Volume 43 1934 > Volume 43, No. 172 > Bark-cloth in Indonesia, by Raymond Kennedy, p 229-243
                                                                                             Previous | Next   

- 229

THE making of bark-cloth is at the present time almost an extinct art in Indonesia, as it must soon become in Polynesia also. Only in the most remote districts is it still manufactured and used, so thoroughly has the use of cotton cloth for dress supplanted it. In these few secluded regions, however, observations have been made which establish the practically absolute identity of Indonesian bark-cloth technique with that of Polynesia. Not only is the process of manufacture precisely the well-known Polynesian felting method, but several of the names of implements and processes, at least in the eastern part of Indonesia, have striking resemblance to, if not identity with, the Polynesian terms.

The term tapa, however, is not used for the material in Indonesia, the general term in use being fuya, a corruption of wuyang, the Minahasa (North Celebes) name for a woman's sarong, or skirt. Until the latter part of the past century, it was an important article of trade from Celebes to Surabaya and Singapore, where it was used as an underlay in coppering the hulls of vessels. 1

Formerly, it was used for clothing throughout all Indonesia, as well as in the Philippines and the Malay peninsula. Even now, in regions where cotton is in general use but paganism still flourishes, bark-cloth is worn at religious ceremonies and in periods of mourning. In many places, also, where it has passed out entirely as a form of clothing, the rougher kinds are used as packing material and the finer sorts for writing purposes. Especially is the latter true of Java, Madura, Sumatra, and the more - 230 advanced regions of Celebes. The bark-paper industry of Java and Madura was flourishing until very recent years, and where conservatism is very strong this ancient kind of paper is still manufactured and used in writing.

The district of Indonesia where the tapa, or, to use the Malaysian term, fuya, industry still flourishes to the greatest extent, is the central part of Celebes, the large island situated between Borneo and New Guinea, just south of Mindanao in the Philippine group. Thanks to the excellent observation and recording of the two Dutch missionaries, N. Adriani and A. C. Kruijt, 2 we have full information regarding the manufacture of bark-cloth among the Toradja tribes of middle Celebes.

The Toradja are a group of small tribes, closely related among themselves, living in scattered communities throughout middle Celebes in the midst of numerous other peoples. They belong to the ancient pre-Malay or so-called proto-Malay stratum of Indonesian population, the stratum from which, evidence points, the Polynesians in large part originated. The name Toradja means “uplander,” but for some time past they have been living in as great numbers along the coast and in the river-basins as in the mountainous interior. They remained practically untouched by Hindu influence, and only a portion of them have embraced Islam.

At the time when Adriani and Kruijt wrote their classic on the Toradja (1912), practically no cotton was being imported into the Toradja country, and weaving was unknown. Bark-cloth was the common dress of the people. P. and F. Sarasin, who spent considerable time among the inland peoples of Celebes, describe a typical early morning scene in the Toradja country in part as follows:

“…From all sides came the melodious sounds of the beating of fuya, all on different levels of tone,—something so characteristic of the Toradja-lands, so reminiscent of Polynesia.” 3

- 231

The trees used by the Toradja as the source of their material for cloth are chiefly the following nine:

  • umayo (Trema amboinensis)
  • ambo (Broussonetia papyrifera)
  • tea (Artocarpus blumei)
  • impo (Antiaris toxicaria)
  • bunta (Sloetia minahassae)
  • leboni (Ficus leucantatoma)
  • kampendo (Ficus sp.)
  • nunu (Urostigma sp.)
  • wantja (?) 4

The most favoured sources of bark for cloth are the first three varieties mentioned above; the last five are used only in the upland lake district, where the first four do not thrive. The best cloth is made from the bast of the ambo, or paper-mulberry, which is cultivated for the purpose by slipping. All the other trees mentioned grow wild. The bast of the umayo and the tea makes up into coarser cloth, but since the former is the most common of all the trees mentioned, the great bulk of Toradja cloth is made from it.

The impo can be used only when young, and the same applies to the nunu, or banyan. The fuya made from the latter bark is long and broad, and is used for blankets, especially in the cold upland regions.

The method of preparation and beating as described by Adriani and Kruijt, 5 is practically indentical with the Polynesian “felting” procedure, as contrasted with the “pasting” technique, a local variation practised in western Polynesia, notably in Samoa and Tonga. 6 The latter process is unknown in Indonesia, and is undoubtedly a western Polynesian or Melanesian invention.

The tree is cut down and the bark stripped off, after which the inner bark, or bast, is separated from the outer bark. If brown or black cloth is desired, the bast is not cooked, but merely beaten (rawalowo), washed (rapaa), and wrung out (rakomo). The paper-mulberry bast is never cooked, it being naturally white in colour. If, however, a fine jacket or headcloth is to be made, or if white cloth is desired and there is no paper mulberry bast available, the bast of the umayo or any other of the coarser varieties - 232 must be boiled (raluwa) for some time before beating. This is done in a large cooking-pot (kura), ashes being added to the water. After boiling, the bast is beaten, washed, and wrung out.

The next procedure is the so-called “fermenting” (raronu) of the bast. The strips are folded up, still damp, in leaves of the kombuno (Livistona rotundifolia), which covering keeps them moist during the period of fermentation. The fermenting usually takes from two to three days. If the bast has been previously cooked, only one day is required. The bast of certain Ficus varieties must ferment for from ten to twelve days. The strips are usually one to one and a half metres long by about ten centimetres wide. Great care is taken not to disturb in any way the bundles while fermenting is in progress. Even a slight touch is considered deleterious to the material.

After the bast has fermented for the required period, it is ready for beating. The term used for beating the prepared bast is different from that used to refer to the beating that takes place on the raw material before it has been washed and put away to ferment. The word for the latter beating is, as noted above, rawalowo, whereas the final beating of the prepared bast is called mondodo, from the stem dodo, which is the same as the Polynesian term for beating, tutu.

The beating takes place in special houses, or on platforms built under the floors of high pile-raised rice-storehouses. The reason given for not carrying on the beating process in the dwelling-house is that the beater might strike and thereby offend some of the invisible spirits that haunt the family domicile.

It should be remarked here that the entire business of cloth-making among the Toradja is carried on by women, they even cutting the trees from which the bast is procured, except for very large trees, which are cut by the men. One exception to this rule occurs among the mountain Toradja, where the priests do a great deal of the painting of the finished fuya. Noteworthy is the fact, however, that these priests dress like women.

Within the beating-house are to be found one or more “anvils,” or beating-planks. These are called totua, which - 233 corresponds directly with the Polynesian term for the same article, tutunga. The beating-plank, made of the hard and springy yellow wood of the wolasi (Lagerstroemia ovalifolia), is approximately one and one-fourth metres long by twenty centimetres wide and has a thickness of six to eight centimetres. It rests on two blocks fifteen to twenty centimetres in height, called tangoni. When struck, the plank gives forth a musical sound, and when several beaters are in action at the same time, the varied pitches of the planks blend into a kind of xylophonic harmony.

Before the bast is laid on for beating, the plank is polished with bamboo or Ficus leaves, and then the woman squats or sits with legs outstretched beside the plank. She grasps the beater with both hands and is careful to aim every stroke just beside the last, thus working gradually and evenly over the whole material. She keeps a calabash or a half coconut-shell containing water at her side, in order to sprinkle the bast from time to time while beating.

The beaters used will be described below in detail. Suffice it here to say that they are of various sorts, running a whole gamut of size, shape, design, and weight. The generic term for beater among the Toradja is ike, which is the name used throughout Polynesia.

The handling of paper-mulberry bast is slightly different from that of the rougher sorts of material. The strips of the former are beaten with a heavy hammer (pombayowo) before they are laid upon each other to be felted together. This is done because the paper-mulberry bast is not cooked before fermenting, and also because it has stiffer fibres than the other kinds of bast. The strips of other kinds are taken from the packages of leaves in which they have fermented and are not subjected to preliminary beating before actual felting begins. They are superimposed one upon the other, in number varying according to the thickness desired in the finished product. Length and width are increased by overlapping the edges of additional strips and felting the whole mass together. Sometimes, since paper-mulberry bast is relatively scarce, strips of it are alternated with strips of other kinds of bast, producing a composite variety of cloth.

- 234

The heaviest hammer, the pombayowo, is used in the first stages of felting. When the strips have been beaten together for some time, the whole sheet is folded over several times and beaten further with finer hammers. This folding-over saves time and also prevents the punching of holes in the rapidly thinning sheet. Care is taken that the folded material does not stick together, and water is sprinkled over it continuously, so that it will not split from dryness.

A fine headcloth may be beaten an entire day to procure the desired thinness. Closed cylindrical skirts, or sarongs, are produced by felting together the longitudinal edges. Sometimes a narrow strip of bast is laid over the juncture and beaten in to strengthen the joining. Such a sarong always shows the reinforced seam-line.

The finished product is hung up in the wind to dry, but before it becomes entirely dried out, it is taken down and beaten further with a smooth, round ebony stick. This ebony beater, illustrated in fig. 2 (b), is called pombobaki or rabobaki, and the final beating with it takes place in the house, on a flat stone, not on the regular beating-plank. Then, after final and complete drying, the cloth is anointed with plant-sap, usually procured from the bitter fruit of the ula (Strychnos ligustrina) in order to make it durable. The sap is applied with a chicken feather or a brush of coconut-fibres. Only one coat of the liquid is applied to the side of the cloth which is to be the interior of the garment, three or four coats to the exterior side.

There are alternative methods of treating the finished cloth. It may be dipped in a decoction made by boiling various kinds of bark with ashes in water. Again, it may be smeared with the juice of the chewed fruit of the bo'e (a Rhododendron) and of the sakoti, a spice which is also chewed by youths whose teeth have recently been filed and are being blackened. After it has been so treated with these juices, it is taken to a swamp and held in the mud for a time. It is then dried and kneaded with the hands to soften it. This method produces a black colour, and such fuya is seldom used by the people of the lowland regions. The process of obtaining a black colour by soaking in a bark-infusion and then submerging the cloth in the - 235 mud of taro-swamps was widely used in Polynesia. There is another method of treatment in vogue, which produces a transparent cloth, used only on festive occasions and so delicate as to be good for only two or three wearings. The cloth is anointed with an oily decoction made by boiling together ula sap and grated coconut meat. Such cloth is made into jackets and headcloths only, is usually dyed in various colours, and is considered the finest form of dress obtainable. The jackets made of this transparent material are worn at feasts by young girls only, the headcloths by both men and women.

The fuya used for ordinary daily wear is brown, somewhat coarse, and practically never decorated in any way, being merely anointed with ula sap for purposes of durability. The cloth used for festive dress, however, is decorated in many designs and colours. The fuya so decorated is always the fine white product made from the paper-mulberry or from the carefully boiled and beaten bast of other trees. It may be dyed in solid colours or painted in various designs. The mountain Toradja produce especially-excellent painted cloth. Among these people, it is the priests and priestesses who do the fuya-decorating, and the designs have symbolic significance. This is not the case among the lowland Toradja, except for men's headcloths, the designs used on which are, or rather were, connected with the head-hunting complex of the region, which has now, of course, been practically stamped out. The type of decoration a man was allowed to have on his headcloth depended upon the number of times he had participated in head-snatching raids.

There are two methods of decorating cloth with colours, both known in Polynesia as well. They are, freehand painting, and stamping. Illustrations of some stamps, made of ebony, in use in Sausu in Celebes, are given in fig. 1. A bamboo pen with three prongs is used to draw parallel lines. Such pens were also used in Polynesia.

In painting, the design is first sketched in, a black infusion of dissolved resin-soot serving as ink. There are four colours used in Celebes for fuya-decoration, obtained as follows:

- 236
FIG. 1.
EBONY STAMPS FROM CELEBES (Actual size). (After Adriani and Kruijt. 7)

1. Carmine-red (malei, mawaa, dolo): Strips of wood of the dolo tree (Morinda bracteata), cut just at the point where the roots enter the trunk, are pounded together with sirih (betel) and lime and then boiled to produce this colour. Red is also obtained by rubbing the pits of the fruit of the alomi (Peristrophe tinctoria) several times over the cloth.

2. Yellow (manggaa, makuni): Usually the roots of the kudu (Morinda citrifolia) are boiled to produce this colour. An alternative source is the kuni (Curcuma longa). Both the Morinda citrifolia (nono) and the Curcuma longa (lenga) were used for yellow dyes in Polynesia.

3. Purple (makodara): The flowers of the lele ngkasa (a Papilionaceous variety), a spice-plant with purple blossoms, are roasted and pressed in water by hand to obtain this colour. This plant is cultivated for the purpose, since many flowers are needed to produce a small amount of the dye.

4. Green (mayawuyu, malumbe): The leaves of the kalamaya (?) are pounded and kneaded by hand in water to procure green dye. This colour is seldom used.

Black designs are usually put on by rubbing the cloth with resin which has been heated and is still liquescent.

- 237

The brushes used are generally made of bamboo or the leaf-stalks of the kalidjawa (Jatrophe curcas).

The type of beater (ike) used among the Toradja has a flat stone head, which is clamped in a holder of bent-over rattan. The rattan is held in place by grooves made in the narrow sides of the stone head and is tied with rope or rattan strips to a short piece of wood which forms the handle. A sketch of such a beater is given in fig. 2 (a). This type of beater is certainly an invention of the Toradja themselves, the other Indonesian beaters being made of wood, like the Polynesian implements. The beaters used in the Javanese and Madurese paper-industry have copper heads as a rule, this being a local Javanese development.

The stone used in the Toradja beaters is serpentine, dark-green in colour. Such stone is always called watu-ike (“beater-stone”) by the Toradja. The To Onda'e, a mountain tribe, are the exclusive possessors of the art of quarrying and manufacturing the stone for beater-heads, and they barter the finished heads for other products with the lowland people. They quarry the stone with axes and make the grooves required with knives. After the grooves have been made, the head is cooked in water containing leaves of the tobalo, or wayowuyu (Bambusa longinodis), and the tetari (Scleria scroliculata), which contain much salicic acid, and are believed to have strong “soul-stuff,” or tanoana, in them. The “strength” of the leaves supposedly passes into the stone in the cooking-process. After this, the stone, while still warm, is rubbed with wax to make it smooth and shiny.

Every beater-head has a different arrangement of grooves on each side, and the fuya-maker has a kit of many beaters, each one used in a different stage of the manufacture of the cloth. The beater with the largest grooves is first used, and as the cloth comes nearer and nearer to completion, beaters with finer and finer grooves are employed. The first and largest beater has three vertical grooves on one side and seven to nine on the reverse. The side with the three grooves, illustrated in fig. 2 (a), has a special name (pombayowo), as does the other side, which is called pongkakagi, meaning “felter.” It is this latter surface which is used in felting the strips

- 238
FIG. 2.
BEATERS FROM CELEBES (About half actual size)., (a) Pombayowo, stone head with rattan holder., (b) Pombobaki, round ebony beater (after Adriani and Kruijt 8).
- 239

together. Then comes the pondeapi, with more numerous vertical grooves, and then the po'opi, with eleven to fifteen vertical grooves, alternating large and small. This is the only beater having unlike grooves. A very small edition of this type is employed in the making of fine headcloths, being called ike rapi.

The reverse side of the po'opi has twenty-one diagonal grooves, and is called pomparo topi, meaning “sarong-maker.” Other beaters are the pontjongi, with thirty-four diagonal grooves, and the pomparo tali, or “headcloth-maker,” with thirty-six diagonal indentations.

The kulamuti is a special sort of beater, having twelve identical indented rosettes, four in length and three in breadth. This is used for the final beating of fine headcloths, and leaves a distinctive watermark on the material. A sketch of such a beater-head is reproduced in fig. 3.

FIG. 3.
BEATER-HEAD (kulamuti) FROM CELEBES (Actual size). (After Adriani and Kruijt. 9)

The distinctive features of the Toradja beaters are the springy rattan holders and the stone heads. Elsewhere in Indonesia, as remarked above, the beaters follow the all-wood Polynesian pattern, the two- and four-sided ones generally having a different arrangement of grooves on - 240 each surface. Many of the wooden beaters are round, and indented in various ways. Sketches of a few representative wooden beaters are given in fig. 4. A variant from the usual type occurs in Java and Madura, where copper-headed beaters are used in the bark-paper industry.

FIG. 4.
INDONESIAN BARK-CLOTH BEATERS (One-sixth actual size)., (a) Four-sided beater from Borneo., (b) Flat two-sided beater from Borneo., (c) Round beater from Sumatra., (After Adriani and Kruijt, and Schmeltz. 10)

The technique of commencing with beaters with more widely-spaced grooves and proceeding with beaters with grooves set closer together was also used in the felting method of Polynesia. The square ironwood beaters of Polynesia provided four surfaces, on each of which the grooves varied in number. As with the Toradja, the Polynesians gave different names to the surfaces according to the number of grooves. 11 Geometrical patterns were also used on the beaters of Hawaii to produce distinctive - 241 watermarks. Even the Toradja method of completing the process by beating with a smooth, round ebony stick finds its Polynesian affinity in the use of a smooth, ungrooved surface on the beater to finish off the cloth.

Space is lacking to pursue in detail the subject of bark-cloth manufacture in the other parts of Indonesia. The Royal Ethnographic Museum of Leiden has numerous specimens of bark-cloth artifacts from practically every region in the Archipelago. 12 Sarongs, jackets with and without sleeves, headcloths, loincloths, shawls, caps, and bags are, or were, all made from bark-cloth in the various islands.

Two articles of clothing made from bark-cloth in Indonesia bear notable similarity to forms of dress in Polynesia. The old type of loincloth, as worn in the Archipelago, is a long strip of fuya wound round the hips and passed between the legs, the ends being allowed to hang down several inches in front and rear. The prototype of the closed jacket, with or without sleeves, which is now generally worn in Indonesia, is the so-called “poncho,” consisting merely of a broad strip of bark-cloth in which a triangular opening is made for the head. This ancient form of jacket, open at the sides and hung on the shoulders like a sandwich-board, is still worn in the Minahasa region of North Celebes. It was formerly used in central Polynesia also. Thoughout most of the regions of Indonesia where bark-cloth is still used, the jackets of fuya are made on the same general pattern as the poncho, but the sides are sewed up, except for the arm-holes. This closed form of bark-cloth jacket in many cases has sleeves attached. The evolution of the jacket from the poncho original is thus well-illustrated in Indonesia. A. B. Meyer's album 13 contains a photograph of a bark-cloth poncho worn by a North Celebes man which hangs from the shoulders to the mid-thigh in front and rear, being held in at the waist by a girdle of bast. An even better illustration, also from Minahasa, is given in P. and F. Sarasin's book on Celebes. 14

- 242

The trees used for bast are the same, generally speaking, in all regions. For instance: the Leiden Museum has cloths made of Antiaris toxicaria and Artocarpus elastica (wild-breadfruit) from Borneo, of Artocarpus elastica and Urostigma benjaminea (banyan) from Nias, of Artocarpus elastica and Artocarpus incisa (breadfruit) from Mentawei, of the same two kinds of bark from the Bataklands of Sumatra, of Broussonetia papyrifera (paper-mulberry) from Java, of Hibiscus tiliaceus (wild-hibiscus) from Celebes, of Ficus indica from Luzon in the Philippines, of Ficus nodosa from Buru in the Moluccas.

From this survey, it is obvious to the student of Polynesian culture that in Indonesia there once flourished, and still does in some places, especially in Celebes, bark-cloth manufacture of a kind identical with that of Polynesia. The similarity of the technique of felting in the two areas is, of course, obvious. The dyeing-methods are also essentially the same, though different materials may be used. When the linguistic correspondences between the terms used for materials and processes in Celebes and those employed in Polynesian districts are added to the technological similarities, the picture becomes complete. We have already mentioned the term ike, used for “beater” in both areas, the correspondence of the Celebes term totua with the Polynesian tutunga for the beating-anvil or -plank, and the practical identity of the Celebes dodo for “beating” with the Polynesian tutu. Adriani and Kruijt 15 note also that the word for the Trema amboinensis, the tree whose bast is most commonly used in Celebes for bark-cloth, though now umayo, has an older form in the Parigi-Kaili dialect, malo, which, they say, is undoubtedly the Polynesian malo or maro(“loincloth”), the Fijian malo, the Massarete Buru (Moluccan) kamaru, the Mafur (New Guinea) mar. It is interesting to note also that the local term for fuya among the Toradja is inodo (sometimes dinodo), meaning “the beaten stuff,” which term is, of course, also from the stem dodo. The same base forms the word for beating-house, pondodoa (also pomdede), meaning “place of beating.” Even more striking is the fact that in the - 243 Parigi-Kaili dialect the word for “bark-cloth beating” is mantutu. “Beating” is in Malay tutuq, in Sundanese (west Java) tutu, in Javanese tutuq, in Bisaya (Philippines) tuktuk, in Tagalog (Philippines) tugtug, in Malagasy (Madagascar) totoka. 16

Undoubtedly, further research and collaboration between Indonesian and Polynesian investigators will result in the scientific establishment of far-reaching correspondences and relationships between the cultures of the two areas, and will throw light upon the ancient history and movements of the Polynesians in their migrations out into the Pacific from the far-western homeland, Hawaiki-pa-mamao, which is certainly to be located in the Indonesian Archipelago. This paper is offered as a sample of what is to be found in the rich store of East Indian ethnographical material of great interest and import to students of Polynesian history and culture. Bark-cloth is just one of innumerable common culture-traits whose future analysis and comparison in the two areas will be of inestimable value in lending objective scientific substantiation to hypotheses of connection and migration which up until now have never been properly corroborated by comparative factual data.

1   Encyclopaedië van Nederlandsch-Indië, 'sGravenhage-Leiden, 1917, vol. 4, 788.
2   De Bare'e-Sprekende Toradja's van Midden-Celebes, 3 vols, Batavia. 1912.
Geklopte Boomschors als Kleedingstof op Midden-Celebes, en Hare Geographische Verspreiding in Indonesië, in Int. Archiv für Ethn., vol. 14 (1901), 139-191.
3   Reisen in Celebes, 2 vols., Wiesbaden, 1905, vol. 2, 93.
4   Adriani and Kruijt, De Bare'e-Sprekende Toradja's, vol. 2, 314.
5   De Bare'e-Sprekende Toradja's, vol. 2, 315-325.
6   Te Rangi Hiroa, Samoan Material Culture, Bishop Museum Bulletin. 75, 1930, p. 294 et seq.
7   Geklopte Boomschors, in Int. Archiv für Ethn., vol. 14, pl. 11.
8   (a) Geklopte Boomschors, in Int. Archiv für Ethn., vol. 14, pl. 11.
(b)De Bare'e-Sprekende Toradja's, 2, 321.
9   Geklopte Boomschors, in Int. Archiv für Ethn., vol. 14, pl. 11.
10   Geklopte Boomschors, in Int. Archiv für Ethn., vol. 14, 168-171, note 4; 172-173, note by J. D. E. Schmeltz.
11   Te Rangi Hiroa, The Material Culture of the Cook Islands, New Plymouth, New Zealand, 1927, 78.
12   Catalogus van 'sRijks Ethnographisch Museum, Leiden, 23 vols., 1909-1932.
13   Album von Celebes-Typen, Dresden, 1889, plate 3, fig. 1.
14   Reisen in Celebes, 1, 50.
15   Geklopte Boomschors, in Int. Archiv für Ethn., vol. 14, 140, note 5.
16   Geklopte Boomschors, in Int. Archiv für Ethn., vol. 14, 142, note 8.