levels in their height, but always "above zero").
Figure 17: Bebitra.
Most of these features have had an effect on the techniques of modeling and casting Pejeng type drums. Most of them, conversely, had their origin in the particular circumstances of the Pejeng type drums' "prenatal history," their being derived from a special prototype (a tifa) that left its characteristic features in the appearance of its successor. The four-sided symmetry and its extension to a larger system represented by the formula 1 + 4 + 4 (in connection with the four handles, the eight (4 x 2) "heads" in the layout and decoration of the Moon), should be understood as part of a well-known Early Indonesian cosmological and philosophical concept. In Bali it is known by the name nawasanga (the two elements of this term refer to "nine"). We will meet this classification system also in Hindu Balinese religion and art, e.g. in connection with the "four-body" figures (catuhkāya and the mukhalingga).
The most essential features briefly mentioned in this list should be more explicitly discussed hereafter.
The Pejeng Moon is the largest and most important representative of the "Pejeng type," which took its name from its home site. There are several tympans, perhaps a few complete drums (if information received but never checked in any detail proves to be correct). There are a few highly interesting related finds, especially the printing mold fragments from Manuaba (that will be discussed presently). More of them exist, and certainly more will be found, which may change the situation and our views. At the moment we have to restrict ourselves to the main members of a "classic" Pejeng type. They are referred to according to the locations where they have been found—if possible. It so happens that for one of them (Leiden) the place it was found is unknown. The second (Tanurejo) was discovered in Central Java and is now in the National Museum, Jakarta. The third (Bebitra) is from a place in Bali, but was presented to the Archaeological Service and is now in the National Research Center in Jakarta. The first named tympans are obviously in the same "classic" category and may be taken as being Balinese in origin.
Figure 18: "Leiden."
The "Leiden" tympan, diameter 55 cm, is illustrated by Figures 18 and 20 B; the Tanurejo tympan from Kedu (Central Java), diameter 51.5 cm, is shown in Figures 19 and 20 C; the Bebitra tympan, from the residence of the former local prince (Puri Gianyar) in Gianyar, discovered in 1962, diameter 64.9 cm (further details will be given in their special context), is illustrated in Figures 17 and 20 A. The Manuaba mold is illustrated in Figures 12-13 and will be discussed in a paragraph of its own. General information is to be found in the present author's book, Kettledrums of Southeast Asia. A Bronze Age World and its Aftermath, 1988.
Figure 19: Tanurejo.
Figure 20: Five versions of the knob-and-loop pattern. A)Bebitra; B) "Leiden" C) Tanurejo D) The Moon E) An early moko from Alor.
Both the Heger I type and the essentially different Pejeng version of metal drum acquired their form and structure by the translation of earlier musical instruments of natural substances—wood, fibres or earthenware) into metal. This material metamorphosis is just one more instance of the wide-spread phenomenon in early metal ages among various peoples. Not unnaturally, therefore, the transformation that led to the kettledrums of Southeast Asia was similar to a degree in both cases referred to, Heger I and Pejeng. Similar, but far from identical. To start with, the non-metal prototypes translated into metal were different. The characteristic make and profile of Heger I drums suggest a combination of, say, a cushion-shaped one-sided membranophone drum resting on a wooden standard of sorts to enhance the effect of the drumming: an aggregate, to put it differently, that before being turned into metal had already undergone one or more earlier changes. The prototype of a comparatively slender, distinctly waisted Pejeng type drum, on the other hand, to all appearances was an equally one-sided membranophone, "hourglass-shaped" hand-drum of a type that is still known in the islands of eastern Indonesia and formerly was used in Java along with gongs and other metal instruments. The type is called tifa referring to a wooden hand drum, elegantly handled by girls when dancing to their own accompaniment. The drum is pressed to their body by the left arm, the membrane (a piece of cowhide) being tapped with both right and left hands. It is a light, manageable drum to be consciously chosen for the prototype for a much larger and heavier metal drum, the Pejeng type drum. The outcome clearly shows several traits that were characteristic of its tifa predecessor. Moreover, instead of being merely musical and sound instruments the Pejeng type drums entered a much wider field of social and religious activities, in what was as yet a "Pre-Hindu Balinese" atmosphere. The form and structure, decoration and technique connected with the Pejeng drums was highly influenced by their prenatal background, and by the tifa itself.
The tifa, open on one end, was closed by the membrane, that was fixed by means of a system of strings, rings and knobs which made it possible to tune and adjust it to the needs of the player. Both the piece of hide and the strings had to be fastened about a hand's width down from the top of the drum's mantle. There may be a connection between the band of concentric circles filling the outside of the cuff and the fastening of the hide. Yet there is a much more suggestive possibility. The four-fold repetition of the decorative pattern on the tympans, consisting of knobs, braids and loops—seemingly a random product of artistic imagination—may well reflect the strings, rings and knots of the tuning gadget that played an important part of the tifa's musical ambience. A good deal later it was still used in some of the early mokos from Alor in a vaguely remembered decorative tradition.
Figure 21: An early moko from Alor. (See also figure 20 E.)
The fragments of a stone printing mold for impressing the design of a Pejeng type drum onto the wax layer prepared for the drum's casting, briefly mentioned previously, were discovered in 1932 at Manuaba in the district of Gianyar.
In the discovery and interpretation of the fragments (three, at first; later a fourth would be added by R. P. Soejono which enabled him to reconstruct part of the mold's outlines), Walter Spies and K. C. Crucq (Archaeological Service) played a pioneer part. The local people regarded the stones as curiously shaped pieces of rock which deserved being venerated as such. When in 1951 Mr. Krijgsman and I visited the pura desa where they were kept we had to study them from a little distance.
The total height of the mantle projected from the Manuaba mold could be calculated by Soejono to be 107.5 cm, which we should increase for the drum itself by perhaps 10 cm. The height of a cuff, 118 cm is far less than the Moon's height, not "extra large" in kettledrum terminology, but "tall" anyhow. The three horizontal sections (A, B and C) were estimated to be 98, 83 and 100 cm in diameter, respectively.
As to what conclusions should be drawn from this mold, it should be remembered that it was a printing mold not a mold for casting, which is quite a different thing. The soft paras stone never would have stood the heat of fluid metal. It should also be understood that for the modeling of an outstanding drum like the Moon much more advanced techniques and molds must have been used than this simple