One of the peculiarities of the Vietnamese land system is the existence of a ‘zero state’ with regard to land institutions: all the country’s existing land institutions were put in place in the last 25 to 30 years. However, this does not mean that there is no history of such bodies; indeed, those that are now emerging carry the traces of each past period. The many local customary institutions reflect the principles underpinning previous systems regulating the social and spatial distribution of resources, and elements of the French land tenure system can be seen in the decision to register land ownership certificates rather than follow the more Anglo Saxon system of using the titles themselves as proof of ownership. Nevertheless, there is a clear synchronic dimension to the process of putting land institutions in place, which is reflected in the role it has played in the profound transformation of the Vietnamese State and society. In the first stage of this process, between 1979 and 1993, one of the primary concerns in designing land institutions was to respond to the high expectations of a deeply rural society without making land an autonomous domain. This period saw the progressive dissolution of the cooperatives through the withdrawal of their land prerogatives. Moving in incremental stages, the State first recognised individuals and households as potential land users (with Decree 100, Decree 10 and the Land Law of 1989), although land use rights were still limited and defined within cooperatives through temporary contracts between the cooperatives, which still held delegated management rights, and these new users. This stage ended with the Land Law of 1993 which, while not openly challenging the cooperatives, paved the way for their disappearance by recognising that individuals and households had fundamental derived management rights in addition to the right to use agricultural lands (rights to exchange, assign, rent, bequeath and mortgage land) for relatively long fixed periods. This gave them significant control over land while dispossessing the cooperatives of any real land management capacities. Since these rights are associated with use rights, it was not the land that could be transferred or mortgaged, but the right to use it and enjoy its produce. However, the very existence of these rights and their fairly long-term allocation to households meant that a land market could develop, and that land tenure seemed to function on the basis of private ownership, even if it was not characterised as such. The second stage was a transition facilitating the ‘smooth’ passage from a land tenure system designed to meet the needs of the rural population to one that could support the drive to make Vietnam a modern industrial and urban country. This stage roughly corresponded to the decade separating the land laws of 1993 and 2003. In this period, the State did little to the rights assigned to individuals and households and hardly changed agricultural land tenure. It did, however, endeavour to put in place the land administration, for which it created an independent organ at the ministerial level in 1994, the General Department of Land Administration (GDLA). For the first time, this brought together its decision-making, operational and technical dimensions (the former General department of land management created in 1979, and the former National department of surveys and cartography), demonstrating the government’s willingness to make this an autonomous domain that carried some weight. The State also progressively regulated modes of access to urban, industrial and commercial lands and increased the rights assigned to private enterprise, thus paving the way for the changes in the next period (albeit rather haphazardly by generating a growing number of texts). The third stage started with a reform of the land administration in 2002 and the publication of a new Land Law in 2003. Land was now becoming a tool to develop the territory for industrialisation and urbanisation. This was made clear by the law of 2003, which incorporated regulations from the previous period and barely touched on rural affairs. Little was done to modify access to agricultural and forest lands, which had been regulated in 1993, or provide more flexible access for rural households. But the other categories of land and land users – some of whom appeared in legislation for the first time – occupied a growing and even dominant place in the law. Thus, the new legislation was full of arrangements to facilitate industrial and commercial investments by private and foreign enterprises, and allowed for the development of markets for land and land use rights. It also specified procedures for cataloguing and planning land use. While land use planning remained a top-down procedure steered by the Land Office at different levels, the legislation made the planning process much more flexible by extending the provinces’ prerogatives and enabling the infra-provincial administrative authorities to change the status of lands. Since 2002, land issues have both multiplied and intensified on several levels. The partial and poorly managed decentralisation of land management increased the shortcomings and tensions between the central and provincial levels. On the one hand, the Land Office, which had been substantially modernised and was responsible for planning at every level, had never had as much potential power. This certainly rattled the central government and probably prompted its demotion in 2002 from a ministry to part of the Ministry of Natural Resources and Environment (MoNRE). On the other hand, the provinces have used even greater rifts within the administrative system to lessen the constraints of centralised planning and work very broadly with the legislation in order to respond to local expectations, and especially those of private interests. The increasing privatisation of land has been another point of tension. Since the Constitution of 1959, the State has owned all land in the name of the entire population, and while individual land rights have constantly been extended, individuals are assigned rights of use and management. However, the growing number of recognised users, more flexible conditions of access to land and the progressive extension of rights associated with use rights have allowed private national and foreign enterprises to become dominant land actors – hence the spectacular growth in the number and size of landholdings reserved for industrial, commercial, real-estate and leisure projects, especially in peri-urban areas. The creation of ‘land fund development organisations’ in 2004 is symptomatic of the problems posed by redefing the role of the provinces and private investors. Modes of expropriation are a recurrent problem with investments, and especially compensation for those whose use rights have been expropriated. This issue was only settled recently, and has been treated on a case-by-case basis by the provinces or the Land Office. The Law of 2003 still presents the State as the principal actor in land distribution insofar as it is the authority that requisitions land in order to immediately reallocate it to investors. However, the State has disengaged from transactions since 2004, creating a new, State-mandated body to intervene when lands are repossessed: ‘land fund development organisations’ whose task is to simplify procedures for investors by offering them a single interface, managing the funds from land recovered by the State in accordance with decisions by the competent bodies, and preparing these lands for reallocation to investors. However, the exact status of these organisations, which are not commercial but also not totally public, is somewhat unclear. They are not financially autonomous, they are not mandatory, and their form and level of competence fluctuates as they can operate at the district or the provincial level. This lack of clarity, which results in the creation of bodies whose nature varies from province to province, suggests that the State is trying to divest itself of the highly sensitive problem of expropriations at the expense of their beneficiaries, rather than seeking to resolve it in the long term. What is the explanation for this disengagement, given that the problems created by the way that land is expropriated for investment projects are some of the thorniest and most intractable for the authorities in Hanoi? One reason is probably the increasing complexity of land management, and the human and financial resources that can be devolved to the administration to carry out the tasks it habeen assigned. These are very substantial needs, especially at the lowest echelons (communes, districts) where staff usually have little or no training. But the State’s disengagement cannot be entirely ascribed to these technical and financial challenges; it is also a manifestation of the difficulties of addressing two very different priorities: leading Vietnam towards modernity by transforming it into an industrial and urban country, and organising a fragile and numerically superior rural population with a long habit of socialist values. One of the factors currently execerbating the question of expropriation is the fact that agricultural and forested lands have been kept in a relatively isolated state of suspension for the last 15 years. One would assume that the State has a duty to protect these lands (and their users), but it is actually making them more vulnerable to the dynamics of urban and/or non-agricultural land use (industrial and commercial, leisure, etc.). Agricultural land has been subject to various changes since 1993, but access to such land is still highly regulated. Maintaining a ceiling on the amount of land and duration of the rights allocated limits the process of land accumulation and ensures that the rural population has egalitarian access to land. By the same token, households that have been allocated rights to agricultural land by the State do not have to pay tax on this land, whose value is set according to the value of its agricultural produce rather than the price of adjacent lands (market price). Although this should mean that such land remains accessible even to poor rural households, this specific status, and especially that of highly protected rice-producing land, works against rural households by trapping them in small, low-value farms and weakening their position when private and/or nonagricultural interests come into play. It seems that rather than being protected, agricultural households – along with agriculture itself – are being sacrificed to industrialisation and urbanisation. However, things are not as simple as the last few lines suggest. On the one hand, rural households’ situations vary greatly from region to region, and there are cases where they may be protected by modes of access to agricultural and forested land, especially the most vulnerable households. Recent events, and the global food crisis in particular, have reminded Vietnam that there is still a role for agriculture and rural producers, and once again put the question of rural land under the spotlight. In response to this crisis (and soaring rice prices), the government decided to freeze more than one million hectares of rice fields and launch a campaign reaffirming the value of rural areas in relation to urban areas (the ‘three nong’). It is too early to know whether the return to ‘rural values’ in 2008 will have a lasting impact on agricultural land, and exactly what this impact will be. But the decisions that have been taken show that agricultural land still constitutes a lever that the government will not hesitate to use when the need arises. For certain national officials, agricultural land remains a strong symbol of socialism, and its regulation a crucial element of social peace in what is still a largely rural society with close attachments to the land. Agricultural land is also an issue that raises questions about the State’s role in the move towards ‘market socialism’, and the legitimacy of the Communist Party. While the State’s indecision (or approximations) with regard to land matters could be interpreted as evidence of a certain pragmatism and determination to work with the legacy of the socialist period, recent developments in this domain are testing the very foundations of the Communist Party’s legitimacy, and it could try to deflect this threat by getting the government to maintain the specific status of rural land. So is Vietnam heading towards a two-tier system where some land – the vast block of agricultural and forest lands allocated free of charge – continues to be managed by central government in the nation’s interest, while other agricultural land can be mobilised at leisure and managed under a liberal regime in order to support the country’s economic development? In order to answer this question we needed to turn to the land actors and seek their opinions. The majority of foreign actors (who were historically excluded from this sensitive strategic domain and whose involvement is therefore relatively recent) view the reform as incomplete and thus a major cause of corruption. They are pushing to divest the law of these ‘socialist archaisms’ and make it even more liberal. As recently as March 2008 the World Bank, which some see as the global symbol of liberalism, and which had until then deliberately avoided land matters, signed up to the highly ambitious Land Administration Project, making it the lead foreign actor in this domain and giving a strong indication of the direction in which land affairs are heading. However, the positions expressed by various national land actors are much less clear. On the one hand, officials in the land administration in particular take a fairly technical approach to land: their main concern is the effectiveness of the administration and legislation, and making land an autonomous domain. Officials working directly with foreign experts tend to take a ‘top-down’ approach, looking at the development of the whole country and seeing the constraints associated with the processes of urbanisation, industrialisation and increasing openness. On the other hand, some of the actors we spoke to from the Ministry of Agriculture and Rural Development (MARD) see land as something that cannot be detached from local and sectoral contexts. Therefore, they believe that agricultural land should respond to the needs of agriculture and rural populations, while forest land should primarily respond to environmental protection objectives. At the moment the first group of actors is much more influential because of their strategic position within the land administration and support from international cooperation; however, in the current economic climate the question of agriculture and rural areas and populations has re-emerged as a priority and is slowing the pace of liberalisation. So far there is no indication that agricultural land will be able to take account of the specificities and great diversity of rural areas, whether or not it is prioritised, planned or liberalised. Agricultural land tenure is controlled by the State, and characterised more by the numerous constraints that it imposes (categories, temporal and spatial limits, etc.) than its capacity to adapt to the problems facing the country’s rural populations, agricultural practices and environment. Liberalising agricultural land tenure would bring it closer to a system of individual ownership, which would make land legislation more onerous in many settings where local rights of access to resources are not managed in this way. The Land Law of 2003 introduced several innovations that are helpful in this respect, mainly by creating a new category of users, ‘residential communities’, which allows groups to collectively hold use rights to unlimited amounts of agricultural and forest land that they are allocated free of charge for unlimited periods. However, this new measure is itself very restrictive in terms of what constitutes a ‘community’, the procedures it entails and the framework it imposes on collective management. So what place do customary land tenure systems have in the emerging land system? Vietnamese land institutions seem to have made little or no attempt to plan for this; and the main reason why there are still such diverse local situations appears to be the government’s hesitant approach to agricultural land tenure. The co-existence of actors with divergent positions on this question and on the role of the State, and the relative abandonment of the rural world (especially remote rural areas) because it is not considered important as long it doesn’t challenge the objectives of urbanisation and industrialisation have left a gap where customary systems can continue to function. The recent resurgence of interest in this domain could revive the debate about systems that are considered incompatible with the establishment of a modern State, either because of agricultural practices such as slash-and-burn or the functioning of longstanding local power systems, but customary systems will continue to survive as long as efforts to develop intensive, industrial-type agriculture are not sustained effectively across the country.