Photo courtesy Sri Lanka Brief

A Tale of Two Constitutions

Every country has two constitutions: a “legal constitution” and a “political constitution”. The focus for the past two years has been on the legal constitution. If the government is serious about introducing a new constitution and winning a referendum it must now shift its focus to the political constitution.

A constitution answers a series of question about the State. Is it unitary or federal? Which is more powerful, Parliament or the Executive? What rights do minorities have? What principles will guide MPs, Cabinet Minister and judges as they serve the people? The answers decide the type of constitution, and the type of country, we have. Sri Lanka’s legal constitution is the text of the 1978 constitution. In the new constitution these questions are in focus once more. Will Sri Lanka be an ekeeya rajjya / orumiththa naadu? Will we have an Executive President? Will we change our electoral system?

The political constitution, on the other hand, sits behind the legal text. If a legal constitution focuses on the structure and form of a constitution – unitary with an executive presidency, and proportional representation – a political constitution explains why the legal constitution has its structure and form. It is a set of assumptions and beliefs that provide the rationale, the legitimating force, the reasons, for the structure of the legal constitution.[1]

A political constitution, then, answers the same questions as a legal constitution. But its answers aren’t from constitutional theory. Instead, it draws on religion, history, culture and social norms. In Sri Lanka, a certain strand of Sinhala-Buddhist nationalist thought dominates the political constitution.[2] The legal constitution tells us that Sri Lanka is a unitary State. The political constitution tells us why we should be one: because the continuation of the Buddha’s teaching depends on the territorial integrity of Sri Lanka and the survival of the Sinhala people.[3] The legal constitution sets up a powerful executive president. The political constitution tells us why we should have one: because this model of political power matches the power and dignity of the pre-colonial Sinhala-Buddhist monarch.[4]

Of course, other arguments from legal and constitutional theory can justify the same provisions. The point is that these arguments, grounded in Sinhala Buddhist nationalist thought, are far more convincing. To say that an executive president will protect Sri Lanka like Dutugemunu did of old, has resonance. It has legitimacy. It’s far more convincing than arguments against it which say an executive president affects the separation of powers, or leads to a personality cult.

If the political constitution legitimises the legal constitution in the minds of the majority, what happens if the legal constitution changes in a way that contradicts the political constitution? The best example of this is the 13th Amendment. A (flawed) devolutionary amendment, introduced by Indian pressure, in a State submerged in unitary logic. The outcome was predictable. Sri Lanka has the formal structures of devolution – provincial councils and a provincial public service – but operates just like a non-devolutionary state. There is illegal encroachment on provincial power at every level – administrative, legislative, judicial, executive. Some parts of the amendment, on land and police powers, have never even been implemented. The spectacle of an “unimplemented constitution” is only possible where the legal constitution changes but the political constitution does not, and the latter is more powerful than the former. If the political constitution dominates how a majority of the people, their representatives and bureaucracy of a state behave, as it does in Sri Lanka, then changing the legal constitution alone will make no difference in how the country is run. Both need to change.

Shifting the Focus

If all this is true, then the constitutional reform process must switch gears. It has focused on the legal constitution. This is important work. But elite consensus on changing legal provisions will be useless without popular consensus on changing the political constitution. The constitutional reform project must now focus on the political constitution. If not, the government will likely lose the referendum. More important, even if it wins, without a change in the political constitution there will be no change in how power is actually used in Sri Lanka. Then, whether we call ourselves an orumiththa naadu or not, will make no difference. We will remain as we are.

How do we change the political constitution? The first step relates to the type of arguments we use to advocate for any constitutional arrangement. If exclusionary Sinhala-Buddhist nationalist thought is the foundation of the political constitution, and if it is at the heart of opposition to power-sharing and changes to the executive presidency, then our response must also be grounded in Sinhala-Buddhist nationalist thought. This is how we convince the majority. Like all older nationalisms, Sinhala-Buddhism has regressive and progressive elements. Reformers must resurrect its progressive elements and inject them into public debate. Anthropologists and historians have already done the research. What remains is for reformers to take it to the people. The inherent tolerance and pluralism of Buddhism, and the asymmetric nature of the pre-colonial Sinhala state must be the foundation of all arguments in favour of the new constitution. This is how we win the referendum.

For example, consider the unitary state. Opponents of power-sharing say that this English legal concept is the modern political form that is closest to the pre-colonial Sinhala state. But is it? Critical historians and anthropologists have shown that the state at the time was “neither unitary nor centralised”. It was devolutionary, asymmetric and gave local rulers, including Tamil chieftains, extensive territorial autonomy. Power-sharing, then, is a return to our glorious past, not a retreat from it.[5]

The first step is to make these kinds of arguments. To respond to opponents of constitutional reform from inside their own intellectual terrain. The usual liberal constitutionalist arguments will not resonate; these will. To work, this needs a collaboration between historians, anthropologists, sociologists, and of course lawyers and legal scholars for this to work. Constitution-making is usually dominated by the latter two, but we simply do not have the cultural and social tools to make these arguments effectively. We must have the humility to defer to others.

The second step relates to the way we make these arguments. Dry, rationalist interventions can secure journal publications and win cases in court, but they cannot change political constitutions. If reformers are serious about winning the referendum – and more than that, changing the way this country uses power – then they must turn to drama, dance street theatre, the arts – to bring their arguments to life. The Sudu Nelum movement led by Mangala Samaraweera is the sort of thing we need. Now that the Budget is out, maybe he can step in!


Winning the referendum needs consensus on the legal text, but it also needs changes to the political constitution. And if it passes, making the new constitution work will definitely need a change in the political constitution. Constitutional reform – its advocacy and its practice – must shift to the terrain of Sinhala-Buddhist nationalism. If reformers stick to the legal, constitutional and liberal arguments for reform they may not even win the referendum – and they will certainly fail to win change. Instead, they need a creative, plural strategy of advocacy that takes the political constitution seriously, is grounded in its terrain, and responds to it on its own terms. If they do, change will finally have a real chance in Sri Lanka.

(This article is based on an earlier publication by the same author titled “Sri Lanka’s Constitutional Reform Project: Constitution-making or Constitution-building?” published in the Junior Bar Law Journal 2017.)

[1] Asanga Welikala, ‘Two Concepts of the Constitution: An Essay in Memory of Chanaka Amaratunga’, Groundviews, 19 April 2010, available at: <> accessed 10 November 2017

[2] I use the word thought to indicate that not all Sinhalese Buddhists adhere to this belief-system, and to acknowledge that certain sections of the Tamil and Muslim communities, do. Of course other factors, including social attitudes to hierarchy, nepotism and power, also influence the content of the political constitution.

[3] This is based on a passage in the Mahavamsa where the Lord Buddha is said to have made the following statement to Sakka, lord of the gods: “Vijaya, son of king Sihabahu, is come to Lanka from the country of Lala, together with seven hundred followers. In Lanka, O lord of gods, will my religion be established, therefore carefully protect him with his followers and Lanka” (emphasis added). See William Giger (trans) The Mahavamsa or the Great Chronicle of Ceylon (Ceylon Government Information Department 1950) Chapter VII, 2-4.

[4] See Asanga Welikala, ‘Nation, State, Sovereignty, and Kingship: The Pre-Modern Antecedents of the Presidential State’; Roshan De Silva Wijeyeratne, ‘Cosmology, Presidentialism and JR Jayawardene’s Constitutional Imaginary’; Michael Roberts, ‘Mahinda Rajapaksa as Modern Mahavasala and Font of Clemency? The Roots of Populist Authoritarianism’ in Asanga Welikala (ed) Reforming Sri Lankan Presidentialism: Provenance, Problems and Prospects (CPA 2015)

[5] See Asanga Welikala, ‘The Idea of Constitutional Incrementalism’ CPA Working Papers on Constitutional Reform No 14 (2016) 21