The British Museum’s Cambodia collection: from colonialism and ethnography to the Khmer ‘art’ market

The BM’s collecting of objects from Cambodia began relatively late in its history, with the first Cambodian object, a carved bamboo flute (BM As1864,1208.9) recorded as entering the museum in 1864, approximately 110 years after its founding. In 1890, two substantial groups of objects entered the collection. The first consisted of 38 ethnographic objects, including a number of pieces of beaded jewellery, donated by Sir Augustus Wollaston Franks, who, as well as being the BM’s first Keeper of British and Medieval Antiquities and Ethnography, was a prolific collector. This group of objects is described as having been collected by C.W. Rosset, a British explorer and collector.

The other group comprised 62 Neolithic objects, including adze blades and earthenware bowls, and human remains, purchased from Henry (Ludovic Henri) Jammes (BM 1890,0208.1-62). Jammes was a Frenchman who seems to have collected the objects in the late 1880s from one or more prehistoric sites in Cambodia, while he was the director of the École royale in Phnom Penh (Finot 1928). No information is provided as to the source or date of acquisition of another Neolithic object, a shouldered celt (BM 1880.1169). The BM accession number suggests that it was acquired in 1880; however, the object entry on the BM website makes reference to it having been “field collected” by Harry Beasley (another British collector of ethnographic objects) and to a sticker that says “1889” suggest that it was in fact part of a group of objects (discussed in the following section) possibly purchased by Beasley from Jammes in 1889, and later donated to the BM. 

The arrival of these two groups of objects, today comprising nearly half of the BM’s Cambodia collection (the purchase from Jammes alone comprising over a quarter of the collection), seems to reflect both the interest and efforts of Franks and Jammes in amassing their own collections, and the BM’s apparent intention to begin to develop this part of its collection. This is further demonstrated by the acquisition of two objects of different types: a bronze statuette of a standing female deity, purchased at a sale of objects from the collection of Stefano Cardu in 1895 (BM 1895,0501.4) (Email correspondence with BM Central Archives, February 2024), and a celadon-glazed ewer donated by Edward Oates (BM 1896.0520.1). The BM records do not hold any information about Oates that has allowed us to find out more about him. Cardu, on the other hand, was an Italian who travelled to Siam and found work in the design and construction of buildings in Bangkok, including the Royal Military College and the Oriental Hotel. In 1914, he donated the majority of the objects he collected during his time in Bangkok to his hometown of Cagliari, Sardinia, leading to the foundation of the Museo d’Arte Siamese. In addition to the bronze statuette described as being from Cambodia, the BM has 14 other objects from Cardu’s collection, including religious statuary and some household items from Siam.

In contrast with what seems to have been a deliberate expansion of the BM’s Cambodia collection at the end of the 19^th century, this period saw a relative decline in the pace of acquisition, with only 24 objects entering the collection between 1900 and 1950. Archaeological and ethnographic objects still made up the majority of acquisitions but religious images, including bronze statues of deities and sandstone Buddha images, made up a far greater proportion than in the preceding period. Continuity can be seen in the similarity of the contexts in which objects that entered the BM’s collection prior to 1950 were collected: like the objects purchased from Jammes or those collected by Rosset and donated by Franks, the majority were acquired from European expats living in Cambodia or Siam and involved in either colonial organisations or influential private enterprises, or from European collectors of ethnographic objects. For example, a sandstone figure of the Buddha seated on the coils of a naga (BM As1924,0714.1) was purchased from Henri Sambuc, a prominent lawyer who worked in French colonial Saigon from 1894 to 1917 (Saada 2007). The BM also purchased a substantial number of objects, including two bronze palanquin hooks possibly made in Cambodia, ceramics and sandstone sculpture, from the collection of Reginald Le May, who worked in Siam first for the British consular service and then for the Thai government. Le May was an influential member of the Siam Society and published on Southeast Asian, and in particular Thai, art. He thus provides an example of the influence and contribution of expat private collectors to knowledge production around Southeast Asian ‘art’, and his collection, like that of Stefano Cardu, attests to the position of Bangkok as a market for Cambodian antiquities.

The majority of objects that entered the collection during this period were donated by Harry Beasley, a British collector and the founder of an ethnographic museum in Kent, who focused predominantly on collecting from the Pacific and the Americas. He donated three religious images, one in 1924 (BM 1924,0707.1) and two in 1925 (BM As1925,0210.1 and 1925,1005.1), and a group of 13 Neolithic artefacts, including bracelets and beads (such as BM 1927,1005.1) in 1927. In the photographs of these objects on the BM’s website, 11 of them can be seen to display or be displayed with stickers or labels that say “CAMBODIA. 1889” and, in the case of nine of them mounted together on a board, the name “H. Jammes”. These labels suggest that the objects were originally collected by Henri Jammes, formed part of the collection from which he sold a larger group to the BM in 1890, and were purchased by Beasley in 1899. The timing of Beasley’s donations of Cambodian objects (made between 1924 and 1927) might make for useful further inquiry. Beasley’s museum, the Cranmore Ethnographical Museum, opened in 1928, and the question therefore arises as to whether the donations he made to the BM prior to this date were connected with aims for the shape of that collection, or his particular collecting focus. (For more on Beasley’s collection and collecting practices, as well as attitudes of institutional collectors towards private collectors, see Carreau 2009 and 2010.)

In 1933, the BM’s Oriental Collections, which included the Cambodia collection and had until that time been housed within the department of Ethnography, were re-organised, with the creation of a new Department of Oriental Art and Antiquities. This new department comprised four sub-departments: Oriental Ceramics; Oriental Sculpture, Metalwork and Antiquities in general; Oriental Paintings, Prints and Drawings; and Ethnography. The BM’s management recommended that the “Oriental Sculpture” then included in the Ethnography section should be transferred to the new Oriental Department, and that “[t]he exhibition of Eastern Religions should be abolished, and the objects constituting it be treated from the point of view of the history of art and culture” (British Museum Trustee Standing Committee Minutes (TSCM), Vol. LXIII (January 10, 1931 to December 9, 1933), 10^th June 1933, p. 4975). Although it is difficult to assess whether this had any immediate impact on the BM’s collecting practices, it is notable that, a few months before, in March 1933, the BM had received a gift from the National Art-Collections Fund, of “the upper part of a stone statue of the eleven-headed Avalokitesvara (BM 1933,0407.1), a fine example of Cambodian work of about the 12^th century” (TSCM, Vol. LXIII (January 10, 1931 to December 9, 1933), 8^th April 1933, p. 4963). We are undertaking further research into this piece, which might be compared to the Metropolitan Museum of Art’s ‘Bust of Hevajra’ (MMA 36.96.4), purchased directly from the École Française d’Extrême-Orient (EFEO), the archaeological service of French Indochina, in 1936. This piece might be described as the first Khmer sculpture of art historical ‘quality’ to enter the BM’s collections, and it is also an important example of the commoditisation of fragmented or broken sculpture as ‘art’.

1951 might be seen as a turning-point in the development of the Cambodia collection: while the overall number of objects accessioned from that year onwards is roughly equal to that for the period pre-1951, two substantial donations of 15 and 21 objects, made in 1951 and 1954 respectively, represent the first instances of objects entering the museum with more than one layer of provenance information (with the exception of the Beasley donations that were possibly purchased from Henri Jammes). That is, this is when we first begin to see that objects initially collected through colonial or expat encounters with the region are circulating within a wider market. 1954 also saw the adoption of the Hague Convention on the Protection of Cultural Property in the Event of Armed Conflict. The BM’s recording of earlier provenance history in relation to some objects might be seen as a side effect of the 1954 Hague Convention, although the other objects accessioned during this period that were not part of the Ridley (in 1951) and Brooke Sewell (in 1954) donations still generally have only limited provenance information.

H.N. Ridley, a botanist and naturalist who worked at the British Museum before the creation of the Natural History Museum, donated 15 archaeological artefacts, including a group of 10 Neolithic shouldered celts and a Neolithic stone disc, all described by the BM as having been given to Ridley by Louis Jacquet, a former French colonial director of agriculture in Annam (Vietnam). The findspot for all 15 objects is recorded as Kompong Thom in Cambodia, and therefore it is possible that the objects formed a single group, all previously owned by Jacquet. The 31 objects donated by P.T. Brooke Sewell at this time, of which 21 are likely to have come from Cambodia, are all described as having been “field collected” by Madame Leopold Robert. The BM’s database does not record her first name, only that of her husband, the first director of the Queen Saovabha Memorial Institute in Bangkok, which, from the end of 1922, housed some of the activities of the Pasteur Institute and the Thai Red Cross Society. BM records do not contain information on when or how Madame Robert’s objects were acquired by Brooke Sewell. Brooke Sewell also made a substantial monetary donation to the BM, establishing the Brooke Sewell Permanent Fund, originally in order to purchase particular Asian material that was of interest to him. It is not clear why Brooke Sewell, as a Norwegian banker living in Switzerland, who never visited London or the BM, should have made such generous donations to the BM in particular.

The scope of the Brooke Sewell Permanent Fund was expanded in the 1950s to cover most of Asia, prompting an increase in purchases of Cambodian objects directly from auction houses and dealers, at a time when the market for Khmer ‘art’ was expanding. The commercialisation of Khmer sculpture had begun with the EFEO, which in the 1920s gave itself the power to sell objects in its care, initially in limited circumstances (Abbe 2021). This power gradually expanded and in 1935, George Cœdès, the director of the EFEO, offered the Metropolitan Museum of Art in New York the opportunity to purchase thirteen sculptures. In a letter to Martin Birnbaum, who represented the Met in these negotiations, Cœdès justified the prices quoted for the pieces by referring to other potential buyers who would be happy to pay those prices and whose offers Cœdès has already declined in favour of the Met (Letter from George Cœdès to Martin Birnbaum, 11 April 1935, AAA_birnmart_3519290. One of the pieces chosen by Birnbaum (‘Buste à 7 têtes’) appears to correspond to the ‘Bust of Hevajra’ mentioned in the preceding section.)

The effect of the expanding market and popularity of Khmer art with museum curators can be seen in discussions around acquisition of a sandstone head of a female deity (BM 1968,0213.1), purchased in 1968 from Spink & Son. The report of the curator recommending the purchase concludes by alluding to the reputational, and even competitive, dimension to institutional collecting at this time: “If refused by the Museum, it will go to America” (BM Archives, Object File – Curator report recommending purchase of BM 1968,0213.1, 26 January 1968). The use of the Brooke Sewell Permanent Fund to purchase 15 Khmer objects (mostly bronze and sandstone religious sculpture but also two ceramic pieces) between 1959 and 2002 thus seems to reflect the general expansion of the market for Khmer art, and sculpture in particular, as well as the increased availability of funding. It was also in the period from 1951 to 1970 that the proportions of archaeological and ethnographic objects entering the collections began to decline, and instead there were more purchases of sacred art and ceramics (eight items from six different dealers), perhaps also reflecting the same strategy that had led to the 1933 reorganisation of the Oriental Collections. It is notable that a shared feature of purchases during this period is the lack of provenance information available about them, despite the adoption of the 1954 Hague Convention and the period leading up to the adoption of the 1970 UNESCO Convention: generally only the identity of the dealer and the date of sale are given.

During this period, the BM received donations from the Bangkok-based dealer Douglas Latchford who was indicted in 2019 for a number of felony charges. Between 1970 and 1975, he donated five objects to the BM, including two lintels discussed in Phuy’s case study (BM 1971,0924.1 &  1971,0924.2).  Latchford began collecting and trafficking Khmer art in earnest in the 1960s (Tabachnik 2022a), developing close ties over the following decades with numerous auction houses, dealers, private collectors, and museums globally. He also amassed a substantial personal collection of Khmer sculpture and jewellery. Latchford came under the scrutiny of law enforcement agencies for his involvement in the illegal trafficking of Khmer art from Cambodia first in 2011 following a dispute between Sotheby’s and US authorities over a 10^th-century statue of Duryodhana from Prasat Chen in the Koh Ker temple complex, in northern Cambodia (Noce 2013), then in 2016 when he was mentioned in the indictment of New York gallery owner Nancy Wiener (Mashberg 2021), and in 2019 with his own indictment (Southern District of New York 2019), which alleged that over a number of years he had conspired with others to create false provenance information, fraudulently obtain export licenses, and deal in statues known to have been looted. The scrutiny of Latchford in this context has spilled over into museums only very slowly, with provenance research generally being reactive. Given the state of knowledge of Latchford’s activities, it would seem reasonable to interrogate provenance information provided by Latchford in relation to all objects donated or sold by him. See Phuy’s case study for an examination of issues around the lintels donated to the BM, which Latchford claimed were from Northeast Thailand.  

In terms of the motivation for Latchford’s donations, which were made relatively early in his art-dealing activities, these can be seen as part of a practice of ‘reputation laundering’, whereby dealers of looted or trafficked objects place pieces with reputable private collectors and institutions in order to enhance their own reputations and ability to sell other pieces. (See, for example, Yates and Graham 2024 for an examination of donations of low-value Latin American antiquities to prominent museums by art dealers later convicted for offences relating to fraud or smuggled objects.) In the 1980s and 1990s, Latchford also donated six pieces to the Metropolitan Museum of Art, New York. These donations included two Khmer pieces that were deaccessioned for return to Cambodia in December 2023 (MMA 1983.551 and 1998.320a-f) after sustained pressure from US and Cambodian authorities (Southern District of New York 2023), and three joint donations (MMA 1989,237.1, 1989,237.2, 1989,237.3) with the London auction house Spink & Son, which US prosecutors alleged was involved in Latchford’s activities (Southern District of New York 2019). Meanwhile, from the 1980s to the early 2000s, Latchford sold five pieces to the Met either directly or through Spink; in another case, he sold the object (MMA 2002.477) to a private collector, Jeff Soref, who immediately or soon after loaned it to the Met, and then donated it three years later. Four of those six objects were deaccessioned in the same 2023 operation (MMA 1983.14, 1992.336, 1999.262 and 2002.477). The two objects that were not deaccessioned are described by the Met as coming from Thailand (MMA 1994.51) and Vietnam (MMA 1992.53).

While Latchford’s donations to the BM did not lead to any subsequent sales to the museum per se, the timing of these donations early in his career place them in line with a longer legacy of using museums’ reputations to enhance his own, and so his trade. It is worth considering whether Khmer lintels similar to those Latchford donated to the BM and now in other museums might be connected with him. The Met has 10 Khmer lintels in its collection, none of which are explicitly connected with Latchford. However, only one of those lintels has provenance information going beyond 1972, having been purchased from the EFEO in 1936 (MMA36.96.6). One is described as being with Spink “until” 1990 (MMA 1994.111), while another was donated by the gallery of Doris Wiener (Nancy Wiener’s mother) (MMA 1992.192). The provenance and ownership history of these lintels might be a useful avenue for future research.

Another way in which Latchford worked to develop his reputation and legitimise looted objects he traded was through the publication of a number of books, co-written with Emma C. Bunker, a consultant and volunteer at the Denver Art Museum (Tabachnik 2022b). These are coffee-table type catalogues of previously unpublished objects with unknown provenance, most of which likely belonged to or were traded by Latchford, and feature discussions of Khmer art with illustrations and examples from museum collections. There is evidence that Latchford used these books “to market his antiquities to wealthy collectors” (Tabachnik 2022a). In his review of Latchford and Bunker’s book on Khmer bronzes (published in 2011), Paul Lavy highlights the potential for such publications to “support questionable acquisition and collecting practices, perhaps stymie heritage management and archaeological research, and […] legitimate objects with potentially contested provenance and/or controversial provenience histories” (Lavy 2015, 334). Reviews calling attention to the problematic nature of these publications and the materials they illustrated had in fact come very early on, in the wake of the first of the Latchford-Bunker volumes, in 2004 (Baptiste and Zéphir 2004, Porte 2004). Despite such understandings, art historians such as Lavy and curators of international collections continued to cite the Latchford-Bunker volumes, contributing to further normalise and even legitimise them and the trafficking they underpinned.

Besides the Latchford donations, the period from 1970 to 2000 also saw a decrease in the proportion of archaeological and ethnographic objects (three in total) entering the collection, compared with previous periods, while the proportion of ceramics increased, through both purchases and donations. In 1979, the BM purchased a brown-glazed pot from Bluett & Sons (BM 1979,1113.1) and in 1993, it purchased three baluster jars from Spink & Son (BM 1993,0417.1-3).  In the late 1990s, two British private collectors of ceramics donated stoneware jars and sherds from Cambodia, as part of larger donations including ceramics from other parts of Asia and from the Middle East. One of these collectors was Angus Forsyth, a Hong Kong-based lawyer and member of the Hong Kong branch of the Oriental Ceramic Society, who was also a friend of Latchford, and attended a trip led by Latchford to the temples of Koh Ker in Cambodia in 2002 (Bunker and Latchford 2003).

The role of social relationships between dealers, curators, and collectors in the circulation of both legitimately sold and looted art cannot be overlooked and is also relevant in the purchase of seven bronze figures (BM 1999,1006.1-7) in 1999 from Dr Henry Ginsburg, a scholar and connoisseur of Southeast Asian arts and literatures and curator of Thai, Lao and Cambodian manuscripts at the British Library. He had begun his career in 1967 as the BM’s Special Assistant responsible for Thai collections, within the Department of Oriental Manuscripts and Printed Books, and in 1973, he moved to the newly-formed British Library, where he worked until his retirement in 2002 (Gallop 2008). On Ginsburg’s death in 2007, his estate made an endowment supporting Southeast Asia curatorial posts at the British Library and the British Museum.

The bronze figures purchased from Ginsburg were acquired by him at an auction of part of the collection of Natasha Eilenberg in 1998. Together with her husband, Samuel (‘Sammy’) Eilenberg, a professor of mathematics at Columbia University, Natasha Eilenberg amassed a vast collection of Asian art, a substantial proportion of which came from Southeast Asia. When the couple divorced, their collection was divided and substantial parts were later separately donated or sold. The Southeast Asian art from their collection that is now in other hands is generally described as having been collected in the 1950s and ‘60s, as is the case for the BM objects. However, the BM’s records contain no further evidence of where and when they were purchased by the Eilenbergs (See Fitzpayne’s case study for more on trying to trace the provenance of these objects). The later acquisition of the objects by Henry Ginsburg is evidence of his own interest in Khmer art, and Southeast Asian art more generally. However, it also speaks to the overlap between the ‘academy’ and the market: between 1991 and 2005, the BM purchased at least 25 objects (including the Eilenberg objects) either from Henry Ginsburg or his company, Hentell Ltd (See, for example, British Museum Archives, Object File 1999,1006.1-7, Memorandum dated July 1999, and British Museum Archives, Object File 2002,0330.1, Invoice dated 15  January 2002). The company also sold manuscripts to the British Library (For example, BL Or 14447, purchased from Hentell Ltd Hong Kong in 1989, discussed in Igunma 2022). From 1987 onwards, he also made several donations to the BM (at least 55 objects, including 20^th century Thai paintings (for example, BM 2003,1027,0.1) and popular prints depicting Buddhist themes (BM 1990,0625.1-13), and after his death six Chinese ceramic pieces were bequeathed to the BM (BM 2008,3018.1-6). While Ginsburg’s archival work and philanthropy have made important contributions to the field of Southeast Asian arts writ large, the possibility of a conflict of interest arising from his management and ownership of a business that sold Southeast Asian art at the same time as holding a post at one of the institutions to which that business sold objects merits greater consideration than it has received to date.

Since the early 2000s, there seems to have been a decline in donations and purchases of Cambodian objects. This might appear particularly marked in the light of the possible impact on collecting of the major touring exhibition ‘Sculpture of Angkor and Ancient Cambodia: Millennium of Glory’ of 1997 (For more on this exhibition, see Jessup and Zéphir 1997). This exhibition, organised by the National Gallery of Art, Washington, the Royal Government of Cambodia and the Réunion des musées nationaux/Musée national des Arts asiatiques-Guimet, Paris included sculpture from the collections of the National Museum of Cambodia and the Parisian Guimet Museum, eschewing inclusion of materials from private collections. The BM’s purchases of the former Eilenberg pieces in 1999 (BM 1999,1006.1-7) and of a monumental sandstone sculpture in 2002 (BM 2002,0330.1) can nonetheless be placed in the wider context of increased interest in Khmer sacred art and monumental sculpture arising out of this exhibition, which visited Washington DC, Paris, Tokyo and Osaka. A relative decline in the collecting of Cambodian objects might also be attributed to increased media coverage of looting or the increasing efficacy of the Cambodian and international measures to curb illegal trafficking and pursue successful restitution claims.

In total, only thirteen objects have been accessioned since the year 2000. These include three sandstone sculptures and a small bronze figure purchased from and donated by Henry Ginsburg, building on the BM’s collection of sacred figural sculpture, which makes up just under a quarter of the total collection. The bronze figure, like the other bronze figures purchased from Ginsburg, had been part of Natasha Eilenberg’s collection and was acquired by Ginsburg at the same auction. The BM’s purchase of this figure was funded by Jeff Soref, mentioned above as having purchased an object from Douglas Latchford before loaning and then donating it to the Met. While there is no suggestion of a connection to Latchford in this case, it seems worth noting the relatively small and ‘connected’ circles of dealers and collectors of Khmer art.

The three sandstone sculptures received from Ginsburg are described as having been acquired by a French archaeologist in the 1920s (BM 2002,0330.1, 2002,0330.2, 2002,0330.3). These are quite large in scale, particularly the relief of a standing apsara, comprising six blocks with a total height of nearly two metres. This prompts questions about how they were removed from Cambodia, even during the colonial period, as well as questions around colonial collecting practices more generally. Some of these issues are considered in Fitzpayne’s case study.

The remaining objects that have entered the BM in the 21^st century are fairly diverse, and include the only paintings in the Cambodia collection: four tall 19^th-20^th century paintings depicting scenes from jataka tales (BM 2010,3018.1-4), purchased from a Portobello Road dealer and attributed to the collection of Henry Ginsburg, and a 20^th-century banner painting (BM 2015,3014.1) purchased from Marani Fine Art, an Australian art dealer. The other objects comprise a brass lime pot (BM 2005,0528.5), also unique within the collection and donated by Richard Blurton, who was at the time Assistant Keeper of the Department of Asia, a silk textile (BM2000,0205,0.1) donated by David Snellgrove, a former professor at SOAS, and two contemporary mouth harps (BM 2020,3017.1-2), commissioned as part of an anthropological research project. Although there has been a relative decline in the present period in the number of objects entering the Cambodia collection, it can be said that the objects that have been accessioned reflect the diversity of the collection as a whole.