Navigation – Plan du site

The Question of Land in Cambodia: Perceptions, Access, and Use since De-collectivization

Anne Y. Guillou
p. 299-324

Texte intégral


  • 1 I am grateful to Mrs Fabienne Luco, associate researcher at EFEO and CKS, who offered valuable data (...)
  • 2 The major features of Cambodian political history are as follow: 1863 –?French protectorate; 1953 – (...)

1Cambodia is one of the countries in the world which experienced the most radical disruption of its land tenure.1 Almost its entire present adult population has been forced at one time or another to leave its lands and houses, whether because of the American bombings (1969-1973)2 which were particularly murderous in the south-eastern part of the country (the bombs’ holes are still visible from an airplane); or on a far greater scale, during the Democratic Kampuchea regime, called the “Khmer Rouge” regime (1975-1979), when the whole urban population was sent to the countryside. A large part of the rural population was also forced to leave its lands and homes in order to take part in the large-scale forced collective labour or to participate in the mobile youth teams (kong chalat). In the years following the breakdown of the Khmer Rouge regime, hundreds of thousands of Cambodians ran away from the Popular Republic of Kampuchea which was supported by the socialist regime of Viêt Nam (1979-1989). The refugees were sheltered in camps built along the Thai border. And year after year, during the dry season, governmental troops launched military offensives against the opposition guerillas (mostly Khmer Rouge soldiers), causing hundreds of additional escapes and destabilizing the northwestern provinces. In addition to these large-scale population movements, smaller, well-established migrations caused by floods or (less often) by drought, regularly occurred. These climatic accidents happen more and more often, almost annually now, and may be due to the massive deforestation in the northeastern provinces.

2Despite all the disruptions over the last thirty years, some features of the Cambodian land tenure system have remained unchanged, principally the perception that the king is the owner of the land, who gives the right of use to the peasants, and also the related idea that each person is entitled to the amount of the land that he/she is capable of farming. These two perceptions still prevail nowadays among the peasants of the rice-growing areas and they still have an impact on the enforcement of the land laws. On the other hand, other features of land tenure are evolving rapidly. The emerging population pressure in some areas is responsible for new attitudes towards land. The Cambodian population is now (2004) close to 12million unequally distributed on a small national territory of 180,000 sq. km. The central plain, the river banks and the border of the Tonle Sap lake have the highest population density (200 to 500 inhabitants per sq. km, 1993). The southeastern provinces are the most populated regions and to a lesser extent, the northwestern provinces (with a richer soil). By contrast, the forested plateau around the central basin and the northeastern part of Cambodia are very poorly populated. The perception, shared by many Cambodians and foreigners, of abundant fertile land that only needs a larger work force is nowadays outdated—if it has ever been true. If the problem is not addressed with a strong policy encouraging the diversification of economic activities in the years to come, it will become one of the most serious challenges faced by the government. On the other hand, the land market has developed after the official de-collectivization process and the redistribution of land lots on an individual basis in 1989. It had first begun to develop under the Lon Nol regime (1970-1975) and was then suspended during the war. De-collectivization itself has led to a wider distribution of land throughout the territory: even uncultivated and uninhabited land is now said to belong to a landlord, as a result of a process of appropriation and concession grants that I will describe below. I will also describe the situation regarding land tenure in the 1990s (access to land, laws, conflicts) and I will show how the present difficulties are partly a result of the past situation but also are caused by the fact that the transition between the old and new systems was not done smoothly.

  • 3 I went to the following villages: in Battambang province: village of Dong (district of Banan, commu (...)
  • 4 I carried out a part of this research when I was a member of a multidisciplinary mission of the Inv (...)

3This article is based on field research conducted in November-December 1993 in the provinces of Battambang, Kandal, Kompong Cham, Kompong Speu, Prey Veng and Pursat. It included interviews with peasants working individually or involved in a development project (such as irrigated areas). I also interviewed peasants involved in land conflicts;3 officials (ministry of Agriculture in Phnom Penh, provincial offices of Cadastre, provincial offices of Agriculture, heads of communes, heads of villages); members of local organizations of rural development; international staff of NGOs working in programmes related to agriculture (mine clearance, rural development); staff of international organizations (High Commissioner of Refugees, United Nations).4 Around forty interviews were conducted in the Khmer language (or in French/English with the international staff) by the author. The possible superficiality of the field research (which lasted a few weeks) is balanced by the great diversity of the individuals interviewed and the places visited, representing a large range of the agro-ecological regions of Cambodia which have different kinds of land issues. The field research was completed with an analysis of the Khmer law texts related to land tenure.

Monarchy and Peasantry: Land Tenure and Traditional Farming

4The pre-colonial conception of land tenure still exists today although it is mixed with other perceptions, as I will show below. The pre-colonial view of land tenure says that whoever farms the land has the right to possess it, but is not a full owner. Indeed, the farmer who leaves the land does not have right to it any more. This specific relationship between Cambodian farmers and their land was made possible because Cambodia did not experience significant population pressure until the end of the Pol Pot regime. Moreover, the farmers’ “relaxed” attitude towards the land should be understood in reference to the relative weakness of village social structures—in comparison with Vietnamese villages, for example. The sense of absolute property did not exist as it does in some other rural traditions in the world. On the contrary, the Cambodian peasantry’s attitude towards its land plots is characterized by a relatively high comfort level with relocation and a relatively low level of work investment.

  • 5 Ghouls (priey) are also responsible for such punishment, for example, when a road cuts into their o (...)

5For the king is the true “master of the land and the water” (mchah teuk dey). During the Angkorian period, he was the go-between for the human and the supernatural worlds and was as such the protector of the kingdom, responsible for its integrity. Before 1970, rites were performed in which the king acted in a symbolic manner as the owner of the land. It was said that he “ate the kingdom” (saoy reach). These rites resumed after 1993. Another force, more familiar to the everyday life of the peasants than the far-off royal palace located at the Four-Faces place in Phnom Penh, also had claims to ownership of land. These are supernatural beings called neak ta, which occupy trees, rice fields and villages. No one can forget their presence without risking punishment. For example, if one violates the rule of the land use (excessive clearing, tree cutting without asking for neak ta’s permission, construction of a fence)5 a neak ta can make one sick. These tutelary supernatural beings are organized in a strict hierarchy. They control and protect villages or parts of territory as large as an entire province. At the village level, when the ancestor who has cleared the forest and established the village dies, he/she is often set up as the protective supernatural entity of a clearly delineated land plot. Neak ta are nonetheless true inhabitants of the villages and human beings must deal with them in their everyday lives.

  • 6 The kind of rice farming currently in use is principally a rainfed lowland rice farming.

6Cambodian farmers, then, were symbolically the tenants of the land of the kingdom and had usufruct of it. The traditional farming system was based on small rice-fields directly farmed,6 with small vegetable gardens around houses. In 1930, 88% of the owners of rice fields possessed from one to five hectares of land (Henry 1932, quoted by Tichit 1981). The figures remained stable during the following thirty years (85% in 1962). But a few years later, an evolution towards tenant farming is discernable in the poorest areas (Tichit 1981). The configuration of land plots was and still remains very typical: small rice fields, spread out and well-separated from the neighbouring ones by small dikes on which sugar palm trees (thnaot) are planted. There is no collective ownership. As early as the pre-colonial period, a clear difference appeared between the relatively poor rice field areas and the richer soils of the banks and back slopes of the Mekong and Bassac rivers. The problems concerning land tenure were more evident on the latter because population density was higher, the farmers were more stable and their production devoted a larger proportion to crops which gave higher incomes than rice, like vegetables, beans, maize, tobacco, mungo beans, peanuts... A portion of these crops was also sold. The population settlement on the river banks offers characteristics which are different from those of the rice field areas: the early inhabitants were businessmen of Chinese ethnic background who married Khmer women and began to settle down (at the end of the 18th century for the Mekong river banks). Large concessions of rubber tree estates on red basaltic soils in Kompong Cham and Kratie, producing rubber-latex of very good quality, appeared in the 1920s. They were also run by foreigners, that is to say, by French people. They could be changed into true freehold, and European banks took the opportunity to own estates in Kompong Cham. The laws which were applied to the estates lent as concessions by the Crown during the French Protectorate were in fact laws governing the foreigners’ rights in Cambodia, and particularly the laws meant to protect the colonial interests; they were not, strictly speaking, land laws. For example, a law enforced in 1926 forbade any foreigners (other than French citizens or subjects of French protectorates [Indochina]) to acquire land through the concessions system.

7In pre-colonial Cambodia, what kind of right did farmers have over land which they did not own although they could use it? The Cambodian codes (kram), which bring together the legal texts concerning land tenure, tell us that the right to land was conditional on the fact of farming it. If farmers discontinued farming for more than three years, they did not have rights to it any more. This law was enforced all over the country, whether on rice fields, river banks or even on housing land. In this way, the king, probably motivated by fiscal considerations, aimed at encouraging farming. Nevertheless, farmers enjoyed extended rights over the land, for example the right to inherit, donate, sell, mortgage, lend and rent it. All these activities and more were taken into account in the legal texts, as for instance the protection against neighbours’ encroachment, etc.

8The traditional system of permanent usufruct allowed the farmers to hand over to their children the right to use the land. The division of the parents’ land was often made when they were still alive, after the children were married. The land was equally divided between sons and daughters. However, matrilocality was the general rule. The new son-in-law came and lived with his wife’s family. During a three-year probationary period, he had to help his father-in-law and was then able to build his own house after clearing new land if necessary. So, it was more useful to dower the daughters with land. On the other hand, parents were more generous with their son or daughter who would stay with them and take care of them when they grew old. The youngest daughter generally took on the duty of taking care of the parents. The principle of undivided ownership which would consist of keeping the family patrimony intact did not exist. These general trends still prevail nowadays although they conform to new conditions. That is to say, when the elder sons and daughters have married and been given a land plot, there is sometimes no more land to share among the younger children. So that the younger married couples have to settle wherever there is a land plot available (or a forest to clear), whether near the bride’s family or the groom’s or even at a new place. Information gathered from field research as well as demographic data lead me to think that an internal migration towards the less populated areas started before the Pol Pot regime. These characteristics of Cambodian land tenure have often been perceived as ambiguous by Western observers. They still prevail in the present land laws, at least until the recent law promulgated in 2001.

9But this relative freedom with regard to land did not prevent attempts at registration, carried on for fiscal reasons. Kings before the French protectorate (1863) had instituted a tax on agricultural yields—but not on land itself, in conformity with the Cambodian perception of land ownership. This tithe (1/10th of the yield) was introduced as the rental price. The king’s envoys (not the local masters) went from village to village in order to receive payment after each harvest. So the Cambodian codes made provision that all cultivation had to be registered. Land plots had to be metered and demarcated with stakes, when they were sold. The police prefect, local authorities and officials of the Governor’s office were responsible for these operations (Kleinpeter 1937). In fact, peasants often neglected this duty and concealed their incomes, sometimes in collusion with the royal envoys themselves. When the king felt that he had to strengthen the tax control process, he first launched operations on river banks areas, for their agricultural potential was higher. A royal ordinance drafted in 1877 strongly reminded the Cambodian subjects that the river banks were a royal property and were just rented (and not given) to their occupants. In the middle of the 19th century, the 10% tax was supplemented by a new annual tax which was proportional to the cultivated expanse.

Establishing Ownership: Land Reform under the French Protectorate

10Land policy in the second part of the 1990s is influenced not only by the traditional practices that I have described, but also by the laws implemented under the French Protectorate (1863-1953). The present government is once again trying to register land plots throughout the country and is working again on a cadastre project, with German assistance (GTZ). I will briefly outline the attempts of the French Protectorate to organize land tenure in Cambodia and the reasons why they failed.

11As the Protectorate extended its control over Cambodia with the establishment of the Civil Service and the Registrar of Civil Status, and the creation of a Civil Code in 1920, it also needed to establish a general system of land registration in the kingdom. This concern was mostly motivated by cultural and “civilizing” reasons because the colonial power thought it had the responsibility for establishing the principle of private property in conformity with the Napoleonic Code in order to protect landlords’ rights and prevent land conflicts. A covenant signed in 1884 paved the way for a change in the traditional land tenure system. It stated that Cambodian soil was no longer the exclusive property of the Crown. It planned a real estate endowment for the royal family, as well as the establishment of a state estate, a reserve estate and a so-called heritable estate which was constituted with the farmers’ land plots. In order to implement the allotment, the covenant envisaged a general census of the farmed lands which would permit the establishment of private ownership, at least in the so-called “usefull Cambodia”, including the border of the Tonle Sap, the central plain of the Four-Faces and the southeastern provinces. This covenant, as well as the 1902, 1908 and 1912 royal ordinances with similar objectives (but different means), were never implemented.

12There are three major reasons for this failure. First, the decrees that were promulgated successively took Cochinchina for a model. But the social and administrative situation in Cambodia was very different at that time, although the two countries were neighbours. Cochinchina experienced great demographic pressure; its people were more stable on their land, and the system of land tenure had been in use for a long time. The second reason is that Khmer peasants did not understand the purpose of the new measure and lost interest in it because it did not bring any substantial change in their land use. Finally, the land registration operations depended upon the local Cambodian civil service which was then in its infancy. For example, the creation of communes (khum) as administrative units occurred as late as 1908.

13Nevertheless, there were major exceptions to the whole picture of land use. Three western provinces (Battambang, Siem Reap and Sisophon—now integrated into Banteay Meanchey province) were given back to Cambodia by the Siamese kingdom in 1907, after being occupied for years. But they remained under the Siamese land law which had begun the land registration process much earlier. The rich soils of the river banks and the land cultivated with crops other than rice (chamcar) remained under the close control of the authorities for financial reasons. Various royal ordinances and ministerial bills tried to remove these land lots from the general land registration system because the ultimate purpose of registration was the establishment of permanent private property which would have deprived the monarchy of income generated by rentals.

14On the eve of independence (1953), the French Protectorate eventually learnt from its previous failures and established a land tenure system, perceived as transitional and supposed to fit well with the disadvantaged social and technical conditions. A decree signed by the Gouverneur Général of Indochina in 1925 fixed the legal framework of land use throughout the entire Indochinese Union (Union indochinoise). Then there arose the idea of creating a cadastre that would help in definitively establishing a system of private property. Registrations began around 1933, using the technique of aerial photography. An observer (Kleinpeter 1937) described this technique as precise, quick and cheaper than surveying. Moreover, the use of aerial photography was easy because Cambodian topography was flat. Nevertheless, priority was given to land with high commercial potential like the pepper plantations of Kampot, the rubber plantations of Kompong Cham, the central province of Kandal (including Phnom Penh) and the Mekong river banks.

15At the same time, land registration using techniques of land measurement was continued. The process was simplified (although still meticulously done) and was free of charge for the farming land plots under 10 hectares. The boundary marking committees were presided over by surveyors from the zoning ordinance administration (Cadastre) and included representatives from tribunals, district (srok) administration and commune (khum) administration who visited each land plot. Minutes of the registration were made public. Then a second administrative committee began the procedure of the final approval of title deeds. Contrary to previous attempts, this new method did not require any action by the farmers so the acquisition of the land title was almost automatic. The only condition was that the land be farmed for five successive years. From 1928 when the first tests were done, until 1936, 572,000 land plots, representing 605,700 hectares were allegedly registered (Kleinpeter 1937). In comparison, the Cadastre Department reported that it issued 350,000 land titles during the 1989-1992 period, or an average of 90,000 or 100,000 land registrations per year compared with an average of 71,600 registration operations during the 1928-1936 period.

16The end of the Prince Sihanouk regime and the pro-American Lon Nol regime (1970-1975) saw the growth of socioeconomic inequality and the deterioration of small farmers’ access to land. Rich civil servants and businessmen, who had not been interested in accumulating land capital before, began buying large amounts of land, notably the newly available “red soils”, the Battambang rich rice fields and the irrigated areas (Tichit 1981). Then the tenant farming system developed. Peasants became increasingly indebted and were often forced to sell or mortgage their land; the Khmer Rouge propaganda would make use of the poor farmers’ discontentment as early as the 1960s and it would help the guerillas in gaining progressive control over the country and in establishing farmers’ cooperatives in 1973. But the beginning of the war sharply worsened the land situation. The United States bombings during the Viêt Nam war spread in 1969-1970 over to the eastern Cambodian provinces and killed from 200,000 to 400,000 persons. Refugees fled the countryside and rushed into the cities, notably the capital Phnom Penh which became overcrowded. Insecurity caused by the proximity of the Khmer Rouge guerilla battlefront and by highway robber groups lasted until the 1990s. The land plots located far from the houses were abandoned and the farmers preferred to cluster along the roads. These regroupings started during the colonial period, when the nationalist guerillas (Khmer Issarak) fought against the French rule. The policy of consolidating the population aimed at controlling as well as protecting the scattered inhabitants of isolated hamlets. Later, under the Sangkum regime, Prince Sihanouk also implemented the same policy when he tried to sedentarize the ethnic minorities of the northeastern provinces of Rattanakiri and Mondolkiri.

Two Successive Collectivizations (1975 and 1988)

The Agrarian State of the Khmer Rouge Regime

  • 7 Recent research give a death toll of nearly two million people during the Khmer Rouge regime. This (...)

17The Khmer Rouge regime (1975-1979) is a radical break in every area. The brutality of this totalitarian regime and its desire to have total control over the population and to achieve its ideological objectives, notably in the area of farming, make this regime an almost unique case in human history. Large parts of this dreadful history strewn with corpses7 are now well known, especially the huge deportations which aimed at profoundly destabilizing social organization. City dwellers and everyone living in areas which were not under the Khmer Rouge control prior to 1975 (they were called “new people”, pracheachun thmey) suffered most heavily from the regime. The cities were emptied of their inhabitants. Large-scale deportations brought hundreds of thousands of people to the rice fields because the improvement of rice production was the first goal of the Khmer Rouge, and this had to be achieved through collectivization of capital goods and the abolition of private property. As for the “old people” (pracheachun chah) –those who lived under the Khmer Rouge administration prior to its victory over the Lon Nol regime, they often stayed not far from their native village. Young people were organized in working mobile squads (kong chalat) and worked on the big irrigation projects all over the country. From 1978 on, this work was concentrated in the northwestern provinces (Pursat, Battambang, Siem Reap) and Kompong Thom (Grunewald 1992).

18The landscape was thus permanently reshaped by the frenzied policy of rice production. Its major creed was the improvement of the traditional production and outputs and the new standard was three tons per hectare and up, and two—and even three—crops a year. Most of the efforts concerned the development of an irrigation system which would make possible a dry season crop and thereby improve the system of cultivating rice which was based on a rainy season crop. An initial assessment of the efficiency of these massive efforts was made in 1981 (Martin 1981), although it does not cover the whole Cambodian territory, its careful attention to detail makes it very valuable. It shows that while “spectacular” (Martin 1981: 23) efforts had been made to build dams and dykes and to dig reservoirs and canals, the outcome was disappointing and disproportionate with the human sacrifices involved. At the national level, there was no increase in the amount of cultivated land, and much of the public works fell apart and had to be rebuilt several times because they had not been designed by skilled technicians. Some of the dams are nevertheless still working. Production levels were barely improved in relation to the national average. Innovations were introduced with no other rationale than the desire to erase the past. For example, some land plots were abandoned and others, sometimes less fertile, were farmed instead. The rice farming land was reorganized in regular 100 meter-long squares and the destruction of the small dykes separating rice fields was still visible ten years after. The transplanting of rice seedlings in staggered rows was abandoned for linear transplanting, supposedly imitated from the Chinese model. But this technique proved to have a poor output when applied to Cambodian seedlings (Martin 1981: 21).

19All these comments should of course be nuanced. They should take into account the chronological steps of the organization of the totalitarian power: it did not uniformly impose its rule and the military-administrative areas which composed Democratic Kampuchea kept some autonomy. The Eastern Zone cadres (Prey Veng, Svay Rieng, Kompong Cham, Kratie) were more moderate in enforcing orders and were therefore purged. The new government installed in 1979 with the help of the Vietnamese was composed of survivors of this purge (Heder 1980) and was mostly influenced by the Vietnamese political framework.

The Solidarity Groups (1980-1989)

20When Vietnamese soldiers invaded Cambodia at the end of 1978, they found a lawless and disastrous situation. When they had fled, the Khmer Rouge troops had pushed the civilian population which was still under their control towards the Thai border and the Cardamom Mountains (the western forested mountains). Freed Cambodians begun to search for their family members from whom they had been separated. Some of them still began to prepare their escape to Thailand or went to the Thai border in order to get international food aid. But most went back to their native villages or the villages where they lived before the Khmer Rouge regime. The city dwellers were the most decimated and also went back to the cities when that was permitted by the new government. The great majority of the present inhabitants of Phnom Penh settled down in the capital after the fall of the Pol Pot regime. The peasant migrations to the cities was either permanent—and the biggest flow occurred in the resettlement period of 1979-1980—or lasted a few months, when the food supply dwindled in the countryside at the end of the harvest year. Then, the Cambodian farmers went to the cities to work as rickshaw or motorcycle-taximen. This time-honoured seasonal migration seems to have decreased recently (1995-2000) and has been replaced by a more permanent and massive migration of farmers towards the cities, producing an underclass of construction workers and beggars. A second type of permanent settlement developed in areas where natural resources were abundant, as in the case of the Tonle Sap Lake which was full of fish. The population density around the so-called Big Lake dramatically increased in the 1980s whereas earlier migration had only been seasonal (for the making of the prahoc, the popular fish paste). But let us come back to the 1980s, when the new Popular Republic of Kampuchea was put in place and carefully organized and the rural population was slowly reinstalled in the rice fields.

21The policy of “Solidarity groups” (krom samaki) was established in 1980. Its objectives were to compensate for the paucity of the material that is needed for farming (that is, land, farm animals, seeds, fertilizer), to help the poorest people like widows and, finally, to contribute to the “revolutionary effort”, by setting aside a portion of the rice crop for the military and civil servants. The krom samaki brought together between 10 and 20 households. They were classified into three categories according to the level of collectivization of work and land use. In all cases however, draft animals remained private property—an informal arrangement which proved to be crucial some years later, when the land was de-collectivized, because the amount of land allotted to each household was calculated according to the number of draft animals owned. The type 1 krom samaki was a land farmed collectively, whose yield was shared according to the amount of work done by each member. The owners of draft animals got an additional share per head of animal. The type 2 krom samaki had a half-collective and half-private organization. Major farming work was done in common and the daily care of the rice fields and the vegetable gardens, as well as the harvesting, was done on a household basis. Lastly, the type 3 krom samaki was very similar to traditional household farms.

22In many areas, the krom policy was not popular and not actually implemented and, as an old farmer explains it:

  • 8 This village was located in a district where the average population density was 100-200 inhabitants (...)

After a while [we understood that] the hard workers got the same share as the lazy people. The households with lots of children had lots of yield shares. It was not fair. Lot of people showed a medical certificate and pretended to be sick. But when the time for sharing the yield came, they were not sick any more. So we shared the fields in 1987 (Okrang village, Ombel, Kompong Speu).8

23The land plots were discreetly shared among the villagers before the official de-collectivization of 1989. In some places I visited (Kompong Cham, Chamcar Loeu district), the krom samaki never even existed. In other places, the allotment was organized step by step from 1981-1982 (Kompeang Svay village, Pursat; Nikum Knong village, Battambang) to 1987 (Okrang Ombel village, Kompong Speu). At the national level however, type 2 krom samaki were always the most common type (more than 50% from 1980 to 1983; 80% in 1986). The number of highly collectivized type 1 krom samaki progressively decreased although the government tried to re-impose them as the major farming system around 1983-1984 (Grunewald 1993: 58).

24The traditional opposition between rice fields and chamcar (river banks, vegetables) also played a role. The policy of very collective krom was implemented mostly in the poorest areas (rainfed lowland rice fields of Kompong Chhnang, Takeo and Kompong Speu), whereas the river banks (Kandal, Kompong Cham) often had the more supple system of type 3 krom samaki (Grunewald 1993: 57). The organization of power also played a role in the differential application of the krom samaki policy. The village is the lowest administrative level and therefore poorly controlled by the central power (troops and Vietnamese advisors did not “come down” to this level). At the village level, the authorities were organized in popular committees whose political backgrounds were far more heterogeneous than at higher levels. Moreover, committee members were co-opted by the “masses” and not appointed by the Party, which only appointed heads of districts (sometimes) and heads of provinces (always). This political policy gave some freedom to the farmers in organizing the production process. This freedom was abolished in the 2000s because all the heads of villages have become local representatives of the central power and, from now on, play an important role during the elections, as the party’s lowest-level propaganda agents.

New Trends since De-Collectivization

Allotment of Land Plots

25Field work in several provinces suggests that the allotment of land plots has been relatively fair. It was done with the goal of respecting the farmers’ practice of land use: land plots of different types of soil available in the solidarity groups (housing land, rice fields, chamcar, proximity to sources of water) were allotted to each household according to size. A share was also given to those who did not participate in farming, such as members of the military and civil servants. But the most important criterion was the size of the work force available in each household. For in the minds of peasants as well as policy makers, it was not necessary to obtain more plots of land than one could actually farm.

26The abolition of the solidarity groups ended experiments with collectivization. Peasants do not have a positive enough memory of the krom samaki to continue with it, even when it would be essential, for example in the management of irrigated areas. So, for example, in Anlong Riet village (Prey Veng), the irrigated system supported by a French NGO has management problems. Only 49 out of 79 households that own a land plot inside the area have agreed to join the association in charge of water management. Everybody mistrusts the system of sharing work and fees. Some owners, for example, obey the rule and clean out the canals every week while others do not. And everybody prefers to operate the water pump individually for his/her own parcels of land rather than to use it collectively although the latter would be less demanding in water and petrol. A second example is provided by the warping canals (prek). They are poorly looked after or simply abandoned because their maintenance calls for a collective organization which is neither provided by the state (administration at the district and commune levels does not work well) nor by the individuals who only farm their own land running along the prek (I observed three cases of abandoned canals in Samrong Thom commune, district of Kien Svay, Kandal).

  • 9 This figure includes widows (10.8?%) and divorced and separated women (4.2?%). The last category is (...)

27Several factors allowed a differential access to land which produced a social stratification that only got stronger over time. Firstly, female-headed households got a smaller piece of land because the allotment was based on the number of family members and also because their social position was lower than the men’s, so that they could not make their voice heard equally. Moreover, men are traditionally in charge of ploughing, a major part of farm work. This is not a small problem if one considers that the proportion of Cambodian women heading a household ranged from 30% to35% of the Cambodian population in 1990 (Sonnois 1990: 1) with figures as high as 40%-50% in some villages (Ledgerwood 1992: 7 and field research). It is a consequence of the abnormally high male death rate due to the war; the percentage then decreased but still remained high in the 1990s (15% according to the latest population census in 1998) (National Institute of Statistics 1999: 14).9

  • 10 The bao is a local weight unit, equivalent to 80-100 kg.

28Secondly, the ownership of farm implements in areas where there was still some land available was crucial. In the highly populated areas that I visited, where allotted land plots were small (around one hectare per household), farmers seemed to be more satisfied with the allocation. On the contrary, in areas where some large pieces of land were still available, people were more often unhappy because the differences between the villagers were more visible. For example, in the north of Battambang province (Nikum Knong village), the four farmers who owned tractors could extend their property up to 20 or 30 hectares. The first private tractors were those of the Ministry of Agriculture, resold to private owners during the de-collectivization process. Social disabilities, then, work cumulatively against female-headed households because they have fewer draft animals than others and thus are less capable of obtaining large amounts of farming land which they could not, in any case, farm by themselves. The salary of a ploughman is surprisingly high in Cambodia, from 30 to 35 thang (which means 720 to 840 kilos) of unhusked rice, according to a provincial representative of the Women’s Association of Pursat province. Finally, the political and military instability allowed for the development of specific strategies in order to get land. The first one was developed in front line areas of the Khmer Rouge guerrilla. Until the total collapse of the Khmer Rouge guerillas in 1998, many villages, particularly those located in the northwestern part of Cambodia, were not far from these frontlines (a few kilometres). The guerrillas had de facto control of the land and demanded a “tax” to let the peasants farm it (Nikum Knong village, Battambang). This way of getting land was not safe because the farmers had to rub elbows with the fighters who themselves needed rice fields for their military movement so that the permission was always temporary. The Khmer Rouge guerillas also provided “protection” in exchange for a large part of the yield (from 3 to 5 bao per hectare10). Similarly, the farmers living near the forests were used to cut wood and this activity was also “taxed” by the guerrillas (Chamcar Loeu district).

29The gradual sale of state properties when public administration began to decline also contributed to the unequal access to land. A text called “Political directions concerning the public administration and land use”, passed in the Council of Ministers in June 1989, recalled that each village should keep free land for future needs (for example, settlement of new households, use by poor households). Sparsely populated villages had to keep 5 hectares and densely populated villages and towns, 3 hectares. But in most cases, this land disappeared as soon as the official allocation was made; it was kept by the local representatives. Several scandals were made public, particularly in towns—where the price of the land dramatically increased after 1993: building and housing land of the public domain were sold and the money was more or less fairly shared between the staff of the concerned ministries. This was made legal by a law passed in 1992 which specifies that “public property which does not have social use any more can be converted into private property and sold” (article 17).

Unequal Access to Land in the 1990s

30One of the most eventful facts of the de-collectivization process was that all good farming land, officially remaining free, was said to be occupied in the 1990s. This was a total break with the previous conception of land tenure which had prevailed during the Sangkum regime of Prince Sihanouk, when the population density was lower. During my field research, there was no village where I was told that any farming land of good or even fair quality was available and could be allotted to new households, for example.

  • 11 The chi is a weight unit, equivalent to 3,37 grams. One hun is the 1/10th part of a chi.

31In the village, we only shared the rice fields, not the chamcar. The 2 hectares of chamcar remained collective property. Those who could farm it, just borrowed it free of charge. We also put cows on it. But in 1992, we sold this land for one chi and two hun [of gold11] and we villagers, all shared the gold (head of village, Okrang Ombel village, Kompong Speu).

32Some free land does remain in Kompong Speu province, but it is far from roads, schools and health centers, and the area is insecure and unhealthy (malaria in forests).

  • 12 The usual measure unit is the thang which is equivalent to 2 bags, i.e., 24 kg.
  • 13 The best quality wood is a scented wood. The oil extracted from it (khleum chan, “heart of the moon (...)

33So, if a person wants to extend his/her landholding, he/she has three options: the clearing of forests, the clearing of mines and the purchase of land. During the 1980s, in areas near the border of the forest, like the district of Kravanh (Pursat), the clearing of forests was still the most common way of getting new land (when the official allotment did not meet all the farmers’ needs) or of increasing the output which was reduced due to the overexploitation of rice fields (60 bags12 of rice was the yield on 30 ares of newly cleared land versus 40 bags on old land). New villages, which split off from old ones, are thus built. In addition, in this area of primeval forest with high priced wood13, woodcutting provides an important supplementary income to peasants living at the border of the forest. They are included in a reciprocal system in which they work for groups of businessmen, themselves associated with soldiers who provide protection against gunmen as well as information about mined areas for the encampments in the forest. Farmers work in groups of 20 to 30 near the encampments and then sell their harvest on site before walking back home.

  • 14 The most heavily mined srok are Thmar Pouk and Serei Sophorn (Banteay Meanchey province), and Banan (...)

34The mining of the land by the regular army of the successive regimes as well as the guerillas since the beginning of the 1970s has considerably reduced the area of cultivation. The most heavily mined provinces (Battambang, Banteay Meanchey, Siem Reap, Kampot14) were those which had the best soil. Estimates vary from ten million (Aitkin 1993) to five million landmines (interview with a member of the Cambodian Mine Action Centre, CMAC) and the number of anti-tank mines and non-exploded bombs must be added to this figure. They all cover an area estimated by the CMAC at 3,200 sq. km. Accidents have decreased since the end of the armed conflict but they still remain high, from 300 to 700 monthly in the 1980s (Aitkin 1993) to 100-200 victims (Gould 1993) in 1993. There were still about 100 known accidents every month in 1998 (Landmine Monitor 1999, quoting CMAC), peaking during the dry season –these figures include deaths as well as injuries which most often lead to amputations of legs. This considerably weakens the rural work force because the great majority of the disabled are young men (Gould 1993).

35Mine clearing is carried out by trained teams of NGOs and the Cambodian Mines Action Center. But private mine clearing is also performed by the villagers themselves or by the military (for example in Dong village, Battambang). According to an interview with a mine-clearing expert working for Mine Advisory Group, a Western NGO, these mine clearances are done by driving a tank on the mined land. They cost the owners of the fields 120 dollars per hectare. But this method does not offer guarantees because it is too rough and not as precise as the “frying-pan” method which tests every square centimeter of land. In any event, the CMAC, which was created in 1992 by the Cambodian government and is funded by international donors, is the biggest mine clearing organization but was still unable to clear mines from more than 106 square kilometers of land in ten years (1993-May 2003). One can thus have an idea of the huge work still to be done. The government was concerned that the cleared land would be handed back to its original “owners”, and in 1999 it established a Land Access Planning Unit to prevent the newly available land from remaining in the hands of influential individuals in the administration; the media had denounced such practices.

  • 15 In another village (Nikum Knong, Battambang province) the rates are 3 bags of rice paid back for 1 (...)

36There are two very different types of land purchase. The first is practiced by the richest farmers who want to extend their land. The second is the purchase of land for commercial or speculative purposes by businessmen. At the time of my field research in 1993, few farmers had sold their land and definitively left their village. Nevertheless I met several cases of people who had in each village I visited. Moreover, the socioeconomic status of many villagers was precarious. If they fell sick or had an accident, they could easily run into debt with moneylenders whose interest rates were very high. In Kompeang Svay village, Pursat province, one had to pay back 360 kilograms of paddy (15 thang) for 100 kilograms of paddy borrowed. The rate was lower when the payback was done in money but it was still high, between 15% and 20%.15 Farmers who were too deeply indebted then had to sell their land and become day laborers in their villages or go to the cities. The rural exodus has dramatically increased since the end of the 1990s and it has led to the development of shantytowns and slums in Phnom Penh, as well as the development of bands composed of several hundred street children whose parents could no longer support them.

37But the sale of land can also be an economic strategy for farmers, when they need cash in order to start a small business, build a house, buy draft animals or even a motorcycle. In areas where the price of land is increasing, farmers are easily convinced by the salesmen.

The broker came to our village in order to find land to buy. He then put together one hectare by buying pieces of land from four neighbouring households. He paid one chi and six hun per are. It was in July [1993]. Now [November 1993], this land costs three chi and four hun and the price is still increasing. The three other families have sold their land. I also sold my land to buy cows and to create a stock breeding business. I bought new land which is farther from my house but cheaper. In our village, twenty households have already sold their land along the side of the road to the same buyer (old farmer, Okrang Ombel village, Kompong Speu province).

38Among the buyers are substantial Cambodian or foreign businessmen. Foreigners buy the land with the help of a figurehead.

39In addition to purchasing land, businessmen can rent state concessions for large-scale commercial crops. For example, I met a civil servant from the Ministry of Trade who rented 100 hectares for a period of 45 years. At the end of the rental period, he will have to give the land and equipment back to the state. He is planning to enlarge his cashew nut tree plantation to 200 hectares. He obtained the remaining 100 hectares by clearing the forest (Chheu Prey district, Kompong Cham province).

40Some social groups are becoming particularly vulnerable in this new context of increasing land scarcity. In addition to the indebted peasants mentioned above and the farmers despoiled by powerful groups or individuals, people who were displaced during fighting or flooding as well as the refugees in the camps at the Thai border were the first victims of the change in land tenure.

41At the end of 1989, heavy fighting led to the flight of tens of thousands of people from their villages. As time passed, the insecurity continued and 190,000 people settled down along the roads with their meagre belongings and a plastic tarpaulin as a roof. There they waited for better days. In 1992, 60% were resettled either in their former villages or in new locations. In addition to the land mines which prevented 45,000 of these displaced persons from farming their land (1992), there was also the problem of right over the land because the existing land law, based on the extended usufruct principle, specified that the land had to be farmed without interruption by the usufructuary (Teas 1993). In particular, the military coming from the densely populated southwestern areas who fought in the northwestern areas sometimes took this fertile land over by cultivating it and did not want to give it back. Negotiations were held between the government and international organizations whose duty was to protect and support displaced persons. The problem was easier to solve when the displaced persons continued some farming activity on the land (it was sometimes risky to do so) by leaving their villages only when the fighting occurred and returning to farm as soon as the situation calmed down. The number of despoliations is increasing as time passes and tens of peasants often demonstrate peacefully in front of the National Assembly building in Phnom Penh.

42Managing the return of the 365,000 refugees, repatriated by the High Commissioner of Refugees from the camps in Thailand was even harder. They mostly resettled in their native provinces, in the West (one third of the repatriates in Battambang province, one third in Banteay Meanchey province, the remaining third mostly in Siem Reap and Pursat provinces). Although theses provinces had a low population density, access to land was difficult for the repatriates.

43The HCR had offered three resettlement options to the repatriates. Option A offered farm land (srae chamcar); option B offered housing land (dey phum) and a house; option C gave money. The projections concerning the allocation of land were too optimistic and very little was actually given to the returnees. Because of the ill will of the local authorities and because of landmines, option A was chosen by only 3% while 11% took option B and 84% option C (UNHCR 1993). Surveys of the repatriates show that the land was considered to be the property of its owners. (This applied to all land whether farmed or not, including the borders of the forest in Pursat, srok Kravanh.) They informally shared the forest area among themselves and prevented newcomers from clearing it, possibly by the threat of weapons. Commercial woodcutting, which is of vital interest for the landless returnees, is allowed on condition that the newcomers go deep into the forest (i.e., several days of walking, or 30 to 35 km, according to an interview with a repatriate). The newcomers who do not have relatives in the village must rent or buy land. Some of them also get rice in exchange for helping former villagers with their harvest work.

44In the best cases, local authorities gave housing land and poor quality farm land located far from the village to the returnees. In the worst cases (Tuol Ngan village, Battambang province), the repatriates were resettled in a new village built not far from the old one and decided to leave when the HCR stopped its food aid to them. When I went to this village, one-third of the new villagers had already left (according to the officer in charge of the follow-up of the support to households). They had sold their land and their house to old villagers. It thus became quickly obvious that the recipients of option C (money) would not be able to rely upon the rehabilitation program. According the HCR officer quoted above, from 20% to 30% of the repatriates had already left their resettlement area by the end of 1993.

45The problems related to land tenure that stemmed from population pressure were worsened by the political and military instability from the 1970s to the 1990s. Despite the fact that the vast majority of the Cambodian population rely upon food-producing rice fields, the problems were not addressed by the legislation that was adopted.

A New Land Law Inherited from the Past

46The current law promulgated in 2001 should help clarify the land tenure situation over the next few years. It is the result of previous juridical efforts (the 1989 law, the 1992 law, various decrees and instructions) characterized by deep ambiguity and therefore, rather ineffective. Previous land laws had been influenced by three factors that the lawmakers could not or did not know how to integrate: socialist ideology, ancient Khmer codes and French law. These laws also showed that the lawmakers wanted to take into account new situations such as the new land and real estate market in a liberalizing economic context, but the law could not openly express this will because it was politically incorrect at that time. This is why the previous land laws are so confusing and difficult to implement. I will go back to the circumstances in which these laws were written in order to clarify matters.

47The Constitution of 1989 is the benchmark of political and economic liberalization in Cambodia. Regarding land tenure, it officially abolishes the solidarity groups. However, all the natural resources of the country as well as the country’s land remain state property (article 14). The state gives land and houses to the Cambodian people “in order to live and farm there” (article 15). Nevertheless, it unequivocally forbids the selling and the rental of land to be used exclusively for fallow (article 16) with the obvious and reasonable purpose of preventing the development of the tenant farming system. The intent of this law is thus not far from the mind of the old Cambodian codes. But a sub-decree passed in the Council of Ministers a few days before the Constitution gives the right to sell and to give houses, with a 10% tax; the sub-decree could indicate ambivalence or juridical inexperience on the part of the ruling powers. The sub-decree nevertheless reminds us that land and houses remain the collective ownership of the state, standing for the people.

48The 1992 land law promulgated by the National Assembly offers more freedom regarding land transactions. It stipulates again that land is state property and does not acknowledge titles to landed property prior to 1979 (article 1). The Cambodian People’s Party has always been tough on this point when conflicts occurred between old and new owners, as the example given above shows. This firm policy has succeeded in calming down the innumerable land conflicts which have mushroomed in the shattered land context of the post-Pol Pot era. According to this law, land is divided into three categories: the “ownership” (kamaseth aekchun ptoal khlouen mneakmneak) of housing land, “possession” (phokeah) of farming land and the “concession” (sompatien) of land larger than 5 hectares. The land thus belongs to those who farm it. The owner has broader rights upon his land as many transactions are allowed such as handing over by inheritance, mortgaging, etc. However, rights are still strictly restricted because the question of selling is eluded and the owner must farm or inhabit his/her land to maintain his/her rights on it. Ownership is withdrawn from the owner who does not stay/farm his/her land or does not pay his/her land tax during five successive years.

49The text of the law is further complicated by the French influence which is seen in the various words used for the “owner of the land”. He/she is sometimes called “possessor”, or “owner” or “user”. It shows that the lawmakers always hesitate between the Khmer perception of the land law (i.e., the broader right of access to land) and the French perception of the permanent ownership. The ambiguity is so deep that the lawmaker needs to explain that “possession means that the possessor (phoki) has an exclusive right on his/her property and must fulfill all the duties of an owner regarding his/her property” (article 61). In conclusion, the text of the law is difficult to understand and implement, and a lawyer wrote that it was an “unattainable and superficial” law because it tried to cover too many cases (Grève 1993). But the lawmakers’ trouble that one can feel in the text of this law has another cause, which is also influenced by French law: the possibility of establishing a permanent private ownership directly depends on technical considerations, i.e., the drawing up of a cadastre and the land registration process. The 1992 law is thus supposed to be provisional and stipulates that land in usufruct will be upgraded to freehold when all the land is registered and land titles distributed.

50But the technical and administrative processes which would allow the establishment of permanent private ownership in Cambodia have hardly begun. The Department of Cadastre of the Ministry of Agriculture was created in 1989. It includes the head office in Phnom Penh, provincial offices and district offices. Cambodian people were asked to fill out a land registration form. The programme was a success as 4 200 000 forms, representing 80% to 90% of the land, were sent back to the district offices. This success shows that the cadastre has become a broadly shared concern among Cambodians. In villages where land conflicts occurred, surveying was carried out as early as 1983-1984 (Anlong Riet village, Prey Veng). Rough sketches of the land were attached to the land registration forms and signed by the head of village and the head of commune and then filed in the district Cadastre office.

51The next step of the registration programme will be the issuing of temporary land titles after the land survey and field investigation by civil servants of the Cadastre Department. But this work is progressing very slowly. Less than 10% of the programme had been carried out by the end of 1993 (6% in Pursat province for example) and, in 2000, only 15% of the land titles had be issued (Medicam 2000:3).

  • 16 For example, according to the chief of Cadastre Department of Kompong Cham province, the 65 staff o (...)
  • 17 In Kompong Cham province, in 1993, only four employees out of 65 staff of all ranks working at the (...)

52The reasons are numerous, such as the lack of trained staff16, in a country where the “elite” had suffered the most from the Khmer Rouge regime17; the lack of equipment (transportation, surveyor rings, adding machines for calculating areas); perhaps also the lack of staff (the cadastre Department employed 760 persons in 1993 of whom 500 were not technicians); the fact that registration forms are sold instead of provided without charge and bribery is rampant in all ranks of the administrative hierarchy; and finally the difficulty of surveying insecure areas. Thus, the third phase, i.e., the official registration of the land, is a long-term objective. All these problems have led to a dramatic increase of land conflicts since de-collectivization.

Land Conflicts

53Land conflicts are not new in Cambodia, as can be seen in the many provisions of the old Khmer Codes. During my field research, all the interviewees agreed that land conflicts were by far the most numerous official complaints. The provincial Cadastre office of Pursat—where land tenure is not the worst—said that about 50 complaints were ongoing. Among the associations in charge of conciliation with which I spoke, land conflicts were also the major part of their activity (88 land conflicts out of a total of 255 conflicts mediated by the Human Rights Commission of the National Assembly in two months, at the end of 1993; 40 complaints registered by the provincial Department of Cadastre of Kompong Cham with the less serious cases adjudicated at the district or village level). The number of land conflicts has increased in ten years, since in 2000, one household out of 30 was involved in such a dispute (Medicam 2000). They are sometimes violent, in a country where firearms are readily available and military groups often involved in such problems.

54In addition to the usual reasons (conflicts about land delimitations, family quarrels such as divorces) there are more specific reasons related to the Khmer tradition of access to land. Quite often, land lent to a family member is not returned to the main user when he/she wants to get it back. These conflicts mainly involve “village land” (dey phum) and riverbanks (chamcar), whose prices are higher than rice fields. But it seems that for the last few years rice fields have also been subject to conflict as their prices increased because of the development of the network of secondary roads. Finally, since 1979, new reasons for conflicts have been added to the old ones, like the serious problem of the refusal to acknowledge the abolition of private property by the Khmer Rouge in 1975. In the village where I saw this kind of conflict, many peasants were involved. The local authorities responded quickly and firmly because they were aware of the potential explosiveness of the situation and the danger of its being exploited politically during the electoral campaign.

In our village, lots of people died under the Pol Pot regime. In 1979, only four or five households came back and settled down again. The other survivors followed their wives to their native villages. But the old owners, more than thirty households, who had settled down in the West were thrown off their land by the Khmer Rouge guerrillas. They returned and claimed their former land. There were no murders here, but there were several related to the same problem in neighbouring villages. During the quarrel, a committee composed of commune representatives, district representatives and representatives of the provincial police, was created. When a former owner came and ploughed his land, he was caught, his hands bound and he was brought to the police office. Then the former owners used the troubles caused by the elections in June [1993] to get their land back. The FUNCINPEC [Prince Ranariddh’s party] had promised to give their land back to the old owners, but since the Cambodian People’s Party won the elections, the problem is solved (a witness, Preah Khae village, Kompong Speu).

55A second thorny problem is the fraudulent sale of state land, that I have already mentioned. Yet a third problem is caused by the Khmer Rouge guerrillas who fuelled the conflicts (several cases in Kompong Cham, for example) by allotting land that had already been distributed during the de-collectivization, to new people in the areas they controlled. After the end of the Pol Pot guerrillas in 1998, other armed groups have despoiled peasants of their land by threatening to use force and by taking advantage of the confusing land law.

56The phenomenon of landless peasants has increased during the last few years and in 2000, one household out of six (17%) had never owned land or had recently lost it, while 80% of the Cambodian population live from agriculture (Medicam 2000).

57In these cases, the usual mediators (old villagers, monks…) are relatively powerless to help. It seems that heads of villages have neither enough power nor legal knowledge to solve (somrohsomruel, “conciliate”) land conflicts, all the more so because they sometimes are themselves the very source of the conflict because they use their personal power to seize land for themselves or their relatives. Recourse could be had from the chief of commune, the Casdastre Department which brings the case in front of the district tribunal and, in last resort (if the two opponents have land titles), the provincial tribunal. But the complainants are not confident about these entities, especially the tribunal, because it asks for informal supplementary “fees”. In the wave of democracy which developed right after the first free elections of 1993, several new associations have made efforts to offer support to despoiled peasants. They include an association of legal practitioners in Phnom Penh; several human rights associations and a human rights committee in the National Assembly, authorized to act as mediator between complainants, on the one hand, and courts of justice or the Ministry of the Interior (in charge of the land conflicts), on the other hand.

A Historic Break and an Uncertain Future: The New Land Law and the Large Estates

58In the 1990s, land conflicts developed at the same time as sale of land by the most vulnerable part of the Cambodian peasantry. Many international organizations—which are very influential in Cambodia because their financial support is a large part of the national budget—informed by NGOs working in the field, have been concerned with the land situation and have urged the government to draft a new land law in a timely fashion. The international organizations were concerned with the drafting process for several reasons. To begin with, they called for the end of the self-subsistence economy in Cambodia and elsewhere. But the process of integrating a vulnerable economy like Cambodia’s into the international economy can be dangerous if not carefully done. Second, the international organizations wanted the development of private commercial agriculture (because the new creed of development is privatization and liberalization), notably by the establishment of concessions given to large companies in the sectors of agriculture and forestry. These factors underscored the need for a new law which would reassure potential investors and take into account the increase in the value of land caused by its scarcity in relation to population pressure, the increase in speculation, and the increase of the amount of cultivated land reserved for commercial crops.

59This law was finally voted in 2001. It corrects most of the ambiguity of the previous laws. Although many sub-decrees still had to be promulgated in order to make the law implementable, it marks the end of the pre-colonial land tenure system (in theory at least) because it clearly establishes permanent private property. However, it contains some elements of the old land tenure. For example, land can become private property if it is worked during five successive years by the person farming it. In general, access to land is linked to the system of concessions set down by the state; the so-called “social” concessions which allot land plots to the poorest households in order to build a house and engage in subsistence agriculture, and the so-called “commercial” concessions, leased to private companies which aim at developing an agricultural or forestry activity. Holdings are limited to 10000 hectares, leased for 99 years. Committees were established in order to implement this law, such as the committee for the settlement of conflicts (1999) and the district and provincial cadastre committees (July 2002). Some other working groups are still to be set up. The Cadastre committee will have to draw a cadastre following two procedures, i.e., a systematic one for more valuable land, and a sporadic one, for less valuable land.

  • 18 Among the 51 companies which obtained a concession between 1992 and 2004, for which I have detailed (...)

60The social concessions are still to be created, notably with the support of the World Bank which would design a programme to be started by the end of 2005. On the other hand, commercial concessions have already increased in a dramatic and… somewhat alarming manner if one thinks about the lack of transparency. Secondly, the social consequences of those concessions are sometimes tragic. Thus, in ten years (1994-2003), 85 agribusiness programmes were approved by the Ministry of Agriculture. They cover almost all of Cambodia, the most populated provinces (2 concessions in Takeo, 17 in Kampot, Kompong Cham and Krong Preah Sihanouk), as well as the provinces with the lowest population density (Stung Treng, Mondulkiri, Ratanakiri, Koh Kong) (see Fig. 2). Among those companies, 26 have already signed a contract (the majority did so between 1998 and 2000), renting a total area of more than 700,000 hectares; and 12 companies have already begun to work, some of them in an illegal manner, since they actually rent more than 10,000 hectares. The area of potentially rentable land is huge. In February 2003, the ministry made out a list of 40 areas, totalling 800,000 hectares (i.e., 8% of the national territory), which could be rented as commercial concessions for commercial crops such as cassava roots, cashew nuts, cotton, paddy, and tree plantations (including palm oil, coconut, rubber, teak, eucalyptus, and fruit trees.) At this time, 66 companies were already registered, mainly from mainland China and Taïwan (but also from Korea and Malaysia) in association with Cambodian partners.18

  • 19 Informations provided by Mrs Fabienne Luco, based on figures of the organizations and ministries qu (...)

61If forestry concessions are added to these agricultural concessions, 20% of Cambodian land could be considered susceptible to be rental by commercial private companies.19 Government policy limits deforestation, known to be a major environmental risk in Cambodia; this has reduced the area of forest land allotted as forestry concessions from 6,400,000 hectares in 1995 (32 companies) to 900,000 hectares in 1999.

62On the other hand, some concessions have more “social” purposes. Thus, the state maintains control over the large rubber tree plantations which, before the Khmer Rouge regime, were known all over the world for their high quality latex. In addition to these seven state plantations (32,000 hectares), there are households plantations that the government wants to develop. It has also tried to address the difficult problem of the professional rehabilitation of the demobilised soldiers by creating military development zone.

63If the phenomenon of commercial concessions continues, it will tend to drastically modify the main features of Cambodian agriculture in the coming decades. Several problems must be addressed by the government now. These include the problem of the implementation of a very new law not congruent with the traditional Cambodian view of land tenure. They also include the need to improve the technical means of land measurement as well as land management. Problems in the power structure also need to be worked out. Then, the commercial concessions are often poorly demarcated and encroach upon protected forest areas, and they also often encroach upon farmers’ land, leading to conflicts. The neighbouring farmers are in fact not allowed to farm their land (sometimes by armed guards hired by the companies), while the area in which they can gather medicinal plants and harvest resin from trees (that represents a large part of their annual incomes) is cut off by the companies. On the other hand, the rent price does not always go to the state’s coffers although the coffers are empty, because the negotiations are sometimes done in an informal manner, sometimes involving powerful government representatives at national or provincial levels. In this unequal fight between peasants who do not know the new law and continue to possess the land by virtue of farming it, and large companies backed by powerful civil servants, the peasants will not have the last word. This could lead in the coming decades to the development of an under-class of underpaid agricultural workers in a country which for hundreds of years was characterized by a land tenure system that made for a self-sufficient peasantry.

Fig. 1: Cambodia:...Fig. 1: Cambodia: Population Density by Province

Fig. 1: Cambodia: Population Density by Province

Source: National Institute of Statistics, Ministry of Planning, General Population Census, 1998.

Fig. 2: Agricultural Concessions Granted to Commercial Companies (1992-2004)


Average provincial density of population (inhabitants/sq. km)

Number of companies

Average areas in ha (minimum & maximum)

Kompong Chhnang



176,065 ha

Stung Treng



54,126 ha (7,400 to 100,852)




47,421 ha (300 to 138,963)

Koh Kong



27,120 ha (7,500 to 60,200)




20,000 ha

Kompong Thom



18,250 ha (10,000 to 26,500)

Kompong Speu


12 (including 8 in Phnom Srouch district)

11,245 (2,040 to 26,550)




7,500 ha (16,400 to 3,800)




6,550 ha (200 to 20,000)




6,400 ha (1,800 to 11,000)




5,000 ha (2,000 to 8,000)

Kompong Cham



3,624 ha (500 to 9,900)




775 ha (500 to 1,050)



Source: Information from the Ministry of Agriculture, Forestry and Fishery and the Council for the Development of Cambodia

Note: This table is probably not exhaustive as fourteen of the contracts have been canceled by the government between 2000 and 2003, because they were not legal. Moreover, the table does not list forest concessions.

Haut de page


Aitkin, Susan, 1993, Getting the Message about Mines, Report, CMAC/UNESCO.

Delvert, Jean, 1994, Le paysan cambodgien, Paris : L’Harmattan [first ed. 1961].

Forest, Alain, 1992, Le culte des génies protecteurs au Cambodge, Paris : L’Harmattan.

Greve, Hanne Sophie, 1993, Land Tenure and Property Rights in Cambodia, Report, Phnom Penh.

Gould, David, 1993, Landmine Victim Survey, Report, Cambodian Mine Action Centre.

Grunewald, François, 1992, Essai d’analyse historique de l’évolution de l’agriculture cambodgienne, Document de présentation pour le séminaire “Politiques agricoles et agriculture paysanne” organisé à Phnom Penh par le GRET et la FPH avec l’aide du ministère de l’Agriculture.

Heder, Stephen, 1983, “From Pol Pot to Pen Sovann to the Villages”, in Indochina and Problems of Security and Stability in Southeast Asia, Theeravit Khien & MacAlister Brown (eds), Bangkok: Chulalongkorn Univ. Press.

Henry, Yves, 1932, Economie agricole de l’Indochine, Hanoï: IDEO.

Kleinpeter, Roger, 1937, Le problème foncier au Cambodge, Paris: Domat-Montchrestien.

Lamballe, Patrice (traduit par), 1993, Evolution de la situation foncière durant les différents régimes en prenant l’exemple du district de Ba Phnom – Prey Veng. Transformatioon du foncier et de la société de 1965 à 1992, Rapport, Phnom Penh, GRET.

Landmine, Monitor, 1999, Cambodia Country Report prepared by the Cambodia Campaign to Ban Landmines,

Ledgerwood, Judy, 1992, Analysis of the Situation of Women in Cambodia, Report, Phnom Penh, UNICEF.

Martin, Marie Alexandrine, 1981, “La riziculture et la maîtrise de l’eau dans le Kampuchea Démocratique”, Etudes Rurales, 83 : 7-44.

Medicam/NGO Forum on Cambodia/Cooperation Committee for Cambodia, 2000 NGO Statement to the 2000 Consultative Group Meeting on Cambodia.

National Institute of Statistics, 1999, Ministry of Planning/United Nations Population Fund, General Population Census of Cambodia 1998. Final Census Results.

Robequain, Charles, 1939, L’évolution économique de l’Indochine, Paris.

Sonnois, Brigitte, 1990, Women in Cambodia. Overview of the Situation and Suggestions for Development Programmes, Report, Phnom Penh, Redd Barna-Cambodia.

Teas, Martha, 1993, Needs Assessment for internally displaced Persons in Cambodia, Report, Phnom Penh, PNUD/CARERE.

Tichit, Lucien, 1981, L’agriculture au Cambodge, Paris: Agence de coopération culturelle et technique.

UNHCR, 1993, Programme and Technical Support Section – Cambodia – Land Identification for Settlement of Returnees in Cambodia, PTSS Mission Report.

Haut de page


1 I am grateful to Mrs Fabienne Luco, associate researcher at EFEO and CKS, who offered valuable data drawn from her own recent field research (writing in progress) concerning concessions of agricultural and forest land given to private commercial companies. I am also deeply indebted to Pr Evelyn Ackerman, City University of New York, who kindly reviewed my translation from the French into English.

2 The major features of Cambodian political history are as follow: 1863 –?French protectorate; 1953 – Independance of the kingdom, the so-called Sangkum Reah Niyum regime is established by Prince Sihanouk (who gave the crown to his father Norodom Suramarit); 1970 – Impeachment of Norodom Sihanouk and establishment of the pro-American Khmer Republic; 1975 – Victory of the “Khmers Rouges”; 1979 – “Khmers Rouges” are defeated by the Vietnamese army which established a Popular Republic of Kampuchea; 1988 – New Constitution instituting the State of Cambodia as well as some political and economic liberalization; 1991 – Paris Agreement preparing the settlement of the conflict among the four Cambodian fighting factions; 1993 – Elections organized by the United Nations Transitional Authority in Cambodia. They re-establish the constitutional monarchy and Norodom Sihanouk as the King; 1998 – Death of Pol Pot in the western forests of Cambodia and end of the Khmer Rouge guerilla.

3 I went to the following villages: in Battambang province: village of Dong (district of Banan, commune of Chay Meanchey); village of Touel Ngan (district of Battambang, commune of Chrouy Sdav). In Kandal province, village of Srae Ampel (district of Kien Svay, commune of Chheu Tiel); several prek in Samrong Thom commune. In Kompong Speu province: village of Okrang Ombel (district of Samrong Tong, commune of Trapeang Kong); village of Preah Khae (commune of Kaheng); village of Steung (commune of Trapeang Kong). In Kompong Cham province, agricultural holding of 50 ha (cashew nuts), district of Chheu Prey; commune of Bok Knor (district of Chamcar Loeu). In Prey Veng province: village of Anlong Riet (district of Kompong Trabek), village of Dey Srot (district of Peam Chor). In Pursat province: village of La (Mol) (district of Kravanh, commune of Rokat); village of Kompeang Svay (district of Bakan, commune of Snam Preah).

4 I carried out a part of this research when I was a member of a multidisciplinary mission of the Investment Center of the FAO (“Agricultural Development Options Review for the Royal Government of Cambodia”), headed by Mrs Loretta Sonn and funded by the UNDP in 1993. The final report is titled Cambodia. Agricultural Development Options Review (Phase I). Sector Review. See Annex 11, Aspects démographiques et sociologiques, 40 p. and Annex 12, La question foncière, 25 p.

5 Ghouls (priey) are also responsible for such punishment, for example, when a road cuts into their own paths.

6 The kind of rice farming currently in use is principally a rainfed lowland rice farming.

7 Recent research give a death toll of nearly two million people during the Khmer Rouge regime. This “excess” of deaths with respect to the usual death rate is caused by starvation, denial of health care, ill-treatment and executions.

8 This village was located in a district where the average population density was 100-200 inhabitants/sq. km.

9 This figure includes widows (10.8?%) and divorced and separated women (4.2?%). The last category is increasing.

10 The bao is a local weight unit, equivalent to 80-100 kg.

11 The chi is a weight unit, equivalent to 3,37 grams. One hun is the 1/10th part of a chi.

12 The usual measure unit is the thang which is equivalent to 2 bags, i.e., 24 kg.

13 The best quality wood is a scented wood. The oil extracted from it (khleum chan, “heart of the moon”), costs around 300,000 riels for 100 grams. The lowest quality of oil costs 2,000 riels per kg. One US dollar was worth 2,200 riels (November 1993) and subsequently was virtually stable after the huge inflation caused by the presence of thousands of United Nations staff.

14 The most heavily mined srok are Thmar Pouk and Serei Sophorn (Banteay Meanchey province), and Banan and Rattanak Mondul districts (Battambang province).

15 In another village (Nikum Knong, Battambang province) the rates are 3 bags of rice paid back for 1 bag borrowed and 30?% interest for a money loan. The lowest rate was observed in Dey Srot village (Prey Veng province) where 2 bags were given back for one bag borrowed and a 10?% monthly interest rate was charged. Programmes of rural credit at low rates were put in place by NGOs in order to prevent excessive indebtedness which has been lamented by observers since the Sangkum.

16 For example, according to the chief of Cadastre Department of Kompong Cham province, the 65 staff of his office had been trained in 1991 during a ten-days training organized in the provincial town. Eleven of them (including four staff of the provincial office and three of the commune office) had then a one-month course in Phnom Penh.

17 In Kompong Cham province, in 1993, only four employees out of 65 staff of all ranks working at the Cadastre Department were surveyors who occupied the post since the Sangkum.

18 Among the 51 companies which obtained a concession between 1992 and 2004, for which I have detailed informations (see Fig. 2), 15 have officially a capital whose shares are Cambodian in majority (from 6% to 100% of the shares for all the companies).

19 Informations provided by Mrs Fabienne Luco, based on figures of the organizations and ministries quoted in the text.

Haut de page

Table des illustrations

Titre Fig. 1: Cambodia: Population Density by Province
Crédits Source: National Institute of Statistics, Ministry of Planning, General Population Census, 1998.
Fichier image/png, 192k
Haut de page

Pour citer cet article

Référence papier

Anne Y. Guillou, « The Question of Land in Cambodia: Perceptions, Access, and Use since De-collectivization »Moussons, 9-10 | 2006, 299-324.

Référence électronique

Anne Y. Guillou, « The Question of Land in Cambodia: Perceptions, Access, and Use since De-collectivization »Moussons [En ligne], 9-10 | 2006, mis en ligne le 06 mai 2013, consulté le 09 août 2020. URL :; DOI:

Haut de page


Anne Y. Guillou

Anthropologist, Centre Asie du Sud-Est, CNRS-EHESS.

Haut de page

Droits d’auteur

Licence Creative Commons
Les contenus de la revue Moussons sont mis à disposition selon les termes de la Licence Creative Commons Attribution - Pas d’Utilisation Commerciale - Pas de Modification 4.0 International.

Haut de page
  • Logo Presses Universitaires de Provence
  • Logo Irasia – Institut de recherches asiatiques
  • Logo Aix Marseille Université
  • Logo CNRS – Institut des sciences humaines et sociales
  • Logo DOAJ - Directory of Open Access Journals
  • OpenEdition Journals