It is difficult to criticise the 1972 Constitution without first conceding the case for it, because the case was strong.
Sri Lanka became independent in 1948 under an instrument drafted in London and brought into force by Order in Council. Whatever its merits, it was granted rather than made, and a country that has just stopped being governed from elsewhere has a reasonable objection to being governed under someone else's document. That objection was not the invention of one party. S. W. R. D. Bandaranaike put it to the State Council in 1945; the Communist Party argued the same year that Ceylonese should have the right to frame their own constitution; the Lanka Sama Samaja Party called the Soulbury settlement a limitation on the freedom of the country.
So when the United Front sought a mandate in 1970 to let the members elected to Parliament sit simultaneously as a Constituent Assembly, and won it, the mandate was real. The Assembly was inaugurated in July 1970. The Constitution it produced was adopted on 22 May 1972 by 119 votes to 16. Section 3 placed sovereignty in the people and declared it inalienable.
The criticism that follows is not that Sri Lanka wrote its own constitution. It is what was written.
A legislature that answered to nobody
The central decision was to make the National State Assembly the supreme instrument of State power, exercising the legislative, executive and judicial power of the people.
The consequence is set out plainly in section 5, and it is the consequence that matters: no institution administering justice could pronounce upon the validity of a law passed by the Assembly. The power the ordinary courts had exercised to test legislation against the constitution was removed.
This was not an oversight. It followed from a particular theory of sovereignty, in which an elected chamber embodies the people and a court reviewing its work is an unelected brake on them. The theory has respectable ancestry. The difficulty is practical rather than philosophical: a constitution that cannot be enforced against the body that makes the law is a description of how power is meant to behave rather than a constraint on how it does.
Every other guarantee in the document inherited that weakness.
Rights that gave way
The 1972 Constitution did something the Soulbury Constitution had never done. It set out a charter of fundamental rights: equality before the law, freedom of thought, religion, speech and expression, peaceful assembly and association, movement, and equal consideration for appointment to the State services.
Then it qualified them. The Assembly could legislate against those rights where it considered the law necessary in the interests of national security, national unity or integrity, the national economy, public safety, public order, the protection of public health or morals, or the rights and freedoms of others.
Read together with the removal of judicial review, the qualification is close to complete. A right that may be overridden whenever the legislature considers the override necessary, in a system where no court may ask whether it was, is a right held at the pleasure of the majority. The contemporary criticism — that what one section granted another took away — was not rhetorical.
Existing law was left in place regardless of inconsistency, so the chapter did not reach backwards either. And there was no straightforward remedy attached to it: a charter without a route into court is a statement of intent.
A court built not to bite
The Constitution did create a Constitutional Court. Its weaknesses are worth setting out precisely, because they compound.
| Feature | Effect |
|---|---|
| Jurisdiction over Bills only | A law could be tested before enactment and never afterwards, so a provision that survived the Bill stage was beyond challenge for good |
| Members appointed by the President | And the President was himself nominated by the Prime Minister, so the appointing chain ran back to the political executive |
| No security of tenure | The one protection that lets a judge rule against the government that appointed him was absent |
| Short time limits in practice | Scrutiny had to be compressed into the legislative timetable rather than taken at the pace a constitutional question deserves |
The Court was constituted in November 1972 and the Press Bill was challenged before it almost immediately, which tells you the mechanism was used. It does not tell you the mechanism was strong.
The clause that was not rebuilt
The most consequential omission is the easiest to miss, because it is an absence.
Section 29 of the Soulbury Constitution had prohibited legislation that made persons of any community liable to disabilities to which others were not, or conferred advantages not conferred on others. It was an imperfect safeguard and it had been read narrowly. The framers of 1972 knew it was inadequate.
Their answer was to drop it and put the rights chapter in its place. That trade looks even only if the rights chapter is strong, and it was not: the override clause and the absence of judicial review left the minorities with a protection weaker than the one they had lost. A general guarantee subject to legislative override is not a substitute for a specific prohibition on discriminatory legislation.
Section 6 ran the other way, giving Buddhism the foremost place and placing the State under a duty to protect and foster it. One may hold any view about the merits of that provision and still notice the asymmetry: the document made a positive commitment on religion in the same breath as it weakened the safeguard the religious and linguistic minorities had relied on.
The response was not slow. The Tamil United Front was formed on 14 May 1972, days before adoption. It resolved to seek amendment, took its seats in June to argue for it, and when amendment did not come, began a campaign of protest on 2 October 1972. S. J. V. Chelvanayakam resigned his seat the following day. Whatever one concludes about what came afterwards, the constitutional grievance was stated at the time, through parliamentary channels, and was not met.
Two provisions about power keeping itself
Two further features are harder to defend than the rest, because neither is a matter of constitutional theory.
The Head of State was nominated by the Prime Minister. In nearly every other republic the office is filled by election, direct or indirect, precisely so that the person who may have to act as a check is not the appointee of the person to be checked. Making the nomination the Prime Minister's removes the independence the office exists to supply.
And the first Assembly was given five years running from the adoption of the Constitution rather than from the 1970 election. A government returned in May 1970 would therefore sit until 1977, seven years, on a mandate given for a shorter term. No referendum was put. Whatever the merits of a longer parliamentary term in the abstract, extending your own is the one constitutional change that cannot be made by the body that benefits without the objection arising.
Alongside this, the electoral system was left alone, and ministerial control was extended over public service appointments and the subordinate judiciary. Each of these moved authority towards the political executive and away from the institutions built to be insulated from it.
What it got right
A fair account has to record the other side.
The Constitution was made here, by an assembly with a mandate to make it, and that settled a question that had been open since 1945. It set out fundamental rights in a written charter for the first time. It abolished the Senate and the appointed members, both of which were defensible reforms. It stated directive principles that were honest about the economic programme the government had been elected on, rather than concealing them.
The impulse was democratic. The design was not, and the two are separable.
Why it still matters
The 1972 Constitution lasted six years. The 1978 Constitution replaced it, and the replacement answered several of these criticisms rather than defending them: fundamental rights became justiciable with a route to the Supreme Court, and the judiciary recovered ground the 1972 settlement had taken.
That is the argument for reading it now. It is the clearest domestic demonstration that a constitution's legitimacy of origin and its quality of design are different things. A document can be made by the right people, for the right reasons, with a genuine mandate, and still concentrate power in a way that its drafters would not have accepted in anyone else's hands.
The structure of the courts that emerged from these settlements is set out in how Sri Lanka's court system works, and the profession that argued through them in the oldest law firms in Sri Lanka.
