CHAPTER FIVE Paper Landscapes
State Formation and Spatial Legibility in Postwar Laos
The understanding of the authorities and the people … is not deep, leading to the delay in implementation, creating opportunities for illegal land occupation and leaving the deforestation issue unaddressed in many locations.
—PRIME-MINISTERIAL “INSTRUCTION ON THE EXPANSION OF LAND MANAGEMENT AND LAND AND FOREST ALLOCATION,” JUNE 1996
IN mid-2007 the residents of a village in southwestern Vieng Phoukha planted twenty-four thousand rubber seedlings on the hillsides behind their houses, the site of the previous season’s upland rice crop. Their work, visible in panorama from the nearby NEC, was the result of a new initiative designed and promoted by the district governor. Located high in the uplands near where the road passes into Bokeo, the village had originally been part of the planned Sino-Lao rubber project. When that deal had fallen through, district authorities enrolled another company, a local sawmill, to finance a village-scale pilot scheme for rubber development. The sawmill owner had provided seedlings to the village on the same terms as Bolisat Ltd.’s “3 + 2” scheme: a 39–61 percent sharecropping split where the company provided the inputs and the villagers the labor and the land. In return, the governor had granted the sawmill owner a new logging concession in a remote part of the village, located over the hill a kilometer or so down a small dirt road.
Village authorities’ explanation of the project contained many familiar tropes: the poverty alleviation and forest protection that would result from replacing shifting cultivation with permanent crops; the same strained effort I had heard elsewhere to skirt the fact that the scheme had been imposed unilaterally on the village. (The governor had “promoted” it personally, and villagers had “agreed.”) More interesting, however, was the new logging concession. Exemplifying the adaptation of Lao land policy to “local interests and power struggles”1 that we have seen before, the deal had emerged from a suite of challenges that confronted local authorities in the wake of the Sino-Lao collapse: the loss of potential rubber financing, the ongoing challenge of upland livelihoods, and growing pressure on local authorities to rein in the extractive industries on which they often relied for economic development. The collapse of the Sino-Lao deal had created both the need for an alternate source of financing and an opportunity to close the loop, so to speak, with the sawmill. By financing reforestation with rubber, village leaders explained, the sawmill was addressing its own negative impact on the forest, not to mention giving something back to the village.
This narrative was supported by a revision of the village’s LFA map, which had been updated the same year (fig. 5.1). The new map showed a shuffling of land and forest categories, including most importantly the creation of a new “production forest” (the logging concession) and the reclassification of a sizable chunk of the village’s “agricultural land” to “protection forest,” a forest category that in some cases refers to military areas but in this case referred to watershed protection. This cartographic sleight of hand more than offset the new concession, and the revised map thus increased the amount of forest (on paper at least) contained in the village’s territory. It had the added bonus of pushing the deforestation out of view, displacing it from the swidden fields that had previously lined the NEC (and were now a maturing rubber plantation) to the new logging area over the hill.
The sawmill scheme exemplifies two key ways that zoning maps have become important tools of population management in Laos’s northwestern uplands. The first concerns the work of legitimation. As noted already, Lao officials engaged in rural development schemes often invoke the government’s exclusive authority to manage or zone land (chatsan thi din) as outlined in the country’s constitution and land laws. This is clearly at stake in the example above, where LFA seeks to draw on maps’ persuasive power to assert state authority over village-scale land use. In this sense, LFA is analogous to the Bolisat Ltd. project map described in this book’s introduction (and elaborated below); both use formal cartography to try to impose a particular vision—and indeed a particular spatial plan—on upland landscapes and communities.2 These efforts are never just about maps, though. We have seen LFA maps a number of times in earlier chapters, and taken together, their role is ambivalent at best: for every rationalization of a land grab using the tool of mapping, it is possible to find a “paper tiger” like the LFA map in Ban Deng, for instance, whose presence in the village defied rather than reinforced the dominant strand of population management work. The techniques of upland governance, after all, were developed in the 1980s, a full decade or so before the rollout of the LFA program. Zoning maps, in other words, figure into larger efforts to get upland citizens to “do as they ought,” but they are hardly recipes for success in themselves. For every case where they appear as an agent of enclosure, another example testifies to the weakness of land-use interventions that are merely formal.
FIGURE 5.1 Rubber, logging, and forest classification in southwestern Vieng Phoukha. Maps by Ben Pease based on author’s data.
A second dimension of village-scale zoning work is less obvious and yet maybe even more important. This is the use of zoning maps as a political technology directed not just downward at local land users but also upward at central-level authorities. In recent years, “illegal” logging practices by local authorities have become a political issue in Laos on par with foreign land deals, stimulating a range of state, donor, and civil society interventions.3 Meanwhile, timber concessions have remained a key resource for local authorities, funding outright corruption (as is often pointed out) but also paying for local budgetary expenses like new infrastructure and salary supplements that are seen as locally legitimate, if not entirely within the letter of the law.4 The example above shows district authorities attempting to embed an increasingly embattled status quo—the granting of a new timber concession—both within the rhetoric of economic and environmental sustainability and, more substantively, within the technical, grounded work of land-use planning and smallholder rubber development. It shows, in sum, how formal zoning has become a tool for managing not just local populations but faraway authorities as well.
The evolution of village-scale mapping in Laos exemplifies the articulation of transnational land deals and LFA in the early 2000s, as the former were on the rise and the latter was coming to a close. This is a story of localized control over the power of maps, and it ultimately helps explain the ongoing lack of central-level regulatory leverage over many rubber and other land deals despite the fact that village-scale LFA began as a bid to take power away from local authorities. The effort to use formalized, step-by-step mapping as a means to break local authorities’ monopoly over land-use planning has received far less attention than its role in facilitating enclosure and, more generally, in what I have called population management work.5 Yet as Peter Vandergeest points out in his foundational work on the LFA program, zoning projects need to be understood historically, as the concrete outcomes of particular conjunctures rather than an inevitably taken step on the road to state territorialization.6 Most accounts of LFA, in contrast, do not even consider the question of timing; they take it simply as natural that Lao authorities would pursue a program like LFA because that is just what states do.7
My approach here focuses more explicitly on what scholars call ongoing state formation, or the struggles and fragmentations within the state that center on the control over spatial information as a way to control the value, or rent, that comes from land.8 Looking at the paper landscapes created by LFA and derivative land-deal maps allows me to dig into these politics of internal regulation, which continue to plague transnational land deals more than a decade and a half after they were first officially named as a problem. This requires a lot of reading against the grain, including narratives like the epigraph above that emphasize the lack of capacity or understanding among not just local citizens but also local authorities, whose “lack of capacity” is a familiar theme in Lao (and other) development discourse.9 In contrast, my emphasis is on elucidating the competing interests and, more specifically, the high-stakes resource politics that underlie land deals. Ultimately, this fight over regulatory authority and jurisdiction helps to explain why maps have proliferated at the local level at the same time that central-level regulatory capacity has continued to falter.
To adequately grasp the full trajectory of relevant events, we need to return (again) to Laos’s market transition efforts of the late 1980s and early 1990s, and specifically to the decentralization of state control over industrial forestry. Doing so facilitates a focus on how and why information about land deals travels (or does not travel, as the case may be) through bureaucratic channels. Even as transnational land deals have become formalized on paper, this formality has not necessarily translated into legibility at the national scale. This is because of internal and ongoing state formation, which itself turns on a political economy of resource control that stems back to the late 1980s, as the Cold War was ending and the still-struggling Lao government was deciding how to recruit “investment” for purposes of national economic development. But this chapter is not merely about the weight of the past. Amid these internal struggles over legibility, we also see the emergence of new forms of transnational legibility, which in turn raise questions about the “global” nature of the contemporary moment.
“SOCIALIST BUSINESS ACCOUNTING” IN THE FORESTRY SECTOR
The late 1980s were a time of intense economic and political disruption. The height of neoliberalism in the United States and Britain, this was the era of perestroika and dissolution in the Soviet Union and of economic reforms across the socialist world, from those of Deng Xiaoping in China to Doi Moi in Vietnam.10 Under Laos’s New Economic Mechanism, it was the era of “stepping up population management work,” “heightened vigilance,” and territorial experimentation with focal sites and managed enclosure in the uplands (ch. 3). In 1986 Laos’s Fourth Party Congress adopted the slogan of “everything for the socialist fatherland and for the plentiful and happy life of the people of all ethnic minorities.”11 Yet this was aspirational at best. Behind the scenes, internal communications reflected the structural difficulties inherent in the economy, from the prevalence of “backward” and “traditional” production to the increasingly predatory actions by a range of state actors whose loss of Soviet and Eastern Bloc aid meant a direct threat to their own subsistence. In 1987 a prime-ministerial “Order on the Free Market” echoed the French rhetoric of “unblocking Laos” from a century earlier (ch. 1), but pulled no punches in criticizing the heavy hand of the government’s own “administration” efforts, including the disruptive “inspection of goods” and ongoing predatory interference in economic production by “provinces, districts, tassengs [subdistricts],” the Ministries of Defense and Interior, and various “military units, police and militia units.”12
Amid these difficulties, central-level authorities launched a “socialist business accounting” initiative that aimed to make state-owned enterprises—at the time, the pillars of the Lao economy—more productive and accountable to market-based calculations. Piloted in 1985 and expanded in 1987, these reforms sought to replace heavy-handed administration with market-based production one enterprise at a time.13 On one level, the statistics sounded almost impressive. By the time of a March 1988 national meeting on the topic, at least 105 provincial and 38 central-level state-owned enterprises had been “granted full autonomy”; many other businesses had “also actively made efforts to fulfill the 1987 plans” for reform.14 Yet the transition to a “free market” was largely acknowledged as a failure. As in the above-quoted “Order on the Free Market,” interference remained widespread. A 1988 reflection by Lao president Kaysone Phomvihane tried to put a positive spin on the recent “renovation and development,” claiming to have “modified the attitude of Left-leaning and haste” and “stopped abolishing private and individual rights to ownership and markets.” But even this public-facing account, published in the Soviet newspaper Pravda, acknowledged “abnormalities” in the economy that had resulted “because the production and circulation of goods had come to a halt.”15
Forestry was the exception that proved the rule. Contra the difficulties earlier in the decade, the late 1980s are widely remembered as a boom era for the Lao timber sector due to a mix of increased “business autonomy” and rising external demand (in particular from Thailand, whose 1989 domestic logging ban helped spur exports). Lao provincial administrations, which used forestry revenues to support their long-neglected budgets, benefited significantly.16 The transition to provincial control over SFE 1 that took place during the late 1980s and early 1990s (ch. 3) exemplified this, but Bolikhamxai was hardly unique. During this period provincial authorities across the country gained increasing control over activities like roadbuilding, logging, and sawmilling, in part owing to the creation of provincial forestry enterprises alongside and in some cases from the older SFEs.17 A parallel dimension concerned “strategic” state enterprises like the infamous Mountainous Areas Development Company (BPKP in its Lao acronym), whose mix of military affiliation and newfound “business autonomy” made it especially powerful.18
Between 1987 and 1989, Lao timber exports increased to the point where they became the country’s chief source of foreign exchange, accounting for roughly a third of annual exports.19 The adoption of “market principles” was so successful that in 1988 central-level authorities issued a logging ban, and when this failed to be heeded, forestry officials attempted to at least regulate logging through the creation of a provincial quota system.20 According to one well-placed observer, these quotas were issued, in 1990, “largely to combat the widespread overharvesting that had occurred in the late 1980s when provinces exercised almost complete control over the industry.”21
A 1989 editorial in the Lao People’s Daily hinted at the regulatory challenges that confronted officials in Vientiane who were increasingly realizing the unaccountable and indiscriminate nature of the logging operations they had unleashed. Targeting “those units running businesses of agricultural and forestry production”—a veiled reference to state and provincial forest enterprises, as well as strategic enterprises like the BPKP—the editorial insisted that the Ministry of Agriculture and Forestry “ha[d] the right to supervise” these units so that they could “correctly carry out their activities in accordance with the line and policy of the party and state.” In the language that followed, one gets the distinct impression that there is more than just the conceptual distinction between necessary regulatory oversight and unhelpful interference that would stifle productivity: “But this does not mean that the said ministry has the right to interfere in the work of those business production units. By saying this, it does not mean that the ministry has to support or allow the said business units to carry out their businesses any way they wish. In this regard, the various business production units must also directly take responsibility over the implementation of various regulations and laws adopted by the party and the state.”22
In such assertions of regulatory right and calls for responsible action on the part of “business production units,” the multiple moving pieces of the state appear to be increasingly in conflict. In the years that followed, the nexus of extractive entrepreneurialism and local authority would only continue to snowball. In the process, it would inspire a whole policy apparatus aimed against it.
“THEN MAKE LAND MAPS”: LFA AND THE PROBLEM OF LOCAL AUTHORITY
In an effort to rationalize and expand the capacity for long-term forest management, Lao officials held the country’s first National Forestry Conference in May 1989. The conference set the stage for much of what would follow, including the creation of a national system of protected areas in 1993 and, in the shorter term, a series of logging bans, the first of which was issued the same year as the conference. As part of this immediate action, the conference launched an effort to address the “indiscriminate logging” of the later 1980s by “dissolving” the SFEs and associated provincial forestry enterprises. Yet despite some impressive policy language—a prime-ministerial decree ordered that logging “should only take place in inventoried production forests” and had to involve “approved economic-technical plans,” “full payment of log royalties,” and “compensatory tree planting”—the rationalization of the sector stumbled when the putative dissolving of the forest enterprises turned out to be simply a mix of devolution and privatization.23 As one official source admitted years later, the lowering of accountability over the forest enterprises meant that during this period, “in many cases, Provinces harvested above the allocated plan to create additional fund[ing] for development projects, and there were many irregularities concerning log sales contracts as well as logging, grading and sales themselves.”24 A second logging ban in mid-1991 also accomplished little, as noted by the admission that the “ban” had, in fact, been little more than a slowdown, and that log sales—a staple of provincial operating budgets—had remained roughly the same and even increased in 1993 as supplies of “old logs” ran low and local authorities began harvesting anew.25
It was in this context that the LFA program was born. Initially outlined as part of the forestry-sector rationalization decreed in the wake of the first National Forestry Conference, LFA was piloted at the district scale in the early 1990s and scaled up to a nationwide program beginning in 1996. LFA is one of the most studied government land policies in contemporary Laos, and rightly so. Between the late 1990s and mid-2000s, the program oversaw village-scale zoning in roughly half of the rural villages in Laos; estimates vary—between 5,400 and 8,000 villages are reported to have received LFA implementation during this period, often in connection with foreign development projects26—but they collectively point to LFA’s role in intensified state efforts to manage land use at the village scale.27 As chapter 3 illustrates, LFA was hardly the beginning of such efforts; resettlement and other forms of population management work were features of upland reorganization since the late 1970s. But LFA’s launch in the 1990s added a new level of intervention in village-scale agrarian affairs.28
Coming in the wake of the relatively ineffective forest management reforms described above, LFA also brought a more resource-intensive and micromanagerial approach to the problem of local authority. Broadcast in intent via the 1993 decree that announced LFA’s scaling up to a nationwide program, the project took substantive shape in mid-1996 with the beginning of a concerted effort to formalize the practice of granting land for “development.” This was an explicit play against the land-allocating prerogative of local officials, and it centered on a prime-ministerial decree (issued in June) and a conference (in July) that brought provincial-level officials to Vientiane for instruction.
Titled “Instruction on the expansion of land management and land and forest allocation,” the decree focused on land administration broadly and was explicit about central-level intentions. It began with the epigraph above, describing “the understanding of the authorities and people” as “not deep,” and thus both “creating opportunities for illegal land occupation” and “leaving the deforestation issue unaddressed in many locations.”29 Under the “activities” portion of the decree, two articles merit special attention. The first mandated the establishment of local-level committees to review all instances when “the acquisition or possession of land” by private actors had occurred “due to village administrative authorities’ decision to exchange it for constructions of public interest, such as roads, electricity, schools, hospitals, and so on.” These committees were instructed to figure out whether these concessions (“exchanges”) had been reasonable, and where they had not been, to withdraw part or all of the allocated land.
Second and more importantly, the decree banned local authorities from granting concessions outright in the future: “Starting from this instruction, the administrative authorities at each level shall be strictly forbidden to further exchange land for constructions.”30 In doing so, it then specified an alternative instrument for exchanging land for development that would be formalized and centrally administered: the state land concession (sampathan thi din lat). Focused on “building confidence in investment for land development, as well as orderliness in the Lao population’s livelihood,” the formalization of land-concession practices underlay an effort that would be expanded the following year in Laos’s first Land Law, part of what Lao officials would eventually refer to as the transition to being a “rule-of-law state.”31 Specifically, this sought to replace the arbitrariness of local authority with a formalized mechanism grounded in the law: “State and social organizations receiving the right to use and conserve land shall not be entitled to transfer, lease, grant as concession or evaluate such land as collateral for loans. The right to use land by foreign residents, persons without nationality living in the Lao PDR, [or] aliens investing or conducting other activities shall be acknowledged by the State under the form of lawful lease or concession only.”32
The First Nationwide Review Conference on Land Management and Land-Forest Allocation was held in July 1996, and emphasized this agenda of formalization and legalization as a pathway to better governance. The conference’s resolutions document emphasized the “necessity to fully continue the work of gradually putting land management into a legal framework, settling cases of illegal occupation of land, the falsification of documents, the illegal purchase/sale or occupation of state land …, [and] at the same time allowing the state to collect taxes and fees from land, which constitute considerable income.”33 Framing the “illegal” disposal of state land assets as something that only local authorities did was a stretch, but it was a useful simplification in that it allowed the issue to be broached and discussed didactically rather than simply ignored.34
After outlining various “weaknesses and shortcomings” and lamenting that the “absorption” of government policies to date was “not yet appropriately profound,”35 the resolutions document outlined the pathway to a landscape that would be legible to regulatory oversight: “Plan for land surveying and management in order to collect data, work out management plans, and classify land types based on sectoral and regional development plans based on scientific use of land and actual local conditions. Then make land maps; particularly in the coming years the focus should be firstly on priority development areas, districts and suburban areas, which will be the basis for the delegation of management responsibility to concerned sectors as well as the allocation of land and forest to villages and families to manage and use.”36
The document twice mentioned the goal of creating a nationwide set of fine-scale zoning maps by the year 2000, and it laid out a series of plans to facilitate central-level control over this process.37 These plans converged on the need to reverse the existing pattern of central-level instructions going “down” to the local level but little coming back “up”: in the resolutions’ specification of “regional plans,” it was made clear that geographically specific planning documents needed to travel to Vientiane for approval.38 A newly created central-level Committee for Land Management and Land-Forest Allocation was also announced, which would oversee a series of regional pilot projects around the country.39 Again, the message was that reforms would not just exist on paper.40 The document promised visits from this committee “to the grassroots to direct the experimentation in some provinces or open up training on concepts, directives and various technical topics for local personnel.”41 Echoing the instructions on population management work from almost a decade earlier, it also reminded conference participants that “land management, land-forest allocation, surveys to develop land title registration, and the management of state land and housing” entailed “delicate, comprehensive work, linked to political, economic, social, environmental, and security aspects of the population.”42 As provincial officials left the conference for home, they did so with the knowledge that their independence over the disposal of land was officially a problem, and its days officially numbered.
REPURPOSING LFA
In the half decade that followed, LFA turned into a state-territorial juggernaut; as already noted, the program intervened in the land-use affairs of thousands of villages between the late 1990s and when it was officially suspended in 2003. But as LFA was taken into the field at scale, it was widely repurposed by local authorities for their own territorial ends, as in the rubber-for-logging scheme described above. As a result, LFA ended up as anything but the centralization mechanism its designers had initially envisioned. Upward reporting of land-use plans failed to materialize, and this caused widespread (and ongoing) challenges for large-scale projects like dams, which require locally detailed data to plan and budget resettlement infrastructure, social services, and livelihood restitution.43 LFA maps were never collected in a centralized repository, and the summary statistics given above are basically all there is. These numbers testify to the program’s proliferation, but also to its almost complete lack of success in creating the detailed spatial legibility that central-level authorities demanded of it. In summarizing and aggregating the data about how many villages were included in LFA, as well as in various summaries of how much of each forest category was allocated, local officials left out what mattered most: precise spatial data about what was supposed to happen where.
When my colleagues and I first encountered the Bolisat Ltd. project map in 2007, much of its illegibility stemmed from the fact that we had not seen the LFA maps on which it was based. In time, we discovered not only the story of uneven enclosure that underlay the map (chs. 1 and 4) but also the map’s own legal-geographic origins—its cartographic genealogy, so to speak. These were, in many ways, totally different investigations. The former lay largely out in the field, in the villages shown on the map and in the various (and often place-specific, historically inflected) details of population management work that facilitated the project’s enclosures. The latter, on the other hand, began in the map room at the district Agriculture and Forestry office, where the original paper version of the nursery map was kept alongside the earlier generation of LFA maps. On one level, the project map (bottom of fig. 5.2, below) was of only minor help in understanding Bolisat Ltd.’s activities on the ground: it represented all of the project’s target villages equally, and neither its cartography nor the village-by-village statistics that accompanied it gave any hint of why Khet Nam Fa received so much enclosure relative to the other villages. In this sense, the enclosure story was the result of digging at the reality underneath the map, of getting beyond the reality of equal treatment conjured on the map’s surface. It was, in short, born of ignoring the map and looking elsewhere.
On another level, however, the formal geography shown on the map was far from meaningless—it just told a different piece of the story. The Bolisat Ltd. project map had been derived from the earlier generation of LFA maps created in Vieng Phoukha between 1997 and 2004 (fig. 5.2).44 In making both generations of maps, district-government technicians had been attempting to follow various directives to create “permanent livelihoods” for upland residents. The LFA maps tried to do this by categorizing certain areas—often the most intensively or recently used shifting-cultivation areas, which forestry officials saw as the most degraded—as agriculture land (din kasikam), and classifying the rest of the village as various administrative forest categories. These almost always included restricted-use categories like conservation forest and watershed protection forest, as well as some form of forest for village utilization; other categories, for forest regeneration or commercial (“production”) forestry, were sometimes included as well (as in the opening sketch above). This was a sort of triage: an attempt to restrict shifting cultivation into consolidated areas where it could ideally be “stabilized” through intensification, or at least corralled away from other types of (more valuable) forest. When district Agriculture and Forestry technicians made the Bolisat Ltd. project map, they followed both this same logic and this earlier geography, directing the company to target its plantation operations into areas that had been zoned for agricultural use under the LFA process. This would both this minimize the conversion of natural forest into rubber plantation and help operationalize the project-level agreements between companies and provincial officials that “the target is shifting-cultivation fallow land” (ch. 1), replacing upland cultivation with rubber at the scale of individual fields.
In the transition from individual LFA maps to the rubber-project map, we thus see a second example of how local authorities have continued to use the earlier generation of LFA data for their own state-territorial purposes. These uses differ by case and context, of course: the opening example showed a rezoning in which the earlier LFA map had been deemed to have outlived its use, while the Bolisat Ltd. case shows an example of the selective reproduction of one part of the earlier maps. But collectively, these examples contradict the position, often voiced in the international donor community, that old LFA maps have been left to rot because they are inaccurate and out of date.45 In contrast, the examples shown here illustrate the dynamism and ongoing life of this earlier generation of maps. Even if they are out of date, their adaptation to changing circumstances and ongoing opportunities hardly means that they have been left to rot.
FIGURE 5.2 Cartographic genealogy of the Bolisat Ltd. map. Diagram by Ben Pease based on author’s data.
Although the distinction between Bolisat Ltd.’s “3 + 2” and “4 + 1” villages did not appear on the project map above, there is an important clue in the earlier generation of LFA maps. As a district that was both fairly small and staffed with competent technicians, Vieng Phoukha was almost fully mapped by LFA between 1997 and 2002. After they mapped the boundaries and laid down internal zoning on each village, district Agriculture and Forestry staff were faced with the question of how much additional mapping work to do in each village. They tended to distinguish between villages that had at least some upland plots that were deemed to be individually developed and those that did not. These two types of villages are plainly visible in the district’s LFA maps; the former show the results of what district technicians described as “full” LFA—zoning plus the cadastral mapping of recognized (“allocated”) individual upland plots—while the latter show the “short” version only: village zoning but no plot-level mapping. At the time, this reflected local authorities’ efforts to create land-use plans that would help facilitate various land-governance goals, including agricultural intensification and sedentarization, forest triage of the sort mentioned above, and, where applicable, the taxation of “developed” land. In doing so, these efforts also operationalized the distinction between lowlands and uplands in a cartographically explicit way that created space, quite literally, for precisely the types of managed enclosures that Bolisat Ltd. and district authorities would develop in the years that followed.
In practice, this distinction was indicative rather than predictive; the key to Bolisat Ltd.’s differential geography of enclosure, as noted already, lay largely off the map. This extended to LFA as well. Perhaps unsurprisingly, the villages of Khet Nam Fa had received only the “short” version of LFA, and they were thus well positioned for their later inclusion in the “4 + 1” scheme. But a number of other villages in the project, including Ban Deng and a few of the other villages whose LFA maps are shown in figure 5.2, had also received only the “short” version of LFA; they nonetheless ended up with “3 + 2” arrangements. Still, the distinction was invoked by Bolisat Ltd. representatives to justify the project’s uneven enclosures. This was laid out especially clearly by a former provincial Industry and Commerce official whom the company had hired to liaise with local officials.46 Our interview, in late 2007, illustrated a key way that fixers like this man help companies translate their business interests into development narratives that have the trappings of legality. As he explained it, in the “3 + 2” villages, the company was working with farmers who had individual plots of land, while in the “4 + 1” villages, agricultural land was managed collectively and organized by the district under the state’s mandate to coordinate (chatsan) land use. This land would thus still technically belong to the village, he explained, even as it was allocated into company-owned plantations for the next thirty years. This final qualification indicated the man’s full awareness of the sleight of hand he was describing. Formally, the agricultural land allocated to Bolisat Ltd.’s plantations was still “locally owned”; substantively, as chapter 4 showed, the managed enclosures guaranteed that it was anything but.
Like the rubber-for-timber scheme profiled in the opening sketch, the “on-paper” geography of the Bolisat Ltd. project was thus formal yet flexible, capable of being manipulated to accommodate particular interests and changing circumstances while also providing an appearance of formal continuity with state-originated land-use planning and law. But despite their selectiveness and opportunism, both projects were, in a very real sense, a continuation of the transformation in land management begun by LFA. Long before Bolisat Ltd.’s arrival, LFA had begun creating “purified” categories of agricultural land and forest, anticipating on paper the enclosure of both forest and farmland out of a messy landscape where the two had been previously intertwined.47 But as noted in chapter 1, LFA on its own had been largely aspirational—lines on the map, in most cases, and little more. By injecting capital in the form of hundreds of thousands of rubber seedlings and, in the years that followed, a multimillion-dollar rubber-processing facility, Bolisat Ltd. had helped to change all of this. The partnership between the company and district authorities had actually started to produce the landscape that LFA had dreamed into existence years earlier. Even if the spatial details varied, the new landscape was coming to be one where land was purified into distinct parcels of agriculture and forestland, where lines between state and private property existed, and where there was increasingly less space for the mixed and mobile forms of production that had predominated in the years prior.
GEOGRAPHIES OF OPACITY AND LEGIBILITY
Land deals leave extensive paper trails. While details vary by case, the rubber projects I studied exemplified the back-and-forth between processes involving various state authorities and multiple possible field locations that plantation developers traverse as they go through the land-finding process.48 Maps like the ones shown above sit atop piles of project proposals, letters of introduction, MOUs, meeting minutes, and one-sided “agreements” (kho toklong) that show the backing of provincial and district governors—to name just a few of the documents involved. Yet despite all of these documents, much of the detail about the land access they have facilitated has remained illegible to central-level authorities. One of my colleagues from the NLMA once joked, in reflecting on his experience collecting land-deal information from provincial and district authorities, that despite his agency’s mandate to coordinate between various ministries and administrative levels on issues of land governance, it was “policy” not to share documents between agencies.49
Often this is framed, even by local-government staff themselves, in terms of low capacity—both in making maps and in caring for them. My colleagues and I encountered this kind of dissembling in 2008, in the lead-up to viewing the cache of maps that included Vieng Phoukha’s original LFA maps, the Bolisat Ltd. map, and a handful of others—including the map described in the opening sketch—that had been recently updated as part of the district’s effort to formalize its ongoing forestry operations. We were told that the maps were a mess (bo ngam), and it was only with obvious reluctance that we were eventually allowed to examine them. Practically speaking, the maps were a bit of a mess, but as the examples above imply, this was not the only reason local authorities were likely reluctant to share them.
We were hardly alone. Throughout 2005, 2006, and 2007, anecdotal evidence about concessions’ on-the-ground problems had been trickling into policymakers’ offices in Vientiane, and this ultimately culminated in a May 2007 prime-ministerial moratorium on new agribusiness and mining concessions.50 Yet this was merely an effort to pause the “trading of land for development” that had been ongoing for well over a decade. As one of my informants in Vientiane put it at the time, “Our central government here is very weak. We don’t have the tools to check whether the provincial governors—and the provincial governments more generally—are following the national laws.”51 Another informant, also a central-government employee, explained how these problems of vertical information access created horizontal ones as well out in the field: “It’s very difficult for parallel departments”—for example, a provincial Land Management Authority office and a provincial Agriculture and Forestry office—“to get information from one another at the provincial level. Usually, it has to go to the top [central level] and come back down.”52 Provincial land management officials I spoke to in Luang Namtha were experiencing this acutely in their efforts to collect specific details about land deals in the area from the provincial Finance Department. “We don’t know [the details]! We don’t have area statistics, we don’t have maps. When a company is interested in a concession, they [go through various steps to obtain various permission documents]. These documents exist, but our office doesn’t have them. Where are they? In the provincial Finance Department. But just on paper, not in the computer—and they’re all mixed up [sapson].”53
Since the late 2000s, a few different efforts have aimed at assembling systematic data sets about land deals but have run into similar challenges as the 1990s-era promises to “go [down] to the grassroots” in order to improve land-use planning and governance. Researchers working with the Lao Ministry of Environment and Natural Resources on a later iteration of the NLMA inventory that had taken my colleagues and me to Vieng Phoukha echoed this sentiment in 2012 when describing the constraints on their own (ongoing) regulatory efforts: “Because state land can be granted at multiple levels and across different [sectoral] ministries within the government, land [deals] have been particularly difficult to measure and monitor. Data collection has been ad hoc in nature and transparency or dissemination of records across sectors and levels of government have been limited. As a result, aggregated data was often available from provincial or district level offices, but less on an individual project basis and often not in formats available for or compatible with other administrative areas.”54
The distributed geography and “ad hoc nature” of land-deal data is a problem for planners as well. As explained by a researcher at the Ministry of Planning and Investment, because of the recent and significant growth of the Lao plantation sector, “specific information about the size, location and types of plantations and investors … remains limited and dispersed across various government agencies,” and “information on projects approved and implemented at the provincial level is often not available at the national level. Total plantation investment data is [thus] not currently held by the Ministry.” As a result, “the overall scope of investment in the plantation sector is unclear, which has hindered planning efforts.”55
The common themes here—multiple land-granting authorities, limited transparency and cooperation—are not just about information. Statistical knowledge plays a key role in planning and executing large-scale projects, as noted above, but it also underlies an activity that is even more fundamental to state practice: the control over land-based taxation and, more generally, over resource rents. One reason that the timber concession featured in the chapter’s opening sketch was so important to paper over was that it would have been looked at askance by central-level authorities who, as noted above, have been trying since the early 1990s to make logging more accountable to the capturing of royalties. As these efforts have become increasingly public in recent years, the magnitude of untaxed timber allocations—often by local authorities and involving special timber quotas for infrastructure or unspecified “development projects”—has been acknowledged to dwarf officially collected timber royalties by roughly an order of magnitude, leaving hundreds of millions of dollars unaccounted for.56 A 2016 prime-ministerial order addressed to ministers and provincial governors once again forbade “the direct trade or exchange of timber for infrastructure development projects” and, just to make sure its intentions were completely clear, explicitly nullified any “previous orders, decisions, notices and legislations, including those of local authorities” that conflicted with its instructions.57 A suite of mapping efforts by multiple government agencies, meanwhile—including revised forest categorization by the Ministry of Agriculture and Forestry; cadastral land titling under the Department of Lands; and the concession-inventory efforts by the Planning and Environment ministries quoted above—continue to gesture to both the economic stakes and the institutional scope of the issue.58
The question of Bolisat Ltd.’s precise plantation geography illustrates the extension of this politics of rent in the concession era, as agribusiness has emerged alongside forestry as a key economic activity. The rubber taxation policy issued by Luang Namtha’s provincial governor in 2006 (ch. 4) aimed to capture a fraction of the anticipated production rents by taxing rubber trees on the basis of owners’ plantation holdings: smallholders with less than one hectare would pay RMB 1 per tree per year, while larger-holders would pay RMB 3 per tree for plantations between two and six hectares, and RMB 6 per tree for plantations larger than six hectares.59 At the time it was issued, the policy was largely hypothetical since it applied only to trees that were already being tapped; most rubber trees in the province then were still a few years away from tapping. But in the terms it set, it made explicit the stakes of the mapping that would need to be done in the coming years if the province was to realize the taxation potential of its new plantation resource.
The project map shown at the bottom of figure 5.2 is doubly interesting in this regard. On the one hand, it says very little about the company’s tax burden. As a plantation-targeting map, it does not show where the company’s plantations actually were; it only showed the larger area—over 9,000 hectares (compared to the company’s 3,000-hectare quota)—within which they were supposed to be developed. Moreover, in delineating this extent, it did not distinguish between areas where the company would own the plantations directly under the “4 + 1” scheme (and thus generate tax liability for the company) versus areas where the rubber trees would be owned by smallholders working with the company as contract farmers. Since taxation rates differed according to plantation size by owner (rather than simply aggregate area), the map only gestured to the wide range of potential tax rents.
Thus far, these numbers have remained largely hypothetical. After global rubber prices fell in 2011, Chinese rubber companies negotiated provincial tax waivers with Luang Namtha officials—allegedly in order to keep company operations viable—and provincial authorities have reported an ongoing “reluctance” to enforce the policy.60 Nonetheless, information about project-specific plantation areas remains closely guarded. During my 2018 revisit to Vieng Phoukha, I had an interesting interaction with government staff-people who were unable to reconcile the obvious extent of Bolisat Ltd.’s “4 + 1” plantations in Khet Nam Fa with statistics held by their office, which listed the company’s plantation holdings at a mere thirty-three hectares.61 Given the scale of the company’s actual plantations—most likely in the range of 1,000–1,500 hectares—this would have translated into a few hundred thousand dollars’ difference in annual tax revenues had the provincial policy been enforced.62 Even in the policy’s absence, however, the company remained wary; after registering confusion over the mismatch in the numbers, the staff-person acknowledged that the company had been reluctant to share its precise area with local officials “because they don’t want to pay tax.”63
Nevertheless, the Bolisat Ltd. map did create an important type of legibility, albeit of a different and expressly transnational sort. As noted in earlier chapters, projects like Bolisat Ltd.’s were in part financed by Chinese government subsidies that, in their stated purpose to replace opium-poppy cultivation with legal agricultural crops, articulated with and grounded the NEC’s wider vision as a licit, integrated, and development-oriented transformation of the former Golden Triangle borderlands. The poppy-replacement subsidies were generous, offering reimbursement for 80–90 percent of exploration and insurance costs as well as cash payments of USD 25–70 per hectare developed, depending on crop.64 To access these payments, however, companies were required to provide evidence of feasibility and permission such as contracts, official letters and MOUs, feasibility studies, and, yes, maps. Even if it did not show the location of Bolisat Ltd.’s actual plantations, the map shown above was a more localized version of the proposal recounted in chapter 1: it effectively said that, from the district government’s perspective, providing 3,000 hectares to the company’s collaborative plantation efforts “would not be a problem.”
The fact that, in hindsight, this full quota seems not to have been achieved is almost beside the point. Maps are pictures that travel—“immutable mobiles” in the language of Bruno Latour. But while this traveling can be a source of power when accurate knowledge is carried across distance—in Latour’s classic example, this concerns naval supremacy because one side has the foreknowledge of an area thanks to a good map65—the mobility can be more important than the immutability. As critical cartographers have shown, the power of maps is not merely to carry realistic representations to faraway places but to carry ideas and claims, whatever their relationship to reality.66 Nikolas Rose, for example, describes maps as “little machine[s] for producing conviction in others,”67 and indeed, the less the viewer knows about the area being shown, the more easily their conviction can be manipulated. Maps like Bolisat Ltd.’s would have run into problems if they had been taken too literally, and it should be no surprise that the land acquisition process itself relied on other means (ch. 4). But farther away the map would have worked like a charm, since a combination of unfamiliarity with the terrain and an interest in supporting new Chinese agribusiness made the map’s potential audience unlikely to object. Like the signs that advertise Chinese poppy-replacement development projects across northern Laos (fig. 5.3), the map would have functioned—much like it did in my initial encounter—to signify approval without getting into the details.
Reading the historical literature on Laos, it can be tempting to view the persistence of local authority over the disposal of land as a direct descendant of Southeast Asian (and in particular “Tai”) political geography. Martin Stuart-Fox, a prominent historian of Laos, describes Lao political culture today as the result of French, American, and then socialist rule being layered “on top of, rather than in place of, traditional Lao social and political relationships.”68 The persistence of long-held and locally oriented patron-clientelism makes provincial administrations, for Stuart-Fox, “the modern muang—enjoying considerable independence from the center in return for a tribute (tax) which is often [in actuality] withheld.”69 Others, too, have looked to precolonial times to explain contemporary resource politics, such as this anonymous development professional quoted in a study of timber politics in southern Laos: “In other countries, the political culture surrounding timber is often termed a ‘kleptocracy.’ But [here], it is more of a ‘tributocracy.’ The way that it is handled is very much in keeping with both the traditions of Lao politics and the long history of paying tribute to more powerful entities, be they internal or external.”70
Without denying the clear echoes of historical muang politics in contemporary events, a series of distinctly contemporary processes arguably play an even more definitive role in shaping local governments’ prerogatives when it comes to allocating land “for development.” From the decentralization of forestry capital in the late 1980s to the distinctly modern story of postwar legacy told in chapters 2, 3, and 4, land governance in the Lao uplands must be pulled out of the mists of time and tradition in order to grasp its particular form of territoriality. Focusing on the articulation of village-scale mapping and transnational land deals, this chapter examines the way that Laos’s LFA program spawned an arena of “techno-politics” that remains relevant today even as the program’s heyday is in the past.71 The ongoing politics of post–Cold War state formation figure centrally here, undermining—at least from the perspective of central-level planners and regulators—the ability to control the economic dynamism that has been unleashed at the upland frontier.
FIGURE 5.3 Poppy-replacement project (“6,700 hectares”), Oudomxai province.
Spatial abstraction, as many scholars have noted, is fundamental to capitalist planning.72 Yet the ways in which this occurs—the level of precision, the various actors involved, and the strength of the abstractions when it comes to the interactions between markets and law—is anything but straightforward. If transnational land deals have been embraced by governments like Laos because they offered to help strengthen their abilities to effectively govern their territories and populations, they have also articulated with ongoing struggles within the state over the ability to direct and benefit from that process. Territoriality in practice depends on how space itself is rendered calculable.73 Part of the challenge of regulating transnational land access is thus not merely the emergence of new, cross-border legibility regimes like those created by China’s opium-replacement subsidies. It is also that these emerging regimes remain hard to see, and thus difficult to govern, because of the persistent domestic-territorial politics of ongoing state formation.