Born to be a Butcher?
A Study of Social Mobility and Symbolic Struggles of Low Castes in the Kathmandu Valley

Benedicte Lie

Department of Social Anthropology, University of Bergen
Thesis submitted for the Cand. Polit. degree in Social Anthropology, 1999

Contents

Acknowledgements

I would like to thank my thesis supervisor Gunnar Haaland, whom I admire for his extensive knowledge on people and society, for his theoretical insights, valuable suggestions, and not least his enthusiasm and interest in my work. He has advanced my knowledge and understanding through questions and  suggested readings, and always made time for me in his busy schedule.

I also owe my thanks to the Nepalese government and Tribhuvan University for granting me a research-visa. At Tribhuvan University, I would like to thank in particular Dr. Prem Kattri, head of Centre for Nepal and Asian Studies, for welcoming me and assisting me.

I want to thank my interpreter, Shamsher Bahadur Nuchhen Pradhan for working with me, for his flexibility, and patience, during the field work, and during all the long hours translating and transcribing the recorded interviews, and for continuously updating me on events of relevance to my thesis after the fieldwork.

I must acknowledge a dept of gratitude that can never be repaid to all the Khadgis I have got to know; Thank you for your hospitality, for letting me into your lives, for giving me generously of your time, and for putting up with all my questions.

I would also like to thank the three fellow student who have been my discussion partners in a working group. You have offered valuable suggestions, comments and criticisms.

Last but not least, I would like to thank my family: my son, my husband, my sister, and my mother and father, who have all shared my enthusiasm for Nepal, and helped me with caring for my son, both during fieldwork and when writing the thesis. A special thanks to my father who has corrected the English, a precondition for writing this thesis in English, and making it available to those it is about.

Chapter 1: Introduction

Nepal has seen a lot of changes over the last half century, changes in many different sectors, such as communication, tourism, technology, politics, economy and trade. The country is increasingly influenced by foreign impulses through television, tourism and foreign aid agencies. During the last decades a liberal market economy has been introduced in Nepal. The introduction of market economy and wage labour is affecting the system of production.

Within a society where caste is an important social institution, many of a person's attributes are defined by what caste that person is born into. The ascribed caste status is closely connected with ritual status, occupation, economic status and power. Although the caste system was legally abandoned in 1951, caste is still an evident feature in Nepalese society. A society where division of labour is related to caste status, is often referred to as Jajmani systems. The Jajmani system is thought to be a relatively complex division of labour where people are tied to each other in patron-client relations. Although it is unclear to what extent a Jajmani system has characterised the division of labour in Nepal, it is clear that new occupational and distributional relations are developing. Contractual patron-client relations are being challenged by market forces of supply and demand. Changes in the system of production and the division of labour have turned many people's lives upside down. Occupational caste groups are faced with new situations. Some are benefiting from the changes, others losing out. Some have completely changed their occupational habits, by choice or force. Below are three examples of changing labour relations, one of which will be followed throughout this thesis.

The Manandhars are known as oil pressers. With the opening of the market a new oil was introduced, and the imported product completely ousted the Manandhar oil. Consequently, their caste based profession was wiped out by international trade, mass production and new requirements of efficiency. The Manandhars had to abandon their traditional profession and find other means of making a living. My second example is an occupational group that has benefited from the changes. The sweeper caste was ranked at the very bottom of the caste hierarchy. They were a deprived group with few prospects of a comfortable life. Urbanisation and the increasing waste problems in the city forced the government to employ street sweepers on a regular basis. Cleaning streets is considered a defiling activity and not many people were willing to take these jobs. Thus, the people from the sweeper caste have been provided secure and well paid jobs working for the government and private companies, taking jobs that other people hesitate to take. Cases like these are much discussed among Nepalese.

There is a lot of talk on the matter of low castes aiming at social mobility, and on the strategies they employ. The third example is at the apex of such discussions. According to caste rules the Khadgis should be butchers. The Khadgis rank as low caste and are, accordingly, by some considered untouchable. Gangalal Shresta has published an article on the history of the Khadgis. He notes the following:

The Khadgis were to have cowsheds, rare the cows and sell the milk from them. They were titled Kasahi; butcher, and forced by royal order to do the butcher job. This was a big insult to the Newar community. According to the rules of Jayasthiti Malla the Khadgis should not wear shirts with sleeves (that cover the arm) and not wear shoes with sole. They should slaughter buffalo, and in return get payments. They should do the business of meat (Gangalal Shresta, undated).

In the Kathmandu Valley the demand for meat has grown enormously during the last years. Many factors have contributed to this increase in demand. They include general population growth, urbanisation, changing food habits, increased purchasing power, and immigration. The demand has provided the butchers with a valuable source of income. Allegedly, some have become extremely wealthy and now enjoy a lifestyle traditionally unthinkable to a caste of such low ritual rank. The Khadgis are often held out as an example of how low castes may attain upwards social mobility, and how the caste status and economic status no longer coincide. The phenomenon stimulated my curiosity.

1.1 Working Hypothesis and Objectivities

I decided to study the Khadgis first to establish the facts: what processes of change were taking place: How wealthy had they become? Then I would try to see what had made the acquisition of wealth possible, and how they had acquired it. I would also try to find out how they used their acquired capital. How did they spend their money? What did they invest in? Would they move to fashionable areas, buy expensive cars, and get an education? Would they use a lot of money on religious participation? Finally, I wanted to try to see how the economic capital was affecting their overall rank and position in society. If their rank is affected, is it due to changes in their economic status, in their occupational choices, or in ritual practices? What is the relation between economic status, social status and caste status? How are the changes perceived by the Khadgis, and how are they perceived by other groups?

A change in rank and social status might imply that new criteria for evaluating and acquiring status are gaining foothold. The situation and questions touch on issues of modernisation, development and change. According to the new laws of Nepal no one should be discriminated on the grounds of caste, religion, race or gender. This represents an egalitarian way of thinking quite different from the hierarchical ideology dominant in the caste society. Enforcing these principles would in theory give individuals equal opportunities. What then becomes of the ascribed status of caste, that has provided different groups with such different opportunities?

This thesis is a study of the Khadgis in the context of a changing society in the Kathmandu Valley. The Khadgis are affected by, and are part of the changes that are taking place. Changes in their frames for action, their limitations and opportunities will be dealt with. The information may reveal aspects of the changing significance of caste, norms and values, both within Khadgi communities, and in the society as a whole. Changes in the Khadgis' lifestyle and strategies must be seen as part of the wider process of globalisation and modernisation that is affecting the Nepalese society as a whole, and the Khadgi socio-economic position in particular. The working hypothesis will be as follows:

How do changes in present contexts (economic, technological, political and communicative) open for possibilities for members of certain low castes in the Kathmandu Valley to engage in new economic strategies which can be expected to have significant repercussions on their socio-economic position in Nepal society, as well as on the symbolic legitimacy of caste hierarchy?

1.2 The Fieldwork

The questions presented above guided my preparations for the fieldwork. There is a lot of information available on Nepal. The Khadgis are Newars, a fairly large ethnic group, that are considered the original inhabitants of the Kathmandu Valley. Quite a bit of research has been done on the Newars, but not much on the Newar low castes per se. Gellner and Quigley have published several books and articles on the Newars. The writings on the low castes, however, they admit, is not written on the basis of an in-depth study or prolonged personal experience. Gellner comments: "There is a considerable need still for fieldwork on Newar low castes, and, if it is ever done, it will certainly modify and quite possibly overturn some of the assertions of this chapter" (Gellner 1995:265). Quigley has done a comparative study of Khadgis and Shrestas in Dhulikhel (Quigley 1985), and Levy has written an extensive ethnography on Newars in Bhaktapur, dealing briefly with the Khadgis (Levy 1992). There is not much literature available on the Khadgi group, and when preparing for the fieldwork I had to limit my preparations to basic reading on Nepal and the Newars at large, and to improve my general knowledge on regional ethnography, on anthropological theory and methodology.

By the beginning of January, 1998, the fieldwork could begin. Scott Cook gives a schematical description of the process of anthropological fieldwork and research (Honigman 1973:805). This schema is presented in box 1.1; "The epistemological situation of the economic anthropologist".

Box 1.1 The epistemological situation of the economic anthropologist

The table is developed for use in economic anthropology, but might be applicable to other research as well. The figure describes how the fieldworker from his/her own culture (Matrix Society) brings with her on the one hand, (A) popular culture, personal opinions, impressions and assets, and on the other hand (B) received wisdom as a student and scholar. Both these aspects affect the way the student accumulates knowledge, and how he/she interprets and perceives phenomena in the social context in which the fieldwork is carried out (Context Society).

During fieldwork external social phenomena are observed and experienced trying to learn the action perspective (C), and information is gathered through interviews and participation, to ascertain the actor's view of the field (D). Through the fieldwork, there is an ongoing dialectical process that includes all these relations (between A, B, C and D) through which the researcher tries to gain successive approximations and successive abstractions.

When doing participant observation the researcher inevitably becomes part of the research material. It is important to reflect on and try to understand what role one has, and how one affects the research material through interaction, personal experiences and reflections. Anthropologists, particularly in the post-modern period, argue that reflexivity should be present at all stages of the research work, when gathering data, analysing data, and when transforming data into ethnographic texts. According to Nader the representation of "the other" should include reflections on one's own participation. (Nader 1994:88). I will throughout my exploration, try to show how my perceptions, my accumulated knowledge from my Matrix Society (A + B) are gradually altered and expanded by experiences and discoveries during fieldwork. Some of my revised or achieved knowledge will be presented in this thesis, as a result of the analysis. The "answers" I provide here are mine, limited by my views, my questions and by how I go about exploring the field. The product must be seen as a result of the dialectic process (between A, B, C and D), and may hopefully provide a new point of departure for researchers and for others interested in these questions and people.

The fieldwork was carried out over a period of eight months, from January, to the end of august, 1998. The first two months were spent taking intensive language courses in Nepali, getting to know the area, people, acquiring general knowledge of the society and culture, and applying for research visa (a very time consuming task, nevertheless instructive as it gave me insights into Nepalese bureaucracy and hierarchy). As I started contacting people from the Khadgi caste it turned out that my Nepali skills were not sufficient for conducting in depth interviews, neither was I learning enough through participation. I kept up my Nepali lessons, but I realised that I would have to get someone to help me with the interpretation. After a few months, and some considerations, I engaged an interpreter. There are some classical problems involved in doing fieldwork with the use of an interpreter, and they are elegantly described by Berreman (Berreman 1962). The social positions of both the interpreter and the fieldworker affect the situation and the information revealed. The social position does, however, not only limit access to information. Social position is a necessary requirement for access to information all together. It is important to try to understand what characterises the relation between the fieldworker and the informants, and how it affects the information conveyed. For several reasons the use of an interpreter, in my case, was a success. The relation between the interpreter and my informants provided me with considerable information. This will be further explored in section 3.2, 3.3, and 7.5.3.

Participant observation includes as the word implies both participation and observation. The anthropologist observes events. The events are analysed, interpreted and abstracted. The events that the anthropologist observes must be put into a context, a frame of reference. Events and actions must be analysed as meaningful acts, meanings that are reproduced within and across social fields. The anthropologist tries to find out what meaning is imbedded in the text and make out of what he or she observes. Part of understanding what the events mean, and how they are perceived, is to see the intentions. What values, interpretations and possibilities do the actors see in the moment of action? What social constraints and limitations are there? How do events and actions form the basis for new events and actions? In order to understand the individuals and their frames of action, empathy and involvement are needed. It is at the same time important to keep a certain analytical distance. It is the selection, abstraction, and presentation of events that makes the observed events into data. The requirements of distance must be seen in relation to requirements for empathy, sympathy and engagement, that is the basis for understanding "the other" (point taken from Borchgrevink 1997:27).

Throughout the fieldwork, I participated in many of the Khadgis' professional activities, following their daily routines. I participated in various informal social gatherings and special activities like marriages, private rituals, festivals, and organisational activities. I also carried out a number of formal and informal interviews. The Khadgis have an organisation established for the improvement of their community. It is called the Nepal Khadgi Sewa Somity (The Nepal Social Services Committee, hereafter abbreviated NKSC). The fact that the committee celebrated its 25th anniversary, or silver jubilee, during the year of my fieldwork, was strongly in my favour. The occasion provided me with the opportunity to participate in their meetings, in the elaborate celebrations and the activities organised on that occasion. I was introduced to a large number of Khadgis, otherwise difficult to meet, within the time-limits of a fieldwork of eight months.

Doing fieldwork and ethnographic analysis includes many processes of selection; selecting hypothesis, questions, informants, data, and interpretations. Interviews and participant observation with Khadgis form the main body of data. Informants have been chosen by different criteria. About half of the informants were randomly chosen, the other half were chosen to cover the variety in occupational activities, economic status, levels of education and political involvement, in order to cover the diversity of the group and their different experiences. The impressions and experiences of similarities and diversities that I encountered, had to be interpreted, selected, noted down and presented as data. I focused on collecting many types of data, quantitative and qualitative, oral and written, life stories and careers, mythologies and historical stories, economic situations and commodities used, speech and action, self-presentations, formal speeches and social interaction. Some information has been drawn from interaction with people outside the Khadgi group to get an idea of the nature of their relation to and perception of the Khadgi caste. My interpreter represented a valuable source of information. Other sources of information have also been useful. These are listed in box 1.2.

I have gathered quantitative data that can give valuable insights to qualitative data and vice versa. It is my opinion that anthropology, in its basic nature, should be exploratory in the use of approaches and material. It opens up rather than closes off the continuos process of understanding and ascribing meaning.

Box 1.2. Additional Sources of Data

1.3 Bourdieu, Theoretical Framework and Analysis

It is traditional to make analytical distinctions between culture and society, organisation and structure, patterns for and patterns of behaviour etc. Culture is often described as a system of ideas or meanings, that a person needs to know to operate in a society. Society has, on the other hand, been described as what actually manifests itself empirically. Overemphasising culture as a meaning system that structures society, might lead to a static view of culture, explaining only the recurrence of order in everyday life, not changes and deviations. Some anthropologists focus more on society, on the empirical level, and the complexity of the agents' everyday life. Too much of an emphasis on behaviour and the manifestations of social life may lead to problems of generalisations. Anthropologists commonly try to capture the dynamics between these two analytical dimension.

Bourdieu seeks to avoid or transcend the oppositions both between culture and society and between objectivism and subjectivism. He calls his approach "stucturalist constructivism" or "constructivist structuralism" (Bourdieu 1990:123, 124). By constructivism Bourdieu means a social genesis that is constructive on two dimensions: on the one hand a social genesis of patterns of perception - on the other hand a social genesis of social structures, life fields and groups, social classes. By structuralism he does not mean structure in the sense of classical structuralism à la Levi-Strauss and Saussure.

Bourdieu holds that in the social world itself objective structures exist that are independent of agents and are capable of guiding and or constraining their practices and representations. This is how he describes objective structures:

"The objective structures which the sociologist constructs in the objective moment, by setting aside the subjective representations of the agents, are the basis for subjective representations and they constitute the structural constraints which influence interactions. These representations have also to be remembered if one wants to account above all for the daily individual and collective struggles which aim at transforming these structures." (Bourdieu 1990:125,126).

In this way he proposes to explain both stability and change. I will not elaborate on the implementation of Bourdieu's focus of objectivity and subjectivity, although it reaches into the many anthropological meta-theoretical and methodological discussions on epistemology, interpretation, relativism, universalism and comparison. Jenkins criticises Bourdieu for not transcending the objectivist-subjectivist dichotomy but remaining firmly rooted in objectivism (Jenkins 1992:175).

Anthropology, Bourdieu holds, should be an analysis, a systematic way of thinking about social life, and more than merely a representation of representations, and thus a secondary representation. One must go beyond actions and events and see patterns or parameters that structure people's lives, and structure structures. One should discover and suggest possible connections. Bourdieu, like Lucien Goldman, suggests a sociology of world views: There are world views or common sense constructions of social reality that have constructive power, he holds.

Bourdieu has made important contributions to social science, although not all of his arguments and insights are new. I shall make use of his substantial theories, and use some of his concepts as a framework for interpretation and presentation. I find some of Bourdieu's theoretical framework useful for describing and analysing my research material. The concepts I have found particularly useful and will apply to my analysis are the concepts social space, social groups and social fields, the various concepts of capital; both symbolic, cultural, social and economic, and the concepts of symbolic power and symbolic struggles.

To conceptualise culture and society Bourdieu introduces the concepts of social space and social fields. Through these concepts he is able to describe social units and their distribution in the social "topography". A social field is made up of agents sharing many of the same properties. The agents, groups or institutions "…have more properties in common the closer they are to each other in this space; and fewer common properties, the further they are away from each other… People close to each other in social space tend to be closer together - by choice or necessity - in the geographical space" (Bourdieu 1990:127). Interactions that take place in geographical space must not be taken as face value, since interaction itself is not a sign of proximity in social space, and might conceal the structures that are realised within them.

According to Bourdieu there exist objective homogenities of conditions and dispositions that form a basis for the formation of social fields. "A field, therefore, is a structured system of social positions - occupied either by individuals or institutions - the nature of which defines the situation for their occupants." (Jenkins 1992:85) The individual's position in a field is determined by internal constructions of value within that field. Jenkins interprets Bourdieu's social field in this way;

A field, in Bourdieu's sense, is a social arena within which struggles or manoeuvres take place over specific resources or stakes and access to them. Fields are defined by the stakes that are at stake - cultural goods (life-style), housing, intellectual distinction (education), employment, land, power (politics), social class, prestige or whatever - and may be of differing degrees of specificity and concreteness. Each field, by virtue of its defining content, has a different logic and taken-for-granted structure of necessity and relevance which is both the product and the producer of the habitus which is specific and appropriate to the field. (Jenkins 1992:84) (I will not apply Bourdieu's concept "habitus", as it in my opinion obscures more than it clarifies)

A field is structured internally in terms of power relations. There also exist power relations between fields, powers that govern the field's structure and define the relation and rank between them. Some fields are more dominant than others. The power to define world order, to impose this order (and thereby the formation of groups) and define values, Bourdieu calls symbolic power. The formation of groups is only likely to succeed when they are based on reality; that is, on the objective homogenities that exist in the world. The dominant fields' view of world order is more likely to gain dominance the more it corresponds to the experiences and situations of the individuals and of groups (or fields). Construction of a world order is not made in a vacuum, but is developed in a process, along with existing views of the social space, and by concepts such as high/low, pure/polluted, male/female.

Social structures are internalised and tend to be seen as "natural". However, perceptions of social reality vary with the agent's perception. The perceptions of the social world always include a degree of indeterminacy and vagueness. The social world does not present itself as totally structured. The social world may be described and constructed in different ways in accordance with different principles of division - e.g. economic, ethnic, religious, or national principles. The uncertainty provides a basis for the plurality of world views and thereby a base for symbolic struggles. These struggles may be individual or collective. I will return to these struggles and the forms they can take in chapter 9 and 10, when discussing the strategies and symbolic struggles of the Khadgis.

Power and domination, in Bourdieu's work, are closely associated with different forms of capital. Capital is seen as different goods or resources that can be used to maximise wants and needs. The different forms of capital that Bourdieu distinguish are; economic capital, social capital (various kinds of relations to significant others), cultural capital (legitimate knowledge), and symbolic capital (prestige and social honour). Cultural capital is understood as legitimate knowledge on areas of importance and value. Cultural capital involves power over words, pictures, opinions, appearances and behaviour. The uneven distribution of cultural and economic capital results, according to Bourdieu, in two fractions, one economic based position and one cultural based position. Important social relations such as contacts, networks, family, acquaintances form the social capital. Capital is not given, but constructed differently in various social fields. According to Bourdieu symbolic capital is different from the other forms of capital in the sense that it is unquestioned, and therefore more powerful. Symbolic capital is the sum of economic, cultural and social capital.

According to Bourdieu agents are distributed in the overall social space, in the first dimension in accordance with the overall volume of capital that they possess, in the second dimension in accordance with the structure of the capital. Some of the capital is inherited, some capital is achieved or lost during that actor's life.

These concepts seem to be useful in identifying the individual's position in the social structure and for thinking systematically on what types of strategy are available for maximising this position, or what on the other hand minimises the position in the social field. Strategy will here be understood as conscious or unconscious attempt to protect, convert or gain capital on the part of individuals or groups. The concepts are also useful for discovering what kind of capital is approved within and between social fields. The concept of capital can be used to discover how power and status is constructed within and across social fields.

Capital is to be seen as properties and assets that agents possess. Bourdieu's concepts of properties and capital are very similar to the analytical concepts and framework developed by anthropologists in Bergen, e.g. Haaland, Barth, Grønhaug and others (e.g. Barth 1992, Barth 1994). They were concerned with distinguishing assets such as social relations, economic capital, expertise, honour and so on. They tried to discover and distinguish the conversion barriers between these different spheres of activa, noting that the conversion of capital from one form to an other is not a simple one-to-one exchange, but guarded by complex mechanisms. Bourdieu's use of the different terms of capital opens for discovering conversion barriers, although this is not something he elaborates on. I will use Bourdieu's concepts of capital to see what type of capital the Khadgis acquire, how it is acknowledged and what difficulties or barriers they face when trying to convert one type of capital to the other.

Within social fields, Bourdieu holds, people have "a sense of one's place", "a sense of other's place". These notions are the basis for all forms of co-operation; friendship, love-affairs, marriages, associations etc. Bourdieu also discusses how markers are used for identifying and distancing, displaying sympathies and antipathies; markers such as clothes, habits, ways of acting etc. The strategies agents employ for gaining and converting capital are affected by the agents' position in social reality, to what extent the social structures are internalised, and to what extent they find the world structured differently than within the dominant "world view". The Khadgis' actions may be seen as expressions of their sense of place and as attempts to manipulate their position in the social space. The actions either confirm or contest the dominant world view and the classifications, and as Marcus suggests, they take the form of resistance or accommodation (Marcus 1992).

I shall throughout the presentation and analysis, return to Bourdieu's concepts, and use them to present and order the data. Although the theoretical framework is developed in a very different social context (France and Algeria) it seems helpful when analysing aspects of the social life of the Khadgis.

1.4 Caste, Jati and Varna

Before I move on, I find it necessary to comment on the very complex and delicate phenomenon of caste. To deal properly with caste, a number of clarifications need to be made. This is essential in order to get a proper understanding of the analytical problems involved in the study of caste, and the practical problems involved in doing fieldwork in a "modernising" society where caste is an important social institution.

The word caste was first used by the Portuguese travellers to describe a social and cultural complexity they encountered when they first came to India:

The Portuguese seafarers who traded mainly on the west coast of India in the 16^th and 17^th centuries described groups they called castas (from which derive the English and French word caste), meaning "species" or "breeds" of animals or plants and "tribes", "races", "clans" or "linages" among men (Marriott and Inden 1985:348).

The term "caste" must be seen as concept from my matrix society, as scholarly knowledge (B), and is to be distinguished from the term "jati", that can be said to be "host" concepts, from the Context Society.

The nearest equivalent in the indigenous concepts of India and Nepal are jati (or jat) and varna. The term jati derives from an Indo-European verbal root and can be used to designate a distinct sex, a race, a caste, or a tribe, a population, the followers of an occupation or a religion, or a nation (Marriott and Inden 1985:349) or birth, origin or genesis, and refers to a strictly regulated social group or community into which one is born. Each jati has its own customs that restrict the occupation and dietary habits of its members and their social contact with members of other castes and outsiders. In general, a person is expected to marry within the jati, and follow a particular set of rules for proper behaviour according to one's jati.

The meaning of varna is quite different. In India four varnas are distinguished; Brahman, Kshatriya, Vishya and Sudra. Hindu myths tells how Shiva created the four varnas from different parts of his body; 1)The Brahmans were created from the mouth, and are thus associated with the head, and function as thinking and speaking; 2)The Kshatriya were created from the shoulders, and are associated with physical power and fighting; 3)The Vishya were created from the hips, and are associated with lust and trade; and 4)The Sudra were created from the feet, making them the service people. Untouchables or "outcasts" fall outside the varnas and can be said to make up a fifth group, coming from underneath Shiva's feet. The basic idea is one of function, functions that are needed to ensure that the social harmony and cosmic stability is maintained (Srinivas 1967).

While jati can be said to be the empirically identifiable unit, varna usually operates on an ideological level. The varnas do not form the basis of interactional and operationally joint groups. All essential units of caste operate on a jati level; as for instance marriages, commensal taboos, rituals and occupations. Football might serve as a metaphor for understanding the differences between jati and varna. While a jati can be said to form the different teams that are actually competing against each other, varna can be said to be the leagues or divisions they compete within. Rank between leagues, and varnas, is given and not contested. Leagues do not form a basis for co-operative action against other leagues. The teams, or jats, compete within the different leagues or varnas, and barriers between the leagues and varnas are relatively stable and hard to cross.

Social scientists have been criticised of putting too much emphasise on caste. This was also my concern when planning my fieldwork in Nepal. I was afraid to give it significance in situations where it might not be relevant. I did not want to make assumptions about its role and importance, but gradually let its importance unfold where I was made aware of it. However, caste for more than one reason became a very central theme in my fieldwork and in this thesis. The fact that the empirical unit I had chosen to study was in fact a specific caste group made the focus inevitable. I had to confront caste issues all the time. Secondly, I soon realised that it was important in an overwhelming number of situations. Caste in Nepal is such an obvious feature that it is impossible to ignore it. Even if I did not bring up caste issues, they were constantly brought up by others, in all types of situation and in discussions. Thirdly, I had read a lot about caste, but I had to find out how it functioned in practical life. To understand its importance in different situations and contexts, I had to put it on the agenda. I wish to let the role and importance of caste to the Khadgis unfold throughout the presentation of my ethnographic data. I shall, however, give a brief presentation of some different approaches to the study of caste, as it forms the scholarly knowledge (B) available in Matrix Society, and is important to my perception of the phenomenon in Context Society, and subsequently, to the result of the analysis.

Social scientists have proposed different ways of analysing and studying caste. There has been, and still is, a lot of disagreement on the issue. Definitions of caste vary with theoretical positions and what aspects one chooses to emphasise. Early studies focused on explaining and describing the caste system (Dumont 1970). The descriptive studies basically try to account for one aspect of the system, either it's hereditary character and the clear-cut separation between groups, or the division of labour, or hierarchy, or much more rarely, the religious aspect. Dumont criticises them for trying to grasp caste in some immediate way by starting from our own civilisation, and for trying to derive the whole from the part. The period 1900-1945 is, according to Dumont, characterised by progress in comparison, understanding and analysis, and by better appreciation of the place of religion. With Hocart the place of religion was heavily emphasised. Bouglé insisted on the presence of three characteristics: hierarchy, separation and interdependence. He based all three aspects on the opposition between the pure and the impure. Max Weber's work on caste was focused on the distinctions between economic class and status group, the nature of the division of labour, and comparison between the western and the Hindu universe.

Later approaches can be sorted in two groups, based on the focus, the aspects, and the problems they are concerned with. "Ideological" approaches and "sociological" approaches can be seen as two main tendencies, under which much of the work on caste can be categorised. I hold Dumont and Pocock as the most outspoken representatives for what I will call the "ideological approach". Ideological or "cultural" approaches to the study of caste emphasise ideology and religion. Many of the scholars of this tradition are of the conviction that the phenomenon of caste has to be linked to Hinduism. Dumont holds that caste is structured by an underlying and all-encompassing system of ideas and values, whose basic element is the principle of ritual purity, rather than secularism, politics and economy. In his view caste was a unique Hindu phenomenon that was to be understood on the basis of Hindu ideology. According to Dumont it is easy to isolate the principles behind the Hindu's view. Central to Hindu philosophy is the thought that all living beings are differentiated into genera, or classes, each of which is thought to possess a defining coded substance embodying its code of conduct or moral action (dharma). Dumont's analysis of caste is consistent and logical, and describes a system of ideas, but his theory entails a static view of caste society. Dumont is not concerned with the individuals and their actions, but with a system of ideas, and the relation between ideas.

Other social scientists, such as Barth and Berreman, argue for using the term caste sociologically, as a term applicable to all societies sharing certain elements of social structure. They argue for understanding caste as fundamental to social organisation, a hierarchical structure, where the term can designate similarly divided societies. The focus is more on empirical manifestations and variations. Sociological approaches to caste often focus on power and control of goods. Some have conceived of politics and power in non-monetary economies as the prime movers in the emergence of a caste system. It is often held that a caste hierarchy is an interaction in the form of exchange of goods and services, between inferior and superior parties. Scholars of this tradition emphasise the importance of elaborate jajmani systems. It is commonly argued that ritual rank and power has been over-emphasised at the expense of secular, economic and political rank and power. Béteille argues that caste is not only a ritual phenomenon where dominance is manifested through levels of purity and pollution (Béteille 1991). He stresses the importance of material dominance and exploitation. Caste should, according to the sociological tradition, be considered a system of social stratification where statuses appear in clusters. Berreman holds that caste should be used as a cross-cultural tool, applied without sacrificing the cultural context; "Caste systems are rigid systems of social stratification, but they are also systems of socio-cultural pluralism and both of these facts can best be understood in terms of distinctive patterns of social interaction" (Berreman 1985:333). McKim Marriott & Inden (1985) see caste as a moral system that defines rank, descent, marriage and occupation. They argue for using the concept of caste for comparative purposes, but not without taking the cultural context into consideration. While the ideological approaches focus on the relation between ideas, the sociological approaches focus on relations between people, but not without also considering ideas. In the study of caste, scholars have attempted to transcend the earlier dichotomies by focusing both on the ideological and the empirical levels. As I see it, caste can not be seen purely as a fundamentally religious or ideological phenomenon, nor as a purely political, occupational or organisational phenomenon. It is both at once. Ideology is available, produced and reproduced through social life, and social action is also motivated by ideology. As Bourdieu and many other scholars have pointed out; the distinction is useful, but the two levels must be seen in a dialectic relation to each other. In my opinion Bourdieu's theoretical framework lends itself to encompassing both levels, and the dynamics between them. With this framework we can identify the ideology and the practice of the Brahman, as well as the ideology and practice of other groups, and so the relations between them. Bourdieu's concepts of social fields and different constructions of capital allow us to see the competing ideological views, and the relation between them, and so discern the caste ranking system as contested.

At this point the term "caste" should be interpreted very loosely. I will introduce Bouglé's definition of caste to point to the complexity of the phenomenon. It covers what is often held to be the main characteristics of caste:

To fix our ideas we must have an initial definition. Let us take Bouglé's and say that the caste system divides the whole society into a large number of hereditary groups, distinguished from one another and connected together by three main characteristics: separation in matters of marriage and contact, whether direct or indirect (food); division of labour, each group having, in theory or by tradition, a profession from which their members can depart only within certain limits; and finally hierarchy, which ranks the groups as relatively superior or inferior to one another. (Bouglé in Dumont, 1970: 21)

Throughout the following chapters when dealing with the Khadgis, caste will refer to the indigenous term jati or jat, and should thus be considered as such when used by informants. I seek to use the word caste as they use the word jat by trying to find out what it means to them. It should therefore be seen as a concept from Context Society. In the concluding theoretical and comparative discussion, caste will be used as an analytic anthropological concept, and should be seen as a concept taken from Matrix Society, as scholarly knowledge (B).

1.5 Disposition

This introduction presents the focus of the research. I have presented the starting point for the fieldwork, and some methodological problems related to fieldwork and analysis. The theoretical framework and important concepts that shall be applied throughout the analysis have been presented. A clarification of the terms caste, jati and varna has also been given. Some main trends in the anthropological debate on caste have been dealt with, to give an indication of the analytical problems involved in the study of caste.

Chapter two deals with ethnographic contextualisation. Various context are described, first situating Nepal in a global context, and then closing in gradually on a local level, the Kathmandu Valley. Aspects of the different levels are described to identify the frames for action the different people in Nepal have to operate within. Some general characteristics of Nepalese society are presented briefly, and the Newars and the Khadgis are introduced. At the end of chapter two I discuss how the Khadgis can be said to form a group, and how I wish to apply the concept of field to the group and to caste. The last part of the chapter also seeks to apply Bourdieu's concepts of social space, capital and symbolic power to the phenomenon of caste.

In chapter three I have attempted to show how my first encounters gave me a basis for understanding some aspects of social life in Nepal, insights that only gradually became clear to me. These encounters were crucial to the focus of my approach and my exploration of the field. I attempt to give a preliminary understanding of how objects, persons, activities and places, and particularly in our case, the butcher, the butcher activities, animals and the animal's bodily substances, are associated with the concepts of purity and pollution, touchability and untouchability. I attempt to show how these concepts have had, and still have, a profound impact on the structures of the old city and the settlement patterns of the Khadgis.

The concepts dealt with in chapter three are crucial for understanding the constraints and situations under which the Khadgis live their lives, and among other things, for understanding their occupational choices. But in order to understand the occupational choices strategies it is essential to take many practical considerations into account, and try to see what their caste ascribed occupation, the butcher occupation, implies. Chapter four seeks to uncover what it implies for a Khadgi to be a butcher by occupation, and what advantages and disadvantages the butcher occupation face. This will be discussed in some detail.

In chapter five I shall present the occupational distribution of the Khadgi group, and what characterises it. Many factors affect the Khadgis' occupational choices. Among these are their experiences of opportunities and discrimination, and their attitudes towards education. Social rank, capital, and assets are crucial to individuals' and groups' opportunities. This is discussed in the concluding part of the chapter.

Chapter six takes up the issue of the Khadgis' role of being service providers towards other castes. Their "traditional" duties are discussed, and the extent to which these duties are performed today. The discussion includes the duties and traditions of drumming, toenail cutting, milk selling, and midwife services, among others. The Khadgis' attitudes towards the "traditional" duties, and other types of services, are discussed and analysed with the use of the concepts of cultural and economic capital.

Chapter seven takes account of the Khadgis' social organisation, describing the most important units for interaction and what criteria they are based on. Different levels and units of social organisations are discussed, from family and household to local groups, and larger organisations based on caste identity. The chapter seeks to discover to what extent inter caste relations are dominant and what characterises the inter caste relations. In the concluding part of the chapter social relations are analysed in relation to the concepts of capital.

In chapter eight I will deal with the Khadgis' construction of identity, traditions and ritual practice. History and various myths are presented to explain how the Khadgis through misfortune got low status, and how they became associated with the butcher occupation, stories through which they claim royal descent. Naming or names are central to caste identification, and is a very intricate matter. The chapter also deals with variations in ritual practice; what rituals are performed at present, how they are performed, how the present practice departs from the "traditional" practice, and how new ritual practices have developed. Religious practice, and strategies for gaining religious merit can be seen as strategies for converting and gaining cultural capital.

In chapter nine a further exploration of Bourdieu's theory on symbolic power and symbolic struggles is given, before reviewing the institutionalisation of caste hierarchy. Scholars have used several concepts when describing change and structures of power in societies with caste institutions. Some of these concepts will be reviewed critically with reference to my empirical data. "Westernisation" is a concept often applied. It is a problematic one, but nevertheless relevant. In the last part of the chapter I shall show how it is relevant to change and the symbolic struggle in Nepal.

Chapter ten attempts to show how the symbolic struggle is taking place in the Kathmandu Valley, and how the Khadgis' strategies for social mobility are part of this symbolic struggle. Their strategies, individual and collective, are summed up and discussed in relation to the concepts and the discussion introduced in chapter nine. The strategies are formed in the process of the struggle, and as battles are won and lost the strategies change and become more or less effective. Throughout the process of the struggle the Khadgis social position is affected. New criteria for rank and for defining and evaluating capital are introduced. The changes have consequences for the categorisation of social fields, for the role of caste and the legislation of caste hierarchy, and it has significant repercussions on the social position of the Khadgis.

Chapter 2: Orientations and Context

This chapter seeks to situate the Khadgis' lives in a wider context and to describe some conditions that frame the lives of the Khadgis and other people in Nepal. The chapter deals with the various contexts, from the global to the local. After placing Nepal in a global context, I will close in on the Kathmandu Valley, the Newars and the Khadgis. A short political history of Nepal is given, describing some of the periods that are of particular importance and interest for this study. The ethnographic description of Nepal will be brief and somewhat superficial, but it will function as a description to those not familiar with the ethnographic area. It represents the knowledge I had acquired when starting my ethnographic research, and the scholarly knowledge that was available to me in Matrix Society. Bourdieu's concepts of capital will be applied throughout the chapter, and in the last section I shall apply his concepts of social group and social field to the Khadgis.

Map 1. Nepal
(http://travel.yahoo.com/Destination/Asia/Countries/Nepal//map.html)

2.1 A Global and Comparative Perspective

The world is undergoing rapid change due to processes of globalisation. The world becomes a "smaller" place, as people from different parts of the world are brought in closer contact with each other. As Hannertz puts it: "Culture travels in our time quickly and over great distances." (Hannertz 1992:43). Even though Nepal has many remote corners, all Nepalese people relate to different levels in various ways; the nation state, world economy, world ecology, global forces of power, etc. Their lives are affected by the country's financial situation, the country's development strategies, the foreign political situation, imported technology, and so on.

Nepal is undergoing, and is about to undergo, enormous changes. There are many factors contributing to this: modern technology, improved communication (e.g. roads, air traffic, media), and tourism, migration (e.g. Gurka military service, labour and other migration, urbanisation), foreign and domestic investments. Nepal has established many global and international relations during the last decades. The engagement of other nations and international organisations in Nepal has integrated Nepal in the world economy. The majority of the population still, however, live in a traditional way without access to modern communication, technology and transport. There are innumerable remote villages with many days' marches to the nearest road. You can find children who have never seen a car, but who see jumbo jets daily.

Nepal is classified by the United Nations as one of the least developed nations, with a per capita income of US$ 210. This is among the six lowest in the world (Nepal Human Development Report 1998, hereafter abbreviated NHDR). This reflects the fact that there is little circulation of capital, little production in the service- and industrial sectors, and little wage labour. The growth in domestic production is less then the annual population growth. Nepal is poor in economic capital, in the terms Gross Domestic Product (GDP). It also has few natural resources and is dependent on economic aid. The population have a potential in hydroelectric power, but do not have the means of utilising it without assistance. The human development index (HDI), was 0.378 in 1996 (see box 2.1). The Human Development Index can be seen as an indicator of the level of social and cultural capital, as it takes into account not only economic development, but health, knowledge, social and political participation and gender empowerment. In a regional perspective, all countries, except Bhutan and Bangladesh, are better off than Nepal. The HDI values for Pakistan and India are higher than for Nepal by nearly one fourth (ibid.). The table in box 2.1 also shows Nepal's capability poverty in a comparative perspective (CMP). The level of CMP reflects deprivation due to mortality, undernourishment, lack of information and knowledge. "CMP draws attention to a set of capabilities which are basic for human functioning: capabilities to lead a life free of avoidable morbidity, to be informed and educated and to be well nourished and able for healthy reproduction." (NHDR 1998: http://www.nepali.net/undp/keydoc/nhdr-c03.html#3.2) The level of capability poverty in Nepal is very high compared to the levels for Sri Lanka and the Maldives. Only Bangladesh is poorer in capability than Nepal. Many Nepalese lack education and technology, and much of the population is poor, undernourished and without access to resources.

Box 2.1 Nepal Human Development Report

Both HDI and the CMP (Box 2.1) may serve as indicators of the level of cultural capital. The small amount of social capital can also be seen in the country's unfavourable position with regard to foreign policy. They do not have strong alliances, as they are relatively powerless in relation to neighbouring countries in the South Asia region. For some indicators on human development and lack of capital see Box 2.2 and Appendix.

It can be said that Nepal, in a global and regional perspective, is a country with an extremely small amount of capital, both economic, cultural and social, in the way these forms of capital are defined in Western terms. To this comes the problem that what Nepal has of capital, economic or otherwise, is very unevenly distributed, and concentrated in the hands of a very few of the country's 21 million inhabitants.

Box 2.2 Indicators on Human Development in Nepal

Development measurements raise as many critical questions as they answer, questions of both quantitative and normative character; What is the relation between the local and the global? What kind of influence and exchange is taking place and through what channels? What kind of power is maintaining the relations? How relate to the asymmetrical stream of information and influence? What kind of modernisation is favourable, and to whom? Who lays down the premises? Such questions represent a great challenge for researchers in many fields, both thematically and methodologically. In my opinion, anthropology as a social science has an important role to play in these matters. While other sciences can contribute to understanding some aspects of the development, anthropology with its wide and comparative approach can tell us something about how different forces work together, and often reveal connections beyond disciplinary boundaries. The global context and global issues represent more than one challenge. Macro questions are often hard to operationalise and limit. Many anthropologists hesitate to work with macro questions for fear of becoming too theoretical, too abstract and too general, and so lose the grip on concrete reality. Anthropologists' fear of macro questions might lead to too much concern with micro-studies. The research may become too concrete, limited and thereby lose some of its relevance. The challenge is, as I see it, going from the local to the global, or as Hylland-Eriksen puts it; "Fra små steder til store spørsmål" or "from small places to big questions" (Hylland-Eriksen 1994, my translation). One should raise topical hypotheses and problems, and answer them with solid empirically grounded arguments.

It is my hope that this thesis can be of importance and relevance to anyone interested in development problems, questions on cultural diversity, aspects of power and influence, caste and modernisation, as well as to the Khadgis and people in the Kathmandu Valley.

I will return to the discussion of the importance of the global context and macro-questions in the final chapter.

2.2 Diversity, the Land and the People

Nepal is situated in South Asia, between India and Tibet. Although it is a small Himalayan state of 147,181 square km, it possesses a striking cultural and geographical diversity. Nepal has one of the worlds steepest landscape profiles, reaching from the altitude of 8849 meters, to the tropical plains at 100 meters above the sea level. 15% of the land is cultivated, 13% is forest, 31% is forest, and the rest is mountain area (Lauridsen 1995). The Kathmandu Valley is situated at 1350 meters above sea level. Temperature and climate varies immensely with altitude and topography.

The population is no less varied than the country's nature. People are socially segmented along lines of caste and ethnicity. Nepal has been a gathering point for many different peoples, in particular Tibeto-Burman and Indo-Aryan. The largest Tibeto-Burman groups are Gurung, Limbu, Rai, Magar, Newar, Sherpa, and Tamang. The largest Indo-Aryan groups are Indian, Brahmin (Bahun), Chetri, pilgrims, travellers, traders, and service-providing low caste groups. The National Ethnic Groups Development Committee has identified 61 different groups, and the National Language Policy Advisory Commission lists 60 living languages (NHDR 1998: chapter 1.2.4). The languages and dialects can be traced to four language families; Tibeto-Burman, Indo-Aryan, Austro-Asiatic and Dravidian. Through the establishment of the Nepalese Kingdom, all the various groups have been subsumed to the nation state, and Nepali as a unifying language has been introduced.

Although Nepal has a strong development of local communities, there are many overarching cultural themes. The cultural features briefly presented in the following, touch different aspects of social organisation, such as caste, kinship, religion, system of production, economy and political organisation. The description of social elements is, as mentioned, very general, but none the less relevant for the fieldwork and analysis, as it represents the knowledge of Context Society, available in Matrix Society.

2.3 Descriptive Elements of Social Organisation

In Nepal people commonly order and group people into castes, tribes and ethnic groups. There is a complex relation between the categories. The Newars are considered an ethnic group, but they have their own hierarchy of caste groups.

In Nepal there is a tradition of listing castes in order. Hierarchical and ladder-like presentations and symbolism of caste is thus a native, and not only an anthropological construction. The division of the people into castes can be seen in the country's old laws. Box 2.3 shows the ordering of castes and ethnic groups according to the 1854 Law Code (Muluki Ain). This law code has been studied by Höfer, and is presented by Gellner (Gellner 1995:11). Other schemes are more detailed and have other ways of separating "clean", "unclean" and "untouchable" castes. The hierarchical order of society applies to most fields of life, within the family, in rituals, in organisations, at work, in companies, institutes and universities, the government, and so on. Castes, tribes and individuals are ranked according to several principles of inequality; purity and pollution, religion, economy, political power, seniority and gender.

As castes are thought to consist of exogamous patrilinages, kinship and caste are closely connected. Caste ranking is further closely associated with religious concepts and religious practice, and very often with Hinduism. In Nepal's case the phenomenon of caste is complicated by the fact that Buddhists and various ethnic groups are incorporated into the caste system. There is a syncretism of religious practice and traditions. Some hold that, in Nepal, Hinduism is concentrated in the south and Buddhism in the north, but they are tied together and interspersed by tantrism, animistic rituals and shamanic traditions. Officially 90% are Hindu, 8% Buddhists and 2% Muslim (Lauridsen 1995), but these numbers vary due to the complex relation between the different religious practices. The major cults are Vajrayana Buddhism and Tantric Hinduism. Both creeds have been established since antiquity in the Valley.

Nepal is sometimes characterised as having the world's most ritual culture. Religion is a dominant cultural trait that penetrates a majority of social institutions in Nepal. An enormous number of rituals, festivals, religious statuses and religiously motivated tasks contribute to the rich religious life and cultural variation of Nepal. Through religion frames for action, identity, world view and moral codes are constructed. In western societies it is possible, both analytically and practically, to draw a line between many religious, political and economic activities. In Nepalese society these lines can not be easily drawn. This is reflected in the Nepalese word for religion, dharma, which also means morale and duty. Different people have different levels of ritual purity assigned to them on the basis of their caste identity, and have different codes for proper moral behaviour.

Box 2.3 All-Nepal Caste Hierarchy of the 1854 Law Code

The ordering of people into castes is also important to the distribution of rights and duties. According to the old laws each caste group had prescribed duties and occupational specialities. As mentioned in the introduction, such a division of labour is often referred to as a jajmani system. In a jajmani system services and goods are thought to be exchanged within a local context, in contractual, personal and hereditary relations, between a patron and a client. The payment is made in grain or other products. This form of distribution has stronger resemblance to feudalism than to a modern market economy. Many anthropologists hold that the jajmani system has been overemphasised and simplified, stating that it is not at all a pan-Indian phenomenon, that it does not refer to a systematic set of relations that may be isolated from others, and that it is seldom found in a "pure" form. Jana Fortier holds that it should not be considered a system, but merely an important part of a network of labour exchange practices. The jajmani system, she continues, is bound up with other labour exchange practices; reciprocal labour, sharecropping, labour parties, or group labour, bonded labour and so on (Fortier 1993). Critics hold that the term obstructs rather than facilitates a comparative framework. The caste prescribed division of labour and exchange of services have characterised many villages and communities in Nepal. However, in Nepal, the system of exchange is complex and woven with many threads. Jajmani systems are, in Nepal, part of larger systems, and there are many examples of this type of "closed" exchange systems existing side by side, and integrated into, modern market systems. New modes of production, technology and regulations introduce different kinds of relation between people.

The dominating modes of production are farming, crafts and industry. At least 80% of the population engage in farming. Many live on an existential minimum, often by cultivating small terraces carved out of the hillsides. Most of the businesses are situated in the cities. These are mostly small companies with only a few employees. Subsistence production where goods and services were exchanged through contractual relations, is slowly being replaced by larger units of production, with wage labour and production for a market. The carpet industry and tourism are Nepal's largest industries.

Formally, Nepal is a constitutional monarchy. Traditionally the power of the king was ritually legitimised. The king was thought to be a reincarnation of Vishnu. King Birendra Bir Bikram Shah Dev has somewhat more power then e.g. the British Queen. After the revolution in 1990, Nepal has been struggling to established a workable democracy. Nepal has a parliamentary multi-party system. The Nepalese Communist party and the Congress party are the major fractions. As Gellner remarks, caste has been little politicised.

Although caste no longer has the force of law in Nepal, there has been nothing like Mahatma Gandhi's campaign to overcome prejudice. Nepal has no equivalent to the Indian Constitution of 1950, which supports positive discrimination, or of the different states' Temple Entry Acts of the 1930s and 1940s, forcing clean castes to admit Untouchables to their temples (Galanter 1972). Low castes in Nepal have until now remained very little politicized. Politicians have refrained from advocating positive discrimination, presumably from fear of antagonizing their clean-caste supporters, but also from fear of promoting divisiveness (Höfer 1979:205-7). (Gellner 1995:289)

With democracy and a multi-party system, where also low castes hold mandates, this is liable to change. New ideas on equality, human worth, and equality before the law are emerging.

Box 2.4 A Brief History

Plates 2-1 -  2-3:
Plate 2-1 The Colourful Culture of Nepal
Plate 2-2 Newar Stucco Work
Plate 2-3 A Typical Newar House

2.4 The Kathmandu Valley and the Newars

The Kathmandu Valley houses the capital and 1 million people. This number is growing rapidly, due to the general population growth and the process of urbanisation. Gellner gives the following description of the situation in the Kathmandu Valley:

Since 1951 the Kathmandu Valley has experienced rapid development, benefiting far more than the rest of the country because it houses the capital. The Valley now has electricity, drains, metalled roads, hospitals, schools, colleges, a university, numerous businesses, scores of foreign aid agencies, hundreds of hotels, digital telephones, cinemas, video parlours, and television. There has been a massive immigration by Nepalese from the hills, by refugees from Tibet, and more recently by poor Indians from Bihar. Overcrowding, inadequate drainage, pollution, traffic jams, water and electricity shortage have all worsened since the end of the 1980s. Land prices have spiralled, especially within or near the Kathmandu-Lalitpur conurbation enclosed by the ring road. Tourism, "development" and entrepôt business between Hong Kong and India have become major "industries". (Gellner 1995:13.)

There is a fascinating mixture of old and new, and the contrasts are striking. Internet cafes and traditional clay houses stand side by side. Many people who live in the city reside in traditional houses, without running water and electricity, others live in new armoured houses with all the modern equipment imaginable. One sees people wearing saris and traditional dresses, teenagers wearing jeans and t-shirts with bragging western brands, local tea-shops and American styled fast food restaurants. The sounds of traditional ritual music and western pop music drown each other out. One can see young ladies on new scooters with their hair and the saris fluttering, and old men carrying loads for a living, without other means of transportation. The city also displays the contrasts between rich and poor. A couple of blocks away from the enormous Royal Palace, poor children spend the night sleeping on cardboard and the day begging or searching the garbage piles.

There is, furthermore, a striking contrast between urban and rural areas. Kathmandu is the political, economic, educational and communicational centre of Nepal. It is considered the most developed part of Nepal along with some areas of Terai. Box 2.5 shows how human development is unevenly distributed according to geographical districts. The human development index for the Kathmandu Valley is considerably higher than for instance that of the hills. Most of the economic and social and cultural capital is concentrated in the urban areas, and the urban areas are the locus of symbolic power.

The Newars are considered the original inhabitants of the Kathmandu Valley. After the caste system was abandoned there is no statistic record covering and classifying the different caste groups. The Newars form a very large group and it is estimated that they make up 60 % of the population in the Kathmandu Valley (NHDR 1998). The Newari culture is notably urban. The Newars have a high degree of specialisation of labour. There was and still is to a large extent, various specialists who inherit their profession according to caste: e.g. priests, astrologers, masons, painters, stone workers, bronze workers, copper workers, silver- and goldsmiths, carpenters, traders, gardeners, dyers, washer men, tailors, butchers and sweepers. There is often a concentration of certain specialists in certain localities (tole).

The Newars are proud of their culture and tend to think of themselves as being the true culture bearers of Nepal, referring to the development of crafts, the cities, and the complicated network of kin and Guthi relationships. They form a highly self aware group, with many distinct traditions that mark them off from other ethnic groups and peoples. They have their own calendar, festivals, food traditions, clothing, house style and settlement pattern. They even have their own Tibeto-Burman language, Newari. Most Newars speak Newari in addition to the official language of Nepal (Nepali), but very few can write it using the original script (Løwdin 1986:7).