Puginier, Oliver: Hill tribes struggling for a land deal: Participatory land use planning in northern Thailand amid controversial policies


Kapitel 3. Research Methodology

Based on the four problem complexes identified in the introduction, namely environmental, agricultural/livelihood, policy/institutional and data availability, the methodology also takes into account the failure of the TG-HDP Land Use Planning Teams, so as not to re-invent the wheel. It seems obvious to propagate planning teams composed of stakeholders and government agencies, and the theory clearly advocates such an approach, yet in this case it did not work. This necessitated a re-examination of the two key conditions stated by FAO, namely changes in land use agreed by all parties involved as well as the political will to implement a land use plan. Looking back at the TG-HDP, neither of the two conditions for success applied, as forest department officials stayed away from planning meetings and therefore there was never a land use plan agreed by all parties, let alone a political will to implement it. The consequences of this situation were only realised after fieldwork had started and led to a complete reorientation of the research plan in order to assess the obstacles to a participatory planning approach. Therefore the analysis of the policy situation occupied a major part of the thesis, with a focus on the sub-district (Tambon) level in the current process of decentralisation and options for local initiatives in spite of a lack of political framework. Within this context, the research approach followed a call for the application of “hybrid research“, whereby natural and social environmental science are integrated into a holistic perception of land use planning (forsyth 1998,113). The above components were examined in chronology, yet since planning is an iterative process, the research followed the same path.

3.1 Components of the research approach

3.1.1 Main objectives and local realities

The very complex problem situation shows the diversity of issues and opposing interests in natural resource management, particularly when the competition for land and water are increasing, yet at the same time a centralised government struggles to take a unified planning approach. Therefore the challenges represent an ideal case study for land use planning in the hope to identify ways to overcome this difficult situation. In this context the original technical planning approach was changed to a modified objective once the local conditions were witnessed. The original objective was:

The development of a method to combine the “top-down approach“ of land use planning with remote sensing tools with the “bottom-up approach“ of full integration and participation of local communities, in order to maintain natural resources and to safeguard sustainable, ecological farming systems.

Due to several new developments after fieldwork began, underestimated policy impacts and false assumptions, the objective was modified to:

The assessment of participatory land use planning in the highlands of northern Thailand, with hill tribes as the primary stakeholders and government agencies as the secondary stakeholders, based on the agricultural systems of the hill tribes, the policy framework, as well as institutional platforms for communication.

The objectives were formulated into leading research questions, and two hypotheses. A research plan was formulated as a guideline to examine the following (see Annex IV):

Based on the above objectives, two hypotheses are verified, one of a technical nature and one of an institutional/application nature:

  1. While hill tribe farmers are adapting to permanent agriculture and settlement, they preserve their natural resources at the same time.
  2. In spite of a lacking policy framework, the Tambon will become a communication platform for planning by various stakeholders in the highlands at local level, either formally or informally.

The research was conducted towards the end of the longest running foreign funded development programme in northern Thailand (1981-1998) and had to acknowledge modified local realities of Community Based Land Use Planning and Local Watershed Management (CLM) as well as recent policy developments, summarised below:

Formal or informal organisations: During the process of scaling up land use planning activities, three neighbouring villages initiated a hill tribe network in 1996 (Jantakad 1998, vol.2,54). The network took over some functions of the abandoned Land Use Planning Teams (LUPT), yet with little support from government agencies‘ field staff and no formal official mandate. However, the implementation of decentralisation at sub-district level, with newly forming Tambon Administrative Organisations (TAO), could mean an uncertain future for the hill tribe network.

The above conditions are described in some detail, for they had quite an impact on the research design and the collection of data. What initially focused on the technical support of an on-going extension and planning process changed to examine the impact of incoherent policy and decentralisation much more. Then came the research needs expressed by TG-HDP staff, which were important as the enterprise was set up as project accompanying research. While it is very useful to consider research in terms of the application of results in target areas, this approach also bears the inherent danger of a project evaluation or close identification with project activities, to the extent that the researcher may lose the necessary distance for the assessment of the research plan. It was at times very difficult to keep track of the middle path without too much divergence towards project implementation or the application of an abstract planning model. The resulting outlook of who can use research results in the absence of a project gains importance, particularly since GTZ withdrew from natural resource management assistance in Thailand,


with the exception of the Chiang Mai branch of the Sustainable Management of Resources in the Lower Mekong Basin Project (anonymous 1999). Perhaps this was premature given the current focus on “Good Governance“ that increasingly influences the development debate. Therefore it was important to consider the post-project institutional environment at village and sub-district level to produce results pertinent to local stakeholders in the planning process.

Even though GTZ has more or less withdrawn from natural resource management in Thailand, other organisations continue the participatory land use planning approach. The furthest steps have been taken by CARE Thailand with the establishment of Village Forest Conservation and Watershed Management Committees in Mae Chaem district of Chiang Mai (Anonymous 1997), in which government and village representatives are members and sign land use agreements that use digitised maps as baseline information. So far this is the only documented case where this has led to written documents and could serve as a model to be followed. In a continuation of this approach, a new project phase focuses on the empowerment of these watershed networks as part of the Tambon Administrative Organisations (TAO) under the ongoing decentralisation (srimongkontip 2000). This shows that land use planning has progressed to institutional implementation by local organisations, with ever-stronger demands for a reformed and unified policy framework for the highlands.

3.1.2 Resulting focus of the research approach

The focus of the research approach goes back to the unclear policy situation or lack of a framework as the starting point for the failure of the TG-HDP land use planning teams. The policy development has indeed been very confusing and contradictory, hence it is important to understand the history in order to get a better picture of how it affects the present situation. The chronology of policy development also goes beyond a description in terms of a unisectoral approach like forestry, agriculture or social integration, but shows the interactions and sometimes concurrent realities affecting more than one sector. Particularly from the point of view of hill tribes, issues like permanent settlement, forest use as well as protection, agricultural sustainability, village boundaries and Thai nationality are interrelated. A historical overview therefore serves to better illustrate the state of uncertainty prevalent among the primary stakeholders in the highlands.

A second focus lies on the effects of the CLM approach in terms of agricultural zonation, the identification of watershed forest areas, conflicts with neighbouring villages, and the potential of sustainable planning approaches. Of particular importance is the village boundary, which is a new concept to most hill tribes, and the lack of acceptance of own demarcations at the time of village registration. When it comes to decisions, villagers‘ own demarcations are not accepted by government agencies and can easily be overturned. This also affects areas under shifting cultivation, for long term fallow areas are often considered as protected forest land by the Royal Forest Department (RFD) with its mandate of forest protection. In this context the use of Geographic Information Systems (GIS) was considered to display achievements to date as well as to show inconsistencies in the recognition of village boundaries by government agencies. The appropriateness of digitised maps is also examined in comparison to topographic models at village and Tambon level, including an assessment of their further use in planning negotiations.

The third focus lies on the institutional environment at Tambon level as the platform where centralised planning comes into direct contact with society under the current decentralisation process, represented by elected village leaders. In this context the issue of the viability of formal and informal organisations is of importance, for in Pang Ma Pha district hill tribe leaders founded a hill tribe network among themselves to deal with conflicts over natural resource management. The future of the network is uncertain, given the establishment of Tambon Administrative Organisations (TAO) with often the same members as representatives. For village leaders official government functions are very new and they have to get used to new responsibilities as well as rights as communal speakers with Thai nationality. Additionally, linkages to district planning are examined as well as the extent to which national policy allows or inhibits such approaches.


In an acknowledgement of the local situation, four components have been identified as the cornerstones for a land use planning model pertinent to this thesis (Table 3-1). There have indeed been a number of changes and very promising developments that deserve acknowledgement, most notably in the formation of inter-village land use planning networks. This development has at the same time taken natural resource management issues out of the village and into contact with first neighbouring villages, and later with government planning bodies. It is there that the top-down influences of national policy can be combined with technical means of presentation and display, such as topographic models and digitised land use maps to set priorities and implementation means. By placing a planning platform at the centre this also considers the issue of aggregation and up scaling of information to meet regional planning priorities of the government. Starting with individual villages affected by development activities, a further step was the creation of inter-village networks that later have come into contact with decentralisation policy of the government in the form of TAO membership or sub-district organisations. The key for future planning lies in these communal organisations in the post-CLM period.

Table 3-1: Components of the land use planning model

Farming systems and adaptations

Policy framework development

Community networks and organisations

Planning tools

Representative farming systems in selected villages

History of highland development policies

Village level planning organisations, formal and informal

Past land use perception and display means

Effects of CLM interventions and changes towards permanent farming

Policy changes during the research phase at various government levels

Hill tribe networks and formal government decentralisation by means of TAOs

Topographic models used by the TG-HDP for planning

Future scenarios for agriculture and diversification

Necessary changes for future planning, stumbling blocks

Sub-district level organisations and links to district planning

Digitised maps and 5-year land use plans

3.1.3 Diamond model for planning

The components described above should not be seen as separate entities, for there are constant interactions that have reciprocal effects. Nevertheless, the model starts with an overview of highland development policies to show some of the trends and contradictions that have influenced natural resource management. This is followed by interlinkages between the effects of the TG-HDP planning activities as well as planning tools in the present situation, whereby a planning platform at local level stands in the centre of the model. This represents the process of decentralisation that has increasing influences on highland development. The planning model may thus appropriately have the form of a diamond with interlinkages as shown below (Figure 3-1). The purpose of showing it as a diamond model is a consideration of the various activities that have taken place in the past in terms of agricultural and livelihood changes, policy modifications and effects of the TG-HDP approach of CLM. These activities and developments have affected planning from various sides, often more concurrently instead of sequentially, thus reflecting real life realities. These realities will also continue to affect planning in future, hence they are shown as external influences that will shape agreements and approaches emerging from the central communication platform. A proposed land use planning approach is finally suggested, focusing on the necessary commitments of government agencies for effective implementation at sub-district or TAO level, yet with caution given many uncertainties.


Figure 3-1: Diamond model for land use planning

3.1.4 Challenges for participatory GIS

With the aim to combine technical and participatory means of survey for land use planning, it was attempted to combine Community Based Land Use Planning and Local Watershed Management (CLM) with Geographic Information Systems (GIS). The advantages and risks of participatory GIS for hill tribe farmers as the primary stakeholders and for government agencies are manifold. The concept of a "Participatory GIS" has even been labelled an "Oxymoron" or contradiction in terms, implying that participation of local communities and high technology tools are mutually exclusive. There are several challenges when combining participatory approaches and GIS (abbot et al. 1998,30):

  1. Scaling up to show local concerns as well as broad regional or national perspectives, so that local priorities can be integrated into regional plans.
  2. The access of local people to decision making power through the ownership and use of data, since in the past this access was limited to a few high-level decision makers and thus constituted a merely extractive extension tool.
  3. A land use model or GIS turns local knowledge into public knowledge and out of local control, and can be used to locate resources or extract more taxes.

The application of GIS in Thailand goes back to a World Bank land policy analysis in 1985 (ongsomwang 1993,15), though it was carried out in the USA by a consulting firm. Since then the United Nations Environment Programme (UNEP) carried out an environmental case study in 1987 and the Thailand Development Research Institute (TDRI) in 1988. A more recent assessment of forest inventory methods recommends the combination of remote sensing and GIS for forest management planning, and even proposes a decentralised national forest inventory (ongsomwang 1993,195). GIS was even extended to land use changes in the highlands in order to monitor the effects of population pressure on forest cover and productivity, combining spatial information and socio-economic factors (ekasingh et al. 1996,402). These issues have also been considered in more detail for northern Thailand including areas settled by hill tribes by scientists from the local office of the International Centre for Research in Agroforestry (ICRAF,


Saipothong et al., 1999). The extractive aspects can thus already be analysed in case studies, so the challenge remains for the inclusion of land use classification by hill tribes in order to include their priorities in joint planning approaches.

Given the problems prevalent in the highlands and building on the CLM approach, it was important to document the project achievements and to integrate the land demarcations into a computer database for modification and upgrading for future planning to produce:

The use of a GIS may help to overcome the lack of a common map base for the assessment and management of natural resources such as forests, water resources, protected areas, agricultural land and village locations. The issue of data management and local political interests is a crucial one in the unclear policy framework for highland development. It is has the potential to help in conflict resolution between villagers and government agencies, the assignment of land titles and the determination of sustainable forms of agriculture. Yet at the same time, the threat that the revelation of land use to authorities can backfire for farmers in the form of land confiscation is also very real.

3.2 The field research

3.2.1 Introductory phase

The fieldwork began in March 1997 with a three months introductory phase of familiarisation with the TG-HDP infrastructure, activities and objectives for the remaining final project phase, which included an identification of what contributions the research could make. This included many visits to the various project areas with staff, so as to understand the diversity and status quo of the CLM approach and how the project defines its Participatory Working Approach (PWA). It was also important to become familiar with the highland environment, the farming systems, the culture of the hill tribes and their understanding of the project driven land use planning. There were individual and group interviews in villages by TG-HDP staff and short-term consultants, which were used for mutual familiarisation with farmers. Interviews were conducted with the help of translators in Lahu and Karen language.

3.2.2 Selection of target villages

The next phase was the selection of six target villages (underlined in bold) for detailed surveys of representative farming systems under the CLM approach in both project areas, namely Nam Lang (Figure 3-2) and Huai Poo Ling (Figure 3-3). Nam Lang is the name of the watershed area and gained district status in 1996, upon which it was renamed Pang Ma Pha district and was subdivided into the four Tambon or sub-districts of Pang Ma Pha, Tham Lod, Na Pu Pom and Soppong, inhabited by Shan, Karen, Black Lahu, Red Lahu and Lisu. In contrast, Huai Poo Ling sub-district (Tambon) is inhabited mainly by Karen. Thereby the two traditional shifting cultivation systems in transition to permanent farming were represented, namely pioneer swiddening by Lahus in Pang Ma Pha and rotational swiddening by Karen in Huai Poo Ling Sub-District (Muang District). In Tambon Wawi project activities had ceased and CLM was never practised.


Figure 3-2: Map of Pang Ma Pha district (Nam Lang)


Figure 3-3: Map of Huai Poo Ling sub-district


The following main village selection criteria were used (Table 3-2):

  1. Overlap with areas of the two Thai MSc students as well as the German MSc student (the research grant from TÖB stipulated the inclusion of Thai and German students);
  2. Survey areas located in both TG-HDP project areas to obtain a representation of two different swiddening systems as well as different ethnicities;
  3. Same administrative level of Tambon (sub-district) for data aggregation and assessment of administrative planning structure;
  4. Inclusion of a village outside the CLM target villages for comparison;
  5. Overlap of village areas with protected conservation forests to assess conflicts with the Royal Forest Department (RFD);
  6. Contrast between “key“ and “satellite“ villages to assess differences in decision making on natural resource management. Key villages are officially registered in the government gazette, whereas satellite villages do not have village status and depend on the key village committee they belong to for any government service;
  7. Farming systems diversity in relation to rice, where villagers differentiate between pure paddy farmers, mixed cropping and pure highland farmers.

Table 3-2: Characteristics of surveyed villages (source: TAO offices and TG-HDP)

Village profiles

Tambon Pang Ma Pha (population 2,600)

Tambon Huai Poo Ling (population 3,500)


Huai Hea

Bor Krai (Cha-Aeu)

Pa Charoen

Luk Kao Lam

Huai Tong

Huai Hee


Lahu Sheleh

Lahu Sheleh

Lahu Nyi

Lahu Sheleh

















Area (ha)







Density people/km2








key village 1987, No 8

key village 1996 No 11

satellite of Ya Pa Nae

key village 1988, No 9

key village 1964, No 5

key village 1983, No 8


more than 50 years

20 years

11 years

more than 10 years

more than 100 years

more than 170 years















Group type2











Ger.+1 Thai


2 Thai

  1. Upon registration with the Department of Local Administration (DOLA), a village is given a number, such as No. 8 for Huai Hea in ascending order by registration date. A village is then also given a Thai name like Bor Krai, which had the name of the Village Headman Cha-Aeu. The satellite village of Pa Charoen is only referred to by its key village Ya Pa Nae (No.5). Village registration is a precondition for the provision of government extension services.
  2. Refers to the First Highland Master Plan (1992-1996), in which the Department of Land Development classified hill tribe villages according to their potential for permanent settlement and farming, where type 1 is a permanent village. Type 1 villages receive the most government support, while type 3 may face relocation. This classification was conducted in a very top-down manner, without the involvement of local communities living there.


3.2.3 Digitisation of land use maps

Towards the end of 1997, all available maps at village and sub-district were collected for digitisation and inclusion into a Geographic Information System (GIS). The digitisation was done at the Survey Section of the Northern Narcotics Control Office (NNCO) as well as at the Department of Geography of Chiang Mai University. NNCO is the Chiang Mai branch of the Office for Narcotics Control Board (ONCB). The GIS programme Arc View 3 with baseline data on Mae Hong Son province was obtained from the Department of Urban and Regional Planning of Chulalongkorn University and was presented by Dr. Saengsuan in a workshop at the TG-HDP office in October 1997. The purpose was to assess possibilities to integrate hand-drawn village maps into the programme and to think of ways to use this data for future land use planning activities.

Digitised land use maps were produced using the following procedure. Hand-drawn land use maps were collected in all the 10 CLM target villages of Huai Poo Ling and in three villages in Pang Ma Pha, as only three villages have transferred their land use models onto maps. For Pang Ma Pha it was thus not possible to aggregate the maps into Tambon maps. The village maps were digitised using a hand digitiser into the GIS programme Arc Info and then converted into maps using the map-drawing programme Arc View 3. Contour lines were obtained from the Remote Sensing Centre of Chiang Mai University (CMU) to give a three-dimensional perspective, with 20 m intervals for the village maps and 100 m intervals at Sub-District level. The roads and streams, as well as the Tambon boundaries for Huai Poo Ling were obtained from the Survey section of NNCO in digitised form and overlaid with the remaining data. The different land categories were then colour coded using the same colours as on village maps. Maps were displayed using the Universal Transverse Mercator (UTM) coordinates with grid points in steps of 1 km² for village maps and 5 km² for the sub-district map. The polygons for different land categories were added for area calculations. The same procedure was applied to Tambon Huai Poo Ling and aggregated. Adjacent villages often had overlapping outer user boundaries (marked in pink on the map), which is significant in the case of land disputes and official village registration.

The next step intended to overlay village land use maps with the watershed classification for both project areas, but the Royal Forest Department in Bangkok claimed that this information was not available in digitised form, which seemed strange given the proposal for GIS use in forest management (ongsomwang 1993). At district level it was possible to see the maps, but not to make copies, probably for reasons of political nature. A third component of inclusion of village registration data was more successful, though it was not possible to obtain written criteria from the Department of Local Administration (DOLA) in Mae Hong Son town of how land is allocated to one or the other village. However, village registration documents were obtained for 1995 when Huai Poo Loei separated from Huai Tong as well as Phapuak from Huai Hea to become key villages, and for 1996 when Bor Krai separated from Cho Bo as a key village. For the inclusion of demarcation points on digitised maps, DOLA vectors were converted to UTM coordinates as below:

i. M is the first

ii. B is the second

iii. 215 is the third

iv. 755 is the fourth

Where L = 3, M = 4, N = 5, P = 6, Q = 7 etc. Then the equivalent value for the first part (4) is joined with the third part (215) and 00, resulting in: 4 215 00.

Where S = 16, T = 17, U = 18, V = 19, A = 20, B = 21 etc. Then the equivalent value for the second part (21) is joined with the fourth part (755) and 00, resulting in: 21 755 00.

4 215 00 (X-coordinate) and 21 755 00 (Y-coordinate)

The conversion of vectors to resulting coordinates is summarised (Table 3-3).

Table 3-3: Conversion of vectors used by DOLA to UTM coordinates (based on village registration documents of 27 April 1995 and 2 May 1996)

Huai Hea village (1995)

Bor Krai village (1996)

Huai Tong village (1995)

DOLA vector



DOLA vector



DOLA vector













4229 00

21759 00








4246 00

21762 00








4234 00

21736 00








4208 00

21722 00








4211 00

21730 00








4185 00

21745 00








4195 00

21758 00







Once the maps had been digitised and printed in poster size (A1) at the ONCB Survey Section in Chiang Mai, they were taken back to villages and TAO members for modifications or corrections, before they were shared with government officials. With the closure of the TG-HDP it is likely that the models will not be updated and will disintegrate over time, and the hand drawn maps may suffer the same fate. Digitised printouts can also be distributed to other agencies and can be taken to network or district meetings to discuss land use issues. Maps were also distributed to district forest officials to facilitate their work in land use monitoring. The results from mapping were first presented at the TG-HDP final workshop on natural resource management in June 1998, and eventually the data and the GIS software were then transferred to the Survey Section of ONCB as well as to the ICRAF office in Chiang Mai at the final workshop in February 1999 (see Annex I for events).

3.2.4 Field surveys

From July 1997, joint field surveys with TG-HDP staff were conducted at village level and at site offices for the process of topographic modelling of land use and villagers‘ perception regarding the usefulness of models and hand-drawn maps for planning (Photo 3-1). A series of interviews were conducted without project staff to give villagers the chance to speak more freely on communication and planning problems (see questionnaires in Annex II and III). There were also visits to various government agencies in Bangkok in April 1997 for data collection on policies and mapping approaches, and again in June 1998 for the collection of aerial photographs at the Royal Survey Department. In September 1997 and December 1998 it was possible to join a helicopter flight with the Survey Section of the Office of Narcotics Control Board (ONCB) in order to take pictures of land use from above.

In preparation for the final TG-HDP workshop on Natural Resource Management in June 1998 was the 5-day Community Leaders Cross Visit Programme supported by the Highland Peoples Programme of the UNDP, for which 17 village representatives from project areas of GO and NGO highland development programmes were taken as a group to the various project areas in Chiang Mai and Mae Hong Son province in order to exchange different work approaches and discuss problems. The first village in the Doi Inthanon


National Park (Chomthong district, Chiang Mai) was caught in the land use conflict with lowland Thais that staged a roadblock the day after the group passed (see Chapter 1.1). Other project areas included Mae Chaem district (Chiang Mai) where CARE operates, Pang Ma Pha district in Mae Hong Son (TG-HDP), and Mae Taeng district in Chiang Mai where the Thai-UN Sam Muen Development Programme (SMDP) operated till 1994, as well as a project area of the Thai NGO Hill Area Development Foundation (HADF). These village leaders also participated in the TG-HDP workshop and shared their experiences.

Photo 3-1: Manual transfer of Tambon map to the model in Huai Poo Ling

After the closure of the TG-HDP in September 1998, a final series of interviews were made about the future of CLM under the current decentralisation. This time it was possible to go without project staff, which enabled a much more direct and detailed communication than with project staff, enhanced by communication in the villagers' own language (Photo 3-2). A survey of numbers and types of cultivated plots was conducted for all inhabitants of the 4 target villages in Pang Ma Pha for the purpose of triangulation with areas on maps. The Pang Ma Pha Hill tribe Network was also interviewed about its purpose and operation, particularly on its future given the formation of Tambon Administrative Organisations (TAO). In Huai Poo Ling the TAO is at a very early stage of formation, so the focus was on land use changes and the relevance of land use models. Interviews were carried out on:

Resource availability and food sufficiency;

Land use changes during the last 10 years;

Land categories and land use based on villagers assessments;

Village boundaries and conflicts;

Farmers strategies to solve land scarcity problems;

Village regulations for natural resource management and land use.


Photo 3-2: Interviewing a village elder with a Karen interpreter

The field work ended with a final workshop on 18 February 1999 at Chiang Mai University with 20 participants to present preliminary research results, followed by a four-day field trip to the project areas to discuss the impact of CLM with hill tribe farmers and to assess the future of land use planning.


© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.

DiML DTD Version 2.0
Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML - Version erstellt am:
Wed Aug 28 15:14:57 2002