Kamasan Art in Museum Collections ‘Entangled’ Histories of Art Collecting in Bali

Museum collections of art from Kamasan village attest to historical engagements between Balinese subjects and colonizers in the context of colonial state formation. Paintings circulated as war booty, acts of prestation and gift giving, souvenirs, and as the subjects of academic study, motivated by colonial concerns to preserve authen-tic, traditional Bali. Although slanted towards the perspectives of colonial collectors, this survey allows the muted voices of colonized subjects to emerge. Exchanges of people and goods reconfigure linear views of Balinese art history by challenging presuppositions about the cultural value of art being corrupted by exchange in a market economy.

Bijdragen tot de Taal-, Land-en Volkenkunde 170 (2014)  There are several public collections of Kamasan art in museum institutions around the world, most notably in the Netherlands.1 Given that museums housing collections of material culture now define their mission in terms of the relationships that lie behind collections, knowing how objects came into their care, how they relate to the culture of the people who made them, and what cultural-heritage significance they have today is paramount. This survey of collecting Kamasan art proceeds from the understanding that collecting art involves overlapping relationships between people and objects. Using the notion of 'entanglement' proposed by Thomas (1991), it explores the interdependencies, mutual influencing, and social relationships embodied in museum collections. Although the term 'entanglement' is more a descriptive term than a theoretical model, it has the analytical potential to shape the way we think about how cultural and social change occurred in this period. Thomas argues that 'entanglement' is a way to incorporate local perspectives in wider world histories. In doing so, he places the study of museum collections within a bigger picture of global interactions, involving people, ideas, and material goods. This article considers how collecting art might be located within a broader narrative of colonial expansion by exploring the collection of Kamasan art in Bali during the colonial period.
Kamasan paintings depict stories from epics of Indian and indigenous origin, relating the lives of the deities, the royal courts, and-sometimes-even commoner families. When displayed in temples and family compounds, paintings are hung only for the duration of festivals or ceremonies. At other times they are folded, or rolled, and stored alongside other ceremonial equipment. Paintings are normally produced on cotton cloth (kain belacu), made from cotton grown on the island of Nusa Penida or cotton imported from Europe. They are also painted on bark cloth (daluwang) sourced from around the archipelago and on wooden boards or panels (parba) that form the back of bed-like offering platforms (taban). The narratives depicted on them serve a didactic and devotional function and the paintings acquire many layers of meaning in the context of their display. They are intended to gratify and entertain the gods during their visits to the temple, as well as the human participants in ritual activities.
Not a lot is known about the origins of Kamasan art, since the earliest physical examples of cloth paintings go back only as far as the early nineteenth 1 Throughout this article frequent reference is made to paintings, photographs, and other objects in museum collections. Where possible the object reference or accession number appears in square brackets. Many of these are accessible via the digital collection and photographic archives on the websites of the institutions concerned.
century. Although there is little material basis to illustrate precisely how Kamasan art came into being, it is generally associated with the Majapahit court culture, which came to Bali in the late fifteenth century. A similar style of painting was practised throughout Bali at least until the early twentieth century and continues to be produced by small numbers of artists in other Balinese villages; Kamasan, however, is the only village in Bali where this style predominates.2 The most important distinction between Kamasan art and other visual art traditions of Bali is that the style of Kamasan has not been superseded by the newer styles that took root in other villages. This is not to imply that artists in Kamasan adhere to an unchanging model, though the art is highly conventionalized in that artists work according to certain parameters (pakem) and adhere to strict proscriptions in terms of iconography.
Artists interpret these stylistic boundaries in different ways, introducing innovations while maintaining that they belong to an unchanging tradition of great antiquity.
The style of art produced in Kamasan is often called 'wayang painting' with reference to its shared roots with the shadow-puppet (wayang) theatre. Artists use the same term for the figures (wayang) they paint, which are depicted in almost the same manner as flat Balinese shadow puppets except in threequarter view. The term 'narrative art' also describes Kamasan paintings because they depict versions of stories and myths found in written, oral, and performance genres. Other terms, including 'temple art' , 'traditional art' , and 'classical art' , are also applied to Kamasan painting. Although artists in Kamasan have no standard description for their art, they usually include the name of the village as a referent. They refer to the paintings as 'Kamasan paintings' (lukisan Kamasan) or as painted in the 'Kamasan style' (gaya Kamasan) or use other similar phrases to identify them.

Kamasan and the Court of the Dewa Agung
Today Kamasan is a village of four thousand people located between the east coast and the mountain ranges of Gunung Agung on the island of Bali. It is part of Klungkung, the smallest of eight administrative districts (kabupaten) in the province of Bali. Although now incorporated within the Indonesian state, the administrative arrangement of Kamasan village reflects the historical role Bijdragen tot de Taal-, Land-en Volkenkunde 170 (2014) 250-280 of the village as a community of artisans working in the service of Balinese rulers. The village is divided into wards (banjar) based on the specialized services once provided by artists to the court, including goldsmiths (pande mas), smiths (pande), and painters (sangging). The artists were members of the commoner caste ( jaba), and in exchange for decorating royal palaces and temples they were provided with tracts of farming land (sawah ngayah). In the arrangement of customary villages (desa adat), which coexist with the administrative villages, Kamasan belongs to Gelgel, the former seat of royal power. From the late fifteenth or early sixteenth century, Gelgel was home to the court of the Dalem, a dynasty established by the Javanese Hindu-Buddhist kingdom of Majapahit. The following two centuries, when Bali was united under one ruler, are widely perceived as a 'golden age' . Although this may be an imagined ideal rather than a political reality, the glory and prosperity associated with Gelgel is evoked in Western and Balinese histories of the island alike (Creese 1991a:255).
The 'golden age' ended around the late seventeenth or early eighteenth century, when the reigning king, Dalem Di Made, was removed from power and chief minister (patih) Gusti Agung Maruti assumed power. In around 1686, nobles loyal to the Dalem attacked Gelgel and reclaimed the throne from Maruti on behalf of his son Dewa Agung Jambe.3 Dewa Agung Jambe re-established the dynasty; however, he relocated his palace from Gelgel to the town of Klungkung.4 The dynasty adopted the reign title Dewa Agung at the same time. The new palace, called Smarapura and meaning abode (pura) of the God of Love (Smara), was the residence of the ruling dynasty until 1908, when Klungkung became the last Balinese kingdom to succumb to colonial rule. At this point the entire island of Bali was incorporated into the colonial administration of the Netherlands East Indies. This event is known by the Balinese as the Puputan, meaning the 'finishing' or 'the end' , which 3 There are unresolved questions about the re-establishment of the Klungkung dynasty as the successor dynasty to Gelgel, particularly about the amount of power vested in the Dewa Agung of Klungkung as new kingdoms emerged under local rulers who had been ministers to the king of Gelgel (see Creese 1991a;Vickers 2005;Wiener 1995). When Dewa Agung Jambe (1686-1722) decided to relocate the palace to Klungkung, the Gelgel site was left in ruins; however, the Gelgel palace was re-established in the eighteenth century as a branch line of the Klungkung royal family when Dewa Agung Ketut Agung (Wuruju Wirya) sought permission to settle there (Creese 1991b:406-9).
Bijdragen tot de Taal-, Land-en Volkenkunde 170 (2014) 250-280 The painted ceilings of the Kerta Gosa, 2011 refers to the slaughter of the Klungkung royal family by (or their ritual surrender to) the Dutch on 28 April 1908. The two pavilions (bale) that survived the Dutch attack feature spectacular painted ceiling panels produced by the artists of Kamasan. These are regarded by present-day artists as the most enduring testimony of their relationship with the court (Fig. 1). The pavilion on the north-east corner of the site, known as the Kerta Gosa, features nine levels of paintings. These include the story of the mythical bird Garuda from the Adiparwa; stories from the animal kingdom (tantri) referring to the social and moral duties of leaders; and scenes from the 12-month prognostic calendars (palindon). The most well-known story associated with the Kerta Gosa is Bimaswarga, depicting Bima's mission to release the souls of his step-parents, Pandu and Madri, from the Balinese realm of punishment. On this journey Bima witnesses an extraordinary range of sinners receiving their punishments, depicted in the scenes along the lower levels of the ceiling.
The second pavilion is the Bale Kambang, set amidst a water garden (taman gili) and accessed by a pathway built across the pond. The panels of the Bale Kambang relate scenes from the life of a commoner couple with eighteen children, known as Pan and Men Brayut; the story of the Buddhist Sutasoma, Bijdragen tot de Taal-, Land-en Volkenkunde 170 (2014) 250-280 who renounced kingship in favour of a spiritual path in life; and scenes from the star calendars (palelintangan) depicting the various named days and their attributes.
Although Kamasan artists believe that the paintings at the Kerta Gosa are part of a much older tradition, the earliest reference to the painted ceilings dates to 1849, during the joint reign of Dewa Agung Isteri Kanya and Dewa Agung Putra ii (Vickers 1991). Balinese oral histories also relate the story of the nineteenth-century artist I Wayan Gede Mersadi (1771-1830), said to have been held in great esteem by the Dewa Agung (Kanta 1978). This artist belonged to the Pulesari descent (dadia) group, and is said to be the first of the Pulesari in Kamasan to become a painter under royal patronage. The Dewa Agung instructed the artists of Kamasan to depict the figure of Modara, known from the epic Bomantaka, relating the death of Boma. The rendition produced by Wayan Gede Mersadi made such an impression that the Dewa Agung began to associate the artist with the figure he had painted, affectionately bestowing the name Modara on him. Many old paintings in collections around Bali are attributed to Modara. The descendents of Modara in Kamasan have a small painting attributed to the artist on a decorative wooden stand, depicting the figures of Suprabha and Arjuna and their servants (Fig. 2).
The reputation of Modara, and the continued attribution of paintings to him, demonstrates that courtly patronage remains an important pedigree amongst Kamasan artists. Although the current Dewa Agung, Ida Dalem Smara Putra, does not maintain a collection of Kamasan paintings at the palace, as a figure he is still regarded as an important patron of Kamasan art.5 Nevertheless, as colonial powers encroached on Balinese kingdoms, Kamasan artists and their paintings were invariably caught up in new relationships. The following section locates Kamasan paintings in the development of ethnographic collections in the Netherlands East Indies. After describing specific instances of acquisition, the motivation and circumstances of particular collectors will be considered.

Collecting Art in the Colony
Collecting material culture in the colony for the national museums established after the unification of the Netherlands in the nineteenth century has been described as more opportunistic than systematic (Ter Keurs 2011). The collecting policy of the colonial government was to delegate collecting to officers in the field, whether they be civil servants, military officers, missionaries, or private individuals. In this way museums established in the colony and the Netherlands accumulated sizeable collections of Balinese material through arbitrary donations by serving officials, bequests of private collections by colonial officers, as well as the seizure of material goods during attacks on Balinese property. These museums played important roles in developing and imparting knowledge about the colonies, the most significant being the Rijksmuseum Volkenkunde (National Museum of Ethnology) in Leiden and the Tropenmuseum in Amsterdam.6 In the colony, the Koninklijk Bataviaasch Genootschap van Kunsten en Wetenschappen (Royal Batavian Society for Arts and Sciences) was established in 1778 in Batavia, the capital of the Dutch East Indies. Although the Society was a private organization, during the administration of Governor General J.C. Baud (1833-1836) officials posted around the archipelago were instructed to assist in the creation of a museum within the Society, and an annual budget was allocated to run it (Djojonegoro 2006). Throughout the history of the Society, which devolved to the Museum Nasional Indonesia (National Museum of Indonesia) in 1963, individual and board members were drawn from prominent members of the colonial government and a few members of the indigenous elite.
The Batavian Society was on hand as the Dutch tried to conclude treaties with Balinese rulers to recognize Dutch sovereignty during the nineteenth century. When the Dutch launched the First Bali Expedition against the ruler of Buleleng in North Bali in 1846, German Sanskrit scholar R.T. Friederich, an 6 The National Museum of Ethnology in Leiden was established in 1837 to house the royal cabinets of King Willem i. Over time the collection came to be associated with the development of Indies studies at the State University of Leiden. The colonial civil-service education programme, which had been based at the Royal Academy in Delft, was transferred to Leiden in 1864 and taught linguistics, geography, and ethnology of the Indies. From 1892 it was formally attached to the State University of Leiden, which then emerged as the principal centre for Indies studies. From 1902 the training of civil servants was based entirely in Leiden. The Tropenmuseum traces its founding to 1864, with the establishment of a Colonial Museum in Haarlem dedicated to research on the natural resources and products of the Indonesian archipelago and Suriname. In 1910 the museum was incorporated into a new Colonial Institute in Amsterdam; in 1926 it was relocated to its current premises, a grand building jointly funded by colonial enterprise, private business, the Dutch government, and the city of Amsterdam. Among other things, the museum component of the Colonial Institute was charged with preparing civil administrators, scientists, business people, teachers, nurses, and soldiers for service in the colonies. After Indonesian independence, the Colonial Institute was initially known as the Indies Institute before, in 1950, it became the Koninklijk Instituut voor de Tropen (kit, Royal Tropical Institute).
Bijdragen tot de Taal-, Land-en Volkenkunde 170 (2014) 250-280 assistant-librarian at the Batavian Society, joined the military mission to Bali, charged with 'bringing back any manuscripts, inscriptions or images he managed to collect' (Wiener 1994:354). Interest in manuscripts flourished under the directorship of W.R. van Hoëvell during the 1840s, as part of a growing interest in ancient Hindu-Buddhist Javanese culture. As it happened, during the sacking of the royal palace in Singaraja, Friederich failed to secure manuscripts because he was beaten to them by Heinrich Zollinger, another member of the Society who had joined the trip. Instead, he went to stay in Badung with the Danish trader Mads Lange. Thanks largely to the networks of Lange, he obtained several kakawin manuscripts and learnt how to read them from brahmana informants (Vickers 1989:81-3). Friederich's collecting method exemplifies a pattern for official collecting in which paintings were of lesser interest than texts and, if they were collected at all, seem to have been acquired as mementos rather than as subjects worthy of academic interest. Mads Lange (1807-1856), whose connections enabled Friederich to collect and study manuscripts, is the earliest documented European collector of Kamasan paintings. Appointed by the Dutch as their agent on Bali in 1844, one of his many roles was to persuade Balinese nobles to sign treaties with the Dutch (Schulte Nordholt 1981: 38). In 1851 Mads Lange gifted about twenty objects to his country of birth, including three manuscripts and two paintings, now in the collection of the National Museum in Copenhagen. His one-time assistant, Ludvig Verner Helms, donated another painting to the same collection (Wulff 2002:116). Little is known about how the two men acquired their paintings, although a Balinese portrait of Mads Lange hints at possible relationships with artists (illustrated in Schulte Nordholt 1981: 17). In the course of his work as a trader and intermediary between the Dutch and the Balinese nobles of South Bali, Mads Lange became a rich man in an environment where 'gifts and profits went hand in hand' (Schulte Nordholt 1981:31). In the context of these diplomatic missions it is possible that he was gifted the paintings.
Meanwhile, the ruler of Buleleng continued to reject Dutch demands to enter into treaties and a Second Bali Expedition was launched in June 1848. Once again, the Dutch failed in their mission to subjugate North Bali. This time, when the royal palace at Singaraja was demolished, officers helped themselves to souvenirs, including a number of paintings (Hinzler 1986:10). One of the Kamasan paintings removed from the palace found its way into the collection of the Utrechtse Zendings Vereniging (Utrecht Missionary Society), reproduced and described by Cornelis Marinus Pleyte (1901).7 The second painting Bijdragen tot de Taal-, Land-en Volkenkunde 170 (2014) 250-280 connected to this expedition is a flag (kober) in the National Museum of Ethnology in Leiden , said to have been carried by Balinese opponents during battle. Both sides of the cloth bear the image of a demon general holding a club and standing between two trees, a common subject for flags.
The following year the Dutch enlisted more personnel when they embarked on the Third Bali Expedition. The town of Singaraja was occupied and the population fled, leaving soldiers and officers free to march through the city, looking for valuables. The commander of the force, General Michiels, took up residence in the palace. One of the officers, a Captain Van der Hart, describes wandering the city. He reports that anything of value was removed by departing residents, with the exception of the palace, where there were paintings worth taking (Drieënhuizen 2012:55). As the palace had been occupied, it was presumably not possible for departing Balinese to take the paintings with them. Although the Dutch did not regard them as high-value lootable goods, they took them nonetheless.
Naval officer brothers Henri and William Quarles van Ufford acquired a painting which they later donated to the National Museum of Ethnology in Leiden [3457-1]. As naval officers they were probably not part of the land combat, but were aboard the ship docked at Padang Cove, in May 1849, from where the Dutch occupied Pura Gua Lawah and planned their move on Klungkung. The painting may have been traded or swapped onboard the vessel, although the brothers reported that wherever they went, there were countless opportunities to either loot or offer gifts in exchange for goods (Drieënhuizen 2012:55). On the Klungkung coast, Dutch land officers engaged with Balinese troops at Kusamba. Once the Balinese retreated, General Michiels set up camp in the marketplace besides the royal palace. In a much celebrated night attack, Balinese forces fatally wounded the general, forcing the Dutch to abandon their planned assault on Klungkung and sign a peace treaty with the Dewa Agung (Wiener 1995:178-81). Nonetheless, Kamasan paintings were looted from the Kusamba palace and found their way to Berlin. Ten Kamasan paintings were donated to the Museum für Völkerkunde (Berlin Ethnographic Museum) by an E. Mayer some time before 1894 (Bastian 1894).8 who collected material in the Indies in 1898-99 for the 1900 World Exhibition in Paris. He was curator of the museum of the Batavia Society from 1902 until his death in 1917. The narrative subject of the Kamasan painting from the Singaraja palace is the temptation of Arjuna (Arjunawiwaha). 8 Three of these paintings are illustrated in Vickers (2012:34-5, 48, 86) The Kusamba palace was established during the late-eighteenth-century reign of the mad king Dewa Agung Sakti. His son, Dewa Agung Panji, became the de facto ruler and built the Kusamba palace to distance The peace settlement held until 1908, when Klungkung became the last of the Balinese kingdoms to succumb to colonial rule. After the Dutch attack of 28 April 1908 the royal treasures were systematically removed and assessed by H.J.E.F. Schwartz, then Assistant Resident of South Bali and later curator of the museum of the Batavian Society (Brinkgreve 2010:120). The majority of the treasures were sent to Batavia and many are now part of the National Museum collection in Jakarta (Sitowati and Miksic 2006:152-4). Objects of lesser value remained in Bali and were either sold or kept by the Resident to establish a small museum collection.9 According to a former palace servant there was minimal local looting of the royal treasures before the Dutch occupied the palace, because Klungkung's queen had promised a life of misery for any of her subjects who dared to loot her things after her death (Wiener 1994:367 n. 8). Although local looting was negligible, some objects found their way into Balinese hands. For instance, local villagers who had served the court as musicians took the palace gamelan into custody (Toth 1975:65).
Various objects and fittings said to have been salvaged from the palace were also sold to foreigners in the years following. It is possible that Balinese agents fabricated these connections to boost their value, since it was known that objects associated with the Balinese courts were highly sought after. For instance, in 1924 the Dutch collector W.O.J. Nieuwenkamp purchased a winged lion from a Balinese seller who claimed that the Dewa Agung had used it to take offerings to the temple (Brinkgreve 2010:119). In the 1930s anthropologist Jane Belo and her husband, Colin McPhee, purchased a wooden door from the Ubud artist Gusti Nyoman Lempad. He told them that he had purchased the door in 1919 from the Klungkung palace.10 Kamasan paintings were acquired by colonial officers as gifts. Members of the Raad Kerta, the court which sat at the Kerta Gosa, presented a painting himself from his father (Vickers 1991:66). The battle referred to here took place during the joint reign of Dewa Agung Isteri Kanya and Dewa Agung Putra II. 9 Wiener (

Maintaining Balinese Tradition
Maintaining Balinese tradition was a major contradiction of Dutch colonial policy. The Dutch administration aspired to keep the Balinese as traditional as possible, and perhaps make them 'even more Balinese than they ever had been' (Pollmann 1990:14). These policies were informed by a circle of Dutch experts on Balinese culture, committed to ensuring that the 'basic values' of Balinese society remained 'intact' (Ramseyer 1977:239). The Dutch justified their conquest of the royal courts on the grounds that they were ridding Bali of despotic rulers and simplifying village administration, yet they had to contend with the demise of traditional arts brought about by the removal of courtly patronage. from other forms of compulsory labour so that they might use their time to pass on their skills to other Balinese and thus ensure that the painting tradition of Kamasan did not die out.12 Artists were commissioned to produce a new set of paintings for the Kerta Gosa in around 1918 as part of widespread restoration projects conducted around the island in the wake of a devastating earthquake in 1917. The Dutch administration also maintained the former function of the Kerta Gosa as a court of law (Raad Kerta), though with different arrangements for the hearings. Klungkung rulers reportedly used the Kerta Gosa for an annual court session of the rulers of Bali and for monthly meetings between the Dewa Agung and local officials, as well as to receive foreign visitors and palace priests (Warsika 1979:11). Under the Dutch the court dealt with civil and criminal matters, hearing a total of 183 cases in 1928 and 678 in 1937 (Van der Kaaden 1938:37-9). A Dutch official led the court with a deputy and other members, including brahmana priests (pedanda) and administrators (punggawa). From at least 1929 onwards the deputy was the Dutch-appointed regent (negarabestuurder) Dewa Agung Oka Geg , nephew of the former Dewa Agung.13 A photograph of the 1920s depicts a court session with five Balinese and one European man seated on ornately carved chairs around a table (Fig. 3). The Balinese appellants sat on the floor facing them, with the panels of the Bimaswarga looming above.

Tourism in Klungkung
The Kerta Gosa site was also transformed into a tourist attraction. Just a few years after the kingdom was destroyed, a Dutch travel guide recommended that visitors spend a few days in Klungkung (Pareau 1913). The armed conflict between the Dutch and the Balinese was part of the travel narrative, with visitors advised that they would be able to observe 'clear evidence of the effects of our artillery' on the remains of the Gelgel palace (Pareau 1913:75). The three-day island tours promoted by the Dutch shipping line Koninklijke Paketvaart-Maatschappij (kpm) in the 1920s and 1930s included a half-day visit to Klungkung during which visitors went to the Kerta Gosa, Gelgel, and Kusamba, before being served lunch at the government guest house on the street opposite the Kerta Gosa (Morzer Bruyns 1934:79). At this time Klungkung became the centre of a lively trade in art. During the 1930s two art shops were established on the main street, east of the Kerta Gosa, by families from the nearby village of Sampalan. They had sales networks extending around the island and as far as Singapore. The presence of the art shops and a profusion of itinerant vendors meant that tourists had no need to go searching through the villages of Klungkung to buy art. The Sanskritist Bijdragen tot de Taal-, Land-en Volkenkunde 170 (2014) 250-280 Arnold A. Bake described the transformation of the square in front of the guest house 'into a market' .14 One of his photographs depicts two Balinese women with objects for sale-including paintings, wooden masks, folded cloths, and bells-on the ground in front of them [kitlv 17541]. They are engaged in a transaction with a foreign woman dressed in white, whose back appears partially in the foreground. At some distance behind the three women a crowd of onlookers is watching the interaction. Another photograph, taken by a Dutch traveller in the 1930s, shows the space directly in front of the Kerta Gosa being used as a trading ground: women sit with paintings and painted hats (capil) spread out on the ground in front of them (Fig. 4).
Travel accounts listed Klungkung as one of the best shopping destinations in Bali, describing the women that 'come to your door and drop on their knees before you, unrolling a store of treasures before your eyes' (Yates 1933). Another traveller of 1928 wrote of the 'keen traders' who gathered on the porch of the guest house each afternoon, carrying 'baskets filled with the manifold artistic products of the island' (Zobel in Vickers 1994:197). Visitors portrayed the women as skilled transactors, noting that they 'hate to lose a sale' (Zobel in Vickers 1994: 197). Yates (1933) declared that the goods sold had little intrinsic value, being 'taken from a native hut in a rush to swell the family finances when a foreigner appears on the scene' . Although these remarks convey almost nothing about the intentions of the Balinese women, the same assumptions were made by other foreign collectors, namely that the Balinese could find a value for anything.
Many of the Kamasan paintings in museum collections were acquired as tourism gained momentum. In addition to paintings on cloth, many objects painted in Kamasan style-including coconut water-containers, fans, hats, and bamboo food covers-entered museum collections in the Netherlands from the 1920s onwards.15 Tourist collectors are generally regarded with contempt by 14 Bake was part of the entourage accompanying the Indian poet Rabindranath Tagore on his tour to Bali in August 1927. The description of Klungkung comes from a letter he wrote to his mother: 'Touristen komen er heel weinig, afgeschrikt door den tocht erheen. Om de een of andere reden kan je in Kloengkoeng de meeste dingen zoo van oudheden krijgen. Nauwelijks waren we in de pasangrahan aangekomen of er was een hele markt van allerlei verkoopende vrouwen voor het huis op het pleintje het was een gekoop en geding van je welste. Vooral Suniti kocht los en vast aan elkaar.' Letter from A.A. Bake to his mother, dated August 1927 For examples, refer to the following painted objects in the Tropenmuseum [1927-2; 77-22; 432-4; 809-59; 941-39, 1772-550a; 2746-3] other collectors because the material culture that they collected is considered to exist in 'a very different context to systematically collected objects' (Harrison 2011:61). Ignoring the implications of their own art-buying projects, other collectors often held tourists accountable for the commodification of Balinese art. This attitude is exemplified in the collector W.O.J. Nieuwenkamp, despite the fact that he made a professional career out of collecting in the Netherlands East Indies and supplied Kamasan paintings and objects painted in Kamasan style to several institutions in Europe. For artists without independent means collecting was a source of income, and Nieuwenkamp's collecting financed a total of five visits to Bali over the period 1904-1937, including numerous visits to Klungkung. During his second trip to Bali, in 1906Bali, in -1907, he stayed near the palace and visited the Dewa Agung (Nieuwenkamp 1910:179-84). He stayed at the government guest house during later visits, from where he reported being offered all manner of old things for sale (Brinkgreve 2010:118-9). Nieuwenkamp was generally scornful of things he believed had been produced for the tourist trade. He tried to acquire painted household and ceremonial objects associated with the Balinese courts, like the winged lion already mentioned.16 As well as being a visual artist, Nieuwenkamp produced a vast number of ethnographic reports and publications of his travels (see Carpenter 1997).17 This places him within a wider group of scholarly collectors who distinguished themselves from tourist collectors. The following section looks at the circulation of paintings amongst this group to emphasize the relationships between colonial policy, collecting, and scholarly investigations of art. 16 See Reichle (2010:237-9, 273) for other examples of painted objects associated with the palaces, including storage boxes for textiles and palanquins to carry members of the royal family. 17 The initial impetus to visit Bali probably came from Gerrit Pieter Rouffaer, a scholar of Indonesian arts and crafts, who spent six weeks on the island in 1888. Before his first trip to the Indies, in 1904, Nieuwenkamp saw Balinese drawings from the collection of Van der Tuuk in Leiden, and reportedly studied them to familiarize himself with Balinese painting and taught himself to paint in Balinese style. Hendrik Neubronner Van der Tuuk (1824-94) assembled a major collection of Balinese drawings on paper, now held in the library of Leiden University (Hinzler 1986). Van der Tuuk was probably the first scholar collector to commission work from Balinese artists for scholarly purposes.

Scholar Collectors
Once Bali was fully colonized the population of resident ethnographers increased. Most were in the employ of the colonial government, although they also collaborated and competed with a more eccentric circle of artists and scholars that had Walter spies at its centre. Moving between Bali and Java to study archaeology, the art of temple reliefs, and Old Javanese and Balinese texts, the resident ethnographers regarded Bali as a repository of classical Javanese culture and civilization, partly 'pure Old-Javanese' and partly an adaption of it (Hooykaas 1973:1). Kamasan paintings were of particular interest to scholars trying to identify the narrative reliefs of Javanese temples and match them to extant literary sources. One member of this set, Thomas Anne Resink (1902-1970, assembled a personal collection of about one hundred classical Balinese paintings, many from Kamasan, acquired by the National Museum of Ethnology in Leiden shortly after his death.18 Formally trained as a hydraulic engineer, Resink was born into a prominent Eurasian family in Yogyakarta. His mother, Anna Jacoba Resink-Wilkens, was well known as a collector of Javanese antiquities. Resink went to Bali in the late 1920s, probably to flee a difficult relationship with his mother who had difficulty accepting his homosexuality. Resink had many contacts in Bali, including the artist Walter Spies, who had been employed as a piano teacher in the Resink household.19 Resink was initially accommodated in the Ubud palace (puri) of Cokorda Agung Sukawati and later moved to Tabanan, in close proximity to the traditional artists of Krambitan (Stuart-Fox n.d.). Although few specific details are known of how Resink acquired his paintings, he had ample opportunities to buy art through his involvement in the management of the Bali Museum and during an assignment to survey the art and craft industry on behalf of the government. The latter was part of Dutch initiatives to improve the economic well-being of their Balinese subjects by finding new markets for art. In 1936 Resink reported that art sales in Klungkung averaged 1,200 guilders per month, based on street sales and the earnings of two 18 The major study to date on the Resink paintings is Miedema (2004), a thematic discussion of eighteen of the Ramayana paintings. Brief descriptions of his other paintings are found in an exhibition catalogue, including a short essay by Resink (1961)  art shops (Soekawati 1941). Although the Klungkung figure was less than a third of total sales in Denpasar and less than half of the sales attributed to European art dealers, it was similar to the colonial capital Singaraja, the entrance port for tourists to Bali. Other hints about Resink's motivation can be gleaned from his publications. Although he wrote mainly on the narrative reliefs of East Javanese temples, Resink referred to Balinese paintings in the context of his scholarly arguments. With reference to a Balinese painting of the Sudamala, Resink (1939) argued that the order of the narrative reliefs at Candi Tigowangi in East Java had been incorrectly described. Hidden in this argument, Resink stated that what many scholars regarded as a 'deviation' (afwijking) in Balinese art was not due to ignorance, but was intentional on the part of the artists (Resink 1939:284). In a later publication, Resink (1965:466) compared Javanese sculptors of the past with the Balinese artists he knew, saying that they enjoyed 'great personal freedom' with regard to what they depicted. In the same article Resink (1965:439) explained that he had identified the narrative reliefs of Candi Kedaton after showing photographs of the Javanese site to the uncle and father-inlaw of Cokorda Agung Sukawati, who recognized the story depicted on them. As the above makes clear, different members of the Balinese aristocracy took on patronage functions previously associated with the Klungkung rulers by facilitating the access of scholars and officials to various sites and objects related to Kamasan art. In particular, members of the Ubud royal family actively cooperated with the Dutch regime, playing host to influential and affluent European visitors.20 In addition to having their own collections of Kamasan art and displaying it around their homes, the Ubud royals took foreign guests to visit Kamasan (Fig. 5).21 The collector J.L. Moens (1887Moens ( -1954 similarly exemplifies this set of 'entangled' relationships. Like Resink, Moens was trained as a hydraulic engineer and spent most of his career based around Yogyakarta. Moens is best known for his collection of Javanese manuscripts, though he collected widely from around the archipelago and was a member of various committees devoted to the study and preservation of Indonesian art (Djajadiningrat 1952(Djajadiningrat -1957. 20 Cokorda Raka Sukawati, who trained as a civil servant in Java, was responsible for bringing the artist Walter Spies to Bali in 1927. After his move into national politics, he was replaced by his younger brother, Cokorda Agung Sukawati, as head of the palace. 21 Long-term European guests were housed in a house set beside a pond adjacent to Puri Saren Kauh, furnished with classical paintings. For one example, see Tropenmuseum image number 60026304. Moens belonged to the Batavian Society and was an active board member of the Java Instituut, founded in Surakarta in 1919. The Java Instituut established the Sonobudoyo Museum in Yogyakarta in 1935, to which Moens served as vicepresident and curatorial advisor. Moens was also briefly employed as curator of the Bali Museum prior to his internment during the Japanese occupation.
Moens was probably involved in the circulation of Kamasan paintings through the Mangkunegara vii in Surakarta. The Dutch-educated Mangkunegara vii was a committed sponsor of classical Javanese culture and a patron of the Java Instituut. He promoted dance and music performances at the court and played host to the European visitors who studied them.23 A large Kamasan painting appears in a formal portrait of the Mangkunegara vii and his wife Ratu Timur (Fig. 6) Maha 1936aMaha , 1936b. Moens also purchased paintings directly from the Dewa Agung in Klungkung between 1925 and1927.25 At this time, Dewa Agung Oka Geg was a clerk in the colonial administration and was being groomed by the Dutch in preparation for his appointment as a regent. Moens had an  Korawa, c. 1922 (Photo Tassilo Adam, Tropenmuseum image 60035949) ongoing relationship with the Dewa Agung because he, along with several other of the Balinese regents, regularly contributed objects to the Sonobudoyo Museum. By far the most active scholar of Kamasan paintings during this period was Theodoor Paul Galestin (1907Galestin ( -1980, whose academic career was devoted to the study of visual depictions on the reliefs of Javanese temples and in Balinese paintings.26 Although Galestin assembled a collection of Kamasan art, which he used for teaching purposes, his collected paintings are not held in a public institution. Almost all the Kamasan paintings Galestin referred to in his published work are found in museum collections, principally the Tropenmuseum in Amsterdam (see Galestin 1939Galestin , 1947Galestin , 1956Galestin , 1969 and the National Ethnographic Museum in Leiden (see Galestin 1943Galestin , 1953. It is largely thanks to his work that paintings gained admission into the hallowed tradition of philology associated with Leiden University. This academic milieu, with an assumed hierarchy of text over performance and art, led Galestin to compare the visual depiction of narratives with manuscript literature. Galestin identified the stories in Kamasan paintings according to translated and edited versions of texts, explaining where painted versions of narratives corresponded to, or differed from, textual literary sources. In 1934 Galestin travelled to Bali and joined government officer (ambtenaar) Roelof Goris on an official study tour of the island, along with the journalist and art dealer J.A. Houboult.27 Their tour included a day trip to Kamasan, where they visited an artist called I Rene, a client of Houboult. Galestin made only brief comments on the visit as he never regarded the contemporary paintings produced in Kamasan as fitting subjects of academic study, in part because new paintings were too far removed from the ancient past. Galestin (1962) argued that artists simplified their paintings to make them appeal to a market unfamiliar with the subtleties of narrative composition. His views were also informed by the conviction that production for tourists marred their value. This is clear from the disparaging remarks he made on a 1937 visit to Klungkung, as part of a tour organized by the Java Instituut, when he dismissed the Kamasan paintings offered for sale to the group because they had been 'manufactured exclusively for tourists' (Galestin 1938:92-5). When Galestin returned to Klungkung in 1953 (Fig. 7), this time Cokorda Anom Putera, the son of Dewa Agung Oka Geg, escorted his party to Kamasan village.28 The circumstances of the latter visit 27 Between 1928  leave no doubt that the Dewa Agung was still acting as a patron to artists, facilitating their access to foreigners and new markets. While Galestin did not collect art on the same scale as his colleagues Moens and Resink, all three moved in interdependent circles relying on intermediaries like the Dewa Agung and other aristocratic Balinese to facilitate their projects. The trio crossed paths in 1953, when Galestin examined the collections of Moens and Resink in Jakarta, and this probably explains why both eventually went to the National Museum of Ethnology in Leiden. The collections of Moens and Resink represent a different mode of collecting to the single works, or small assemblages, constituting the mainstay of Dutch collections which were acquired as gifts or souvenirs. There was a stronger rationale behind their collecting in that they were interested in the narrative content of paintings and in linking these to manuscripts and to the stories illustrated on the temple reliefs of East Java. In part, this drove them to seek a representative selection of narratives to keep pace with the projects of their philologist colleagues.
Not all scholarly collectors in the colonial period disregarded new Kamasan paintings. In addition to the old Kamasan works collected by ethnomusicologist Colin McPhee, the American Museum of Natural History has several paintings collected by anthropologists Gregory Bateson and Margaret Mead.29 Better known for the large collection of Batuan paintings they commissioned, in 1938 Bateson and Mead purchased six paintings from the compound of Pan Putera (I Ketut Rabeg) in Kamasan. Pan Putera (1915-2011) was a regular host to European visitors, including artists Walter Spies and Rudolf Bonnet, who sometimes stayed overnight in his compound when they brought guests to the village.30 The paintings, including one by Pan Putera and four by his wife Ni Ketut (Made) Remi, all relate scenes from the Bimaswarga.
Although it is not surprising that Kamasan artists were reproducing the most visually striking narrative from the Kerta Gosa, with which foreign eyes would be familiar, the subject also fitted into the Bateson and Mead research project. The shaking kris tree by Ni Ketut Remi [amnh 8308] was also produced filled by various members of the Klungkung royal family since independence. He was also a leader of the Partai Nasional Indonesia (pni, Indonesian National Party) in Klungkung, along with his father. 29 McPhee collected about forty paintings from Kamasan through his employee, I Wayan Ledang, a Kamasan resident with an antique business and silversmith workshop. Mc-Phee's main interest in Kamasan was the gamelan ensemble (semar pagulingan), a musical tradition associated with the Balinese courts, an interest he shared with artist Walter Spies. 30 Interview with Pan Putera and his son I Ketut Widastra (Kamasan, 17-8-2010).
Bijdragen tot de Taal-, Land-en Volkenkunde 170 (2014) 250-280 by Balinese artists outside of Kamasan, including Ida Bagus Made Togog in Batuan, when the anthropologists asked him to depict his 'dreams' (Geertz 1994:90). At the same time Pan Putera's depiction of the various punishments associated with Bimaswarga bears the traits of experiments with new styles of composition. Artists did find ways to simplify their compositions, making them more appealing to foreigners, and in this case Pan Putera divided the cloth into a rectangular grid of three horizontal and four vertical rows, in the manner of a palelintangan.31 Calendars, with their visually simple pictorial compositions, were regarded by some foreigners as more realistic and thus of greater interest than the 'fantastic and highly stylized representations' of other narrative paintings (Holt 1967:175). Although used as charts by the Balinese to determine the intersections of different weeks and days, foreigners regarded the images on them as depictions of everyday life.

Conclusion
The search for the ancient, and for the related presuppositions about authenticity, has been the resounding legacy of colonial-era collectors and writers.
Adopting the Western tradition of art-historical writing, artists like Rudolf Bonnet took a linear view of art history and assumed that traditional painting in Kamasan was poised to die out and to be replaced by the new styles developing in villages like Ubud and Sanur. Bonnet (1936) described lifeless and 'moribund' painting and made explicit the connection between uninspired art and the production of art for shops, while scholars like Galestin (1962) implicitly accepted that contemporary Kamasan paintings were inferior to anything produced in the past. Yet the interdependencies described here destabilize the notion that Kamasan art underwent an uncomplicated progression from a ritual obligation undertaken for gods and kings into tourist product. Although these vignettes may seem devoid of Balinese voices and slanted towards the colonial collectors, the sources available nonetheless convey a sense of the interdependencies and mutual influencing between foreigners and Balinese. Several types of cross-cultural negotiation and exchange took place as Kamasan paintings circulated through networks of royal patronage, foreign collectors, and art dealers. While many early instances of collecting were based 31 Only a small proportion of the paintings collected by Bateson and Mead went to the amnh Collection. This painting was acquired by Leo Haks, when parts of the Bateson and Mead collection were sold, and is illustrated in the auction catalogue of Borobudur Fine Art (Pre-war Balinese 2011:43). on opportune moments resulting from the colonization of the island, they were also motivated by tourism and scholarly interests in Balinese culture. Although the political authority of the Dewa Agung was curtailed during the colonial and independence periods, he continued to be an active patron to artists. In this sense, colonial encounters resulted in shifts but not ruptures in exchange relationships. Foreigners, sometimes described as replacing the Balinese palaces as the patrons of art, literally did fulfil this role when they flocked to the Kerta Gosa to admire the paintings of the former court. At the same time, a direct consequence of the transformation of the Kerta Gosa into the foremost venue for viewing Kamasan art meant that tourists continued to associate Kamasan art with the royal courts. By understanding collections of Kamasan art as products of these 'entanglements' and as reflections of the motivation and circumstances of particular collectors we can re-envisage Kamasan paintings in museums not as remnants of disappearing traditions, but as part of a tradition with a long history of articulation. Artists in Kamasan today produce paintings for traditional ritual purposes as well as for sale to local and foreign markets. The circulation of paintings within different markets does not affect how artists define the parameters of traditional practice, and their art continues to be regarded as traditional, despite innovations in materials, techniques, and the narrative subjects. The paradigms of authenticity associated with earlier collectors are a product of contemporary attitudes to cultural and social change that are challenged by the continued reworking of traditional art in villages like Kamasan.