Project Management Discussion

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=tsdw20

The International Journal of Sustainable Development and World Ecology

ISSN: 1350-4509 (Print) 1745-2627 (Online) Journal homepage: http://www.tandfonline.com/loi/tsdw20

Why did the project fail to achieve its objectives in some villages? The experience of the Upland Settlement Project (USP) in Bangladesh

Tapan Kumar Nath & Makoto Inoue

To cite this article: Tapan Kumar Nath & Makoto Inoue (2008) Why did the project fail to achieve its objectives in some villages? The experience of the Upland Settlement Project (USP) in Bangladesh, The International Journal of Sustainable Development and World Ecology, 15:2, 153-169, DOI: 10.1080/13504500809469780

To link to this article: https://doi.org/10.1080/13504500809469780

Published online: 20 Sep 2010.

Submit your article to this journal

Article views: 103

Citing articles: 3 View citing articles

Why did the project fail to achieve its objectives in some villages? The experience of the Upland Settlement Project (USP) in Bangladesh†

Tapan Kumar Nath1 and Makoto Inoue2

1Institute of Forestry and Environmental Sciences, University of Chittagong, Bangladesh 2Department of Global Agricultural Sciences, Graduate School of Agricultural and Life Sciences, University of Tokyo, Tokyo, Japan

Key words: Chittagong Hill Tract, Upland Settlement Project, planters, livelihoods, social capital, NGOs

SUMMARY By sampling an upland settlement project (USP) village, this study examined why the project failed to achieve the planned objectives in some villages, and what livelihood strategies villagers (planters) have adopted to sustain their livelihoods. The paper uses data obtained by qualitative and quantitative methods. The USP was formulated with the objectives of settling 50 landless and marginal tribal families in a project village and intro- ducing agroforestry and rubber plantations for short- and long-term socioeconomic development of planters, in a participatory manner. There are now only 15 planters in the project village and, although agroforestry has progressed well, only 30 to 40% of poor quality rubber trees remain. Planter participation in project activities was limited to wage labour, indicating that the USP failed to adequately achieve all its proposed objectives. In the face of project failure, planters have adopted diversified strategies to maintain their livelihoods. Several shortcomings caused failure of the project, such as the inability of project staff members to organize planters, failure to create awareness among planters about project benefits, lack of regular project money disbursement, sociocultural igno- rance, and lack of social capital among planters. Recommendations are made for securing effective participation of local people to assure successful and sustainable project out- comes. Lessons learned will be helpful in formulating future development programmes in this country and elsewhere.

INTRODUCTION

Since the 1980s, many state-sponsored and donor- funded development projects have been imple- mented in developing countries for the welfare of

rural people and the sustainable management of natural resources. Even though political and financial support for development exists, and the

International Journal of Sustainable Development & World Ecology 15 (2008) 153–169

Correspondence: Tapan Kumar Nath, Department of Global Agricultural Sciences, Graduate School of Agricultural and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, 113-8657 Tokyo, Japan. Email: [email protected], [email protected]

153

†Based on a paper presented at the 117th Japanese Forest Society conference, Tokyo Agricultural University, Tokyo, Japan on 2–4 April 2006.

best policies have been advocated, such develop- ment intervention has a high incidence of failure (Agunga 1992: 3). Moreover, development practi- tioners are well aware that even if programmes have the same level of overall assistance, the results vary considerably from location to location (Krishna 2004: 292). Several reasons, such as leadership and programme staff effectiveness, have been suggested to explain the observed differences, and social capital is one other possible explanation that must be considered (ibid.: 292).

Social capital is a prerequisite for sustainable management and development of natural resources (Pretty 2003: 1). Many donor-funded projects have sought to improve their effectiveness and efficiency by sponsoring the formation of social capital in the form of groups to meet immediate project objectives (Carney 1996). Groups or organizations provide the entry point for efforts to work with community members (Schneider 2004). For sustainability, rural development programmes can be specifically designed to promote emergence of organizations and to enhance their effectiveness (de Janvry 2003). Farmers’ organizations may engage community members in a common income-generating activity to meet individual and social needs and to develop procedures to ensure the accountability of their leaders (Bingen and Munyankusi 2002). It is argued that external factors and interventions might be in- effective in marginal areas unless farmer communi- cation and cooperation relationships are given more attention (Wu et al. 2002: 60). Social capital and the capacity of village leaders have a synergistic impact in influencing development performance (Krishna 2004: 300).

In rural areas, overall development success or failure is often an outcome of what happens in agriculture, which contributes significantly to livelihoods by raising income and employment (Addison 2005). Sustained livelihoods for the poor ensure environmental sustainability, which judi- ciously combines their use of natural resources with alternative opportunities for employment and income (Osmani 2005). However, outcomes of agricultural projects are unsustainable when no attention has been paid to institutional develop- ment and local participation (Pretty 2003: 6). Sustained community participation in develop- ment interventions helps to maintain resources and the flow of accrued benefits over time (Sinha and Suar 2005: 126). The success or failure of projects in

any particular community is in the hands of com- munity members and, for projects to be sustainable, all community members must be involved, or at least considered, in the decision-making process, and all must be able to share in the benefits (Ikerd 2001). Investment is needed in participatory pro- cesses to bring people together to deliberate on common problems, and to form new groups or associations capable of developing practices of common benefit (Pretty 2003: 21).

In this paper, we focus on a government- financed development project, the Upland Settle- ment Project (USP) in the Chittagong Hill Tracts (CHT), Bangladesh, which was intended to rehabil- itate degraded forests in the region and improve the livelihoods of landless and marginal hill people. Taking a sample project village as a case study, this research attempts to answer two questions: 1) why the USP failed to achieve target objectives in some villages, and 2) what the villagers (officially known as planters) are doing to sustain their livelihood in the face of project failure. The answers to these questions are described both qualitatively and quantitatively, and analysed followed DFID’s sus- tainable livelihood (SL) framework (DFID 2001).

METHODS

The project

Increasing population pressure due to settlement of non-tribal lands, construction of development infrastructure (e.g. Kaptai hydroelectric dam), indiscriminate deforestation depriving tribal people of their source of income, and other so- called development projects have had profound effects on CHT indigenous communities (Dasgupta and Ahmed 1998; Khan and Khisa 2000; Nath et al. 2005). The government of Bangladesh therefore initiated multi-sectoral development projects for the socioeconomic advancement of tribal people through community empowerment and capacity building to strengthen resident local institutions and create new infrastructure (Dasgupta and Ahmed 1998). USPs were ranked top as they have been successful in rehabilitating landless and mar- ginal jhumias (shifting cultivators) through agro- forestry technologies and social development activities such as capacity building and infrastruc- ture development (Dasgupta and Ahmed 1998; Khan and Khisa 2000; Khan et al. 2003). The Asian

Evaluation of an upland settlement project Nath and Inoue

154 International Journal of Sustainable Development & World Ecology

Development Bank funded the first phase of the USP between 1985 and 1993, settling 2000 tribal families and establishing 1,620 ha of homestead agroforestry and 3,240 ha of rubber plantations in 39 project villages in the Khagrachari district, Bangladesh. Considering the success of the first phase, the second phase of the USP was formulated with the following objectives (Anon 2001):

1. Settlement of landless and marginal tribal farmers/jhumias in suitable areas in Khagrachari and Bandarban districts;

2. Development of suitable upland areas (which are currently unused) for diversified cropping with horticulture, agroforestry, and rubber plantations; and

3. Long-term socioeconomic development of target groups.

The short-term objective is to improve economic conditions of the landless and marginal tribal farmers of the area through:

1. Increased horticultural and rubber production;

2. Raising farm income and creation of rural employment; and

3. Enhancing local participation in project implementation activities.

This phase is being financed by a grant from the Ministry of CHT Affairs. The project has been

extended to 2004–2005, and the project manager is trying to extend it to 2006–2007 in order to complete the establishment of rubber processing units. In this phase, 506 ha of homestead agro- forestry and 1,620 ha of rubber plantations have been established. Each family received 2.1 ha of degraded forest land, of which 0.5 ha is for a home- stead and agroforestry and 1.6 ha is for a rubber plantation. All families in every project village developed a rubber plantation as a contiguous block of 81 ha (1.6 ha per family).

Even though earlier studies claimed successful intervention, our recent studies (Nath et al. 2005) and very recent field visits (May through August 2005) show that the USP has had both successes and failures. Based on the opinions of the project manager and other staff members, discussion with some project villagers, and reviews of the project objectives, we defined the villages as follows:

1. Relatively successful: villages where the pro- ject objectives are adequately achieved, and

2. Relatively unsuccessful: villages with inade- quate progress in achieving the project objectives.

Study village, data and methods

We deliberately conducted the study in the second phase of the USP in Bandarban because: a) we assumed that planters would not recall project his- tory and activities from the first phase that ended in

Evaluation of an upland settlement project Nath and Inoue

International Journal of Sustainable Development & World Ecology 155

Topographic feature: Hilly topography Land categories: i. ii. iii.

First class: Plain land used for irrigated rice cultivation, very limited Second class: Somewhat elevated land used for rainfed rice cultivation, fairly available Third class: Upland suitable for forestry practices, available

Vehicle accessibility: Pitched approach road linked to main road Distance to main road: 2.5 km Distance to market: 8–9 km Distance to district headquarters: 9 km Distance to nearest medical facilities: 8 km Electric power: Available where absent planters live Source of lighting: Mostly kerosene, and electricity in some absent planters’ residences Fuel source: Wood Drinking water source: i. ii.

Major source: Communal water supply in absent planters’ village, and streams for resident planters’ village Minor source: Streams/ring well and tube well for resident planters

Distance to government schools: 1–9 km

Table 1 Typology and features of the study USP village, Bandarban, Bangladesh

1993; b) the project authority applied their experi- ences in making the second phase successful; and c) as the second phase is ongoing, planters can provide recent and accurate information.

Selection of study village

Due to the remote locations, lack of accommoda- tion, poor communications and social unrest, the project manager suggested that we conduct the research in Bandarban instead of the Khagrachari district. We asked the project manager to categorize the 10 project villages into relatively successful and relatively unsuccessful villages based on their attain- ment to date of project objectives. The manager provided a list categorizing the villages, which we informally discussed with field staff members. Finally, we visited villages and discussed this catego- rization with villagers. There were similarities in the opinions of the project manager and staff members; however, our discussions with villagers and one project staff member uncovered differ- ences. For confirmation, we talked with more villagers and found that their observations con- trasted with those of the project staff. Where the manager and staff had divided villages into seven relatively successful and three relatively unsuccess- ful villages, our findings were that four villages were relatively successful and six villages were rela- tively unsuccessful. From the six relatively un- successful villages, we selected one for in-depth study (Table 1).

Livelihood analysis

The concept of livelihood has been widely dis- cussed (e.g. DFID 2001; Cramb et al. 2004; Soini 2005) in terms of poverty issues, livelihood strate- gies and diversification, and impacts of develop- ment intervention on the rural poor. This paper follows the SL framework of the Department for International Development (DFID 2001) in study- ing the livelihoods of the planters. The SL frame- work provides new ways to address rural poverty issues beyond the traditional methods of employ- ment generation and target group programmes, and helps to examine livelihood issues in relation to five capital types (physical, financial, natural, human and social) identified in the framework (Messer and Townsley 2003; Reddy et al. 2004).

The framework is useful for explaining interrelationships among different livelihood assets, and utilization of assets in diversifying liveli- hood strategies to attain desired outcomes (e.g. increased income, stable natural resource base, etc.) in the enabling environment (e.g. policies, institutions, etc.).

Data and methods

We collected quantitative data and qualitative infor- mation for pluralism of the methodology and to minimize weakness in the methods. For quantita- tive data, we conducted household interviews, which we originally intended would entail a partici- patory wealth ranking exercise to stratify all house- holds into three categories: wealthy, medium and poor. However, our discussions with key infor- mants (four persons) and project staff members explored other potential criteria for household classification:

1. Resident planters: those presently living within the project village (15 planters);

2. Absent planters: those who had left the pro- ject village, but are still officially recognized as planters and live in nearby villages (23 plant- ers); and

3. Missing planters: those who have permanently left the project village and whose where- abouts are not known by the authorities (12 planters).

Key informants also said that all resident planters are very poor in comparison with absent planters, because the former only have project land and all are wage labourers. We therefore divided all planters into two groups for household interviews:

1. Resident planters: eight households,

2. Absent planters: 11 households.

Quantitative data

For quantitative data collection we used a question- naire containing some structured questions, and a checklist for semi-structured interviews. The ques- tions were designed in a way that helped collect data on all five assets of the livelihood framework as

Evaluation of an upland settlement project Nath and Inoue

156 International Journal of Sustainable Development & World Ecology

well as some variables on governance issues. We used one questionnaire for each household and carried out interviews over several weeks in August to September 2005.

To assess household expenditure, we asked about weekly expenditure on items such as food. We found that interviewees could recall weekly expenditure more accurately than monthly or yearly. Weekly figures were converted to an annual basis. For income, we asked how much they had earned in the previous year from different sources. Even though this income-expenditure estimation could not produce exact figures, it provided a general picture. To calculate the costs of agricul- tural activities, we asked how much they spent annually for planting materials, fertilizer, labour and other inputs. To calculate the monetary value of physical assets (e.g. televisions, cassette players and water pumps), we asked what available physical assets they have, and the estimated monetary value, and summarized these to obtain the mean equiva- lent value. To measure variables of governance and social capital, we used ordered scales.

Qualitative information

We separately conducted two group discussions with two planter groups, talked with people of different ages, held informal discussions with pro- ject staff, and made personal observations. Group discussions had four or five participants and high- lighted issues such as problems, local organizations and activities, conditions of rubber plantations, role in rubber management, and livelihood diversi- fication strategies. A separate semi-structured ques- tionnaire was used to facilitate the discussion. Talks with project staff dealt with project issues such as planter selection, land tenure, rubber plantation management, present plantation condition and benefit sharing. We talked with people of different ages about social issues such as social relations, con- flicts, the leader’s activities, education and daily activities. We visited their agroforestry homesteads and checked species composition and the number of trees including saplings. We summarized all quantitative data into averages and percentages, determined standard deviations and conducted an unpaired t test to find statistically significant differ- ences among means of different variables of two categories of planters.

RESULTS

How and why the project failed

Objective 1: Settlement of landless and marginal tribal farmers/jhumias

According to official records, 38 Marma and 12 Tonchangya planters were settled in 1994–95 in this project village. Only eight Tonchangya and seven Marma planters are now living in the project village. The other planters left the project village and returned to their original village for the following reasons:

1. Lack of employment and agricultural land. Three to four years into the project there were almost no employment opportunities. More- over, there was little demand for wage labour in neighbouring areas. Due to a land shortage in the project village, they could not produce rice. Planters who had no land and/or who could not practice jhum (shifting cultivation) for the production of staple food (rice) and other agricultural products left the project village to find livelihoods in their original villages.

2. Cultural problems. Marma and Tonchangya speak different languages and have dissimilar cultures. For example, Marma people live closely in a hamlet; do not allow the corpses from outside the hamlet to re-enter; and feed mothers only vegetables for 3 months after childbirth. These practices conflict with Tongchangya culture. Even Marma people from different localities have divergent cul- tures and languages. In the project village, houses are sparsely located in hilly areas of more than 100 ha. Even though there were no distinct conflicts, planters left the project village to avoid possible future conflicts. However, some Marma planters who came from other areas still live in the project village because they have no land outside the project.

3. Child education. There is a primary school about 1 km from both planter groups. How- ever, there is a steep slope from the school to the village that is difficult for the children to climb. This is another reason why planters left the village.

Evaluation of an upland settlement project Nath and Inoue

International Journal of Sustainable Development & World Ecology 157

Hence, the project could not attain its first objective of settling nomadic tribal people in a permanent village. Moreover, like other USP villages, planters do not receive permanent land title (kabul). With- out this, they feel unsettled and believe the govern- ment may take the land in future.

Objective 2: Development of degraded upland through homestead agroforestry and rubber plantation

The project authority supplied all initial planting materials and fertilizer. Subsequently, planters invested their own resources for its development. Our observations indicate successful establishment of agroforestry in homestead areas. We identified 30 and 23 different plant species in homesteads of resident and absent planters, respectively, and the mean numbers of trees per ha were 1154 and 804, respectively. Planters now sell fruit and trees in markets. Many young saplings have been planted in the last two or three years. Planters are aware of the importance of homestead agroforestry that provides them not only with produce for house- hold consumption but also cash.

However, the project authority raised 81 ha of block rubber plantation intercropped with bananas, papayas and pineapples in the five years from 1995 to 1999. Planters earned Tk. 47,000 (US$1 = Tk. 65 in 2005) from selling intercrops and deposited this in a bank account controlled by the project village committee. Three to four years after planting, an accidental fire burned a large portion of the rubber plantation. Moreover, rats damaged roots, and young plants died. Because project money is not regularly disbursed, the authority could not re- plant the damaged area. Even though initially pro- ject staff visited the area frequently, the rubber plantation was later further destroyed due to lack of regular monitoring and maintenance. Our observa- tions and the opinions of planters and project staff reveal that now 30 to 40% of the rubber plantation is in very poor condition. Planters (50% resident and 18% absent) practice jhum inside the planta- tion. Planters told us that they are not interested in rubber because they realized no benefit from rub- ber despite wage labour, and think that the autho- rity will take all benefits from the rubber. This shows that the authority could not motivate planters on the importance of rubber. The authority plans to start tapping rubber this year (2006), but the pres- ent condition of the rubber trees fuels doubts about

latex production and, hence, increased income and employment for planters through rubber pro- duction. The authority did not adequately achieve its second objective, through which it hoped to improve the long-term socioeconomic circum- stances of the planters.

Objective 3: Local participation in project implementation activities

The USP’s intent was planter capacity-building through participation in project activities. How- ever, in reality planter participation was nominal, and their involvement was in the form of wage labour. The manager made all decisions related to project activities in the villages. A three-member executive committee, field superintendent as co- ordinator, field assistant and project village leader as members, develops work plans and implements all village-level activities according to the manager’s decisions. Other than leaders, no planters are con- sulted in developing work plans. Even leaders do not discuss matters with other planters, but decide the number of workers to be engaged.

Selected governance issues

Equity

During the early stages of the project, all planters, their wives and other work-capable family members had jobs (as wage labour) in project activities. After establishment of the rubber plantation there were, in a real sense, no jobs. Under the project proposal, only landless and marginal farmers were supposed to be selected as planters by a specialized selection committee formed by executives from the project authority, the sub-district officer, settlement offi- cer, local government council chairman, headman (leader of mouza), and village leader. However, some better-off farmers among the absent planters who possessed the best agricultural land were also selected. One project staff member told us:

‘Planters were chosen hurriedly and hence real landless people could not be selected. Many people had no idea how they were selected as planters. The headman and leader selected planters who were known to them . . . As the project started and its budget was released, we had to complete the work

Evaluation of an upland settlement project Nath and Inoue

158 International Journal of Sustainable Development & World Ecology

quickly. If we judge the actual criteria of planters, it would take a long time to select planters for 10 project villages. Ultimately this would hamper the completion of other project activities such as infrastructure devel- opment and the rubber plantation.’

Because of such biased selection, when there were full-stream project activities, planters took their benefits and then left the village, while still possess- ing project land. If planters had been properly selected, real landless farmers could have had land and project benefits.

Accountability and transparency

Our discussion with the project staff and planters explored the lack of accountability in handling project money. They commented:

‘If the budget for a job was Tk. 1000, the man- ager allocates 60% of that money to field staff. The field staff then gave 40% of that money to the village leader to perform all activities by employing planters. The leader, however, did not spend more than 20 to 30% of the allo- cated money.’

This shows that much of the project money was mis- handled, and the project could not fully benefit the targeted poor planters. Leaders said that after the completion of jobs, all remaining money should be deposited in the village common fund, but planters feel that this was done only rarely.

We observed a gap in information flow among planters, leaders and project staff. Planters were ignorant of project activities, budget allocations and expenditures. Only leaders and staff handled these functions. Planters did not even know about rubber garden benefits, or who would manage plantations after the project period, and how. The project manager and the staff said that a central management unit (CMU) would be formed to look after the rubber plantations, production and over- all management after the stipulated project period. This management unit will manage the rubber plantation for 40 years and then hand the land over to the planters. The benefits, after deducting all management costs, will be distributed to the planters. However, only project staff know about this policy, while planters remain in the dark.

Responsiveness

In accordance with the project plan, planters were supposed to be encouraged to develop horticulture on their homesteads to fulfil short-term objectives. There was no specific policy on species selection or planting design. The authority supplied planting materials based on the planters’ choices. Planters said that due to inappropriate planting, most of the plants died after one year. As these planters were jhumias, they were not skilled in horticulture. They opined that the project staff visited the rubber plantations frequently, but very rarely visited their agroforestry homesteads. As a result, planters realized no short-term benefits from their agro- forestry. However, as time passed, planters who stayed with the project gained experience, and the conditions of homestead agroforestry improved.

Participation

Even though planters had no part in project design and implementation, some attended project meetings several times. Over the project period’s three to four years, 50% of resident and 55% of absent planters attended 6.63 and 3.33 times, respectively, meetings held in a temporary office near the village. These were to motivate planters to live together harmoniously and plant trees on their homesteads. If we compare project meeting indices (measured on a five-point scale: 1 = disagree strongly to 5 = agree strongly) with past local meet- ing indices, we find that planters were not satisfied with project meetings and the deliberative quality (Figure 1). As time passed, absent planters orga- nized themselves with some other villagers for their own interests and led strong collective efforts (see below). All absent planters are very conscious of what is shown by their local meeting indices, and all three indices are significantly higher than those of resident planters (Figure 1), signifying that all members of their organization have almost equal access in meetings, and are satisfied with meeting discussions, feel comfortable, sense convenience, and enjoy good quality deliberations.

Resident planters are not involved with this organization, but are members of the project village committee. As a member of the project village, the leader (also the leader of the absent planter organization) invites them to attend meet- ings such as those for religious festivals. We found

Evaluation of an upland settlement project Nath and Inoue

International Journal of Sustainable Development & World Ecology 159

that resident planters attended 2.81 times whereas absent members attended 18 times in the most recent year. The variation was highly significant (P < 0.0000). However, the role of resident planters in decision-making is negligible: 25% are somewhat active, but 75% do not take part in decision-making processes. Only 18% of absent planters do not participate, but other people participate actively in reaching decisions. All absent planters are satisfied with the leader’s activities and responsiveness, whereas more than 75% of resident planters mention their dissatisfaction, saying:

‘The leader does not come to our village, and he does not involve us in any social develop- ment work except religious festivals. When rubber garden work starts, he employs more labour from his own village. He employed a rubber garden guard from his village even though we live very near to the plantation. He does not inform us about the committee funds, and we don’t know how much money remains in the bank account or what he is doing with this money.’

They point out bias in allocating project jobs, and a lack of transparency in common fund manage- ment. We finally asked all sampled planters why they joined the project. Across two categories, all planters cited land, inspiration by project staff, and leaders’ suggestions. All resident planters gave equal weight to perceived benefits from rubber and improved livelihood through project functions, but 64 and 72% of absent planters, respectively, mentioned these benefits. Moreover, among

resident planters, 88% also cited money and houses as other reasons for joining the project, whereas 54% of absent planters cited these reasons. There- fore, it seems that some absent planters were well-off because they had less interest in project money and houses, rubber plantations and liveli- hood improvement.

Livelihood analysis

We analyzed the state of livelihood capital (human, physical, natural, financial and social), livelihood diversification and strategies, mean annual income- expenditure and the role of social organization and NGOs in livelihood improvement.

State of livelihood capital

We examined five different capital types in the light of a sustainable livelihood framework (Table 2). Household members of sampled resident and absent planters were 51 and 57, respectively, with an average of 6.38 and 5.18, respectively. Male-female ratios were 1:1.31 and 1:1.28, respectively. Total literacy, secondary education, female education and current school enrolment were higher among absent planters than resident planters. Cash- contributing members were almost equal in both groups. Findings on human capital indicate that absent planters are more conscious of popula- tion growth and female education. Government- sponsored fellowships and World Vision (an NGO) assistance motivate them to educate girls. However, most boys quit after primary school and become involved in income-generating occupations such as agriculture and small village tea stalls.

For physical capital, we assessed housing condi- tions, household appliances such as televisions, water pumps and agricultural implements, and then estimated their equivalent monetary value. All houses are built with bamboo mats and walls, and roofs thatched with sun grass (88% of resident and 36% of absent planters’ houses) or tin (12 and 64%, respectively). Two absent sampled households have a television and one resident has a cassette player. One absent planter also has a water pump and a rice mill. All planters possess agricultural tools. Though there are remarkable differences in equivalent monetary values, this was not statistically significant; higher prices of water pumps and rice mills make the difference.

Evaluation of an upland settlement project Nath and Inoue

160 International Journal of Sustainable Development & World Ecology

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

S a ti

sf a c ti

o n

C o

m fo

rt

D e li

b e ra

ti o

n

S a ti

sf a c ti

o n

C o

m fo

rt

D e li

b e ra

ti o

n

S a ti

sf a c ti

o n

C o

m fo

rt

D e li

b e ra

ti o

n

Pas t local meeting Project meeting Pres ent local

meeting

M ee tings

In d

ic e s

Res ident planters

Abs ent planters

Figure 1 Indices of meetings in the past, during the early project period, and current local meetings

Evaluation of an upland settlement project Nath and Inoue

International Journal of Sustainable Development & World Ecology 161

Planters

Capitals Resident Absent Sig. level

Human capital Mean household size (no.)a

Male: female Members ≤ 11 years (no.) Members 11–59 years (no.) Literacy (%)

Total Primary Secondary

Male/female literacy (%) Current school enrollment (No.)

Primary (Male/female) Secondary (Male/female)

Mean members contributing Cash income (no.)a

6.38 (± 2.82) 1:1.31 18 30

27 22 12 30/25

6/4 1/– 2.6 (± 1.4)

5.18 (± 1.40) 1:1.28 12 44

37 16 29 36/41

3/5 2/8 2.73 (± 0.78)

Physical capital Bamboo-matt wall (% of households) Sungrass roof (% of households) Tin roof Television (% of households) Cassette player (% of households) Rice mill (% of household) Water pump (% of households) Agricultural tools (% of households) Mean equivalent money (Tk.)a

100 88 12 – 12 – – 100 875 (± 710)

100 36 64 18 – 9 9 100 22513 (± 3430) 0.20

Natural capital Mean landholding (ha)

Own farmland (ha) Leased farmland (ha) Jhum plots on khas land (ha) Jhum plots on project’s Rubber garden (ha) Project land (ha)

2.79 – – 0.75 0.85

2.1

2.76 1.58 0.49 0.28 0.16

2.1

0.90

Financial capital Livestock (no.)b

Cows Goats Chickens Pigs

Food security (% of household) Surplus Sufficient Shortage

2 (13) 2 (25) 12 (88) 1 (25)

– 12 88

2.25 (36) 2 (45) 3.2 (82) 1.33 (55)

28 36 36

Note:aFigures in parenthesis indicate standard deviations bFigures in parenthesis (other than above notes) indicate the percentage of households

Table 2 Household livelihood capital in the study USP village, Bandarban, Bangladesh (Source: Field survey, Aug.–Sept. 2005)

Although there is no significant difference in mean landholding, in measuring natural capital, resident planters have no leased or personal agri- cultural land (Table 2). Besides project land, they occupy government khas land (50% resident and 27% absent planters) and project land for jhum. This reveals that all resident planters practice jhum. On the other hand, 73% of absent planters practice irrigated agriculture and some (45%) carry out jhum on khas as well as project land. This suggests that the USP could not prevent jhum and contra- dicts the observation of Khan and Khisa (2000) that ‘the project [USP] attempts to rehabilitate and develop some impoverished indigenous ethnic people, who have hitherto been engaged in shifting cultivation.’ However, it supports the findings of our earlier study (Nath et al. 2005), in which records from 15 USP villages indicate that planters still practice jhum in both khas and project land.

We measured financial capital in terms of live- stock, loans and food security. As with other capital sources, absent planters have adequate financial capital. In the case of cows, some planters were owners and others had access through different institutional arrangements. For example, one resi- dent cares for two cows of a village trader and for this receives Tk. 50 per month and can use the manure for fertilizer. Moreover, in emergencies he can get a loan from the trader. Some absent planters have cows from GROUS (a UNDP funded project). Absent planters had fewer chickens than resident planters, and said that before the monsoon season, most/many chickens die from diseases. Credit facilities are not available in the project village, but four absent planters (36%) obtained credit and a loan from the IDF (NGO), with an aver- age loan size of Tk. 6750. When we asked about food security, 28% of absent planters reported that they have substantial surplus rice each year that they can sell in markets, while 12% of absent and 36% of resident planters have sufficient food, and the remaining 88% of resident and 36% of absent planters face food shortages for 4 to 8 months of the year. Food security status indicates that jhum could not produce enough food for absent planters; how- ever, one family with 11 members can practice jhum on large areas and produce sufficient rice and other products.

In measuring social capital, we considered groups and networks, trust and solidarity, social cohesion and inclusion, information and

communication, and empowerment. The differ- ences in these variables between resident and absent planters were most striking. For groups and networks, we used six variables (Table 3). We found highly significant differences between resident and absent planters in terms of number of organiza- tions involved, and number of household members belonging to organizations and household commu- nication networks (HCNs). Only one member of the resident planter group was involved in the pro- ject village committee. In addition to this com- mittee, an average of 1.27 members of the absent planter group were involved in four other organiza- tions (IDF, GROUS, World Vision and Green Hill). HCNs (i.e. number of households, including rela- tives or neighbours with whom a family could share problems, request help or seek emergency loans) were more than twice as high in absent planters than in resident planters. These variations indicate that absent planters have more connections not only with their relatives and neighbours, but also with different organizations. As a result, they have access to more social services than resident plant- ers. All absent planters believe that at least three households (mostly from their village) are willing to help, and 1.86 households are currently able to loan money (Table 3), whereas only 63% of resi- dent planters think that 2.5 households could help and 1.6 households could provide loans. These people, however, are local traders and provide loans to resident planters on the basis that planters will supply labour for their businesses. For example, one resident planter obtained a loan of Tk. 1100 from a timber trader in 2004 provided that he and his wife supply labour in collecting wood. As of August 30 2005, they have repaid Tk. 800 with labour and have to supply labour for another Tk. 300. These planters are poor and cannot help each other. Moreover, they lack feelings of connectedness and trust.

We found distinct variations in trust and solid- arity between the two groups of planters (Table 3), e.g. trust among people living in the villages, trust in lending or borrowing money, trust of people in the same ethnic group, trust of the village leader and his responsiveness, and of village group mem- bers are significantly different between resident and absent planters. The resident planters have come here from different places and therefore have no family relationships, and hence have little faith in others, even people of the same ethnic

Evaluation of an upland settlement project Nath and Inoue

162 International Journal of Sustainable Development & World Ecology

group. On the other hand, absent planters have been living in their village from childhood, know each other, have family relations and are organized among themselves and place more trust in these

relationships. Further, due to increased access to social services from different NGOs, active partici- pation of villagers in social development projects, and people’s willingness to help each other, the

Evaluation of an upland settlement project Nath and Inoue

International Journal of Sustainable Development & World Ecology 163

Planters

Variables Resident Absent Sig. level

Groups and Networks No. of organizations involved No. of members involved Household communication network No. of people willing to help by giving money in emergency No. of people currently able to provide this money No. of people who helped solve personal problems last year

1 1 2.25 2.5(63) 1.6(63) 1(25)

5 2.27 5.41 3.0(100) 1.86(100) 1.5(36)

0.0000 0.0035 0.019

Trust and Solidarity Opinion on statementsa

Most people living here can be trusted One has to be alert or someone is likely to take advantage of you Most people are willing to help if needed People generally do not trust each other in lending or borrowing

money Trust in following peopleb

People of same ethnic group People from other ethnic groups Village leader Leader responsiveness Project staff Project staff responsiveness Village group members

Level of trust in last five yearsc

Getting better Getting worse Unchanged

2.87 4.13 3.00 3.50

3.13 3.00 2.50 2.50 4.00 4.00 3.00

– 87 13

2.09 4.09 3.00 4.09

4.18 3.00 4.18 4.18 4.00 4.00 3.91

91 – 9

0.0001 0.82

0.0069

0.0000

0.0000 0.0000

0.0002

Social Cohesion and Inclusionc

Feeling of togetherness or closeness Somewhat distant Neither distant nor close Somewhat close Very close

Social stratification Neither great nor small extent Small extent

Times visited relatives last year

38 62 – –

– 100 4.19

– – 55 45

100 – 6.42 0.011

Note: a = Five-level scale [1 = agree strongly, 2 = agree somewhat, 3 = neither agree nor disagree, 4 = disagree somewhat, 5 = disagree strongly]

b = Five-level scale [1 = to a very small extent, 2 = to a small extent, 3 = neither small nor great extent, 4 = to a great extent, 5 = to a very great extent]

c = Figures indicate percentages of households; Figures in parentheses indicate percentages of households

Table 3 Some selected social capital variables of the study USP-village, Bandarban, Bangladesh (Source: Field survey, Aug.–Sept. 2005)

level of trust among absent planters is improving, whereas the level is falling among resident planters. Most resident planters feel neither close nor distant and some said, ‘we all think for ourselves, nobody communicates with others, and there are no relations among us.’ They therefore rely mostly on project staff for information related to farming systems, while absent planters get farming-related information from relatives and neighbours as well as from NGOs.

We then asked whether members are happy, the impact of the project on their livelihoods, and their power to make decisions that change livelihoods. Even though they face economic hardship after 3 to 4 years of the project, more than 75% of sampled planters feel moderately happy and the project has had a big impact on their livelihoods. They obtain land (though without formal title), perform agro- forestry, and rubber plantations will benefit them through employment and benefit-sharing in the future. Due to insecure employment opportunities and a lack of irrigated agricultural land, 62% of resident planters cannot make decisions that might change their livelihoods, and 38% are unable to make such decisions. On the other hand, 36% of absent planters are mostly able to make decisions about their livelihoods and 64% are neither unable nor able.

Social organization, NGOs and collective action

Resident planters have no organizations. Absent planters along with other villagers formed a social organization, Tunkhyan Para Samaj Unnayan Samiti, or village development committee in 2001–2002. These villagers are planters from different USP villages. Previously there were Kheyang (Buddhist temple) social organizations related mostly to reli- gious festivals. The present organization (popularly known as a committee) consists of members mostly of adults (20 to 40 years old), and has a 13-member functional committee. All villagers are general members. A woman is the vice-president and two other women belong to the functional committee. In an annual general meeting, all villagers select functional committee members based on educa- tion, character, responsiveness, sincerity and other attributes. The committee calls monthly meetings and informs villagers about the agenda in advance. The president (who is also the USP village leader)

presides over meetings. All villagers can equally participate in discussion on agenda items, after which consensus decisions are reached. However, if a fair decision is not reached, villagers request another committee meeting to discuss the matter and then reach a decision after discussion and con- sensus-building. General meeting agendas include social development projects in collaboration with NGOs, maintenance of village infrastructure such as roads, kheyang, water supply channels and child education.

Currently, four NGOs are operating in this village. The villagers are pleased with their activi- ties. All villagers are very happy with the Green Hill project, which ensures a safe water supply and has helped prevent water-borne diseases such as diarrhoea and jaundice that killed three or four people in the dry season. Women are happier as they do not need to go far (> 1 km) to get water, and villagers receive water from four locations. Absent planters worked collectively 5 or 6 days a year on village road repair and kheyang development. More- over, the committee employed four volunteers for 2 months for the care and protection of water supply infrastructure. After 2 months, another four volunteers are employed. This way they keep the village water supply safe.

Therefore, absent planters, through their orga- nization, create linkages with many other formal organizations in order to receive social services. Their connectedness has helped to expand their networks to local government councils and political leaders. One absent planter is a member of the local union council and can influence the council chair- man to provide necessary village support. They also have close contact with political leaders, e.g. villagers support the former hill tracts council chairman and a national parliament member who provided funding to construct a brick road con- nected to a paved secondary road. Therefore, it is clear that collective efforts of villagers help to elect local union council members, and their vertical relationships with politicians facilitate the develop- ment of common village infrastructure. We asked why many NGOs are operating in their village and not in the resident planters’ village. They said that families in a village living in a cluster, with a strong committee and communications, attract NGO activi- ties. Although communication is good for resident planters, they live sparsely and have no committee, and hence no NGOs are active in their village.

Evaluation of an upland settlement project Nath and Inoue

164 International Journal of Sustainable Development & World Ecology

Livelihood strategies and diversification, income and expenditure

Currently, due to poor employment opportunities in project activities, planters find other ways to make a living. There are remarkable differences between the two groups regarding livelihood diver- sification. Resident planters include homestead agroforestry along with livestock, jhum, wage labour, and NTFPs (non-timber forest products) collection and sale, while absent planters include irrigated agriculture, homestead agroforestry, wage labour, jhum, NTFPs, services, village busi- nesses and vegetable farming. Differences in liveli- hood strategies are due to village location and availability of irrigated agricultural land. Many absent planters possess agricultural land and others lease it from landlords who live in town. Being near

to agricultural land, absent planters have better access to wage labour. The many active NGOs employ local people as part-time workers. Some relatively well-off planters run small businesses in the village. In contrast, being on hilltops, resident planters have no alternative other than jhum, wage labour (extraction of bamboo and wood for timber traders and pulp mills), and NTFPs collection and sale.

In terms of household income and income sources, resident planters had seven income sources while absent planters had 13 (Table 4). In both groups, wage labour provides more than 60% of total household income to relatively poor planters who worked 3 to 4 days per week. Absent planters depend almost entirely on wage labour and NTFPs for income, but income comes from

Evaluation of an upland settlement project Nath and Inoue

International Journal of Sustainable Development & World Ecology 165

Planters

Variables Resident Absent

Income sources Rice Wood Fruit Chickens Cows Goats Carpentry Wage labor

Men Women

NTFPs sale Businesses Services Jhum Vegetable farming Income from land rent

– 1375(75) 950(75) 584(63) 600(13)a

– –

12000(88) 6400(75)

13200(100) – –

919(75) – –

31477(27) 2100(45) 350(27) 550(9)

– 8500(9)

24000(9)

15200(27) 6600(18) 8817(55)

13476(55) 11200(27) 4500(9) 3500(9)

13000(9)

Mean annual income 31373 ± 4129 35262 ± 6002

Expenditures Food Education Agriculture Social work

25200 5500 (13) 760 97

23018 5563(73)

11469 360

Mean annual expenditures 25984± 3134 27425 ± 3734

Note: Figures in parentheses indicate percentage of household a: This family looks after two cows of a nearby village trader and gets Tk. 50 per month for this service

Table 4 Mean annual income (Tk.), income sources, and mean annual expenditures (Tk.) of planters in the study USP village, Bandarban, Bangladesh (Source: Field survey, Aug.–Sept. 2005)

many sources and both men and women have access to off-farm jobs, though at different salary levels (Tk. 80 to 100 for men, and Tk. 50 to 60 for women). The variation is due to differences in jobs: men generally do hard work while women do relatively easy work such as weeding, as also noted by Bleie (2005: 181). Most NTFPs collectors are women. After completing household chores, women go to collect different NTFPs such as bam- boo shoots and wild vegetables, return home at noon, have lunch, then sell the products at the nearby market. They collect and sell 3 days a week, and earn about Tk. 80 a day, but do not extract NTFPs while working in jhum and during the rainy season. They reported that, due to uncontrolled harvesting, it is now very hard to find NTFPs and they walk long distances (3 to 4 km) to collect a shoulder-load. They also feel that without regula- tion, NTFPs will lose regenerative capacity and their future is bleak. Women with infants who cannot collect NTFPs prepare traditional wine (made from rice) at home. Many other families also make wine for their own consumption. It takes 3 or 4 days to make 1 l of grade-1 wine at a cost of Tk. 60 and a sale price of Tk. 200. During religious festivals, they make more wine to sell and earn a handsome income. Considering all income sources, we found only small differences in mean annual income of the two planter groups. However, due to diversified income sources and less dependency on jhum and forest resources, the incomes of absent planters are more reliable than those of resident planters.

Considering selected items (food, education, agriculture, social work), we found that both groups of planters spend most money on food (Table 4). They collect more than 70% of food items such as vegetables, wild animals, small fish and crabs from forests. They purchase rice, fish, oil, spices, nappi (fermented dried fish) and other items from markets. Nappi is very popular among ethnic people, and a small amount is added to almost all curries. Resident planters spend more than 96% of their total expenditures on food, and absent planters around 84%. More than 70% of absent planters spend some of their income on child education, whereas only one resident planter does. It is clear that very little money remains to cover other expenses such as healthcare or clothing. However, relatively well-off absent planters, who have surplus rice for sale and who run businesses, have substantial savings.

Problems of the planters

Even though resident planters now have better water supply facilities, both groups ranked the lack of safe drinking water as the principal problem. Absent planters wonder if the disappearance of forests, where water collection tanks were estab- lished by Green Hill, limit the availability of seepage water. The project authority installed two ring-wells for drinking water and one cross-dam (660 m long x 28 m wide) that hold water year-round for house- hold uses and aquaculture. Planters can collect water for 6 months from ring-wells, but during summer and winter very little water is available. In those months, they get water from about 1 km downstream from their houses. Men, women and children bring water for drinking and household uses when they go to bathe in the stream. Some- times a water crisis also creates a family crisis. For example, one woman told us that the long distance to collect water delayed food preparation; her husband returns from work at noon, finds the meal uncooked, becomes angry and quarrels with her. She also said that sometimes she has to get water at night and wait in a long queue. If she does not get water at night, she cannot cook rice early in the morning. Stream water causes water-related dis- eases every year and sometimes deaths. Three years after the cross-dam (gudha) was constructed, it was damaged and washed out by heavy rain and could not be rebuilt due to the high cost (Tk. 150,000). A request was made to the project authority, but the response was negative due to budget constraints. Planters claimed that if the gudha was small, they could maintain it themselves. They also earned Tk. 7000 from the gudha through aquaculture. This money was deposited into the common fund.

A lack of jobs lead planters to practice jhum and harvest trees and other NTFPs. Yields from jhum feed planters for only 2 or 3 months a year. Unregu- lated extraction of bamboo and other NTFPs has put pressure on the sustainability of these products. The project authority believes that when rubber tapping starts, it will create job opportunities, but with the present condition of the rubber plants, one doubts how many jobs it may offer.

Due to the lack of first-class land, planters cannot produce enough rice for consumption, and face food shortages all year. A shortage of suitable hilly land prevents them from practicing jhum with a long fallow period. Currently, they allow 1 or 2 years

Evaluation of an upland settlement project Nath and Inoue

166 International Journal of Sustainable Development & World Ecology

of fallow. Villagers used to open-graze their cattle, but the project prevented this practice because it would damage rubber plantations. As a result, planters cannot raise cows or goats. Seasonal prob- lems also exist, e.g. in the rainy season they have almost no jobs and cannot sell their products. In winter, they face a shortage of warm clothes, and in summer, they suffer from water-borne diseases.

CONCLUSIONS AND RECOMMENDATIONS

Since the early 1960s, several development projects, with variable results, have been implemented in the CHT with a view to the rehabilitation and socio- economic development of landless people through the introduction of different farming systems in degraded hilly land (Khan and Khisa 2000; Nath et al. 2005). This study has identified several reasons for the failure of USP, the most recent development programme in the CHT, in achieving its proposed objectives. The inability of the project manager and staff to identify planters’ needs, to understand their cultural problems, to provide appropriate training for agroforestry development, and to raise aware- ness of perceived benefits from rubber have created conditions for project failure. Most project staff are agriculturists by training but could not under- stand planters’ behaviour and needs. There were no communication experts in the project team. Agunga (1992: 4) noted that project implementa- tion teams rarely include communication experts capable of assisting project managers in addressing communication shortcomings, such as mobiliza- tion, organization and training of local people for participatory decision-making, design and conduct of the development campaign, or the facilitation of coordination and linkages among participating institutions. Understanding the psychology of people is important if one is to succeed in winning their cooperation in a project (Blay et al. 2004: 29). Poor selection of planters, lack of responsiveness, lack of accountability and transparency in handling project money and decision-making processes, and lack of participation of planters in technology development equally contributed to project failure. More importantly, USP failure is largely attribut- able to managerial inability to organize the planters on a common platform, failure in leadership development and capacity-building and failure to create a sense of ownership. Lambin (2005:

177–180) suggested three factors — information, motivation and capacity — are equally essential for sustainable management of natural resources. Effective leadership can augment collective action by inspiring people, enforcing institutional norms, resolving conflicts, networking with development partners and guaranteeing expected benefits to people (Sinha and Suar 2005: 127).

The period between rubber establishment and production is critical for planters. The authority could provide interim off-farm employment oppor- tunities through cottage industries such as textiles, bamboo cane-based enterprises, or homestead agroforestry production of fruit such as papayas and bananas. Tribal people are skilled in tradi- tional textiles, handicrafts and making household appliances. These products have buoyant markets at both local and regional levels. Smallholders may be unable or unwilling to do anything other than try to improve their immediate financial circum- stances (Lamb et al. 2005: 1632).

Our findings suggest that absent planters could improve their livelihood through their strong social cohesion and collective efforts. Social capital is a resource that needs to be marshalled carefully by agents who can bring it to fruition based on incen- tives available in the external environment of the state and market; where such agents are present, social capital produces beneficial development outcomes (Krishna 2004: 300). The social develop- ment committee, backed by a strong and acceptable leader, can create vertical relationships with other formal organizations, such as NGOs and the local union council, enabling them to obtain different social services. Formal organizations such as union councils, NGOs, political parties and local leaders are particularly important in poor people’s lives in rural areas (Khan and Seeley 2005: 77). In Bangla- desh, NGOs provide social and economic support like non-formal education, women’s employment, environmental protection and child healthcare. (ibid: 78). By establishing and using good relation- ships with different stakeholders, the poor can enjoy social safety and security that helps them to reduce tensions confronting their livelihoods (Purvez 2005: 104). Considering the foregoing discussion, we rec- ommend the following measures for formulating future development interventions:

1. Apart from technical staff, a project team should include a communication expert to

Evaluation of an upland settlement project Nath and Inoue

International Journal of Sustainable Development & World Ecology 167

help organize beneficiaries, considering their beliefs, culture and needs.

2. A consultation workshop should be held in every project village, involving stakeholders such as the project authority, local govern- ment, NGOs, local organizations and local people before the project begins.

3. All arrangements, such as benefit-sharing, and roles and responsibilities of different stake- holders, should be clarified, defined, written and signed.

4. Without collective action, it is impossible to achieve the objectives of participatory devel- opment projects. Measures to create aware- ness of the benefits of collective action should be taken to create locally accepted organiza- tions to oversee collective activities. The pres- ence of a strong organization would much improve rubber garden conditions.

5. For capacity-building, planters should receive necessary training in technology development.

6. NGOs play an important role, not only in social capital formation but also in income- generation. Projects should involve local NGOs. All stakeholders, including local people, should be effectively involved in pro- ject planning and implementation.

7. To avoid misuse of project resources, provi- sions should be made for maintaining account- ability, transparency and equity. Provisions should be made to enforce rules and regula- tions and invoke sanctions if necessary.

8. Issues of land tenure should be solved as quickly as possible to ensure effective local participation. In USP, providing collective tenure for rubber plantations would ensure collective action for their smooth manage- ment and development.

None of these are difficult tasks, but require commitment, enthusiasm, strong integrated efforts and political support for the development of uplands and their indigenous peoples.

REFERENCES Addison T. Agricultural development for peace. Research

Paper No. 2005/07. Helsinki: UNU-WIDER; 2005 Agunga R. Development by rules: an ethical reflection

on the high failure rate of development projects and implications for communication. Africa Media Review 1992;6(1):1–13

Anonymous. Ministry of Chittagong Hill Tracts Affairs project proforma (revised) of Upland Settlement Project (2nd phase). Chittagong Hill Tracts Development Boards, Rangamati, Chittagong Hill Tracts; 2001

Bingen J and Munyankusi L. Farmer associations, decentralization and development in Rwanda: chal- lenges ahead. Agricultural Policy Synthesis: Rwanda Food Security Research Project/MINAGRI. www.aec.msu.edu/agecon/fs2/rwanda/index. htm; 2002

Blay D, Bonkoungou E, Chamshama SAO and Chikamai B. Joint management as an option for rehabilitating degraded forests: the case of the Gwira Banso Project in the Wassa west district of the western region of Ghana. In Wood P and Yapi AM (eds), Rehabilitation of degraded lands in sub-Saharan Africa: lessons learned from selected case studies. Austria: FORNESSA and IUFRO-SPDC; 2004:25–9

Bleie T. Tribal peoples, nationalism and the human rights challenge: the Adivasis of Bangladesh. Dhaka, Bangladesh: The University Press; 2005

Carney D. Formal farmers’ organizations in the agri- cultural technology system: current roles and future challenges. Natural Resources Perspectives No. 14. London: Overseas Development Institute; 1996

Cramb RA, Purcell T and Ho TCS. Participatory assess- ment of rural livelihoods in the central highlands of Vietnam. Agricultural Systems 2004;81:255–72

Dasgupta S and Ahmed FU. Natural resources manage- ment by tribal community: a case study of Bangladesh. The World Bank/WBI’s CBNRM Initiative. http://srdis.ciesin.columbia.edu/cases/bangla- desh-002.html; 1998

de Janvry A. Achieving success in rural development: tools and approaches for implementation of an integrated approach. ECOSOC Brainstorming Dialogue Sessions held at UN headquarters. www. un.org/esa/coordination/ecossoc/hl2003/ deJanvry.pdf ; 2003

DFID (Department for International Development). Sustainable livelihoods guidance sheets–comparing development approaches. London: DFID; 2001

Evaluation of an upland settlement project Nath and Inoue

168 International Journal of Sustainable Development & World Ecology

Ikerd J. Sustaining rural communities through sus- tainable agriculture. Southern Rural Sociology Association Annual Meeting, Fort Worth, TX. www.ssu.missouri.edu/faculty/jikerd/papers/ SustainableCommunities.htm; 2001

Khan NA and Khisa SK. Sustainable land management with rubber-based agroforestry: a Bangladesh example of uplands community development. Sus- tainable Development 2000;8:1–10

Khan NA, Emran AM and Al-Mamun CMA. A holistic approach towards participatory forestry: means of poverty reduction and upland development in Chittagong Hill Tracts, Bangladesh. Journal of Forest Economics 2003;49:1–10

Khan SM and Seeley J. Accessing a living: the roles organizations and institutions play in poor people’s livelihoods. In Khan IA and Seeley J (eds), Making a living: the livelihoods of the rural poor in Bangladesh. Dhaka, Bangladesh: The University Press Ltd.; 2005:75–92

Krishna A. Understanding, measuring and utilizing social capital: clarifying concepts and presenting a field application from India. Agricultural Systems 2004;82:291–305

Lamb D, Erskine PD and Parrotta JA. Restoration of degraded tropical forest landscapes. Science 2005; 310:1628–32

Lambin EF. Conditions for sustainability of human- environment systems: information, motivation, and capacity (editorial). Global Environmental Change 2005;15:177–80

Messer N and Townsley P. Local institutions and liveli- hoods: guidelines for analysis. Rome: Rural Devel- opment Division, FAO; 2003

Nath TK, Inoue M and Myant H. Small-scale agro- forestry for upland community development: a

case study from Chittagong Hill Tracts, Bangla- desh. Journal of Forest Research 2005;10:443–52

Osmani SR. The role of employment in promoting the millen- nium development goals. Issue in Employment and Poverty Discussion Paper 18. ILO-UNDP; 2005

Pretty J. Social capital and connectedness: issues and impli- cations for agriculture, rural development and natural resource management in ACP countries. Review Paper for CTA. CTA Working Document No. 8032. The ACP-EU Technical Centre for Agricultural and Rural Cooperation (CTA); 2003

Purvez MSA. Building support for a living: the impor- tance of social networks for the livelihoods of the poor. In Khan IA and Seeley J. (eds), Making a living: the livelihoods of the rural poor in Bangladesh. Dhaka, Bangladesh: The University Press Ltd.; 2005:93–107

Reddy VR, Reddy MG, Galab S, Soussan J and Springate-Baginski O. Participatory watershed development in India: can it sustain rural liveli- hoods? Development and Change 2004;35:297–326

Schneider JA. The role of social capital in building healthy communities. Policy paper. Annie E. Casey Foundation, Catholic University of America; 2004

Sinha H and Suar D. Leadership and people’s partici- pation in community forestry. International Journal of Rural Management 2005;1(1):125–43

Soini E. Changing livelihoods on the slopes of Mt. Kilimanjaro, Tanzania: Challenges and opportuni- ties in the Chagga homegarden system. Agroforestry Systems 2005;64:157–67

Wu B, Parnwell M and Bradley P. Farmer self- organizing innovation in the marginal areas of China: a study of farmer communication networks in Zhidan, Loess Plateau.Journal of Rural Coopera- tion, 2002;30(1):43–63

Evaluation of an upland settlement project Nath and Inoue

International Journal of Sustainable Development & World Ecology 169