Survival of a Weaving Tradition in New Guinea

By: Michael C. Howard and Naffi Sanggenafa

Originally Published in 1999

View PDF

Traditional dress in Irian Jaya, in the Pacific archipelago nation of Indonesia, is usually associated with penis gourds, string bags, and grass skirts. The province (which occupies the western half of the island of New Guinea) has universally had the reputation among textile spe­cialists of a place “where weaving is not practiced” (Gitlow 1992:15). It would no doubt come as a surprise to many, therefore, to learn that there is a type of woven cloth produced in Irian Jaya. The term terfo refers to cloth woven from nibung palm fiber by the Sobei, who live in the vicinity of the present-day town of Sarmi (Figs. 1, 2).

This relatively simple and little-known cloth is of significance for two reasons: it represents the only loom-woven cloth known to have been produced in Irian Jaya; and it is one of the last surviving examples in Southeast Asia of what was probably the initial style of weaving employed by the Austronesian peoples as they migrated across Island Southeast Asia thousands of years ago. Moreover, as we describe below, that terfo weaving has survived to the present day is in itself a remarkable story.

Our own interest in terfo was sparked when we came across examples of the cloth and a loom on display in Irian Jaya’s provincial museum in the capital, Jayapura. The information available on these items was sketchy so we decided to visit the source, making an initial trip to the village of Sawar (near Sarmi) in July 1995. On our return to Jayapura, we stopped at various islands with a history of terfo production and use. Sawar is one of two communities where terfo is still being produced. The other is the adjacent village of Bagaiserwar. Unfortunately, because of the intense rivalry that exists between its inhabitants and those of Sawar, it has not been possible to conduct much research among the people of Bagaiserwar, and our account is largely from the perspective of the people of Sawar.


Let us begin with a brief description of terfo production. The thread for making the cloth comes from the fronds of the nibung palm (Oncosparmo tigillorium), found growing a little inland from the coast. The fronds are cut from the tree and the individual leaves or pinnae removed from the midrib and tied into bundles (Fig. 3). The bundles are taken back to the vil­lage and the leaves are boiled in a pot of water (Fig. 4). The boiled pinnae are then soaked in sea water, scraped until they are white, and hung to dry (Figs. 5, 6). They are then pulled apart into thin strands which are rolled on the thighs to form thread.

While some strands of thread are left plain, others are dyed. Undyed threads are said to be white. The colors most commonly encountered are red, yellow, blue, black, and green. Red is made from the root of a fruit tree known as mare which grows around the villages. The root is scraped and then mixed with coral in water. Yellow is made either from saffron, which is grown in household gardens or from the more root unmixed with coral (Figs. 7-9). Blue is made from the fruit of a tree known as manoafo. Black is made from the heart of the fruit of the menoerto tree, and green from the bemotepori tree. All of these trees are found in and around the villages. These days dyes are also sometimes obtained from colored paper.

The Sobei employ two types of simple single loop-heddle loom (Yashimoto 1990; Fig. to). The weaving technique is simple and patterns of lines and checks are formed by employing differ­ent colored threads. It usually takes about four to five days to weave a piece of terfo.

Traditionally, the size of the cloth varied according to its intended use. Because of the served as the medical officer of the 1903 Neth­erlands New Guinea Expedition, produced what is widely regarded as the most important Dutch account of the material culture of t he north coast of Irian Jaya (Sande 1907). He does not, how­ever. make any mention of terfo.

The only weaving cited by van der Sande comes from Tarfia, at the eastern end of the cul­tural group that includes the Sobei. He describes a “weaving frame” from Tarfia, which is said to be “used in the manufacture of brow bands. girdles. armlets, etc., which form the monopoly of this village” (Sande 1907:237).

Terfo reappears in the literature in the 1920s in an article by B.M. Goslings, in which he discusses a report by the Assistant Resident of Northern New Guinea. L.A. van Oosterzee (Goslings 1928-29:119-20). On a brief visit to Wakde island in 1900 van Oosterzee found that weaving had been practiced on Wakde a short time before, but that it had ceased as a result of the importation of cotton cloth, as was also true of Yamna island. Thus, he stated, at this time weaving was found only among the Sobei speakers of the Kumamba islands and the neighboring mainland. Goslings’ article also includes a photo taken in 1922 of a Sobei woman weaving.

We have found no further mention of terfo until Yoshimoto’s 1990 study of Indonesian looms mentioned above. With so little written on the subject, especially in English, it is little wonder that the surveys of textiles in Indonesia ignore weaving in Irian Jaya. Interestingly, even the recent, and in many ways encyclopedic, Art of Northwest New Guineo. although covering the “Wakde-Yamna” area, makes no mention of tofu (Kooijmann and Hoogerbrugge 1992).

Contemporary Terfo Weaving

Our initial 1995 survey indicated that terfo weaving is now restricted to the two villages of Sawa r and Bagaiserwar. Terfo weaving and most other handicraft production had long since van­ished from Sobei (now part of Sarmi) and Like. In fact, when Goslings published his article in the 1920s. from what we could gather, terfo weav­ing was on the verge of extinction in Sawar and Bagaiserwar as well.

Two factors in particular had led to this situ­ation. The first was the increasing availability of commercial cotton cloth for clothing. By the 1920s terfo was no longer used for clothing. Moreover, it had been largely replaced by cotton

cloth or simply fallen out of use for many types of exchange as well and was no longer traded within the region. The second factor was the conversion of the Sobei to Christianity. The destruction of the men’s house and various ceremonial objects put an end to the need for terfotopo. Christian­ization also served to enhance the value placed on modern commercial, rather than primitive handwoven, cloth.

The woman featured in the photo in Gos­lings’ article is Pueba Merne. Informants told us that at the time she was the only woman in Sawar who was still weaving. It is interesting to note that Nenek (Grandmother) Pueba was from the Merne clan, the highest-ranking clan in Sawar and the one traditionally responsible for sacred affairs, including care of the men’s house and the sacred flutes in Christianity times. Along with a number of other artisans from Irian Jaya she was selected by Dutch officials to take part in the 1929 Arts and Crafts Exhibition in Batavia ( Jakarta) (see Howard 1998:56). Contemporary  limitations placed on the width of the cloth by the loom, sometimes wider cloths were made by sewing two pieces of cloth together. In the past there were mainly three types of cloth (Figs. 11-13) a narrow type used for a man’s waist-or loincloth (sober). another narrow one used as a cheesecloth (adze). and larger pieces used for women’s skirts (memdo).

Terfo was not only used as clothing. It also served important religious and social functions. It was sometimes imbued with special sacred sta­tus and hung in the ceremonial men’s house or was displayed by the village head or his sons. Such cloth was referred to as terfotopo. It looked no different from other terfo, but was assigned a special status. Terfo was included in traditional payments associated with rites of passage, in par­ticular the sorani weyan (milk price) that takes place at the time of marriage. It was also pre­sented as a reward for merit within the village and to help seal bonds of friendship and other social relations to non-villagers. In addition, it could he bartered, but only for certain goods such as particular bodily adornments and stone axes, never for everyday goods such as food. 

Surprisingly enough, considering the scarcity of materials from the north coast of Irian Jaya in North America, the collections of the University of Pennsylvania Museum include an early example of terfo. This piece is not part of a systematic collection, but one of a group of miscellaneous objects given in 1897 to what was then called the Free Museum of Science and Art of the University of Pennsylvania by Enrico H. Giglioli.

Giglioli (1845-1909), a zoologist by profession, was also an enthusiastic ethnoanthropologist and a founder and leading member of the Società Italian di Antropologia ed Etnologia. In 1883 he conceived the idea of creating an ethnographic collection that would illustrate, by means of examples both ancient and modern, that “primitive stage of human culture” known as the “Stone Age.” As his collection grew, he was able to arrange exchanges with museums all over the world: “no opportunity was lost, no person who could help me forgotten.”

From 1894 to 1902 he corresponded with University Museum director Stewart R. Cuiin, asking if Cuiin could send him examples of types of implements still missing from his collection. Cuiin sent Giglioli a few specimens in 1894 and another, larger installment in 1897. For his part, Giglioli sent to Cuiin in 1894 three specimens of Tuscan game disks called ruzzoie and then, in 1897. a shipment of over seventy objects—”gleanings from all parts of the World,” “odds and ends” which Giglioli hoped would “fill up gaps in your grand collections.” In his handwritten inventory of the 1897 shipment, Giglioli lists the New Guinea weaving as “I Fine small mat, Merat or Djamma (Tabi Is) N. W. New Guinea.”

The University Museum’s terfo has a maximum width of 67 centimeters, and a length (excluding fringes) varying from 58 to 65 centimeters. The length of the fringes varies from up to 14 centimeters at one end to up to 18 centimeters at the other. The piece is in a plain weave, with a plaid pattern formed by palm fibers dyed red-brown and blue-gray. In measurements, colors, and pattern, it appears to resemble de Clercq’s No. 261 (the woman’s “sarong,” worn front and back, measuring 66 by 63 cm), and was possibly collected at about the same time. it is shorter than the modern woman’s skirt (memda, Fig. 13) collected by Michael Howard, and unlike that example consists of a single loom width.

The Sober of Northern Irian Jaya

The Sobel, at present numbering around 2.000. are part of a small group of Austronesian speakers living along the coast and on offshore islands from the Kumamba islands (which include Liki and Armo) just west of Sarmi to the coastal villages of Tarfia and Arnbora in the east. This group constitutes a fairly cohesive cultural unit with a traditional network of social and eco­nomic relations that once included the produc­tion and exchange of ofterfo.

Sobei is the language of the original inhabit­ants of the town of Sarmi. It is also spoken in the adjacent villages of Sawar and IBagaiserwar. It is difficult to tell just how long the Sobei have lived where they are at present. Austronesians migrat­ing from mainland Southeast Asia via the Philip­pines reached the eastern Indonesian islands such as Timor some time around 2000 BC. Over the next millennium they settled along northern New Guinea and out into the Pacific. Today there are over 80.000 speakers of Austronesian languages in Irian Jaya, out of a total population in the province of over one million.

The Mamberamo river provides an impor­tant dividing line for the Austronesians of Irian Jaya. with marked linguistic and cultural dif­ferences distinguishing those living to t he west of the river from those, like the Sobel, living to the east. The differences may in part reflect differences among the initial migrants, but are also to some extent the result of distinct histo­ries. Compared to the other Austronesians to the west, the Sobei and their neighbors appear to have been considerably more isolated from out­side contact before the coming of Dutch colonial rule during the latter half of the 19th century.

The Sobei have a history of relatively close contact with the Dutch dating from the early part of the 20th century when the Dutch established a military base east of the Mamberamo river, cen­tered on Jayapura. Their influence in the Sobei area proper began primarily with the founding of a mission by the Utrecht Mission Society on Yamna in 1911. (During the Second World War the area was occupied by Japanese and then Amer­icans; the Dutch returned after the war.) The Sobei readily accepted many aspects of this new culture. Thus, when Dutch missionaries arrived in 1912 “the Sobei accepted Christianitywithout hesitation, destroying the men’s ceremonial house and stopping initiations” (Sterner 1992:5S).

While retaining some animistic beliefs and practices, the Sobei consider themselves to be Christians. A similar point can be made regard­ing other aspects of their culture. Their eco­nomic life, social and political organization, and many of their other beliefs and practices have changed a great deal as a result of external influ­ences, but they have retained significant aspects of their traditional way of life as well. Thus, the woven cloth they call terfo has survived through centuries of relative isolation (other aspects of Sobei culture are discussed in Sterner 1992; and Sterner and Sterner 1991).

Prehistoric Woven Cloth in Island Southeast Asia

Terfo represents perhaps the purest surviving example of what was probably the first form of weaving practiced by the Austronesian peoples. The early development and spread of weaving in Southeast Asia remains one of the least known parts of the region’s prehistory. Much of the writing on the topic is little more than sup­position based on very little data. Climate is the main culprit. Cloth and the mainly wooden implements used to make it tend not to last long on humid tropical islands.

Despite the often repeated assertion that early Austronesian migrants were weavers, evi­dence of this is scant. Instead, the evidence seems to support Bellwood’s argument in favor of “a strong Austronesian tradition of barkcloth rather than woven clothing” among the early Austro-nesians (1978:173 Weaving was, how­ever, probably practiced by at least some of the early migrants to Island Southeast Asia. Bellwood suggests that loom weaving may have spread into the region during the 1500-1000 BC period (1997:150). In the case of Austrone­sian peoples of Irian Jaya. while making bark-cloth appears to have been an almost universal tradition (see Howard 1996), weaving seems to have been limited to the Sobei and their imme­diate neighbors. The implication is that only some of these early migrants brought the knowl­edge of weaving with them, and for some reason or reasons the skill was not subsequently widely dispersed among others.

Efforts to reconstruct the prehistory and evo­lution of weaving in Southeast Asia and elsewhere have paid particular attention to the types of looms employed. For example, what Yoshimoto refers to as the Sobel type I loom is placed first in his exhaustive study of Indonesian looms, suggesting it is the precursor of all other types (Yoshirnoto 1990:4).

The study of weaving techniques has been of interest to students of the prehistory of weaving in Southeast Asia and particularly those inves­tigating the diffusion or evolution of ikat tech- niques. The relatively plain appearance of the cloth woven by the Sobei is in sharp contrast to the far more intricate ikot cloth woven by oth­ers who still use non-cotton fibers_ The simple weaving technique of the Sobei is probably rep­resentative of the techniques used by the first Austronesian weavers. Thus, it can be argued that terfo provides us with a niques view of the type of cloth made by the earliest Austronesians, repre­senting a survival of pre-ikat, as well as pre-cotton weaving and dyeing on the earliest types of looms known in Southeast Asia.

Previous Studies of Weaving in the Salmi Area

The literature on weaving in the Sarmi area is very limited and at times confusing. The first published description of ofterfo comes from F.S.A. de Clercq’s account of his travels to northern Irian Jaya in 1887 and i888 (Clercq and Schmeltz 1893). De Clercq collected a large number of items from along the coast, most of which came to be housed in the Rijksmuseum voor Volkenkunde in Leiden in 1893. His report describes three pieces of ofterfo, made of palm leaf thread in various colors, that would have been worn by a woman as a sarong, a hip-cloth, and a girdle, respectively (Clercq and Schmeltz 1893: nos. 261, 262, 263a). A few years later, G.A.J. van der Sande. who informants told us that this more than anything else convinced Nenek Pueba of the importance of her craft.

Pueba Merne eventually passed her knowledge on to another woman, Nebamer Sefa, and for a number of years Nenek Nebamer was the only person who knew how to weave in Sawar. Fortu­nately, by the early 1970s interest in lerfo weaving apparently underwent a revival. Writing in their field journal in 1974, the Sterners note that “the lady who can make the handmade cloth is quite honored and now there are several young people who want to learn to make it. Older people are also starting to want to learn it (Sterner and Sterner 1991:105). The year before, three pieces of lerfo had been presented to a visiting govern­ment official modern adaptation of the old practice of using lerfo to help cement relation­ships with significant outsiders.

By the early 1980s a handful of women in Sawar had learned how to make lerfo. One of the young women who was particularly keen on weav­ing was Cornelia Zeifran (b. 1966). In 1982 the Sterners remarked in their field notes that the people of Sawar were anxious not to let the tradi­tion of weaving terfo die out. even though “they do not use it for anything functional any longer,” and noted that they hoped to make money from it (Sterner and Sterner 1991:66).

Selling lerfo is an entirely new function for the cloth, but to date it remains a largely elusive goal. Sawar has had relatively few visitors in recent years, both because of the difficulties in getting there and security-related restrictions on travel in the area, and because few other market oppor­tunities have emerged. Nevertheless, the desire to capitalize on what they perceive to be an impor­tant cultural commodity persists—spurred on in part, so informants’ comments have indicated to us. by encouragement from outsiders who have highlighted the unique nature of this product and its overall significance as the only woven cloth from Irian Jaya.

A combination of cultural pride and the search for additional sources of income led to the founding in 1992 of a village cooperative to sup­port local arts. The cooperative (known as Taman Seni Karya Bahagia) was to promote traditional carving, music, and weaving, with weavingbeing its primary activity. Its members included about ten weavers, but since its founding production of cloth has been limited. Traditional carving (e.g., of canoe decorations) had died out quite a few years earlier and the group sought with only limited success to revive this craft.

Over the past few years most of the pro­duction of terfo in Sawar and Bagaiserwar has been associated with performances at govern­ment-sponsored traditional dance fetes, where villagers wear the cloth as part of their dance costumes and sometimes try to sell pieces. Thus, both Sawar and Bagaiserwar have been invited on separate occasions to participate in the tradi­tional dance competition sponsored by the kobu­palen (an administrative region) of Jayapura each August (Fig. 14). Such activities have played an important role in keeping the tradition of lerfo weaving alive. Without further incentive, how­ever, the momentum of the past two decades may be lost and the craft may once again find itself on the verge of extinction.


The authors would like to thank the CIDA­funded Eastern Indonesia University Develop­ment Project for support of their research. A return trip to Sawar in 1966 included on the research team Marlina Classy, Yosephine Gri­apon, and Hanro Yonathan Lekitoo from Candle rawasih University, with supervision provided by Michael Howard and Emmed Prioharyono of the University of Indonesia. Since 1995 the people of Sawar have visited Naafi Sanggenafa in Jayapura on numerous occasions. We are grateful for their interest and support. Photographs are by Michael Howard, unless otherwise indicated.

Cite This Article

Howard, Michael C. and Sanggenafa, Naffi. "Terfo." Expedition Magazine 41, no. 3 (November, 1999): -. Accessed February 29, 2024. https://www.penn.museum/sites/expedition/terfo/

This digitized article is presented here as a historical reference and may not reflect the current views of the Penn Museum.

Report problems and issues to digitalmedia@pennmuseum.org.