Federalism became integral to Nepal’s political discourse soon after the success of the Second People’s Movement of April 2006. The idea of federalism in Nepal is, however, not that new. The Tarai Congress, a Madhes-based regional political party, first floated it in the early 1950s with the advent of democracy following the overthrow of the Rana oligarchy. The call for federalism was revived in 1990 after the downfall of the partyless Panchayat political system and the restoration of democracy. Once again, the call came from a Madhes-based political party, the Nepal Sadbhavana Party (Nepal Goodwill Party), but the Janajati-oriented Rastriya Janamukti Party (National People’s Liberation Party) and other Janajati (indigenous peoples) groups also raised similar demands. While these efforts failed to capture the political imagination, the 1990 Constitution did at least recognise the country’s ethnic and linguistic diversity.

The current debate on federalism gained momentum from 1996 with the Maoist insurgency. The rebels’ political mobilisation centred on the promise of ethnic self-determination, regional autonomy and an end to discrimination based on caste, ethnicity, language and religion. In 2004, at the height of the insurgency, the Maoists actually declared the division of the country into nine ‘autonomous regions’. Seven of these were named after the ethnicity or languages of the largest local communities and the Maoists even formed ‘Autonomous People’s Governments’.

But none of the Maoist documents during the insurgency period refers explicitly to federalism, indicating that they had not given it much thought as an organising principle of the state. Nor was federalism even mentioned in the core documents of the peace process, such as the Comprehensive Peace Accord 2006 (CPA) and the first iteration of the Interim Constitution 2007.

The decision to adopt federalism as the way ahead resulted from Madhes concerns. The Madhesi Janadhikar Forum (Madhesi People’s Rights Forum), at the time a rights-based NGO working for the Madhesi community, protested against the Interim Constitution’s failure to spell out federalism or to grant the Tarai proportional representation in the proposed Constituent Assembly (CA). The resulting Madhes Movement (along with a movement by the Janajatis) ultimately forced the state to accept federalism as the basis of state restructuring, and to revise the constituencies in the Tarai as per the population ratio [see article on social movements, p.97].

The Fifth Amendment to the Interim Constitution in July 2008 further clarified the modality of the federal restructuring of the state, as follows:

Accepting the aspirations of indigenous ethnic groups and the people of the backward and other regions, and the people of Madhes, for autonomous provinces, Nepal shall be a Federal Democratic Republic. The provinces shall be autonomous with full rights. The Constituent Assembly shall determine the number, boundary, names and structures of the autonomous provinces and the distribution of powers and resources, while maintaining the sovereignty, unity and integrity of Nepal. (Article 138–1A)

Political parties, Janajati groups, civil society organisations, and even individual citizens all floated different models of federalism, ranging from extremist ethno-regional, to much milder approaches. The Unified Communist Party of Nepal-Maoist (UCPN-M) and Janajati activists demanded the formation of provinces along ethnic lines – to establish some sort of ‘ethnic homelands’ based on their historic territories and with the provision of agradhikar (the right of the titular ethnic community to head the government in the respective provinces). Others argued that the provinces should reflect the mixed settlement patterns of the population and Nepal’s economic and development reality.

There were three major challenges to incorporating federalism in Nepal’s new constitution. First, there were no clear regional blocks for political and administrative purposes. Second, a multi-level division of power and distribution of jurisdictional authority had to be created. And third, the new federal structure needed to maintain coherence with the broader principles and values of a democratic polity, namely that everybody has the same rights

When the Committee on Restructuring the State   and Distribution of State Power (State Restructuring Committee) of CA I prepared its preliminary draft on federalism, it had no problem dealing with principles.   It unanimously approved the concept of a federal Nepal that underlined ‘identity’ and ‘viability’ as the two main foundations for carving out provinces. ‘Identity’ was defined as having five components: ethnicity, language, culture, contiguous geography and territorial alignment of the ethnic population, and historical continuity. ‘Viability’ consisted of economic interrelationships, status of infrastructure development, availability of natural resources, and administrative convenience. The Committee agreed on a three-tier structure of government – federal, provincial and local – and also listed their respective powers and responsibilities. Other modalities of federal polity such as division of power, inter-provincial relationship and dispute resolution procedures, were agreed as well.

However, the parties were strongly divided over the provinces’ names and boundaries. The State Restructuring Committee’s report, adopted by majority vote, provided for 14 provinces based largely on the ethnic criteria. Even a group with a population of one per cent of the national total was linked to a potential province while groups that would not be able feasibly to administer their own province were to be given autonomous regions with independent jurisdiction for self-governance. The committee also proposed special preferential rights to the principal ethnic group to head the government in its respective province and autonomous region. The 14-province model was heavily criticised as propagating ethnic federalism in Nepal and was rejected by the more traditional political parties, the Nepali Congress (NC) and the Communist Party of Nepal–Unified Marxist-Leninist (UML).

The High Level State Restructuring Commission (HLSRC) provided for by the Interim Constitution was formed very late, long after the State Restructuring Committee had submitted its report. The HLSRC members were selected along party lines and they remained divided accordingly. The nine-member commission consisted of two each from the major political parties, UCPN-M, NC, UML and the Madhesi Front, while the chair was selected by consensus. In terms of social background, HLSRC comprised three from the Khas Arya group (‘upper caste’ Hindu communities with origins in the hills), three Janajatis, two Madhesis and one Dalit (‘low caste’).

This composition left the three Khas Arya members, representing the NC and the UML (the second UML nominee was a Madhesi), in a minority. As a result, the HLSRC could not produce a consensus report and its recommendations were also highly controversial. Indeed, the minority group from within the HLSRC submitted a separate report. The majority report had proposed 10 provinces to be based on identity and one non-territorial province for the Dalit community, to be spread across all 75 Nepali districts. The majority report reflected the UCPN-M and the Madhesi parties’ positions, while the minority report mirrored the NC’s and the UML’s. Although the majority report had dropped the idea of preferential rights in heading the provincial governments, the boundaries and names of the proposed provinces along with the idea of a non-territorial one were still much disputed. This sparked prolonged unrest in the far-western region and a nation-wide protest by the ‘upper caste’ Bahun, Chhetri, and Thakuri, while Madhesi and Janajati groups also held counter-protests in favour of identity-based provinces as recommended by the State Restructuring Committee and HLSRC.

The leaders of the major political parties met several times to resolve these disputes. Neither the CA Committee’s 14-province model nor the HLSRC’s 10-province version was acceptable to all the parties. At the same time, a caucus of Janajatis within the CA had been exerting pressure on the party leaders to adopt one or the other. Towards the end of the first CA’s tenure, the three big parties – UCPN-M, NC and UML – agreed on an 11-province model with the boundaries to be decided on the recommendations of a future government commission. But the insistence on a complete provincial design by Madhesi and Janajati representatives remained the major obstacle and CA I came to an end without delivering a constitution.

In the second Constituent Assembly, elected in November 2013, the NC emerged as the largest party followed by the UML, with the UCPN-M a distant third and the Madhesi parties further behind. The first CA had seen strong ethnic caucuses, even if only informally, whereas the second CA did not allow caucuses at all. The changed political equation had a visible impact on constitutional negotiation although the differences on federal design continued to persist.

Responding to the devastation caused by the earthquake in April 2015, the major political parties, including UCPN-M, felt the urgency to end the uncertainty in constitution- making in order to facilitate reconstruction. The NC, UML, UCPN-M and a faction of the Madhesi Front reached a 16-Point Agreement in June that settled the differences on the major contents of the constitution. It agreed that there would be eight provinces, but with the boundaries to be settled later through a commission.

Following public consultations, which were more of a formality than a substantive and systematic exercise, and a Supreme Court decision that the constitution would have to demarcate the provinces prior to adoption, the party leaders came up with a six-province model. But, following protests from the most ‘backward’ (disadvantaged) Karnali region and a few districts from the Mid-western Development Region, they revised it to seven provinces. The Madhesi Janadhikar Forum-Democratic, which had been party to the 16-Point Agreement, walked out on that decision and boycotted the CA process thereafter, including the adoption of the constitution.

Despite protests by Madhes-based parties and a few Janajati groups, the constitution was passed by an overwhelming majority of the CA. The seven provinces keep almost all the current 75 districts intact, while the naming of the provinces has been devolved to the provincial assemblies. It has adopted a three-tier state structure – federal, provincial and local – with residual power kept at the centre. In general, the distribution of powers is tilted in favour of a strong centre and the polity may thus be termed a form of centralised federalism. However, accepting a vital principle of an advanced federal system, the constitution has provided for a double majority while amending the constitution on matters affecting provincial boundaries and powers – ie a simple majority in the respective provincial assembly and a two-thirds majority in the federal parliament.

When it was promulgated on 20 September 2015, the constitution received divided support. The Madhesi Front rejected it principally on three grounds: 1) on the demarcation of provinces; 2) on constituency delineation; and 3) on the issue of inclusion.

The demarcation divided the 20 districts of the Tarai among five provinces. The eastern Tarai districts were placed in Province 1, which is dominated by the eastern hill districts. Likewise, Province 3, principally of the hills of the central region, included one Tarai district. Two districts were joined with the hill districts of the far-west region that made up Province 7, while Province 5 coupled Tarai and hill districts from the mid-western and western regions. Only Province 2 consisted exclusively of Tarai districts [see special feature mapping the evolution of federal design, p.80].

The Madhesi parties viewed such a division as perpetuating the domination of the Tarai by the hill populations, and therefore not meeting the aspirations   of the Madhesi people. The Madhesi Front had been demanding that none of the Tarai districts should be placed in hill provinces. For public consumption they had been calling for one single province to be made up of all the 20 districts of the Tarai, but were in fact prepared to negotiate the number.

Similarly, the constitution favoured geography over population in terms of electoral constituencies. This raised fears among the Madhesi Front that the representation of Madhesis in the federal parliament would not be proportionate to their population strength, given that the Tarai accounts for only 17 per cent of Nepal’s territory but is home to half the population. As for inclusion, the Front took issue with the fact that the new constitution had backtracked even from the provisions in the Interim Constitution.

Federalism remains the most contested post-war political issue in Nepal. Opinion has been divided between ‘ethnic’ versus ‘non-ethnic’ federalism, with the split largely mirroring progressive and conservative blocs – although, as explained below, there were significant differences within these. Where Madhesis and Janajatis saw ethnic federalism as the answer to all their woes through their ‘right to self-determination’, the dominant Khas Arya saw it as a threat. The Khas Arya are scattered across the country and the prevailing (non-ethnic) district boundaries allowed them a sizeable presence in most provinces. The status quo effectively enabled them to retain their hold over the power structure of the provinces. The Khas Arya preference, therefore, was for fewer provinces, keeping the districts intact, and preferably following a north-south geographic alignment. They pointed to the development potential of larger provinces and also to the reduced cost of governance as their rationale.

Although not the first to do so, it was the UCPN-M that forcefully propagated the ethnicity-based approach to province formation, a position it held throughout the CA deliberations. But it had failed to grasp the fundamental concepts of federalism and was attracted more by the popular appeal of such a concept among certain communities. As communists, the Maoists were manifestly opposed to pluralism, but they also claimed to be the most ardent advocates of federalism. In Nepal, federalism equates with diversity, and so without a commitment to pluralism federalism has no substantive meaning. Meanwhile Madhes-centric parties proposed one Madhes province for the entire Tarai region, from east to west, and were not really concerned with whatever happened in the hills and the mountains.

The more traditional NC and UML stood on the opposite side. The NC sought a small number of provinces (six or seven) to be based on geographic features and demography, not ethnicity. The party could just as easily have argued the case for federalism, the principles of which align well with its democratic foundations. But it never encouraged its cadres and supporters in this direction, despite the electoral potential of upholding the interests of its traditional Madhesi support base. The UML was divided, but the dominant leadership opted for the NC approach. Although initially appearing sympathetic to the identity concerns of hill Janjatis, the senior UML leadership ultimately united to oppose any consideration of ethnicity in the federalism debate. Right-wing parties either demanded that the issue be referred to a national referendum or aligned with the NC.

The reluctant embrace of federalism by the NC and the UML can be explained on a number of grounds. First, the idea of federalism was completely new for both since, despite sporadic calls, the issue had never been seriously considered in the context of Nepal before the war. Even after the Interim Constitution was amended to that effect, neither party could think beyond regional devolution of powers such as reorganising the existing five development regions. Neither party held any internal debate on why federalism had been demanded so strongly or on how its objective conditions could help consolidate nationalism by engendering a wider sense of belonging. Second, the leadership of both parties has always been from the dominant hill Bahun and Chhetri communities and they believed that carving out provinces with identity as the principal criteria would come at a political cost of losing support from their own communities. And, third, these leaders took the very mixed settlement patterns, particularly in the hills, as a strength that negated identity as the basis for provinces.

The international community has provided technical and other expertise to support the constitutional process in Nepal. But, encouraged partly by Nepali ethnic activists in Kathmandu, it has tended to see Nepali society as deeply divided, and federalism as the panacea. In fact, Nepali society has never experienced high levels of ethnic conflict. Grievances have been focused on state policy and behaviour, not against any community per se. This failure to distinguish between diversity and division played into internal political schisms in Nepal, and NC and UML leaders were able to blame the international community for provoking ethnic tensions. Moreover, the international community in Kathmandu did not have a uniform view. Influential donors such as Denmark, Norway and the United Kingdom did not prioritise federalism, and this enabled the NC and the UML to adopt a more conservative stance.

Geopolitics has been influential in the recent developments in federalism and both India and China have had a much more visible impact. Given its sensitivity on Tibet, China had always warned against ethnic provinces and the constitution’s seven-province model seems satisfactory to China. India, on the other hand, has had strong interests on federalism in relation to the Tarai, as well as on other aspects of state restructuring such as preferring a parliamentary form of government. India has not outlined its position with regard to the provinces, only stressing that all stakeholders should be taken into consideration and openly supporting the Madhesi Front and its demands on that score. In January 2016, the Nepali Government amended the constitution to make some changes on issues of constituency delimitation and inclusion, following which India seemed relatively satisfied and the blockade that had been imposed along the border in support of the demands of the Madhesi Front immediately after the promulgation of the new constitution was lifted.

Much could have been done at the beginning of the peace process to set the pace of the transition towards federalism. The CPA and the Interim Constitution called for restructuring of the state in order to end inequality and discrimination existent in the Nepali polity. Had the State Restructuring Commission been formed soon after the promulgation of the Interim Constitution, it would have steered the federalism debate along a more coherent and professional basis while also allowing the CA to focus on the framework of federalism.

Many Nepalis have also seen the federal project as zero-sum. Janajati and Madhesi activists felt that anything less than achieving their full aspirations would represent total failure. On the other side, the conservative Khas Arya political elite and intellectuals believed that federalism would undermine national unity and interests if the identity aspirations of Janajatis and Madhesis were accepted.

Federalism has represented the key means by which to change the power structure of the Nepali polity. The first amendment to the new constitution has gone some way to address Madhesi and Janajati concerns over inclusion and representation, but not provincial demarcation. The second amendment proposal brought by the government in November 2016 consists of three major points relating to federalism: 1) revision of the boundary of Province 5, ie, to detach the six Tarai districts from the hills; 2) revision of the basis of representation in the Upper House so that that 35 of the 56 seats would be divided among the provinces in proportion to population, instead of the existing provision for equal representation; and 3) listing national languages and official languages as per the recommendations of the National Language Commission.

The proposed amendment seems the most that can be expected in the context of the present political equation in the legislature, and however minimal it may look from the perspective of the Madhesi Front, it may just satisfy the Madhesis to begin with. But much depends on where it goes next. Nepal cannot afford the newly declared constitution to be a failure or marred by mutually defeating positions. Reviewing and reorganising provinces is a natural course for any emerging federal polity, and if that is what it takes to bring on board all Nepalis, then the principal architects of this constitution, the NC, the UML and the UCPN-M, have no real choice but to follow that path.