Before the dust settled: is Nepal’s 2015 settlement a seismic constitution?

ABSTRACT Two significant institutional developments occurred in the aftermath of the major earthquakes that struck Nepal in 2015: a new national constitution was drafted and promulgated and a National Reconstruction Authority was established. The constitution had been promised for over seven years, and was now completed within just over three months, while it took seven months for a Bill establishing the NRA to be passed in parliament. Many commentators have posited a direct causal relationship between the earthquake and the ‘fast-tracking’ of what was in certain respects a contentious constitution. Drawing upon conversations and interviews conducted in Nepal over the winter of 2017–18 and a close reading of media discourse and political analysis from 2015, this article will examine and assess the extent of this supposed causality. Given that the most radical and contentious change ushered in by the new constitution was the introduction of a federal structure for the state, particular attention will be paid to the evolution of the debate on this issue.


Introduction 1
What effect does a major disaster have on the politics of the society that suffers its impacts? More particularly, if a state is undergoing a process of political change or transition when a disaster strikes, will that disaster necessarily affect the direction, speed or general inflection of that process? If we think for a moment of a political transition, or indeed of history more generally, as a text that is in the course of being written, we might ask whether and in what circumstances a disaster will serve as a full stop, bringing a grammatical end to a particular line of political discourse or institutional development. Alternatively, and especially with the benefit of hindsight, we might also wonder whether a disaster that appeared at first sight to be bringing an end to something might actually turn out to represent something more akin to a comma or a semi-colon: a mere pause, a temporary distraction, after which the process of change and the discourse that accompanies it carry on much as before.
Some years ago, Disaster Studies specialists began to express concern that insufficient attention was being paid to such questions. Albala-Bertrand, for example, wrote that the 'endogenous response' to disasters was being disregarded, feeding a 'technologically biased approach to disaster response, at the expense of political and socio-economic structure and dynamics'. 2 In their widely-cited article on disasters, critical junctures and 'tipping points', Mark Pelling and Kathleen Dill complain of the 'exceptionalism and apolitical construction of disasters', about which most knowledge is generated by 'practitioners, humanitarian agencies and donors' and 'coloured by agency viewpoints'. 3 They identify three 'discursive moments' in a typical aftermath. The first moment, they say, focuses attention on the unequal distribution of losses and can lead to a questioning of development failures and asymmetry in the social contract; the second draws attention to the mobilisation of state and non-state actors to champion, direct, counter or capture evolving critical discourses; the third sees the discourse being institutionalised into policy. They discuss the potential for a disaster to provide either a 'critical juncture' (a contestation of established political, economic and cultural power) or an 'accelerated status quo' (a successful concentration of that power -'a concentration or speeding up of pre-disaster trajectories which remain under the control of powerful elites . . . ') 4 Capoccia and Kelemen have described how the concept of critical junctures has been used to construct causal arguments in historical institutionalist literature, defining critical junctures as 'brief phases of institutional flux' which punctuate 'relatively long periods of path-dependent institutional stability and reproduction'. Critical junctures are characterized by a situation in which the structural (that is, economic, cultural, ideological, organizational) influences on political action are significantly relaxed for a relatively short period, with two main consequences: the range of plausible choices open to powerful political actors expands substantially and the consequences of their decisions for the outcome of interest are potentially much more momentous.' 5 Pelling and Dill believe that change initiated at the moment of a critical juncture is 'made concrete in a revision of the social contract, or at least in the balance of underlying institutions.' 6 Olson and Gawronski (2003) discussed the application of the term 'critical juncture' to the aftermaths of the 1972 Nicaragua and 1985 Mexico City earthquakes: Simply put, both Nicaragua and Mexico were "different countries" ten years after their disasters, and the underlying question kept coming up: What was the role of each country's disaster in the national transformation? Applying the critical juncture framework comparatively to the two events, their aftermaths, and regime change then brought up a second question: Is there a broader utility for critical juncture analysis in disaster research? That is, can and should the framework be further adapted and applied to other major disasters? Would such an effort be fruitful beyond merely better description? 7 While long term political effects may well be argued for in these and other cases, it is surely generally true that if a disaster possesses any causal force in the political sphere it is during its immediate aftermath that this will be most clearly apparent. As time passes and an aftermath lengthens, a 'time decay effect' 8 inevitably sets in: attempts to trace causal links between disasters and subsequent developments become increasingly speculative, and the risk of teleological historicism grows. This article will explore these questions with reference to the first six months of the aftermath of the earthquakes that struck central Nepal in April and May 2015, killing nearly 9000 people and damaging or destroying homes, schools, infrastructure and religious and cultural heritage sites across 31 of the country's 75 districts. To what extent does the 'critical juncture' framework help us to understand what happened in the immediate aftermath of this catastrophe? Do the empirical specificities, the 'better description' of this particular aftermath, accord with or challenge this model?
The 2015 Nepal earthquake aftermath It is not difficult to identify two key political moments in the immediate aftermath of the major earthquakes that struck Nepal in 2015: (1) The leaders of Nepal's four largest political parties signed an agreement on the basic principles of a new constitution for the country 45 days after the April earthquake, and the constitution that emerged from the drafting process that ensued was promulgated just over three months later, on 20 September. Thus, a constitution that had been promised but not delivered for seven years (despite agreement between the parties on many of its key features and principles) was eventually completed and promulgated within a timeframe of just over 100 days. (2) In contrast, the National Reconstruction Authority (NRA) came into being much more slowly. The President of Nepal approved an ordinance to establish this body three months before the constitution was promulgated, but the Bill to establish the NRA was not passed in parliament until nearly three months after the promulgation, and no staff (other than its CEO) were appointed to it until January 2016, seven months after the first earthquake. The slow progress of the reconstruction effort that year was parodied in many cartoons in the Nepali news media, including that shown in Figure 1.
The drafting of the new constitution took place in a context of resurgent post-earthquake patriotism and calls for national unity, particularly among Nepal's hill populations, most of whom celebrated its eventual promulgation. Other groups, notably most of the political leadership of the Madhesi population of the southern Tarai lowlands, considered the new constitution to be a regressive document. They therefore rebelled against it, with consequences that included over fifty deaths in clashes between security forces and protestors, and a grievous deterioration in Nepal's relationship with India, which was accused of mounting a six-month economic blockade in support of the Madhesi agitation. On 10 November 2017 a Nepali parody of the popular song 'Despacito' was posted on YouTube with the title 'Desh Sakkiyo' [The Country is Finished] and quickly went viral. This song conflates the three major events of 2015 (the earthquake, the new constitution, and the blockade) as follows: ho bhuinchaloko pida ajhaisamma ni yo manma thiyo sarkarle naya samvidhan po diyo bharatle nakabandi garidiyo Yes, the pain of the earthquake was still there in this mind, [When] the government gave a new constitution [And] India blockaded us. 9 When these developments are viewed in outlinefrom a distance, as it werethere would indeed seem to be a direct causal relationship between the 2015 earthquakes, the prioritisation and 'fast-tracking' of the constitution, the Madhesi uprising against it, and the 'blockade' (nakabandi), which lasted from shortly after promulgation until early February 2016. The relationship between the earthquake and the constitution, at least, has been presented in precisely these terms in many narrative accounts. 10 The following discussion draws primarily upon the Nepali media discourse of the period in question, 11 supplemented by close readings of published political analyses and conversations and interviews conducted in Nepal over the winter of 2017-18 and during subsequent visits in 2018 and 2019, to examine and assess the extent of the 2015 earthquakes' supposed political causality. Given that the most radical change ushered in by the new constitution was the introduction of a new federal structure for the Nepali state, particular attention will be paid to the evolution of the debate on this key issue.

The political context
The Interim Constitution and the first Constituent Assembly  The Comprehensive Peace Agreement (CPA), concluded in November 2006 at the end of the ten-year war between the then Communist Party of Nepal (Maoist) and the Nepali state, included commitments to draft and promulgate an interim constitution and then to elect a Constituent Assembly to draft a constitution that would permanently replace it. The CPA also committed its signatories to engage in a 'forward-looking restructuring of the state by resolving the prevailing problems related to class, ethnicity, regional and gender differences'.
The Preamble and Articles 33(d) and 138 of the Interim Constitution (2007) maintained this commitment to a 'progressive restructuring of the state'. However, the publication of a draft of this document provoked political protests in Nepal's southern Tarai districts in December 2006, and its promulgation in January 2007 was followed by mass civil unrest across the region, in what is now commonly referred to as the first madhes andolan [Madhes Movement]. This continued until the government conceded the movement's principal demands. Before proceeding any further, a brief excursion is necessary here to explain the historical bases of Madhesi discontent.
Madhesi is the term used for a heterogenous 33% of Nepal's population who are defined variously in terms of region (they reside in or hail from the Tarai lowlands, the Madhes); language (their mother tongues are cross-border languages such as Maithili, Bhojpuri or Awadhi, and Hindi serves as a lingua franca for many of them); and ethnicity (mostly in a negative sense: Madhesis are explicitly not Pahadis, hill people.) 12 Thus, Madhesi is not an ethnic or caste label: it might be most accurately described as a political or even an experiental term: 'the people living in a particular region (the Tarai or Madhes) who feel discriminated against by the Pahadi-dominated state' 13though even this definition excludes others from the same region such as the Tharus or the Tarai Muslims, who may feel the same way but hold themselves to be ethnically distinct.
The Anglo-Nepal war of 1814-16 was the starting point of the discrimination against the Madhesis by Nepal's Gorkhali rulers, and a prohibition on Madhesis entering the state's security forces was its legacy. Under the Rana family autocracy , the people of the Tarai had to obtain written permission to enter the Kathmandu Valley, and during the Panchayat regime  the Madhesis fell outside the frame of national identity encapsulated by the palace-sponsored nationalist slogan ek desh, ek bhesh, ek bhasa ('one country, one dress, one language'). Until 2006, the widespread absence of birth certificates and other documents to prove their Nepali origin made it almost impossible for many Madhesis to acquire citizenship, without which they could not own land, apply for government jobs, register births or marriages, obtain a passport, stand for elections, register a business, get bank loans, or access government benefits. For about thirty years from the 1950s onward the government encouraged hill people to migrate to the plains, partly in order to keep Madhesis, perceived by many in Kathmandu as sympathetic to India, in check. 14 Despite all of the above, academia and the media paid scant attention to Madhesi concerns until the 2006-7 andolan.
Madhesi political leaders argued that they had long been systematically underrepresented in Nepal's electoral system, and protested that the commitment to 'restructuring' made in the Interim Constitution was too vague. The amendment that was made to the constitution in April 2007 therefore made an explicit commitment to a federal structure, promised under-represented groups access to state bodies on a proportional basis, and re-delineated a number of Tarai constituencies to take account of population growth. The number of federal units, the basis on which they would be created, and the delineation of their boundaries would subsequently become highly contested matters. 15 Nepal conducted elections to a 601-seat Constituent Assembly (hereafter 'CA') in April 2008. The Unified Communist Party of Nepal-Maoist (UCPN-M) emerged as the largest single party, with 238 seats (39.6% of the total). The Nepali Congress (NC) and the Communist Party of Nepal (Unified Marxist-Leninistreferred to as 'the UML' hereafter), gained only 114 and 109 seats respectively, while three Madhes-based parties shared a total of 83 seats. Compared with the three parliaments of the 1990s, the representation of indigenous [Adibasi] Janajati groups rose from 25 to 36 per cent, of Madhesis from 12 to 24 per cent, of Dalits from practically nil to 8 per cent, and of women from 5 to 33 per cent. 16 The Khas-Arya who have historically constituted the dominant group achieved a representation of 33%, roughly in line with their share of the total population. All of this represented a dramatic and largely unexpected shift in the balance of power, and was reflected in the agenda of the first CA, which was dominated by debates on 'secularism, multilingualism, identity-based federalism, self-determination, collective rights for ethnic groups, preferential rights for marginalised groups, inclusive electoral design and proportional representation'. 17 According to Article 70 of the 2007 Interim Constitution, the primary means by which the Preamble and each article of the new constitution was to be approved was through a unanimous vote at a meeting of the CA attended by at least two-thirds of its membership. Without a numerical majority, the political forces that favoured radical structural change 18 the Maoists and the Madhes-based partieswere not able to prevail over the more conservative parties. There were also disagreements on the form of government (presidential vs parliamentary; bicameral vs unicameral); the judiciary (constitutional court vs Supreme Court); electoral systems, and so on. 19 Thus, despite four extensions to its term of office and the installation and dismissal of five different coalition governments, the first CA failed to produce a constitution and was dissolved in May 2012.
Its legacy was a 22-point agreement between the three main parties (UCPN-M, NC, UML) dated 15 May 2012 which sought compromises on key disputed issues. However, Madhesi organisations began agitating against this agreement even before it was finally ratified, and the UCPN-M backtracked from it quickly thereafter. This eventually led to the demise of the first CA, creating a state of constitutional and political stalemate.

The second Constituent Assembly (2014-15)
The second Constituent Assembly was elected in November 2013. The NC and the UML emerged from these elections as the two largest parties, the UCPN-M came a poor third, losing 154 of its seats to finish with only 84, and the Madhes-based parties won only 50 seats. The Maoists and Madhes-based parties boycotted the newly-elected CA at first, claiming that the elections had been rigged against them, but their complaints were given little credence.
The caste and ethnic composition of the second CA was also somewhat different from that of the first: for example, the historically dominant Khas-Arya increased their share of seats to 41 per cent. Perhaps more importantly, the three largest parties disallowed the formation of cross-party caucuses, such as the Janajati Caucus that had been so influential in the first CA. The NC and the UML maintained similar positions on most of the more contentious constitutional issues, and although their combined majority fell just short of the two-thirds needed to pass a constitution through a vote, they were confident of gaining enough support from smaller parties to obviate the need for any negotiations with the Maoists or the Madhes-based parties.
Sushil Koirala came to power on 22 January 2014 as the head of a new NC-UML coalition government, and promised that his government would complete the drafting of the constitution within one year. The coalition parties' leaders decided not to bother to seek a broad consensus, which would inevitably lead to some measure of compromise on issues of federal structure and inclusivity, but to proceed on the basis of majority voting. 20 Meanwhile, a Maoist-led alliance of thirty smaller parties (of which only nineteen were represented in the CA) advanced a common position on key constitutional issues, including an insistence on identity-based federal provinces, and urged that the constitution should be approved through consensus.
The NC and UML were hostile to the idea of identity-based federalism, which they saw as 'an external agenda, as weakening Nepali sovereignty, and as nationally divisive'. 21 They sought to minimise the number of provinces, and insisted that these would have to be based on 'geography, historical continuity and economic viability' and preferably carved out along north-south lines so that they included both hill and Tarai districts, which was precisely the opposite of what the Madhesbased parties were demanding. Although their predominance in the second CA meant that they did not need to take the Maoists' views into account, the NC and UML believed that the Maoists still had enough support to disrupt the process through street protests and agitations. They therefore decided to offer them a share in power after the constitution had been promulgated, but they also sought an arrangement that 'would not disturb the prevailing dominance of the upper-caste hill groups . . . in the political order'; 22 this dominance was caricatured in a famous cartoon by Rabin Sayami, reproduced here as Figure 2. The Madhes-based parties were not especially concerned about the total number of provinces, but they insisted that the new structure must include provinces that consisted solely of Tarai districts, entirely segregated from the hills.
As the deadline of 22 January 2015 loomed, tension between the two power blocs within the CA grew, with the Maoist-Madhesi coalition enforcing strikes in various locations. When the CA failed to deliver the constitution as promised, there was pressure on Sushil Koirala to resign. As the co-leader of the coalition government, the UML leader Khadga Prasad Sharma Oli wanted a turn as Prime Minister, but Sushil Koirala's publicly stated position was that he would not step aside for anyone until after the constitution had been promulgated. The main parties had settled many contentious issues, but the question of the federal structure remained unresolved.
Meanwhile, the Maoist-led 30-Party Alliance continued its agitation and declared a Nepal Band (nationwide shutdown) for 7-9 April 2015. However, this was suddenly and unexpectedly called off by the UCPN-M's leader Pushpa Kamal Dahal (a.k.a. 'Prachanda'), at the end of the first day. Dahal cited a minimal presence of activists on the streets and the danger of an adverse reaction to a strike, but analysts suspected other reasons, including a lack of support from China, whence Dahal had very recently returned, and the Maoist leadership's possible vulnerability to prosecution for war crimes under transitional justice legislation if they remained on the political margins. 23 Realising that their movement was not going to gather force and that the CA was likely to simply proceed without them, the Maoists adopted a more conciliatory approach. The NC and UML also conceded that consensus needed to be sought on certain points. Dahal and the UML leader Khadga Prasad Sharma Oli began to meet regularly from this point onward to work towards forging a working alliance, probably because they both knew that the NC-UML coalition would not survive long after the promulgation of the new constitution. 24 Seismic change?
The situation on the eve of 25 April earthquake was as follows. The NC-UML coalition was determined to get a constitution voted through the CA with a twothirds majority, and had vowed to achieve this by the end of June. 25 This would exclude the UCPN-M and most of the Madhes-based parties, mainly because it would introduce the bare bones of a federal structure that had a smaller number of provinces than the Maoists had argued for, none of which would be named or demarcated on the basis of minority identities, and a map of federal Nepal with provincial boundaries that paid little heed to Madhesi demands. Oli was anxious that this should happen as quickly as possible so that Koirala would retire from the premiership and open the way to him becoming Prime Minister (see Figure 3). For its part, the Maoist leadership feared that it was at risk of becoming marginal or even irrelevant. It urgently needed to find a way of compromising without losing face in order to re-enter the political mainstream, and was clearly prepared to offer concessions. 26 Many political commentators write of the earthquake of 25 April 2015 bringing the parties 'to the same place' (ekai thauma). 27 Amid calls for political and societal unity from almost every quarter, many op-eds in leading Nepali newspapers in the early weeks of the aftermath debated the merits of forming a 'national government'. Supporters of this idea argued that it would protect Nepal's national interests at a time when so many foreign powers were active within its borders, and compensate for the weakness of the sitting Prime Minister. A national government would be better at coordinating disaster management and reconstruction, building a new constitution, and organising long-awaited local elections. The fear (or alleged fear) that the lack of a constitution at a time of national disaster would undermine or even threaten Nepal's national sovereignty may have had some part to play in determining political priorities at national leadership level. Within two days of the 25 April earthquake, the BBC reported that rescue and relief teams had already been dispatched by India, China, Pakistan, Israel, Japan, the USA, and the UK. 28 After two weeks, it was reported that about 4900 military personnel from 34 different countries were engaged in relief operations in Nepal. 29 Although initially welcomed, the Indian intervention quickly became controversial, largely because of what was perceived as the insensitive behaviour of many Indian news reporters. Concerns were also expressed about Indian aircraft flying over sensitive border regions to the north. Although the Nepal government gratefully accepted pledges from the international community amounting to a total of 4.4 USD billion (see Figure 4 for a sardonic representation of this), it turned away relief aid in a number of instances. Its rejection of an offer of help from Taiwan is readily explicable in terms of Nepal's reluctance to offend China, but the decision not to allow the UK government to deploy Chinook helicopters into Nepali airspace led to the circulation of conspiracy theories concerning the legacy of the 1996-2006 'People's War'. 30 In an interview given to OnlineKhabar.com on 13 June 2015, the Maoist leader Pushpa Kamal Dahal made a clear reference to the dangers he believed Nepal's status as a sovereign country would face if its leaders took another eight years to agree a new constitution, saying 'there will no longer be any national party, no leader, the leaders' credibility will fall below zero, and the reputation and independence of the country will be in danger.' 31 A number of other politicians were consulted off the record during the course of this research, and I found general agreement among them that there was a fear that the earthquake might lead to a higher level of international interferencenot due to the lack of a constitution per se, but because the state was weak.
Opponents of a national government argued that the inter-party negotiations that would be a necessary preliminary to its establishment would waste time that would be better spent on relief work, and that a unitary government with no opposition would be like a return to the undemocratic Panchayat period. 32 In the event, a Nepali  constitution. According to the Agreement, the constitution would enshrine an eightprovince federal structure, based on five criteria of identity and four criteria of capability. 33 The provinces would be named by a two-thirds majority in each provincial assembly and their geographical demarcation would be recommended by a federal commission. The form of government had long been a point of contention, but now, instead of actively opposing the idea of a parliamentary system, the UCPN(M) simply registered a note of dissent (pharak mat) in order to allow the constitution writing to proceed. It was agreed that the constitution writing process would be taken forward in the spirit of the Agreement and that local bodies would be elected as soon as possible. 34 Informally, it was understood that Sushil Koirala would vacate the office of Prime Minister in favour of a UML leader after the promulgation of the constitution. 35 The only Madhes-based party to sign up to this Agreement was the MJF-D. As a consequence of the Maoists' and the MJF-D's signing of the Agreement, the Samyukta Loktantrik Madhesi Morcha, a united front of all of the Madhes-based parties, broke off its alliance with the Maoists, expelled the MJF-D, and rose up in opposition to the constitutiondrafting process. 36 Pushpa Kamal Dahal defended his party's decision to sign the 16-Point Agreement and participate in the fast-track process in his 13 June interview with Online Khabar. 37 He argued that the republic, federalism, secularism and the principles of proportional representation and inclusivity that were going to be enshrined in the new constitution were all achievements of the 'People's War' of which the Maoists should be proud, and he claimed credit for bringing the parties together 'at a time of crisis when the people are so greatly in need of relief.' He argued that if the Maoists sat 'outside' the process with the Adibasi Janajati and Madhes-based parties and simply left the NC and UML to make the constitution on their own, the people would be 'confused'. However, he warned that if the other parties reneged on the promise of identity-based provinces or tried to create any provinces containing both hill and Tarai districts, then the Maoists would launch a 'powerful movement' against them. This sense of having to settle for a less than perfect constitution quickly, because the political gains that had been made were at risk of being further eroded in the earthquake aftermath, was also very clear in a conversation this author had with the erstwhile Maoist leader and former Prime Minister Baburam Bhattarai in February 2018: Then at exactly that moment on 12 gate the earthquake struck. And attention was diverted elsewhere. Such a massive national calamity. My view was that at least the people should be sovereign. At least there should be a constitution . . . At least the people should be sovereign, even if the constitution is weak. Getting sovereignty for the people was a big achievement after the abolition of the monarchy in 2008. There was a danger of a reaction from the right after the earthquake. And a danger from within, because the leaderships of NC and UML were not so very committed to change. It was not their agenda. They only compromised with us because of circumstances, when they were marginalised after [King] Gyanendra's coup [in 2005] . . . As everything was against us and there was a risk of more loss, so [we thought] let us compromise as much as possible to make a constitution. 38 However, the political scientist Krishna Hachhethu believes that it would be naive to take these representations of the Maoists' decision at face value. His reading is that the Maoist senior leadership's main concern during the months leading up to the earthquake was how to regain the political space they had lost since the election of the second CA. They had already decided to give the NC and UML free rein in the drafting of the constitution because they believed that the alliance between these two parties would break once it had been promulgated, providing them with new leverage. Until the earthquake struck, the public perception of the Maoists' position on the constitution was that they were committed to identity-based federalism, and to continuing to lead the 30-party alliance in opposition to the other parties' attempts to dilute thisa somewhat hopeless cause, in the face of falling public support for their strikes and agitations. The earthquake gave the Maoists a 'good excuse' to change their position and compromise on the content of the constitution, on the understanding that it would thereby be promulgated speedily, leading to the break-up of the ruling coalition. 39

Drafting controversies
The 16-Point Agreement was widely welcomed in the mainstream Nepali media. For example, Suman Khadka stated that the earthquake would provide 'a new reference point' by replacing the 1996-2006 war as 'the most defining event in contemporary Nepal.' From the earthquake he saw re-emerging 'a Nepali identity', which he described as 'a conflicted term in recent times' (see Figure 5). 40 The Agreement was viewed very differently in Madhesi political circles, where it was seen as a ploy to renege on the commitments of the 2007 Interim Constitution and thus circumvent Madhesi demands, postpone federalism, and reinstate the dominance of the Khas-Arya establishment. 41 The Madhesi lawyer Dipendra Jha played a leading role in an attempt to have the Agreement declared unconstitutional. A writ petition to the Supreme Court pleaded that the decision to leave the names and boundaries of the federal provinces as matters which would be resolved after promulgation was in contravention of Articles 82 and 138 of the Interim Constitution. 42 The Supreme Court responded by issuing a stay order, directing the CA to include federal boundaries and power sharing provisions in the constitution at the point of promulgation. 43 In response, the 16-Point Agreement's signatory parties issued a defiant statement, asserting that the Supreme Court's stay order 'stands Figure 5. The constitution, shown as a means of national unity and reconciliation, and as the basis for Nepal's post-earthquake reconstruction. From newspaper op-eds published in the Governmentowned newspaper Gorkhapatra by Acharya (2015) and Aryal (2015). Reproduced with the permission of the Gorkhapatra Corporation.
against the sovereign right of the people to frame their Constitution through the CA.' The CA was 'an independent and capable body as far as the making of the Constitution is concerned' and its decision in this matter would be final.
So the drafting continued, in defiance of the Supreme Court order, and with virtually no input from the wider CA membership. A first draft was presented on 30 June by the NC's Krishna Prasad Sitaula, who stated that it would be promulgated by 31 July, 'provided there was no political earthquake'. 44 In July, the draft was made available for a public consultation exercise across all 240 electoral constituencies. This exercise lasted for just 15 days, with a mere two days allocated for public feedback, but over 150,000 suggestions were received from the constituencies none the less.
The senior journalist Kishor Nepal observed the process of gathering suggestions from the public in the mountainous Helambu district, and reported that most villagers were too busy working in their fields or reconstructing earthquake damage to contribute: Even leaders at village level knew the constitution was already complete, and that the collection of suggestions was just for the sake of delivering the process. However, setting the people to one side might have aroused challenges to the legality of the constitution, and the pretence (natak) of collecting suggestions was also necessary for international consumption. 45 Ujjwal Prasai provides a vivid account of the same process in the Siraha, Morang and Jhapa districts of the Tarai. Here the situation was very different, with boycott movements and security crackdowns in several locations, and a clear division of opinion between the Madhesi and Pahadi populations. When he asked a local activist to identify the demands of the Madhes that had not been addressed by the draft constitution, he received the following reply: 'You did not need to come all this way to understand that. Doesn't Kathmandu understand? It's clear. Create two provinces in the Madhes. Provide the proper level of representation in all government bodies, in proportion to the population. Consider the daughters in law who come to marry us, and our mothers who came before, as equal Nepali citizens. That's it (bas). ' 46 According to Khanal, the most common suggestions were 1) declaring Nepal a Hindu state; 2) delineation of the boundary of the provinces before the promulgation of the constitution; (3) simpler provision for citizenshipeither by the name of mother or father; 4) keeping the number of provinces to the minimum; etc. 47 There appears to have been very little concern about matters such as how the new provinces would be delineated or which cities should be made their capitals: 'when the constitution itself was confused (anyolma) on such matters, it was not hard to understand that those giving suggestions would be confused too'. 48 On the basis of the suggestions received, the four parties announced that they would consider seven main issues: directly elected executive head; rights of Dalits and women; citizenship; religion; electoral provisions for local bodies; demarcation of federal units; and the judiciary. 49 They came up with a new sixprovince model, and also added a clarificatory note on the definition of secularism.
Public protests against the six-province model erupted in the Karnali region and the mid-West, and a concession was granted in response to these: the mid-west development region would become a separate, seventh province. In stark contrast, the Tharuhat movement, launched at about the same time to demand a separate province for the Tharu people of the West Tarai, degenerated into violence and was suppressed with the use of lethal force. The MJF-D demanded a review of provincial boundaries and either a move to an eight-province model, including both Tharuhat and the Karnali province, or a reversion to the original six-province model. Once it became clear that the three other parties were not ready to accommodate this demand, the MJF-D, which had been one of the signatories to the original 16-Point Agreement, walked out and joined the Madhesi Morcha in boycotting the process thereafter.
Thus, no Madhes-based party was involved in the final stages of the adoption of the constitution. A plenary meeting of the CA attended by more than 90% of its members passed the Constitution Bill clause by clause between 13 and 16 September. Meanwhile, 'the entire southern plain (Kailali to Morang) was engulfed in violent protests'. 50 The Indian Foreign Secretary S. Jayashankar came to Kathmandu to try to persuade the political leadership to delay promulgation 'so that Madhesi parties could also be brought on board' but was rebuffed. 51 When the Constitution Bill as a whole was put to the vote, it was passed by an 85% majority: of the 598 members in the House, eight abstained, 25 voted against, and 57 boycotted the vote. Most of those who did not vote in favour were Madhesi members, meaning that 'a significant chunk of the population did not own [the constitution] when it was proclaimed'. 52

Criticisms of the new constitution
Julia Strasheim lists three factors that can limit the legitimacy of territorial reform: (1) the dominance of elite actors; (2) the effect of tight deadlines on these actors' decisionmaking quality; and (3) the risks posed by embedding territorial reforms into a wider 'concert' of institutional reforms that, as a whole, sparks fears of marginalisation. 53 All of these factors were clearly at play in the drafting of Nepal's 2015 constitution. The greatest challenge to its legitimacy arose from the fact that the federal structure did not locate the whole of the country's Madhesi population in purely Madhesi provinces, as per Madhesi demands. The twenty districts of the Tarai were distributed among five provinces: four of these also contained hill districts, and the only purely Tarai province was Province no. 2. In addition to this, although the 2015 Constitution retained many of the progressive aspects of the 2007 Interim Constitution, it was also seen to have either reneged on or diluted some of them, as follows: (i) Certain of its provisions for the acquisition of citizenship were widely perceived to discriminate against women, and against Madhesi women in particular, who are more likely to marry across the national border with India. Whereas the foreign wife of a Nepali man could begin the process of acquiring citizenship immediately, the foreign husband of a Nepali woman would have to wait for fifteen years for naturalisation, granted at the state's discretion. 54 Thus, the power of an individual to confer citizenship upon his or her spouse was unequal and dependent upon that individual's gender. Similarly, a Nepali woman married to a foreign man could only confer naturalised citizenship to her child, whereas a Nepali man married to a foreign woman could confer citizenship by descent 55 Naturalised citizens would not be eligible for appointments to positions of high public office at national or provincial level.
(ii) The new constitution reduced the proportion of the legislature elected through a PR system from 58% to 40% for both national and provincial legislatures and gave the population an unequal measure of representation at both provincial and national levels. Each province would send eight members to the upper house of parliament, regardless of significant differences in the total population of the provinces, and a change in the relative weighting of population and geography as factors determining the delineation of electoral constituencies raised the fear that the people of the Tarai would not be represented in proportion to their overall population. Moreover, whereas the Interim Constitution of 2007 allowed for the re-delineation of constituency boundaries every ten years, in the 2015 constitution this was extended to 20 years. This was likely to work against the interests of Tarai districts, where in-migration from the hills still continued. (iii) The Interim Constitution had specified that marginalised groups would have 'the right to participate in state structures on the basis of principles of proportional inclusion' (Article 21). The 2015 constitution maintained this principle, but omitted the word 'proportional' (samanupatik) in the relevant Article (Article 42 (1)) and extended the list of social categories to which it promised a greater measure of inclusivity to include the poorer members of the dominant upper caste grouping: 'economically poor (aarthikruple vipanna) Khas-Arya'. 56 (iv) The addition of a clarificatory note to the definition of secularism which used the term sanatan, with its whiff of Hindu conservatism, instead of the more neutral parapurva of the 2007 Interim Constitution, was widely seen as a concession to those who still conceived of Nepal as originally and essentially Hindu. 57 The Madhesi rebellion continued to gather force and attracted Indian support, leading to a blockading of the main overland supply routes into Kathmandu. A total of 58 people, including both civilians and security personnel, were killed in violence in the Tarai The earthquakes were quickly framed by both government and media as a national trauma and a cause for national mobilisation, although only 31 of the country's 75 districts had been directly affected. Much of the public discourse in the immediate aftermath focused on the need for political and social unity across the nation, as evidenced not only in political speeches, media editorials and news coverage, but also in popular culture. Harrowell and Özerdem record a number of comments made by their informants about the earthquake bringing renewed 'social harmony' in its immediate aftermath, 'when Nepalese society came together across political, social or ethnic lines to work together'. 59 Nepali poets and songwriters amplified this message, representing the earthquakes as a socially levelling and unifying force. 60 'Social harmony' (samajik sadbhavna) has a specific meaning in Nepali political discourse, and it is often used in opposition to the idea of 'inclusion' (samaveshikaran). For those who had long opposed the inclusion agenda, construing it as a threat to 'national unity' (rastriya ekta), the earthquake aftermath presented an opportunity. The national character of the post-disaster narrative favoured and empowered majoritarian conservative forces, which were particularly prominent among the hill populations. Moreover, the 2015 earthquakes did not strike areas where federalism was a burning issue, and the threat to the historical status quo represented by the post-2006 agenda of minority inclusion was already receding in the Nepali domestic political sphere, because its proponents were electorally weak. Thus, those who had advanced the inclusion agenda saw the ground slipping away from beneath their feet, and decided that it was necessary to secure whatever gains they could, before the dust settled and the rising hill nationalism of the aftermath swept even these away. Hence the Maoists' surprising flexibility, and, initially at first, the MJF-D's willingness to be a part of the process.
Some might argue that what has been described here is a classic instance of Pelling and Dill's 'accelerated status quo' in which the dominant use the chaos of a disaster aftermath to ram through their own agenda at the expense of the marginalised. To assess this claim, it is of course necessary to compare the post-earthquake and pre-earthquake status quos, but it is also important to ask exactly how these status quos are being defined. What was the status quo in Nepal prior to the 2015 earthquakes? One answer might be that it was the complicated, stalemated post-1990 struggle between classes, ethnicities, and regions that was most dramatically represented by the ten-year-long People's War, and then persisted in the form of a political deadlock in the CA. 61 For the purposes of the present discussion, however, I have taken it to be the much longer-standing, historical status quo of Khas-Arya hegemony which has been the backdrop for almost every field of political contestation in Nepal for generations.
Despite its dilution of the post-1990 inclusion agenda, the 2015 constitution does not represent a complete reinstatement of this Khas-Arya hegemony, or a complete closure by the political elites of avenues leading to adjustments of the political order. It has not denied all of the demands of marginalised groups: most significantly, it has granted federalism, albeit not in the shape and form that the Madhes-based parties had demanded. Equally, however, it could be said to be regressive in respect of matters such as proportionality of representation, gender and citizenship, its definition of secularism, and the introduction of a quota for a section of the historically dominant section of the population. Elements of the status quo ante of Khas-Arya dominance could be said to have been reinforced -'accelerated' in Pelling and Dill's formulationbut they now sit alongside other provisions that represent partial fulfilments of longstanding minority demands. These concede enough to reduce long-term support for continued oppositional activity, without providing much of a foundation for building provincial units that might challenge the federal state. So the constitution that was promulgated does not measure up very well to either side of the binary choice offered us by Pelling and Dill: oppositional forces were not able to use the opportunity presented by the chaotic aftermath of the earthquakes to radically alter the terms of the social contract, nor did the established elite use it as an opportunity to deny all of their demands.
Capoccia and Kelemen believe that change is not a necessary element of a critical juncture. They argue that there is no reason to discard cases in which change is considered but ultimately rejected as 'non-critical' junctures because ' . . . ignoring the near misses of history would actually deprive scholars of important and interesting negative cases with regard to the outcomes they seek to explain.' 62 So what effect did the 2015 earthquakes have on the constitutional process in Nepal? Returning to the punctuation metaphor with which I began, I would argue that the disaster was neither a full stop, nor a comma. It did not bring a complete end to any line of political discourse or institutional development: the constitution was probably coming, and come it certainly did. But the disaster was not merely a pause or a temporary distraction either, because the process of change and the discourse that accompanied it did not carry on into the aftermath exactly as before. Instead, I would suggest that if the earthquake was a critical junctureand this will always remain a matter of opinion and argument, not of objective, verifiable factits effect was to modify the terms of Nepal's internal political conversation. Rather like a lens bending the light that passes through it, it refracted the debate, strengthening some lines of argumentation and weakening others. Fears of threats to national sovereignty could be stirred up more readily in the earthquake aftermath, minority demands could more easily be framed as divisive. Moreover, contra Harrowell and Özerdem, the key political players within Nepal did not see the process of institutionalising the Comprehensive Peace Agreement simply as 'post-conflict reconstruction' (which may reflect more of a donor/INGO agenda), but also as an ongoing struggle for access to power and resources. The earthquake aftermath provided the section of the political elite that had the upper hand at that juncture with an opportunity to deliver a constitutional settlement that might not otherwise have been supported across the country, and oppositionists with an opportunity to extricate themselves from politically unprofitable positions in which they had hitherto been locked, in order to rejoin the political mainstream.