The process of Nepal’s restructuring from a unitary to a federal state is creating new urgencies. The crux of the matter is how the sub-national units of federation —call them provinces—should be formed; should they be along territorial lines, ethnic lines or a combination thereof. Looking at the proposals and counter-proposals put forward by the political parties, it is clear they have failed to grasp the gravity of the problem and are devoid of a vision. Their solution will lead to a deeper rift tomorrow. The challenge is to steer through the intense and perplexing divides on ethnic issues and unite the country with a framework that would uphold democracy, and a road to economic prosperity, inclusiveness, and equality of opportunities.

The aspirations that people have shown for their rightful recognition is commendable; such alertness and agility is what make a nation strong. Unfortunately, peoples’ grand vision—desire to see all identities as part of nation building—has been high-jacked by destructive debate on which group was in Nepal from time immemorial, what is the chronological order of settlement and why one should have more rights than others. The implications of these two mindsets cannot be starker. State restructuring that reflects people’s desire would make all Nepalese proud citizens of a just society whereas state restructuring that justifies irresponsible activism would lead to irreversible course of action creating fault lines that benefit no one in the long run.

Nepali people have shown remarkable humility in the past; they might have to show it again. The sacrifice required of them is to be mindful of one’s unique beam of identity while recognizing the rainbow of diversities in the neighborhood.  They must rise above what is defined by caste, language, religion, region and family and commit themselves as one people of a singular nation.

The mosaic of different ethnic, linguistic and caste (ELC) groups in Nepal and its economic gravitas dictate against ethnic federalism. The rainbow of population settlement defies all the principles required for ethnic federalism. The way forward is to combine ethnicity with capability as pillars of federation.

Principles for ethnic federalism: learning from the world

Broadly defined, there are two bases for federation: (1) territorial which considers geography and economy and (2) ethnicity which considers ethnic, linguistic or religious criteria (“ethnic”). Both types of federation are territorial but in the former the territory is neutral, whereas in the latter, it is recognized as a “homeland” of a single group. The rationales for the two types of systems are different. The territorial one is associated with individual rights, whereas the ethnic one is related to self-determination with right to autonomy for a group. In a territorial federation, there are equal rights for all; personal characteristics (language, religion, culture) are not that relevant. In an ethnic federation, however, persons of one characteristic enjoy benefits that others do not.

With different kind of objectives they produce different kinds of federation. Everything else—the size of a province, the number of provinces, the division of powers across different levels of government and the judiciary system—depends on which of the two bases is taken. Hence, the basis of federation is a key factor in shaping the types of institutions.

These are two extreme forms. Not all federations can be fitted neatly into on or the other category. It is possible to mix territory and ethnicity. Belgium, Canada, India, Malaysia and Switzerland have elements of both. Analyzing these countries’ settlement, some stylized facts—I call them principles—emerge which gave rationale for mixing ethnicity and territory as the bases of federation. I describe them below.

First, people of one ethnicity are in overwhelming majority in a province—principle of democracy. For example, in Switzerland, there are 26 cantons (provinces) and each of them has overwhelming majority of people that practice one religion or speak one language. All cantons have unilingual people except only four that are multi-lingual. The other example is Belgium; the north province is almost all Dutch-speaking and the south one is French-speaking. In India also, each state (province) has majority of one language group: in Karnataka, 65% speaks Kannada; in Gujarat, 75% speak Gujarati and so on.

Second, if a geographic region populated by people of the same ethnicity is too large in terms of providing services, then it is divided into more than one adjoining province—principle of economic efficiency. In Switzerland, few German-speaking cantons are adjacent; so are French-speaking ones. Same is in Canada. If adjoining land of common ethnicity would make one province, all English-speaking Canada would have only two provinces instead of eight.

Third, the provinces are formed in such a way that the majority of the target population resides within the province(s) where it is in majority—principle of inclusion. As a result, people of the same ethnicity do not have the feeling of “in” or “out” of its focus province.

Fourth, each major population group has its defined province(s)—principle of equality.

Fifth, if people of the same identity are not settled in an adjoining region, then their representation in accommodated using electoral process—principle of flexibility. In Belgium, to accommodate Brussels where majority (85%) is French speaking and is inside the Dutch-speaking north, the provincial parliaments are supplemented by community parliaments of Dutch-speaking and French-speaking people.

Sixth, people of different identities have the same rights in all aspects, except that the language (or religion) of the dominant group could be sole or one of the official languages (religions) in the province—principle of pragmatism. For example, in a French-speaking Swiss canton, all non-francophone will have the same rights as that of the francophone.

Reality check: a different report card

Next, I evaluate which of these principles will be fulfilled in Nepal. For that I take the proposal put forward by High Level State Restructuring Commission (HLSRC). The assessment of this report provides a valid test for the feasibility and rationality of ethnic federation, as it distributes villages and districts in such a way that the share of target ethnic group in its “homeland” province’s population is maximized. Therefore, if the principles are not met by the distribution made in this report, no alternative distribution would. The verdict will also be applicable to proposals of the Constituent Assembly Committee on Restructuring and Distribution of Power (CRCSD), United Communist Party of Nepal (Maoist), and 12 provinces version of United Marxists Leninist Party. All these proposals are based on the same logic as of HLSRC.

The HLSRC has proposed 11, including one non-territorial for Dalits, provinces. Using information given in this report about which villages and districts are allocated under which provinces, I calculate the share of 10 major population groups in each of 10 proposed provinces in Table 1.  In this 10 × 10 matrix, the first column is the names of the proposed provinces and the first row is the names of ELC groups. The provinces’ names and respective target groups’ names are placed as mirror image, except for Narayani and Karnali (with no target groups) and hill Brahmin and Chettri (BC) and Dalit (with no “homeland”), which I put respectively in the mirror-image. Although the HLSRC report combines Tharu with Madhesi, I report Tharu as a separate group. As Tharus are mainly in LAT, I put them correspondingly.

Table 1. Composition of ELC groups in proposed provinces (percent)

* MBKM stands for Mithila-Bhojpur-Koch-Madhesh and LAT stands for Lumbini-Awadh-Tharuwan
** Remaining 3.4% of Nepal’s population is from other groups outside of these 10 ELC groups.
+ Dalits include both hill and Terai Dalits. Hence, “Madhesi” does not include Terai Dalit.

By construct, the diagonal entries (in bold) are the shares of the corresponding ELC group in the province that is named after its ethnicity or regional identity. The off-diagonal entries in each row show the shares of other groups in that province’s population. The first diagonal entry (of 26.1%) corresponding to “Limbuan” province and “Limbu” means that only 26% of the population in that province—about one in four—is Limbu. Among 74% non-Limbu, 15.6% are Rai, 27.5% are hill BC, 6.5% are Dalit and so on (reading through first row). Not only are the diagonal entries small, for each province, many ELC groups have non-zero entries, meaning that almost all groups reside in that province, a rainbow settlement. If one was looking similar Swiss table, the diagonal entries would have been very high and the off-diagonal almost zero.

Just by eyeballing the diagonal entries, we see that except in two Terai provinces (for LAT one needs to add Madhesi and Tharu), the share of “homeland” group is in minority. In all six provinces with ethnicity (Limbuwan, Kirat, Tamsaling, Newa Tamuwan and Magarat) the share of corresponding target ethnic group is at most one-third, ranging as low as one-quarter. In most cases, they are not even the largest fraction in their “homeland”; hill BC are larger in number than the respective target group. In Terai, the majority of “homeland” group is the artifact of highly aggregate denominator “Madhesi”. In reality it is composed of several heterogeneous groups (in terms of social, linguistic and cultural aspects). Had we applied more disaggregation either by language (Maithali, Bhojpuri and Koch), or by ethnicity (Janajatis) or religion, there would not have been majority of any particular group. So, forming provinces in ethnic line violates even a simple majority rule—the principle of democracy.

Furthermore, in case of MBKM having 30.5% of the population (Table 2) is not desirable on ground of economic efficiency. It is simply too big. Other provinces also violate the second principle, that of economic efficiency, but from another side, by being too small.

Table 2. Share of area and population in proposed provinces (percent)

In Table 3, I examine the distribution of each ELC group across provinces. For each ELC group (reading column-wise), the diagonal entry is the share of its total population in Nepal that resides in its “homeland” province and the off-diagonal entries are shares of this group that resides in “others homeland”. For example, the number of Limbus that resides in “homeland” province “Limbuwan” makes up 68% of total Limbu population in Nepal. The remaining 32% of Limbus would be outside their “homeland” province (including 26.9% in MBKM). Except Limbu and Tamang, for other four ethnic groups, majority of them reside outside “homeland”. So the ethnic federalism would also violate the third principle, the inclusiveness. A quick scan of Table 3 shows that each ethnic group will be scattered throughout all provinces as a small group and will be excluded minorities in “others homeland”. Again, for countries like Switzerland and India, the diagonal entries would have been much larger and the off-diagonal negligible or zeros.

Table 3. Distribution of an ELC group population by proposed provinces

MBKM stands for Mithila-Bhojpur-Koch-Madhesh and LAT stands for Lumbini-Awadh-Tharuwan

If we impose prior-rights, meaning that only a member of the “homeland” group has the right to be the leader of the provincial government, then the exclusion is even more daunting. There will be rules by “lucky” minorities who happens to be in their “homeland” by coincidence over “unlucky” majorities who reside in “others” provinces. Take the case of Newa. Newars will rule this province although Newars are only 34% of the province population (episode 1 of anti-democracy). Only 46.7% of Newar live in Newa and the remaining majority (52.3%) has to accept the rule of another group depending on where they reside (episode 2 of anti-democracy). Similarly, Gurungs will rule in Tamuwan even if they are only 30% of the province’s population and since 74% of Gurungs live outside Tamuwan, other people’s rule will be imposed on them.

Finally, ignoring the two largest sections of the population, 30.9% of hill BC and 11.8% of Dalit (with total of 42.7%), which is what ethnic federalism boils down to, in unthinkable. It violates every single democratic principle that one can think of; it’s against the principle of equality under the constitution.

Amicable solution: combining beam and rainbow

A more prudent, farsighted and reasonable approach is needed in deciding the bases of sub-national units of federation so that it can address the aspiration, culture and dream of all citizens. The fact that four principles would be violated in ethnic federalism and the consideration of the remaining two principles— flexibility and pragmatism—led me to propose that ethnicity should be combined with territorial consideration (Acharya, 2007). I still think what I proposed then is the best option: combining individual’s beam of identity with group’s rainbow of identity. I provide a heuristic sketch of the proposal below.

Nepal will be a federation of four provinces—extending north-south—and one territory (with more central government role than in others). Each province will have three ELC regions one each from mountain, hill and Terai. The most eastern province, Mechi, combines Limbuwan, Kirant and the far eastern Terai region mainly of Koch language area (Jhapa, Morang and Sunsari). Next, Koshi province combines Tamsaling, Newa and Maithali speaking Madhes (Saptari, Siraha, Dhanusa, Mahottari and Sarlahi districts). The third province, Gandak, includes Tamuwan, Magarat and Bhojpuri speaking Madhes (Rautahat, Bara, Parsa and Chitwan districts). The far west province, Karnali, includes Awadhi speaking Madhes, Tharuwan and western hill where the concentration of Dalits is very high. Finally, Rara territory includes Dolpa, Jumla, Humla and Mugu districts.

With this structure in mind, I propose a three-tier government: local, provincial and national. Both at the provincial and the national levels, I propose bicameral parliaments. The bicameral system at the national level will be a method of combining the principle of democratic equality with the principle of federalism—all citizens are equal in the lower house, while all provinces (and ELC regions) are equal in the upper house. Similarly, at the provincial level, the lower house will maintain the equality of individuals at the lower house and equality of ELC regions at the upper house.

The representation of different ELC groups at the upper houses of both provincial and national levels would guarantee they have equal sharing of political power—the main intended outcome of making Nepal an ethnic federation. This mechanism can be supplemented by educational policies, social policies and cultural policies that are conducive for all ELC groups. Dalit, Muslim and other marginalized groups should also have the proportionate representation as other ELC groups in both upper houses.

This will be a win-win situation—lifting the boat of downtrodden while avoiding the negative consequences that federalism based on only ethnicity would generate. It also provides the best possible economic structure with North-South provinces that are both reasonably large in size and house people of all groups.

Conclusion

Nepal is a house of intermingled communities; but while we might be different in sociological and anthropological sense, we share common desires and aspirations. Each of us wants the freedom to determine own destiny, express own views and participate in the decisions that shape our lives. Let us not place extra burden to the society by trying to make one identity superior to other. After all, identity is not a weapon of the weak; it is something to be proud of and respectful for others’. In history great people have dedicated their lives for the betterment of humanity, irrespective of race, gender, geography and ethnicity. Even if we cannot match their deeds, can we just stop disrespecting them?

(The writer can be reached at: This e-mail address is being protected from spambots. You need JavaScript enabled to view it )