But a mos t important and often ove rlooked feature perlaIns
exclusive ly to candi Siwa. On the outer b alustrade there are 62 reliefs showing
dancers and musicians , together with 70 oth er relie fs showing celestial beings in
gro ups of three and in a standing posture (Fig . 2).
The 62 dance panels have been the foc us of my own in-depth study and I have
bee n able to ident ify eac h one of the m as a representation of a karana mo vemenL
A karana is a small movement sequence wh ich forme-the basis of a technique of

Song." Natak No. 119 January - March 1996

 Fig. I. View of candi
Siwa, Prambanan.
Photo: John' s Photo

:
-...• ',;
.. Fig. 2. View of karana
.. ---: ..
reliefs and groups of"
apsarilwidyiidham in
niches capped by ratna .
" - ,--." . Photo: author
 26 ALESSANDRA IYER

dance described in the Nii!yasiistra' , the foremost work on dance, music and'
drama known from India. i am not the first to suggest that the dance reliefs ,£
Prambanan are karana' but by virtue of this recent research work it is the first time'
that all the reliefs ha~e been identified as karana on the basis of movem ent analy-]
sis and comparison with related material fro"m India. Indeed the identification;
" would not have been possible without the previous research on the karana seen
on South Indian temples carried out by Padma Subrahmanyam', whose recon-i
struction I have used as model (see Fig. 3 and Fig. 4 for examples of karana:,
reconstruction). 1
In this article my objective is to touch upon issues which are an integral partof
my research and are of more general interest. Th is means that for reasons of space:
it will not be possible to discuss the karana of Prambanan from a technic al point.
of view, nor will it be possible to go into the details of how the complex was,
restored and what went wrong during this process. The sequence of reliefs seen
on the temple today is definitely not the original one and it could well be that
those reliefs showing only drummers may have had a separate placing and that the
apsarilwidyadhara groups in niches did not belong there at all. There is also the
question of identification of the performers which, at this stage, cannot be brought
to a definitive conclusion'. However these questions cannot be dealt with here and
I shall have to gloss them over, reserving to write about them in a different con-
text",
Researching the dance scupltures of Prambanan involved more than a melt
identification of dance movements. It raised questions relating to history and its
perception. It involved a critique of South-ea st Asia as the 'Orient' of India, an
idea that still has wide currency, as well as a critique of the fragmentation of much
research carried out within artifical geographico-political boundari es. Historical
perspectives need re-evaluating and this will require the concerted effort of joint
scholarship. I do not claim to have all the answers; all I can do at this stage is ask
questions and offer possible lines of enquiry to test well established beliefs.

Dance and the Niityaiiistra in An cient Java

The most important issue is why karaTJa sculptures should be found in JaVa.
Can the Prambanan series tell us anything about dance performance in ancient
Java? What is the relation ship between performance practice and the visual arts?
Is the Niuyasiistra of any relevance to the Javanese context? "
In the art of the post-Central Javanese period Wayang Kulit (shadow theatre),
sculpture and painting mutually influenced one another'. Old Javanese literatute
suggests that there 'are a number of correspondances between literary and plastic
art tradition s, for example , in the way deities are repre sented'. So it seems rea·
sonable to suppose that such art interre lationship existed earlier than the post.
Central Javanese period . "
It is self-ev ident, by the very nature of karana, that unless the dance technique"
 Fig. 3. Candi Siwa, relief 14,
Karana Bhujan gatriisita
(no . 24) . Photo: author
The hands in abha va and
varada in Fig. I (left of
viewer) show the begi nning
of the basta see n in the Devi
figure in the drawi ng. The
central figure has me right
ha nd gracefully stretched on
the left and the left hand
near the head . The right foot
is at the front in kuh cita
poda . The foot is damaged.
The third fig ure is also q uite
damaged. Only the chest is
visible and part of the legs.
The right hand is in
ham scpaksa near the chest.
The move me nt performed by
this figure equates the very
beginning of the karana
sequence . Compare this
relief with Fig. 4.

Fig. 4. Bhuj ahgalrasita ka tana. Author' s copy of preliminary drawing for Satara relief by
Tirugnanam. Devf's movement follo ws that of Fig . I in the Pramban an relief an d Siva's
I S close to the central fig ure . De vf s movement is performed on the other side.
28 ALESSANDRA IYER

seen in the reliefs had been in pract ice they could not have been sculpted
Sculptors would have had to have seen karana performed to create their likeness
in stone. A karan a cannot be represented if its dynami cs are not known becaus
each karana is not a pose but a small sequence of movement. Any portion of tha
small sequence can be shown through a static medium, allowi ng for multip'~
visual interpretations. The Prambanan reliefs are the only kata na series in Java;
the existence of individual karal.la-inspired sculptures elsewhere is conclusin
evidence that the karana dance technique was known . These are found in Centr:ll
Java and in others areas, including Sumatra, and consist of dance scenes or pet>
pIe in a dance-rel ated altitudes'. If the movement captured in these sculptures ~
analyzed it can be clearly identified as a ka rana or its sub-unit, the ciiri, move
rnent of the lower limb.
Dance forms not based on karana were also practised. At Borobudur, the
famous Buddhist temple in Central Java that dates fro m the 8th century A.D., dif
ferent dance forms appear in the reliefs. The sculptures of ' Indian' dance move-
ments show the dance that would have been practised by the Javanese elite".
There is no dispute that in ancient Java there was a strong assimilation of
Indian cultural elements and Indian ideology, particularly in relati on to ideas of
kingship, in the court environs. There is also a degree of consensu s that Indian
~class ic al Hindu) dance was performed in ancient Java".
The first example of the Niityasiistra text was rediscovered in north India jll~
over 150 years ago". It is a sima, that is, it gives the rule s that govern (in this
case) performance-music, dance and drama conceived as a whole. Drama is seen
as an act of religious worship, the equivalent of a sacrifice. The text dwells upon
the divine origin of drama. In its fourt~ chapter it describes ka rana which arc the
elements of the cosmic dance of Siva. Siva teaches Tandu the karana so that they
may be used for the purpose of bestowing the same f~it as that o'f yajiia, sacri-
fice; this gives a metaphysical dimension to each karana" . We do not know for
sure when the Niityasastra was compiled, nor is the exi ~tence of a written text so
important in terms of transmission and teaching of its content. The tradition it
embod ies is known as miirgi in post-Niityasiistra times-literally. ' of the high
way' . though this nomenclature should not be mistaken to imply any type of hier-
archy". The NiiD'asiistra did not create the tradition anew. Th e tradition grew over
the centuries before it was finally systematized. In all likelihood it embrac~
material contri~uted from different areas. The Natyasastra take s stock of what11
,,!ready hap~mn~ and form ulates a performance code and a system for the nnaly'
SIS and cl':5slficatlOn of dance movements, which are seen to be an integral part
of dramatic performan ce. Everywhere the sastra provided a clear refere~
fra~ework yet miirgi also varied greatly in terms of v rtti, modes , and p rol'[flt,
reg~onal .styles, showing a dynamic relationship with desi , literall y local or
regional, The latter were the forms and styles which were not codified in the
Niityaiiistra but had their own independ ent codifications, as Bose points out".
 KARAJ':lA SCULPTURES AND NATYAs As TRA IN INDONESIA 29

This way every region and area lent a different flavour to milrgi dance and the
NilD'asilstra itself records this by listing the modes of the North, South, West and
East and describing briefly the regional preferences. The keywords to understand
this complex situation are multiplicity and coexistence.
The geographical content of the Niitya sds tra deserves a separate study. The
text claims the status of a fifth Veda, i.e. Niuvaveda, as well as of siistra, and
specifies the application of its rules in the Saptadvipa, 'seven continents' of the
earth, The geographic divisions of the Niitya iii stra are in terms of directions
(East, West, South, North). Some specific areas are mentioned but their identifi-
cation is fraught and the list is by no means exhaustive". Trade routes had criss-
crossed the Indian subcontinent and South-east Asia (maritime and overland trav-
el) and had been active long before the beginning of the Christian era". An aware-
ness of contemp orary geographical knowledge would, quite likely, be displayed
in encyclopaedic works such as the silstra , One wonder whether the Saptadvipa
is the subcontinent alone or a wider geographical area.

The Katana technique: lndia~ or Asian ? ,

The karana technique of dance was the foundation of the miirgi tradition but
the choreographies- the way the katana were used-and even the 'actual num-
ber of karana seen in every region was markedly different. Anukarana, imitation,
is the concept behind katana . Thu s karan a often imitate animal movements or are
inspired by natural phenomena". There are 108 nrtta katana which constitute the
dance of Siwa. There are 32 main combinations of nrtta karana known as
angahiira, which make use of only 89 karana. There are also many more karana
based on combinations of ciiri, that is, movements of the lower limbs". Ma rgi
seems to have been pan-Asian dance. We find karana sculptures on Khmer and
Cham temples, at Angkor and at Trah Kieu (Fig. 5). Trace of karana technique
and its sub-units can still be seen in the living performing arts of Asia, although
their practitioners do not call the movements katana. They can be see for exam-
ple in present-day Thai and Khmer dance", In Javanese coun dance as practiced
today, one can see the use of three Niuyai iistra sthiina in the female Putri and in
male Alus and Gagah style. These are the iiyata, vaiwava and vaiiakha sthiina (a
description of these sthiina is in NS 11 ,51 -62; 13, 161). These observations can
be best verified thro ugh a practical demonstration.
Present-day traditions of dance from the Indian subcontinent are also some-
what removed from the katana technique , which makes little use of foot-stamp-
ing-unlike such living traditions. Traces of karana can be seen in all Indian
dance styles but they are just traces, of which not even the name is remembered.
For example, the Manipuri style, from the state of Manipur in eastern India, uses
a movement taxonomy that is clearly local in terminology. But some nrtta hasta,
the dance actions for hands, wrists and arms (not to be confused with the denota-
tive hand gestures used in all Indian styles and popularly known as mudriiy which
30 ALESSANDRAIYER

are a component of the nrtta karana are still practised".

There are hundreds of scholarly works on the Niuyasastra and its origin. It is
regrettable that the Niuyasiistra section on karana and its sub-units has not
received much attention (with a few notable exceptions). compared to other por-
tions of the text. There is still great controversy over the date of the work. Wheo
examining the complex issue of dating the text it may help to take into account
sculptural evidence from all over the subcontinent together with the South-east
Asian sculptural material for further clues. If one looks at the dance sculptures
appearing all over the subcontinent one notices that from the 5th/7th century to
roughly the 12th/13th/14th (depending on the area) sculptural representations of
dance seem to follow very closely the karana technique which would indicate that
the technique was definitely practised during that period. There is a true attempt
at conveying karana movements through animation and at indicating the move-
ment path through the use of a number of devices. This is also apparent in South-
east Asian examples. After this time one still sees in India dance-related attitudes
and poses but they are crystallized into iconographic conventions of a static
nature. In the centuries prior to the Gupta period one sees a great number of
attempts aimed at showing dance through sculpture but these do not seem to fol-
low the karana codification too closely although they certainly bear a relationship
with it.
This earlier movement material in stone seems to be prefiguring the codified
movements. not quite coinciding with them. It is not possible to use evidence as
proof of a Gupta date for the Niuyasastra-unese can never be such a thing as a
proof. given our present state of knowledge-but the sculptural material certainly
points to the fact that the practice of a dance closely following the Niuyasiistra
condification seems to have been faithfully recorded in stone from Gupta times
onwards. In this connection Vatsyayan writes: "The earlier examples of the rep-
resentation of Indian dance in sculpture up to the Gupta period can be analysed
only generally in terms of the movements of the feet, hands etc. Only the later
examples can be analysed in terms of the appropriate ciiri and karana?", This
seems to support the view I am putting across.
Another point to be considered is that katana consist of such a variety and
range of movement patterns that one cannot but wonder whether some of these
movement patterns could have originated in areas outside India entering the sub-
continent prior to their codification in the Niuvasiistra, only to be relayed outside
India at a later stage. After all. mterborrowing of movement patterns is a very
common practice that occurs and recurs in the history of dance all over the
world".

Karana Sculptures and Cultural/Artistic Recycling

An important and very puzzling fact is that the Prambanan karana series is
about 100 to 200 years older than the first known karana series in India, which
Fig 5. Tra Kieu, Vietnam, Karana vrtcikam no. 47. Photo: Hilary Smith

Fig. 6. Karana Yricikam. Author's copy of preliminary drawing for Satara relief by
Tirugnanam.
32 ALESSAl'DRA IYER

can be seen at the Cola Brhadi svara temple in Tanj ore . Thus there does not seem
to have been prototypes or samples for the karana of Prambanan-the images
wer e entirely a local concep t. Would it be far-fetched to speculate that the ideaof
representing katana in a serial order in Cola times was recycled into India after
Prambanan was carved? This does not me an that the concept of katana was not
known in Indi a before Pramb anan nor that the Niuyaiastra is of Javanese origin.
It also does not imply that katana co uld not ha ve already been a pa rt of the dance
practice of Cola times prior to their rendering in stone o n the Brhadisvara. I am
o nly saying that the idea of putting karana in a sc ulpted se ries m ay have origi-
nated in Java " . I wou ld ca ll this an instance of recycling. The inevitab le question
is: were there other instances of recy cli ng?
Most resear ch on South-east As ia since the 1950s has sought to re-evaluate the
indigenou s element, reversing earlier scho lars hip wh ich searched excl usively for
the Indian and/or Chinese influence, whi ch saw the indige no us cultures as mere
passive rece ivers" . This different way of looking at histo ry has also resulted how-
eve r in a po larity bet ween the ca tegory of foreign (in th is in stance Indian) and
indigeno us, viewed as fixed entities. But wh at is or is not indigenou s is never so
clear-cut and static.
Th e whole question of Ind ian influence, wh eth er it is the now outmo ded con-
cept of Indianization or the more modern one of locali sation of Ind ian influence.
needs rethinking. The travelling of people- and of ideas-must have bee n a two-
way traffic. It would be misleadi ng to th ink that So uth-East Asian Hindus depend-
ed entirely o n Indian input. Present politi cal bo undaries cannot be projected on to
pas t history . Th e cultu ral recycling and relaying that took pl ace amo ng the
Hind uized south-east Asia n regions"'-here Hindu ized refers to both Hinduism
and BU dd~i sm~may well have involved India . The ph eno me no n of cu ltural my-
.cling within ASia shou ld thus be investigated fu rther. It may yield some surpns-
mg results.

Evidence f rom Old Javanese Literature

. It may ~e objected th at ~cause no co py of the Niuyaiastra h as eve r been found
in Indonesia, It IS most unlikel y th at this text was kn own there, T he fact that there
is no trace of a cop y does not ho wever mean th at there never was one. But. as I
have alrea dy pointed out, Hindu sastra were usually orally transmitted and there-
fore the presence or ~bse~c~ of a written ver sion is perhaps only signi ficant to the
Western hi storian, Lingui stic and literary evidence supports the view that there
was knowledge of the Niityai iistra in ancient Java.
Th e Old Javane se langu age has words such as nrtta to refer to dance, rasa to
ref~r to fe~l in!l and bhii.wa wh ich may refer to ei th~r temperament o r manner of
ac tIng- thIS links up with abhinaya whic h amo ng other meanings is glossed as
'dramatic:" , These. are k~y word s in the Niitya iiistra. Admittedl y this mig ht ~Ol
be seen as conclusive eviden ce , However, in the Old Jav anese text Nawanal)G,
 KARAt:/A SCULPrURES AND NATYASASTRA IN INDONESIA 33

composed in the 13th/14th century A.D •• we find a passage relating to music and
dance in which variou s musical instruments, among them a drum called mrdang-
ga , are listed and activities such as gitiinrtta are mentioned". Nrtta is the dance
based on karana; as we know fro m the Niityasiistra. gi fel is song or singing. Siva
talks of both gitii and nrtta when referring to his dance". The word karanii is also
ment ioned in the Old Javanese text and from the context it can be interpreted in
the technical sense of dance", The Nawanatya text is not the Niitya iastra trans-
lated into Kawi. But Niitya iiistra technical terminology relating to dance and
music in a work of Majapahit times suggests a prior knowledge of that tradition
with traces of that knowledge reflected in the contemporary vocabulary of music
and dance.
The text is about the ways of ex pression of the court ier, said to be nine. Robson
and Zoetmul der do not feel that the word natya has any connection with the
Sanskrit homonym because it is written with a dent al l. In this they follow the
view of Pigeaud who translated fragmen ts of this manu script". This interpretation
is however unsatisfactory. Pigeaud takes natya to mean physiognomy. which is a
rather co mplicated translation . If natya appears in the manu script with a dental t,
this is not sufficient proof of its bei ng unrelated to the Sanskrit niuya , as indeed
on many occasions borrowings from Sanskrit do not follow the original spelling
and variant spellings for several words appe ar all through the very manuscript in
question. .
The translation "physiognomies" is not followed by Kern and by Poerbatjarak a
and Slarnetm ulj ana, who prefer "dance of nine" and " nine dances" respectively".
The work read s at times like some sort of acting manual, with specific instructions
on how to direct the g lance. sit. stand. etc. Perhaps life at co urt as described in this
work followed such an elaborate etiquette that any real distinction between the-
atre and life became somewhat blurred. with stylised behavior as the norm. But
Nawanatya may have really been a text important to actors in terms of character-
ization. The nine na tya could well be nine basic situations to be enacted through
dance and song. Niigaraketagama refers to nawanatya in connection with the
Raker play and thi s would support the view of its being concerned with nine basic
situations for enactment" . The Nawanatya text ought to be critically edited and
translated in full. as the translation found in Pigeaud is incomplete.

Conclusions
The dance sculptures found on the outer balustrade of candi Siwa, belon ging
to the 9th century temple complex of Pra mbanan in Central Java, can be identi-
fied as karana of the Niuyaiastra, the Indian text on dance and drama. The karan a
series of Prambanan is 100 to 200 years earlier than any other equivalent repre-
sentation found in India. There is therefore a distinct possibility that the idea of
representing karana in a series could have been recycled into India after
Pramban an. This raises the issue of wheth er there was other such artistic/cultural
34 ALESSANDRA IYER

recycling. Further investigations are necessary which may change our present per'
ceptions.
Although so far no Niuyasiistra manuscript has been found in Indonesia, both
the sculptural and literary evidence suggests that the dance and drama tradition
embodied in this text was known here. Transmission may have been entirely oral-
ly based. Keywords from the Niityasiistra can be found in the Old Javanese WOIT.
Nawanatya, in passages relating to music and dance. The text awaits a new criti·
cal edition and translation. 0

NOTES AND REFERENCES

(In this article the sound 'w' is rendered with 'v' when referring to an Indian context, with 'w' when
referring to a Javanese context following the conventional transliteration of Sanskrit and Old
Javanese.)
1. See E. Roy Jordaan, Imagine Buddha in Prambanan: Reconsidering the Buddhist Background
a/the Loro Jonggrang Temple Complex, Leiden 1993.
2. The Sanskrit work Niiryafiistra is ascribed to the sage Bharata. Its date cannot be fixed with
any certainty. Scholars vary in their opinions but it is generally thought that the upper and
lower limits would be 1st c. B.C. and 4th c. AD. Bose prefers a 3rd c.A.D. dating and Nagar
is prepared to considers a 5th or 6th c.A.D. dating. Dr Prem Lata Sharma, at present working
together with Professor K.D. Tripathi of Benares Hindu University in Varanasi on a new edi-
tion and translation of the Niityasiistra, under the auspices of the Indira Gandhi National Centre
for the Arts, is of the opinion that the precise dating of the Niityaiiistra is impossible to o~tain
for a number of reasons, including the important fact that the siistra was orally transmitted
from guru to disciple. This way the tradition was augmented and refined over a long period of
time. The Niityaiiistra cannot be fully understood without reference to the commentary writ-
ten to it by Abhinavagupta in the l lth century (Abhiniivabharat!). This Is a most difficult and
somewhat corrupt text but extremely precise in terms of reference to sources now entirely lost
to us, with extended quotations of other commentators' views (Prem Lata Sharma, personal
comment 1995). Also see M. Bose, Classical Indian Dancing: A Glossary, Calcutta, 1970;
R.S. Nagar, Niityasiistra of Bharatamuni with the Commentary Abhinavabhiiraii, Vol. 1-4,
Delhi 1988 (2nd edition).
3. Previous notices of karana at Prambanan are in Oudheidkundige Dienst 1937-48 Verslagen,
The Hague; K. Vatsyayan, 'Dance Sculptures of Lara Jonggrang (Prambanan)', Quarterly
Journal of the National Centre/or the Performing Arts, VI, 91,1977, and E. Sedyawati. 'The
Question of Indian Influence on Ancient Javanese Dance', RiMA 16, 2, 1982. The
Oudheidkundige Dienst identified the sculptures as representations of ahgahiira (sequences
~a.de up ~f se~era1 ~rafJa) without however providing an adequate analysis and without spec-
ifying wh~ch IS which; Vatsyayan remarked on the presence of the urdhvajiinu karafJa n. ~~
and the WIdespread use in South-east Asia of the knee-bent position; Sedyawati lists the c~n.
~een in the reliefs w~thout mentioning karana as such and without specifying where each can
IS seen. In my thesis I gave a reconstruction and a classification of karana based on the
Niiryasdstra couplets but this is now superseded by my more recent research.' See A. Lopez Y
Royo-Iyer, 'Dance Iconography and the "Indianization' of Southeast Asia, with SpeCial
Reference to Ancient Indonesian Temple Sculpture' (unpublished Ph. D. thesis, SOAS,
University of London, 1990).
 KARA~A SCULPTURES AND NATYASASTRA IN INDONESIA 3S

4. Dr Padma Subrahmanyam. Director of Nrithyodaya, Madras, is a leading Indian dancer.

Originally trained in Bharatanatyam, she worked under Professor T.N. Ramachandran.
Director of the Archaeological Survey of India, in the 19605, on the problem of identifying and
reconstructing the karana seen at the Brhadisvara (Tanjore), Sarangapani (Kumbakonam) and
Siva Nataraja (Chidambaram) temples in Tamil Nadu. The reconstruction took some 25 to 30
years to be finalized, involving a thorough study of variant manuscripts of the Niiryasiistra.
Abhinavagupta's commentary, and all the katana sculptures found in India for the purpose of
cross-reference. Unlike other reconstructions, based entirely on translations of the relevant
sloka of the Niuyasiistra, Dr Padma Subralimanyam's reconstruction involved accurate testing
and recreation of the movement sequences from a practical point of view, to eliminate inaccu-
racies based on anatomical misunderstandings. Thus her karana reconstruction can be a refer-
ence tool and model for comparing karana representations wherever these are found. See
Padma Subrahmanyam, 'Karana in Indian Sculpture', (unpublished Ph.D. thesis, Annamalai
University 1978).
S. My view is that at least some of these reliefs would show Siwa performing his tiindawa. An
inscription from the Ratu Boko plateau referring to a dancing Siwa seems to confirm that the
concept of a dancing Siwa was not unknown in Java, nor can the fact that the images are here
seen with only two arms be considered as a serious objection, as two-armed Siwa images are known
from India, the rest of South-east Asia and Prambanan itself. See]. de Casparis, Praiasti Indonesia
II, Bandung, 1956, and R.E. Jordaan and Edi Sedyawati, 'Some Iconographic Notes on the Siwa
Temple at Prambanan, Archipe/40, 1991.
6. See my forthcoming article, 'Prambanan Revisited: a Fresh Perspective on the Dance Sculptures of
Candi Siwa', Bulletin de l'Ecole Francaise d'Extrime Orient, Paris 1996.
7. This view is expressed by E. Sedyawati, 'The Dramatic Principles of Javanese Narrative Temple
Reliefs' in B. Arps (ed.) Performance in Java and Bali: Studies of Narrative Theatre Music and
Dance, London, 1993; Soedarsono, Wayang Wong, the Stare Ritual Dance Drama in the Court of
Yogyakarta, Yogyakarta, 1984; and G. Marrison, 'Balinese Classical Painting: its Literary and
Artistic Themes', Indonesia Circle, SOAS 65,1995.
8. See E. Sedyawati, 'Iconographical Data from Old Javanese Kakawin', Arkeologi II (I) 1978.
9. Dance sequences can be seen at the Biaro Bahal in Sumatra and people in dance-related attitude are
found in the iconography of all the Central Javanese temples as well as in some statuary from east
Java and Bali. See A. Lopez y Royo-Iyer, 'Dance iconography. . . . ' (Ph.D. thesis, SOAS, 1990).
10. See E. Sedyawati, 'The Making of Indonesian Art' in J. Fontein (ed.) The Sculpture of Indonesia,
WashingtonlNew York 1990: 108;
11. Such consensus is by no means unanimous. See Clara Brakel, 'The Court Dances of Central Java
and their Relationship to Classical Indian Dance'. Archipel 11, 1976, and Felicia Hughes-Freeland,
'Javanese Visual Performance and the Indian Mystique', The Art and Culture of Southeast Asia,
Lokesh Chandra (ed.), New Delhi 1991.
12. The first notice of the Niuyaidstra was by Sir William Jones in the 18th century. Through the work
of Regnaud and Grosset in the 19th century, Sanskritists became aware of the importance of this
text. The fust complete edition of the text was that of Sivadatta and Kasinatha Panduranga Parab in
Bombay, 1894. This was followed by the Varanasi edition of Batukanath Sharma and Baladeva
Upadhyaya in 1929. A major breakthrough was the discovery of the commentary on the Niuvaiastra
by the 11th-century scholar Abhinavagupta. This was first brought out in 1926 together with the
Niityasiistra in the Gaekwad Sanskrit series. Abhinavagupta's work has not yet been translated in
its entirety, only portions of it are available, as well as summaries. The Niityasastra's standard trans-
lation is that of Manmohan Ghose in 1951, in two volumes. The chapter on karana (Chapter 4) was
translated in 1936 by Naidu Naidu and Pantulu with the title Tli.fJ4avalalqanam. See K. Vatsyayan,
The Natyasastra: A History of Criticism' in Anna Dellapiccola (ed.) Shastric Traditions in Indian
36 ALESSANDRA IYER

Arts. Stuttgart 1989.

13. This corresponds to the subject matter of Chapter 4 of the Niityaiiistra. It is interesting to note thai
the concept of performance as yajiia is very important in Hindu Balinese culture as highlighted by
Ramstedt in 'Traditional Balinese performing arts as Yajnya' in B. Arps (ed.) Performance inJa'IfJ
and Bali, London 1993.
14. It must be reiterated, for clarity, that the Niityasiistra does not mention either miirg"i or its opposite
deii (roughly equivalent to 'local'). This terminology seems to be a later one. See M. Bose,
Classical Indian Dancing: A Glossary, Calcutta, 1970:9-11.
15. See Bose ibid.
16. See NS 14,36-46.
17. See I. Glover and B. Syme, 'The Bronze Age in Southeast Asia: its Recognition, Dating andRecent
Research', Man and the Environment, 18,2, 1993.
18. This is apparent when their colourful names are considered. Karana n.38, 'bee', is a fast turn
resembling the flight of a bee, the 'scorpion' group of katana are inspired by the movement of
a scorpion's tail, karana n.68 imitates the play of the elephant and karana n.70 imitates the
flight of the Garuda bird. Nos. 74 and 80 imitate the landing of the eagle and the sporting of
the peacock respectively, Nos.89 and 90 show the lion playing and hunting its prey and soon.
19. A karana is defined as being made up of three elements: sthdna, posture, ciiri, movement of
the lower limb, and hasta, hand gesture. There are 32 ciiri divided into earthly and aerial move-
ments. That is, the way the lower limbs move involves several types of jumps and skips.
20. In Khmer dance the kbach (variously translated by practitioners and roughly equivalent10
cadence) is most reminiscent, conceptually, of the karana, which is also a cadence of move
ment. Like the karana. kbach imitates animals or forces ~f nature. In Thai dance there are sim-
ilar cadences which constitute the "alphabet of Thai dancing". See Dhanit Yupho, Tht
Preliminary Course of Training in the Thai Theatrical Art, Bangkok 1956:26-28.
21. A comparative analysis of nrtsa hasta and Manipuri equivalent gestures can be seen in V,
Balakrishnan's documentary film Bharatiya Ndtya Siistra (Doordarshan 1992).
22. See K. Vatsyayan, Classical Indian Dance in Literature and the Arts, SangeetNatak Akademi.
Delhi 1968:270.
23. For example, during the Tang dynasty in China, a great number of artists from Central Asia
went along the silk route. Some were slaves, others (like the Tashkent dancers) simply trav-
elled in search of riches. Many Tang poets celebrated the beauty of the green-eyed Westen
dancers who sang and danced in wayside inns. Several Chinese dances of the Tang period sholl'
a remarkable Central Asian origin, in the name of the dances and in the type of costume
dancers were supposed to wear. Likewise, in more modem times it is worth remembering that
Fokine's ballet Cygne exhibits dance movements borrowed from Siamese dancing, whichhe
saw at the court of the Tsar where a troupe of Siamese dancers had been sent by the king of
Siam. See N. Savarese, Teatro e Spettacolo tra Oriente e Occidente, RomaIBari 1992: 33; 397,
24. I am indebted to Dr Padma Subrahmanyam for this insight which she first put forward inher
1993 paper on the dance art associated with the Brhadisvara temple, presented in Thanjavurat
the Seminar on the Brhadlsvara.
25. See J. Wisseman Christie, Theatre States and Oriental Despotism: Early Southeast Asia in tht
Eyes of the West, Hull 1985, for a discussion of Indianization theories.
26. See Wang Gungwu, 'Introduction' in D.O. Marr and AC. Milner, Southeast Asia in the 9thl{/
the 14th Centuries, Singapore/Canberra, 1986:xv.
27. See P. Zoetmuller and S. Robson, Old Javanese-English Dictionary, Gravenhage 1982:4;226;
1203; 1514.
 KA~A SCULPTURES A ND NATYASASTRA I:-II:-IDONESIA 37

28. Nawanatya (cod. BCB ptf 8. n. 6 and 7, Leiden Universi ty Library) lOb.
29. NaD'ai astra 4,12-16,
30. Nawaruu ya 2a
31. See ' Nawanatya' (appendix) in Th . G. Th. Pigeaud, Java in the 14th Century: A Study in
Cultural History. Th e Nagara Kerta garna by Rakawi Prapafica of Majapahit 1365 A. D.• 5 vols.
Th e lI ague 1960-63.
32. See Claire Holt. Art in Indonesia: Continuities and Changes, Ithaca J967:286.
33. Niigaral.:irtiigama 91.8.

Acknowledgements
I gratefully acknow ledge the support of the British Academy and of the Indian Council for Cultural
Relations which atlowed me to spend one year in Madras at Nrithyodaya, the institute directed by
Dr Padma Subrahmanyam, to whom I am greatly indebted. I am especially gratefu l to leeR as its
funding allowed an addit ional field work trip to Central Java in Aug. 1994 for the purpo se of visual
documentation of the temple complex. I also thank the staff of the Suaka Peninggalan Sejarah dan
Purbakala, Jawa Ten gah, for their assistance as well as Dr Padma Subrahmanyam, Shri V.
Balakrishnan and Professor Soedarsono for shari ng their knowledge with me on that occasion. This
article is based on a paper I presented at the seminar 'C on vergences and Continuities between the
Performing Art s of South and Southeast Asia', The Asiatic Society, Calcutta 11-12 Dec.1995. It
presents the resu lts of research carried out at Nrithy odaya during the period Jan-Dec. 1994 and pre-
vious research in Indonesia during 1992-93 , also carried out under the auspices of the British
Academy as well as the British Academy Committee for South-East Asian Studies, the University
of London Convocation Trust, the Ancient India and Iran Trust and the Central Research Fund of
the University of Lo ndon. I thank Dr Kapila Vatsyayan. Professor Lokesh Chandra and Professor
Edi Sedyawati for sparing their time to meet me and advise me. The views expressed in this paper
are entirely my own . l also acknowledge the considerable support given to the project by Miss
Hilary Sm ith, Dr and Mrs jR. Marr and my ow n fami ly.