There has recently been a flurry of comment and speculation about whether, due to a technical flaw in the Nineteenth Amendment to the Constitution (2015), former President Rajapaksa may actually be eligible to offer himself for election at the next presidential election, although it had so far been universally assumed that he was disqualified by virtue of having been twice elected previously to the presidency.
A legal argument to this effect has most completely been made by the distinguished jurist, Dr. Nihal Jayawickrama. In his recent article for The Colombo Telegraph/Sunday Island, Dr. Jayawickrama argues that the disqualification from running for presidential office imposed on persons who have been twice elected to the presidency applies only to the future, and not retrospectively to those who were twice elected to that office prior to the changes introduced by the Nineteenth Amendment. This argument rests on the following claims:
- The Nineteenth Amendment repealed and replaced, and not merely amended, the office of President in the 1978 Constitution, as borne out by the particular drafting technique adopted in the Nineteenth Amendment in relation to the provisions on the executive, and in its transitional provisions;
- The powers, functions, and duties of the new office of President are “fundamentally different” from those of its predecessor, and the post-Nineteenth Amendment presidency is “essentially non-executive, primarily symbolic and ceremonial”;
- Article 31(2), which provides for the two-term limit, is “an entirely new prohibition on a citizen’s right to be nominated to stand for election and to be elected”;
- Since the Nineteenth Amendment does not, with express words, show the intention of Parliament to make this prohibition applicable retrospectively, the prohibition must be treated as only applying to the future;
- For these reasons, the two living former Presidents, Kumaratunga and Rajapaksa, both previously twice-elected to the pre-Nineteenth Amendment presidency and not legally made subject to the prohibition, are eligible to stand for presidential office.
With a startlingly counterintuitive premise, Dr. Jayawickrama’s article has certainly succeeded in creating a minor sensation due to its boldness, originality, and timing. In passing, he makes some telling points: for example, the absurdity of the findings about the ‘symbol of national unity’ by the Supreme Court, or the ad hominem use of constitutional amendment legislation to assign ministries to the President. If his main legal argument is accepted, however, the political consequences are potentially enormous; not least by paving the way for former President Rajapaksa to contest again (it is not apparent that former President Kumaratunga has any intention of doing so).
In this essay, speculating on such political outcomes is not my concern, but to carefully – and I hope, constructively – evaluate the legal merits of the argument as an approach to constitutional interpretation. I have both a narrow and a broader purpose. In the former sense, I want to assess the argument purely from a juristic point of view, to understand if the legal principles that are implicated in Dr. Jayawickrama’s argument are correctly applied.
My broader purpose is this: how persuasive is Dr Jayawickrama’s argument, viewed against all the factors that we must consider in expounding our constitution, in a way that is consistent with both its text and underlying values, as well as respectful of the basic intuitions and credulity of citizens in a long-established democracy? In other words, does it comport with common sense as well as legal sense? Are there salient issues that perhaps Dr. Jayawickrama has failed to consider, or weighted insufficiently, in coming to his conclusions? What public interest does it serve, aside from showcasing its exponent’s undoubted talents as a clever and resourceful lawyer? What to make of several awkward infelicities such as, “In 1978, what the National State Assembly established, through the new Constitution, was the office of Executive President”, when we know that it was not the new constitution but an amendment to the 1972 Constitution that established the executive presidency?
As I will show, when analysed in the light of all these relevant considerations, Dr. Jayawickrama’s highly legalistic argument fails to impress either as an explanatory account of the Nineteenth Amendment and the constitutional changes it introduced on the back of one of the most compelling electoral mandates for democratisation in recent memory, or as a recommendation of how to interpret the constitution consistently with the values of constitutional democracy. Unfortunately – and I very much doubt if this was Dr. Jayawickrama’s intention – the account he has offered ultimately emboldens populism and authoritarianism rather than the democratic ethics of limited government. That is not what most people would expect from someone who is, among many other national and international accomplishments, the author of a major Cambridge University Press textbook on the international and comparative law of human rights.
In arguing that the nature of the presidency was changed fundamentally by the Nineteenth Amendment, Dr. Jayawickrama makes much of the fact that the drafters chose to repeal and replace rather than merely amend Article 30, which establishes the office of President. Since, however, the only change made by the substituted Article 30 is to reduce the presidential term of office from six to five years – simply by changing one word – Dr. Jayawickrama believes that this must mean that the drafters intended the new provision to establish an entirely new constitutional office, rather than only amend the term of the existing and continuing presidency. The logic underlying this claim is not immediately clear to me, and I am not sure if so much can or ought to be made of a drafting device. Surely if the aim was to render toothless the over-mighty post-Eighteenth Amendment executive presidency – as Dr. Jayawickrama claims is the effect of the Nineteenth Amendment – then there are far more direct and obvious ways of doing so than using a particular drafting technique. No doubt the Legal Draughtsman’s Department would be able to provide a straightforward clarification of the drafting policies and protocols they use, but an examination of the nineteenth amendment Acts to the 1978 Constitution does not readily reveal when or why replacement is used over amendment. Without further information of the kind that can only be provided by the Legal Draughtsman’s Department, the choice of one or other device, as can be gathered from the face of the amending Acts, is not a reliable indicator of legislative intent.
If so much is to be made of this, then what is the effect of ‘insertions’, which is another drafting device? Do insertions amend the constitution? Can they impliedly repeal older inconsistent provisions? Aside from changing the duration of the presidential term, the other difference between the old and new Article 30 is that the latter drops the comma after the word ‘Government’? What are we to make of this? Is this a mere lapse, or is something more sinister going on? Pointing out these abstruse possibilities is not to ridicule a serious argument, but to demonstrate the manifest absurdities that we can be led to when we adopt excessively technical and legalistic attitudes to making sense of law. As with the numerous apocryphal tales in the lore of Sri Lankan law (e.g., that a ‘vangediya’ and not a human being killed Mrs. Sathasivam), there may have been a time when lawyerly casuistry was the object of lay admiration, but no longer and quite properly so.
The argument that the Nineteenth Amendment wholly reconfigured the presidency flies in the face not only of the structure and text of the 1978 Constitution before and after it, but also of the entire set of circumstances that obtained between the presidential election of January and the certification of the Nineteenth Amendment in May 2015. The early radicalism of the proposal for the total abolition of the executive presidency when the popular campaign against the Rajapaksa regime began in 2014 was gradually and consistently diluted by the need for political consensus and compromise. The multiple compromises that made the necessary legislative majority possible are reflected in the text of the Nineteenth Amendment as enacted, including several incoherent features (e.g., that the President does not require prime ministerial advice in changing subjects and reallocating ministries, but does so in relation to ministerial appointments and dismissals).
Politically, this complicated context was determined by the demands of coalition building around the common candidacy of Maithripala Sirisena, the ambiguities of political agreements and promises within this wide and disparate coalition, the resulting equivocality of Sirisena’s mandate as between reform or abolition, the schism in the Sri Lanka Freedom Party, and the constraints on the new President and his minority government by a hostile majority in Parliament. Legally, continuity and consistency with the existing constitution rather than radical change were also dictated by the new government’s decision to avoid a referendum. The Supreme Court’s pre-enactment determination on the Nineteenth Amendment Bill, among other things, disallowed the Bill’s original proposition that the Prime Minister be made the Head of Government as that would breach the entrenched provisions of the semi-presidential 1978 Constitution. The entire purpose of the determination is to make the Bill consistent with the constitution, and to point out those clauses that exceed the limits set by the constitution. Nothing in this political and legal background supports the conclusion that the Nineteenth Amendment was a drastic change of the sort Dr. Jayawickrama now says it is.
On the other hand, there were some elements of reform that remained constant throughout this turbulent political process. While there was obvious disagreement within the winning coalition and beyond as to whether the executive presidency should be abolished in toto or merely reformed, there was no disagreement whatsoever about reducing the term of the President or re-imposing the two-term limit. The abolition of the two-term limit by the Eighteenth Amendment was one of the more disquieting aspects of the Rajapaksa regime that directly contributed to its electoral downfall in 2015. There seems to have been no dispute on these matters in the pre-enactment proceedings before the Supreme Court and there is no mention of them in the determination. If the effect of these two changes contributed to a change as fundamental as Dr. Jayawickrama alleges it to be, then it is very unlikely that they would have escaped attention as they did in the parliamentary debate on the Nineteenth Amendment Bill and in the Supreme Court proceedings. Rather, the two-term limit was widely seen as simply the reintroduction of a principle that had found expression in the 1978 Constitution for 32 years before the aberration of the Eighteenth Amendment extinguished it for a relatively short period between 2010 and 2015. Article 31(2) – “No person who has been twice elected to the office of President by the People, shall be qualified thereafter to be elected to such office by the People” – is not only a verbatim reproduction of the old provision that was repealed by the Eighteenth Amendment (except for the insertion of the comma now separating the two limbs of the formulation), but it also has the identical Article number as in the pre-2010 constitution. It was not, as Dr. Jayawickrama states, a “new Article” that brings in “an entirely new prohibition on a citizen’s right to be nominated to stand for election and to be elected.” This claim is not so much ahistorical as it is a complete denial of the broader constitutional reality, but Dr. Jayawickrama is compelled to be selective with the facts in this way so as to contrive the basis for applying the rule against retrospectivity to his argument.
The essential legal principle that is relied on by Dr. Jayawickrama to lay down some sort of normative foundation for his other claims is the well-established rule concerning the retrospective application of legislation. This is sometimes expressed in the Latin maxim lex prospicit non respicit (law looks forward not back). Of many definitions and judicial dicta, the authoritative Bennion on Statutory Interpretation (7th Ed., 2017) is one of the most succinct: “It is a principle of legal policy that, except in relation to procedural matters, changes in the law should not take effect retrospectively. Legislation is retrospective, if it alters the legal consequences of things that happened before it came into force” (p.181). It can thus be understood as a substantive common law principle of general fairness as well as a canon of interpretation laying down a rebuttable presumption against retrospective legal effects. In Dr. Jayawickrama’s words, “A law is not considered as imposing a disqualification retrospectively unless Parliament has clearly stated that to be its intention.” But this old and well-known principle is more nuanced than his bald statement of it suggests, and thus his application of it to the meaning of Article 31(2) can also be questioned.
The first thing to note is that the retrospectivity rule must not be overstated. As the legal theorist, Lon L. Fuller noted in his The Morality of Law (1969), “If every time a man relied on existing law in arranging his affairs, he were made secure against any change in legal rules, the whole body of our law would be ossified forever” (p.60, cited with approval by the UK Supreme Court in the recent constitutional case of AXA General Insurance Ltd v The Lord Advocate and Others  UKSC 46 at ). This is all the more important in relation to constitutional change. Of course, most constitutions including our own impose procedural constraints on constitutional amendments, and some like Germany and India also impose absolute substantive limits on the scope of constitutional change based on various conceptions of the general social good. But if a constitutional change motivated by constitutionalist values to impose presidential term-limits were to be frustrated by a theory of retrospectivity about the rights of former Presidents, then that would surely be an inappropriate and perverse misapplication of the principle.
Why this is a misapplication of the rule in the particular circumstances with which we are concerned here becomes clearer when we consider the nature of the rule more closely. Dr. Jayawickrama is correct in implying that the protection of individual rights is one of the rationales for the retrospectivity rule. It would be an infringement of liberty if legislatures were allowed to retrospectively remove or limit rights, to impose disqualifications, or criminalise acts that were not criminal at the time they were carried out (the prohibition of retrospective criminal liability is a fundamental right under the 1978 Constitution: Article 13 (6) and provisos). As Bennion states: “The essential idea of a legal system is that current law should govern current activities. If we do something today, we feel that the law applying to it should be the law in force today, not tomorrow’s backward adjustment of it” (p.181). But the statement of the principle in this essential form does not reveal the full story of how it is applied judicially. The courts’ approach is well illustrated in the following passage:
It is not simply a question of classifying an enactment as retrospective or not retrospective. Rather it may well be a matter of degree – the greater the unfairness, the more it is to be expected that Parliament will make it clear if that is intended (Secretary of State for Social Security v Tunnicliffe  2 All ER 712 at 724 per Staughton LJ)
This view was approved in the House of Lords by Lord Mustill in L’Office Cherifien des Phosphates v Yamashita-Shinnihon Steamship Co Ltd  1 All ER 20 at 29:
I must own to reservations about the reliability of generalised presumptions and maxims when engaged in the task of finding out what Parliament intended by a particular form of words, for they too readily confine the courts to a perspective which treats all statutes, and all situations to which they apply, as if they were the same. This is misleading, for the basis of the rule is no more than simple fairness, which ought to be the basis of every legal rule.
In discussing Lord Mustill’s judgment, Cross on Statutory Interpretation (3rd Ed., 1995) states:
For him, the correct approach was to weigh the various factors, such as the value of the rights, the unfairness of adversely affecting those rights, and the clarity of the statutory language, before answering the question ‘whether the consequences of reading the statute with the suggested degree of retrospectivity is so unfair that the words used by Parliament cannot have been intended to mean what they might appear to say’ (p.188)
What all this tells us, in short, is that the rule is flexible and competing values are weighed by the courts in applying it appropriately to the circumstances of each case; in no sense is it a rule of rigid application admitting of binary conclusions only. When the rule is applied in this way to the situation under discussion, we find that the vested rights that Dr. Jayawickrama is so concerned to protect are not of an order that merits a rigid and undiscriminating application of the rule. That is, we would not be setting out to protect the general interests of society as a whole or even the rights of a substantial section of citizens if we are to hold that Article 31 (2) cannot have retrospective effect, but the political rights of a very small group of individuals who have been twice elected to the presidency. That number currently is two persons – Mrs. Kumaratunga and Mr. Rajapaksa – out of a population of 20,359,439 Sri Lankans, and it is not ever likely to be more than a handful of persons. Not only is the number of affected individuals infinitesimal, but the persons concerned are remarkably privileged in having enjoyed, for a decade or more, the highest political office in the land, which, at the time they held it, was “the fount of all power and patronage” in Dr. Jayawickrama’s words. It is therefore ludicrous to argue that the application of one of the most important principles of power limitation in a presidential state should be abridged, simply to protect the right of ex-Presidents to run for office in perpetuity. In these circumstances, it is little wonder Parliament, and the drafters of the Nineteenth Amendment, felt no necessity whatsoever to affirm the retrospective effect of Article 31 (2) by the express words that Dr. Jayawickrama demands. The balance is so weighted in favour of retrospectivity, and the intention of the legislature so apparent, in the restoration (not the introduction) of Article 31 (2) to our constitutional scheme, that express words would be superfluous.
I described Dr. Jayawickrama’s method at the outset as excessively legalistic. There are many juristic variants of legalism, but for our purposes what is important to highlight about the analytical approach that underpins his article is the assumption that the law is objective, determinate, and comprehensive, and that it can provide answers to any and every dispute without having to draw on extra-legal sources. In this instance, thus, Dr. Jayawickrama asks us to believe that the text of the constitution together with established principles of positive law (e.g., the rule against retrospectivity) say everything we need to know about our constitutional system. I eschew this approach almost entirely, not only because I think the law cannot be made sense of without recourse to meta-constitutional normative principles and public policy considerations, but more fundamentally, without quite involved excursions into the history and culture of society (a type and method of constitutional enquiry that legalists, for more-or-less good reasons, are repelled by). While we do not need extra-legal historical or cultural enquiries for the purposes of this discussion, we certainly do need a certain ‘normativism’ as defined by Jeffrey Goldsworthy in his chapter on ‘Constitutional Interpretation’ in the Oxford Handbook of Comparative Constitutional Law (2012): “a holistic conception of a constitution as more than the sum of its written provisions: as a normative structure whose provisions are, either explicitly or implicitly, based on deeper principles, and ultimately on abstract principles of political morality that are the deepest sources of its authority” (p.691). And it is also in this sense that the task of constitutional construction is distinct from statutory interpretation simpliciter. As Bindra’s Interpretation of Statutes (8th Ed., 1998) says,
[a] democratic constitution cannot be interpreted in a narrow and pedantic (in the sense of strictly literal) sense. It is the basic and cardinal principle of interpretation of a democratic constitution that it is interpreted to foster, develop and enrich democratic institutions. To interpret a democratic constitution so as to squeeze the democratic institutions of their life giving essence is to deny to the people or a section thereof the full benefit of the institutions which they have established for their benefit. (p.871)
It is not that Dr. Jayawickrama’s argument is entirely devoid of any deeper normative content beyond the principle of legal policy embodied in the retrospectivity rule. There is a certain theory of democracy unlimited by constitutionalism that underpins observations like “a citizen’s right to be nominated to stand for election and to be elected.” But it is a perspective on democracy that is both profoundly shallow and deeply problematic. One of the best illustrations of this view is in the parliamentary speech of Professor G.L. Peiris MP, the then Minister of External Affairs, defending the Eighteenth Amendment Bill. In repealing Article 31 (2), he argued,
What we are doing is giving the people of Sri Lanka the opportunity of electing the candidate of their choice. There is no question of extending the term of office of any particular President. To our mind, Sir, that is an enlargement; a strengthening of the franchise of the people of this country. It is not a restriction; it is not a curtailment of the franchise, but precisely the opposite of that. What is wrong in allowing a person to be elected a third time if that is the declared wish of the sovereign people of this land? That is all we are doing; we are paving the way for the sovereign electorate of this country to exercise their freedom of choice in a manner that is not trammelled or restricted by the law” (Parliamentary Debates, 8th September 2010: Col.281.)
This anaemic and procedural conception of democracy is bereft of any constitutional values – for example, that political power must be limited, that public power must not be excessively personalised, and that political accountability in a republican democracy demands rotation of governments – all safeguards against elective tyranny which are denuded in presidential systems without term limits. The rationale for term limits has been a concern for political and constitutional theory from the time of the ancients. Aristotle identified as a key feature of democracy that no man should hold the same office twice. The cursus honorum in the constitution of the Roman Republic all had elaborate rules concerning re-election and term limits. In a speech written for George Washington by James Madison, it was said that the “rotation in office” that comes with term limits would “…accord with the republican spirit of our Constitution, and the ideas of liberty and safety entertained by the people.” Although he did not eventually follow his own advice, Simon Bolivar argued, “Nothing is more perilous than to permit one citizen to retain power for an extended period. The people become accustomed to obeying him, and he forms the habit of commanding them; herein lie the origins of usurpation and tyranny … Our citizens must with good reason learn to fear lest the magistrate who has governed them long will govern them forever.” Bolivar might as well have been describing the predicament we faced after the Eighteenth Amendment was pushed through by the Rajapaksa regime.
To these axiomatic republican norms must be added the tenets of a good political culture, chief among which is the proposition that responsive elected representatives must carry out promises in good faith. The reform of the executive presidency including through the reintroduction of term limits was the reason that Maithripala Sirisena was elected in 2015. For his first constitutional initiative to do anything other than that would have been a gross betrayal of the people’s trust. Moreover, as President Sirisena, opening the debate on the second reading of the Nineteenth Amendment Bill reminded Parliament, not only his own platform in 2015, but from 1994 every presidential candidate’s manifesto – including the three manifestos of Mahinda Rajapaksa in 2005, 2010, and 2015 – has promised reforms to cut back the powers of the executive presidency (Parliamentary Debates, 27th April 2015: Col.513). There has thus been a demonstrable and consistent social consensus on these issues and it is only the deeply problematic character of our political culture that has enabled those mandated by the electorate to implement constitutional change to abandon reform without any accountability. This is not a question we need to explore at more length here, but these observations are more than sufficient to underscore the point that, in contrast to the view presented by Dr. Jayawickrama, the restoration of term limits is not some sudden and illegitimate imposition from above, but a measure for which there are both sound normative reasons of compelling force as well as an empirical reality of social consensus.
The final point I wish to make is easily made and it relates to the most astonishing and least persuasive aspect of Dr. Jayawickrama’s article. He goes to considerable pains to list the changes that cumulatively mean, in his view, that “The office of President established by the 19th Amendment is fundamentally different from its predecessor.” But this is a highly idiosyncratic conclusion that no scholar with even a passing familiarity with the voluminous politics and law literature on comparative government would arrive at, it being already established that his view militates against what most informed Sri Lankans thought had happened to the executive presidency after the Nineteenth Amendment.
We can agree that the changes to the institutional form of executive power made by the Nineteenth Amendment were substantial, and that it restored or introduced a range of temporal, procedural, substantive, and institutional checks on the office of the President. But there is no proper basis whatsoever to conclude from this that the “office of President established under the 19th Amendment is now more in the nature of a constitutional Head of State.” Again, I do not understand how Dr. Jayawickrama squares this theory with the reality of presidential power we have seen President Sirisena exercise since May 2015. It may not be the Rajapaksa-style hyper-presidentialism that it was our collective misfortune to experience under the Eighteenth Amendment, but no one can, with a straight face maintain the contention that the presidency is now a mere figurehead. The President continues to enjoy an independent democratic legitimacy by virtue of direct statewide election, which he deploys in exercising his powers as “the Head of the State, the Head of the Executive and of the Government, and the Commander-in-Chief of the Armed Forces” (Article 30 (1)) and as a member and Head of the Cabinet of Ministers (Article 42 (3)), and while he is now responsible to Parliament (Article 33A, formerly Article 42) he is not answerable to it and can be removed only by the special procedure of impeachment. I can go on, but the point is perhaps sufficiently made.
The transitional provision in Section 49 of the Nineteenth Amendment Act was not necessitated because the offices which Maithripala Sirisena and Ranil Wickremesinghe “held since January 2015 had ceased to exist”, but because, out of an abundance of caution, the drafters were seeking to ensure that the incumbents in the offices of President and Prime Minister at the time the Nineteenth Amendment came into force were made subject to the reforms done to their offices. In other words, it was to guard against a Rajapaksa-style attempt to perpetuate the Eighteenth Amendment powers because they were originally elected or appointed before the Nineteenth Amendment.
The 1978 Constitution, before and after the Nineteenth Amendment, remains a classic representation of Gaullist semi-presidentialism. The Nineteenth Amendment was clearly a significant recalibration of the institutional balance of power in the state, so much so that it can be plausibly characterised as transforming the 1978 Constitution from a ‘president-parliamentary’ to a ‘premier-presidential’ model of semi-presidentialism (to use the canonical typology of semi-presidential sub-types developed by Matthew Shugart and John Carey). One of the foremost academic authorities in this area, Robert Elgie, defines the essence of semi-presidentialism as “where a constitution includes a popularly elected fixed-term president and a prime minister and cabinet who are collectively responsible to the legislature.” That describes the essential features of the Sri Lankan system from 1978 to date, and in their diametrically opposed ways, neither the Eighteenth nor the Nineteenth Amendments altered this basic structure. To state otherwise in the face of this reality is to engage in the sort of sophistic dissimulation that gives lawyers a bad name everywhere.
I would like to conclude if I may on a personal note. Dr. Nihal Jayawickrama is an eminent lawyer and scholar with a stellar record of over half a century of service to the law in Sri Lanka and abroad. As a gracious and erudite contributor to volumes on Sri Lankan constitutional law I have edited, I have valued his acquaintance and his genial company whenever we happen to be in the same country at the same time. I have no doubt he was trying to make a useful contribution to public debate when he wrote the article I have critiqued here. While it is something of a pity that I must disagree with him so fundamentally on this occasion, I am sure he will be the first to recognise that this type of critical dialogue is one of the foremost responsibilities of the academic profession that we share as public law scholars.