Dance in 9th century Java: a methodology for the analysis and reconstitution of the dance

Dr. Alessandra Lopez y Royo

 This short essay  presents a case study – that of the dance reliefs of the Prambanan

complex in Central Java,  aiming  to steer the  discussion around  an important aspect

of any archaeological investigation of  dance.  Rather than focusing solely on

contextual issues, such as the nature and function of dance at a particular point in time

and in a specific socio-cultural context, the Prambanan case study questions  how to

engage with the archaeological dance record from a dancer’s  point of view, in other

words in terms of movement reconstitution and its re-embodiment.  It is almost

tautological to say that dance is practice based and performance oriented. However it

is often the case that it is precisely this aspect of dance which is neglected in

archaeological accounts and no methodologies are being developed to deal with such

issues.  My work on the Prambanan dance reliefs attempts to bridge this gap.

A brief history of candi Prambanan

Prambanan is a village located at about fifteen kilometers from Yogyakarta.  It  boasts

a now famous temple complex known as candi Prambanan (candi being the

Indonesian word for temple ) or candi Loro Jonggrang, from the name of a mythical

princess associated with the complex.  Candi Prambanan is in fact made up of more

than one complex – the Loro Jonggrang proper, the candi Sewu complex and the

smaller temples of Bubrah and Lumbung.  All these are within the Taman Wisata

Prambanan, an archaeological park which is a major tourist attraction in the island of

Java, second perhaps only to Borobudur, also in Central Java, and a major source of

revenue for the local inhabitants.

The area surrounding the Prambanan complex is historically one of the most

important regions of Java.  The Prambanan complex was unparalleled  in its heyday in

the ninth century CE..  It had over two hundred temples built within a system of

concentric courts.  The whole area surrounding Prambanan is  regarded as sacred  to

this day. Ruling dynasties in succession built numerous temples in the area with the

result that this is where one finds the highest concentration of Hindu/Buddhist temples

in the whole island. 

Towards the middle of the tenth  century the royal court moved from Central to East

Java, for reasons that are as yet unclear. This has been seen as an indication that the

whole area was abandoned following a disastrous earthquake , though there is not

sufficient evidence to support this theory of massive evacuation.  In the sixteenth

century a major earthquake, this time  documented in the  writings of the time,

devastated the complex.

Rediscovered in the nineteenth century, Prambanan attracted the attention of Sir

Stanford Raffles who had enthusiastically begun to record the antiquities of Java

during his brief military campaign in the island.  The complex subsequently

underwent several restoration phases and work on it is still going on.  All the sixteen

temples in the inner court have now been completely rebuilt and on my last visit to the

complex in 2001, the reconstruction of the temples in the outer courts had begun.

The Siva temple is the tallest, situated in the middle of the inner courtyard and

flanked by other two tall structures, candi Visnu and candi Brahma.  Occasionally

candi Siva is referred to as candi Loro Jonggrang,  because of the Durga statue it

houses in the north facing chamber, which  is said to portray   princess Loro 

Jonggrang.  This generates some confusion between the main temple and the

complex. Candi Siva ‘s main chamber contains a colossal statue of Siva Mahadeva;

the other chambers have statues of Ganesa, the elephant headed god, of Agastya, the

Saivite guru par excellence and the already mentioned Durga in her

Mahisasuramardini  aspect – killer of the demon Mahisa.  The temple is famous

worldwide for the reliefs on  the inner side of the balustrade, which depict the

Ramayana story.  Of Indian origin, the Ramayana or story of prince Rama is well

known in the whole of Southeast Asia, where it became  a symbol of ideal kingship. 

The narrative continues on the panels on the inner side of the balustrade of candi

Brahma. 

The outer side of the balustrade of candi Siva has sixty-two reliefs which depict

dancers and musicians in action .  Each relief is a self contained, short  dance

sequence whose movements can be identified and reconstituted using the

codifications found in the Indian work Natyasastra, a Sanskrit text on dance , music

and drama dating  back to the fifth century CE. 

The dance reliefs of candi Siva

The sixty-two reliefs alternate with other reliefs, seventy in number, showing celestial

beings in groups of three in a standing posture.  By having three figures in most of the

sixty-two dance  reliefs performing segments of one entire movement we see the use

of an iconographic animation device. Such  small dance sequences, shown in fifty of

the sixty-two reliefs , can be taken to be representations of a karana movement.  (The

nomenclature used here is that found in the Natyasastra).  A karana is a small

movement sequence which forms the basic unit of a now obsolete technique of dance

but also, and perhaps more importantly,  it offers a valuable tool for the analysis of

dance movements, due to the way  the dance unit is conceptualised.  The  Natyasastra

is a seminal work  because it  gives an account of the dancing and acting body  -

conceived as one and partitioned across the text in the way it is described -  which

does not superimpose a dance model  external to each  dance form that is  being

considered but offers the opportunity to create a relevant dance model out of the

dance itself.  Through a study of the  Natyasastra  it is possible to retrieve  a  concept

of dance technique which incorporates a set of criteria through which one can

“extract” the characteristics of “a system of bodily actions” (O’Shea 2000:82) and the

way these are systematically  joined. 

Thus the Natyasastra  classifications and codifications of movement  can be applied

cross-culturally to a variety of dance forms, without necessarily having   to postulate

a derivation of such forms from ancient Indian dancing. The technique can be

organised  out of  breaking up the dance body of each dance form, segmenting and

fragmenting the dance actions (and corresponding body parts which initiate the

action) of each specific dance and  slotting them into a category, corresponding to a

karana,  arranging each category on the basis of three body elements: the upper and

lower body and the body’s  position in relation to  a central axis (and any deviation

from it).  The reassembled dance actions  constitute a dance phrase which   in turn

becomes a dance unit, susceptible of  further subdivision –  a dance karana .  In

practice,  I have used this model  for a theory of corporeality  of the ‘Prambanan

dance’,  organising  the   analysis around the concept of technique, centred on the

dance karana.  By arranging the movements into karanas,  I have in fact    used  a tool

which emerges from the dance itself - here I specifically mean the  dance movements

of the Prambanan reliefs. This occurred as the dance was being re-embodied: 

 the  re-embodiment was led by  the segmentation process and subsequent  co-ordination of

the limbs’ movements. 

Karanas are not static poses; they are movements.  However, the Prambanan reliefs

show us poses, movements frozen in time .  These are in themselves insufficient  to

reconstruct a dance sequence.  This is where the concept of karana can be usefully

applied to the pose.  Through an analysis of the smaller movement elements visible in

each relief  it is possible to understand the movement dynamic of the sequence and

assign   to it a specific karana which is based on that movement dynamic. Once the

sequence is matched with a karana – in other words, “identified”   as a  karana - it is

possible to supply the missing movement and the sequence can be re-embodied.  This

of course does not constitute a whole dance; however one has the movement

vocabulary of a dance and can proceed to use such movement vocabulary in new

choreographies, if so desired. 

  A linguistic and literary analogy might help to understand this process: let us think

of dances – choreographies – as being   literary compositions such as poems.  These ‘poems’ of 9th

century Java are irretrievably lost.  What we have however is a series

of words, even phrases – the small dance sequences – which of course can be strung

together to make new “sentences”. These “sentences” will never be the original

“sentences” of the original dance compositions.  The reconstitution of the dance

movements seen in the Prambanan reliefs does not amount to a reconstruction of the

dance forms of ninth century Java.  But it is a reconstitution of some of its basic dance

“vocabulary”. 

The reliefs showing musicians give us some clues as to the kind of musical

instruments employed.  There are flutes and drums.  The reliefs do not show us the

context of the dance: the focus is entirely on the dance figures.  It is the Ramayana

reliefs that give us some clues as to the context of dance performances. In one of the

Ramayana reliefs we see a street dance  in connection with the celebrations for the

coronation of prince Bharata, thus we can infer that dance  was performed at major

festivities.  From other dance reliefs found at other temples in Central Java such as

Borobudur we can infer that dance had a role to play in a ritual context.  Some of the

Borobudur reliefs show dancing activities near a stupa (a Buddhist burial mound with

a relic of a holy perso, such as the Buddha himself) .

Analyzing dance

One question springs immediately to mind: can the Natyasastra be of help in

analysing all kinds of dance?  Here one needs to exercise some caution.  Conceptually

the notion of technique retrievable from this work is not tied to any specific dance

form, because the technique arises from the form itself.  However, I would be wary of

advocating the karana as a universally valid tool for analysis, without having the

opportunity of testing it further.  One thing that certainly helped the analysis of the

dance of the Prambanan reliefs was the awareness that there was an affinity between

Indian dance and Javanese (and also Balinese)  dance, in the way the dancing body is

constructed around the articulation of joints (in relation to one another), and in terms

of alignment and medians .  For example,  the Legong dance of Bali  uses extreme

deflection of the hip  from a central axis and extreme angularity of the arms to be held

at shoulder level with a sharp bend  of the elbows, the forearms facing forward and

the hands thrust upwards,  perpendicular to the respective wrist joint.  Deflection of

the hips is also found in Indian dance, especially in Odissi.

 Culturally, ancient India and ancient Java were very aware of each other and Indian

high-culture was imported and consumed in ancient Java.  Thus reference tothe 

Natyasastra system seemed to be contextually appropriate in this research.

Nevertheless, scholars such as Kapila Vatsyayan have highlighted the analytical

properties of the Natyasastra and its potential  use in a cross-cultural context

(Vatsyayan 1983a;1983b).  It remains open to further enquiry and research how far

this can be stretched and how useful such an endeavour would be.