Abstract

The main objective of this paper is to discuss the importance of looking at households as basic analytical categories in studies of migration in the Third World. The paper points to the difficulties of defining the household, and in particular the assumption thai the household can be treated as a strategic decision making unit, is challenged. The importance of taking a critical stand towards the household, as the basic analytical category, is illustrated by the results from research into the migration trajectories ofHill migrants settled in the Terai region of Nepal, h will be demonstrated how previous research has overlooked the importance of the split-up of extended households in relation to the migration and settlement process. It is concluded that future studies should continue to consider households as important elements in migration and settlement processes, however, they should be treated with many more analytical precautions.

Keywords

Nepal, migration and settlement, micro-level perspective, households,
methodology.

Jytte Agergaard: Department of Geography, University of Copenhagen,
Øster Voldgade 10, 1350 Copenhagen K., Denmark. Email:
ja@geogr.ku.dk

GeografiskTidsskrift, Danish Journal of Geography 99:101101-

Since the beginning of the 1980's, theoretical awareness has gradually emerged, so that micro-level approaches to migration and settlement in developing countries should now consider the importance of households, the close family and other family and kinship relations (i.e. Harbinson 1981, Hugo 1981, Wood 1982, Bach & Schraml 1982, Halfacree & Boyle 1993, Goss & Lindquist 1995, Agergaard 1998, Lawson 1998). It is acknowledged that households åre important agents to consider when explaining the reasons for migration and settlement, and when developing the analyses of the specific migration and settlement trajectories. It is assumed that the decision making process on economic matters (i.e. migration and settlement) in developing countries is less an individual question than a process where household members negotiate on a joint strategy. Households may tentatively be defined as identifiable groupings basedon some form of kinship relations within which people live. Furthermore, rural households may be characterized by joint ownership, production, consumption and reproduction (Crehan 1992), hence, household members have to negotiate on the economic and productive dispositions to retain rights of the

joint assets. Accordingly, the settlers' allocation of money and labour power, in order to cope with the difficulties of converting virgin forest land into productive assets, is highly dependent on these household dispositions. Households åre, therefore, crucial for the analysis of the process of frontier settlement and planned settlement in rural areas.

The process of migration from the Hill region and settlement in the Terai region (the plain, southern East-West extending zone) of Nepal has resulted in a considerable relocation of the Nepalese population. Settlement in the Terai region, which started to take place due to stateinduced planned settlement in the 1950'5, has taken on many forms and changed over time. However, a considerable number of settlers first settled as agriculturalists in the region. The process, since the first preliminary studies of the 1960's and 1970's (i.e. Elder et.al. 1974, Rana & Thapa 1974, Sainju 1974, Dahal et al. 1977, Kansakar 1979) and specifically following the first in-depth analysis (Conway & Shrestha 1981) has been characterized as one of relocation of Hill households in need of land.

The relocation hypothesis is supported by a survey, asking
the Chitwan head-of-households about their strategic

considerations and reasons for the decision to migrate (Conway & Shrestha 1981). One problem of Conway and Shrestha's (1981) study is that they did not investigate the actual links between the sending (Hill), and the receiving (Terai), communities, but base their conclusions on randomly administrated questionnaires from both regions. Hence, the analysis builds on an un-qualified assumption, that the Chitwan households equal the Hill households that took the decision to migrate. With the limited space available, an extended criticism of the methodological approaches of the former studies would be far too ambitious (for a discussion, see Agergaard 1998). In the present context I will narrow my focus to the misinterpretation of the term 'relocation' of households and the implications of this mistake.

The discussion draws on research of the migration and settlement processes from Lamjung District to Chitwan District in Nepal. Fieldwork was carried out in both districts from September 1993 to July 1994. In Chitwan District two wards/village communities, Indrapuri and Hanuman Nagar, approx. 10 km south of the district capital Bharatpur, were selected as the starting point for the investigation. The settlers' villages of origin and family/ relatives in Lamjung District were then researched.

This paper is basedon an analysis of in-depth case studies of 22 settler-households in Chitwan, focussing on their migration/settlement trajectories. The households were selected on the basis of a preliminary wealth and status ranking of all 320 households residing in the two localities. During the ranking procedure baseline data was obtained, which made it possible to delimit the selection of households to those stemming from the Lamjung District and where the head-of-household was old enough to obtain entitlement to land on his own while settling in Chitwan (first generation). Furthermore the selection of households covered differences in three social criteria: time of arrival in Chitwan, size of landholdings in proportion to the number of dependents, and caste/ethnic group affiliation. Migration history interviews were conducted among various members of each settler household, their relatives in Chitwan, relatives still living in their previous home village in Lamjung District, as well as selected key informants in both the Chitwan and Lamjung communities.

The paper begins with a discussion of the theoretical basis for selecting the household as the central analytical unit in migration and settlement studies, with specific focus on the Nepalese/Hindu household. The paper then turns to a brief review of the trends in migration and spatial relocation in Nepal. This section is followed by a presentation of the colonization of the Terai region, with specific focus on the Chitwan district, which was the first to be exposed to state-led planned settlement schemes. The the idea that household settlement in Chitwan which is characterized by a relocation of Hill households will then be discussed. It will be concluded that the Hill households seldom move jointly, rather, the settlement process in Chitwan coincides with the process of household separation of the 'original' Hill household. The final conclusion sums up the preceding discussion and emphasizes the theoretical and methodological challenges faced by migration and settlement research, not least in Nepal.

Researching the household

Theories and approaches to migration that have been inspired by neoclassical economics, as well as politicaleconomical thinking, åre mainly concerned with either the economic and structural explanations of migration or the dynamics at micro-_level, looking at the individual as either a rational decision maker or as an individual constrained by his (seldom her) income or class position (Halfacree & Boyle 1993, Goss & Lindquist 1995, Agergaard 1998, Lawson 1998). Hence, considerations of households/families (there is seldom a clear definition of the concept used) as contexts of migration and settlement processes, have been left out of much the theory about the process of migration and settlement. However, in the beginning of the 1980's, this obvious shortcoming led to the first attempts to develop a household approach to migration studies (Harbinson 1981, Hugo 1981, Wood 1982, Bach &

Schraml 1982). While being a welcome response to the impasse in migration theory, this early household approach has not gained much ground. Not least from feminists has the implicit assumption of these studies, that households take joint strategic action and have coherent migration strategies, been disputed (Chant & Radcliffe 1992, Hondagneau-Sotelo 1994, Wright 1995, Gardner 1995, Lawson 1998).

As Netting (1993) points out, households åre social groups so übiquitous in human society that taking them for granted is easy. Although the household is acknowledged to be a central or maybe the most important context of human behaviour, we find no easy definition of the household (Netting et.al. 1984, Netting 1993, Varley 1994). Considerable efforts have been made to make the term household more universally applicable. However, one central problem of defming the household is that the meaning of family and kinship, and how they åre translated into institutions of everyday life, vary widely between different cultures. Netting et al. (1984) argue that a distinction between household and family is an important step towards universally applicable definition since family is a much more culturally defined term than household and differs considerably among, and even within, cultures. However, this demarcation may not solve our problem because, as Lardinous (1992) underlines, the common distinction within European wri_tings between family, household and housefull (people living in the same house) does not mitigate the ambiguity of the definition of the domestic group, cross-culturally (see also Robertson 1991, Varley 1994).

Another discussion within household studies has been whether to focus on the household form or function (Wilk & Netting 1984, Lardinous 1992, Varley 1994). Netting et.al (1984) suggest that rather than putting boundaries around households, a definition should be made with respect to the different types of household activities (i.e. production, distribution, transmission, reproduction and coresidence) (Wilk & Netting 1984, Netting 1993). Hence, households åre characterized as social units, with a considerable overlap of activities. It is surely important to focus on how households function, rather than how they åre formally constituted. Nevertheless, it has been questioned whether or not different household activities or functions neatly overlap (Campbell 1996). Furthermore, an emphasis on the household form may lead to an overly static perspective on the household, which may neglect the faet that

households undergo constant transformations associated with household development cycles and households interface with changing social and economic environments (Varley 1994).

To challenge the obvious difficulties regarding the priori boundaries and functions of the household, it has been suggested that what is meant by the family/kinship basis, joint ownership, consumption, production and reproduction, must be empirically determined in each case (Netting et.al. 1994, Varley 1994, Campbell 1996). Consequently, some central aspects of how these characteristics of the household may be understood, in a Nepalese context, will be discussed.

The ideal Hindu joint family is characterized as an extended household, comprising two or more generations of a patri-local family, with the oldest active male as the household head. This family type plays a central role in the Nepalese society (Berreman (1963) 1997, Bennett 1983, Campbell 1996), but although this ideal may serve as an important point of reference, many Nepalese households differ from this ideal (Campbell 1996). First, it should be considered that the Brahmin orthodoxy is not always known and/or acknowledged by the people, and likewise the written and unwritten prescriptions åre open to symbolic and strategic manipulation (Lardinous 1992). Secondly, it should be acknowledged that many households åre in some sort of transition phase, where they develop into various nuclear households due to the split up of household assets among brothers or/and father and sons. Hence, despite the faet that households åre basedon some kind of family and kinship relationship, households åre subjects to continous changes.

Rural households in Nepal åre to a large extent characterized by joint ownership, production and consumption, although, a range of variations can be found among the different ethnic groups of the Nepalese society (Rao 1992, Campbell 1996). However, a number of intermediate household forms occur. For instance where joint ownership continues while production and consumption is split between new household units. This can be explained by the dynamic characteristics of the joint family, characterized by various separation processes of the household, for example when sons separate from their parents, brothers separate or a co-wife separates from her husband etc. Hence, the relationship between coresidence, joint ownership, production and consumption undergoes obvious changes, which means that thorough consideration of hov/ household

separation influences the unit of analysis (the household)
is needed.

Finally, the implicit assumption, that households take joint action should be considered. Campbell (1996) wams against the tendency among social and cultural studies of the Nepalese households, to suppress the differences and conflicts over various economic matters within the household. When researchers go behind the domestic facade of the Nepalese households, it has been revealed that the disagreements on economic matters do not necessarily result in joint action, as often proposed by advocates of the cooperative conflict approach (i.e. Sen 1990; see also Lawson 1998), but can end in conflict and unrest (Campbell 1996). It is, therefore, reasonable to suggest that decisions over migration and settlement åre also objects of conflict and contest, and not necessarily an outcome of joint strategic action.

The purpose of the previous discussion has not been to launch a new definition of the household within migration studies. Emphasis is placed on the methodological challenge of combining two such dynamic concepts as migration and household. It should be remembered that migration is a process that happens in and to a household, and that households åre themselves transitional units. More focus is needed on the dynamic characteristics of the migration processes, acknowledging that migration can be seen as a process by which household activities åre reorganized and often split between sites (Sharma 1987).

To take a critical stand towards the household as the basic analytical unit within migration studies, it is necessary to make use of more qualitative research methods. In this respect, a case study approach building on longitudinal studies of actual migration histories has been suggested (Sharma 1987, Halfacree & Boyle 1993, Hondagneau 1994, Wright 1995, Gardner 1995, Lawson 1998). As Lawson (1998) concludes, aggregate statistical analysis of migration processes 'lose the connection between individual experiences of migration and the household and community structures within which migrants åre embedded'. On the contrary, the migration history approach can reveal the dynamic and context-specific power relations operating within households, and at the same time recognizing the incomplete knowledge of the migration trajectories. Hence, the case study approach is selected for its explanatory power, rather than for its typicality or representativeness since statistical inference is not invoked at all. However, deciding which research methodology to adopt is not a question of good or bad methodology, but rests on the individual researches main research objectives (see also the discussion in Agergaard 1998).

Mobility and settlement in Nepal

Migration is an integral part of everyday life of the Nepalese peasantry. The current migration pattern in Nepal is a result of a long tradition for spatial mobility that dates back from the 18th century. Until the unification of Nepal in the middle of the 18th century, Nepal was divided into more than 60 minor principalities primarily based in the Hill areas. Between the principalities there was little social, economic or political interaction. However, during the process of unification they were gradually incorporated into what is now known as the Kingdom of Nepal (Shrestha 1990).

The period of unification of Nepal is often characterized by the establishment of socioeconomic inequalities that were further developed during the subsequent Rana period. Furthermore, the gradual process of Hinduization during the 19th century dramatically changed the land tenure rights in favour of the Hindu aristocracy and bureaucracy (Gurung 1989, Shrestha 1990, Ghimire 1992, Karan & Ishii 1996). An understanding of these socioeconomic and spatial disparities is an important background for the comprehension of the macro-scale driving forces behind migration in Nepal.

Among the farmer population of Nepal, a number of labour migration strategies have been employed. These include recruitment to the Nepalese army or Indian Gurkha regiments (a strategy that can be followed up by labour migration to the Middle East or semiskilled employment in India), labour migration to the industrial areas of Northern India (Bombay and New Delhi), and employment in Kathmandu (the capital) as a construction worker, vendor, tourist guide, hotel boy etc. Army recruitment is very popular, not least because 15 years of employment qualifies for a lifelong pension. Employment as civil servant is another strategy, only obtainable by educated people with the appropriate social contacts (Bistå 1991, Khadka 1994). Today, an increasing proportion of young people have moved to the urban areas for high school education. Estimates of the number of labour migrants in Nepal åre very difficult to obtain, since there åre no exact census data covering this matter. However, anthropological data supports

the idea that various combinations of these strategies
åre employed in almost every household (for a discussion,
see Subedi 1993).

However, a permanent relocation of the Nepalese population is a more recent phenomenon. During the first half of this century, the North Indian states and Bhutan attracted many Hill households for permanent settlement, whilst internal migration was limited (Gurung 1989). Until the middle of this century, the majority (approx. 70 percent) of the Nepalese population had permanent residence in the Mountain-Hill region (including Kathmandu) (Gurung 1989). However, since the 1950'5, migration from the Hill region towards the agricultural region of Terai, and later towards the expanding city areas of Kathmandu and district centres, has resulted in a considerable redistribution of the Nepalese population. At the same time, the Nepalese population has doubled from approximately 9.5 million in 1952, to approximately 18 million in 1991, the population of Terai has increased from approximately 25 to 55 per cent of the total population (Gurung 1989, CBS 1993). Accordingly, it is no wonder that the process of migration from the Hill region and settlement in the Terai has attracted considerable interest frompoliticians, planners and researchers.

Household settlement in Terai

The rapid increase in the population and the growing regional and social disparities åre among the reasons for the higher mobility among the Hill population, this century (Shrestha 1990). However, the state-led planned efforts to resettle Hill farmers in Terai seems to be the starting signal for the regional redistribution previously mentioned. The Chitwan district was the first district to experience the Nepalese govemment's resettlement programme, the Rapti Valley Development Programme, and has attracted Hill farmers for settlement way past the termination of the official programme in 1964.

The process of state-induced settlement has often been characterized as land colonisation and is a strategy wellknown from other regions with frontier land, for example the Amazonas. The original idea behind settlement in Terai was to convert virgin land into productive farming units. The state legalized the cultivation of the land and the farmers contributed with their labour. Household settlement was an almost inevitable requirement for land coloni

zation, since the success of this type of migration and settlement is closely related to the availability of labour. That labour power was a crucial parameter during the process of settlement in Chitwan, has been confirmed by official documents (Dussenbery 1958, Donner 1967) as well as interviews with settlers that settled within the resettlement schemes. However, it has also been noted by those that have settled subsequently to the termination of the official programmes in 1964 and åre, therefore, formal buyers of land, that they were in need of labour in order to succeed as an agriculturalist on frontier land (for more details see Agergaard 1998).

Far from all Hill migrants, having Chitwan as their destination, have managed to settle as agriculturalists. However, in the current context, the trajectories of the many provisional settlers, temporary farm hånds and squatters åre not considered. The analysis of migration and settlement trajectories is basedon the segment of settlers that åre officially registered as residents in the two researched villages, a group, which nevertheless is rather diverse. To some extent this diversity can be explained by changing options and constraints to settlement: entitlement to land in the migrants home area, availability of land and land prices in the settlement areas, availability of employment in the Terai agriculture etc., but also differences in caste and economic position prior to settlement, influences the process of migration and settlement (Agergaard 1998, Agergaard 1999). These structural differences and changes definitely shape the specific migration trajectories. However, what the majority of cases have in common, is the coincidence between major household processes and migration and settlement.

The general explanation among the Chitwan settlers as to why people left the Hills and decided to settle in Chitwan District, is that the settlers were attracted by the excellent facilities of the district. It was often noted that it was easy to explain why people moved, because in Chitwan one had easy access to water, irrigation, electricity, transportation, and since the area is flat, the carrying of loads is much easier than in the Hills. Nevertheless, everyday life was not easy straitforeward when the migrants from Hills started to settle in Chitwan. Further questioning revealed a much more complex set of answers and explanations of why and how people decided to move and settle in Indrapuri and Hanuman Nagar. As has been discussed in Agergaard (1998) the migration/settlement trajectories should be seen in a number of contexts that åre not mutually exclusive. Household dynamics is one of those contexts.

Relocation of households?

The focus of previous studies on households as primary contexts for understanding of the settlement process in Terai have been well founded. The following discussion, however, questions the assumption that household settlements in Chitwan should be characterized by a relocation of Hill households. It is anticipated that the focus on the relocation of households indicates that well-defined households remain unchanged through migration and settlement, hence, it may be overlooked that household composition changes during the process of migration and settlement. An examination of the timing of migration and settlement in relation to household dynamics has proved to be an important source for the understanding of who decides to migrate from Lamjung and manages to settle in Chitwan.

Information about the timing of the migration and settlement
processes developed from an analysis is often contradictory
to the opinions gained from interviews with the

various members of the Chitwan households, as well as interviews among their relatives still residing in Lamjung. Figure 5 is a review of the timing of household separation in relation to the process of migration and settlement. The figure shows that of the 22 migration histories investigated, only five of them can be characterized as a relocation of a Hill household. In the remaining cases, household separation has or was taking place during the process of migration and settlement. In the five cases of relocation, household separation was not a possibility and åre, therefore, exceptional cases for various reasons. In case l, 2 and 21, the head-of-household was the only son of parents who died before household separation could become a possibility. The two Gurung households (12 and 13) åre part of a more specific migration and settlement history. They åre among the considerable number of Gurungs who decided to settle in Chitwan to support the central leader of the Nepali Congress Party, Bakan Sing Gurung, in his struggle for democracy in the beginning of the 1950'5. Hence, they decided

to relocate and settle in a group of fellow Curlings. This makes it reasonable to conclude that household separation played a significant role in the context of, and perhaps was the reason for the decision to migrate from Lamjung and settle in Chitwan. This conclusion is further supported by a more detailed analysis of three cases outlined below and summarized in figure 6.

Gokarna Koirala left Koirala Phand and settled in Chitwan in 1969, five years after the termination of the RVDP. Thus, his chance of obtaining land depended on the availability of land for sale and sufficient cash for the purchase. Until Gokarna left Lamjung with his wife, children and old parents, they lived in a joint household with Gokarna's eldest brother, while the youngest brother (Lekh Nath) separated from the Lamjung household in 1965 and settled in Chitwan. Separation from the joint household in Koirala Phand was a prerequisite for the release of enough money to buy land in Chitwan. However, Gokarna's elder brother and his parents were against separation and migration, and Gokarna's wife was also reluctant to move. Although Gokarna visited Gitanagar several times to meet his relatives and inquire about the prospects of settlement, he hesitated to turn himself against the rest of the family, in the way his second brother, Lekh Nath had done. When the eldest brother died, a new basis for negotiation developed. The household was formally separated and all assets in Koirala Phand were sold. Receiving his share ofthe assets in Koirala Phand, Gokarna could sell the land in Koirala Phand and buy new land in Chitwan.

In the case of the Thapa family from Kaole, the sequence was rather different. The Thapa family settled jointly in Chitwan in 1956. Thus, they were among the first settler families in Chitwan. They were four brothers with wives, a sister-son and their old mother. Five years before migrating from Lamjung, two of the brothers had tried to cultivate new land in a village near to Kaole, however,

without much success. In the middle of the 1950's the Kaole household had an abundance of labour power. The four brothers as well as their sisters were all married, though there were only two or three babies. Furthermore, the son of the eldest sister, who was adopted by the household, had grown up to be a young and energetic man. Some members of the household made exploratory visits to Chitwan while the others looked after the assets in Kaole.

After making the decision to move, the household moved jointly, andeach brotherreceived one plot (approx. 3.4 ha) of land while their mother received only half a plot, which was the size of land provided for widows. The adopted nephew was too young to demand land of his own. In the beginning, the Kaole household continued to operate as one household in Chitwan. Nevertheless, tensions between the brothers developed and after approx. two years the younger brothers demanded separation, sold their land and moved to their present home area (Indrapuri) together with their mother and nephew. At that time land was available in Indrapuri at a lower price than in their first home, Kansan Basti, so they were able to buy a reasonable piece of land.

Makan Singh BK settled in Chitwan in 1957 when he was only 17 years old. The availability of land in the the Chitwan District provided favourable conditions for himself and his family to settle. He was not able to rnake his own application for land at that time, since applicants had to be at least 18 years old, and have a responsibility for

their own family. He settled together with his parents and younger brothers and sisters, and managed within a short time to make a successful application for land of his own. Nevertheless, he continued to live jointly with his parents and younger brothers and sisters until some time later when he married. When married he separated form his family, although it is unclear how much land Makan received at that time. Generally, the family had not been very successful in keeping their total land, which at some time, may have been more than eight hectares. As Makan Singh explained, his father was not very good at managing the new land.

These three examples illustrate that important household processes take place before, during, and after the process of migration and settlement. In many cases the settler households were not identical with the Lamjung households: the migrants left and after five to six years of settlement and even more households have separated. Accordingly, it should be acknowledged that the links between households in Chitwan and the former households in the Hills åre much more complicated than the relocation thesis assumes.

However, the separation of households often takes place in relation to the process of migration and settlement, this does not explain why this relationship exists. One probable explanation is that the households have reached a phase in their development cycle, where household separation is most favourable for all the parts of the family. These households have the necessary resources for a relatively successful settlement, such as livestock, money and labour force. Likewise, critical tensions between household members seem to have arisen when the sons and in-laws of the joint households have come into a position where they can pursue their own ambitions.

Conclusion

Judging from the cases investigated in Indrapuri and Hanuman Nagar, the process of household settlement in the agricultural districts of Chitwan cannot be conceptualized as a relocation of Hill households. The preceding analysis of 22 migration trajectories (migration histories) has illustrated how household separation due to joint strategic action or because of conflict and disagreement between members of the Hill household coincides, with specific

therefore, appears that previous studies of the micro-dynamics of the migration process from the Hill regions and settlement in Terai has overlooked that settlement in Chitwan is closely related to the household cyclus, nuclear - joint - nuclear. Migration has been attractive and possible, in particular, to households or members of households wishing for household separation.

The empirical investigation also supports the criticism of the formal questionnaire approach to a micro-_le vel investigation of the migration and settlement process in Chitwan. The methodological alternative followed here, as in the case study approach to migration and settlement processes, has revealed important knowledge of the context of migration and its character as a process. Important knowledge about the reasoning behind, and explanation of the migration and settlement process has emerged from the exposure of the actual history of migration and settlement.

Despite the great difficulties in defming households as empirical units of investigation, it is concluded that microlevel research of migration and settlement processes, has to cope with this challenge. In Nepal households and families åre perhaps the most important social units for the majority of the population, and the migration and settlement processes cannot be understood without a proper investigation of the interface between the individual migrants and their household and family/kinship ties. To look uncritically at Nepalese households as homogeneous entities that take strategic decisions regarding migration and settlement can no longer be defended.

Acknowledgements

I would like to thank the Council for Development Research for providing financial support for my PhD research and senior researcher Steen Folke for his supervision of the overall project. I would also like to thank Jens Lerche and Chris Jensen-Butler for their valuable comments. Naturally, I alone am responsible for the content of this paper.

References

Agergaard, J. (1998): Unsettled Settlement: A study of the Process of Migration from Lamjung District and Settlement in Chitwan District of Nepal. Unpublished PhD Thesis, Department of Geography, University of Copenhagen.

Agergaard, J. (1999): Settlement and changing land use in the

Chitwan District of Nepal. Danish Journal of Geography,
Speciallssue 1: 11-19.

Bach, R. L. & Schraml, L. A. (1982): Migration, crisis and
theoretical conflict. International Migration Review, 16 (2):
320-341.

Bennett, L. (1983): Dangerous wives and sacred sisters. Social
and symbolic roles of high-caste women in Nepal. New York,
Columbia University Press.

Berreman, G. D. ((1963) 1997): Hindus of the Himalayas. Ethnography
and Change. Oxford University Press, Delhi.

Campbell, B. (1996): Notes om some studies of Himalayan
households. Bulletin ofßesearch, (11).

European Himalayan
Central Bureau of Statistics (CBS) (1993): Statistical Year Book
1993. HMG Commission Secretariat, Kathmandu.

Planning Crehan, K. (1992): Rural households: making a living. Bernstein, H. et.al., Rural Livelihoods. Grises and Responses. Oxford, Oxford University Press.

Conway, D. & Shrestha, N. R. (1981): Causes and Consequences
of Rural-to-rual Migration in Nepal. Final Report. Indiana,
Department of Geography, Indiana University.

Dahal, D. R. et.al. (1977): Land and migration in Far-Western
Nepal. Institute of Nepal & Asian Studies, Tribhuvan University.

Donner, W. (1967): Remarks on Rapti Valley Development Project.
Kathmandu, HMG.

Dussenberry, H. L. (1958): Report on Rapti Valley Development
Project. Kathmandu, HMG.

Elder, J. W. et.al. (1974): Planned settlement in Terai. Kathmandu.
Institute of and Asian Studies, Tribhuvan

Nepal University.
Gardner, K. (1995): Global migrants, local lives. Travel and
transformation in rural Oxford: Claredon Press.

Bangladesh.
Ghimire, K. (1992): Forest or farm? The politics of poverty and
land hunger in Nepal. Delhi, Oxford University Press.

Goss, J. & Lindquist, B. (1995): Conceprualizing international
labor migration: a structuration perspective. International Migration
Review, 29 (2): 317-351.

Gurung, H. (1989): Regional parterns of migration in Nepal.
Paper of the East-West Population Centre, No. 113. Hawaii, the
East-West Population Centre.

Harbison, S.F. (1981): Family structure and family strategy in migration decision making. Gordan F. DEJong and Robert W Gardner (eds.): Migration Decision Making: Multidisciplinary Approaches to Microlevel Studies in Developed and Developing Counrries. New York: Pergamon Press.

Halfacree, K.H. & Boyle, P.J. (1993): The challenge facing
migration research: the case of a biographical approach. Progress
in Human Geography, 17 (3): 333-348.

Hondagneu-Sotefo, P. (1994): Gendered transitions. Mexican

experiences of immigration. Berkeley, Los Angles and London,
University of California Press.

Hugo, G. J. (1981): Village-Community ties, village norms, and ethnic and Social networks: A review of evidence from the Third World. DeJong, G. F. & Gardner, R. W. (eds.), Migration Decision Making: Multidisciplinary Approaches to Microlevel Studies in Developed and Developing Countries. New York, Pergamon Press.

Kansakar, V. B. S. (1979): Effectiveness of planned settlement
programmes in Nepal. Kathmandu, CEDA.

Karan, P. P. & Ishii, H. (1996): Nepal: A Himalayan Kingdom in
transition. Tokyo, United Nations University Press.

Lardinous, R. (1992): Family and household as practical groups: preliminary reflections on the Hindu joint family. Saradamoni, K. (ed.), Finding the household. Conceptual and methodological issues. New Delhi, Newbury Park, London, Sage.

Lawson, V.A. (1998): Hierarchical households and gendered
migration in Latin America: feminist extensions to migration
research. Progress in Human Geography, 22 (1): 39-53.

Muller-Boker, U. (1995): Die Tharu in Chitwan. Kenntnis, bewertung
and nutzung der natiirlichen umwelt in siidlichen
Nepal. Stuttgart, Franz Steiner Verlag.

Netting, R.M. (1993): Smallholder, Householders. FarmFamilies
and the Ecology of Intensive, Sustainable Agriculture. Stanford,
Stanford University Press.

Netting, R.M.,Wilk, R.R. & Arnould, EJ. (eds.) (1984):
Households. Comparative and historical studies of the domestic
group. Berkeley, University of California Press.

Rana, R. S. J. B. & Thapa, Y. S. (1974): Population migration in
Nepal. Kathmandu, CEDA.

Rao, N. J. U. (1992): Gaps in definitions and analysis: A sociological perspective. Saradamoni, K. (ed.), Finding the household. Conceptual and methodological issues. New Delhi, Newbury Park, London, Sage.

Robertson, A. F. (1991): Beyond the Family: The Social Organization
of Human Reproduction. Cambridge, Polity Press.

Sen, A. (1990): Gender and cooperative conflicts. Tinker, I. (ed.),
Persistent inequalities. Women and world development. New
York, Oxford University Press.

Sharma, U. M. (1987): Migration as a household process: data
from Himachal Pradesh. Sociological Bulletin: Journal of
Indian Sociological Studies, 36 (2): 61-79.

Shrestha, N. R. (1990): Landlessness and Migration in Nepal.
Boulder, San Francisco, London, Westview Press.

Subedi, B. P. (1993): Continuity and change in population movements:
From inside a rural Nepali community. Unpublished
PhD thesis, University of Hawaii.

Varley. A. (1994): Housing the household, holding the house.
Gareth Jones & Peter M. Ward (eds.), Methodology for land
and housing market analysis. UCL Press: 120-134.

Wilk, R. R. & Netting, R. M. (1984): Households: Changing forms and functions. Netting, R. M., Wilk, R. R. & Arnould, E. J. (eds.), Households. Comparative and historical studies of the domestic group. Berkeley, University of California Press.

Wood, C. H. (1982): Equilibrium and historical-structural perspectives
on migration. International Migration Review, 16 (2):
298-319.

Wright, C. (1995): Gender awareness in migration theory: Synthesizing
actor and structure in southern Africa. Development
and Change, 26 (4): 771-791.