Introduction

Beginning as early as 1907, French scholars, architects, and archaeologists—together with a handful of colorful adventurers—dedicated themselves to the clearing, interpretation, and eventual restoration (principally stabilization) of the great stone and brick monuments at Angkor. How this venture transpired has been the subject of numerous histories and incidental treatments of the French presence in Cambodia. What have been less examined are the deeper cultural and institutional roots of this effort and the preliminary work done by many of the same scholars and practitioners in Vietnam, then French-ruled Annam and Cochin (and Tonkin to the north).

The remnants of ancient Champa, a civilization that thrived along the coast of south central Vietnam from around the fourth century CE until well into the seventeenth century, served the purposes of French scholarly inquiry and also provided the basis for a more sustained institutional involvement. The excavation, repair, and presentation of Champa ruins were a rehearsal for the better-known work in Cambodia, Laos, and even Northeast Thailand, which came to define France’s cultural preeminence in the region. Champa, as an inarticulate partner in this project, proved the perfect focus of these efforts.

The École Française d’Extrême-Oriente (EFEO) was perhaps the key cultural institution in France’s colonial project. Founded at the turn of the last century as an ostensibly disinterested scholarly organization, the ‘French School of the Far East’ came to serve as an intellectual ally to France’s imperial ambitions in the region (Clémentin-Ojha and Manguin 2007). Focusing initially on architectural ruins, the EFEO eventually extended its purview to Vietnamese history and language, as well as to prehistoric archeology. However, ruins—in the instance of Vietnam, nearly all associated with ancient Champa—and their conservation were a significant interest of EFEO scholars well into the twentieth century.

Emulating Napoleon’s Egyptian expedition of a century before, scholarship worked hand-in-hand with military, administrative, and economic expansion (Osborne 1969, 1999). Early explorations, such as that of Francis Garnier (1839–1872), extolled the wonders of ancient civilizations. The 1866–1868 Mekong Expedition of Ernest Doudard de Lagrée (1823–1868) and Francis Garnier (1839–1872) stopped to record Angkor and other monuments, eventually visiting Vat Phu in Laos and Vientiane, the former capital of one of Laos’s three kingdoms, as did the subsequent expedition of their artist, Louis Delaporte (1842–1925), who returned to Angkor in 1873 and again in 1881 (Osborne 1969, 1999). Throughout these missions, antiquities served as touchstones for France’s ‘civilizing mission’—an expression of the country’s unique understanding of the importance of the past and of ancient civilizations, whose own efforts France was then duplicating. In many ways the very course of empire was demarcated by the distribution of ruins. And it was eventually the EFEO that would provide the expertise for their identification.

Intrigued by the apparent ‘Indian’ origins of ancient shrines and city remains, scholars with the EFEO collected the materials for what was to become the meta-theory of ‘Indianization,’ a term coined by the EFEO’s onetime director George Cœdès (1886–1969). Cœdès’s theory, published in complete form only in the 1940s, was that the region’s assemblage of cities and states, as well as ideas about finance, military organization, and apotheosized rule, had all been the result of a protracted series of interactions with Indian traders and priests (Cœdès 1944, 1948). ‘Indianization,’ as expounded by Cœdès, had lifted the indigenous peoples of the Mekong Region above their former level, bringing advanced ideas of rule and religion and new kinds of religious monuments, based on Hindu and Buddhist ideals.

This theme of Indian conquest, whether intellectual and religious or actual, had important ramifications for the French presence in Southeast Asia, reinforcing, in a sense, the persistent trope of France’s civilizing mission. If the ancient Indians provided models for religion, politics, architecture, and much else, France provided the template for the demarcation and preservation of the remnants of that earlier cultural transmission. The ancient sites of Vietnam would provide an initial foray into this effort.

Antiquities in Vietnam

Vietnam’s prehistory extends back to the Paleolithic period, probably as early as 30,000 years ago. Bảo Sơn and Hòa Bình Cultures (referred to commonly as Hoabinhian, or Văn hóa Hòa Bình in Vietnamese) provide evidence of the use of stone tools followed by ceramics and the introduction of agriculture around 2000 BCE (Higham 2002; Hà Văn Tấn 1997; Phạm Đức Mạnh 2000). Within a thousand years a significant rice-growing culture emerged in the Ma River (Sông Mã) and Red River Plains in the north, ultimately coalescing to form the Đông Sơn culture, so famous for its use of bronze drums and its influence throughout both mainland and island Southeast Asia (Taylor 1991).

Paralleling these northern developments, central and southern Vietnam was home to several small kingdoms during the early part of the Common Era. Early cultures included what the Chinese identified as Lin Yi (Lâm Ấp in Vietnamese). Farther south in the Mekong Delta was the Óc-Eo or Funan Civilization, a culture with close connections to eventual Khmer states in present-day Cambodia. By the fourth century CE, Lin Yi in central Vietnam was a major urban complex, with mud-brick walls, temples, and a palace. Possible adherents of Buddhism, Lin Yi may also have incorporated Hindu deities into their ritual practice, though much about this early civilization remains speculative (O’Reilly 2007: 142–44).

Most early scholars believed Lin Yi referred to Champa and the Cham; most now agree they were separate cultures (Coedès 1968: 65, 247; Stein 1947; Lafont 1991; Vickery 2005; Nguyễn Kim Dùng, Glover and Yamagata 2006). The Cham were an Austronesian-speaking people, with origins probably in Borneo. They first rose to prominence in the southern part of coastal Vietnam around the beginning of the Common Era. Their capital was near present-day Danang, where they were an important trading power by the end of the third century (Figure 1) At the height of their powers, the Cham people occupied an area stretching from the Đồng Nai River in the south to a point about 180 miles (290 kilometers) north of Huế (Ngô Văn Doanh 2005, 2006; Nguyễn Văn Kự, Ngô Văn Doanh, and Hardy 2005; Vickery 2005, 2009; Andaya 2008: 44–45).

Throughout the early centuries of the Common Era, Champa, as the civilization is known, spread over five centers of power, probably forming more a confederation than a centralized empire—although historians still argue this fact (Maspéro 1928; Glover and Yamagata 1995; Southworth 2000). The five centers were, from the north, Indrapura (Đông Dương), Amaravati (Trà Kiệu), Vijaya (Chà Bàn, or more straightforwardly, Quy Nhơn), Kauthara (Nha Trang) and Panduranga (Phan Rang). Adherents of both Śaivite Hinduism and Mahayana Buddhism (Schweyer 2009; Mabbett 1986), the Cham created their own unique writing system based on characters derived from the Brahmi script of South India. They also built sacred shrines dedicated to Shiva and other Hindu gods. By the end of the seventh century, Cham leaders began to replace earlier wood temples with permanent structures of stone and brick. Inscriptions point to dynastic ties to the Khmer as well, a culture with which they also competed.

Champa never had a single ruling dynasty or capital. During the eighth century, Cham power shifted in part to the south central area of Vietnam, at Quảng Nam, Khánh Hòa and Ninh Thuận. The temple complex of Pô Nagar, near coastal Nha Trang and dedicated originally to the earth goddess Yan Pô Nagar, served as the centerpiece of one Champa polity. Beginning in the late eighth century, the regional Cham ruler enlarged the city and temple complex to include statues of Shiva and other Indian gods. In 875 CE, the Cham king Indravarman II founded a new dynasty closer to Danang. (O’Reilly 2007: Asia, 139–41; Po Dharma Quang 2001; Vickery 2009). This was the city of Indrapura that Cham sovereigns equated with stories from the Indian Mahābhārata saga. Indravarman adopted Mahāyāna Buddhism as a state religion and built temples to commemorate this shift.

During the tenth and eleventh centuries, the Cham at Indrapura faced increasing competition from other states in the region. By end of the twelfth century, Champa engaged in successive wars with both Khmer kings to the west and the Vietnamese to the north, ultimately losing a war of attrition. During the thirteenth century, the Khmer occupied large parts of Champa territory. Despite a temporary return to power in the fourteenth century, under the leader Chế Bồng Nga, Champa was increasingly subject to Vietnamese hegemony. Adopting a plethora of Islamic ideas, merged with both old native beliefs (animism) and Hindu and Buddhist ideas, Champa society had changed significantly by the fifteenth century (Tarling 2000; Vickery 2009; Taylor 2007).

Champa Architecture

Historically, the typical Champa temple complex included a central sanctuary, called a kalan (from Cham language), surrounded by small towers, various ancillary buildings and then an outer wall. As with other Indian-inspired sacred sites, the Champa temple complex formed a microcosm of the world, with the kalan standing for Mount Meru, and the outer walls and subsidiary shrines representing secondary peaks and mountains. The kalan itself had three conceptual elements: the base (bhurloka), symbolizing the human world; the body of the tower (bhurvaloka), standing for the spiritual world; and the pyramidal roof (svarloka), representing the sacred universe (Le Bonheur 1998; Boisselier 2001; also Leuba 1923; Trần Kỳ Phương 2009).

The kalan also served as an outdoor altar and played important ritual functions. The typical kalan featured an interior room containing a phallic lin·gam (linga or lingga) image on a dais. A yoni, the stylized vagina and symbol of the female deity Shakti, provided a platform for the lin·gam. Water flowed over the dais or platform as part of the ritual of washing the lin·gam. A kind of corridor usually surrounded the altar. This permitted worshipers to access the shrine. For Buddhist shrines, the priests placed a different kind of altar against the west wall (Jansen 2003: 46, 119; Schumacher and Woerner 1994).

The temple precinct had several specific features. A gate tower or gopura stood before the central kalan. Near this was a mandapa, a long tile-roofed structure, with multiple doors and windows opening to the east and west (and similar in appearance to the so-called ‘libraries’ of Khmer sanctuaries). In some instances the mandapa included windowed outer walls; in other cases they were open pavilions, supported by brick or stone columns. To the immediate right (south) of the kalan was a repository for offerings. Called a kose grha, the repository generally had two rooms and a gate on the north side. In most instances, a low brick wall surrounded the kalan and other structures. The Cham placed stele immediately outside the enclosure; these usually held inscriptions relating the shine’s construction and the builder (Figure 2).

The primary construction material of Champa temples is kiln-fired brick, with sandstone serving as door and window surrounds and for elements of the roof structure. Carvings included both figurative and repetitive designs, the latter alternatively naturalistic and geometric in style. The bricks are broad and thin by modern standards, measuring about 12 by 8 by 4 inches (30 by 20 by 10 centimeters). The composition of the mortars is still a matter of speculation, although recent research suggests that the clay mortars included organic additives, particularly a resin, and possibly by ground oyster shells as well (Ballio, Baronio and Binda 2001). Workers completed the carvings, both in brick and stone, after completing the wall.

Mỹ Sơn and Other Sites

Subject to neglect and abandonment, most of the Champa sites were archaeological ruins by the time of French involvements in the region. It is likely too that the earliest shrines were of wood and, therefore, they have not survived. The oldest masonry buildings date to around 600 CE. Most are of brick or of combined brick and stone with stone embellishments and sculpture. Again, as in other ‘Indianized’ states, successive rulers added to existing shrines, altering their character in a cumulative way. Unlike the Khmer architecture, however, few shrines were rebuilt in their entirety, so that each Champa site tends to be a record of successive additions more than a single artistic statement.

One of the earliest known Champa sites is at Pô Nagar, near the city of Nha Trang. Also called Tháp Bà (the Lady of the City), the Pô Nagar complex was built between the eighth and thirteenth centuries, replacing an original wood shrine burned by the Javanese in 774. Pô Nagar was site of worship as early as the second century CE, although only limited archaeological evidence survives from this initial period. At one point the complex covered an area of a little over an acre (4047 square meters) included seven or eight towers. Scholars later gave a date of 817 and attributed the shrine to a minister of the king Harivarman I (Ngô Văn Doanh 2006: 192–98; Trần Kỳ Phương 2000: 71–75). Pô Nagar was the subject of one of Henri Parmentier’s earliest archaeological efforts (Parmentier 1902). Numerous stone carvings still remain at the site, including dancing, four-armed Shiva guarding the entrance to the interior vestibule. There are also images of musicians and the head of Shiva’s mount, the bull sometimes referred to as Nandi.

There were about 20 remaining Champa sites at the time French scholars began their investigations. These spread from Quảng Trị in the north to just east of modern Ho Chi Minh City (Saigon) in the south. Most of the larger sites clustered around the city of Danang. All were in coastal locations, usually situated on hilly areas along the coastal plain. The area near Danang included Đông Dương, historically Indrapura, a mostly ninth-century complex built by the Champa king Indravarman; Trà Kiệu, the historic city of Simhapura, or the ‘Lion City,’ with origins to the fourth century; Bằng An, a solitary, octagonal tower; Chiên Đàn, three kalan, dating to the eleventh and twelfth centuries; and Khương Mỹ, a highly decorated grouping of three towers.

A second major grouping of Champa sites was located in Bình Định Province, near the city of Quy Nhơn, almost 200 miles (320 kilometers) south of Danang. Among the most prominent sites was Tháp Đôi, two Champa towers with unusual pyramidal roofs (rather than the usual stepped form). There were also the ruins of Chà Bàn (also, and more commonly, known as Đồ Bàn), located north of Quy Nhơn; Tháp Bạc (Silver) Tower, also known as Bánh Ít; Bình Lâm Tower; and Thi Nại Citadel. Farther south from the Quy Nhơn sites is the solitary Nhan (Swallow) Tower (Tháp Nhạn) near Tuy Hòa, on the northern bank of the Danang River (Trần Kỳ Phương 2009: 155–195).

One of the most important Champa sites was Mỹ Sơn, located in Quảng Nam Province, about 40 miles (64 kilometers) southwest of Danang (Figure 3). Once the center of a Hindu city, the original complex included more than 70 towers built between the seventh and thirteenth centuries. Building began at Mỹ Sơn as early as the fourth century. Starting in the seventh century, Champa builders replaced older wood temples with stone and brick versions. In the tenth century, the Champa kings, beginning with Indravarman II, embraced Mahāyāna Buddhism, and Mỹ Sơn became a Buddhist shrine. A resurgence of Hinduism under the ruler Paramesvaravarman in the twelfth century resulted in a final building campaign. This lasted until nearly the end of the century, after which time little new occurred at the site. By the end of that period, Mỹ Sơn had at least 25 towers, each surrounded by low walls (Ngô Văn Doanh 2005; Trần Kỳ Phương 2000: 19–30).

Close by Mỹ Sơn is the ninth-century site of Đông Dương, historically Indrapura. Erected during the reign of king Indravarman, Đông Dương was a Buddhist shrine, probably, though not conclusively, dedicated to the king’s protector—and possibly ‘alter ego,’ as in the case of Jayavarman VII in Cambodia—the Bodhisattva Lokeśvara (Avalokiteśvara in Cambodia) (Holt 1991; Studholme 2002; Nguyen 2005). Largely destroyed during the Vietnam War, Đông Dương was comprised of three concentric ‘shrine circles,’ labeled by EFEO archaeologist Henri Parmentier (1871–1949) as circle I, II and III. The towers and enclosures follow a linear sequence, revealing a sense of axial progression. Đông Dương or Indrapura served as the capital of Champa from 860 to 986 (‘Dong Duong’ 2011; Hoàng Đạo Kính 2009: 123–25).

Trà Kiệu is also in the vicinity of Mỹ Sơn and Đông Dương. Originally called Simhapura, or ‘the Lion City,’ the beginnings of Trà Kiệu date to the fourth century CE. It was the first known capital of the Champa confederacy and is associated with the reign of the king Bhadravarman. Seventh-century Chinese travelers described the site as a fortified citadel, with walls and a perimeter moat (Trần Kỳ Phương 2000: 31). Excavated in the 1920s by French archaeologist Jean-Yves Claeys, Trà Kiệu was clearly an important military and administrative center (Claeys 1934; Trần Thị Thúy Điể m 2001: 11; Boisselier 2001: 56).

Just south of Phan Rang and Tháp Chàm is Pô Đam, a grouping of six towers, sited on the slope of the Ông Xiêm Hills. Dating to the early period of Champa art, the late eighth and early ninth centuries, Pô Đam had still a range of bas-reliefs and other carvings. Also in Thuận Hải Province was the complex of Phú Hài, a small shrine with two kalan. Much influenced by Khmer art of the ninth century, Phú Hài’s retained only fragments of its original sculptural program, although most would soon sit in museums (‘Phu Hai (Po Xahnu) Towers’ 2011; Trần Kỳ Phương 2000: 91–92). Other significant sites include Hòa Lai Po, Klong (Klaung) Garai (Poklonggarai, My Dragon King), Po Rômê (Rô Mê), and Phú Hài (Chapman 2013: 105–107).

The École française d’Extrême-Orient

The French dedication to Asian studies, focused on ancient Champa and other civilizations, had its origins in late nineteenth-century metropolitan France. A Société des Études Indochinoises, founded in 1865, shortly after the French conquest of Saigon and a year after the recognition of French treaty ports at Annam and Tonkin (and all of French-held Cochinchina), published a number of works on antiquities in present-day Cambodia and Vietnam. They also established a regular Bulletin and a separate review called Excusions et reconnaissances, both of which helped establish the pattern for later scholarly publications.

In addition to scholarly pressure in France, Asian studies had its champions in the field. The most important of these for Southeast Asia and both Khmer and Cham sites was Étienne Aymonier (1844–1929), a linguist, archaeologist, and explorer—and occasional contributor to scholarly publications such as the Société des Études Indochinoises’s Bulletin. An officer in the French military, Aymonier had first traveled to Southeast Asia in 1869, landing in Saigon to serve in the occupation of Chochinchine. Over the next two years, he took an increasing interest in the history and ethnography of the region, particularly focusing on a Khmer minority in southern Vietnam.

In 1872, Aymonier moved to Phnom Penh, serving as the assistant to the French representative to the Protectorate of Cambodia. Shortly afterward he was appointed inspector for the Ha Tien region, on the Gulf of Thailand. From there, following a return to France and further assignments, he rose to additional governmental posts in Saigon and then back to Cambodia, in 1879, the acting representative to the Protectorate of Cambodia. This was followed by a protracted period of exploration, including a survey of the region around Angkor, Laos, and parts of Siam (the results of which were published in a series of publications over the next twenty years). In early 1885, relying on advice of local Cham people, he catalogued known sites in the area of Hué, returning later that year to complete a survey of Champa antiquities in Bình Thuận province and surrounding areas. Living with a Cham wife, he devoted particular energy to the study of Cham language and culture, later serving as the Annam-Tonkin delegate to the Universal Exhibition of 1889 and serving on the board of the Asian Society, an organization dating from 1822 and a center of French study of Asian languages.

While Aymonier and his fellow scholars initiated studies in Southeast Asia, other scholars pressed for greater institutional backing at home. Seeking to emulate national scholarly institutions such as those at Athens (founded 1846) or Rome (1875), philologists Auguste Barth (1834–1916), Émile Senart (1847–1928), and Michel Bréal (1832–1915) pressed for a similar school in India. However, when that initiative stalled in part due to the dwindling significance of French possessions there, several leading Orientalists began to press for a greater commitment to France’s new colonies in Southeast Asia (Clémentin-Ojha and Manguin 2007: 18–35; Singaravélou 2001). In 1897, former civil servant Charles Lemire and military physician Pierre Lefèvre-Pontalis complained of the poor condition of ancient sites in Indochina, petitioning the Eleventh Congress of Orientalists to commit to a project of recordation and protection, a challenge soon accepted by the recently appointed Governor-General of Indochina, Paul Doumer (1857–1932).

Encouraged further by Indian scholar Sylvain Levi (1863–1935), who stopped in Saigon in the latter part of 1897, Doumier instituted a permanent Mission archéologique d’Indo-chine the following year (replacing a more informal organization of the same name). This included the beginnings of a library and museum and an outline of potential projects to be conducted in the region. In 1900, the Mission archéologique d’Indo-chine was renamed the École française d’Extrême-Orient (EFEO); two years later it moved to Hanoi (the site of the French capital of Indochina after 1900). Its mission was laid out as the collection of manuscripts, the preservation of monuments, completion of inventories of ethnic groups, linguistic studies, and the history of Asian societies in general (École française d’Extrême-Orient 2017).

The first director of the new organization was the epigrapher Louis Finot (1864–1935). Finot had made his name as a paleographer, with an expertise in Sanskrit (Finot 1896). Finot’s counterpart was Henri Maspéro (1882–1945), a Sinologist who quickly turned to the study of Indochinese languages. The third original member was Paul Pelliot (1878–1945), who immediately left on a ‘mission’ to China on behalf of the organization (Hopkirk 1980).

Champa and other ancient sites in French-controlled Indochina would become the immediate subject matter of the recently founded EFEO. From its base in Saigon and then, after 1902, in Hanoi, the EFEO became involved in a range of projects, from the study of local customs through to philosophy and language. Archaeology, nonetheless, formed a key feature of the EFEO’s work, building directly on early efforts of Aymonier, Doumer and other administrators beginning in the 1880s and 1890s. The study and later preservation of Champa sites would stand at the center of this enterprise.

The first great exemplar of the scholar-explorer was Étienne-Edmond Lunet de Lajonquière (1861–1933) – the ‘bushman,’ as his colleagues called him. Beginning as a member of the French marines, he joined the infantry in 1882, rising to lieutenant in 1885 and captain in 1892. In 1898, he was seconded to the Archaeological Mission of Indochina (la Mission archéologique de l’Indochine) and soon afterward began a survey of Champa sites in Annam and then a reconnaissance of Khmer sites in then Siamese-ruled northern Cambodia. His first survey, published in 1901 as the Atlas archéologique de l’Indo-Chine monuments du Champa et du Cambodge, set out what would become a template for French publications on the historic remains of the region. He later carried out a mission in Tonkin, reporting on the existing population, returning to Cambodian antiquities, beginning in 1902 with the publication of the first of a three-volume survey of Khmer remains at Angkor, a project initiated in 1900 (Lunet de Lajonquière 1902).

Throughout this period, Lunet de Lajonquière worked closely with the recently designated director of the EFEO, Louis Finot. Despite his institutional locus, Finot was very much an ‘activist’ administrator (Figure 4). Following his arrival in Saigon in 1899, he joined the surveyor Lunet de Lajonquière on expeditions throughout Cochin and Annam and went on to investigate Angkor and then the ancient sites of Laos. Relying on his work with Lunet de Lajonquière, Finot published a first study of Champa sites in Annam, Inventaire sommaire des Monuments chams de l’Annam, in 1901 (Brown 2013: 31–32). As with Lunet de Lajonquière’s separate publication of the same year, Finot’s study enumerated the key features of each site and introduced a classification scheme based on key features and locations.

Lunet de Lajonquière would continue his work in Cambodia throughout the first decade of the twentieth century, venturing into Laos as part of his survey with Finot in 1900, returning again to Cambodia for photographs and additional mapping in 1904, just prior to the retrocession of three provinces, including Siem Reap (Siam Nakhon to the Thai), to Cambodia as part of an agreement with the Kingdom of Siam in 1906 (Clémentin-Ojha, Catherine and Manguin 2007: 79–81). Between 1904 and 1909 he was periodically in Cambodia, Siam, and Malaysia, again creating inventories of ancient sites. The results of his Cambodian travels and investigations would be published in the second and third of his three volume Inventaire descriptif des monuments du Cambodge (1907, 1911). Those in Siam and the Malay Peninsula followed in 1906 and 1910, respectively.

Henri Parmentier joined the newly organized archaeological team in 1900, working primarily on an inventory of Champa antiquities (Figure 8). His architectural training and prior archaeological experience in Tunisia gave him special insights into the construction of temples, resulting eventually in a comprehensive two-volume inventory of Champa sites (based in large part on Finot and Lunet de Lajonquière’s earlier work), Inventaire descriptif des monuments čams de l’Annam, published in 1909 and 1918, respectively (Parmentier 1909–1918). Parmentier was the acknowledged head of the EFEO’s archaeological efforts from 1907 (Clémentin-Ojha and Manguin 2007: 103).

In 1920, the EFEO created an official Archaeological Service, in many ways following the example in the Dutch colonies and the British in India. Parmentier, by then a veteran field worker, was the first ‘head’ of the service, a position he held until his retirement in 1933 (Clémentin-Ojha and Manguin 2007: 81–83). His successors included Henri Marchal (1876–1970), who did so much work at Angkor, and Jean-Yves Claeys (1896–1979), who served as head from 1937 to 1946. With its small staff, the Archaeological Service depended heavily on the contributions of ‘Corresponding Members’ throughout Indochina. In Angkor, the separate Conservation d’Angkor took on the task of clearing and surveying the site.

Parmentier, due to his position in Hanoi, focused primarily on the Champa remains. Accompanied at first by his assistant Charles Carpeaux, Parmentier established a standardized approach to the inventories of the Champa monuments. The first step was clearing. He then drew up site plans, elevations and sections of individual features. During the process, he gathered fragments, inscribed stele and sculptures and shipped them into safekeeping, at first in storage and later in the several museums created to hold Champa and other art (Parmentier 1909–18).

Rudimentary restorations accompanied the surveys. With small annual budgets and small numbers of laborers, Parmentier began to stabilize the Champa sites he had surveyed. In 1901, just a year after beginning work, he carried out the initial restoration of Mỹ Sơn (Figure 6) and Đông Dương (Figures 5 and 7) near Huế (Carpeaux 1908). Finishing there in 1903, he turned to Pô Nagar and Nha Trang in the south, beginning his work at Pô Nagar in 1906 and Nha Trang in 1908. Between times, he visited the Dutch colonies to see the work there; and although Dutch conservators had not yet introduced the technique of anastylosis (the disassembly and reassembly of stones and other components) that would provide so important to the work in Angkor, he still gained much from the experience (Baptiste 2009).

Parmentier continued his work both in Vietnam and in other parts of Indochina—including a visit to the Laotian site of Vat Phu—throughout the early decades of the twentieth century. In Vietnam, he organized the first EFEO museum in Hanoi in 1909. He also filled a new museum in Phnom Penh with Khmer antiquities, writing a catalogue for the collection that remained the standard guidebook well into the middle of the century (Parmentier 1912). As a contribution to French displays, he supervised the making of casts of Champa and Khmer sculpture for the Musée du Trocadéro in Paris and, in 1918, and oversaw the completion of the Cham Museum, replacing the small display at Le Jardin de Tourane. Named the Musée Henri Parmentier after his retirement, this small museum became one of the principal bases of Champa scholarship in the 1920s and 1930s (Trần Thị Thúy Điểm 2001; Kelly 2001: 156–61).

Parmentier, like Finot, enjoyed being in the field. In Cambodia, he supervised much of the early work; and in 1914, he oversaw the first investigations of the ten-century site of Banteay Srei, publishing his findings in the EFEO’s Bulletin (Parmentier, ‘L’Art d’Indravarman’). This was the unfortunate source of André Malraux’s knowledge of the site, which resulted in André Malraux and his wife’s famous larcenous adventures in Cambodia (Langois 1966). In the 1920s, Parmentier provided evidence for the date of the Bayon temple and in 1926 published, together with Louis Finot and Cambodia-based amateur archaeologist Victor Goloubew (1878–1945), a definitive study and guide to Banteay Srei, the site of Malraux’s theft (Finot, Parmentier and Goloubew 1926).

Parmentier consistently gave encouragement and guidance to younger members of the staff. In Cambodia, Parmentier supported the young architects and archaeologists Henri Marchal and Georges Trouvé (1902–1935). It was with Parmentier’s approval that Marchal traveled to Java in 1930 to witness the work of the Oudheidkundige Dienst at Borobudur, the result of which was the introduction of the method of anastylosis to work in Vietnam and Cambodia, initially at Banteay Srei in 1931 (‘Henri Marchal’ 2011). Trouve’s work also benefitted from Parmentier’s continual support and advice. Despite his distance from many of the EFEO’s operations, Parmentier had an important influence on much that took place throughout France’s colonial empire.

In the late 1920s, Parmentier assigned the young architect Jean-Yves Claeys to undertake a new study of the Champa remains at Trà Kiệu. Claeys, who had worked with the city-planner Ernest Hébrard (1875–1933) on the design of the Hanoi museum, later the Musée Louis Finot, then took on more ambitious stabilization projects at Pô Nagar and Bằng An in Quảng Nam. Working with a shoestring budget, Claeys depended on the help of the public works department from Danang for the work at Bằng An. In 1937, he started an even more ambitious campaign of repair and restoration at Mỹ Sơn, which he continued until the Japanese takeover of Indochina in 1945 (Baptiste 2009; Claeys 1934; see Wright 1991 on Claeys’ planning career).

Epigraphers and art historians

Parmentier and other members of the Archaeological service of the EFEO depended upon a historians and epigraphers—the latter specialists in ancient texts—to give context to their work. The principal contributors to these efforts were the school’s original philologists. These included the Sanskritist Louis Finot, Alfred Foucher (1865–1952), director of the school in 1905, and Jules Bloch (1880–1953), the first scholar trained in multiple Indian languages on the staff. Building on the initial work of the French colonial officer Étienne Aymonier (1844–1929) on Khmer and Champa inscriptions, Finot and Bloch began to reconstruct the sequence of Southeast Asian history and the place of ancient monuments in that sequence.

In 1913, the young ‘Orientalist’ George Cœdès joined the school. A student of Auguste Barth, Cœdès studied the Khmer stele in the Musée du Trocodéro and published his first epigraphical study in the EFEO Bulletin at the age of eighteen (Coedès 1906; Nugent 1966). In Hanoi, he joined the young Chinese specialist Henri Maspéro and Maspéro’s brother Georges Maspéro (1872–1942), a French colonial officer and Corresponding Member of the EFEO, who had become an expert in Champa civilization (Maspéro 1928; Porée-Maspéro 2006).

Cœdès quickly took on the Khmer inscriptions, attempting to decipher some of the more complicated texts. Drawn to the comparable ancient civilization of the Dvāravatī, which flourished in Thailand during the seventh to tenth centuries CE, Cœdès expanded the scope of his studies throughout the region, adding to his initial work on the Srivijaya kingdoms in Sumatra. Recognizing the importance of his work, the EFEO Director Claude-Eugène Maitre (1876–1925) assigned Cœdès to Bangkok in 1918, where he worked alongside the Thai Prince Damrong (1862–1943) in sorting out the inscriptions and historical sequence of ancient Thailand. Still employed by the EFEO, he retuned to Hanoi in 1929, when he became director of the school following then-director Leonard Aurousseau’s retirement (‘George Coedès’ 2011).

The work of art historian Philippe Stern (1895–1979) was the final prop in the structure of French scholarship in the region. Trained in art history, Stern wrote on a wide range of art historical topics. Before his career ended in the 1970s, he had published fully 152 works in 263 publications and in 9 languages (‘Stern, Philippe’ 2011). Serving as a curator at the Musée Guimet in Paris from 1929 to1965, he produced many museum catalogs, writing on subjects ranging from musical instruments to Khmer sculpture. It was Stern who revised the date of the Bayon temple, placing it securely in the reign of Jayavarman VII, an attribution supported by Cœdès’s own studies of ancient inscriptions. Stern’s work took place mainly in France. He traveled to Indochina only in 1936, where he looked in detail at stylistic features of Champa and Khmer temples, refining earlier chronologies and assigning new periods and dates. He summed up his work in Vietnam with an ambitious monograph L’art du Champa, published in 1942 (Stern 1942).

Other scholars working in Southeast Asia included Pierre Dupont (1908–1955) and Louis Malleret (1901–1970) and at a later date, Jean Boisselier (1912–1996; see Boisselier 1963). Dupont, who arrived in Indochina in 1936, focused on the art of the Mōn-Dvāravatī and Pre-Angkorian sculptures. Malleret first went to Vietnam in 1929, serving as curator of the Saigon museum. He became a Corresponding Member of the EFEO in 1936 and a permanent member in 1942. Although his trained as a historian, he too turned to archaeology, undertaking an extensive excavation of the Funan site of Óc-Eo in the southern Mekong Delta area. However, the end of World War II and the start of the long Indochinese War interrupted his work. When he left the site for the last time in 1944, all that remained was the exposed temple platforms of the ancient city and an incomplete archaeological record. Bosselier’s career followed in the postwar period, beginning in 1949.

The EFEO was a transparently colonial institution. Photographs of the staff invariably feature the director and his French scholars surrounded by unidentified Vietnamese men and women. Only one Vietnamese scholar became a full-fledged member of the organization before the events of the 1940s brought the school to a temporary halt. He was Nguyẽ̂nn Văn Huyên (1908–1975), an aristocratic young student of geography and Vietnamese traditional culture who Cœdès appointed as a member in 1939 (Phan Hữu Dật and Bể Viểt Đẳng 1996).

Long a bastion of gentlemen-scholars—or wealthy adventurers, such as Victor Goloubew (1878–1945)—and well-meaning French priests or bureaucrats, the EFEO produced at best a spotty survey of the past. By the late 1930s, this had begun to change with the advent of more professionally trained archaeologists and historians (and art historians), but a sense of amateurism still colored the organization until well into the postwar period. One-time member Émile Gaspardone (1895–1982), writing in the Revue de Paris struck at the ‘Indian’ bias especially, suggesting that French and Dutch scholars had pushed aside the independent achievements of the indigenous people of the region in favor of a grander vision of Hindu conquest (Gaspardone 1936). A first chip at the fortress of ‘Indianization,’ Gaspardone’s remarks presaged the tumultuous events of the next three decades.

Conclusions

As with Cambodia, Vietnam and Laos were under French control during much of World War II. After 1942, the Vichy government’s representative in Vietnam, Admiral Jean Decoux, allowed Japanese troops into former French territory, but retained the day-to-day management of the French colonies (Ruane 1998). The Japanese eventually overthrew the Decoux government, setting up a puppet regime in Vietnam headed by the Emperor Bảo Đại (1913–1997, reigned 1926–1945). In 1945, the Viet Minh resistance under Ho Chi Minh (1890–1969) founded the independent Democratic Republic of Vietnam. Shortly afterwards, the French General Jacques Philippe entered the country and declared a resumption of French colonial rule.

The EFEO continued to operate under these unsettling conditions. Led temporarily by Louis Malleret, the school reopened its offices in Hanoi in the immediate post-World War II period. There, Paul Lévy took over the directorship from Cœdès, who had returned to Paris in 1944. Shocked by departure of the Vietnamese staff, most members of which joined Ho Chi Minh’s opposition, the French staff struggled on, first in Hanoi and then in Saigon, where they began to publish the interrupted Bulletin once again. In 1960, the EFEO transferred the office in Saigon and the museum in Danang to the South Vietnamese government (Clémentin-Ojha and Manguin 2007: 42). Although French scholars would continue to aid the new governments of Cambodia and Laos, their role in the history of Vietnam’s past shifted from the field to the library.

The closure of the EFEO’s Vietnam offices marked an end of much research in Southeast Asia. The EFEO resettled in Pondicherry and Indonesia, eventually establishing a branch in Bangkok. But the great days of French-dominated studies had come to a close. Although most members of the EFEO’s small staff in Hanoi continued to focus on the evidence of ancient Champa and Khmer civilizations, a few the organization’s members had turned to Chinese and later Vietnamese contributions to the region’s history and culture. A new school of Vietnamese Studies, initiated by Georges Dumoutier (1850–1904), in fact, emerged in the early part of the twentieth century, emphasizing the importance of Chinese language on the country of Vietnam and the place of a ‘Sinicized’ culture in the historical development of Vietnam (Clémentin-Ojha and Manguin 2007: 140–54).

However, these developments did not reach fruition until the 1940s, by which time France’s colonial presence was winding down as well. When not studying the Khmer, French scholars concentrated on ancient Champa, the evidence of their inscriptions, their sculpture, and most of all their ruined temples and cities. Champa, in fact, became the linchpin of grander ideas about ‘Indianization’ in Southeast Asia.

French antiquarians were in a real way the ‘advanced guard’ or, in some instances, the ‘flankers’ of colonial expansion; administrator Étienne Aymonier and military surveyor Étienne-Edmond Lunet de Lajonquière both relied on evidence of ancient ‘Indianized’ civilizations to demarcate French interests, with Aymonier charting the ruins of Siam’s ‘Cambodian’ provinces and both he and Lunet de Lajonquière conducting reconnaissance into the mostly Lao-speaking provinces of northeast Siam. Some of this enterprise occurred after the French had already asserted a presence in the region—the Protectorate of Cambodia began in 1863, though Siam did not submit to the loss of Angkor until 1906, long after French scholars and administrators had begun their study of the region. (Laos finally became a Protectorate beginning in 1893, slightly ahead of the EFEO’s own mission but after those of Aymonier and other adventurers, notably Auguste Pavie (1847–1925) in the 1880s.) Northeast Thailand, though in the scope of French imperialists, remained part of Siam, despite the presence of ancient Khmer sites; support by the British and Siam’s own decision to assist the French in the Great War no doubt helped prevent absorption of these essentially Lao provinces into French Indo-chine (Hell 2015; also, Wyatt 2003: 217).

The members of the EFEO contributed significantly to this imperial thrust. Henri Parmentier was surveying in Cambodia even before the Thai relinquished their claims to three northern provinces. Following in the footsteps of Étienne Aymonier and in lockstep with French expansion, Lunet de Lajonquière, Finot, and Parmentier further explored ruins in Laos, demarcating the ruins at Vat Phu in Champassak before recording temples and temple ruins in Vientiane. Nearly always, the emphasis was on the structural remains of ancient civilizations, although eventually other aspects of antiquity, such as palm-leaf scripts and other kinds of material artifacts came also to become grist for the EFEO’s scholarly mill. The Cham, then the Khmer, and finally the several Laotian kingdoms, provided the markers for French involvement and expansion. The Champa sites had served as models not only for conservation practice, but also as signposts in a larger schema of cultural absorption and arrogation.