Photo by AFP courtesy Asia One

Amidst strong indications following the Uva provincial election result that a panicked regime is planning an early presidential election, there has been a vigorous debate in the media about the consequences of the Eighteenth Amendment to the Constitution, which abolished the presidential two-term limit and removed other procedural restraints on presidential power. There was of course no comparable opposition or even public interest in the issue when that amendment was enacted in 2010. In the triumphalist afterglow of war victory, this blatant attempt at entrenching presidential authoritarianism and dynastic consolidation was ignored by most and even welcomed by some. Only an infinitesimal minority in the parliamentary opposition, the media, and civil society opposed this pernicious measure then, only to be roundly condemned as traitors for their trouble. Still, better late than never, and it appears that the Southern electorate is at last waking up after the excesses of the long post-war party, albeit with a nasty hangover in the form of a constitutional and democratic crisis.

In recent days, there has been a specific legal dispute as to the legality of the incumbent president seeking a third term within the broader debates about the abolition of the term limit and the neutralisation of the independence of the judiciary, the public service, and law enforcement. It has been sparked by the innovative interpretation of the relevant constitutional provisions propounded by the former Chief Justice Sarath N. Silva, P.C. – and perhaps as importantly, the adoption of that position by the JVP as their policy stance – that both a third term for President Rajapaksa and the proposed early election are illegal and unconstitutional. There has been a plethora of responses to this by a number of lawyers and commentators supportive of the regime, of which perhaps the best articulated viewpoint has been expressed by Chinthaka Mendis, an attorney-at-law.

The Silva view has now been given major support and credence by the written legal opinion of Professor Suri Ratnapala commissioned by the Bar Association of Sri Lanka. Professor Ratnapala of Queensland University, who as a State Counsel in the Attorney General’s Department was involved in virtually all the key constitutional cases in the late 1970s and early 80s, is an internationally respected academic authority on constitutional law and political theory. His independent views therefore have massive persuasive value. Likewise, Dr Reeza Hameed, a long-standing member of the Civil Rights Movement, has published a legal opinion that is complementary in important respects to the views of Silva and Ratnapala elsewhere in these pages. Readers will recall his series of articles that added much clarity and principle during the impeachment of the 43rd Chief Justice.

Both the Silva and Mendis views have their strengths and weaknesses, which I will summarise and assess below. It is likely that more will be said by proponents of both sides of this argument in the coming days. The concern I want to raise in this essay is not so much about the legal merits of the argument, although those are relevant, but about the deeper questions regarding the norms and principles of constitutional democracy that are at play in this debate. Despite their diametrically opposed conclusions as to the main issue, the Silva and Mendis views share an interpretative approach reflecting two schools – textualism and intentionalism – within the same tradition of legal positivism. Confronted with the dilemma of interpreting two sets of highly unsatisfactory constitutional rules introduced by questionable processes and even more questionable motives as reflected in President Jayewardene’s Third Amendment and President Rajapaksa’s Eighteenth Amendment, it seems neither of these positions adequately addresses the deeper normative concerns that ought to inform the task of constitutional interpretation.

In at least one sense, they seem like legal arguments constructed in support of preconceived political positions with regard to the incumbent president’s political future, and they do not seem to be, at least directly, concerned with the purposes and values that constitute, or ought to constitute, the constitutional foundations of the Sri Lankan state. Determining the legality or constitutionality of any governmental action is obviously important, and to that extent the debate between the Silva and Mendis views is illuminating. But in a more broadly political sense of the constitution, we find ourselves today in a difficult and disappointing situation, with extremely grave challenges for the future of democracy and constitutionalism. This seems to require an alternative way of approaching this matter, which places the imperative goal of re-democratising the state at its heart.

The Issue

The 1978 Constitution in its original form reflected two key principles of democratic presidentialism in providing temporal limitations on executive power. These were the principle of the fixed term and the principle of the term limit. The first established a fixed number of years that constitutes a single presidential term of office (in our case, a relatively long six years). The second established the number of such terms that one person could be elected to hold presidential office (in our case, two terms). In addition to substantive limitations on executive power such as a strong separation of powers, there is a logical rationale for such temporal limitations in a presidential democracy. Unlike in a parliamentary system in which the life of the government is usually dependent upon the retention of the confidence of parliament through responsibility and answerability to it, under a presidential system, the conferral of relatively autonomous power on the directly elected executive must be balanced by other limitations. Without these key limitations a presidential system becomes an elected monarchy rather than a republican democracy.

While therefore the 1978 Constitution began by conforming to these requirements, very soon the temptations of party advantage and executive convenience overtook the commitment to constitutional principle. Thus in 1982, the Third Amendment was passed so as to allow the incumbent to seek early re-election after the passage of four years of his first term. At the time, this did not apply to a second term because the two-term limitation was still in place, but after the Eighteenth Amendment, this is available during any term. This removal of the fixed term principle was intended to give the incumbent the maximum advantage over his opponents by allowing him to choose the most beneficial timing for his re-election. The dreadfully convoluted wording of Article 31(3A) introduced by this amendment demonstrates the extent to which it was tailored to benefit a sitting president as well as the specific circumstances of President Jayewardene in 1982.

The absence of coherence and principle in this provision was the subject of another intervention previously by Chief Justice Silva in Ven. Omalpe Sobhitha’s Case in 2005. In that decision, he cut short President Kumaratunga’s second and last term by a year, by giving what is semantically a rather elastic interpretation to that provision. Some intervening petitioners in that case suggested that, while the provision was complicated and repugnant to democratic principle, it was capable of a clear construction. It was argued that where it was possible, the clear meaning of the text should be upheld as the least objectionable option, rather than a strained interpretation to serve immediate objectives with unforeseeable consequences. Chief Justice Silva, who in relation to the present third term issue is advocating a highly textualist interpretation, strongly rejected such an approach in 2005, preferring instead an activist – some might say adventurist – interpretation. The more restrained interpretation would also have meant that President Kumaratunga could hold office until 2006, which was an outcome that seemed at the time that Chief Justice Silva was determined to avoid.

The surviving principle of the two-term limit was then done away with by the Eighteenth Amendment in September 2010, moving Sri Lanka further towards hyper-presidentialism. In the preceding presidential and parliamentary elections of January and April 2010, there had been absolutely no indication given to the electorate about the impending changes, and in any case, President Rajapaksa’s two-thirds majority that ensured the passage of the Eighteenth Amendment was secured not by election but by post-election crossovers of MPs elected from opposition parties, in many cases facilitated by allegedly corrupt methods.

Both amendments were enacted at the behest of presidents controlling large parliamentary majorities, with little or no legislative and public deliberation as to the rationales or the consequences of the measures. Certain MPs and even constituent parties of the ruling coalition are now regretting their support for the Eighteenth Amendment. These expressions of contrition for their sycophantic conduct are sadly inadequate, for it did not take much intelligence or prescience to predict what would ensue from the face of the Bill in 2010. On both occasions, moreover, the Supreme Court timidly acquiesced with the government in power, abandoning its role as a check on the executive. At least in 2010, the court had the benefit of extensive representations from interveners such as Dr Jayampathy Wickramaratne and Rohan Edrisinha as to the possible effects of the amendment bill, and why these would be so far-reaching as to require the consent of the people at a referendum. Chief Justice Shirani Bandaranayake of course subsequently paid a very high personal price for her capitulation on that occasion.

The specific issue about whether or not President Rajapaksa is legally entitled to contest an election for a third term arises within this context and history of undemocratic constitutional manipulation and judicial failure. It is important to stress that there is no disagreement that the Eighteenth Amendment has validly repealed the two-term limit; any person elected in the future to the presidential office would have no term limits whatsoever. Rather, it is about whether the incumbent, who was twice elected under the previous framework that contained the two-term limit, can benefit from the subsequent abolition of that limit. Moreover, if it is accepted that the incumbent is bound by the pre-Eighteenth Amendment rules, then it follows that he is also not entitled to call an early presidential election as is now proposed, because he enjoyed that right only during his first term.

The Silva View

The gist of the constitutional interpretation advanced by the former Chief Justice is that the abolition of the term limit effected by the repeal of Article 31(2) by the Eighteenth Amendment does not apply to the incumbent president. President Rajapaksa had already been twice elected by the time the Eighteenth Amendment was enacted. The rules that apply to him therefore are those that prevailed before in terms of Article 31(2) that: “No person who has been twice elected to the office of President by the People shall be qualified thereafter to be elected such office by the People.” That rule applied the moment he was elected for the second time in January 2010, thereby disqualifying him from any future candidacy for the office of president. If Parliament intended the abolition to apply to him, then it should have done so expressly in the Eighteenth Amendment Act. In the absence of such an express provision, the rule of interpretation set out in Section 6 of the Interpretation Ordinance applies to deny the retroactive application of the abolition. Section 6(3) provides:

“Whenever any written law repeals in whole or part a former written law, such repeal shall not, in the absence of any express provision to that effect, affect or be deemed to have affected – (a) the past operation of or anything duly done or suffered under the repealed written law; … (c) any action, proceeding, or thing pending or incompleted when the repealing written law comes into operation, but every action, proceeding, or thing may be carried on and completed as if there had been no such repeal.”

When this rule of interpretation is applied to Article 31(2), and its repeal by the Eighteenth Amendment without an express provision to make the repeal applicable to the incumbent president, the disqualification that prevailed prior to the repeal continues to apply to the incumbent. The key point here is the requirement of an express provision: for the abolition of the term limit to be applicable to the incumbent, the Eighteenth Amendment should expressly have provided for it. In its absence, the previous legal provision and its effects apply to the incumbent.

This then is a highly textualist position, in the sense that it insists that the correct legal answer to this question can be derived solely from the text of the constitution, i.e., the text of the pre-repeal Article 31(2) and the text of the Eighteenth Amendment, construed in the light of the general principle of interpretation laid down in Section 6 of the Interpretation Ordinance.

The Mendis View

Chinthaka Mendis rejects this interpretation on a number of grounds. He argues that the principles of interpretation laid down in the Interpretation Ordinance concern only legislation subordinate to the constitution; accordingly these principles do not apply to the interpretation of the constitution itself. In any case, the principle in Section 6 of the Interpretation Ordinance is not an absolute rule but merely a presumption that could be rebutted by demonstrating the contrary intention behind the repealing legislation, even in the absence of the required express provision. Furthermore, Section 6 only refers and applies to ‘written law’, which is defined in Section 2(kk) of the Interpretation Ordinance as ordinary legislation and in terms that do not include the constitution. Rather, the constitution is included in the definition of ‘enactment’ in Section 2(f), which means that Silva’s contention regarding the applicability of Section 6 to the interpretation of the constitution is flawed. Mendis’s most substantial argument is that the textual defect of the absence of the express provision required by Section 6 of the Interpretation Ordinance in the Eighteenth Amendment seized upon by Silva cannot defeat the clear intention of the legislature in enacting the Eighteenth Amendment. That is, in enacting the Eighteenth Amendment, Parliament not only abolished the term limit (as well as repealing the reference to the two-term limit as a substantive disqualification for presidential office in Article 92(c)) but also intended the abolition to apply to the president in office at the time. This is an entirely intuitive argument to the extent that no one, until Sarath Silva came out with his novel argument a few weeks ago, ever thought that the Eighteenth Amendment was intended to benefit anyone other than the incumbent. This would be amply borne out by an examination of the parliamentary debate on the Eighteenth Amendment Bill.

Mendis’s view can thus be characterised as an intentionalist approach, in the sense that its key criterion is the intention of the legislature in enacting the Eighteenth Amendment. While he rejects the applicability of the Interpretation Ordinance to the interpretation of the constitution (although judicial authority on this point is against him, and both Ratnapala and Hameed have persuasively argued the contrary), we can assume that even if it did, his point would be that legislative intent must prevail notwithstanding the textual defect of the express provision required by Section 6 being absent in the Eighteenth Amendment.


We thus have two seemingly incommensurable views as to the meaning of the relevant constitutional provisions with regard to the legality of President Rajapaksa’s third term. First let us assess these solely on the merits within a legal positivist framework. The Silva view strictly follows the text of the constitution read in the light of the interpretational principles applicable to the legal system as a whole. It is not for judicial or other interpreters of the constitution to read in an express clause required by the legal system into the Eighteenth Amendment if it is to have a certain effect, when the drafters of that instrument have excluded it, even if unwittingly. In the absence of this express requirement – statutorily established in the legal system for well over a century and by the common law well before that – the only reasonable conclusion that a court can come to is that Parliament did not intend President Rajapaksa to benefit from the abolition of the term limit. The provisions of the Eighteenth Amendment with regard to this specific matter therefore apply exclusively prospectively to future presidents elected in and after 2016, when the incumbent’s second, last, and full term has reached conclusion.

However, the applicability of Section 6 of the Interpretation Ordinance is pivotal to this interpretation, and as Mendis has pointed out, this provision only speaks about ‘written law’ and not ‘enactments’, which is the category in which the constitution is included in Section 2 of the Interpretation Ordinance. If Section 2(f) had not mentioned the constitution expressly and specifically, then there was a possibility that ‘written law’ could be interpreted, internally pari materia, to include the constitution. But given that it does, under the well-established common law canon of interpretation known as expressio unius est exclusio alterius (the expression of one thing is the exclusion of another), it would seem that Section 6 cannot be extended to the constitution. Even if therefore it is conceded by Mendis that the Ordinance applies to the interpretation of the constitution, the Silva view has to answer how and why Section 6 must apply to former Article 31(2) and the Eighteenth Amendment. Silva simply cannot dismiss this line of attack: he cannot adopt a strictly textualist attitude to the text of the constitution without also adopting the same approach to the Interpretation Ordinance. Ironically, that requirement of textualist consistency is what is potentially fatal to his argument if the Mendis critique is accepted. This point has now however been extensively canvassed by Ratnapala and Hameed. Among other arguments, they point out that since the 1978 Constitution was validly enacted as a Law of the National State Assembly under the 1972 Constitution, and the ‘Laws’ of the NSA are included within the definition of ‘written law’ in the Interpretation Ordinance, the latter applies to the interpretation of the constitution.

On the other hand, as already noted, the Mendis view is the more politically intuitive one to the extent that supporters and critics alike of the Eighteenth Amendment have consistently assumed that the abolition of the term limit was always intended to apply to President Rajapaksa. Within the positivist framework, external political commentary including parliamentary debates on the nature and scope of the Eighteenth Amendment would be irrelevant for the purposes of judicial interpretation of the constitution, and the only way in which a court can gather the intention of the legislature in amending the constitution is by the text of the constitutional amendment itself. Given that Mendis places great stock on legislative intention, the question arises as to how this intention is to be discovered judicially.

It is of course entirely possible that a court could simply ignore Silva’s argument (and by extension those of Ratnapala and Hameed) and proceed on the basis of the commonly held perception about the effect of the Eighteenth Amendment. Ignoring or dismissing without proper reasons large parts of submissions in this type of constitutional proceeding has regrettably become a common feature of the Supreme Court’s case law. If it decides to address the matter head-on, however, the court would have to adopt a principle akin to that established in the UK House of Lords decision in Pepper v Hart (1992) – although the UK courts have gradually retreated from it to the extent that it is virtually meaningless in that jurisdiction now – whereby it could examine Hansard so as to discern a legislative intention that President Rajapaksa is not to be bound by the two-term limit. This is a highly controversial proposition with major risks for the separation of powers, and more specifically, one that goes against the grain of the commitment to parliamentary supremacy implicit in the Mendis view (i.e., strictly-speaking, Pepper v Hart involves a rejection or at least a dilution of the Enrolled Bill Rule central to the doctrine of parliamentary sovereignty). The supremacy of parliament and its immunity from judicial interference was the regime’s attitude – in defiance of the Court of Appeal and the Supreme Court – in the impeachment proceedings against Chief Justice Shirani Bandaranayake. While not beyond its ken, it would therefore be both ironic and inconsistent for the regime to invite judicial inspection of parliamentary proceedings in this matter.

It can thus be seen that both the Silva and Mendis views have their strengths and weaknesses, but with the addition of the views of Ratnapala and Hameed, it can be concluded that the Silva view is the substantially stronger interpretation. While this dispute might be settled through legal principles alone, the more serious problem with narrowly positivist approaches is that they can only be resolved in either/or terms. One or the other has to prevail, and because they are also closely allied to opposed views in a contentious political debate concerning the career of the sitting president and his regime, this in turn transforms the judicial role in constitutional adjudication into a politicised process. The paradox of positivism in this regard therefore is that its pretence to the autonomy of law from politics only serves to undermine that very claim by politicising law in contentious ways. This to me points to a need for another interpretational method of resolving this problem, one that is capable of resolving the functional crisis generated by the fundamental interpretational disagreement as between the Silva and Mendis views in a normatively defensible way.

An Alternative Approach

The Silva versus Mendis debate is conducted in the adversarial terms that come naturally to lawyers trained in the common law tradition, but even within this tradition, it is important to remember the distinctiveness of constitutional interpretation and adjudication from ordinary legal disputes. Someone like Professor Ratnapala of course does not need to be reminded of this, but on this occasion the Bar Association has sought his opinion on a point of public law, not his reflections as a constitutional theorist. In ideal terms, constitutional law and politics usually engage higher order moral and philosophical norms, they involve the basic values upon which the constitutional community is constituted, they invite reflection about our past, present, and future, and they call for reasonable and inevitable disagreements about all these things to be addressed at high levels of abstraction so as to generate the widest possible level of consensus within the constraints of democratic pluralism. In short, constitutional discourse is not about short-term goals, doing the best for one’s client, and winning the case at hand, but about how to reinforce and elaborate upon the fundaments of the democratic community. It calls for deliberative debate and conversational constitutionalism, not zero-sum combat. This is neither as naïve nor as impractical as it might sound to those jaded by the cynicism engendered by the nature of our post-colonial constitutional history, but a pressing necessity if we are to ever recover from the pervasive erosion of constitutional democracy throughout that era.

The third term issue illustrates this point well. As I mentioned at the beginning, the constitutional rules engaged in this dispute derive from two constitutional amendments that have been enacted for the executive convenience and partisan advantage of two presidents with a proven authoritarian bent. It is easy in these circumstances to either disengage from the constitutional conversation, or to use the rules purely instrumentally, of which Sarath Silva’s rather clever argument is a good example. Some will no doubt feel satisfaction in the discomfiture Silva’s intervention will cause President Rajapaksa, although given his insouciant attitude to constitutional propriety, this is likely to be relatively minor and short-lived. In my view, however, the thoroughly unsatisfactory nature of these rules demand not disengagement or short-term point-scoring, but a much deeper commitment on our part to interpreting them with a clear view to diminishing their more corrosive effects, and through the reinterpretation process, to augment whatever democratic potential of the broader constitution of which they are a part, until such time as they and it can be formally replaced with something better. And this is matter for a social constitutional conversation: in a republic, ideally, constitutional discourse ought to be a democratic activity and not the exclusive preserve of the courts or elites.

If all of this is true, then it begs the question of how we can transcend the limits of the positivist arguments. The subjective view of the legislator is hardly an appropriate standard to use in these circumstances, unless what we wish for the government of the country is a presidential monarchy, with unlimited terms, weak or no checks and balances, perpetual immunity, a general breakdown of constitutional government, and for the general will to be supplanted by an individual will. Such a regime will be marked by impunity, corruption and waste, and a growing and comprehensive erosion of democracy. As these dangers are becoming more and more visible in Sri Lanka, it is apparent that the electorate’s credulity and goodwill towards the regime is wearing thin. When the public’s withdrawal of consent becomes more and more difficult for the regime to countenance, it will seek refuge in more and more authoritarianism, including through the expansion of militarisation.

Avoiding this disastrous downward spiral is in the overwhelming public interest. We therefore need more objective standards against which to construe specific constitutional rules. Such objective standards are available within the norms embedded within the constitution itself, and through the universal character of those norms, to broader comparative and philosophical resources that we can bring to bear on constitutional interpretation. The key principles here are set out in the preamble to the 1978 Constitution as ‘the immutable republican principles’ of representative democracy, freedom, equality, justice, fundamental human rights, and the independence of the judiciary. The preamble also characterises these values as ‘the intangible heritage that guarantees the dignity and well-being of succeeding generations of the people of Sri Lanka, and of all the people of the world.’ This places our country firmly within the community of democratic states. The standards and principles of government adhered to by this category of countries in the world are therefore directly applicable to the way in which we interpret our constitution. It is both necessary and appropriate that these values and norms are set out in the preamble in open-textured language. They are resplendent generalisations derived from ‘the intangible heritage’ common to humanity. And it is only by reference to the normative foundations set out by these values that we can articulate the concrete meaning of specific rules. When applied to undesirable rules such as those established by the Eighteenth Amendment or the Third Amendment, this approach enables us to restrict the harmful effects of those rules.

As I have suggested before, if this approach was adopted by the Supreme Court in the pre-enactment proceedings before the Eighteenth Amendment, then it is certain to have required a referendum. Whether in the political context of 2010 this would have been won by the regime is not the point; the point is that, at least notionally, the people would have had to consider the implications of the far-reaching changes and then provided or withheld their consent. Rather than rise to the occasion, the Supreme Court delivered itself of a shallow and pusillanimous determination that basically gave the green light to the regime to do as it pleased. In this context, before applying these principles, it is necessary to recall the following relevant facts regarding the substance and the process used for the enactment of the Eighteenth Amendment.

  • The Mahinda Chinthana, Mr Mahinda Rajapaksa’s manifesto for the 2005 presidential election, at which he was first elected to the presidency, stated the following at p.97: “With the consensus of all, I expect to present a Constitution that will propose the abolition of the Executive Presidency … In the interim, I propose to present a Constitutional amendment through which the Executive President will be made answerable to Parliament …”
  • The Mahinda Chinthana Idiri Dekma, President Rajapaksa’s manifesto for the 2010 presidential election, at which he secured re-election, stated the following at p.56: “I was particularly careful when exercising the powers of the Executive Presidency. In the past, the Executive Presidency was used to postpone elections, to topple elected governments, to disrupt the judiciary, to ban political parties, to suppress demonstrations and lead the country towards a violent culture, to sell state institutions at under-valued prices, to defend criminals and to grant concessions to unscrupulous businessmen. Agreements that betrayed the country were entered into using the powers of the Executive Presidency. I used such powers to unite the country. An open discussion on the Executive Presidency will be held with all parties. The Executive Presidency will be converted into a Trusteeship which honours the mandate given to Parliament by being accountable to parliament, establishes equality before the law, is accountable to the judiciary and enacts laws that are accountable to the judiciary, and is not in conflict with the judiciary.”
  • Thus President Rajapaksa promised in his first election to abolish the executive presidency, whereas in the second there is no such promise but a largely meaningless mass of words about consultations for reform. In addition to the broken promise of abolition, his account of how presidential powers have been abused in the past would accurately describe how he and his government have exercised executive power since 2005.
  • In the presidential election of January and the parliamentary election of April 2010, which preceded the Eighteenth Amendment in September, there was no indication whatsoever that there would be a constitutional amendment to abolish the presidential two-term limit or to dilute the Seventeenth Amendment and re-politicise independent institutions and services.
  • In the weeks preceding the introduction of the Eighteenth Amendment Bill, just like how in the present speculation is taking place about an early presidential election in 2015, there were rumours about impending constitutional change. In the absence of clear and transparent proposals by the government or any attempt to set out a proper process, confusion reigned in public discourse fed by conjecture, rumours and gossip. Media speculation about the proposed changes ranged from the removal of the two-term limit, to the abolition of the executive presidency and its replacement with an ‘executive prime minister’ (no one knew what this meant), and the introduction of a second chamber.
  • According to Rohan Edrisinha and Aruni Jayakody, here is how the process of its enactment panned out:

“The President placed the 18th Amendment before the Cabinet on Monday 30 August. Since the Cabinet declared it as urgent in the national interest the Amendment bill was automatically referred to the Supreme Court, the following day on Tuesday 31 August. Those who were fortunate enough to have had access to a copy of the Amendment, intervened before the Supreme Court. At the Supreme Court hearing, it became apparent during the Attorney General’s submissions that the version of the bill in possession of the intervening petitioners was different to the version relied on by the Attorney General [Mohan Peiris, P.C., the current Chief Justice]. When the intervenient petitioners objected in court, the Attorney General turned to them and stated “This is what happens when you have documents you are not supposed to have.” He subsequently relented however, and gave a copy of the Amendment to one of the intervenient petitioners who then asked his junior to rush out into the streets of Hulftsdorp to make photocopies of the Amendment for the others who were challenging the bill. Thus, the intervening petitioners were only given accurate copies of the proposed changes after the Attorney General had commenced his submissions.

Within a day of the hearing the Supreme Court issued its opinion in a judgment consisting merely of few pages, holding that the Amendment did not affect the entrenched provisions of the Constitution, and thus, did not require a referendum. Though a parliamentary debate took place, it did so without the participation of the main opposition party, which with its now characteristic irresponsibility boycotted the debate, and with little contribution from the smaller opposition parties. The Government was able to secure a two- third majority and passed the bill in to law few days after the Supreme Court hearing.”

This policy and legislative history of the Eighteenth Amendment makes it abundantly clear the extent to which it is contrary not only to the fundamental constitutional values described above, but also to the practices of democratic government including truthfulness, transparency of intent, consultation, choice and consent. This demands that we interpret the Eighteenth Amendment in the most restrictive way possible, which in turn means a preference for the Silva rather than the Mendis view on the specific issue of the incumbent’s third term. But for the reasons I have mentioned, that strictly legal interpretation must be complemented and reinforced by a full normative defence of that interpretational preference if we are to make this into an exercise in re-democratising the state, rather than a brief legal skirmish that will soon be forgotten. That is the constitutional conversation that it is the duty of the opposition parties, the independent media, and civil society to start and sustain. If the electorate is not engaged in this way, then of course we will go back to business as usual in which charismatic populists rule the roost and democratic politics is solely about base instincts and emotions, and not reason and ideals.