4 posts

Vol. 10, No. 1, Shimojo Hisashi


Contents>> Vol. 10, No. 1

Local Politics in the Migration between Vietnam and Cambodia: Mobility in a Multiethnic Society in the Mekong Delta since 1975

Shimojo Hisashi*

*下條尚志, Graduate School of Intercultural Studies, Kobe University, 1-2-1, Tsurukabuto, Nada-ku, Kobe 657-8501, Japan
e-mail: shimojoenator[at]

DOI: 10.20495/seas.10.1_89

This paper examines the history of cross-border migration by (primarily) Khmer residents of Vietnam’s Mekong Delta since 1975. Using a multiethnic village as an example case, it follows the changes in migratory patterns and control of crossborder migration by the Vietnamese state, from the collectivization era to the early Đổi Mới reforms and into the post-Cold War era. In so doing, it demonstrates that while negotiations between border crossers and the state around the social acceptance (“licit-ness”) and illegality of cross-border migration were invisible during the 1980s and 1990s, they have come to the fore since the 2000s. Despite the border being legally closed in the 1980s and 1990s, undocumented migration as a livelihood strategy was rampant due to the porous nature of the border. In the 2000s, the border began to be officially opened to local people and simultaneously function as a political boundary to regulate belonging and identities. The changes in migratory patterns indicate that the mapping and establishment of a national border alone is not enough to etch it in the minds of people, especially minorities who have connections with people in the neighboring country. Rather, a border “hardens” through continuous negotiations between state actors, who become suspicious of influence from a foreign country, and cross-border migrants, who become dependent on the state for their needs.

Keywords: cross-border migration, licit-ness, illegality, collectivization era, post-Cold War era, Khmer, Vietnam, Cambodia


This paper examines the history of cross-border migration by (primarily) Khmer residents of Vietnam’s Mekong Delta since 1975. Using a multiethnic village in a coastal province of Vietnam as an example case, it follows the changes in migratory patterns and control of cross-border migration by the Vietnamese state, from the collectivization era (1975–86)1) to the early Đổi Mới (renovation) reform era (1986–mid-1990s) and into the post-Cold War era (from the late 1990s). In so doing, it demonstrates that although migrants have crossed the border to take advantage of political and economic differences between Cambodia and Vietnam as a livelihood strategy for decades, the Vietnam-Cambodia border has begun to effectively function as a political boundary to regulate their belonging and identities only since the early 2000s. This regulation of cross-border migration is a result of a new political calculus on the part of (1) local residents, who have become dependent on the Vietnamese state for economic and religious needs, (2) local governments in the delta, which are suspicious of human and information flows from Cambodia, and (3) the Vietnamese state, which officially protects ethnic Khmer.

Ambiguities and the political instability of the boundary between Vietnam’s Mekong Delta and Cambodia have abounded since before colonization. Nguyen Phuc Anh began to govern the delta at the end of the eighteenth century and eventually unified almost all provinces of today’s Vietnam under a single state, officially declaring the foundation of the Nguyen Dynasty in 1802. The Nguyen Dynasty indirectly began to govern the Mekong Delta through local political leaders, but it lost control of many parts of the delta when Khmer rebellions against the centralization erupted in the 1830s (Kitagawa 2006, 182–183, 190, 221–222; Vũ Đức Liêm 2016, 89–91). As Vũ Đức Liêm observes, the Nguyen Dynasty attempted to demarcate boundaries (giới) by digging canals such as the Vinh Te canal, which is located not far from today’s national border, and to translate those physical markers into cartography. It did not, however, differentiate provincial boundaries from the national border when managing the “Khmer world” (Vũ Đức Liêm 2016, 93), and in that sense it can be said that the modern national border did not emerge during the Nguyen Dynasty. In the late nineteenth century, the delta and Cambodia were integrated into a federation of states known as French Indochina.2) Although the administrative line drawn between the two regions during the French era began to work as a national border in 1954, intensification of the Vietnam War during the 1960s, when several political forces collided,3) brought disorder to the borderland (Chandler 2008, 242–249). Even during wartime, people created many open-air marketplaces along the border in areas uncontrolled by the states (Lê Hương 1970, 10), and people migrated from Southern Vietnam to Cambodia to seek a higher Buddhist education or livelihood. However, as the Vietnam War expanded into Cambodia, notably in 1970—when the Norodom Sihanouk regime was overthrown in a coup by general Lon Nol, whose army units massacred hundreds of unarmed Vietnamese civilians near Phnom Penh on the dubious grounds that they were allied with the Communists (Chandler 2008, 251)—a large number of Vietnamese citizens in Cambodia returned to Vietnam.4)

Following the end of the Vietnam War in 1975, conflict between the unified Vietnamese government and the Khmer Rouge increased political tensions in the borderland. These tensions eased when Vietnam invaded Cambodia and established a pro-Vietnam government in Phnom Penh in 1979. Starting in 1975, the Vietnamese socialist economy stagnated, making it difficult for people to make a living. This, combined with the easing of border tensions, spurred people in the delta not only to flee Vietnam by sea as “boat people” but also to migrate overland to Cambodia and then to Thailand as refugees, laborers, or traders to access means of livelihood. Migration continued throughout the 1980s and increased in the early 1990s, when the economic boom in Phnom Penh under the governance of the United Nations offered promise in the face of the social dislocation and economic difficulty caused by Đổi Mới reforms and a new market-oriented economy in Vietnam.

According to Evan Gottesman (2003, 165–168), during the Vietnamese occupation of Cambodia, the Vietnamese military established checkpoints along the border; and anyone from Vietnam who attempted to flee to Thailand via Cambodia was detained, disciplined, and forced to return to Vietnam. Gottesman also notes, however, that cross-border smuggling may have been secretly permitted by both the Hanoi and Phnom Penh governments during the Vietnamese occupation of Cambodia. Indeed, we know that Cambodian timber and rubber were smuggled into Vietnam, and beer and Vietnamese electric generators were smuggled into Cambodia (Gottesman 2003, 311–314). Research conducted for this paper confirms that from 1979 through the 1990s local migrants crossed the border using informal routes to avoid checkpoints, suggesting that border control was ineffective and undocumented migration occurred during that time.

It is generally assumed that the decollectivization and market liberalization stimulated by Đổi Mới reforms in 1986 eased economic difficulties in Vietnam on the whole, though there were still many poor in rural areas, especially areas with high ethnic minority populations, and the gap between the rich and the poor tended to widen (see Vo Tri Thanh and Pham Hoang Ha 2004, 83–84). According to Philip Taylor (2004), the result of decollectivization and land liberalization policies was that increasing numbers of Mekong Delta farmers had to sell their land and thus had—and still have—less access than ever to the profits from the delta’s agricultural economy.5) Taylor also mentions briefly that many farmers in the delta, especially Khmer people who missed out on many benefits of the liberal reforms, sought work in the urban centers of the delta, or in nearby cities such as Phnom Penh and Ho Chi Minh City (Taylor 2004, 237, 245, 252). Taking into consideration these arguments over Đổi Mới, this paper puts into focus the pattern of those who, failing to cope with the rapid penetration of the market economy, began to migrate to Cambodia in order to seek a livelihood.

Andrew Hardy observes, “at the heart of the Đổi Mới transition lay the process of land decollectivisation and it was this, above all, that undermined the state’s control over population mobility” (Hardy 2003a, 124). While Hardy’s (2003a; 2003b) research focuses mainly on the contemporary history of government-sponsored and spontaneous internal migration from the northern Red River Delta to Vietnam’s highlands, Iwai, Ono, and Ota analyze migrations from the Red River Delta to the floodplain of the southern Mekong Delta since the Đổi Mới reforms (Iwai et al. 2016). This paper deals with cross-border migration between Southern Vietnam and Cambodia as an extension of such internal migrations.

Beginning in the late 1990s, Vietnam gradually returned to international society, accelerating its market economy. Subsequent economic growth during the 2000s witnessed an explosion of industries around Ho Chi Minh City, which began to absorb the rural population. Reflecting this pattern, some people who had moved to Cambodia returned to their home in Vietnam due to its improving social and economic situation. In tandem with this phenomenon, control of migrants crossing the Vietnam-Cambodia border intensified and became more firmly institutionalized from the 2000s, as the government became more suspicious of the human and information flows from Cambodia.

One notable exception to this pattern is the regulation of Khmer monks from Vietnam crossing the border to study in Cambodia. This movement has been strictly regulated since 1975, due to Vietnam’s concern over those monks being influenced by the environment of anti-Vietnamese politics in Cambodia, as Taylor’s ethnographic studies observe (Taylor 2014, 59; 2016, 282–285).6) While Taylor’s research stresses Khmer monks’ migration to and from Cambodia, this paper focuses on the interaction between laypeople’s migration and state control, whose pattern is different from that of the monks.

Timothy Gorman and Alice Beban (2016) reveal that contemporary migrant workers from Vietnam cross the Cambodian border as farmers to cultivate shrimp, and Sango Mahanty (2019) analyzes traders circulating between Vietnam and Cambodia to buy and transport cassava. According to these studies, in order to mitigate their legally unstable position or to facilitate the transnational trade of cash crops, these border crossers voluntarily attempt to establish ties with state actors, such as border guards, military personnel, or local authorities.

Some migrants circulating between both countries today survive by making use of the social insurance provided to the poor in Vietnam and the economic boom of Cambodia’s borderlands. In so doing, they rely not only on the state’s insurance programs but also on middlemen in their homeland to arrange official documents such as passports and ID cards.

By explaining the historical changes in cross-border migration trends, this paper betters our understanding of why migrants from Vietnam have come to depend more on state actors since the 2000s. The paper is structured as follows. The first section presents an overview of the migration between Vietnam and Cambodia while reviewing some literature on borders and migration. After explaining the field site of P. village in the second section, the paper details the patterns of undocumented migration during Vietnam’s collectivization era (1975–86) in the third section, and during the early Đổi Mới reform era (1986–mid-1990s) in the fourth section. The fifth section discusses the circulation of migrants and Vietnamese regulation of the border and migrants in the post-Cold War era (from the late 1990s). In examining the changing migration trends and state regulations, this paper concludes that while interactions between border crossers and the state around the social acceptance (or “licit-ness”) and illegality of cross-border migration were invisible during the 1980s and 1990s (when undocumented migration was rampant despite the border being legally closed), they have come to the fore since the 2000s, when the border was institutionally opened to local people.

I Creation of a Border

In Southeast Asian historical studies, it is assumed that since the mapping and demarcation of modern national borders, the formation of the state and creation of nationhood have been actively pursued (Thongchai 1994). The border gradually emerges in people’s imagination of a nation through migration control policies such as exclusion or greater scrutiny (Osada 2011). Given this “hardening” of borders, some studies focus on how migrants strategically take advantage of social and economic differences on either side of the border, engaging in sophisticated cross-border smuggling using various, often corrupt, networks (Tagliacozzo 2005; Ishikawa 2010).

Itty Abraham and Willem van Schendel (2005) refer to the borderland as both a “licit” and an illegal space. They define licit activities not as permissible by law, but rather in contrast to the popular sense of “illicit” activities; in other words, licit activities are legally banned but socially sanctioned and protected. The borderland is a space formed by the intersection of multiple competing authorities and enforcements. Neighboring states often hold different views on both the law and licit-ness. As a result, what is considered licit, or what may be allowed on one side of the border, may be considered strictly illegal and not allowed on the other side. As people cross borders to work in sweatshops and brothels to avoid labor regulations or the vice police, strategic interactions or “border games” ensue between border enforcers and unauthorized border crossers (Abraham and Schendel 2005, 22–23).

In light of the border and migrant studies mentioned above, the case of P. village in Vietnam raises the question of how concretely migrants have etched the national border in their minds as a “boundary” to differentiate their own society from the society on the other side of the border. Many people in P. village migrated to Cambodia during the 1980s and 1990s to escape economic difficulty. This undocumented migration was not fully controlled by either of the two states, whose lax governments did not trace the mobility of border crossers after their initial border crossing or during long stays. This enabled the people in P. village to continue to recognize their own homeland and Cambodia as a mutually connected cultural and economic space, even after the establishment of the Vietnam-Cambodia border. Border crossing through informal routes was frequent at least until the 1990s, as border crossers relied on backstreets along the border, long-distance family or relative networks, personal experiences, and middlemen preparing land or sea transport. The cases from P. village demonstrate that the Vietnam-Cambodia border was not clearly “etched in the minds of people” as a boundary to regulate mobility or identities, and it did not strictly function as a political institution, as it was virtually porous and permeable until the late 1990s. Schendel (2005, 52) notes that the permeability of a border can differ along its length based on physical features, intensive policing of a particular section, cross-border agreements, and the varying degrees of physical or linguistic difference between borderlanders on either side. Due to a combination of these factors, the Vietnam-Cambodia border was very porous, and although not strictly legal, migration between the countries was habitual and licit, in that it was socially sanctioned among P. villagers and even overlooked and tolerated by the states.

However, the Vietnamese state has paid increasing attention to this undocumented migration and lack of border governance, especially since the early 2000s. Today the Vietnamese government pays particular attention to Khmer people crossing the border, due to the sensitivity of the territorial politics of the Mekong Delta, which in Cambodia is still called Lower Cambodia (Kampuchea Kr[a]om) and considered a territory taken by Vietnam. In the late 1970s, the Pol Pot regime raided the borderland on the Vietnam side with the aim of reclaiming the Mekong Delta (Chanda 1986, 96–98). However, the political presence of Khmer in the delta was temporarily forgotten when the pro-Vietnam government was installed in Phnom Penh after the Vietnamese invasion of Cambodia. Following Vietnam’s withdrawal of troops from Cambodia, the Vietnamese government lacked the ability to prevent local residents from crossing the border, and serious political problems relating to the border, migrants, and Khmer ethnicity did not surface until the late 1990s. However, since the early 2000s, as people, goods, and information from Cambodia have continued to flow into Vietnam, the Vietnamese government has become concerned that if the Khmer people have more contact with Cambodian society, previous political issues might resurface. The regulation of migrants from P. village, many of whom identify as ethnic Khmer, is closely connected with Vietnam-Cambodia border governance. As explained below, only since the 2000s have interactions between unauthorized migrants and territorial states around the licit-ness and illegality of cross-border migrations started to make themselves visible. This has happened even in P. village, which is located quite far from the border.

II Methodology and Field Site

This paper is based on oral histories and ethnographic data collected in P. village ()7) between December 2010 and March 2012. I lived and conducted fieldwork in the village for 15 months, collecting ethnographic data about society-state relations in everyday lives and oral histories of people’s survival strategies to avoid subsistence crises during wartime, the collectivization era, the early Đổi Mới reform era, and the post-Cold War era. I also conducted several one- to two-week research trips in the borderlands of Cambodia and Vietnam, as well as Phnom Penh, to gather information related to migration trends.

P. village is located 150 km from the Cambodian border on the east bank of the Bassac River in Soc Trang, a coastal province in the Mekong Delta of Vietnam. Soc Trang is home to many ethnic Khmer who have remained connected with Cambodian society through their language and Theravada Buddhism even after national border demarcation in the mid-twentieth century. The ethnic Khmer have lived together and intermarried with ethnic Chinese and Viets (ethnic Kinh, the majority in Vietnam).8)

According to Vietnamese government statistics from 2009, ethnic Viets (Kinh) are the largest ethnic group in the Mekong Delta, accounting for about 92 percent of the population; ethnic Khmer account for around 7 percent and ethnic Chinese just 1 percent (see Table 1). Within the delta, Soc Trang is ethnically diverse, with 64 percent of the population ethnic Viet, 31 percent Khmer, and 5 percent Chinese. The province is also recognized as having the largest Khmer population in Vietnam and the largest Chinese population in the delta. Within Soc Trang Province, P. village has a higher percentage of Khmer residents. Government statistics from 2011 show that 79 percent of P. village is ethnic Khmer, followed by ethnic Viet (19 percent) and ethnic Chinese (2 percent).9) In my survey, according to the ethnicity registered with the government, of the 395 people living in “Samrong ward (khu),”10) Q. hamlet (ấp), P. village, 390 were ethnic Khmer, 3 were Viet, and 2 were Chinese. However, 170 of the 390 people registered as Khmer had Chinese (154), Viet (13), or both Chinese and Viet (3) kinship ties among their parents, grandparents, or great-grandparents on their father’s, mother’s, or both sides. Intermarriage between different ethnic groups is common, and as a result many villagers today identify themselves as ethnically mixed, or métis (in Vietnamese [VN], lai; in Khmer [KH], cat), and speak multiple languages.


Table 1 Populations of Vietnam, Mekong Delta, Soc Trang, and P. Village



Most residents of P. village maintain connections with Cambodian society through the Khmer language and Theravada Buddhism. Despite its distance from the Cambodian border, the village has two Khmer temples, from which monks, since at least the early twentieth century, have traveled to temples in Cambodia to undertake higher religious and secular education or to engage in meditation practices (Shimojo 2015, 24–33). The majority of villagers have relatives and acquaintances who travel to Cambodia for work, and many routinely cross the Vietnam-Cambodia border.

Examining changes in the local politics of P. villagers’ migration patterns enhances our understanding of the Vietnam-Cambodia border and cross-border migration for two reasons. First, the mobility of ethnic Khmer, who are the majority in P. village and in Cambodia, often leads the Vietnamese state to give importance to regulating the border and cross-border migration. Second, the village’s location far from the border exemplifies broader changes in migratory patterns in the Mekong Delta more accurately than a location on the border, reflecting which state (Cambodia or Vietnam) or which economic center (Phnom Penh or Ho Chi Minh City) historically attracted people in the delta in response to the economic and political conditions of each era. Analysis of societies located in the borderlands, where people routinely cross the border for their economic and social needs, may not reveal such historical migratory patterns.

III Migration during the Collectivization Era (1975–86)

III-1 Refugee Migration

Refugees who fled Vietnam from 1975 to 1986 during collectivization were generally regarded as “boat people” who crossed the ocean (vượt biển). However, many migrants in P. village evacuated over land, relying on their local knowledge of Cambodian society to move from Vietnam to Thailand via Cambodia. The number of such people from P. village increased after early 1979.

Following the reunification of North and South Vietnam, the socialist regimes of Vietnam, China, and Cambodia entered a new phase in their interrelations, shifting from an outward “fraternal” cooperation (in support of Communist revolutionary movements) during the Vietnam War era to “fratricidal” conflict in the post-Vietnam War era. The Pol Pot regime spread anti-Vietnam sentiments, insisted on claiming territorial rights to the Mekong Delta, and frequently collided with Vietnamese military personnel stationed in the borderlands. Ultimately, Vietnam invaded Cambodia in December 1978 and occupied Phnom Penh in January 1979. To support the Pol Pot regime, China invaded Vietnam in February 1979, fighting in the Sino-Vietnam borderland (Elliott 1981, 8–10; Evans and Rowley 1984, 39, 45–57; Chanda 1986, 231–262; Kimura 1996, 76–90).

Due to the two international wars, ethnic Khmer and Chinese, and métis Khmer- and Chinese-Viets in P. village—whose loyalty to Vietnam was now suspect—were placed in a politically sensitive situation. In addition to the worsening of China-Vietnam state relations, most had lost their livelihoods due to the Communist government’s promotion of collective farming and redistribution of land, its direct purchase of food, and its establishment of a rationing system. Landlords, rice merchants, millers, and shopkeepers who had engaged in the production and distribution of exported rice were forced to close their businesses.

In response to the political and economic situation, some people decided to flee by land. Com (a Chinese-métis Khmer), a family member of a once-rich landlord in P. village, described the journeys of his younger brother and sister:

In Vietnam at that time, people were forced to participate in labor service [VN: công tác, KH: polakam], but they had no work of their own and no way to earn money. My younger brother crossed the Cambodia border and arrived in Thailand in 1979. He moved from Thailand to the Philippines by boat, and in the end he migrated to Canada. My younger sister and her husband arrived at a refugee camp in Thailand via Cambodia and migrated to Canada. Her husband had been engaged in the “revolution’s work,” but he had a difficult life.11)

As Com’s brother-in-law’s case reveals, the refugees included those who had participated in the “revolution’s work” as cadres of the new government, indicating that economic difficulty was a more serious factor than political positions when deciding to flee. Having local knowledge of Cambodia was also an important factor for those choosing to flee by land. Com’s eldest stepsister, Sang (an ethnic Khmer), pointed out that her younger brother had been sent to Phnom Penh as a South Vietnamese soldier during the Vietnam War, and therefore he was familiar with the city.12)

Upon Vietnam’s successful invasion of Cambodia, tensions in the borderlands caused by the international conflict rapidly eased and more people fled overland, defying the border and making use of their own language and past experiences.

III-2 Circulating Migration

Beginning in 1979, migrant workers and traders as well as refugees repeatedly passed thorough the porous border. Some people in P. village circulated repeatedly between Vietnam and Cambodia. Rat, an ethnic Khmer, said:

After the liberation, I didn’t engage in collective farming because it was troublesome. After the Pol Pot regime collapsed in 1979, I traveled every two weeks between Phnom Penh and P. village as a ku roup [painter]. I sold pictures for 5,000 riel each until I stopped going to Phnom Penh one year later. In Vietnam fertilizer and agricultural chemicals were scarce and the land was not fertile, but business was free in Cambodia. I passed through Kompot in Cambodia via Ha Tien by car to go to Phnom Penh. I didn’t cross the formal border but entered Cambodia by an informal route.13)

Although ku roup in Khmer means “painter,” strictly speaking Rat was not a painter. To earn money in Cambodia, he collected pictures of deceased people from bereaved families and brought those pictures to a painter in P. village. The painter created portraits of the deceased, and Rat then returned to Cambodia to sell them to the bereaved families.

Rat went to Cambodia not only because the restrictions on private activities in the black economy there were comparatively lax, but also because he was accustomed to Cambodian society. He had been to Phnom Penh as a monk to study Pali at Unalom Temple for a year in the late 1960s, as many Khmer monks in the delta had in the past. Rat could also take advantage of the informal car route connecting P. village with Phnom Penh via Ha Tien and Kompot (see Fig. 1) that was being created at the time. Rat’s narrative shows that repeated border crossing was possible because of the porous border.



Fig. 1 Mekong Delta and Vietnam-Cambodia Border

Source: Esri, GEBCO, DeLorme, NaturalVue | Esri, GEBCO, IHO-IOC GEBCO, DeLorme, NGS.


In the mid-1980s, some people even migrated to Thailand via Cambodia to engage in undocumented work. Han said:

I went to Thailand by car without a passport to work as a laborer [KH: kammakor], from 1986 to 1992. First I stayed in Phnom Penh for half a month, and after that I moved to Thailand. I did physical labor like loading and earned about 20 baht per day. The pay was a little bit better in Cambodia than in Vietnam at that time. Both Vietnam and Cambodia were under Communist rule, but Thailand was a liberal country. I stayed in Surin Province, Thailand, and I worked on a cassava farm for 20 baht per day. I could understand 80 percent of the Surin Khmer dialect. I don’t know which road I used to cross the border, because I just went along with others and got into a car. I was a manual laborer [KH: si chhunual ke]. Vietnam was very poor, and furthermore, we were obligated to participate in labor service for one month per year. Although I had farmland in 1985, we only produced one annual rice crop, which was not profitable.14)

It was possible for P. villagers, many of whom could speak Khmer, to work not only in Cambodia but also in Surin Province, Thailand, where many ethnic Khmer resided. Han’s narrative reveals that groups of migrant workers from Vietnam crossed the border via informal routes with the help of mediators who prepared transportation for workers’ groups to Cambodia and Thailand.

The two narratives above show that people from Vietnam went to Cambodia seeking less intervention in their lives and a freer economic space. After the Pol Pot regime was driven out of Phnom Penh in 1979, many people in the delta, especially Khmer speakers, crossed the Vietnam-Cambodia border to escape difficulties in their lives brought about by state policies. Migration from Vietnam to Cambodia and Thailand during collectivization, not only to escape the political situation but also to partake in cross-border trade and migrant work, was possible because in the face of political chaos, the state powers could not control the porous border and the human flows searching for informal routes and work. During this time, although cross-border and internal migration was legally banned or restricted, it was socially licit in P. village and carried on across informal routes stretching from the village to Cambodia and on to Thailand. In that sense, the Vietnam-Cambodia border at the time did not function as a political apparatus to regulate interactions between migrants and the state or to functionally regulate the licit-ness or illegality of migrations.

IV Migration during the Early Đổi Mới Era (1986–mid-1990s)

IV-1 Withdrawal of the Vietnamese Military from Cambodia

At the beginning of the 1990s, the political situation in Cambodia began to change, prompted by the withdrawal of the Vietnamese military from that country. For a decade, Vietnam and the pro-Vietnam government in Cambodia (the Cambodian People’s Party) had fought with Pol Pot’s guerrillas and other anti-Vietnam forces that were supported by China and several Western countries. However, after mediation by the Soviet Union, which hoped to normalize its relations with China, Vietnam began to withdraw its military from Cambodia in June 1988, completing its withdrawal in September 1989 (Kimura 1996, 222–232; Gottesman 2003, 336–350). After the withdrawal, the political situation in Cambodia began to rapidly and drastically change. In 1991 the pro-Vietnam government and anti-Vietnam forces in Cambodia concluded the Paris Peace Accords. In 1993 the United Nations Transitional Authority in Cambodia (UNTAC) implemented a parliamentary election, which resulted in the establishment of the Kingdom of Cambodia and major changes in the country’s politics and economy. By 1989 the Cambodian People’s Party government had all but abandoned the socialist economic system, and in 1993 the new Kingdom of Cambodia government officially introduced a market economy system (Gottesman 2003, 280–300; Amakawa 2004, 4–14).

The change in the military power balance, political pluralization, and the introduction of a market economy in the early 1990s further motivated people in P. village to attempt to cross the border to escape economic difficulties.

IV-2 Undocumented Migration of the Poor into Cambodia

During the period of political change in Cambodia, the number of migrant workers from P. village seeking to escape economic difficulties increased to levels higher than the 1980s. The villagers I interviewed said that in the early 1990s the Cambodian economy was better than the Vietnamese economy, and that a significant number of Vietnamese citizens lived in Phnom Penh. In Samrong ward, Q. hamlet of P. village, the number of migrants going to Phnom Penh to work was markedly increased from 1992 to 2002 (see Table 2). To determine how and why people went to Phnom Penh, I analyzed the oral histories of several people in P. village who had made border crossings.


Table 2 Reasons for Migrating to Phnom Penh from Samrong Ward, 1979–2011 (unit: one person)



The main reason why people migrated to Cambodia was their inability to cope with the rapid development of the market economy in Vietnam after the initiation of Đổi Mới in 1986. Đổi Mới reforms brought private economic activities that had been part of the black economy during the Collectivization Era out into the open. As people began to engage in a wider range of crop productions, commercial production was diversified. In P. village, the cultivation of watermelon was popular at the end of the 1980s. According to an official monograph published by the communist party in Soc Trang province in 1988, trucks from Ho Chi Minh City and Can Tho would gather at P. village just before Tet (the lunar new year), and merchants would purchase watermelon produced in the village (Vũ Lân et al. 1988, 61).

Than, a Khmer-métis Chinese villager who lived in Samrong ward, Q. hamlet of P. village, told me that he cultivated watermelons from 1988 to 1990. He and his wife transported the harvest by bus to sell in Ho Chi Minh City. The variety of watermelon he grew was large but did not taste very good, so he could sell the fruit only as decoration for the Tet celebration. Since he could harvest only once per year, Than found it increasingly difficult to make sufficient profit from this endeavor and eventually stopped. After that he mortgaged his farmland to Hang, who was also a resident of Samrong ward, for 100 kg of rice at twofold interest before going to Phnom Penh (rice was desired because it was a valuable currency during the economic dislocation of the time).15) Although some people, such as Than, attempted to produce new agricultural products, they did not receive any state support in the form of technical assistance, investments, loans, etc. Thus, they were at high risk of failure or becoming dependent on private moneylenders such as Hang, who often charged high interest rates.

Than said that he went to Phnom Penh for a year and a half from 1991 to 1992 with his wife and youngest daughter. Without a passport, he crossed the border not through a border gate but via an informal route. He worked as a traveling wharf laborer in a town near Phnom Penh, loading commodities onto a ship and traveling on the ship to Saigon, where he unloaded the items, and then repeating the cycle.16)

People moved to Cambodia also to be closer to relatives. According to Than, his adopted brother (his parents’ adopted child) had lived in Phnom Penh since 1982, so Than went to Phnom Penh relying on his relationship with his brother. His brother earned enough to live on, working in a factory manufacturing bottle lids. Than said that when he lived in Phnom Penh, he met many ethnic Viets who had migrated from Vietnam.17)

Don, another P. villager, went to Phnom Penh with his family to work during the 1990s due to food shortages, insufficient profit from agriculture, and lack of agricultural expertise. He explained that he had sold all of his land because his wife was sick and his five children were very young.18) According to his acquaintance, Don sold his land because, like many other villagers, he had failed in his watermelon cultivation enterprise.19)

Thus, Don went to Phnom Penh to escape the economic difficulties experienced by farmers in rural villages in the Mekong Delta and to take advantage of the economic boom in Cambodia. Under UNTAC, Phnom Penh offered a significant economic incentive that attracted poor wage laborers: US dollars were used by the foreign militaries stationed there, and the value of the US dollar was much more stable than that of Vietnamese đồng or Cambodian riel.

Don’s wife said that during the 1990s only agricultural wage labor such as rice harvesting was available in Vietnam. The average daily wage of a manual laborer in Cambodia was 5,000 riel for a woman and 8,000 riel for a man, but it was just 4,000 riel in Vietnam.20) In Phnom Penh, Don’s wife took care of housework while Don worked as a carpenter for 150,000 đồng per day.21) According to Don’s wife, the potential earnings in Phnom Penh meant that not only Khmer people born in Vietnam, but also many ethnic Viets, lived in Cambodia.22)

The final reason why Don’s family chose to go to Phnom Penh was that Don’s older sister had moved there before the Pol Pot regime was established,23) although, according to Don,24) she died during the Pol Pot era. Don’s wife explained that one reason they chose not to go to Ho Chi Minh City was because they had no acquaintances there. Thus, Don’s family chose to go to Phnom Penh to work because the city was not an unknown place. The experience of Don and others highlights the importance of family or acquaintance networks in choosing the routes migrants traveled along and their destinations.

During the 1990s, like the 1980s, circulating migrants traded commodities and worked in the borderlands of the two countries. Con, who lived in P. village, told me that in the early 1990s there was no work in Vietnam and wages were extremely low, even in Ho Chi Minh City. To earn money, he traveled repeatedly between Cambodia and Vietnam from 1992 to 1993. His father, Rat, as mentioned above, also repeatedly crossed between Cambodia and Vietnam as a ku roup (painter) around 1979, after the collapse of the Pol Pot regime. Con said that he also went twice a month to Takeo Province in Cambodia, near the Vietnam-Cambodia border, by local bus (xe đò) and collected pictures of deceased people by going door to door. Without a passport, he crossed the border either near Cam Mountain in Tinh Bien District or Sam Mountain in Chau Doc city in An Giang Province (see Fig. 1). His maternal uncle had migrated to Cambodia in the 1980s and worked as a goldsmith in Takeo Province.25) Con stayed at his uncle’s house.

The photographs in Fig. 2 are of the area surrounding a rice packaging factory located near the border in Tinh Bien District, An Giang Province in Vietnam, which is adjacent to Takeo Province in Cambodia. The borderland is located in a vast floodplain area, which is often inundated during the rainy season. The photographs show that there is no natural geography or official facilities to prevent people from crossing the border. In terms of intensive policing (Schendel 2005, 52), the Vietnam-Cambodia border, notably the Tinh Bien-Takeo borderlands, was more porous and permeable for people in P. village during the early Đổi Mới era than during the collectivization era. In addition, because internal migration was tolerated by the government after Đổi Mới, more and more people relied on informal and licit routes that had been continuously developed in several places since pre-Đổi Mới times, and where licit middlemen and networks prepared ships or land vehicles to enable cross-border migration.26)



Fig. 2 Landscape of the Vietnam-Cambodia Borderland Visible from Tinh Bien District, An Giang Province

Sources: Left: Satellite photo obtained from DigitalGlobe, Earthstar Geographics | NOSTRA, Esri, HERE, Garmin, METI/NASA, USGS.
Right: Photograph looking toward the Vietnam-Cambodia border taken by the author at the location of the white spot in the photograph on the left.


IV-3 Economic Attraction of Phnom Penh in the Early 1990s

Whereas some people, such as Than, Don, and Con, relied on connections with family or relatives to go to Cambodia, others went based on their own experiences. For example, Thu, a Chinese-métis Khmer woman from P. village, had lived in Phnom Penh with her late husband from 1968 to 1973 or 1974, during the Vietnam War. After the initiation of Đổi Mới she was too poor to keep her land, so she sold it and returned with her five children to Phnom Penh, where she lived from 1992 to 2010. Without passports, Thu’s family took a local bus with what little money they had and crossed the border via an informal route in Tinh Bien District. When the family arrived in Phnom Penh, Thu found that all of her husband’s brothers had died during the Pol Pot era, and therefore she had no acquaintances left in the city. According to Thu, the prices of commodities in Cambodia in the 1990s were higher than those in Vietnam, but wages were better because Cambodia was economically more liberal. In 1993, when the United Nations established UNTAC, US currency became widely used in Cambodia. Thu opened a kiosk and sold food, while her children worked as laborers. She said that many ethnic Viets lived in Phnom Penh at that time. According to her, ethnic Viets, who comprised a larger population than Kampuchea Krom (Khmer born in Vietnam),27) spoke the Khmer language fluently and operated street stalls in the city.28)

During the early 1990s, people migrated to Cambodia because it offered better economic opportunities (including relatively higher wages). Although economically motivated, such migration relied on long-distance family or relative networks and personal experiences. The number of people crossing the border from Vietnam to Cambodia increased sharply in the early 1990s due to the rural economic depression in the Mekong Delta and political and economic transformation in Cambodia. Finally, undocumented border crossings were not fully controlled by the government because border governance at that time had not been sufficiently developed, and people noticed that many points along the border were still porous, especially in Tinh Bien District. Depending on family and other networks, and past memories and experiences connecting the two countries, people created informal and socially licit routes stretching out from their village toward the border and beyond.

V Migrants and Border Control in the Post-Cold War Era (from the Late 1990s)

V-1 People Circulating Legally between Vietnam and Cambodia

As mentioned above, the political and economic situation in Vietnam began to change in the 1990s. Vietnam joined ASEAN in 1995, resulting in the removal of an embargo on external assistance, trade, investment, and loans entering the country. The main reasons for this change were the United States beginning to lift its economic sanctions on Vietnam after the signing of the Paris Peace Accords in 1991, the lifting of the US embargo on Vietnam in February 1994, and the US normalizing its diplomatic relations with Vietnam two weeks after the latter joined ASEAN (Kimura 1996, 283–285; Elliott 2012, 147–155). Vietnam became increasingly accessible to international society, and it became easier for Vietnamese who had moved overseas to return home to live.

The flow of people from P. village changed in line with the transformation in international politics. The major destination for people from P. village gradually changed from Phnom Penh to Ho Chi Minh City, the economic center of Vietnam, as the economic lure of Phnom Penh weakened. As shown in Table 3, the number of workers from Samrong ward migrating to Ho Chi Minh City and the surrounding suburbs was markedly increased between 2001 and 2011. Many of the migrants found work at timber- or fruit-processing companies, industries that were growing rapidly around Ho Chi Minh City. In contrast, the number of workers migrating to Phnom Penh was low between 2003 and 2011 (see Table 2). However, the number of people traveling to Phnom Penh to visit family members increased between 2001 and 2010, suggesting that although the transnational relationship between P. village and Phnom Penh remained, the number of people moving to Phnom Penh for economic purposes was decreasing.


Table 3 Migrant Workers from Samrong Ward in Ho Chi Minh City and Its Suburbs by Type of Work, 1997–2011 (unit: one person)



In P. village, I often saw middle-aged or elderly people staying in the village for a few days or months but then suddenly leaving without notice. They would come from Cambodia and spend their time at a coffee shop engaging in enthusiastic chitchat with villagers regarding current news and rumors from Cambodia. They stayed with their children who lived in the village.

Some P. villagers earned enough to live in both countries by making strategic use of each country’s unique political and economic situation. For example, Son, who was staying temporarily in P. village in August 2011, told me:

From 1978 to 1979 I worked as a laborer in Ho Chi Minh City, but from 1983 to 1990 I migrated to Cambodia to work. In Cambodia, I sold bread as a day laborer. After that I returned to Vietnam, but even today I go to Cambodia every four or five months. At the moment I am staying in P. village for three or four days until my passport is renewed, and then I plan to return to Cambodia. Every time I cross the border, I do so in Tinh Bien District. There are three casinos in Ratanakiri [Cambodia]. One of them is managed by Prime Minister Hun Sen. The casino’s customers come mostly from Vietnam, so my daughter works there because she can speak Vietnamese. I have no land in Vietnam, but I have 20–30 công (2.6–3.9 ha) of land in Ratanakiri.29) There are rubber tree plantations, and some foreigners are mining for gold there. In Vietnam, I can use medical services for free because I am recognized as a “poor household” by the government. Every time I get sick, I come back to Vietnam. When I have spent all of my money, I call my daughter and ask her to send me USD200–300.30)

Son survives thanks to the social insurance for the poor in Vietnam and the economic boom in the Cambodian borderlands. He went to Cambodia and worked as a laborer in an undocumented way at least until the 1990s, as did many villagers. As he recently verified his Vietnamese citizenship by getting a passport, he now simultaneously uses the preferential treatment for poor households in Vietnam while depending for money on his daughter who works at a casino near the border in Cambodia. Only foreigners are permitted to gamble at casinos in Vietnam, so many casinos have been built in the borderland on the Cambodian side, catering to customers from Vietnam.

V-2 Government Suspicion of Khmer Cross-border Migrants

Why is Son recognized by the Vietnamese government as belonging to a poor household, and how can he use the Vietnamese social insurance system? Starting in 2001, Vietnam’s government implemented policies to eliminate poor households, especially among ethnic minorities in the Mekong Delta.31) Today the local government in P. village gives high priority to social insurance policies for poor Khmer households. The government exempts these households from school expenses and provides housing or assistance for their living expenses.32) However, this special status is based on stereotypes and suspicion against Khmer within the Vietnamese government and society, which have historically considered Khmer to be poor. An official 1991 Communist Party document reports that ethnic Khmer do not receive enough of an education; therefore, some ethnic Khmer Communist Party members violate national policies because they do not understand the meaning of the ethnic and religious policies of the Party-state. The document also notes that the matter has had a negative impact on society, the economy, sentiment, thought, and politics.33) Therefore, the government now focuses on Khmer people in order to avoid such negative impacts. The document continues:

We must fight against the plots and means of the hostile forces who make use of historical issues, religious ethnic problems, and trifling mistakes, and distort plans, cause splits, stir up ethnic animosity, spread nasty rumors, and disrupt the implementation of the advocacies and policies of the Party-state.
 We organize and support ethnic Khmer in Southern Vietnam who hope to cross the Vietnam-Cambodia border to visit their relatives and acquaintances based on state laws, while restricting border crossings between the two countries. That is, we take people’s convenience into consideration while protecting public order in both our country and our neighboring country.34)

Thus, the Vietnamese government constructed a social insurance system for ethnic Khmer in Vietnam, a border control system, and a system to regulate migrants in order to eliminate negative impacts on the Party-state. In addition, the government, which was not stable enough to realize its political and economic policies until the 1990s, has attempted since the turn of the twenty-first century to integrate ethnic Khmer into the nation-state and cut off influence from Cambodia with the provision of social insurance.

One day in 2011, while staying in Soc Trang, I interviewed an official who recounted the following incident:

In 2009, before you came to Soc Trang, Khmer monks launched a protest against the government in Soc Trang Province. The protest broke out when traffic police arrested two monks for the violation of riding on the backseat of a motorbike. That affair was distorted and spread by evil people, who manipulated the monks into starting the protest. This incident is referred to as the “August Affair” among police.35)

Informants mentioned that the Vietnamese authorities were wary of “evil people,” or members of political organizations formed by Khmer going from Vietnam to Cambodia who identified themselves as “Khmer Kampuchea Krom (Khmer of Lower Cambodia).” Using various forms of media, such organizations accuse the Vietnamese Communist government of ignoring historical problems, human rights violations, and religious oppression of ethnic Khmer in Vietnam.

Taylor analyzes the contesting narratives as follows:

Although these contesting claims on the Khmers of Lower Cambodia/the Khmer minority of Vietnam are held with equal passion, each is coloured by deep ambivalence. Many Cambodians suspect that the Khmer Krom have been subject to Vietnamese assimilatory rule for so long that they are no longer fully Khmer. For their part, Vietnamese officials lament the recalcitrance of a minority group whom they label pejoratively as backward, insular and marginal, and whose continuing identification with Khmer nationalist mythology threatens the integrity of the Vietnamese nation. (Taylor 2014, 252)

How do these nationwide contestations affect local politics? The government has begun implementing conciliatory policies toward ethnic Khmer as it has become more cautious of them. For example, according to residents in P. village, although the government formerly fixed the date of Kathen—a Khmer festival after the end of the rainy season in which laypeople offer donations such as new robes to temples—since 2009 the authorities have allowed each temple to decide its own date on which to hold the festival.36) Also, during my stay in P. village, Communist Party members at the village, district, province, and even central government levels often visited Khmer temples during ritual ceremonies to donate large amounts of money.37)

While Vietnam’s government emphasizes that it is considerate of the Khmer ethnicity and religion in Vietnam, it remains extremely cautious, even when people try to import items for religious activities from Cambodia. A good example is the process of importing books of the Buddhist Pali canon (the representative Buddhist scripture). Theravada Buddhist temples have imported the canon from Cambodia for a long time. Even today, there is no publisher that can print the canon in the Khmer language in Vietnam, so temple representatives must go to Cambodia to buy the books themselves. However, bringing any books into Vietnam is a complicated process and requires support from a local official. Ke, who works as chairman of the Buon Temple Committee in P. village, recalled that in 2006 or 2007 he was able to buy the canon, but:

It took a total of two months to get my passport and complete the procedure. Than [a P. village policeman] offered to help me. It was necessary to get permission from the district committee to start the procedure. If we had imported the Pali canon without permission, it could have been confiscated. My wife and I hired a car and went to Phnom Penh with another policeman, the wife of the Party secretary in the village, and two monks. Although I invited Than, he was too busy with work to come with us. We bought about 100 volumes in a large building in Phnom Penh.38)

V-3 Role of Middlemen in Negotiating between the Local Community and the State

From 1998 to 2008, Than worked as a policeman and then as a member of the village people’s committee, and he has been a clerk for the Buon Temple Committee for a long time. He was a monk in the Buon Temple from 1971 to 1975 and can read and write both Khmer and Vietnamese, so he was responsible for the exchange of documents between the temple and the local government. Because he had experience dealing with affairs of public order and administration, he was the temple’s go-to person for government matters. He was also the preferred person for the temple to negotiate with the local government to obtain permission for religious activities and economic support from the secular authority. Than may be regarded as a middleman between his religious community and the secular authority.

Dealing with religious affairs was not his only work. As a sideline, he also instructed others on how to obtain official documents such as passports and ID cards. When I lived in P. village, people returning from or migrating to Cambodia would visit Than’s house and ask him to arrange the documents they needed. Than said that his monthly salary from the village people’s committee was only 830,000 đồng (USD41 in 2011) but he earned no less than USD50 per month through his side business.39)

The reason why Than could earn so much through this sideline is that the rate of literacy in the Vietnamese language among the villagers is very low, so most people are unable to deal with administrative procedures. The older people are, the more likely it is that they have received no public education.40) To avoid problems when dealing with administrative procedures, people are dependent on Than, who studied to junior high school level and is now a government official.

Than also has knowledge of administrative procedures relating to people who crossed the border without documents, such as those who moved to Cambodia from 1979 to the early 1990s without a passport or ID card and recently returned home. For example, Thu said that although she returned to Vietnam in 2010 after crossing the border without a passport in the early 1990s, she still had not received her Vietnamese citizen ID card as of 2012.41)

Although it seems that Thu does not have any intention of returning to Cambodia, many people still travel back and forth between the two countries. Because the local government exerts strict control over those crossing the border, migrants depend on middlemen such as Than who can help them obtain a passport and ID card. In fact, Than himself had been to Phnom Penh to work without possessing a passport in the early 1990s, and so he had a good understanding of the migrant’s situation.

Both the local community and the state need middlemen like Than. If the Vietnamese government attempts to excessively restrict the flow of migrants, singling out the movement of Khmer people across the border, the state’s legitimacy may be questioned, as it officially claims to be a multiethnic state that treats each of its 54 ethnicities equally. Excessive restrictions may upset residents who have historically been connected with Cambodian society though Theravada Buddhism. Mediated by local middlemen, cross-border migrants and the local government now negotiate over their belonging and identities regulated by the national border. In other words, after migrants and the state started to negotiate around the licit-ness and illegality of cross-border migrations, people had no choice but to be conscious of the border’s existence and their nationalities.


Cross-border movement has long been accepted as a licit (formally illegal but socially acceptable) (Abraham and Schendel 2005, 22–23) activity by the local community in P. village and overlooked by both the Vietnamese and Cambodian states. Even after the Vietnam War ended and North and South Vietnam were unified, state power was too limited to control the flow of refugees and migrant workers across the border, mainly because political control in the borderlands became lax with the Vietnamese occupation of Cambodia in 1979. The Đổi Mới reforms prompted the development of a market-oriented economy and a marked political transformation in both countries from the late 1980s to the mid-1990s. However, the process of rapid market liberalization coupled with the lack of social insurance policies in Vietnam also sent many peasants in Vietnam into economic difficulty. They fled life in Vietnam, traveling without a passport or documents via informal routes and across the porous border to Cambodia, where political changes and the rise of the market economy were more intense, to find work. Licit middlemen and transnational human networks facilitated their movement by preparing sea or land vehicles along these routes, which had developed and adapted since the pre-Đổi Mới era. Although it is generally assumed that the Đổi Mới reforms drastically improved people’s livelihoods, this paper suggests the contrary. Without any official state support, people who decided to migrate due to the rural economic depression in Vietnam and the political and economic transformation in Cambodia had no choice but to rely on local moneylenders, family networks, and past experiences to secure livelihoods.

The situation changed in the early 2000s. Vietnam had joined ASEAN, accelerating its return to international society and the removal of many restrictions on the entry of foreign assistance into the country in the late 1990s. As a result, it experienced rapid economic development in the 2000s. Some people in P. village began to return to Vietnam from Cambodia due to the improving social and economic situation; others circulated between the two countries, taking advantage of better economic opportunities in Cambodia and the social insurance that the Vietnamese government began to provide. However, the establishment of greater political stability and a borderless market-oriented economy made Vietnam’s government recognize the importance of ensuring control of the border.

To this day, many of those who returned from Cambodia do not possess passports or official ID cards because they crossed the Vietnam-Cambodia border during the 1980s and 1990s in an undocumented way. Local authorities in Vietnam continue their efforts to confirm the citizenship of these returnees, particularly because of the government’s concern that Khmer nationalists in Cambodia might instigate unrest among ethnic Khmer in Vietnam. Increasing amounts of information about Cambodia are reaching local communities in Vietnam through these returnees. Therefore, the Vietnamese government has recently attempted to regulate people’s movements and improve their recognition of the state’s territory by confirming citizenship and controlling the flow of information. At the same time, it is implementing insurance policies and conciliatory actions to appease ethnic Khmer.

These policies cut both ways. To deny the ethnicity and religion of the Khmer would call into question the legitimacy of the multiethnic state and produce discontent. Therefore, the government has no choice but to respect the ethnicity of Khmer in Vietnam while simultaneously denying the historical social ties between Khmer in Vietnam and Cambodia. For their part, local people are seeking out information and commodities from Cambodia and crossing the border as part of their everyday lives. The struggle between society and state reveals itself through middlemen who mitigate the concerns of both sides.

The situation in P. village reveals a dichotomy of social acceptance (licit-ness) and illegality with regard to migrants crossing the border. This dichotomy has recently become a political issue: as Vietnam’s government intensifies its control of Khmer migrants who cross the border, the contradiction of the Vietnamese state territory and the principles underlying the government’s ethnic and religious policies becomes more visible. Although the Vietnamese government has attempted to restrict the flow of migrants across the border and integrate them into the nation-state as an “ethnic minority,” excessive regulation of migration would reveal the contradiction with the state principle to equally protect ethnic minority cultures, and in turn create discontent in society. On the other hand, local people must increasingly depend on state actors in order to make strategic use of the social and economic differences between Cambodia and Vietnam while avoiding excessive state control.

Today the Vietnamese state and border crossers continue subtle negotiations over the latter’s belonging and identities, the duality of which has been socially sanctioned (licit) but is now becoming regulated by the national border, particularly through middlemen who meditate between the local government and migrants. Such negotiation was not visible during the 1980s and 1990s, when the border between the two countries was virtually porous despite being legally closed; it has become visible only since the early 2000s, when the political and economic situation of each country began to stabilize and mobility between the two countries was openly institutionalized and legalized. The changes in migratory patterns indicate that the national border, despite being mapped and officially established, is not always etched in the minds of people. This is especially true of minorities who have connections with people in the neighboring country. Rather, the national border begins to be etched through everyday and continuous interactions between state actors, who become suspicious of influence from a foreign country, and cross-border migrants, who become dependent on the state for their needs.

Accepted: August 4, 2020


I began writing this paper based on an earlier paper published in Japanese (Shimojo 2018), but thanks to the constructive criticism by anonymous reviewers, I was able to submit a completely revised version. I would like to thank the reviewers for their assistance related to previous studies and conceptualizations related to migration and borders, and editorial suggestions. In addition, I would like to thank my excellent copy editor, Jackie Imamura, who repeatedly proofread my unclear paper. All errors remain mine.


Published Sources
Abraham, Itty; and Schendel, Willem van. 2005. Introduction: The Making of Illicitness. In Illicit Flows and Criminal Things: States, Borders, and the Other Side of Globalization, edited by Willem van Schendel and Itty Abraham, pp. 1–37. Bloomington: Indiana University Press.

Amakawa Naoko 天川直子. 2004. Josho ASEAN kameika no Kambojia: Shoseido to jittai no henka 序章 ASEAN加盟下のカンボジア―諸制度と実態の変化 [Introduction Cambodia in ASEAN: Changes in institutions and situation]. In Kambojia shinjidai カンボジア新時代 [Cambodia in a new era], edited by Amakawa Naoko 天川直子, pp. 3–47. Chiba: Institute of Developing Economies, JETRO.

Ban Chỉ Đạo Tổng điều tra dân số và nhà ở Trung Ương (BCĐTĐTDSVNƠTƯ) [Central population and housing census steering committee]. 2010. Tổng Điều tra dân số và nhà ở Việt Nam năm 2009: Kết quả toàn bộ [The 2009 Vietnam population and housing census: Completed results]. Hà Nội: Nhà xuất bản Thống kê [Statistical publishing house].

Chanda, Nayan. 1986. Brother Enemy: The War after the War, A History of Indochina since the Fall of Saigon. Orlando: Harcourt Brace Jovanovich.

Chandler, David. 2008. A History of Cambodia. Fourth Edition. Boulder: Westview Press.

Elliott, David W. P. 2012. Changing Worlds: Vietnam’s Transition from Cold War to Globalization. New York: Oxford University Press.

―. 1981. The Third Indochina Conflict: Introduction. In The Third Indochina Conflict, edited by David W. P. Elliott, pp. 1–19. Boulder: Westview Press.

Evans, Grant; and Rowley, Kelvin. 1984. Red Brotherhood at War: Vietnam, Cambodia, and Laos since 1975. London and New York: Verso.

Gorman, Timothy; and Beban, Alice. 2016. Of Migrants and Middlemen: Cultivating Access and Challenging Exclusion along the Vietnam–Cambodia Border. Asia Pacific Viewpoint 57(2): 207–220.

Gottesman, Evan. 2003. Cambodia after the Khmer Rouge: Inside the Politics of Nation Building. New Haven and London: Yale University Press.

Hardy, Andrew. 2003a. State Visions, Migrant Decisions: Population Movements since the End of the Vietnam War. In Postwar Vietnam: Dynamics of a Transforming Society, edited by Hy Van Luong, pp. 107–137. Lanham: Rowman & Littlefield.

―. 2003b. Red Hills: Migrants and the State in the Highlands of Vietnam. Copenhagen: NIAS Press.

Ishikawa Noboru. 2010. Between Frontiers: Nation and Identity in a Southeast Asian Borderland. Singapore: NUS Press.

Iwai Misaki 岩井美佐紀. 2001. Shudan ka kaitai iko no Kain-hau-sha ni okeru nogyo chinrodo no jittai ni kansuru ichi kosatsu 集団化解体以降のカインハウ社における農業賃労働の実態に関する一考察(特集 20世紀メコン・デルタの開拓)[A study of agricultural wage-labor in Khanh Hau Commune since decollectivization (Special issue: Land-use development in the Mekong Delta in the twentieth century)]. Tonan Ajia kenkyu 東南アジア研究 [Southeast Asian studies] 39(1): 120–136.

Iwai Misaki 岩井美佐紀; Ono Mikiko 大野美紀子; and Ota Shoichi 大田省一. 2016. Betonamu “shin-keizai-mura” no tanjo ベトナム「新経済村」の誕生 [The birth of Vietnam’s “new economic village”]. Chiba: Kanda Gaigo Daigaku Shuppankyoku (KUIS Press).

Kimura Tetsusaburo 木村哲三郎. 1996. Betonamu: To kanryo kokka no aratana chosen (Ajia gendaishi shirizu 5) ベトナム―党官僚国家の新たな挑戦(アジア現代史シリーズ5)[Vietnam: New challenge of the party bureaucratic state (Modern and contemporary Asian history series, 5)]. Tokyo: Institute of Developing Economies.

Kitagawa Takako 北川香子. 2006. Kambojia shi saiko カンボジア史再考 [Rethinking Cambodian history]. Tokyo: Rengo Shuppan.

Lê Hương. 1970. Chợ trời biên giới Việt Nam Cao Miên [Border market between Vietnam and Cambodia]. Sài Gòn: Xuất bản Quình Lâm [Quinh Lam Publisher].

Mahanty, Sango. 2019. A Tale of Two Networks: Market Formation on the Cambodia-Vietnam Frontier. Transactions of the Institute of British Geographers 44(2): 1–16.

Murano Tsutomu 村野勉. 1993. 1992 nen kempo o kofu: 1992 nen no Betonamu 1992年憲法を公布―1992年のベトナム [Promulgating the 1992 constitution: Vietnam 1992]. Ajia doko nempo 1993 nemban アジア動向年報1993年版 [1993 yearbook of Asian affairs] edited by Ajia Keizai Kenkyujoアジア経済研究所 [Institute of Developing Economies, JETRO], pp. 191–222. Chiba: Institute of Developing Economies, JETRO.

Osada Noriyuki. 2011. An Embryonic Border: Racial Discourses and Compulsory Vaccination for Indian Immigrants at Ports in Colonial Burma, 1870–1937. Moussons: Recherche en sciences humaines sur l’Asie du Sud-Est [Monsoons: Social science research on Southeast Asia] 17: 145–164.

Schendel, Willem van. 2005. Spaces of Engagement: How Borderlands, Illegal Flows, and Territorial States Interlock. In Illicit Flows and Criminal Things: States, Borders, and the Other Side of Globalization, edited by Willem van Schendel and Itty Abraham, pp. 38–68. Bloomington: Indiana University Press.

Shimojo Hisashi 下條尚志. 2018. Betonamu-Kambojia kokkyo no ekkyo ido o meguru rokaru na seiji: Reisen shuketsu go Mekon Deruta no Kumeru jin ekkyosha to Betonamu kokka ベトナム―カンボジア国境の越境移動をめぐるローカルな政治―冷戦終結後メコンデルタのクメール人越境者とベトナム国家 [The local politics of transnational migration between Vietnam and Cambodia: Khmer border crossers and the Vietnamese state in the Mekong Delta after the end of Cold War]. Ajia Afurika gengo bunka kenkyu アジア・アフリカ言語文化研究 [Journal of Asian and African studies] 95: 152–179.

―. 2015. Datsu shokuminchi ka katei no Mekon Deruta ni okeru Kumeru jin no gengo, bukkyo, kizoku 脱植民地化過程のメコンデルタにおけるクメール人の言語・仏教・帰属 [Language, Buddhism, and belonging during decolonization: The Khmers of the Mekong Delta]. Ajia Afurika chiiki kenkyu アジア・アフリカ地域研究 [Asian and African area studies] 15(1): 20–48.

Tagliacozzo, Eric. 2005. Secret Trades, Porous Borders: Smuggling and States along a Southeast Asian Frontier, 1865–1915. New Haven and London: Yale University Press.

Taylor, Philip. 2016. Searching for a Khmer Monastic Higher Education in Post-socialist Vietnam. In Connected and Disconnected in Viet Nam: Remaking Social Relations in a Post-socialist Nation, edited by Philip Taylor, pp. 273–310. Canberra: ANU Press.

―. 2014. The Khmer Lands of Vietnam: Environment, Cosmology and Sovereignty. Singapore: NUS Press; Copenhagen: NIAS Press.

―. 2004. Redressing Disadvantage or Re-arranging Inequality? Development Interventions and Local Responses in the Mekong Delta. In Social Inequality in Vietnam and the Challenges to Reform, edited by Philip Taylor, pp. 236–269. Singapore: Institute of Southeast Asian Studies.

Thongchai Winichakul. 1994. Siam Mapped: A History of the Geo-body of a Nation. Honolulu: University of Hawai‘i Press.

Tomiyama Yasushi 冨山泰. 1993. Nanko suru wahei kyotei no jikko: 1992 nen no Kambojia 難航する和平協定の実行―1992年のカンボジア [Execution of the difficult peace agreement: Cambodia in 1992]. Ajia doko nempo 1993 nemban アジア動向年報1993年版 [1993 yearbook of Asian affairs] edited by Ajia Keizai Kenkyujo アジア経済研究所 [Institute of Developing Economies, JETRO], pp. 223–236. Chiba: Institute of Developing Economies, JETRO.

Việt Nam Cộng Hòa, Bộ Kế hoạch Phát triển Quốc gia (VNCHBKHPTQG) [The Republic of Vietnam, Ministry of State Planning and Development]. 1973. Niên giám Thống kê Việt Nam 1972 [Statistical yearbook of Vietnam, 1972]. Sài Gòn: Viện Quốc gia Thống kê [Institute of State Statistics].

Vo Tri Thanh; and Pham Hoang Ha. 2004. Vietnam’s Recent Economic Reforms and Developments: Achievements, Paradoxes, and Challenges. In Social Inequality in Vietnam and the Challenges to Reform, edited by Philip Taylor, pp. 63–89. Singapore: Institute of Southeast Asian Studies.

Vũ Đức Liêm. 2016. Vietnam at the Khmer Frontier: Boundary Politics, 1802–1847. Cross-Currents: East Asian History and Culture Review 20: 75–101.

Vũ Lân (Chủ Biên); Phương Hạnh; and Bạch Mai. 1988. Xã P, Đất nước Con người [P Village (or Commune), Country, and People]. Hậu Giang: Ban Chấp hành Đảng bộ và Ủy ban nhân dân Huyện Mỹ Tú – Hậu Giang [Committee of the communist party and people’s committee of My Tu district, Hau Giang].

Unpublished Sources
Công văn UBND Xã P [People’s committee of the P village (or commune)], Bảng Tổng hợp toàn số, khẩu [Table / Data for total population], 2011.

Đảng Cộng sản Việt Nam, Ban Chấp hành Trung ương [The central committee of the communist party of Vietnam], Số: 68CT/TW, Chỉ thị về Công tác ở Vùng Đồng bào Dân tộc Khơ-mer [No. 68 CT/TW, directive on mission in Ethnic Khmer Regions], April 18, 1991.

Thủ tướng Chính phủ [Prime Minister], Số: 74/2008/QĐ-TTg, Quyết định về Một số Chính sách Hỗ trợ giải quyết đất ở, đất sản xuất và Giải quyết việc làm cho Đồng bào dân Tộc thiếu số nghèo, đời sống khó khăn vùng Đồng bằng sống Cửu Long Giai đoạn 2008–2010 [No. 74/2008/QĐ-TTg, decision on compensation, support upon living land and manufacturing land, and employment solving for poor ethics groups in the Mekong Delta from 2008–2010], June 9, 2008.

Usuki Sayaka 薄さやか. 2013. Kambojia kingendaishi ni okeru ningen shudan bunrui gainen カンボジア近現代史における人間集団分類概念 [People-Grouping concepts in the modern history of Cambodia]. Master’s thesis, Graduate School of Asian and African Area Studies, Kyoto University.

DigitalGlobe, Earthstar Geographics | NOSTRA, Esri, HERE, Garmin, METI/NASA, USGS., accessed June 8, 2020.

Esri, GEBCO, DeLorme, NaturalVue | Esri, GEBCO, IHO-IOC GEBCO, DeLorme, NGS., accessed June 8, 2020.

1) The naming of the era is based on local naming practices; those in my field site call the era thời tập đoàn (collectivization era), rarely using the term thời bao cấp (rationing era), which is generally much better known.

2) On the migration between French Cochinchina (Southern Vietnam today) and the Protectorate of Cambodia by Khmer monks, see my paper in Japanese (Shimojo 2015).

3) These were the South Vietnamese and Cambodian governments, the United States, the National Liberation Front, and the Khmer Rouge.

4) The official number of Vietnamese citizens in Cambodia decreased from 400,000 in 1969 to 210,000 in 1970 (VNCHBKHPTQG 1973, 387).

5) In Khanh Hau village, Long An Province in the Mekong Delta, where several Japanese scholars conducted intensive fieldwork in the late 1990s, the farmland shared by the agricultural collectivization was returned to the former owners by Đổi Mới reforms, resulting in large numbers of landless households. Among them, not only the original landless households but also some others became landless because the process of land inheritance between generations did not catch up with the rapid population growth over 40 years (Iwai 2001, 121).

6) When Buddhist and secular educational institutions were revived and restructured in the 1990s in Cambodia, many Khmer monks from Vietnam resumed undocumented travel to Cambodia for study, but the Vietnamese government attempted to restrict them from going (Taylor 2016).

7) Although is often translated as “commune,” this paper translates as village, because the people in P. village do not distinguish between the terms làng (natural village) and (the smallest administrative unit), treating both as the same.

8) The people in P. village use the term “Viet” rather than “Kinh.” Based on the local context, I use “Viet” in this paper.

9) Công văn UBND Xã P, Bảng Tổng hợp toàn số, khẩu, 2011.

10) Samrong ward, which is located in Q. hamlet (ấp) of P. village, is the focus of this study. Although it is not an official place name, in my fieldwork I used this name for convenience of explanation. Q. hamlet is officially separated into three wards. Samrong ward encompasses most of ward 2 and part of ward 3. At the time of my household survey (December 2011–January 2012), Samrong ward had 663 residents (713 including people traveling to work outside of the ward). I lived in an area known in the Khmer language as “Phno (dunes) located on the fringe of the Samrong trees.” Thus, I named the study area “Samrong ward.”

11) Interview with Com (born 1950), male, Chinese-métis Khmer, farmer, living in P. village, March 6, 2012.

12) Interview with Sang (born 1936), female, Khmer, farmer, living in P. village, January 12, 2012.

13) Interview with Rat (born 1948), male, Khmer, farmer, living in P. village, February 28, 2012.

14) Interview with Han (born 1951), male, Chinese-métis Khmer, farmer, living in P. village, February 22, 2012.

15) At that time, Hang had enlarged his farmland by claiming the land of another debtor who was unable to repay his debt. Interview with Than (born 1955), male, Khmer-métis Chinese, member of P. village people’s committee, living in P. village, February 22, 2012.

16) Interview with Than, February 22, 2012.

17) Interview with Than, March 1, 2012.

18) Interview with Don (born 1953), male, Khmer, carpenter, living in P. village, September 9, 2011.

19) Interview with Than, February 22, 2012.

20) The exchange rate at the end of December 1992 was USD1 = 2,310 riel (Tomiyama 1993).

21) The exchange rate at the end of December 1992 was USD1 = 10,505 đồng (Murano 1993).

22) Interview with Don’s wife (born 1954), female, Chinese-métis Khmer, homemaker, living in P. village, February 29, 2012.

23) Interview with Don’s wife, February 29, 2012.

24) Interview with Don, March 4, 2012.

25) Interview with Con (born 1973), male, Khmer, guard at a junior high school, living in P. village, March 4, 2012.

26) The people whom I have interviewed so far mentioned Tinh Bien (a vast floodplain, see Fig. 2), Ha Tien (a port city on the Gulf of Siam), and Long Binh (a city along the Bassac and Binh Di Rivers) as popular informal border crossing areas during the 1980s and 1990s. These areas today have official border gates through which almost all people can cross with a passport.

27) In 1970, when the Cambodian civil war broke out, the number of “Vietnamese citizens” in Cambodia decreased sharply from 400,000 to 210,000 (VNCHBKHPTQG 1973, 387). The number of “Vietnamese” was just 8,200 in 1981, when the Vietnamese military invaded, but had increased to 95,600 by 1995 (Usuki 2013, 26–28). It is unclear whether “Vietnamese” in some of the surveys conducted at that time referred to ethnic “Viets” or “Vietnamese citizens,” including ethnic Khmer born in Vietnam. Furthermore, it is unclear whether people who migrated from Vietnam without possessing passports were counted in these statistics. However, what is certain is that the number of people classified as “Vietnamese” was increasing.

28) Interview with Thu (born 1947), female, Chinese-métis Khmer, seller of cheap candy, living in P. village, February 24 and March 3, 2012. In 2010 Thu moved back to P. village, where she set up a small business selling candy.

29) Công is the local unit for measuring land; 1 công is about 0.13 ha.

30) Interview with Son (born 1952), male, Khmer (his father was born in Cambodia), wage laborer, living in P. village and Ratanakiri, August 11, 2011.

31) Thủ tướng Chính phủ, Số: 74/2008/QĐ-TTg, Quyết định về Một số Chính sách Hỗ trợ giải quyết đất ở, đất sản xuất và Giải quyết việc làm cho Đồng bào dân Tộc thiếu số nghèo, đời sống khó khăn vùng Đồng bằng sống Cửu Long Giai đoạn 2008–2010, June 9, 2008.

32) Interviews with Than, January 13, 2011 and 7 March 7, 2012; Son, August 11, 2011; Cang (born 1950), male, Khmer, traditional medical practitioner, living in P. village, August 18, 2011.

33) Đảng Cộng sản Việt Nam, Ban Chấp hành Trung ương, Số: 68CT/TW, Chỉ thị về Công tác ở Vùng Đồng bào Dân tộc Khơ-mer, April 18, 1991.

34) Đảng Cộng sản Việt Nam, Ban Chấp hành Trung ương, Số: 68CT/TW, Chỉ thị về Công tác ở Vùng Đồng bào Dân tộc Khơ-mer, April 18, 1991.

35) Interview conducted on September 7, 2011. I have anonymized the interviewee in order to protect the interviewee from political persecution.

36) Interview with Pho (born 1971), male, Chinese-métis Khmer, farmer, living in P. village, October 16, 2011.

37) Author’s field notes, September 26–27, 2011.

38) Interview with Ke (born 1940), male, Khmer, farmer, March 6, 2012. It can be assumed that Ke bought the Pali canon at the Buddhist Institute in Phnom Penh.

39) Interview with Than, February 8, 2012.

40) In Samrong ward, P. village, of the 419 people over 20 years old, 89 had never studied at a public school, 172 had studied in a primary school, and 111 had studied in a secondary school. Thus, the majority had never studied in a public school or had studied only up to primary school level (based on the author’s survey conducted from November 2011 to January 2012).

41) Interview with Thu, March 3, 2012.


Vol. 6, No. 3, Will BREHM


Contents>> Vol. 6, No. 3

The Is and the Ought of Knowing: Ontological Observations on Shadow Education Research in Cambodia

Will Brehm*

* Waseda Institute for Advanced Study, Waseda University, 1-6-1 Nishi Waseda, Shinjuku-ku, Tokyo 169-8050, Japan
e-mail: willbrehm[at]

DOI: 10.20495/seas.6.3_485

This article focuses on the limitations of terms and definitions regarding shadow education research in Cambodia. Although shadow education in Cambodia is typically defined as private tutoring taught by mainstream schoolteachers to their own students, other manifestations of it have been missed by most studies on the subject, including my own. By tracing the terms used and the definitions of shadow education in various research studies, I argue that the assumptions made over terms and definitions (i.e., what ought to be the case) limited researchers’ understanding of shadow education in its ontological evolution and complexity (i.e., what is the case). Methodologically, the unintentional recycling of the same definition across time resulted in the epistemic fallacy and concept reification. These outcomes have profound consequences for how the phenomenon may be theorized not only in Cambodia but across the Southeast Asian region. In conclusion, I propose an alternative approach to study shadow education based on critical realism.

Keywords: shadow education, private tutoring, Cambodia, critical realism, methodology


There is a long-standing Western philosophical problem in using descriptive statements (what is) to make prescriptive claims (what ought to be). But do claims of “what ought to be” limit “what is”? This can happen when erroneous assumptions proliferate. Since all research begins with assumptions, it is vital not to be mistaken.

Some of the most common assumptions in research relate to terminology and definitions. The terms and definitions used by researchers help to manage concepts that are difficult to comprehend. To manage a concept so that it can be studied, the terms and definitions employed in research studies necessarily exclude alternative meanings. As such, assigning terms and settling on one definition over another for a given concept is never a neutral process. This struggle over meaning is a central feature of academic debate.

Research on “shadow education” is a case in point.1) Shadow education can be broadly defined as a collection of educational services that are fee based but not public, mainstream schooling.2) The formation and organization of the phenomenon differ across the globe. Within Southeast Asia, the differences are pronounced. At one extreme are Cambodia, Brunei Darussalam, and Laos, where tutoring is commonly initiated by schoolteachers to top up (sometimes substantially) low salaries. In these cases, it is difficult to know when mainstream schooling ends and private tutoring begins. At the other extreme are Singapore, Thailand, and the Philippines, where tutoring has developed into a legitimate and recognized business sector. Students in these countries typically take extra lessons in centers that are organized as for-profit companies, outside the control of education ministries but connected to school curricula and examinations. In Singapore, for instance, 8 out of 10 primary school children attend tutoring (Straits Times-Nexus Link Tuition Survey 2015), and the amount households pay on tutoring increased from S$650 in 2004 to S$1.1 billion in 2014 (Tan 2014).

Across the globe people have their own, evolving terms to describe the activities researchers commonly refer to as shadow education. In Japan the dominant form is termed juku, in England it is called tuition, and in Cambodia it is labeled ɾiən kuə.3) Yet each of these terms misses the complexity of the phenomenon as it is currently understood. There are juku for examination preparation and juku for remedial study (Roesgaard 2006; Watanabe 2013). There are tuition classes in cyberspace and in everyday life (Ventura and Jang 2010). Both public and private school teachers can teach ɾiən kuə classes. As educational spaces evolve and morph into new realities, and as researchers’ understandings deepen, researchers try to refine and add complexity to their terms and definitions.

Yet decisions over the definition of a phenomenon like shadow education often have unintended consequences. One possible consequence of choosing one meaning over another is the assumption that it accurately captures the phenomenon’s existence, its ontology. Another consequence is that operationalized terms and definitions can be normalized and therefore legitimized by future research studies, thus missing possible changes to the phenomenon itself. In these situations, what was excluded from or simply not captured by the definition and term lead to significant gaps in understanding.

This article focuses on the limitations of terms and definitions regarding shadow education research in Cambodia. Although shadow education in Cambodia is typically defined as private tutoring taught by mainstream schoolteachers to their own students (captured by the term ɾiən kuə), my experience suggests that many types of ɾiən kuə comprising the activities of shadow education have been missed by most studies on the subject, including my own.

The missing terms and definitions in the research literature raise a methodological question: Do terms and definitions used in research studies capture, intentionally or not, only part of the multifaceted phenomenon? By tracing the terms used and the definitions of shadow education in various research studies, I argue that the assumptions made over terms and definitions (i.e., what ought to be the case) limited researchers’ understanding of shadow education in its ontological evolution and complexity (i.e., what is the case). This has profound consequences for how the phenomenon may be theorized. I advocate a critical realist approach to the study of shadow education not only in Cambodia but across Southeast Asia to acknowledge the existence of its reality, whether or not researchers can adequately see, name, or define it.

Changing Terms, Static Definitions

In Cambodia people use the term ɾiən kuə to describe what researchers would call shadow education. Like the term juku in Japan, however, ɾiən kuə can embrace multiple types, which are likely evolving and thus should not be taken as static (see Table 1).


Table 1 Types of Tutoring


The most common type of ɾiən kuə is “regular private tutoring,” which is fee-based tutoring in classes taught by mainstream schoolteachers. It is considered “regular” (tʰoəmməɗaː) because it focuses on the mainstream curriculum and resembles mainstream classes (i.e., class sizes and layouts are like those in mainstream schooling). A less common form of ɾiən kuə is “special private tutoring,” which covers individual or small group classes taught by a tutor who may or may not be a student’s mainstream schoolteacher. These classes cost much more than regular private tutoring classes. Some students have the option of attending and paying for “private tutoring during holidays.” These are classes conducted in school or at a teacher’s home, and are held by a student’s current or future teacher when mainstream schooling is not in session. The last type of ɾiən kuə, which appears to be a growing phenomenon especially in city centers, is “private tutoring at private school.” This type of ɾiən kuə covers tutoring classes of various sorts, held by non-mainstream schoolteachers outside public school buildings, and for some cost. The word “school” in this type of tutoring takes on a broad meaning, from registered tutoring centers as businesses to makeshift classrooms inside university students’ homes or apartments. Each type of ɾiən kuə has different causal origins, and future evolutions will likely make this artificial categorization obsolete.4) Nevertheless, this brief orientation of contemporary ɾiən kuə will be useful to the reader going forward.

The transliterated terms I provide above have rarely been used in the English-language research literature. Instead, terms such as private tuition, private coaching, or private tutoring have been used. Quite apart from the loss of meaning when one of these terms is translated into Khmer, it is the evolution of English terminology that interests me here. In this section I trace both the terminology and the definition of the phenomenon referred to broadly as ɾiən kuə. I show that the English terminology has changed within and across research studies reported in the English language, while the definition has stayed roughly the same.

One of the first mentions in Cambodia of “private tuition” was in a 1994 Education Sector Review (Cambodia 1994). The Review’s executive summary stated, “recent surveys suggest that parents pay around R120,000 per annum per primary student for uniforms, private tuition and books” (ibid., Vol. 1, 14; emphasis added).5) This indicates that the researchers who conducted the cited surveys included “private tuition” as a category of possible household expenses.

Closer inspection of the data within the Review reveals that different terms were used to describe the phenomenon in the two surveys to which the Review referred, namely, “private tuition,” “private coaching,” and “private tutoring.” However, both surveys defined the concept in similar ways. In one table, family costs of education per pupil were reported in the main urban centers in six provinces for primary schools and three provinces for secondary schools. The costs were categorized into textbooks and materials, uniforms, contributions to school, transport, and private coaching (ibid., Vol. 2B, Table 75; see Fig. 1). In another table, household expenditures per student were reported from a sample of 126 students and were broken down into different categories: tuition and other charges, books and stationery, private tuition, uniforms, transport, and others (ibid., Vol. 2B, Table 76; see Fig. 2).


Fig. 1 Table 75 of 1994 Education Sector Review, Vol. 2B



Fig. 2 Table 76 of 1994 Education Sector Review, Vol. 2B


Although data reported in the two tables came from different sources, which is likely why there was a slight difference in terminology,6) the terms refer to the same phenomenon. A definition of the interchangeable terms can be ascertained in the main body of the Review (ibid., Vol. 2A, 109). In this description, “private tutoring is not, as one might assume, an opportunity for individual students to get special help on material they might not have understood in class. Instead, it constitutes an extension of the regular curriculum offered by the same teacher in the same large group setting—this time with a user fee attached.” Moreover, the report labeled private tutoring as “part [of] the shadow private system” of education (ibid.). This description was recycled almost verbatim by subsequent Reviews (e.g., Cambodia 1996) as well as in later studies by Mark Bray (e.g., Bray 1996a; 1999), the scholar who has propelled research on shadow education worldwide and who was my PhD adviser. As I will show below, this description has remained the central definition of the phenomenon in the Cambodian context.7)

Reproducing a case study from the Review, Bray (1996a, 16) used the term “private tutoring” to discuss the phenomenon in Cambodia.8) He (1999) also used the terms “supplementary tutoring” (57), “private tutoring” (22), and “private supplementary tutoring” (90) to describe the “shadowy system considered beyond the control and responsibility of government” (90) in Cambodia. Bray (ibid.) pointed out that in the Cambodian context, “much of the tutoring is in the students’ own schools and is given by their own teachers” (21).9) Although Bray (ibid., 57) acknowledged some “pupils made private arrangements for additional tutoring outside the schools,” tutoring was categorized as an in-school expense. The English terms used by Bray and the authors of the Review in Cambodia include “private tuition,” “private tutoring,” “private coaching,” and “private supplementary tutoring.”

Although different combinations of terminology were used from 1994 to 1999, the description of the phenomenon remained relatively constant. Bray (1999) reproduced the description of “supplementary tutoring” in Cambodia in a highlighted box titled “private enterprise in a public system” (22). In this box, which came from the 1996 Education Sector Review (Cambodia 1996, 107), “private tutoring” was described as “an extension of the regular curriculum offered by the same teacher in the same large group setting—this time with a user fee attached.” This is identical language to the 1994 Review cited by Bray (1996a, 16): private tutoring “constitutes an extension of the regular curriculum offered by the same teacher in the same large group setting.” Despite the variable terminology in all the reports mentioned thus far, the descriptions of the actual concept remained nearly identical. The terminology and description used in the Review (Cambodia 1994) were not only reproduced by Bray’s (1996a; 1999) two studies but also were repeated and reused by various authors over the next 15 years.

Before looking at some of these studies, it is important to situate the evolution of terms in their historical context. The historical beginnings of the various terms can be traced, in part, to a separate study by Bray (1996b) for UNESCO’s International Commission on Education for the Twenty-first Century in which he discussed “a general shift in the centre of gravity towards greater private ownership, financing and control of schools” (i).10) Although this report was not about Cambodia per se, it did show Bray’s own process of coming to understand the concept of shadow education and the various terms (and metaphors) that could be used to label it. It also suggests that the concept of shadow education was implicated in the school privatization processes that became popularized within various development organizations and international financial institutions during the 1990s.11)

Bray continued his Cambodian research with another study a few years later. Bray and Bunly (2005) built on Bray’s (1999) study to focus on household costs at the primary and lower secondary levels. The latter grades were unexplored in the earlier study. In the 2005 iteration, supplementary tutoring was described as in earlier studies, and the terminology was again multiple. Bray and Bunly (2005) used “supplementary tutoring” (11), “private tutoring” (75), and the “shadow system” which operates “alongside the mainstream” (40). The description of these various terms remained nearly identical to those in 1994, 1996, and 1999: “in Cambodia, much of the tutoring is in the students’ own schools and is given by their own teachers” (ibid., 11).

The relative stability in the description of “private tuition,” “private tutoring,” “private coaching,” and “private supplementary tutoring” from the 1994 Review to Bray and Bunly’s (2005) study, which may be an outcome of the relatively short time frame and similar authors across the studies, was normalized and legitimized by later studies. In its 2007 report on informal fees to education, the NGO Education Partnership (NEP) wrote of “private tutoring” with the occasional use of “extra tutoring” (NEP 2007, 17, 26). Similar but not identical to Bray and Bunly’s (2005) formulation, the NEP classified tutoring into two types: teachers who “conduct private classes” do so either (1) “on the school premises” and therefore for their own students; or (2) “in private classrooms set up in the community” and therefore open to all students (16). The report went on to state that private tutoring was “often a continuation of the public curriculum rather than supplementary” (ibid., 16), thus disputing—but not elaborating on—one of the key terms used in previous studies. There may have been small revisions and challenges to the terminology used to describe the phenomenon, but the description of “private tutoring” in the NEP report was similar—if not identical—to the 1994 Education Sector Review.

Walter Dawson (2009) was the first to problematize explicitly the terminology used in shadow education research in Cambodia. The data collected by Dawson in 2008 set out to “re-examine the findings” of earlier studies on shadow education in Cambodia (ibid., 55). He preferred to use the terms “private tutoring” and “shadow education,” and critiqued some of the other terms used in earlier studies. When citing data from a government report from 2005, Dawson noted (ibid., 57) its problematic use of “remedial tutoring” as a category of unofficial fees. He went on to explain how tutoring in Cambodia is neither supplementary (because much of it completes the national curriculum), echoing the NEP (2007) study, nor remedial (because high-achieving students attend just as often as low-achieving students).

Despite his critique of the different terms used to describe shadow education in Cambodia, Dawson nevertheless employed the same description as the 1994 Education Sector Review: “This form of shadow education wherein state teachers conduct private tutoring for their own students is well documented by Bray and not unusual to find in many developing countries . . .” (Dawson 2009, 51).

The description used by Dawson (2009) included the phrase “This form of” without exploring alternative forms. In a later article comparing shadow education in Japan, Korea, and Cambodia, Dawson (2010) again suggested that there were multiple forms of tutoring. In the section on Cambodia (ibid., 20), he qualified the term “private tutoring” with the phrase “this brand of,” like the phrase “much of” used by Bray (1999, 21) and Bray and Bunly (2005, 40). Since Dawson did not explain other “brands of” tutoring within the Cambodian context, I read this phrase as drawing a comparison to the tutoring practices in the other two countries. What he did not do, in other words, was suggest there were different “brands of private tutoring” within Cambodia. This is particularly surprising given that Dawson (2010) discussed the many types of juku in the section on Japan (16).

A 2011 study by William Brehm, Iveta Silova, and Mono Tuot (2012; also, Brehm and Silova 2014) continued the trend of challenging terminology while describing the phenomenon as teachers who tutor their own students. Brehm and Silova (2014) described private tutoring thus: “Before or after attending the required four or five hours of public school each day, many students receive, and pay for, extra instruction [by their own teacher]” (95). Brehm et al. (2012) offered the term “hybrid education” in their discussion on “shadow education” and “supplementary tutoring” (14–16). This concept was defined simply as public, mainstream education plus “complementary tutoring,” which was defined as the type of tutoring where teachers tutor their own students. Although they preferred the term “complementary tutoring” to “supplementary tutoring” because, echoing Dawson’s work, the former includes “lessons that are essential [and not extra] to the national curriculum” (ibid., 15), they continued to use the common description of tutoring since 1994 that focused on teachers who “conduct private tutoring lessons with their own students after school hours either in school buildings or in their home” (ibid., 16). Moreover, the authors argued that the hybrid system “casts a shadow of its own” (ibid., 15), meaning other forms of tutoring (e.g., “remedial and/or enrichment education opportunities” [ibid.]) existed because of this hybrid system. They highlighted the different types of tutoring in a table (ibid., 16). The private tutoring commonly referred to in past studies was labeled “extra study” (with an incorrect Latin-script rendering of the Khmer script as rien kuo). They then offered other types of tutoring (and their English translations), such as “extra study during holidays,” “extra special study” (i.e., individualized tutoring), “private (tutoring) school,” and “English/French extra study.” Each was a different type of tutoring conceptualized into two broad categories: “hybrid education” and “shadow education.”

Despite the expansion in their description of hybrid education and its shadow, Brehm et al. (2012) limited their study to “the differences and similarities between private tutoring (Rien Kuo) and government school classes” (17). In other words, they continued to study ɾiən kuə exactly as it had been historically described in the Cambodian context, neglecting its other forms despite recognizing their existence. Although they questioned the terminology used in shadow education research in Cambodia just as the NEP (2007) and Dawson (2009) had, Brehm and colleagues focused on one type of the phenomenon when collecting data.

This historical look at past research studies of ɾiən kuə in Cambodia shows two things. First, the terminology used to describe the phenomenon has changed greatly over the years. The terms private coaching, private tutoring, private tuition, supplementary tutoring, shadow education, extra study, etc., have all been used. Second, the description of these various terms has stayed relatively similar over time. That description is of tutoring given by schoolteachers to their own mainstream school students. Although the different authors recognize other forms of tutoring, rarely are they elaborated. Because of the similar descriptions of the phenomenon, all the research studies have used a similar core definition when collecting data, limiting what is ɾiən kuə to what it ought to be.

The Epistemic Fallacy and Concept Reification

Despite the changing terms, the similar descriptions of ɾiən kuə employed in the various research studies reduced descriptive claims of what is to prescriptive claims of what ought to be. The alternative realities were, in other words, reduced to the definitions employed in data collection methods. Methodologically, the unintentional recycling of the same definition across time resulted in the epistemic fallacy and concept reification.

The epistemic fallacy (Bhaskar 1975) is the confusion over how researchers know things with whether those things exist. The quintessential epistemic fallacy is perhaps best captured by Descartes’ (1637/1960) famous saying, “I think; therefore, I am.” In this example, it is implied that the ability of a subject to think about itself constitutes the self in reality. In effect, Descartes’ existence depends on his ability to think, thus “reduc[ing] reality to [his] knowledge of it” (Dean et al. 2005, 8). This is a fallacy because Descartes’ physical self exists whether or not he can actually think about it. In terms of Western philosophy, which underpinned all the studies discussed in the previous section, research that commits the epistemic fallacy assumes that epistemology comes before ontology. This is analogous to believing that ɾiən kuə exists only in the form that has been empirically captured by researchers.

The second problem of concept reification is the process of taking an abstract concept and turning it into a concrete reality. Shadow education is an abstract concept because the manifestations of its material reality—juku in Japan, tuition in England, or ɾiən kuə in Cambodia—are different depending on space, place, and time. Moreover, the material realities of juku, tuition, or ɾiən kuə are ever changing and therefore require constant revision to terms, descriptions, and definitions. Yet, through the research process where concepts are clearly defined and then operationalized in data collection instruments, the concept of shadow education necessarily goes from being an abstract concept to being a real thing that can be measured and described. The main problems with concept reification are that reified concepts may incorrectly or only partially capture material reality, and the reuse of the same reified concept in later studies decontextualizes the phenomenon from its material reality in specific spaces, places, and times.

To show these two problems in the research, it is necessary to look closely at the methods employed in the various studies on shadow education in Cambodia. What becomes clear across the studies is that the preferred method of data collection has been the survey, often supplemented with interviews and focus groups. It is within the surveys that the constant definition of ɾiən kuə is used and reused. It is precisely here where concept reification and the epistemic fallacy emerge.

Survey research is the quintessential data collection method that reifies concepts. Surveys must operationalize terms—that is, the process of measuring a concept that is not directly measurable—for data to be collected. In survey research, questions are asked to obtain empirical, measurable data that are said to define (often by proxy) an abstract concept. For example, measuring the amount of money students pay teachers for tutoring classes can operationalize the concept of private tutoring. Another possibility for operationalizing private tutoring is to measure the attendance of students in tutoring classes. Still a third way is to simply ask students, parents, or teachers whether private tutoring exists. In these cases, operationalizing essentially takes a concept and reifies it; it assumes one definition and therefore not another, and subsequently operationalizes the assumed definition by asking one set of questions and not another.

Operationalizing terms, however, is a necessary part of survey research. Bray and Bunly (2005, 28) rightly point this out: “Surveys need to set clear definitions and then to communicate those definitions to all relevant people.” A necessary consequence of setting clear definitions is the exclusion of other possible definitions. For example, defining private tutoring as fee-based classes taught by mainstream schoolteachers and then asking students about that may provide descriptive information on this topic, but it certainly will not provide descriptive information on the classes for which students pay (or not) that are taught by teachers other than their own mainstream schoolteachers. As such, to assume ɾiən kuə is captured completely by a set of survey questions reduces the reality of its existence to the knowledge produced by the survey, thus committing the epistemic fallacy.

The 1994 Education Sector Review reported data on private tutoring and private coaching from two different surveys (Cambodia 1994). The Ministry of Education, Youth and Sports and the development mission in charge of writing the Review conducted the two surveys. In effect, the surveys captured what the people constructing the surveys knew at one moment in time. From the two tables where private tuition/coaching are reported, it can be inferred that the surveys operationalized household expenditures into various categories. Private tuition/coaching was one such category. When the survey was carried out, respondents could respond to questions about money spent on private tuition/coaching. As such, private tutoring was operationalized by the amount of money respondents reportedly spent on private tuition/coaching.

The 1994 Review’s definition of private tutoring/coaching was used and reused in later studies. Although it is possible that surveys used in other studies asked questions about different types of tutoring, all the studies reported data on one (or possibly two in the case of Bray and Bunly 2005) type(s) of tutoring. In effect, the phenomenon was reduced—or flattened—to one understanding. The 1994 description of private tutoring, which captured one moment in time, became trans-historical as it was applied in and reported by subsequent studies. Consequently, what existed was what was seen, thus committing the epistemic fallacy. Alternative definitions of private tutoring were excluded even if other types of tutoring were alluded to.

Some of the studies used mixed methodologies to collect data. These studies can mainly be categorized as sequential explanatory design mixed methods (Creswell et al. 2003). In these types of studies, quantitative data are typically collected before qualitative data:

The rationale for this approach is that the quantitative data and their subsequent analysis provide a general understanding of the research problem. The qualitative data and their analysis refine and explain those statistical results by exploring participants’ views in more depth. (Ivankova et al. 2006, 5)

Sequential explanatory design mixed method empirical studies collect data through a survey and then explore that data in greater depth vis-à-vis public opinion interviews or focus groups. The latter provide qualitative details to the former descriptive statistics, and not vice versa.

Sequential explanatory mixed method studies have been the favored approach in shadow education research in Cambodia. Data in Bray’s (1999) study, conducted in two iterative phases in 1997 and 1998, “were collected through questionnaires and follow-up discussion with personnel from nine schools in each location,” which were based on a previous study in Bhutan (37). Discussion workshops were also organized with parents after questionnaires were administered. The notes from the workshop discussions, which were translated into English, were used “to supplement the data contained in the questionnaires” (ibid., 37). In phase two of the study, the questionnaire was revised based on the first phase of data collection and administered in the same manner. In addition, four case studies were conducted in the second phase. Bray and Bunly (2005) used a similar method to Bray (1999): school surveys followed by focus group discussions, followed by “in depth interviews with pupils for information validation” (Bray and Bunly 2005, 32). Similarly, the NEP’s (2007) study used a structured questionnaire followed by focus group discussions. The latter “provided more qualitative information about informal payments and explored in more depth public opinion and perception” (ibid., 9). In Dawson’s (2009) study, a sample of “primary school teachers . . . completed a written questionnaire after which they participated in a 60–90 minute focus group interview” (57). In each case, quantitative data collection preceded qualitative data collection. The one exception is the study by Brehm and colleagues (Brehm et al. 2012; Brehm and Silova 2014) where qualitative data (focus groups and observations) were conducted concurrently with quantitative data collection (grade tracking), while no survey was carried out. Nevertheless, problems remain in Brehm and colleagues’ work where the definition of private tutoring was assumed without question. In effect, the authors were only looking for one type of tutoring without realizing that other types might have existed.

In all the studies, alternative definitions to ɾiən kuə were excluded through the very research methods employed. The sequential explanatory design mixed methods used by most of the studies limited the definition of shadow education to one or two types, which were operationalized by questions in the surveys. When data were reported, the research studies privileged one definition of private tutoring to the exclusion of possible alternatives, even if the researchers themselves knew other types of tutoring existed. The abstract concept of private tutoring was therefore reified in the research literature to one specific type, with only slight variations over time.

Qualitative research in combination with survey research has the potential to overcome some of the inherent problems of survey research, namely, the impossibility of managing context (Burawoy 1998). That the definition of shadow education employed in data collection methods stayed relatively consistent over two decades of research suggests, however, that the qualitative side to the various studies never truly informed the surveys, at least in terms of the definition of the central concept under investigation. It is this methodological shortcoming that has reduced the many meanings of private tutoring to one definition, thus blinkering researchers from employing alternative definitions to capture other facets of the phenomenon within the Cambodian context. In this way, all the research studies committed the epistemic fallacy because they assumed reality was what could be seen and measured through surveys.

This is not to suggest that surveys should not be used in research. Surveys have real value due to their ability to describe certain concepts at one moment in time. However, without historical understandings of the sociocultural structures informing the construction of surveys, researchers are prone to commit the epistemic fallacy and reify concepts that may be fleeting, elusive, and evolving. An alternative starting point assumes reality is more than researchers can empirically observe. It is to this alternative that I now turn in the conclusion.


Whatever terms and definitions are settled upon dictate how researchers see and know the world, knowingly or not. Words and their meanings are the building blocks for theory, or what Western philosophers call epistemology. Moreover—and perhaps harder to grasp—the assumptions made over terms and definitions presuppose a general account of the world, or what Western philosophers call ontology. Terms and definitions not only help social scientists see the world by giving meaning but also help construct the world.

The history of shadow education studies in Cambodia highlights the dangers of assuming and operationalizing definitions. By limiting the definition of shadow education reported in various studies, researchers likely missed myriad experiences students had with tutoring. This was evident in Brehm et al.’s (2012) table of the different types of tutoring, most of which were different from the common definition of tutoring dating from the Review’s 1994 definition (Cambodia 1994). Although there were changes in terminology used to describe shadow education, such as the experiences in Japan and England, the surveys conducted across the studies in Cambodia did not allow for alternative realities to exist. As I attempted to show, it was not only survey research that limited the definitions but also more qualitative-oriented studies (e.g., Brehm and Silova 2014). By operationalizing one (or two) definition(s) of private tutoring into the various surveys, reality was flattened to only what was seen at one moment in time.

There is an alternative research paradigm that conceptualizes reality as stratified. Critical realism begins with the assumption that reality is more than what can be empirically seen. Reality is not limited to experiences but also includes sociocultural structures that do not have a material reality but nevertheless affect human agency through emergent properties. From this critical realist perspective, reality is stratified and not flat. As such, critical realism differentiates reality into three ontological levels.

The first level is the empirical. This is what researchers observe in daily life. It is precisely at this ontological level where the surveys employed in shadow education research in Cambodia exist. The various surveys could capture an empirical reality of a sample of individuals within a specific moment in time. For example, the surveys captured descriptive statistics such as the percentage of students attending one type of private tutoring, the typical cost of one hour of tutoring, the subjects commonly taught during private tutoring, and parental and teacher perspectives on why the classes were held. Critical realists argue this level of reality is true, but that there are likely other empirical realities from other people at the same moment (or different moments) in time that are also true but simply not captured by the research study. This is where the second ontological level exists.

The second level is the actual. This is the “sum total of events that can be said to have taken place” (Graeber 2001, 52). Although all experiences within the actual may not have been observed by a single actor, it is conceivable to accept the premise that experiences other than one’s own could in fact have occurred and could have been observed given different spatiotemporal configurations. For example, students may have attended private tutoring classes taught by teachers other than their own prior to 2005 when Bray and Bunly (2005) first reported data on this type of tutoring. Bray (1999) implied this in his use of the phrase “much of.” This implies that the ways in which researchers know are relative and socially produced: each person experiences different empirical realities, which then change how he or she knows something to be “true.” The level of the actual suggests that ontology exists whether researchers understand reality or not. As such, critical realists argue epistemology does not precede ontology but rather succeeds it.

The third ontological level is the real. This is the level of powers, mechanisms, and potentialities of what may or may not happen and which are irreducible to (patterns of) events. Whereas the levels of the empirical and actual are concerned with “events, states of affairs, experiences, impressions, and discourses,” the real is concerned with “underlying structures, power, and tendencies that exist, whether or not detected or known through experience and/or discourse” (Patomaki and Wight 2000, 223). It is in the level of the real where “a sense of reaching for deeper” explanations of the world appear through “the latent or invisible . . . forces that manifest themselves in everyday life” (Coole 2005, 124). It is at this level that a different conception of causation emerges. Causation is not a correlation between two or more empirical occurrences, but rather an understanding of the historical mechanisms and structures that make what exists possible. This requires more than empirical data that can describe the empirical and actual levels of reality. At the level of the real, social scientists “attempt to identify the relatively enduring structures, powers, and tendencies, and to understand their characteristic ways of acting” (Patomaki and Wight 2000, 223). These sociocultural structures are context specific and based on history.12)

To understand shadow education in Cambodia from a critical realist perspective therefore requires researchers to see it as a system with its own emergent properties and potentials that are irreducible to its constituent parts. Shadow education from this perspective is a social reality created through the interactions of people (students, teachers, parents, government officials, etc.) that embrace or transform (through reflexivity) certain vested interests, opportunity costs, and situational logics that are embedded in social structures and cultural systems (see Archer 2003).

A stratified ontology offers an alternative set of assumptions that can be usefully employed in shadow education research in Cambodia. First, a critical realist approach suggests survey research can inform understandings of empirical reality at certain moments in time, but its explanatory power of the phenomenon is limited. What causes shadow education, therefore, cannot be explained through survey research alone. In addition, placing surveys within the first ontological level of the empirical prevents trans-historicizing data and definitions. Researchers who take a critical realist approach should question definitions used and operationalized in previous empirical studies because the space, place, and time of definitions and terms must be recognized.

Second, qualitative research takes on a different purpose than studies that use sequential explanatory design mixed methods. Whereas the sequential explanatory design mixed methods approach places qualitative data collection after or in iteration with quantitative data to provide more depth to the statistical data, a critical realist approach would use qualitative data not only to inform the collection of empirical data through surveys but also to understand the ontological levels of the actual and the real. Regarding the latter, qualitative research keeps open the possibility of multiple empirical realities (i.e., the actual) without artificially limiting reality to one meaning as is necessary in survey research. Thus, during unstructured interviews, for example, an infinite number of definitions of shadow education could theoretically emerge from participants because they are not limited by a clearly communicated definition made prior to data collection by the researchers. This likely occurred in all of the research studies when the researchers first learned about the phenomenon, not through published articles but through interactions with their colleagues on the ground.

Third, a critical realist approach to shadow education research would incorporate theory differently than has previously been the case. Whereas theory has often been used to help make sense of empirical data collected, a critical realist approach uses theory to understand the ontological level of the real while acknowledging the social construction of theory itself. This is because understanding the real, which is where the causal mechanisms of shadow education are assumed to reside, requires an engagement with various types of theory. Since “widely different theories can interpret the same, unchanging world in radically differently ways,” it is necessary for critical realist researchers to recognize that “knowledge is not totally arbitrary and some claims about the nature of this reality may provide better accounts than others” (Patomaki and Wight 2000, 224). As such, research studies from a critical realist perspective begin with an engagement with the ontological level of the real and work “up” to the ontological level of the empirical. Understanding the real can help researchers operationalize definitions and terms in meaningful ways that can then capture empirical reality in specific places, spaces, and times.

Shadow education is a growing topic of scholarly research across Southeast Asia. Two decades of empirical research makes the case of Cambodia an important location where lessons can be found. The case of Cambodia shows that it is important to recognize the limits of meaning inherent in survey research studies that have dominated the literature on shadow education in Cambodia. Moreover, the research on Cambodia shows the importance of researchers broadening their approaches by conducting research with different sets of assumptions than previous research studies have made. One alternative advocated here is to see reality as stratified and not flat. With a stratified ontology, new meanings of reality open new possibilities for shadow education research not only in Cambodia but also across Southeast Asia and beyond. The reality of shadow education will subsequently overcome what it ought to be.

Accepted: September 12, 2017


I would like to thank Mark Bray, Johannah Fahey, Roger Dale, and the three anonymous reviewers for their helpful comments and edits on earlier drafts of this article.


Archer, Margaret, S. 2003. Agency and the Internal Conversation. Cambridge: Cambridge University Press.

Bhaskar, Roy. 1975. A Realist Theory of Science. New York: Routledge.

Bray, Mark. 1999. The Private Costs of Public Schooling: Household and Community Financing of Primary Education in Cambodia. Paris: UNESCO International Institute for Educational Planning in collaboration with UNICEF.

―. 1996a. Counting the Full Cost: Community and Parental Financing of Education in East Asia. Washington, DC: World Bank in collaboration with UNICEF.

―. 1996b. Privatisation of Secondary Education: Issues and Policy Implications. Paris: UNESCO.

―. 1995a. Community and Parental Contributions to Schooling in Cambodia. Bangkok: UNICEF East Asia and Pacific Regional Office.

―. 1995b. Community and Parental Contributions to Schooling in the Lao People’s Democratic Republic. Bangkok: UNICEF East Asia and Pacific Regional Office.

―. 1995c. The Costs and Financing of Primary Schooling in Bhutan. Thimphu: UNICEF and Government of Bhutan.

―. 1987. Is Free Education in the Third World Either Desirable or Possible? Journal of Education Policy 3(2): 119–129.

Bray, Mark; and Bunly, Seng. 2005. Balancing the Books: Household Financing of Basic Education in Cambodia. Hong Kong: Comparative Education Research Centre, University of Hong Kong; Washington, DC: World Bank.

Brehm, W. C. 2015. Enacting Educational Spaces: A Landscape Portrait of Privatization in Cambodia. Unpublished doctoral dissertation, University of Hong Kong.

Brehm, William C.; and Silova, Iveta. 2014. Hidden Privatization of Public Education in Cambodia: Equity Implications of Private Tutoring. Journal of Educational Research Online 6(1): 94–116.

Brehm, William C.; Silova, Iveta; and Tuot, Mono. 2012. The Public-Private Education System in Cambodia: The Impact and Implications of Complementary Tutoring. London: Open Society Institute.

Burawoy, Michael. 1998. The Extended Case Method. Sociological Theory 16(1): 4–33.

Cambodia. 1996. Education Sector Review. Manila: Asian Development Bank; Brisbane: Queensland Education Consortium.

―. 1994. Education Sector Review, Vols. 1, 2A, 2B, and 3. Manila: Asian Development Bank; Brisbane: Queensland Education Consortium.

Connell, Raewyn. 2007. Southern Theory: The Global Dynamics of Knowledge in Social Science. Cambridge: Polity.

Coole, Diana. 2005. Dialectical Critical Realism and Existential Phenomenology: A Dialogue. New Formations 56(1): 121–132.

Creswell, John W.; Plano Clark, Vicki L.; Gutmann, Michelle L.; and Hanson, William E. 2003. Advanced Mixed Methods Research Designs. In Handbook on Mixed Methods in the Behavioral and Social Sciences, edited by Abbas Tashakkori and Charles Teddlie, pp. 209–240. Thousand Oaks: Sage.

Dawson, Walter. 2010. Private Tutoring and Mass Schooling in East Asia: Reflections of Inequality in Japan, South Korea, and Cambodia. Asia Pacific Education Review 11(1): 14–24.

―. 2009. The Tricks of the Teacher: Shadow Education and Corruption in Cambodia. In Buying Your Way into Heaven: Education and Corruption in International Perspective, edited by Stephen P. Heyneman, pp. 51–74. Rotterdam: Sense.

Dean, Kathryn; Joseph, Jonathan; and Norrie, Alan. 2005. New Essays in Critical Realism. New Formations 56(1): 7–26.

Descartes, René. 1637/1960. Discourse on Method and Meditations. Translated by Laurence J. Lafleur. New York: Liberal Arts Press.

George, C. 1992. Time to Come out of the Shadows. Straits Times [Singapore]. April 4.

Graeber, David. 2001. Toward an Anthropological Theory of Value: The False Coin of Our Own Dreams. New York: Palgrave.

Ivankova, Nataliya V.; Creswell, John W.; and Stick, Sheldon L. 2006. Using Mixed-Methods Sequential Explanatory Design: From Theory to Practice. Field Methods 18(3): 3–20.

Marimuthu, T.; Singh, J. S.; Ahmad, K.; Li, H. K.; Mukherjee, H.; Oman, S.; Chelliah, T.; Sharma, J. R.; Salleh, N. M.; Yong, L.; Leong, L. T.; Sukarman, S.; Thong, L. K.; and Jamaluddin, W. 1991. Extra Instruction, Social Equity and Educational Quality. Report prepared for the International Development Research Centre, Singapore.

NGO Education Partnership (NEP). 2007. The Impact of Informal School Fees on Family Expenditure. Phnom Penh: NGO Education Partnership.

Patomaki, Heikki; and Wight, Colin. 2000. After Postpositivism? The Promises of Critical Realism. International Studies Quarterly 44(2): 213–237.

Roesgaard, Marie H. 2006. Japanese Education and the Cram School Business: Functions, Challenges and Perspectives of the Juku. Copenhagen: Nordic Institute of Asian Studies Press.

Stevenson, David Lee; and Baker, David P. 1992. Shadow Education and Allocation in Formal Schooling: Transition to University in Japan. American Journal of Sociology 97(6): 1639–1657.

Straits Times-Nexus Link Tuition Survey. 2015., accessed September 17, 2017.

Tan, T. 2014. $1 Billion Spent on Tuition in One Year. Straits Times. November 9, p. 90., accessed September 17, 2017.

Ventura, Alexandre; and Jang, Sunhwa. 2010. Private Tutoring through the Internet: Globalization and Offshoring. Asia Pacific Education Review 11(1): 59–68.

Watanabe, Manabu. 2013. Juku: The Stealth Force of Education and the Deterioration of Schools in Japan. Charleston: CreateSpace.

1) I recognize that the term “shadow education” is itself problematic and debated. Nevertheless, I will use it throughout this chapter to describe the body of research that looks at the phenomenon named as such. Although I put the term in quotes at the outset, I will refrain from making similar notations in later uses.

2) The term “private” is also debated between conceptualizations that see it as either a primarily fee-based service or any educational service outside public schooling. I will not address these debates here.

3) The Latin-script rendering of the Khmer script is based on the International Phonetic Alphabet. English translations are my own.

4) Indeed, in my fieldwork (Brehm 2015), which occurred after developing the labels for the different types of tutoring presented here, I came across the term sahlah kuə. This term means, roughly, the “institution of extra class.” This phrase implies a level of institutionalization that the term ɾiən kuə does not.

5) The Cambodian currency is the riel. At that time, US$1 was worth approximately 2,600 riels.

6) It is also possible that the vocabulary employed in the Review (Cambodia 1994) derived from vocabularies in the authors’ home countries.

7) The progression of the static description but changing terminology begins with the 1994 Review (Cambodia 1994). It then moves to Bray’s 1996 comparative study of parental and community financing for education in nine East Asian countries, which included Cambodia. In this report, one of Bray’s (1996a) conclusions was that Cambodian households pay a disproportionate amount of money toward education compared to the government in relation to the other countries. This finding prompted Bray (1999) to explore the Cambodian case of private and community financing of education in more detail.

8) Bray (1996a, 32) did, however, use the terms “private supplementary tutoring” and “supplementary out-of-school tutoring” to describe tutoring in countries other than Cambodia.

9) This contrasts with the typical way Bray (1999) categorized tutoring as an out-of-school expense for families in other countries.

10) The evolution of these ideas in Bray’s work can be traced to various locations, including his reports on Cambodia (1995a), Lao PDR (1995b), and Bhutan (1995c), and an earlier paper on the challenges to fee-free education in poor countries (Bray 1987). These ideas were also undoubtedly influenced by the nascent literature on shadow education in other contexts being published at the time (e.g., Marimuthu et al. 1991; George 1992; and Stevenson and Baker 1992).

11) Bray’s (1996b, 4) study included a matrix that separated the nature of curriculum into either mainstream or alternative, and the nature of schools into elite, standard, second-chance, or supplementary. In the discussion of supplementary private schools Bray included “tuition” or “private tutoring.” Bray wrote that some supplementary private schools “shadow the public system and provide tuition in the same subjects as mainstream schools” (ibid., 20). Moreover, “the scale of private tutoring causes official embarrassment in so far as it reflects shortcomings in the public system and can be a heavy burden on household incomes” (ibid.). In this report therefore, the terms “supplementary” and “shadow” already appeared alongside “private tutoring.”

12) A critical realist approach has its limitations, too—namely, as one of the reviewers correctly pointed out, the level of the real assumes a transcendent pattern that underpins reality, an assumption challenged by philosophers such as Heidegger and Nishitani. This, moreover, says nothing of philosophical systems developed entirely outside of the West (Connell 2007).


Vol. 4, No. 2, Takahashi

Contents>> Vol. 4, No. 2

Food Supply in Cambodian Buddhist Temples: Focusing on the Roles and Practices of Lay Female Ascetics

Takahashi Miwa*

* 高橋美和, College of Human and Cultural Sciences, Aikoku Gakuen University, 1532 Yotsukaido, Yotsukaido City, Chiba 284-0005, Japan

e-mail: miwat[at]

This article, based on field research in temples in urban areas of Cambodia, aims to examine the roles of lay ascetics in Cambodian Buddhist temples from the viewpoint of the food supply system for temple residents. A number of Cambodian Buddhist temples are not only monasteries inhabited by monks but also residential places for laypeople of various categories, including female ascetics called daun chi. Cambodians in general view lay ascetics as needy people who have no family to rely on in their old age; most monastic laypeople are elderly. In reality, if we focus on food, we can see that lay ascetics do not depend entirely on the temples in which they live. From detailed observations at three temples in Phnom Penh, it is clear that temples are supported by the Buddhist community in general but that food for monks and laypeople comes through different systems of supply routes that are partly connected to each other. This article first explores how these two different food supply systems are run and maintained. Second, by depicting how female ascetics get involved in food-related practices, this article examines their dual position: female ascetics are temple residents just like monks but remain in the lay category.

Keywords: Cambodia, Buddhism, temple, food, monk, lay ascetic, daun chi

I Introduction

In Theravada Buddhist societies, which consist of people ordained as monks and laypeople, what kinds of roles are played by lay ascetics living in temples? While there have been countless studies on Theravada regions in Southeast Asia, studies that focus on laypeople in particular, especially those based in Cambodia and Laos, are very scarce. Lay ascetics1) have mostly been described as part of the faithful lay community or just mentioned on a footnote level as temple inhabitants who are subordinate to monks.

While so-called revival movements of the bhikkhunī (fully ordained female monk) order have recently taken place in Sri Lanka and Thailand, no such movements have occurred in Cambodia so far. It is believed that Cambodia has never had bhikkhunī,2) but it has female renouncers in the lay category called daun chi (ដូនជី) or yeay chi (យាយជី).3) There are some research works that focus on daun chi. Nobue Hamaya (2004) has carried out fieldwork on daun chi in Siem Reap and described their temple life, including their social engagement. Elizabeth Guthrie (2004) has provided historical analyses of laypeople’s practices based on the key word “puos (បួស)” and focused on one daun chi, who was a meditation teacher. The Buddhist Institute, Cambodia (2006) has surveyed daun chi life in one district and submitted concrete data on female renouncers’ attributes. All of these works have given us thought-provoking analyses and presented new findings, but we still cannot see clearly who lay ascetics are and what it means to live as a temple resident within the lay category. For example, ANLWC,4) a socially engaged Buddhist group that Hamaya wrote about, is unique to Cambodia, but its activities are very limited in region and even its name is not known widely among daun chi in general. As far as I have observed, most lay ascetics are not social workers or meditation teachers or Buddhist university students but seem simply to live in a temple. Therefore, to understand the general situation of daun chi, focusing on a particular group or prominent persons is probably not enough. What kind of human relationships do they have within a temple community? In order to understand lay ascetics better, we probably need to not only focus on daun chi but also observe a temple as a whole, including monks and other people.

This article focuses on, among various aspects of Cambodian Buddhist temples, food supply for temple residents, placing an emphasis on the roles and practices of lay ascetics. I use the word “practices” here in a broad sense; it means not only, for instance, purely religious practices in search of enlightenment, but also daily activities in Buddhist temples. I try to analyze Cambodian Buddhist temples through the real lives of lay temple inhabitants in order to understand how the religion “lives” in today’s Cambodian society.

It has often been said that in Theravada Buddhist societies, temple going is part of people’s everyday lives—and Cambodia is no exception. More than 90 percent of the total population of Cambodia is thought to believe in Buddhism, which is the national religion as defined in the Cambodian constitution. In fact, people have a number of occasions to visit Buddhist temples—such as during Buddhist festivals, most of which are national holidays. On a “precept (p. sīla)5) day,” which occurs four times in the lunar calendar, temples in both urban and rural areas are crowded with people gathering to “seek precepts (សុំសីល)” from monks. As a daily practice, many families place freshly cooked rice and other food in the alms bowls of monks who come to their threshold in the morning. It is very common for laypeople to invite monks to chant at rituals held in their homes, including funerals, memorial rites for the deceased, and ceremonies wishing for a long life for living parents.

Temple-goers tend to be middle-aged or elderly. The institution of the Buddhist temple has much to do with people’s life stages. In other words, Cambodians have a clear idea that their life as older people will naturally entail activities such as visiting a temple to listen to monks’ dhamma talks (សម្ដែងធម៌), receiving and keeping lay precepts, serving the monks by doing chores in a temple, preparing for festivals, and even becoming lay ascetics or monks living in a temple.

These activities, especially those occasions where laypeople have contact with monks, have one practice in common: offering food to monks. Besides placing food in the monks’ alms bowls, which is the most direct act, people also bring food to monks on precept days. In addition, when people invite monks to their houses to chant, if the monks are there before noon they are always offered a meal.

Today, Buddhism in Cambodia follows the Theravada tradition, which originated in Sri Lanka. Following the teachings of Theravada Buddhism, monks must live only on food offered by others. For laypeople, giving food to monks is one of the most important religious obligations in support of Buddhism. This basic relationship between monks and laypeople has been described by many scholars. For example, Masaki Onozawa ([1982] 1995), who studied Thai Buddhist society, points out that there are two kinds of aspiration among Thai Buddhists: monks aspire to supreme enlightenment (p. nibbāna), while laypeople aspire to make merit (p. puñña) so as to realize a better life in this world and the next. Onozawa explains that monks and laypeople are in an interdependent relationship through the exchange of materials (food and donations) and Buddhist “merit”: material to monks and merit to laypeople.6)

This view of the relationship between monks and laypeople can also, in general, be applied in Cambodia. So what about lay ascetics? They live in temples but are not ordained as monks, so they cannot go out for alms. Sumiko Yamazaki (2011) provides a unique analysis of conflicts in Buddhist communities about food gathering/offering activities in a Lao-speaking area of Stung Treng Province, but she takes into account only food for monks, not food for lay ascetics.

Therefore, this article will examine the food supply in Cambodian Buddhist temples based on information collected in my own fieldwork, focusing on laypeople’s lives and practices. The primary data used in this article come from the first and second sources as follows, but some findings from the third source are also included:

(1) interviews with several lay ascetics at Temple SD, conducted in December 2012;

(2) interviews and questionnaires with lay ascetics at three temples (Temples SD, CK, and NV) in Phnom Penh City,7) conducted in August 2011;

(3) interviews and questionnaires with all inhabitants—including monks and lay ascetics—at all 48 temples in Kien Svay District, Kandal Province, conducted in August 2009 and August 2010.

Although the third set of interviews and questionnaires included rural areas, it should be noted that my research targets in the first and second sets were limited; I took up only the city-type temples in Phnom Penh with a relatively large population of daun chi, so some of the findings shown in this paper might not be observed in rural areas.

II Overview of Cambodian Buddhist Temples

What Is a Temple?

The voatt (វត្ត, p. vatta) is the most common type of Buddhist institution in Cambodia.8) Most voatts are registered by the Ministry of Cults and Religions, but there are some cases where only a district-level office of religious affairs admits their existence. In any case, in this article I take a voatt to be a temple. In 2013 there were 4,676 temples in Cambodia (AKP 2013).9) In the plains region that includes the capital, Phnom Penh, and its surrounding provinces, there is a very high density of temples as well as population; almost every commune has more than one temple.10)

A voatt’s compound normally contains the main hall (ព្រះវិហារ, p. vihāra) with a sacred area (p. sīma) marked by sacred stones buried underground, where ordination ceremonies are held; the monks’ dining hall (សាលាឆាន់11)), where laypeople gather on precept days and for various other ceremonies; and the monks’ huts or dormitories (កុដិ, p. kuṭi). Many voatts have a public school that adjoins the compound or is located inside it. Some, although not many, also have a crematorium.

Residents of Cambodian Buddhist Temples

 1. Monks

A temple has at least two monks who are full-time residents. Monks are classified into two categories: (1) monks who have been fully ordained (p. bhikkhu) through the ordination ceremony (p. uppasampadā) and are older than 20,12) and (2) novice monks or probationers (p. sāmaṇera) to become bhikkhu who are not yet fully ordained. The total number of both categories was 57,573 in 2013.13)

According to the data collected in Kien Svay District, Kandal Province, in 2009—when I visited all 48 temples in the district—there were 1,182 residents. Among them, about 40 percent were born in the 1980s and about 50 percent in the 1990s. In other words, the majority were in their twenties or teens. This age imbalance was observed also in Kampong Thom Province, where Satoru Kobayashi, a member of our joint research group,14) obtained a similar result in his data in the same year; most monks were sāmaṇera or young bhikkhu under the age of 30. Because the same disproportion in age was observed in both semi-urban (Kandal) and rural (Kampong Thom) areas, it is strongly suspected that this is a nationwide trend.

As is well known, no religious activities were allowed during the Democratic Kampuchea period (the so-called Pol Pot regime), and monks were forcibly disrobed. The Cambodian people removed the ban on Buddhism after 1979, when the new socialist regime began. During the 1980s, however, only elderly men were allowed to be ordained as monks because the government needed young men for military service and conscription. Therefore, officially ordained young monks were absent for as long as a decade after 1979. At present, more than half the monks are students who are taking courses from the Buddhist Education (ពុទ្ធិកសិក្សា) curriculum. Education is obviously one motivation for becoming a monk, as middle-level education is not yet nationwide. So far there are no statistics regarding the length of monkhood, but as far as I have observed and heard in Cambodia, few people are monks for their whole lives. Thus, the absence of young monks in the 1980s and the situation of monkhood today have resulted in the present age distribution of monks: most are younger than 30 or very elderly, and monks in their thirties to fifties are quite scarce.15)

2. Lay Residents

In the Cambodian tradition, only men can be ordained as monks. According to the vinaya, the basic rules of monks, monks must refrain not only from marriage and any sexual acts but also from touching women physically. Originally, Buddhist temples were monasteries housing only male monks. The segregation of women is still observed in some areas in Southeast Asian Theravada regions. For example, in Thailand I observed that women were not allowed to enter the main hall (p. vihāra or uposatha) in the ethnic Mon people’s temples, even if they were mothers whose sons’ ordination ceremony was being held inside. The same goes for the compound of a cone-shaped tower (p. cetiya) at some of the temples in Chiang Mai, in northern Thailand. In Cambodia, however, women are not banned from entering any buildings in the temple compound. They are allowed to talk to monks in their room if the door is kept open, hand things to monks directly, and even live inside the temple compound, as long as their residence is clearly separated from the monks’ residential sections. As a matter of fact, Cambodian temples, especially in the urban areas, have many categories of laypeople, both male and female, who reside in the compound:

(1) Ascetics (male/female): There are often more female than male ascetics in a temple;

(2) Volunteers: Monks’ disciples (male only) who do chores for monks (កូនសិស្សលោក); a usually male lay priest (achar អាចារ្យ, p. ācārya);16) male/female members of the temple committee (គណៈកម្មការវត្ត) organized to take care of its financial matters, among other things; temple cooks (male/female); and others;

(3) Students or workers (male only) who stay in vacant rooms or spaces in the monks’ dormitories and commute to their college, office, etc.;

(4) Employees (male/female): Temple cooks, guards, etc., who work in the temple in order to receive a salary;

(5) People with long-term illnesses (male/female).

Not all temples have people in all the above categories. For example, the people in the first category are concentrated in a limited number of temples, especially in urban areas. In the second category, not all achar and temple committee members stay in the temple compound permanently. Most of these people live in their own houses and commute to the temple. People in the third category are commonly seen in urban areas, while people in the fourth category are seen at a limited number of large temples. The fifth category is a rather rare one in Cambodia: such people are observed only in a temple where there is a monk who is famous for his special ability to heal sicknesses.

The first category is taken up for analysis in Section IV, since the people in this category are most numerous among the lay residents in temples.

III Food Supply and Maintenance for Monks

Food and Eating in Compliance with Vinaya

According to vinaya, Theravada monks must live on food offered by others; therefore, they do not purchase food, engage in farming, or hunt animals by themselves. In other words, monks’ lives are entrusted to laypeople.

Usually monks have meals twice a day, with the second one finishing before noon. They have breakfast around 6:30–7am and lunch around 10:30–11am. From noon until the following morning, they are allowed to ingest only liquids. While alcoholic beverages are all prohibited because they can prevent monks from thinking clearly, there is no restriction on the intake of food, including meat and fish. Thus, monks are not allowed to choose what they can eat.

As a general rule, monks as a group go out to practice mendicancy from house to house in the morning, shortly after nine o’clock. The food that they collect in their alms bowls is supposed to be their lunch for the day. Laypeople prepare freshly cooked rice and other dishes for monks; they do not give leftovers from the day before. Rice is put directly into the alms bowls, but other food is usually put in a small plastic bag and then placed into the alms bowls or into stacked metal food containers. Sometimes the monks’ disciples who accompany this alms walk receive food from laypeople and help to carry the food containers. In any case, laypeople give food to all the monks as a group, not to the monks whom they prefer personally. For breakfast, however, monks eat simple rice porridge. The rice for this porridge, donated by laypeople or bought with donated money, is stored in the temple, so monks do not have to go out to collect it.

At mealtimes all monks from the temple, except the sick or elderly, get together in the dining hall and sit at low tables in order of their monkhood status and length of service, with two to four monks at a table.17) The collected food is put onto ordinary plates. Rice is served individually, while other dishes are shared. Thus, the general rule for the monks’ daily meals appears to be: the same food, with everybody, at the same place, at the same time.

On precept days, which occur every lunar week, monks do not practice mendicancy, because—as mentioned earlier—a number of laypeople visit temples not only to receive precepts from monks but also to donate food (praken changhan, 18002.jpg)18) to them. In many cases the amount of food donated on a precept day is more than needed, so the surplus can be distributed to lay ascetics or other temple residents. During annual Buddhist festivals, more food is gathered by temples.19) Particularly during the 15-day phchum ben (ភ្ជុំបិណ្ឌ) period, temples are literally full of rice, various dishes, fruits, and sweets.20)

Monks sometimes take meals at laypeople’s houses when they are invited to chant sutra on occasions such as funerals, memorial services for the deceased, and rituals wishing for a long life for elderly parents. If they are invited in the afternoon, however, only beverages are served. When laypeople invite monks to their homes, in addition to meals or beverages, they often offer cash in an envelope, daily necessities (candles, incense sticks, stationery, etc.), and preserved foods (sugar, canned condensed milk, tea leaves, bottled soft drinks, etc.). People do not name the monks they want to invite but just inform the temple achar in advance as to the number of monks they need, so these meals and gifts are offered collectively to the group of monks who visit, not to preferred individuals.

Food Cooked in Temples

Since Phnom Penh and the surrounding districts in Kandal Province have a concentration of middle- and higher-level schools that are part of the Buddhist Education curriculum, where the students are mostly monks, there is a very high density of monks; Phnom Penh, with the highest density, had 57.4 monks per temple on average in 2008.21) The majority of young monks who reside in temples in the Phnom Penh area are quite busy commuting to and attending these schools, which are located in other temples or in the compound of the Ministry of Cults and Religion. They attend school every day, both in the morning until the before-noon meal and in the afternoon, except on precept days, when Buddhist schools close. Because of this learning schedule, young student monks are simply too busy to practice mendicancy in the morning.

Thus, while at some temples monks take turns going for alms walks to maintain the tradition, more temples in the Phnom Penh area prepare all or most of the food for the monks in the temple kitchen instead. The cooks are laypeople; some commute to the temple and work in the kitchen voluntarily, some are allowed to live in the temple as a reward for labor, and a few others work as temple cooks to earn wages. Some female ascetics also participate in the kitchen work or take responsibility for the overall management of food preparation for monks, as shown below.

Dak Ven in Food Supply

Dak ven (ដាក់វេន)” literally means to assign (dak) duties (ven) to a person or people. This phrase is not particularly Buddhism-limited but is heard very often when talking about monks’ food supplies. Even though monks’ meals are cooked in the temple kitchen, the ingredients essentially have to be offered by laypeople outside the temple. Providing monks’ food every day is a meaningful but heavy task for Buddhist laypeople to fulfill, so the local people share this burden through dak ven (assigning duties) among themselves; they take turns providing prepared food or ingredients on a regular basis so that the everyday food supply for monks is secure.

There are three basic ways of fulfilling the duty: (1) bringing home-cooked dishes to the temple, (2) carrying foodstuffs (meat, fish, vegetables, etc.) to the temple kitchen and asking the temple cooks to prepare meals, and (3) bringing cash for meal expenditure and entrusting the temple cooks with shopping and cooking.

Let us take Temple SD in Phnom Penh as an example. The monks in this temple have not practiced mendicancy for years, because most of them are young student monks and are busy attending school. The temple has its own kitchen and six cooks, of whom five are laywomen who live in the neighborhood and one is a female ascetic who lives in the temple. Their task is to purchase ingredients in the market and cook food in the kitchen. One cook said that the market vendors know that the food purchased by the kitchen staff is for monks, so they sometimes give additional amounts for the same price. And if the cooks need chickens, for example, the vendors choke them to death for the cooks, because they know that the cooks observe Buddhist precepts and must refrain from killing animals.22) The cooks work on a voluntary basis without a salary, but the temple subsidizes their medical fees when necessary because their health is of great concern to the temple in maintaining its meal supplies.

Temple SD accepts any of the three methods of dak ven above, but the third occurs more often than the other two. In this temple, the administrative work related to the kitchen budget and dak ven is managed by the chief of the female ascetics and the vice chief, in consultation with certain temple achar. The chief keeps the collected money received from ven members and hands the daily budget, which averages 60,000 to 70,000 riel,23) to the cooks each day. A day’s ven is fulfilled by one or more families, i.e., ven members, whose leader is called “mchas ven (ម្ចាស់វេន),” literally meaning “owner of ven.” Each ven leader is obliged to fulfill the duty on the same date of the lunar calendar every month. The temple usually needs dak ven every day except precept days.

The total amount of money from the mchas ven varies each day; it can be as much as US$100. As a general rule, each day’s ven money is required to be spent on that day to reflect the wishes of the mchas ven. On the other hand, if a mchas ven is not able to bring food or money in time, the chief of the daun chi temporarily makes up for the shortfall out of her own pocket.24)

Temple SD accepts irregular or partial ven as well as regular ven. For example, there are rich Cambodians living abroad who may visit this temple to fulfill only one day’s ven. On the other hand, sometimes a family is not rich enough to fulfill a full day’s ven and so would like to offer a meal to one or a few monks only. Essentially, according to the chief, any donation or form of fulfilling ven is welcomed.

The 15-day phchum ben in the latter half of lunar October is a special festive period of the year. This is a significant occasion for temple fund-raising, and for laypeople it is the time for remembering ancestors and accumulating a lot of merit by donating food and money to several temples, including those in remote areas. Each temple organizes a special dak ven system for phchum ben, usually assigning duties to local people living near the temple. It is usually the lay committee of a temple that is in charge of dak ven for phchum ben.25) In rural areas, several villages (ភូមិ) may share this duty for one temple. A village is divided into several groups, each consisting of around 10 households, and the group leader plays the role of the temple ven’s leader and takes responsibility in case any group member cannot fulfill their duty.

IV Food Supply and Maintenance for Lay Ascetics

Lay Ascetics

In Cambodia, ordinary laypeople (p. upāsaka [male]/upāsikā [female]) customarily keep lay precepts (កាន់សីល) at home when they get old, whether they live in rural or urban areas. There is no strict rule about the age that they should start keeping precepts, but they generally start around age 50 to 60, when they retire from taking economic or housekeeping responsibility for their family. There are two kinds of precept sets: five precepts and eight precepts. Those who keep precepts are required to refrain from the following activities: (1) killing living things, (2) stealing somebody else’s belongings, (3) having extramarital relationships, (4) telling lies, (5) imbibing liquor or drugs that would affect consciousness, (6) wearing accessories or perfume and enjoying music or dancing, (7) eating at the wrong time, and (8) sleeping on a high bed.

Of the eight, the seventh requires the greatest effort because “wrong time” here means from noon until dawn of the following day; so people who hold the eight precepts have to maintain a half-day fast. Therefore, many laypeople prefer to keep the first five precepts and add the other three only on precept days.

Lay ascetics in Cambodia often identify themselves as upāsaka/upāsikā. Lay ascetics keep the same precepts, but they keep the set of eight precepts every day and live in a temple all the time. In addition, they shave their hair and eyebrows and wear white robes. As mentioned above, female ascetics are called daun chi or yeay chi; male ascetics are called ta chi. As far as I have observed in temples in several provinces in Cambodia, while some temples have only monks, many temples have both daun chi and ta chi, and the ascetic population tends to be concentrated in a few temples in one region, not spread evenly. In addition, if a temple has any ascetics, the population of daun chi always outnumbers that of ta chi. So far, I have not ascertained the exact reason behind this gender imbalance, but I speculate that it might be due to the gender gap in religious practices as well as demography. First, men have the choice to become ordained as monks all through their lifetime regardless of their age if they wish to do so, whereas women do not; there have never been female monks in Cambodia. Second, according to the census, there are fewer widowed men than widowed women in Cambodia in the middle-aged and elderly population, so more men tend to stay at home with their spouses as ordinary laymen.

Attributes of Lay Ascetics

This sub-section and the sub-sections of Section IV that follow are based on my 2011 survey conducted at three temples in Phnom Penh, where a comparatively large number of lay ascetics reside. Using the questionnaire sheets, the members of my survey team interviewed ascetics individually.26) Out of around 250 ascetics in total, I received data from 206: 195 females (95 percent) and 11 males (5 percent). To abstract gender elements from consideration, here I exclusively analyze the women, who predominated. Therefore, the percentages in the following description come from the denominator of 195.

1. Age

As Table 1 shows, 158 of the ascetics were in their sixties and over and accounted for 81 percent of the group. This tendency is identical to that of the lay ascetics as a wider group in Kien Svay District, Kandal Province; most female ascetics are elderly.


Table 1 Age Distribution of Female Ascetics in Three Phnom Penh Temples


2. Marital Status When They Started Ascetic Life in a Temple

Fifteen percent had never been married, while 68 percent had been married at least once but had lost their husbands through either death (58 percent) or divorce (10 percent). Interestingly, 16 percent of them were still married. This means that the husbands were either living at home (13 percent) or residing in a temple (3 percent) as male ascetics (ta chi), lay priests (achar), or monks. In cases where wife and husband resided in the same temple, they lived separately in different sections within the temple compound.

3. Number of Children

Seventy percent had at least one living child, and 32.8 percent had four or more. It should be noted that some of these people may have been able to rely economically on their own child/children or had the choice of living with them in the future. In addressing the question of their future plans, 34 percent said they were planning to leave the temple and live with their close relatives or children when they were older, while 63 percent said they would stay in the temple as long as they lived.

4. Economic Conditions before Entering a Temple

I did not ask specifically about previous annual income and so forth, but most of the lay ascetics were not from rich families, according to the interviews. Their previous occupations were wide ranging: farmers, vendors, public workers, housewives, etc.

5. Literacy

Elderly people in Cambodia in general have little education, and ascetics’ literacy levels reflect this. While 30 percent of them answered that they could “read and write well,” 32 percent of them answered that they could “neither read nor write at all.” The rest were able to “read a little but cannot write” or “read and write a little.” This literacy situation indicates that only a limited number of ascetics would be able to absorb Buddhist teachings through books.

6. Ethnicity

I did not include questions on ethnicity in the questionnaire. Each temple apparently had a number of people of Chinese descent, which we could sometimes tell by a person’s name or appearance—but all were Khmer nationals, and this was how they identified themselves. The ethnic factor did not seem to make any significant difference to the ascetics’ lives or human relationships within a temple, as far as I could observe.

Food Supply and Its Support

In the case of small-scale temples with fewer than 10 female ascetics in rural areas, they cook and eat together. The three temples in this survey are located in an urban environment and have a large number of ascetics. The temple population and the means of food supply are slightly different for the different temples, as shown in Table 2, but the common point is that lay ascetics get everyday food on a self-supply basis, and they essentially eat alone or with a few close ascetic friends. In other words, lay ascetics in the temple do not depend solely on the temple for food.27)


Table 2 Overview of Three Phnom Penh Temples and Their Food Supplies


As Fig. 1 shows, the most frequent answer to the question “How do you supply your everyday food?” is “Self-supply” (either cooking or buying), at 81 percent. The answer “Receive temple food” (16 percent) refers to the case of Temple CK, where polished rice (uncooked) is distributed to ascetics. According to my own observations and interviews, even ascetics who usually cook by themselves eat temple food on precept days and Buddhist festival days, when temples have large amounts of food donated by ordinary laypeople outside.


Fig. 1 Sources of Food Supply

Source: Field survey by the author in 2011

According to Fig. 2, while a limited number of ascetics (9 percent) work as cooks in the temple kitchen every day, the majority do not.


Fig. 2 Labor in the Temple Kitchen

Source: Field survey by the author in 2011

The result I did not anticipate was that many of the ascetics offered food to monks (Fig. 3). As many as 91 percent of them answered that they offered food to monks every day, occasionally, or on precept days. This reveals the ascetics’ dual characteristics: they are temple members, but at the same time they also remain in the lay category, so they still offer food to monks.


Fig. 3 Frequency of Food Offering to Monks

Source: Field survey by the author in 2011

The cash that ascetics need to buy food or materials for cooking, as Fig. 4 shows, comes most often from family or close relatives, and secondarily from donations (p. paccaya) they receive outside the temple when they are invited to funerals or other family rituals (29 percent). They receive donations from ordinary laypeople just as monks do, but they are supported mainly by their own family and relatives.


Fig. 4 Sources of Cash Income

Source: Field survey by the author in 2011

Fig. 5 sheds light on another aspect of lay ascetics’ lives. We can see that the top category of expenditure is everyday food, which is understandable,28) but the second-most frequent expenditure is on donations to monks and temples. Making donations is certainly a major part of their temple life.


Fig. 5 Cash Expenditures (multiple answers)

Source: Field survey by the author in 2011

Cases: As a Food Taker and a Food Giver

Each female lay ascetic has a different way of securing food as well as cash for food and other expenditures. Following are some of the sources of supply among those who reside in Temple SD.

Daun chi A (Age 85; born in Kampong Chhnang Province in the central region, on the southern coast of Tonle Sap Lake)

I have only one daughter and a grandson. She is a single mother, employed as a maid and babysitter by a family in Phnom Penh. Although she has this stable job, her income is barely enough to make ends meet. Her son is grown up but jobless, so I cannot rely on him.

After my daughter left Kampong Chhnang for Phnom Penh to work, I felt lonely and was not sure how to support myself. Sometimes I felt scared of ghosts and such things. I decided to come to Phnom Penh and start living in this temple as an ascetic in order to strengthen my mind and to be near my daughter. My daughter was also happy because this temple is not very far from her workplace. As a matter of fact, she stayed with me in this room for two years as an ascetic. She eventually stopped being an ascetic and went outside to work again because she needed to earn money.

I share this room with several other ascetic women, but we prepare meals separately. Usually, I eat rice porridge for breakfast and cooked rice and a couple of dishes for lunch. I occasionally receive some portion of the dishes prepared for monks in the temple kitchen, but I usually go to the market by motorbike taxi to buy groceries and cook for myself. For my expenditures, my daughter gives me 40,000 to 50,000 riel a month. Another source of income is donation money that laypeople give us, but such donations are not regular. When I was much younger I used to be invited to laypeople’s houses for funeral chanting and received donations, but I do not go nowadays because of my health condition.

I have gallstones. I do not think I will have an operation, because I am too old, but I need medication. This temple does not provide us with any medical support, but my daughter has given me about $500 in total for my medical fees. Her employer is so kind that he drives me to the hospital. My daughter lives a busy life but visits me at least once a month to give me some delicious food, help me bathe, and so forth. She says it is her employer who buys the food for me.

Several monks come to this temple for alms. Only when I have freshly cooked food or a lot of food from my daughter do I give some to them.

Daun chi B (Age 85; born in Kampong Thom Province in the central region, on the east coast of Tonle Sap Lake)

I have lived in this temple since I was over 60 years old. I became an ascetic because I wanted to be able to chant sutras. My husband died during the Pol Pot regime. I had 12 children: seven sons and five daughters. Sad to say, I lost all the sons—but the five daughters survived. My daughters are all married and living in Phnom Penh. This temple is located near their houses. That is why I chose this temple to live in. If I need any help, I can call them and ask them to come here any time.

Of the five, the second and third daughters are richer than the others and give me more financial support. All five visit me, but because they are quite busy, when they come to give me something to eat, they leave the food and return very quickly. If the daughters are too busy my grandchildren come instead. This morning one of my grandchildren came to bring me a pot of soup. The two weeks of the phchum ben period is the exception; all the daughters visit me one after another and stay longer. Anyway, my daughters or grandchildren bring the rice and dishes that are my main food every two or three days. They also buy me the medicine that I need.

Sometimes I cook for myself if necessary. I give some money to a temple cook and ask her to buy groceries when she goes to the market. I just broil fish and cook simple dishes when the food from my daughters is not enough. For this, they give me around $10–20 in cash every month. I do not cook every day, so I can rarely offer food in the alms bowls of the monks who come to this temple.

When I was younger I used to be invited to funerals often, but these days I do not go because I am too old. If I sit on the floor for long while chanting, I cannot stand up by myself. Therefore, I usually have no occasion to receive donation money from laypeople, but the phchum ben period is different. A layperson will give me a donation, on average, of 3,000 to 5,000 riel at a time.

Daun chi C (Age 65; born in Takeo Province in the southern region)

Before I came to this temple to live, around 1993, I stayed home, keeping the lay precepts. In those days, Ven. Pal Haun was the abbot of this temple.29) There were not many monks, but there were more than 100 female ascetics. I often used to come to this temple to do volunteer work. Knowing that I had no husband or children to rely on, Ven. Pal Haun recommended that I live here. I was not yet old, but I decided to become an ascetic and live in this temple because I realized that it was my way of life to serve the three gems (i.e., the Buddha, dhamma, and saṅgha).

When I was very healthy, I was often invited to laypeople’s rituals, including funerals, where I recited chants and received donation money; but now I rarely go to chant because my knees and back ache if I stay seated on the floor for too long.

I have no direct descendants, but I have four nephews and nieces who are related to me by blood. All of them are the children of my sister, who passed away. Among them, one nephew supports me continuously. He lives in Otdar Meanchey Province in the northern region and sends me rice and money via a bus driver whom he trusts. I receive $50 a month.

This nice hut where I live over here was given by its former owner who stayed in this temple as an ascetic but later immigrated to Australia. Now she comes to visit me every year and gives me $20. I expanded this hut into a two-storied one with a ladder and a bigger roof using money donated by my friends and acquaintances, my nephew’s wife, and so on, together with my own savings.

I live alone, but I often host acquaintances who need one night’s stay on a precept day or who want to stay for three months of the Lent season as short-term ascetics. I became acquainted with them in this temple, so even though I have no family I do not feel lonely.

I go to the market for groceries and cook for myself. The market people know me very well, so they sometimes give me extra food for the same price. Almost every day, I offer food to three monks who belong to another temple and commute to this temple for alms. Sometimes I also offer food to the young monks of this temple who have missed breakfast because of their busy study schedules.

I take part in ven for this temple, so I cook in the temple kitchen five times a month. I’ve also enrolled myself in a group of ven with about 300 members that fulfills part of ven for as many as 28 temples. I support the group by paying 10,000 riel per month.

In addition to working to fulfill ven, I keep learning dhamma. I often go to other temples to listen to lectures by a dhamma teacher.

I am deeply satisfied with my temple life and feel very happy.

V Conclusion

Food Supply Systems and the Connections between Monks and Lay Ascetics

Food for monks and lay ascetics is supplied via different systems that are connected to each other by means of various apparatuses—economic necessities and religious values (Fig. 6)—and the food for temple residents as a whole is well provided and maintained.


Fig. 6 Food Supply and Religious Merit

Source: Prepared by the author

Monks, who never get involved in productive labor, entrust laypeople with everything related to food. Aside from direct food offerings to the monks, such as mendicancy and meals served at laypeople’s houses at rituals, daily meals for monks in temples in urban areas are supported by shared ven, for which the labor and management are provided by faithful laypeople and ascetics.

The food supply for lay ascetics is more varied because such people are not restricted by vinaya. A number of needy ascetics depend on the temple kitchen for food to some extent. Some get enough food by providing their own labor in the temple kitchen, but most ascetics secure their food through support from their family and close kin. They also get a cash income in donations from the laypeople who invite both monks and ascetics to their family rituals. It is worth noting that while lay ascetics manage to secure their own food in either of the above ways, many of them still try to offer food to monks by cooking personally, helping in the temple kitchen, or joining a ven group. They are food receivers but at the same time play a part in food provision.

Regarding the relationship between monks and laypeople in general, all the food (and money) for the former comes from the latter, and the donated food is offered to all the monks of the temple, not to individuals. Even the food put into alms bowls during mendicancy is shared at meals. Lay ascetics, too, are sometimes invited to laypeople’s houses together with monks and receive donations, but only as part of their role of being attached to a group of monks. They do not practice mendicancy and are not supported by ven. Therefore, lay ascetics need to seek way(s) of supporting themselves on an individual basis. In the process of doing this, some ascetics, like daun chi C in the previous section, can create new personal relationships with other laypeople outside the temple through their religious practices.

For lay ascetics, the temple is not only a place for religious practices, such as sutra chanting, meditation, and listening to dhamma talks, but also a place where they live a peaceful life during their old age as faithful Buddhists; consequently, many of them include food offerings to monks in their practices. Some offer food to monks once in a while when they receive large amounts of food from their family or kin members, some become involved in temple ven and cook regularly for monks, and some take responsibility in the management of ven, which involves many laypeople outside the temple. All of these practices are considered to accumulate merit, which is quite important for ascetics, who are mostly elderly, because their remaining time will not be long.

The fact that a number of temples in Cambodia have lay residents living in them means that Cambodian Buddhists have multiple choices in their practice all through life, so they always have the opportunity to become involved, in various positions that they can choose, in receiving and supplying food, one of the most fundamental and important practices in the sustenance of Buddhist temples and Buddhism itself.

Lay Buddhists and Temples: Implications for Future Research on Cambodian Buddhism

The discussion above regarding food supply for monks and lay ascetics leads first to rethinking the significance of lay ascetics in Cambodia. A Buddhist temple has been considered to be a monastery where (male only) monks reside. Few other research works on lay ascetics in Cambodia have appeared so far, and the section of the Cambodian administration that is involved in religious affairs did not show any interest until 2009.30) Using the annual statistics of temples and temple residents issued by the Ministry of Cults and Religion in 2009 as an example, it is clear that not all of the provinces reported the population of lay ascetics, which means that the number of lay ascetics was not regarded as requisite data. From the facts I have presented in this article, we can see that female ascetics play a significant role in temples and in the wider Buddhist communities surrounding each temple, especially in urban areas. For female ascetics, a Buddhist temple is relevant for three reasons: (1) being Buddhist disciples, female ascetics can engage in practices such as sutra chanting, meditation, etc.; (2) being advanced-level precept keepers, they can become merit-making targets for laypeople in general, just like monks; and (3) being in the lay category, they can serve monks by becoming involved in food supply in a direct or indirect way. To sum up, lay ascetics have their own roles and activities, including food supply, and they are not completely dependent on monks.

Second, it should be noted that the condition of lay ascetics has been changing and is continuing to change, albeit gradually. Cambodian people in general share a common image of lay ascetics as being pitiful old people who have no family to rely on and no other choice but to stay in the temple for the rest of their lifetime. This was true to some extent in the 1990s, just after the UNTAC period; I personally encountered numerous such cases. However, it is not always the case any more. There are probably more and more ascetics who prefer to stay in a temple for a limited number of years and then move on to their daughter’s (or other kin’s) house to live out their remaining years. In short, most of them are not alone.

Third, I would like to suggest that we should look at temple residents from the perspective of the life course. Being a monk or a layperson is not usually a lifelong status. In today’s Cambodia, it is no longer a social norm for young men to become monks either. The decision to become a monk or a lay ascetic is made by the person himself/herself, not by force. Therefore, people’s choices in Buddhist practices can change in accordance with their life stage, family environment, preferences, etc. A man might become ordained as a monk, stay in the monkhood for several years, disrobe and get married, visit temples occasionally, and start living in a temple again as a lay ascetic in his old age. The statuses of monk, ascetic, and ordinary layperson can be chosen one after another during the course of a person’s life. Although monkhood is not yet open to women in Cambodia, it should be noted that there are some female ascetics who have considerable knowledge of Buddhism and give lectures to monks and laypeople.31) For some of these people, becoming a learned ascetic may be an option in the future.

Accepted: June 25, 2014


This work was supported by the Japanese Society for the Promotion of Sciences KAKENHI, Grant numbers 23520997 and 20251003. I would like to express my sincere gratitude to all those who answered my endless questions and cooperated in the questionnaire, especially the laypeople in Temples SD, CK, and NV.


Buddhist Institute, Cambodia. 2006. The Situation of Daun Chi in Cambodia. Phnom Penh: Buddhist Institute.

Cambodia, Ministry of Cults and Religion. 2013. Kamrong Aekasa Sthete Vat Vihear Sala ning Samanasoes-Nissit កម្រងឯកសារស្ថិតិវត្ត វិហារ សាលា និងសមណសិស្ស-និស្សិត [Compiled statistics of temples, schools, and monk students]. Phnom Penh: Ministry of Cults and Religion, Cambodia.

―. 2009. Sthete Vat Asram Putthamondol Vipassanathurea Preahsang Preahanukon Preahmekon Preahreacheakena Tuteang Preahreacheanachak Kampuchea ស្ថិតិ វត្ត អាស្រម ពុទ្ធមណ្ឌល វិបស្សនាធុរៈ ព្រះសង្ឃ ព្រះអនុគណ ព្រះមេគណ ព្រះរាជាគណៈ ទូទាំងព្រះរាជាណាចក្រកម្ពុជា [Statistics of Buddhist institutions and monks in the Kingdom of Cambodia]. Phnom Penh: Ministry of Cults and Religion, Cambodia. Mimeographed.

Cambodia, Ministry of Cults and Religion, Neayokathan sravchreav phsapphsay putsasna phsaphchoap sangkom នាយកដ្ឋានស្រាវជ្រាវផ្សព្វផ្សាយពុទ្ធសាសនា ផ្សារភ្ជាប់សង្គម [Department of Buddhism Publicity and Social Relations]. 2012. Kram Soelathoa ning Vithi Pratebat robos Daun chi ក្រមសីលធម៌និងវិធីប្រតិបត្តិរបស់ដូនជី [Morals and way of practice of daun chi]. Phnom Penh: Ministry of Cults and Religion, Cambodia.

Crosby, Kate. 2014. Theravada Buddhism: Continuity, Diversity, and Identity. Chichester: Wiley-Blackwell.

Falk, Monica Lindberg. 2007. Making Fields of Merit: Buddhist Female Ascetics and Gendered Orders in Thailand. Seattle: University of Washington Press.

Guthrie, Elizabeth. 2004. Khmer Buddhism, Female Asceticism, and Salvation. In History, Buddhism, and New Religious Movements in Cambodia, edited by John Marston and Elizabeth Guthrie, pp. 133–149. Honolulu: University of Hawai‘i Press.

Hamaya Nobue 浜家伸恵. 2004. “Tera ni Kurasu” to Iu Ikikata: “Seido” no Moto de “Kesson” o Oginau Kambojia Josei Shugyosha no “Kurashi”「寺に暮らす」という生き方―「制度」のもとで「欠損」を補うカンボジア女性修行者の「暮らし」[The way of life of “living in a temple”: “Lives” of Cambodian female ascetics who fulfill “losses” under the “institution”]. MA thesis, Okayama University.

Hansen, Anne Ruth. 2008. How to Behave: Buddhism and Modernity in Colonial Cambodia 1860–1930. Chiang Mai: Silkworm Books (originally published by University of Hawai‘i Press, 2007).

Harris, Ian. 2005. Cambodian Buddhism: History and Practice. Honolulu: University of Hawai‘i Press.

Hayashi Yukio 林行夫. 1998. Kambojia ni okeru Bukkyo Jissen: Ninaite to Jiin no Fukko カンボジアにおける仏教実践―担い手と寺院の復興 [Buddhist practices in Cambodia: People in charge and recovery of temples]. In Kambojia: Shakai to Bunka no Dainamikkusu カンボジア―社会と文化のダイナミックス [Cambodia: Dynamics of its society and culture], edited by Hisatoshi Ohashi 大橋久利, pp. 159–219. Tokyo: Kokonshoin.

Ishii Yoneo 石井米雄, ed. 1991a. Koza, Bukkyo no Juyo to Henyo 2: Tonan Ajia Hen 講座・仏教の受容と変容 2―東南アジア編 [Lecture series, acceptance and transformation of Buddhism 2: Southeast Asia]. Tokyo: Kosei Publishing.

―. 1991b. Tai Bukkyo Nyumon タイ仏教入門 [Introduction to Thai Buddhism]. Tokyo: Mekong Publishing.

Jacobsen, Trude. 2013. In Search of the Khmer Bhikkunī: Reading between the Lines in Late Classical and Early Middle Cambodia (13th–18th Centuries). Journal of the Oxford Centre for Buddhist Studies 4: 75–87.

Khy Sovanratana. 2008. Buddhist Education Today: Progress and Challenges. In People of Virtue: Reconfiguring Religion, Power and Moral Order in Cambodia Today, edited by Alexandra Kent and David Chandler, pp. 257–271. Copenhagen: Nordic Institute of Asian Studies.

Kobayashi Satoru 小林知. 2013. Kambojia Noson ni okeru Bukkyo Shisetsu no Shurui to Keisei Katei カンボジア農村における仏教施設の種類と形成過程 [A study of Buddhist places of worship in rural Cambodia: With a special focus on their differences and formative processes]. Tonan Ajia Kenkyu 東南アジア研究 [Southeast Asian studies] 51(1): 34–69.

―. 2011. Kambojia Sonraku Sekai no Saisei カンボジア村落世界の再生 [Reformation of the rural world in Cambodia]. Kyoto: Kyoto University Press.

―. 2009. Poru Poto Jidai Igo no Kambojia Bukkyo ni okeru So to Zoku ポル・ポト時代以後のカンボジア仏教における僧と俗 [Monks and laymen in Cambodian Buddhism after the Pol Pot era]. In “Kyoiki” no Jissen Shukyo: Tairikubu Tonan Ajia Chiiki to Shukyo no Toporoji〈境域〉の実践宗教―大陸部東南アジア地域と宗教のトポロジー [Practical religions in mainland Southeast Asia: Topology of religion from the region and “in-betweenness”], edited by Yukio Hayashi 林行夫, pp. 27–65. Kyoto: Kyoto University Press.

―. 2008. Reconstructing Buddhist Temple Buildings: An Analysis of Village Buddhism after the Era of Turmoil. In People of Virtue: Reconfiguring Religion, Power and Moral Order in Cambodia Today, edited by Alexandra Kent and David Chandler, pp. 169–194. Copenhagen: Nordic Institute of Asian Studies.

―. 2005. An Ethnographic Study on the Reconstruction of Buddhist Practice in Two Cambodian Temples: With the Special Reference to Buddhist Samay and Boran. Tonan Ajia Kenkyu 東南アジア研究 [Southeast Asian studies] 42(4): 489–518.

Ledgerwood, Judy. 2008. Buddhist Practice in Rural Kandal Province 1960 and 2003: An Essay in Honor of May M. Ebihara. In People of Virtue: Reconfiguring Religion, Power and Moral Order in Cambodia Today, edited by Alexandra Kent and David Chandler, pp. 147–168. Copenhagen: Nordic Institute of Asian Studies.

―. 2002. Cambodia Emerges from the Past: Eight Essays. DeKalb: Southeast Asia Publications, Center for Southeast Asian Studies, Northern Illinois University.

Onozawa Masaki 小野澤正喜. [1982] 1995. Shukyo to Sekaikan 宗教と世界観 [Religion and worldview]. In Motto Shiritai Tai もっと知りたいタイ 第2版 [The Thailand that we want to know more about] (2nd ed.), edited by Tsuneo Ayabe 綾部恒雄 and Yoneo Ishii 石井米雄, pp. 103–147. Tokyo: Koubundou Publishers.

Sakamoto Yasuyuki 坂本恭章. 2001. Kambojia-go Jiten カンボジア語辞典(上)(中)(下) [Cambodian-Japanese Dictionary (I) (II) (III)]. Tokyo: Research Institute for Languages and Cultures of Asia and Africa, Tokyo University of Foreign Studies.

Takahashi Miwa 高橋美和. 2014. Kyodan ni Tatsu Zokujin Josei Shugyosha: Kambojia Bukkyokai ni okeru Josei no Shinshutsu Doko 教壇に立つ俗人女性修行者―カンボジア仏教界における女性の進出動向 [Female lay ascetics as teachers: Current situation of women in Cambodian Buddhist circles]. Tokyogaidai Tonan Ajia Gaku 東京外大東南アジア学 [Tokyo University of Foreign Studies Southeast Asian studies] 19: 128–146.

―. 2013. Shoku no Kyokyu to Jizoku kara Mita Kambojia Bukkyo Jiin: Zokujin Shugyosha ni Chumoku Shite 食の供給と持続から見たカンボジア仏教寺院―俗人修行者に注目して [Cambodian Buddhist temples from the viewpoint of food supply and maintenance: With special reference to lay ascetics]. Sogo Ningen Kagaku 総合人間科学 [General human science] 1: 91–105.

―. 2012. Zokujin Josei ga Jiin Zumai no Shugyosha ni Naru to Iu Koto: Kambojia ni okeru Don Chi to Haha Musume Kankei 俗人女性が寺院住まいの修行者になるということ―カンボジアにおけるドーンチーと母娘関係 [Lay women as Buddhist temple inhabitants: Daun chi in Cambodia from the viewpoint of mother-daughter relationships]. Aikoku Gakuen Daigaku Ningen Bunka Kenkyu Kiyo 愛国学園大学人間文化研究紀要 [Human and cultural sciences: Journal of Aikoku Gakuen University] 14: 1–12.

―. 2011. Josei to Bukkyo Jiin 女性と仏教寺院 [Women and Buddhist temples]. In Sei to Do no Bukkyo 静と動の仏教 [Buddhism of statics and dynamics], edited by Yasuaki Nara 奈良康明, Masahiro Shimoda 下田正弘, and Yukio Hayashi 林行夫, pp. 415–449. Tokyo: Kosei Publishing.

―. 2009. Shukke to Zaike no Kyoiki: Kambojia Bukkyo Jiin ni okeru Zokujin Josei Shugyosha 出家と在家の境域―カンボジア仏教寺院における俗人女性修行者 [Boundary sphere between monks and laypersons: Female lay ascetics in Buddhist temples in Cambodia]. In “Kyoiki” no Jissen Shukyo: Tairikubu Tonan Ajia Chiiki to Shukyo no Toporoji〈境域〉の実践宗教―大陸部東南アジア地域と宗教のトポロジー [Practical religions in mainland Southeast Asia: Topology of religion from the region and “in-betweenness”], edited by Yukio Hayashi 林行夫, pp. 359–409. Kyoto: Kyoto University Press.

―. 2006a. Konnichi no Kambojia Bukkyo Jiin to Zokujin Josei: Nikushin Soshitsu Taiken o Koete 今日のカンボジア仏教寺院と俗人女性―肉親喪失体験をこえて [Cambodian Buddhist temples and women today: Beyond family loss experiences]. Ajia Yugaku アジア遊学 [Exploring in Asia] 89: 60–71.

―. 2006b. Gendai Kambojia ni okeru Zokujin Bukkyoto no Shukyo Jissen to Jinsei: Josei Shugyosha (Don Chi) o Chushin ni 現代カンボジアにおける俗人仏教徒の宗教実践と人生―女性修行者(ドーンチー)を中心に [Religious practices and lives of lay Buddhists in Cambodia today: Focusing on female ascetics (daun chi)]. In Tonan Ajia Tairikubu, Seinan Chugoku no Shukyo to Shakai Henyo: Seido, Kyoiki, Jissen 東南アジア大陸部・西南中国の宗教と社会変容―制度・境域・実践(科学研究費補助金(基盤A)平成15~17年度研究成果報告書)[Religions and social transformation in continental Southeast Asia and Southwestern China: Institutions, boundaries, practices]. Report of the Research Grant-in-aid by the Japanese Society of Promotion of Sciences: Years Heisei 15–17, edited by Yukio Hayashi 林行夫, pp. 616–651.

Takeda Ryuju 武田龍樹. 2010. Gendai Kambojia ni okeru Bukkyo Tokushinka tachi: Porupoto Jidai Igo no Tobu Sonrakubu o Jirei ni 現代カンボジアにおける仏教篤信家たち―ポルポト時代以後の東部村落部を事例に [Lay Buddhist devotees in contemporary Cambodia: A case study of a village in the eastern part of Cambodia after the Pol Pot era]. Shukyo to Shakai 宗教と社会 [Religion and society] 16: 65–87.

Thomson, Ashley. 2006. Buddhism in Cambodia: Rupture and Continuity. In Buddhism in World Cultures: Comparative Perspectives, edited by Stephen C. Berkwitz, pp. 129–167. Santa Barbara: ABC-CLIO.

Yamazaki Sumiko 山崎寿美子. 2011. Kambojia no Rao Sonraku ni okeru Kyogo to Shakai Keisei: Soryo no Ue no Jirei kara カンボジアのラオ村落における競合と社会形成―僧侶の飢えの事例から [Competition and society formation process in a Lao village of Cambodia: A case study of food shortage for monks]. Nempo Tai Kenkyu 年報タイ研究 [Journal of Thai studies] 11: 47–64.


Agence Kampuchea Presse (AKP). 2013. Preah reachdamnang baoek anusang vacchara mohasannibat montrei sang tuteang prates loekti 22 ព្រះរាជតំណាងបើកឣនុសំវច្ឆរៈមហាសន្និបាតមន្ត្រីសង្ឃទូទាំងប្រទេសលើកទី២២ [Royal representative opens 22nd national congress of Buddhist monks], accessed January 3, 2014,

1) Some researchers call female lay ascetics in the Theravada region “nuns” or “precept nuns” (Crosby 2014, 230) in English. Kate Crosby (2014) has given her view on the general condition of female ascetics in the Theravada region.

2) Trude Jacobsen (2013) has studied inscriptions of the thirteenth to eighteenth centuries and provided a historical analysis.

3)Daun chi” is often used as an official term for female lay ascetics.

4) Association of Nuns and Lay Women in Cambodia. By providing selected daun chi with training on Buddhist values and human rights, the association aims to nurture them so that they can contribute to DV counseling, HIV/AIDS prevention, and other solutions to social problems. ANLWC has two training centers, in Kep and Kandal Provinces, and branches in 14 provinces, according to my interview in 2011. Its activities, however, have had to be reduced since aid from the Heinrich Böll Foundation, its main donor, ended in 2005.

5)p.” indicates the Pāli language. In this paper, for terms of Pāli origin related to Buddhism, I give priority to the Pāli spelling over Khmer (Cambodian) transcription, but non-Pāli key terms are transcribed in the Roman alphabet with their original Khmer script (when mentioned for the first time), based on the transcription system used by Judy Ledgerwood (2002).

6) In reality, the relations between the two kinds of aspiration are not that simple. For example, in both Thailand and Cambodia, even becoming a monk is considered an opportunity to make a lot of merit. I should discuss the meaning of “merit making” more in detail and also in the context of gender, but would like to do so in a future article. For the relationship between gender and “fields of merit” in Thailand, see, for example, Falk (2007).

7) Temple SD is located in the central area of Phnom Penh City. In 2009 the addresses of Temple CK and Temple NV were in Kien Svay District, Kandal Province, so both were included in the second as well as third categories of sources. The commune where the two temples are located has since been incorporated into Phnom Penh City due to the provincial border changes of September 2010.

8) For details on the varieties of Cambodian Buddhist institutions, see Kobayashi (2013).

9) This number is taken from the AKP article that reported the opening of the 22nd national congress of executive monks (anusang vacchara mohasannibat montrei sang tuteang prates) held on December 17, 2013.

10) Using the data taken in 2008 (before the provincial border changes) as an example, Kandal Province, which consists of 155 communes, had 395 temples. One commune had an average of 2.5 temples (Cambodia, Ministry of Cults and Religion 2009).

11) ឆាន់ (chan) means that monks “eat”; a different word is used for eating by non-monks. There are many other special verbs, nouns, and pronouns for monks.

12) The ordination to become a bhikkhu can be postponed if the candidate sāmaṇera thinks that he is not yet ready. When an elderly man wants to become a new monk, it is very common for him to be a sāmaṇera for a while to prepare for full ordination.

13) The source is the same as in footnote 9.

14) For this joint research project, supported by JSPS 20251003 “Time-Space Mapping of Buddhist Societies in Mainland South East Asia” (project leader: Yukio Hayashi), Kobayashi covered four districts in Kampong Thom Province, and I was in charge of Kien Svay District in Kandal Province.

15) This disproportionate age distribution is one reason for the lack of teachers in the Buddhist education system today (Khy Sovanratana 2008; Kobayashi 2009).

16) There are several kinds of achar, according to specialty: an achar of wedding ceremonies, an achar of funerals and cremations, and so forth. An achar of a temple is one such specialty.

17) In Cambodia, only monks use low tables for meals. Laypeople sit on the floor or on a chair when eating. Monks have different styles of having meals. For example, when they are invited to a layperson’s house for a family ritual, one tray full of dishes is prepared for each monk. Monks who practice dhūtaṅga (wandering in the forest for spiritual training) eat food from the alms bowl directly by hand without plates or a spoon.

18)Changhan18003.jpg is a special term for monks, meaning “food” or “meal.”

19) The main festivals besides phchum ben are: the māgha pūjā (February or March), the Khmer new year (April), the visākhā pūjā (April or May), the beginning of rainy season retreat (July), the end of rainy season retreat (October), the kaṭhina (October or November), and the “flower festival” held periodically for fund-raising.

20) Both phchum and ben (p. piṇa) mean “to collect.”

21) Calculated by the number of temples and monks in the data taken in 2008 (Cambodia, Ministry of Cults and Religion 2009).

22) In the fresh food market, fish and chickens are usually sold live.

23) The riel is the Cambodian currency, but US dollars are also widely accepted in markets. In 2013, US$1 was equivalent to about 4,000 riel. Polished rice retails for around 2,500 riel per kilogram in Phnom Penh.

24) It should be noted that this ascetic chief is from a rich family, unlike ascetics in general.

25) Temple SD does not have a lay committee. Several achar, the chief, and the vice chief of daun chi fulfill the duties instead.

26) For the interviews, the research team, which consisted of nine native Cambodians, worked with me. This team included graduates of Royal Phnom Penh University, elementary school teachers, and high school teachers.

27) In my interviews, people said that the population of lay ascetics increased sharply from the late 1980s until the early 1990s, at the end of the civil war period. In those days, many people may have begun living in temples in order to secure food.

28) This does not mean the highest in price.

29) Pal Haun was a well-known monk in the rite of “pouring sacred water (ស្រោចទឹក)” for laypeople during the 1990s. He was not only successful in rebuilding Temple SD using money donated for water pouring but also kept supporting laypeople such as daun chi and male students from the provinces by providing accommodation in the temple.

30) The Ministry of Cults and Religion, following the instructions of Prime Minister Hun Sen, started training seminars for achar in 2009. In 2012 the ministry also started another seminar series for female ascetics. These events may show that the Cambodian government began to recognize the significance of lay monastic members. See also Cambodia, Ministry of Cults and Religion, Department of Buddhism Publicity and Social Relations (2012) and Takahashi (2014).

31) I made a preliminary analysis of a sign of change in Cambodian Buddhist society brought by such learned daun chi in another paper (Takahashi 2014).


Vol. 1, No. 1 of Southeast Asian Studies

Published in April, 2012


Caring for the Dead Ritually in Cambodia ・・・ John Clifford HOLT pdficon_large
State Recognition or State Appropriation?
Land Rights and Land Disputes among the Bugkalot/Ilongot
of Northern Luzon, Philippines
・・・ Shu-Yuan YANG pdficon_large
Javanese Women and Islam: Identity Formation since the Twentieth Century ・・・ Kurniawati Hastuti Dewi pdficon_large
Research Report
Decision Support System Research and Development Network
for Agricultural and Natural Resource Management
in Thailand: A TRF-DSS Experience
・・・ Attachai JINTRAWET et al. pdficon_large
Book Reviews
Edward Aspinall and Gerry van Klinken, eds. The State and
Illegality in Indonesia. Leiden: KITLV Press, 2010, 328 p, with bibliography and index.
・・・ Rommel A. CURAMING pdficon_large
Caroline S. Hau and Kasian Tejapira, eds. Traveling Nation-Makers: Transnational Flows and Movements in the Making of Modern Southeast Asia. Singapore: National University of Singapore Press in association with Kyoto University Press, 2011, 318 p. ・・・ Thanet Aphornsuvan pdficon_large
Paulin G. Djité. The Language Difference: Language and Development in the Greater Mekong Sub-Region. Bristol: Multilingual Matters, 2011, 264 p.
Andy Kirkpatrick. English as a Lingua Franca in ASEAN: A Multilingual Model. Hong Kong: Hong Kong University Press, 2010, 236 p.
・・・ Nathan BADENOCH pdficon_large
Kate Lazarus, Nathan Badenoch, Nga Dao and Bernadette P. Resurreccion, eds. Water Rights and Social Justice in the Mekong Region. London and Washington, D.C.: Earthscan, 2011, 285 p. ・・・ CHEN Jianming pdficon_large
Michael R. Dove, Percy E. Sajise and Amity A. Doolittle, eds.
Beyond the Sacred Forest: Complicating Conservation in Southeast Asia. Durham and London: Duke University Press, 2011, 372 p.
・・・ Andreas NEEF pdficon_large
Lye Liang Fook and Chen Gang, eds. Towards a Liveable and Sustainable Urban Environment: Eco-Cities in East Asia. Singapore: World Scientific Publishing, 2010, 222 p. ・・・ Zhiqun ZHU pdficon_large
John Nery. Revolutionary Spirit: Jose Rizal in Southeast Asia. Singapore: Institute of Southeast Asian Studies, 2011, 280 p. ・・・ Erwin S. FERNANDEZ pdficon_large