Transcript of a presentation made at the session on Forms of Government at the International Conference on Dynamics of Constitution Making in Nepal in a Post-conflict Scenario held in Kathmandu, Nepal in January 2010. The other panelists were the Hon. Bob Rae Q.C. M.P.Canada and Justice Pekka Hallberg, President, Supreme Administrative Court, Finland. The conference was co-hosted by the Nepal Constitution Foundation (NCF), Tribhuvan University Faculty of Law and the Supreme Court Bar Association
Let me begin by thanking you Mr. Chairman and the organizers of the conference for inviting me to speak this morning. We in Sri Lanka watch with very keen interest the constitution making process in Nepal. Those of you who are familiar with the constitutional and political developments in Sri Lanka will know that we have been struggling for the last 15 years to change our constitution. Whenever we have tried to reform our Constitution even before that our process has been flawed. Our process has not been half as participatory or as inclusive as your process has been. But we live in hope because fundamental constitutional reform is essential for our country. When another attempt is made, hopefully, in the not too distant future, we shall certainly look to the developments of Nepal in the last two to three years for inspiration and guidance.
I will like to share with you the Sri Lankan experience on one of the topics that I am led to understand the Nepal process is struggling with. I understand that the Constituent Assembly (CA) Committee on Determination of the Form of Government has presented a report in the form of a compilation of the official positions of the major parties rather than as a consensus document agreed to by all the parties. The stand of the UCPN (Maoist) in favour of a “consensual presidential system and multi-member direct proportional election system” comes as the first among them. The Nepali Congress has voted for the continuation of the parliamentary executive system with a nominal head of state (the President) and mixed-member proportional electoral system. The Communist Party of Nepal (UML) has opted for a presidential form of government elected by the legislature and a mixed electoral system. There is a fourth proposal from CA member Pradeep Giri (Nepali Congress) who has espoused the proposal of 25 NC legislators last year pleading for a directly elected prime minister and a ceremonial president elected by provincial and federal legislatures. This proposal is similar to the Israeli experiment in the 1990s, which was subsequently abandoned because of systemic contradictions that emerged in the implementation process.
I would like to, if I may in the same spirit Mr. Bob Rae referred to, highlight the dangers of the presidential system through the Sri Lankan experience. In our 61 years of independence we had the parliamentary executive model for 30 years and it had the presidential executive model for 31 years. I do think Sri Lanka offers a very good case study for you as you debate this question of whether you should have a presidential executive or a parliamentary executive. I might also remind you that Bangladesh had a similar debate and after a brief flirtation with the presidential model, then reverted to the parliamentary executive model.
I have used deliberately in the topic of my presentation the “Perils of Presidentialism” which is the topic of seminal article written by Professor Juan Linz, the Sterling Professor of Political Science at Yale University in a seminal article that he wrote in 1990 in a journal called the Journal of Democracy. Although it was written a long time ago I think the literature and the scholarship that has followed has critiqued the Linz thesis somewhat, but has really not refuted the basic argument that Professor Linz made in the article. I suggest the members of the Constituent Assembly and civil society who are interested in this issue of forms of government read a copy of Linz’s article.
Linz’s basic thesis was that the presidential system generally promoted authoritarianism and undermined liberal democratic values and institutions. I quote from his article.
“A careful comparison of a parliamentarism as such with presidentialism as such, leads to the conclusion that on balance the former i.e. the parliamentary system is more conducive to stable democracy than the latter. This conclusion applies especially to nations with deep political cleavages and numerous political parties.”
Linz focused primarily on several countries of South America and Spanish speaking world. He mentions Sri Lanka in passing at the beginning of the article but I believe that the Linz thesis applies very much given the Sri Lanka experience of presidentialism from 1978 to the present. In the interest of time I shall briefly outline the Sri Lankan experience and then comment on it using where relevant some of Juan Linz’s arguments.
When Ceylon/Sri Lanka became independent in 1948, its first post independence Constitution, the Soulbury Constitution, provided for a British style Parliamentary Executive with a Prime Minister as its head. The Queen of the United Kingdom continued as a constitutional monarch with a local Governor General as the head of state in her absence. When the United Front Government of Sirimavo Bandaranaike took the lead in espousing the need for a homegrown or autochthonous Constitution in the early 1970s, the Constituent Assembly rejected a proposal to introduce a presidential system and opted to continue with a parliamentary executive model. The First Republican Constitution of 1972, therefore, which was operational until 1978, was similar to the Soulbury Constitution in terms of the form of government but replaced the Queen with a nominal President appointed by the Prime Minister.
When the United National Party won a resounding victory at the parliamentary elections of 1977, it decided to introduce a new constitution and largely at the insistence of its leader, J.R. Jayewardene, it proposed the introduction of an executive presidency. Though this was opposed by most of the other political parties since the ruling party could obtain a two-thirds majority vote in Parliament on its own, it was able to introduce the Second Republican Constitution of 1978 with a executive presidency, unilaterally. The dangerous trend of unilateral constitution making rather than consensual constitution making, or the instrumental use of constitutions by governments in power in the interests of governments in power, that had commenced in 1972, continued much to the detriment of constitutionalism in the country.
In 1978 the advocates of presidentialism basically put forward two arguments in favour of the presidential model. The first was the need for stability and strong government. The second was that the presidential system would actually empower the minorities as a President, the main political actor, would have to be elected by the whole country rather than from a small constituency. The justification was that since a successful presidential candidate would have to attract votes from the north and south of the island and from all ethnic and religious, this would also encourage moderation in the policies of the candidates.
One has to seriously reflect upon these two justifications for the presidential system. What do we mean by stability? I think this is a very important question. Prior to 1978, we had in Sri Lanka nearly 50 years of franchise. Every five years we changed our governments at elections that were basically free and fair. While some people may describe this as a sign of instability I would argue that in the long term it was actually a sign of stability. It is significant that post 1978 we have had much more instability in terms of respect for constitutionalism, the rule of law and the legitimacy of democratic processes. I believe that the proponents of the presidency for stability argument really were suggesting that a “third world” country like Sri Lanka needed a strong government, that the political and economic challenges of a developing country required that there should be a strong government that could take tough, unpopular decisions that was in the long term interest of the country. The chief architect of the 1978 Constitution President J.R. Jayewardene famously once said, “We need an executive freed from the whims and fancies of the legislators”. How this statement fits with the principles of liberal democracy is of course is an interesting question.
The second argument that the presidential system empowers the minorities is, in my opinion, the more powerful one. It needs to be examined carefully. I think the experience in Sri Lanka has demonstrated that the minorities are empowered, but only at the time of the presidential election or during the campaign. The empowerment is, therefore, limited and not a continuing one. The problem is that once the President is elected he or she is so secure in power, is so powerful, that s/he becomes virtual elected despot. Thereafter it is extremely difficult, not only for the minorities, but for anyone for that matter, to exert influence or pressure or for public opinion to influence the President. This is a huge problem in Sri Lanka. Indeed, at the forthcoming presidential election in Sri Lanka, the voters face a terrible choice. They have to choose between a very authoritarian incumbent President and a former military commander with no experience or record of commitment to democracy. Many voters agonise over such a horrible choice which highlights, I think, one of the fundamental flaws of our presidential system in particular but also many presidential systems in general. The dilemma faced by the voters in Sri Lanka is that they know that though the candidates may make various positive promises or state all the correct positions or appear to be committed to the rule of law, they also recognize that on the day after the election, on 27th January 2010, whoever is elected will be virtually uncontrollable and unaccountable because no person or institution can exercise an effective check on the office of the President under the Sri Lankan Constitution.
I must emphasise one more feature of the presidential system in Sri Lanka. We have what has been described as a mixed or hybrid Presidential/Parliamentary system or a semi- presidential model. Our system has also been described by a well known political scientist who was a champion of the 1978 Constitution, Professor A.J. Wilson, as Gaullist, because he argued, that it had several features of the French system. It seeks to combine some of the features of purely presidential system with the British parliamentary executive model. So we have a President directly elected by the people for a 6 year term. The President is the head of State, head of government, Command in Chief of the armed forces with special responsibility for the defence of the country. S/he is the head of the Cabinet of Ministers, chooses the Prime Minister and appoints the cabinet ministers who hold office at his/her pleasure.
A feature of the British parliamentary system that we have adopted in our model is that the President has to choose his or her cabinet members from the members of the legislature. The President can also assign to himself any cabinet portfolio and much to our concern the Presidents of Sri Lanka since the early 1990s have also retained the portfolio of finance which in my view is unconstitutional and violates fundamental principles of parliamentary democracy. If one examines the history of parliamentary democracy one sees that the main way by which parliament exercises control over the executive is by controlling the purse strings and through the effective control of public finance. Parliament cannot exercise such control if the Minister of Finance is outside the legislature.
The President of Sri Lanka can dissolve legislature virtually at any time without having to consult anyone. He has wide powers of appointment- of judges, persons to key institutions such as the armed forces and the police, commissions, election commissioners, secretaries to ministries and other public servants, governors of provinces and the Attorney General. In 2001 a constitutional amendment, the 17 Amendment to the Constitution, was adopted to reduce these powers of appointment by requiring them to be exercised on the recommendation of a Constitutional Council with nominees from several political parties and the Leader of the Opposition, a feature that was inspired by the Nepal Constitution of 1990. Though it worked reasonably well from 2001-5, the amendment has been intentionally violated by the government of Sri Lanka since 2005.
The President wields even more power when the country is under a state of emergency. In Sri Lanka, since 1970 a state of emergency is more the norm rather than the exception particularly because of the ethnic conflict we have had for many years. In such circumstances, the President can promulgate emergencies regulations which even trump or override the laws of the parliament. Furthermore, the President has a sweeping, blanket immunity from judicial surveillance. S/he cannot be made a party to any legal proceedings for anything done in his official or even private capacity. So if you are married to the President, you probably won’t be able to divorce the President while he is in office! The President can also appeal to the people directly, over the heads of the legislature, at a referendum seeking popular endorsement for a Bill that has been rejected by the legislature.
The only check that we have in the Sri Lankan constitution and indeed one that Linz argues, is one of the checks in all presidential systems, is impeachment. Under the Sri Lankan Constitution, however, an impeachment is virtually impossible. Two- third majority votes in the legislature on three separate occasions and the finding of guilt by the Supreme Court is needed. The whole process could take months and as indeed happened in Sri Lanka on the one occasion that an impeachment was attempted, the President can make use of the time and of course his enormous powers to “persuade” legislators to change their minds.
We have, therefore, in Sri Lanka what Professor C.R. de Silva, in a critique of the presidential system in Sri Lanka, has called the “overmighty executive”. This is compounded by the fact that the rival centers of political power i.e the legislature and judiciary – have very little countervailing power or control. Such a combination, therefore, is a recipe for the authoritarianism. We have experienced this increasingly in the last four to five years in Sri Lanka. I would like to also take this opportunity to warn against the dangers of hybridity or mixedness. Bob Rae has already referred to this in his presentation this morning. Sometimes when you mix two systems the danger is that you leave out the checks and balances of each system. If you had a pure presidential system where there is strict separation of powers, as our colleague from Finland highlighted this morning, between the legislature and executive, then perhaps the legislature would have functioned more effectively as a watchdog on the executive. Similarly in a parliamentary executive, where the most powerful political actor is a member of the legislature and continues to wield power only so long as s/he commands the confidence of the legislature, this promotes humility, accountability and responsiveness. But since in Sri Lanka, we have neither, there is a serious restraints on power shortfall that undermines good governance and constitutionalism.
Every system has its rationale, logic and its own scheme of regularized restraints. When you mix systems the danger is you might leave out some of the basic checks and balances. This is the reason why in Sri Lanka the presidential system has contributed significantly to the rise of the authoritarianism, the rise of unaccountable and unresponsive government in Sri Lanka. It has also had a rather corrosive and negative impact on liberal democratic institutions and values. I would like to expand on this on the time I have left referring to some of the arguments Juan Linz made in his article.
One of his powerful points in my view is that presidential system encourages a personalized style of politics and a kind of crude populism which is the very antithesis of constitutionalism. It builds on the style of politics which focuses on the individual rather than people with political experience, capacity and commitment to liberal democratic values.
Some scholars have even made the point that the presidential system is better for political outsiders to come in and take charge or take power as opposed to people from within the political elites and they see this as something positive. But I think it could be argued that this is negative. Think of Chavez in Venezuela, Morales in Bolivia, Correa in Ecuador, Estrada in the Phillipines, Takshin Sinawatra in Thailand, Wahid of Indonesia and of course Rajapakse in Sri Lanka. It would probably have been difficult for a Gordon Brown, a John Major, a Manmohan Singh or a Kevin Rudd to have become President if their respective countries had presidential systems. I am sure you have your parallels in Nepal. In Sri Lanka we have the example of one of our most experienced politicians, the perennial leader of the opposition, Ranil Wickremesinghe, who just can’t win a presidential election because he lacks some of those charismatic, populist attributes that has made Rajapakse such a success as President. We need to reflect on whether we need a system that encourages a personalised, charismatic and populist style of politics. We need to be conscious that such a system is likely to promote authoritarianism as it has done in Sri Lanka.
The other danger relates to what the fact that s/he is elected by the whole country does to the President in terms of ego and power and authority. It fosters a mindset where the President tends to think that because s/he is elected by the entire country s/he has the authority and legitimacy to basically do anything. It gives a person an exaggerated sense of is own importance. To use Linz’s own words “the risk that he will tend to conflate his supporters with the” people” as a whole.” For example if you look at Sri Lanka, President Rajapakse has deliberately and intentionally not implemented an entire chapter in our constitution designed to promote good governance, the Seventeenth Amendment to the Constitution. While various bizarre and unconvincing justifications for such violations have been trotted out by friends of the government, the basic reason seems to me the fact that the President just did not want to implement a chapter in the constitution that reduced his discretion, patronage and power. It is almost as if he took the position ” I am the President, I am elected by the people; I should be able to appoint whomever I want to key institutions.” In Sri Lanka certainly therefore, presidentialism has fostered a kind of crude populism that is very dangerous from a liberal democratic perspective. As you know as constitutionalists we are not only concerned about how people are elected, but we also perhaps more concerned about the extent of their power and restraints on their power. Populism as described earlier is the very antithesis of constitutionalism.
Linz also discusses the defence of presidentialism in terms of stability and rigidity. I was struck by the stability argument that is used often in Nepal debate. I think the experience of Sri Lanka is that what is often used to justify stability has resulted in a kind of unresponsiveness and strong government that goes against the interest of the people. The corruption, nepotism and the abuse of power that Sri Lanka has experienced in recent years has created enormous problems with respect to good governance and generated widespread cynicism about politics in the minds of the people. In recent years Sri Lanka had to deal with a strong separatist movement led by the Liberation Tigers of Tamil Eelam or LTTE that promoted, not surprisingly, an obsession with national security. The cumulative effect of all these factors created a negative kind of stability- authoritarianism.
The third argument that I would like to borrow from Linz is that the danger of presidential system is that it promotes a winner takes all or zero sum game outcome. A good example of this will be the forthcoming presidential election in Sri Lanka. The future of the Rajapakse Administration, members of his family who occupy important positions, cronies, party, are all at stake. The stakes are very high. In fact the outcome of the presidential election will impact upon the parliamentary election which is due in April/May. You will recall the argument made last evening by the Minister of Federal Affairs on the need to promote consensual politics, coalition politics and power sharing arrangements. This becomes extremely difficult when you have a presidential model because the President is so powerful that it is very difficult to have a meaningful power sharing arrangement because the Prime Minister and other cabinet ministers will have relatively far less power. In Sri Lanka we had a brief and positive experience of cohabitation- the President from one party with the Prime Minister from another, from 2001 to 2003. It proved extremely difficult to work and lasted for a short time because powers were so loaded in favor of the presidency.
The fourth point that Linz highlights is the danger that the presidential system could devalue democratic institutions. This has certainly happened in Sri Lanka since 1978. It is particularly tragic in the Sri lankan context because as you know we had a very strong tradition of the rule of law, parliamentary democracy and universal franchise since 1931. If you read the Hansards, the debates in the parliament in the 50s, 60s and 70s, the quality of parliamentary debates was very high. Parliament functioned as a deliberative assembly as indeed it is expected to. We had a strong committee system. In fact in the early 1990s, I had a meeting with a delegation from the Nepal parliament that came to Sri Lanka to study our public accounts committee which was considered a model committee. All that has gone today. That tradition has ended and we have had in the last 20 years or so a shift from parliament as the main locus of political power, debate and discussion to the presidential secretariat. The presidential secretariat is full of unelected presidential advisors, subject to little if any parliamentary oversight and is part of the presidential patronage politics that I referred to earlier. This has had a corrosive effect on parliament as an effective democratic institution.
Capable people want to enter parliament any more in Sri Lanka because parliament is no longer the main political institution in our constitutional structure. The quality of parliament has, therefore, suffered. The deliberative and scrutinising functions of Parliament have suffered drastically. Not only has the parliament been devalued but I would argue that the institution of the cabinet of ministers has been devalued as well. It is no longer the focal point for policy debate and formulation. Linz made this point in his article. He says that a presidential cabinet it is less likely to include a strong minded people because all the people appointed to the cabinet hold office at the pleasure of the President. In a parliamentary executive model, on the other hand, you will find that though cabinet ministers and the Prime Minister are sometimes from the same party, there are invariably strong members of the party, persons the Prime Minister has to appoint whether s/he likes them or not, and therefore there is greater likelihood that the quality of the cabinet will be enhanced. Cabinet ministers could resign from the cabinet, go to the back benches and then make matters extremely difficult for the Prime Minister.
A related point that I think is a very important point in the context of South Asian politics is that the parliamentary system promotes a system where there will be less arrogance on the part of the wielders of power. In 1971, one of our Marxist politicians, Colvin R de Silva, defending the parliamentary executive model, stated that a virtue of the parliamentary executive model is that the Prime Minister has to be continuously accountable to the parliament.” The Prime Minister knows that s/he can lose his/her position at anytime if s/he ceases to command the support of the house or the confidence of the house, for instance by a vote of no-confidence. In a presidency it is virtually impossible to remove the President. Furthermore, the Prime Minister is a member of the parliament, sits in parliament with his parliamentary colleagues, is not cocooned in a presidential office far away from parliament. During Question Time, for example, we have all seen how Gordon Brown gets a hard time in the British House of Commons. This is very healthy and it creates a sense of humility, accountability and a healthy sense of irreverence for the person who wields the main political office in the country.
In South Asia where generally our political culture is so hierarchical, where we have very little internal party democracy, where we have a strongly entrenched welfare state with patronage politics etc. the weaknesses, the dangers and the perils of presidentialism that Linz highlighted apply even more forcefully given that political context and reality.
I would like to briefly comment on the proposal that is floating around in Sri Lanka and which I believe has been floated in Nepal as well- the idea of a nationally elected, so called, executive Prime Minister. It was tried in Israel from 1992 to 2001. It was applied or worked for three elections. There were some very Israel specific reasons for the introduction of this strange model. It was felt that the system of proportional representation practiced in Israel gave too much power to small parties. But after three elections it was abandoned and it was acknowledged as a failure. I think it is important to study why Israel flirted with the idea and then abandoned it, before we think of introducing it either in Sri Lanka or in Nepal.
Let me conclude, Mr. Chairman, by summing up my arguments. The Sri Lankan experience with presidentialism has been a negative one when viewed from the perspective of constitutionalism, the rule of law and the protection of liberal democratic values and institutions. Our democratic traditions have been seriously undermined as a result of presidentialism. In 1971 when the proposal to introduce the presidential system in Ceylon/Sri Lanka was made, one of our most distinguished liberal Prime Ministers, Dudley Senanayake made this statement which has turned out to be prophetic-
” The presidential system has worked in the United States where it was the result of a special historic situation. It worked in France for similar reasons. But for Ceylon it would be disastrous. It would create as tradition of Caesarism. It would concentrate power in a leader and undermine parliament and the structure of the political parties. In America and France it has worked but generally it is a system for a Nkrumah or a Nasser, not for a free democracy.”
Secondly I have tried to use Linz’s “Perils of Presidentialism” article as the basis of the critique to suggest to you that perhaps the lessons from Sri Lanka may be of universal application. The particular form of presidentialism in Sri Lanka is problematic but perhaps the Linz thesis highlights the fact that the concept of presidentialism also has serious flaws.
Finally, many constitutional commentators in Sri Lanka have quite often, unfortunately in my view, cited an English poet when they have talked about the subject of forms of government. Alexander Pope said “For forms of government let fools contest, whate’er is best administered is best.” I hope I have convinced you, however, that the form of government is very important. It is a very important topic. It can have profound implications on constitutionalism and constitutional governance as a whole. So in a nutshell, Mr. Chairman, my message is simple. Please don’t follow the advice of Alexander Pope (or any pope for that matter!). Please don’t follow the mistakes of Sri Lanka and beware of the perils of presidentialism. Thank You.