DAVID N GELLNER

I

None the less, in the officially correct sphere, caste has not been a respectable source of identity for quite some time. A shared national identity was supposed to define all Nepalis until 1990. But the kind of national identity that was propagated in schools and through government organisations was experienced as highly exclusionary by lower castes, by ethnic groups (janajatis), by religious minorities (Buddhists, Muslims, and increasingly now by Christians), and by people of Indian ethnicity (madhesis) living in the economically crucial Tarai region in the south of the country. If the period 1960 to 1990 was one of nation-building, the 17 years since then has been a time of ethnicity-building. New identities have been forged, new organisations set up, and new claims made; everything is still in a state of considerable flux.

Weight of the Past

In the 19th century and well into the 1980s wealth was primarily measured in land. Those groups which held large tracts of land were powerful, those who never held much land (dalits) were dependent on others, and those who tended to lose land or had not enough to survive (frequently the janajatis, but many chetris and bahuns were among them) had to leave and look for work and land further east in the Himalayas, with many ending up in Darjeeling, Sikkim, Bhutan, and the northeast of India. Many joined the gurkhas and used their pensions to invest in land in their home village. Today Nepali migrants head into Indian cities, to the Gulf, to south-east Asia, and east Asia. There are supposed to be half a million in the Gulf and another half a million in the developed economies of the north. Combined remittances are worth around a billion dollars a year, something in the region of 12 per cent of gross domestic product. Retired soldiers and workers often no longer retire to their villages, however. If they return to Nepal at all, they are more likely to settle in Kathmandu, Pokhara or Biratnagar.

Villages are places to visit with your children, but not to settle in.

During the autocratic Rana regime (18461951) society was ordered according to orthodox Hindu notions. The national legal code of 1854, the Mulukhi Ain, explicitly attempted to apply the ‘dharmasastras’ to the civil and criminal law of the heterogeneous kingdom [Höfer 1979]. All groups were equally called jat. The key distinctions, supported by law and the judicial system, were between the wearers of the sacred thread, the Tagadhari, on the one side, who were the elite of the society, and the rest, known as the Matwali, or alcohol-consuming classes. The subordinated groups were themselves divided into enslaveable and non-enslaveable categories, and into “clean” castes and “untouchables”. All the present-day “tribal” minorities were in the Matwali, but clean, category. Some were enslaveable and others were not. Some were able to win promotion to the non-enslaveable category if their elites were sufficiently well connected. It was a gradual process, some Matwalis achieving non-enslaveable status (and the legal right to be recruited into the army) earlier than others.

The dominant groups who spread throughout the country as landowners, priests, administrators, soldiers, and policemen, were the bahun (brahman) and chetri (kshatriya) castes. With them went associated low castes, principally kami (blacksmiths), sarki (leather workers), and damai (tailors). Together these groups are called parbatiyas (hill people) or pahadis. (Some refer to them as Indo-Nepalese.) This was a simple but effective caste system and the tribal groups were slotted into the middle, below the chetris, but above the “untouchable” artisans. One way in which the dominant group spread so successfully was through intermarriage. The offspring of chetri men and tribal women were accepted as chetris, though of lower status, and after a number of generations of strategic alliances could even achieve a more respectable full chetri status. The offspring of bahun fathers and tribal mothers also became chetris, in a striking departure from the biologically determined identity logic of modern times. It is no surprise then that the chetris, the broad and dominant category to which the kingly thakuri sub-caste

Economic and Political Weekly May 19, 2007 belongs, is the largest single group in the country as a whole and is found everywhere from west to east.

Accommodating the newars, the indigenous inhabitants of the Kathmandu valley, within the overall hill caste hierarchy was more problematic, because they possessed their own, more complicated, caste system, with about 20 castes, similar to those found in north India. The solution was to split the hierarchy and distribute the castes roughly where their ritual status warranted. In the same way, and even more so, the madhesi world of the Tarai remained apart from this hill synthesis, more attuned to the culture and language of those they intermarried with on the other side of the border with India.

From 1960 (strictly, from 1962, when the new constitution was promulgated) until 1990 Nepal was ruled by a system known as partyless panchayat democracy [Whelpton 2005]. Instituted by the present king’s father, king Mahendra, it was supposed to be more “suited to the soil of Nepal”, and was designed to mobilise the country’s various groups for the “all-round development” of the country. Ethnic and caste affiliations were discouraged, in the name of patriotism and nation-building. Organisations could be formed for cultural purposes, but not in order to advance the cause of particular social or regional interests politically. All political parties were banned.

With no reservations as in India, nor even any development initiatives specifically targeting “backward” groups, the lion’s share of the fruits of development and rapidly expanding educational opportunities and rewards went to those groups who were already well connected and had long established traditions of literacy and academic study, namely, bahuns, some chetris, and some (principally high-caste) newars (BCNs). On the rare occasions when figures were collected on the proportion of BCNs in high education or the professions, they were considered too explosive to publish. People were told there was no ethnicity in Nepal. Gopal Gurung wrote a book about it under the panchayat regime, calling it Hidden Facts in Nepali Politics. He was put in prison for it in 1988 and after 1990 started the Mongol National Organisation, which won a base in local government in east Nepal.

The situation in the Tarai during this period began to develop in ways which would store up trouble for the future. In the 19th century there were some areas of ancient settlement, as around Janakpur, but most of the Tarai region was covered with jungle and was sparsely populated by the tharus and other tribals who had some degree of inherited immunity to malaria. The British colonialists had allowed the Nepalese state to retain this strip of territory in order to ensure Nepal was economically viable, and indeed two districts in the far west, Kanchanpur and Kailali, were returned to Nepal’s rulers for their loyalty after 1857. Being flat and warm, the land was much more fertile than the hills. The Ranas in the 19th century were concerned only with expanding their revenue base, and were not at all interested in what language their tenants spoke. They encouraged settlers to move in from further south with five- and 10-year tax breaks.

In the 1950s and 1960s a programme of malaria eradication meant that the Tarai region became much more hospitable for people from the hills. Many who owned land already, but visited the Tarai only in the cold season, now settled there for good. Many others sold their land in the hills and moved south to the more agriculturally attractive flat land. Often they moved into areas where tribals had practised shifting cultivation and over a matter of decades they were able to acquire the tribals’ land, thanks to tribal innocence in matters of moneylending, land deeds, and dealing with state officials. In some cases whole villages of tharus moved further on, to the far west Tarai of Nepal, for example, where there was still virgin forest.

For the non-tribal (caste-organised) people who have lived in the Tarai for generations, and are called madhesis, there are cultural, kin, educational, and political links with Uttar Pradesh and Bihar. The border with India is therefore a border that is, for many purposes, merely notional. Many of Nepal’s madhesis take their wives from India, and marry their daughters there in return. Many study there, shop there, and move to and fro frequently. Nepali and Indian citizens may move and work across the border without let or hindrance. This makes the Nepal-India border fundamentally different to India’s borders with any of its other neighbours. It means that Nepal cannot, in truth, carry out any economic policy that is different from India’s, because goods will simply flow across the border in whichever is the economically advantageous direction. The current Nepal government’s suicidal policy of not raising petrol and kerosene prices means that Nepalese subsidised petroleum products are smuggled on a huge scale from Nepal into India at enormous daily cost to the Nepalese exchequer.

Nepal is a nation forged in migration: migration of people from west to east through the Himalayas, as well as from south (the plains) to north (the hills), from north to south (to settle the plains), and even (as in the tharus fleeing land appropriation) from east to west. As noted already, many Nepalis did not stop at the borders of Nepal, but migrated further east to Darjeeling, Bhutan, Assam, Burma, and beyond. Some have returned, willingly or unwillingly, and settled in the Tarai. Others, born and bred in Darjeeling, and therefore educationally more advanced than homegrown Nepalis, migrated to Kathmandu to man Nepal’s nascent private sector in the 1970s and 1980s. Trade, labour migration, and land shortage and the need to work, and now the availability of education and opportunity all over the globe, have ensured that Nepalis keep moving.

Growth of the Janajati Movement

In 1990 the panchayat system was overthrown by what has come to be called “the people’s movement” (‘jan andolan’). There

Table 1: Population Breakdown of Nepal (2001 Census) (Total: 23.15 million)with Figures for Hill Minorities

(Per cent)

Notes: Dalit = former untouchables; janajatis, underlined, are mainly those who were formerly called hill tribes; 59 groups were officially designated as janajatis in February 2002, not all of which had been included in the 2001 Census. Estimated figures for language loss are taken from Whelpton (1997, p 59). All figures are likely to be disputed.

Economic and Political Weekly May 19, 2007

were street battles and demonstrations for several months with 41 official martyrs before king Birendra agreed to legalise political parties and allow a new constitution. The new constitution defined Nepal as “a multi-ethnic, multilingual, democratic, independent, indivisible, sovereign, Hindu and Constitutional Monarchical Kingdom”. Ethnic and religious activists were very disappointed that the word “Hindu” was still there; none the less, the addition of the words “multi-ethnic”, “multilingual” and “constitutional” was a genuinely new departure.

The revolution of 1990 ushered in an entirely new situation, and fired the starting gun, so to say, on a rapid and not wholly predictable process of ethnogenesis similar to what happened in India under the Raj and for rather similar reasons [Whelpton 1997; Pfaff-Czarnecka 1999; Gellner 2001]. Many people from minority backgrounds felt that a great weight had been removed from their shoulders, by the demise of the panchayat ideology claiming that all Nepalis were the same while in practice institutionalising parbatiya supremacy. Ethnic difference began to acquire increasing political salience, especially with the publication, for the first time in 1991, of data on caste and ethnic group affiliation. Hitherto hidden strengths and weaknesses were now revealed. Democracy, as all south Asians are aware, is a numbers game.

Nepalis and their friends have had to learn a new word for “tribe” since the beginning of the 1990s, namely, janajati. The term seems to have come into Nepali from Bengali, via Darjeeling. It was completely unknown in the early 1980s, started to be used in activist circles shortly before 1990, and now has wide currency among the political elite, though it is still far from being universally recognised in the wider population. It picks out those groups previously known as “hill tribes” (plus the tharus and similar groups, e g, santals, from the Tarai). The distinction between caste and janajati thus corresponds more or less to the Indian distinction between caste and tribe – with the important proviso that in Nepal the janajatis comprise a much larger percentage of the population (the exact percentage is debated: some claim they are as much as 40 per cent or more of the overall population).

Table 1 does not include the high Himalayan groups, as they represent such small numbers in the population as a whole (Table 2). These groups of Tibetan culture, religion, and language prefer – for strategic reasons – not to be called “Tibetan” (the Nepali “Bhote” is often highly pejorative) but rather to form a large number of very small ethnic groups based on their particular valley or group of villages. These groups are also included in the janajati category.

Just as the second column of Table 1 does not exhaust the janajati category, so column four, the Taraians, cannot be labelled ‘madhesi’. The term ‘madhesi’ literally means a dweller in ‘madhes’ (‘madhyades’) or the plains. In theory anyone living in the Tarai could be considered a madhesi, but it has come to have an ethnic meaning (just like parbatiya or pahade, meaning “hill-dweller”). It refers to plains-dwellers of Indian, Hindu origin. Usages differ, but it seems clear that the tharus and other groups do not wish to be included in the category. Parbatiyas who are settled in the plains are definitely not included, and Muslims are also not full members either.

Janajati Politics

The janajati movement is driven, fundamentally, by anti-brahmanism and the overwhelming presence in public life of bahuns, revealed by the census to be only 13 per cent of the population as a whole (the point can be made more acerbically by saying that bahun men constitute only

6.5 per cent of the population but lead all the major political parties, etc). Politics, the judiciary, the universities, and the civil service are all dominated by bahuns. One attempt to quantify this, looked at the leaders of the judiciary, civil service, trade unions, etc, and found that bahuns and chetris together – just 31 per cent of the population – had two-thirds of the jobs, whereas hill janajatis (i e, excluding newars and tharus), with 22 per cent of the population had just 7 per cent of the jobs, and madhesis, with 31 per cent of the population had only 11 per cent of the jobs. Dalits with nearly 9 per cent of the population had just 0.3 per cent of the jobs.1

Some janajati activists have attempted to build a national alliance around Buddhism, in order to oppose the presumed Hinduism of the parbatiya high castes. The Buddhism they espouse bears a close resemblance to the engaged and politicised Buddhism of B R Ambedkar, though he is not an icon they invoke, because their constituency is not the dalits. The attempt to build a Buddhist coalition founders, however, on the fact that many janajatis have long and deep commitments to some form of sectarian Hinduism, some have their own shamanic traditions, and many others have converted to Christianity.

Statistics on representation in high positions should not be interpreted to mean that all janajatis are equally deprived. Among most minority groups there were elite families holding land or other resources and with connections to the state or even the palace, though more so among some ethnic groups than others. Some groups, notably the newars and thakalis, have done well as traders and businessmen.

On February 10, 2002, the ministry of law, justice, and parliamentary affairs published an official list of 59 janajati groups (unlike in India, it is not called a schedule).2 For those on the list, there will be reserved seats in political, administrative, and educational institutions. In 1990 a new national organisation, the Nepal Janajati Adivasi Mahasangh or Nepal Association of Indigenous Nationalities or NEFIN, was founded (though the term “indigenous” was added only later). In the beginning there were just seven member organisations. Today, of the 59 officially recognised groups, nearly all now have an organisation which is their official representative with one vote in NEFIN.

A few groups had an old (i e, pre-1990) national organisation. The oldest was the Tharus’ Kalyankarini Sabha, founded in 1949 with antecedents as far back as 1922. In those days it was a landlord-dominated caste-reform and sanskritising organisation, much influenced by caste associations in India. It has metamorphosed into a modern indigenous pressure group.3 Most groups have formed a representative organisation only after 1990, and in the case of many smaller groups this has been directly encouraged and assisted by NEFIN.

Under pressure from NEFIN, the government founded a unit called the Nepal Foundation for Indigenous Nationalities (or NFDIN), which has the task of uplifting janajatis. As such it operates under the ministry of local development. Both NEFIN and NFDIN have been successful in attracting large grants from the British Department for International Development (DFID). In attempting to deal with the question how reservations should be distributed between different janajatis, given that some are far more educationally and economically advantaged than others (newars and thakalis in particular being two of the best-off groups in the country), NEFIN came up with the ingenious

Economic and Political Weekly May 19, 2007 five-fold classification illustrated in Table 2. The idea is that the reserved “seat” should only be released for one category if no candidates from more disadvantaged categories of janajati come forward. Thus Newar and Thakali candidates would only qualify for reserved seats in medicine at the Maharajganj Teaching Hospital, for instance, if there were insufficient Gurung, Magar, etc, candidates to fill the available places.

The Communist Party of Nepal (Maoist) [CPN(M)] launched its “People’s War” on February 12, 1996, with attacks on police stations in the districts of Rolpa, Rukum and Sindhuli, and on a agricultural development bank in Gorkha district. Rolpa and Rukum are adjacent districts in the west of the country, and the home of the kham magars.

The Nepalese Maoists had long been aware of ethnic difference and factored it into their strategic calculations, much as Indian Naxalites have sought to establish bases in tribal areas. It was no coincidence that the CPN(M) made the kham magar regions of Rolpa and Rukum districts their base area. These areas had been a target of leftist proselytisation from the 1950s onwards and the village of Thabang was known as a communist stronghold already in the panchayat days. As chairman Prachanda explained to leftist journalist Li Onesto in 1999, “And in western Nepal there are the Mongolian ethnic groups – you saw how all our comrades there look Chinese. These nationalities are so sincere and such brave fighters – historically they have had this kind of culture. And upper caste chauvinism and feudal ties do not prevail in these nationalities.” The magars are, in short, exemplary “primitive communists” who would enable the Maoists to bring communism to the rest of Nepal [cf Lecomte-Tilouine 2004].

In April 2002 the Maoists established a magar-dominated “special district” (‘bishes jilla’) in their heartland in west Nepal. This led the way for the later declaration of

Table 2: Classification of 59 Official Janajatis by NEFIN and NFDIN (2004)

Note: There are now only five janajati groups without their organisation holding membership of NEFIN: Bankariya, Hayu, Kusunda, Raute, and Phree. Figures and percentages (which are the percentage of the total Nepali population of 22,736,934) come courtesy of Sant Bahadur Gurung, director of NFDIN until 2006; these follow the 2001 Census. Where there is no percentage this means that the group was not a category recognised by the 2001 Census; thus the figures are either unknown or are based on an estimate (note that in some cases, e g, Yolmo, the ethnic label is not widely recognised or accepted in the region, and therefore the figures are much lower than might be expected: thus while only 579 people returned ‘Yolmo’ as their ethnicity, 3,986 recorded that they spoke ‘Yolmo’).

Sources: Adapted from NEFIN (2004) and nefin.org.np.

Economic and Political Weekly May 19, 2007

autonomous regions in January and February 2004. The entire country was divided into nine “autonomous” regions, which on paper were the first level of government below the national. Six of the nine were named on an ethnic basis, and the unspoken assumption was that the leader of that region and the majority of the representatives in the regional government must come from the ethnic group so named, though no explicit statement to this effect is to be found in the Maoist rule book.

The prominence which the Maoists have given to the ethnic issue has raised the question among commentators whether the Maoists genuinely seek to address the grievances of ethnic activists, or whether they played the ethnic card because they knew they could outflank the political parties on this issue. They were well aware that the dominance of bahuns in all walks of public life did not lessen, and in some cases actually intensified, after parliamentary multiparty democracy was reintroduced in 1990. The current unrest and ethnic conflict in the Nepalese Tarai would perhaps have been inevitable at some stage or other, but the role of the Maoists in giving ethnic militancy a voice there has certainly been considerable.

Mobilising for Power

With the collapse of king Gyanendra’s attempt to reinstate autocratic monarchical rule in April 2006 following the “second people’s movement”, a new, and revolutionary situation has emerged. In November 2006 the Seven-Party Alliance and the Maoists signed an agreement which would lead to an interim constitution, an interim parliament, and an interim government to oversee the elections to a constituent assembly (CA). The arguments since have been about the degree to which janajati demands for proportional representation in the CA should be met. At the same time, armed clashes of madhesis opposed to the Maoists, or split from the Maoists, have intensified with the Maoists and with others in the Tarai. In the eastern Tarai ethnic cleansing has been going on, with parbatiya households and civil servants being chased out. Elections to the CA, which were forcefully announced for June, have had to be postponed because the country simply is not ready: constituencies, voters’ lists, the exact method of ensuring proportionality

– none of this had been decided.

How a newly decentralised state or a Nepal of federal republics would work is the matter of vociferous debate. (It is interesting in this connection that the Maoists have been very quiet about their autonomous ethnic republics since around August 2006.) There are “homelands”, i e, districts where the hill minorities form a majority, or where madhesis form a majority, but there is also a very large degree of intermixture with people from many different groups living together. In only 15 out of the 75 districts of Nepal is there a single group that makes up more than 50 per cent of the population.

There are Tamangs, for example, to be found virtually all over the country, as well as in India. Normally migration means language loss, in two or more generations, which is why all Nepalis in Darjeeling speak Nepali and why Darjeeling was one of the crucibles of Nepali literature and Nepali national consciousness, even though the majority of Darjeelingites are descended from migrants whose mother tongue was one of the minority languages. Despite this melting pot effect, one consequence of the Indian reservations system is that in recent years tribal and caste groups that long ago lost any cultural distinctiveness are now seeking to re-establish it, sometimes by visiting their “homeland” within Nepal in order to recover their cultural traditions [Shneiderman and Turin 2006].

In one sense all these groups have lived together, if not in harmony, at least without ethnic war breaking out for the vast majority of the time. But the inequalities of the caste system, the privileged access to the state and its benefits that accrued to the parbatiyas, mean that there has long been simmering discontent among janajatis, madhesis, and dalits that the dominant bahuns and chetris are hardly aware of and are ill-equipped to understand or deal with.

The madhesis are particularly bitter about the way they have been treated as a potentially disloyal fifth column within Nepal for so many years, about the facile way in which Nepalese nationalism has been built on the symbols of hill culture and antagonism to India, thus excluding them from full participation in the nation. The bureaucracy and the army are dominated by parbatiyas, and madhesis feel disrespected by both. For many years the Nepali Congress has treated the Tarai as a vote bank without ever offering proportionate leadership positions to madhesis. During the 1990s the government set up commissions to deal with the inequalities faced by women, by janajatis, and by dalits, but no recognition was given to the exclusions faced by madhesis. One madhesi slogan which targeted this cultural inequality, demanded that both the ‘dhoti’ and the ‘topi’ (the hillman’s cap) should receive equal respect.

The recent violence in the Tarai has been concentrated mainly in the eastern Tarai, where madhesis are concentrated. In the western Tarai there are more tharus and fewer long-established areas of settlement. The Maoists recognised these historical and cultural differences when they designated the western Tarai as the autonomous republic of Tharuwan and the eastern Tarai as Madhes. Part of the violence has pitted armed madhesi groups (they are charged with being supplied and recruited in part from over the border, as well as with being supported by revanchists in the palace) against the Maoists, who are identified by many in the Tarai with the hill people. The bottom line is that this is a time of political transition and of a virtually collapsed state. Everything is being renegotiated and the people of the Tarai finally decided that they must act now and at last be heard at the centre.

In all this one group – the most divided, excluded, and disadvantaged of all, i e, the dalits – is in danger of being forgotten. On all criteria, the dalits come out considerably worse off than any other group, though noticeably the Muslims come out on the human development indices as pretty similar. Among the dalits, those in the Tarai are the poorest and most deprived of all.

Nepal faces two main dangers. One is of a return to civil war between the Maoists on the one side and the army on the other, a cowed and frightened middle class being willing to go along with an army-backed regime in return for peace and a semblance of order. The other is of all-out ethnic war breaking out in the Tarai between madhesis and parbatiyas. In most of Nepal there are so many complex and cross-cutting allegiances that a Sri Lankan-type polarisation is unlikely. In the eastern Tarai, however, with its 30 per cent population of parbatiyas, there is a very real possibility that two “majorities with minority complexes” could confront each other in bloody vendettas. The politicians in Kathmandu seem to take an age to decide and agree on anything, but they have pulled back from the brink before, and it is to be hoped that they continue to pull off compromises and solutions, while keeping the peace process moving forwards, over the next couple of years.

EPW

Email: david.gellner@anthro.ox.ac.uk

Economic and Political Weekly May 19, 2007

Notes

Whelpton.]

References