From presidential to parliamentary state? A midterm look at Sri Lanka’s Constitutional Reform Process

By Dr. Asanga Welikala, 31 March 2015
Sri Lankan President Maithripala Sirisena (L) addresses members of parliament at Parliament House in Colombo. (Photo: AFP/President's Office)
Sri Lankan President Maithripala Sirisena (L) addresses members of parliament at Parliament House in Colombo. (Photo: AFP/President's Office)

The on-going Sri Lankan constitutional and governance reforms process has attempted to radically cut back the powers of its executive presidency, and establish independent institutions to depoliticise key state services. After considering a bicephalous model of shared power between the President and the Prime Minister, the process seems now headed in a direction that may likely result – subject to judicial approval and parliamentary support for the constitutional amendment – in a mainly parliamentary executive, although the President will continue to be directly elected.


The Political Dynamics of Sri Lanka’s Constitutional Reform Process

Sri Lanka has undertaken a constitutional and governance reform process since January 2015, when Maithripala Sirisena unexpectedly defeated Mahinda Rajapaksa in the presidential election. Sirisena served in the Rajapaksa government as a minister for over nine years, just before announcing his candidacy in November 2014. President Rajapaksa, who had presided over an authoritarian, corrupt, and nepotistic administration, had abolished the two-term limit on election to the presidency in 2010, through the Eighteenth Amendment to the 1978 Constitution. The 2015 presidential election was therefore the first time that an incumbent was seeking a third term. The Eighteenth Amendment had also removed various restraints on presidential power introduced previously by the Seventeenth Amendment of 2001, further distorting the constitutional imbalances in the separation of powers in Sri Lanka’s executive-dominated constitutional system and political culture.

The broad opposition coalition that supported Sirisena therefore was concerned not merely with evicting the Rajapaksa family from power, but also with introducing reforms to restore constitutional democracy and good governance. The electorate clearly agreed with this vision. Rajapaksa, a divisive ethnonationalist, had enjoyed wide popularity amongst the majority Sinhala-Buddhist community for winning the long-running war against Tamil separatists in 2009 and for his populist style more generally. On the back of the war victory, he had won the 2010 presidential election by nearly 2 million votes. Even against the diminishing returns of incumbency, that he lost the 2015 election at all is testament to the revulsion that the excesses of his regime had provoked, even within his core constituency.

In calling for a presidential election almost two years before it was constitutionally due in November 2016, Rajapaksa was following his customary strategy of destabilising the opposition with a relentless succession of elections. On this occasion, however, what had been a weak and divided opposition, and a severely repressed civil society, were able to build a broad coalition against Rajapaksa. Only five years after ending three decades of ethnic conflict between the Sinhala-Buddhist majority and the Tamil minority, an inclusive opposition alliance seemed improbable. Yet, the National Democratic Front (NDF) managed to bring together, on the one hand, the Jathika Hela Urumaya (JHU), a small but influential party of Sinhala-Buddhist nationalists, and on the other, the Tamil National Alliance (TNA), the conglomeration of Tamil nationalist parties. And it included everything in between; from the main opposition United National Party (UNP) and a sizable section of defectors from Rajapaksa’s own Sri Lanka Freedom Party (SLFP) led by Sirisena himself, to smaller minority parties representing the Muslims and Upcountry Tamils. Liberal civil society, which had done much to uphold Sri Lanka’s democratic traditions of tolerance, pluralism, and constitutionalism in the face of the Rajapaksa assault and the failures of the political opposition, were also key supporters of the Sirisena candidacy.

Sirisena’s promise of a 100-day reform programme to commence immediately upon his election was articulated in this political context. The programme was originally scheduled for 10 January – 23 April 2015, with Parliament to be dissolved after the enactment of the reforms. Four germane consequences flowed from this for the policy content offered by the coalition, which have had significant implications for the way the reform process itself has unfolded after the election. First, the implausibly short timeline to produce, review and enact quite fundamental reforms suggested a necessarily closed and elitist process. The logic seems to have been that the ends justify the means, and that a rapid reform process followed by a parliamentary election to both validate the reforms as well as consolidate the new government would be the best method to prevent a regrouping of anti-reform forces.

Second, the formation of the NDF entailed an implicit agreement to avoid devolution and minority issues altogether, on the basis not only that consensus within the coalition would be impossible on these matters, but also to denude Rajapaksa of the opportunity to portray it as anti-Sinhala. The TNA and other ethnic minority parties, whose constituencies voted overwhelmingly for Sirisena, seem to have calculated that defeating Rajapaksa was imperative to realising any of their power-sharing demands in the future. Minority issues therefore formed no part of the 100-day programme.

Third, Sirisena’s manifesto in general and the 100-day programme in particular, even though they addressed specific governance problems that had arisen during the Rajapaksa era and before, reflected a lack of coherence not only in the choice of reforms but also in their prioritisation and sequence. The extremely loose drafting of documents and glaring inconsistencies between versions evince this where, for example, the Sinhala version of the manifesto promised to ‘change’ the executive presidential system, whereas the English version mentioned ‘abolition.’

Fourth, the most serious weakness which has predictably returned to haunt the process was the absence of clear agreement between coalition partners as to the precise nature of the keystone reform promise: what to do with the executive presidency. As noted, there was an ambiguity as to what the NDF manifesto exactly promised and this was reflected in the campaign rhetoric of the various parties. While the UNP, the SLFP defectors, and the minority parties – in other words, the large majority within the opposition coalition – campaigned for a return to a parliamentary executive, the JHU consistently argued against outright abolition, preferring instead the piecemeal removal of some of the powers, privileges, and immunities of the office. The JHU’s support for presidentialism in principle derives from its commitment to the preservation of the unitary state, and the centralisation of power (and concomitant hostility to devolution and power-sharing with the minorities) that is one of the key postulates of its majoritarian nationalist ideology. In this view, a strong President from the majority community is needed to protect the primacy of the Sinhala-Buddhists within the state. 

More complexity was added by the NDF’s insistence that it would only undertake constitutional changes that would not require a referendum but would only require a two-thirds majority in Parliament. The intention seems to be to prevent anti-reform forces from instantiating former President Rajapaksa as the figurehead of a ‘no’ campaign in a referendum. While there are moves to bring him back to frontline politics, it would be more difficult to focus opposition to the reforms around a single figurehead in a fissiparous parliamentary election with myriad candidates and issues.  

The JHU has only three Members of Parliament and its objections could presumably be overruled within the coalition, if the majority within the NDF wants to abolish rather than reform the presidency. The bigger problem, however, is that the current SLFP majority in the Parliament elected in 2010, holds over from the Rajapaksa administration. Sirisena won the presidency with the support of the UNP, and has therefore appointed his Prime Minister and Cabinet predominantly from that party. The UNP had to necessarily form a minority government. This was justified on the grounds that the minority government was formed solely for the purposes of fulfilling Sirisena’s reform mandate, and that it would only function for the duration of the 100-day programme, after which there would be fresh elections. In any case, constitutional reforms need cross-party support for the two-thirds requirement to be met, but the assumption that the SLFP would fall in line when Sirisena won the presidential election has proved mistaken. Even though Rajapaksa has been ousted from the leadership of the SLFP, Sirisena as the new leader has not been able to convince his parliamentary party to unequivocally support the abolition of the executive presidency.

To overcome this hurdle, the ‘Troika’ of President Sirisena, Prime Minister Wickremesinghe (UNP) and former President Kumaratunga (SLFP) have offered two concessions. First, they have given ministerial appointments to a number of SLFP MPs in the hope of co-opting them and their supporters to the reform programme. It remains to be seen if this strategy succeeds, given that many within the SLFP continue to be Rajapaksa loyalists and the campaign to bring the former President back into active politics continues apace. Second, they have offered to enact electoral reforms demanded by the SLFP in exchange for parliamentary support for the reforms to the executive. The details of this are still unclear, but according to present indications, any electoral reforms to be agreed will be enacted in the form of a constitutional amendment only after the presidential reforms are passed. But more broadly, whether these two strategic concessions are sufficient to stabilise the government and the reforms process remains to be seen, given the notoriously undisciplined nature of Sri Lankan political parties.


The Irish Model v. the French Model? The Substantive Alternatives for Reform

Sri Lanka’s Second Republican Constitution of 1978 is often described as the Gaullist experiment in Asia. This is true in some respects: the strong presidency was intended to provide for stable government against the vagaries of shifting parliamentary majorities. In other respects, such as the institutional and functional separation of powers, and in the normative and cultural justifications for presidentialism, the Sri Lankan Second Republic is very far from the French Fifth Republic. In particular, the deep traditions of democratic republicanism that characterise French constitutional modernity are much weaker in the Sri Lankan case. Thus the republican and secular normative safeguards that act as counterweights to the autocratic potential of the monarque ŕepublicaine in French political culture are substituted in Sri Lanka with the mytho-historical model of Sinhala-Buddhist kingship, a tribute-extracting overlord of an hierarchical polity, epitomised in the Rajapaksa presidency. 

The reform process seeks to address the problems arising from this form of presidentialism by reducing the scope for unilateral action by the President, first, by reducing or removing substantive powers from the office, and second, by introducing procedural requirements on the exercise of presidential powers, especially over key appointments. While the latter set of reforms enjoys wide support, it is the issue of substantive powers and therefore the nature of the executive itself that has created deadlock.

In an initial Concept Paper of February 2015, the government outlined a scheme whereby the President would remain the Head of State and of the Executive. Yet, except in the appointment of the Prime Minister, he would exercise his powers ‘always on the advice of the Prime Minister’, who would now be the Head of the Government. While the President would have the right to ask the Prime Minister to reconsider his advice, he would be obliged to follow it if the same advice is re-tendered. This formulation, widely used in Commonwealth countries that follow the Westminster system, implies that the President for all intents and purposes occupies a ceremonial office, and real executive power in forming, directing and controlling the government lies with the Prime Minister.

The Concept Paper departed from the orthodox Westminster model only in retaining the direct election of the President. This was because the mode of election of the President is considered to be entrenched under the 1978 Constitution, and any attempt to change that would likely necessitate a referendum that the government is intent on avoiding. Although there are key differences (such as in the method of nomination of candidates), this scheme therefore resembles the Irish model of parliamentary government where the ceremonial Head of State is directly elected, rather than the more common method of being appointed or indirectly elected by the legislature in other Westminster countries such as India where there is no constitutional monarch. In the Irish case, moreover, constitutional conventions have developed that reinforce the ceremonial character of the presidency notwithstanding direct election. It is perhaps the hope of Sri Lankan constitutional drafters that the nature of the presidency will evolve in a similar direction. But given the past experience of strong presidentialism, a political culture that incentivises charismatic leadership, the ethnicised nature of electoral politics, and the survival of certain provisions from the 1978 Constitution that potentially allows the President to claim ‘sovereign’ powers, there is also the possibility that some future holder of the office may try to expand its constitutional role.   

A preliminary draft of the constitutional amendment that emerged in early March followed the scheme set out in the Concept Paper, which in turn caused an uproar within the government with the JHU leading the criticism on the ground that this amounted, in effect, not to a reform but to an abolition of presidentialism. In response to this, when the government gazetted the official Nineteenth Amendment to the Constitution Bill on 13th March, the scheme was substantially changed. The key phrase that the President should ‘always act on the advice of the Prime Minister’ was removed, and the President was given a more substantive role. Overall, the Bill envisaged a bicephalous executive where executive powers are shared and must be exercised in cooperation between the President and the Prime Minister and his Cabinet, in a manner that closely resembled the French model.

While the Bill still provided substantial powers to the Prime Minister in terms of government formation, composition, and dismissal, in essence, it retained the hybrid presidential form of government. It maintained the basic character of the 1978 Constitution as a semi-presidential system, in which there is both a directly elected President and a Cabinet of Ministers responsible to Parliament. But it redefined the institutional shape of executive power by reducing the substantive scope of presidential powers without removing the President’s executive functions entirely, and by strengthening the position of the Prime Minister. This key change would alter the system of government from a ‘president-parliamentarist’ form of semi-presidentialism (where the Cabinet is responsible to both the President and Parliament) to one that is ‘premier-presidential’ (where the Cabinet is collectively responsible solely to Parliament). The Bill was laid before Parliament by the Prime Minister on 24th March.   

However, the political pendulum swung back in the days after the publication of the Nineteenth Amendment Bill. As noted, except for the JHU, every other party within the NDF favour abolition in the terms outlined in the Concept Paper. The government is now going to move amendments to the Bill at committee-stage in Parliament that would restore the elements removed on account of the JHU’s opposition; including the requirement that the President acts on the Prime Minister’s advice and that the latter is the Head of the Government.

Ideally these disagreements ought to have been resolved before the Bill was gazetted, so that what is already an extremely opaque and confusing process would not be further complicated by changing and re-changing the provisions of the published Bill. Yet this convoluted process only illustrates the precarious balance of competing forces within and outside the government that the Troika is trying to achieve, with a laudable underlying commitment to delivering the promise of abolishing presidentialism for which the country voted in January.   

Other significant changes that are proposed include the reduction of the terms of both the President and the Parliament to five years from the current six years, the reintroduction of the two-term limit on presidential terms, and a constitutional limitation on the number of Cabinet and other Ministers. Moreover, Parliament’s term is fixed at five years and it may only be dissolved by a two-thirds majority after the expiry of four and a half years. Therefore the President cannot dissolve Parliament at will as before, and neither can the Prime Minister advice dissolution. This means that if a government falls within a parliamentary term, then an alternative government must be formed without recourse to a general election. This may provide stability, but it could also lead to deadlock, especially where the parliamentary majority is hostile to the President, although notionally this situation should not arise given that the President is now expected to be largely a titular figure.

Yet the fact of direct election and the President’s role as the embodiment of the sovereignty of the people in respect of the executive, adds both democratic legitimacy as well as an undefined source of residual power to the office. This assumes major importance in interpreting the limitations on the President’s powers and functions that are contemplated in the Bill. The President may potentially be able to draw upon his legitimacy and his ‘sovereign’ powers to settle any ambiguity, or conflict with other actors such as the Prime Minister or Parliament, in his favour in relation to those duties – such as to ‘ensure respect for the Constitution’, or ‘preserve religious and ethnic harmony’ or ‘promote national reconciliation and integration’ – where his wishes are in conflict with the advice of his Prime Minister. These are widely worded provisions, which are prone to serious interpretative disputes. On the other hand, it is not clear how the President will enforce his directions and orders in the discharge of his constitutional duties if he does not receive the co-operation of Ministers, and this may in practice force him to co-operate with the Cabinet.    

In relation to the Prime Minister and the Cabinet, the reform proposals effectively codify Westminster conventions. It must be added, however, that this amounts to a meaningful constitutional change only if the phrase ‘on the advice of the Prime Minister’ is taken to mean, as it is in other Commonwealth countries, that the President always follows the Prime Minister’s advice except in respect of those limited and mostly ceremonial matters in which he is constitutionally empowered to exercise his own discretion. If in the working of the constitution the President asserts any discretion to reject the Prime Minister’s advice on occasion, then the entire scheme would be defeated. In the Sri Lankan experience, this is not an inconceivable possibility. The office of Prime Minister under the 1978 Constitution has been generally powerless; but if the Nineteenth Amendment is passed, then he will become the most important executive player in the country.

Ultimately, all institutional reforms are only as good as the political will to implement them properly, and this is where Sri Lanka’s problems of political culture become critical. In particular, the pervasive ethnicisation of politics has distorted the promise of constitutional democracy from the very beginning of universal franchise in 1931. As the current calls for the return of Rajapaksa underscore, the ethnonational challenge to civic reformism has not lost all its potency after the 2015 presidential election. 


The Independent Institutions and Good Governance

The principal manner in which the Nineteenth Amendment Bill proposes the de-politicisation of certain key state services is through the establishment of a nonpartisan Constitutional Council comprising political and civil society members. The Council would recommend appointments by the President to the independent commissions that oversee elections, the public service, the judiciary, the police, human rights, bribery and corruption, public procurement and audit. The Bill makes elaborate provision for the composition, powers, and functions of each of these commissions, which would enhance the independence and impartiality of the officials and bodies they oversee. The President would have no discretion to reject the recommendations of the Council. The Council would also have to approve presidential appointments to certain senior positions; including that of the judges of the superior courts, the Attorney General, the Auditor General, and the Inspector General of Police. There is virtually no disagreement among political parties or in the country at large about the desirability of these changes, and if implemented well, they will have a far-reaching impact on improving the quality of governance. Nonetheless, many of these public bodies will take time to recover from the assault on their independence, dignity and professionalism suffered under the previous regime.     


The Pre-Enactment Challenges before the Supreme Court

The 1978 Constitution permits the pre-enactment constitutional review of legislation by application to the Supreme Court. Anyone can challenge parliamentary bills for constitutionality, but in relation to bills expressly classified as constitutional amendments, the Supreme Court is only empowered to determine whether the impugned bill requires to be approved by referendum, in addition to the two-thirds majority in Parliament that all constitutional amendments must secure. The court decides this by reference to the provisions that are expressly entrenched by the constitution, or if in its opinion the proposed amendment affects any impliedly entrenched provision of the constitution, such as the principle of the separation of powers.

While the provisions on the presidential election are not expressly mentioned in the category of entrenched provisions in the constitution, it is widely held by legal experts to be impliedly entrenched as an aspect of the principle of the ‘sovereignty of the people’, which is entrenched. Consequently, changing the mode of election of the President would require a referendum, and this is apparently why the Nineteenth Amendment Bill leaves that untouched.

As expected, a number of pro-Rajapaksa lawyers and politicians have challenged the Nineteenth Amendment Bill. The Supreme Court is scheduled to hear the petitions as well as the government’s arguments against a referendum along with the proposed committee-stage changes to the Bill on 1 – 2 April. Doubtless the government’s stance is supported, among others, by the official views of the Attorney General, but given the extent of the proposed changes and the centrality of executive presidentialism to the structure of government, it is at least debatable whether the Nineteenth Amendment Bill would require a referendum.

This will be the first opportunity for the Supreme Court to demonstrate the extent to which it has – under an untainted new Chief Justice – regained the independence it had lost during the Rajapaksa regime. The Rajapaksa government had illegally impeached a previous Chief Justice for halting legislation concerning one of their major projects by the exercise of this very constitutional jurisdiction, and replaced her with a sycophantic legal advisor. While it would be unwise to speculate on impending judicial proceedings, the petitions do not appear to make a particularly strong or well-founded case in favour of a referendum. For instance, one alleges that the constitutionalisation of the freedom of information in the Nineteenth Amendment Bill adversely impacts national security, and in this way, the principle of the unitary state. Another claim is that even the establishment of the Constitutional Council would affect the power of the presidency in a way that would require a referendum. Unless better arguments are canvassed at the hearing, it would be surprising if any court would uphold such bizarre arguments.


Going Forward

The situation on the ground in Sri Lanka remains fluid and fast-changing. The process has been less transparent and more disorganised than ought to have been the case, and the political class has demonstrated yet again its inability or unwillingness to transcend politics-as-usual in realising the desire for fundamental reform registered by the electorate in the last presidential election. On the other hand, the Troika, and especially President Sirisena, has demonstrated a remarkable commitment to delivering on his mandate, and this has made for a resolute perseverance in trying to ensure the success of the reforms.

In terms of the legal process, the government must persuade the Supreme Court that a referendum is not necessary. This is not a foregone conclusion given the ambitious nature of the reforms. The spectre of a Rajapaksa comeback is heightened if the government has to fight a referendum campaign that it has tried so hard to avoid.

It is undeniable that Rajapaksa continues to enjoy considerable popularity and his nationalist message may have resonance, especially in the rural hinterland. Without controlling a mass-based political party like the SLFP, however, he and his supporters are unlikely to make a significant electoral impact. Rolling back the popular reforms is even more inconceivable.  

If the Nineteenth Amendment does become law, at least institutionally, it would replace the presidential authoritarianism of the unreformed constitution with a more balanced system of government. The reformed constitution would also provide for a more robust framework of good governance through the Constitutional Council and the independent commissions. Much will depend on the political landscape following the parliamentary elections to be analysed in the next part of ConstitutionNet’s Sri Lanka series.



Dr Asanga Welikala is ESRC Teaching Fellow in Public Law at Edinburgh Law School and Associate Director of the Edinburgh Centre for Constitutional Law. He is also a Senior Researcher of the Centre for Policy Alternatives (CPA), Sri Lanka.

Disclaimer: The views expressed in Voices from the Field contributions are the author's own and do not necessarily reflect International IDEA’s positions.


Post new comment