Blog of the International Journal of Constitutional Law

An Unconstitutional Constitutional Amendment in Trinidad & Tobago?

Richard Albert, Boston College Law School

Two days ago, the House of Representatives in Trinidad & Tobago passed the Constitution (Amendment) Bill, 2014 by a simple majority. The bill must still pass the Senate by a simple majority and receive presidential assent before becoming law, but neither step is expected to pose a threat to its eventual entrenchment in the Constitution.

The bill proposes three amendments to the Constitution.

The first creates a new requirement that House candidates be elected with a majority vote.[1] The two highest vote-getters in an electoral district would contest a run-off election if no single candidate earns more than fifty percent of the votes cast. As I explain below, this proposal could be better designed but it does not appear to suffer from any constitutional infirmity.

The second amendment would give voters in an electoral district the power to recall their representative.[2] The bill proposes to authorize voters to recall their representative only during the fourth year of the representative’s term. Like the first proposal, this one also appears constitutionally sound, though I wonder about its recall threshold and the way it currently operationalizes the power of recall, as I discuss below.

The third proposal is constitutionally problematic, and below I explore why. The amendment bill proposes to limit prime ministerial service to no more than two full terms, or 10 years and six months, whether the tenure has been continuous or interrupted.[3] The amendment is worded as a prohibition on the president’s power to appoint as prime minister anyone who has reached the ten-year limit.

The political implications of Trinidad’s proposed reforms are important but, for purposes of this post, less important than the potential unconstitutionality of prime ministerial term limits.

First, a word about the two other proposed amendments.

The Run-Off Election

The Constitution (Amendment) Bill, 2014 requires the run-off election, if necessary, to occur fifteen days after the general election.[4] Yet a run-off election is a less preferable and more costly alternative to an instant-runoff vote, which can fulfill the same democratic objectives as a run-off election without the additional cost and, in the case of these amendment proposals, in fifteen fewer days, which is to say automatically on the day of the general election.

The Recall Power

For its part, the recall proposal seems oddly designed at cross-purposes. The recall initiation threshold is set relatively low at ten percent of registered voters in an electoral district.[5] (For comparison, California’s gubernatorial recall threshold is twelve percent, and that is the lowest in the United States, where it is common for states to require twenty-five percent of registered voters to initiate a recall procedure.)

This low threshold in Trinidad invites disenchanted voters in an electoral district to initiate a recall. But once the process is initiated with ten percent of registered voters, the recall can be achieved only with the support of two-thirds of all registered voters in the district.[6] This supermajority threshold makes it very unlikely that a representative will ever be recalled. That being the case, the low recall initiation threshold raises questions about whether the two thresholds are misaligned. If political actors wish to signal to voters the gravity of recall, then it is better perhaps to raise the recall initiation threshold from ten percent to a level that reflects the difficulty of eventually accomplishing the recall, perhaps to anywhere from twenty to twenty-five percent.

Designing the Recall Power: Constructive or Destructive?

One further point about the proposed recall procedure merits mention.

The recall procedure is designed as a no-confidence vote. Voters are called to decide whether they wish to retain their representative on the basis of her or his record alone. The vote is therefore a referendum on the representative. Where the representative loses the referendum as a result of a successful recall vote, a by-election must be held subsequently to fill the vacant seat. In the interim period between the successful recall and the installation of the new representative, the voters in the electoral district would be left with no representation.

Trinidadian political actors could better serve the interest of institutional stability by designing the recall procedure as a constructive vote of no-confidence. This would make it even harder to recall a representative, which is consistent with what the government apparently intends to do by setting such a high threshold to accomplish, though not to initiate, the recall.

A constructive vote of no-confidence is rare, as it exists only in just over a handful of countries, most notably including Germany, Israel, and Spain. Trinidad could design its recall procedure to require voters to decide whether they wish to retain their elected representative or to replace her or him with another specified candidate. The recall would therefore be a choice as between two candidates, not a referendum on a single political actor. With the power of incumbency so great, this would make it extraordinarily difficult to displace a representative.

A Limited Power of Recall

At the moment, it is unclear from the amendment bill whether the government wishes to make it easy or difficult to actually recall a representative. Perhaps the government wants simply to give voters more democracy, just not too much more. As the bill stands, voters in an electoral district would be given just enough power to initiate the recall process but not enough to go much further than that. If that is the government’s intent, it may have struck the right balance with this bill. But then the question becomes whether the presumably high costs incurred to manage the recall process once successfully initiated is worth the marginally small increase in democratic empowerment the bill offers citizens.

Prime Ministerial Term Limits

This brings us to the problematic part of this three-part constitutional amendment. There may be a serious constitutional deficiency in the proposal to institute prime ministerial term limits.

To restate the key elements of the amendment: the bill proposes to introduce a two-term limit on prime ministerial service, up to ten years and six months.[7] The term limit is to be enforced by the president, who is barred by the constitutional amendment from appointing anyone who has exceeded this period of service.

A bit of background is may be useful. Trinidad is a Commonwealth republic. Its head of state was formerly the Monarch of Trinidad & Tobago (who also doubled as the Monarch of the United Kingdom) But the country’s head of state is now a locally chosen, and largely ceremonial, president. In the Commonwealth tradition of responsible government inherited by Trinidad, the president appoints as prime minister the leader of the party that commands the confidence of the House of Representatives. Trinidad has a written constitution. But its constitution also consists of those unwritten constitutional principles inherited from Commonwealth traditions.

This constitutional amendment to prime ministerial service passed the House of Representatives with 23 affirmative votes, 14 opposing votes, and 1 abstention. This margin represents the support of a simple majority of 56% of all members of the House of Representatives.

The government has insisted that only a simple majority in both houses of Parliament, as well as the assent of the ceremonial president, are required to make all three changes to the Constitution.

But a more nuanced reading of Trinidad’s informal Constitution suggests that the prime ministerial term limits amendment may require special supermajorities of both houses of Parliament as well as the assent of the president, or perhaps even more.

Formal Amendment in Trinidad

The Trinidadian Constitution creates three formal amendment thresholds:

  1. Three-fourths agreement in the House and two-thirds agreement in the Senate plus the assent of the president for changes to important provisions such as the Constitution’s formal amendment rules, specified presidential powers and eligibility, and certain parliamentary powers;[8]
  2. Two-thirds agreement in each house of Parliament and the assent of the president for changes to provisions including human rights and freedoms, citizenship, and certain parliamentary powers;[9]
  3. Simple majority agreement in each house of Parliament, plus the assent of the president, for all changes not specifically assigned to one of the two higher amendment thresholds.[10]

In a recent paper, I have described this structure of formal amendment rules as escalating. It is escalating because it creates multiple amendment thresholds of increasing difficulty, and in turn identifies which constitutional provisions and underlying political commitments are deserving of heightened constitutional entrenchment protection relative to other provisions and commitments, at least according to the constitution’s authors.

Amending Unwritten Constitutional Principles

Trinidad’s formal amendment rules demonstrate a major weakness of this escalating structure of formal amendment: while it is possible to assign written constitutional provisions to specified amendment thresholds, unwritten constitutional principles are not usually identified in the constitutional text as amendable by any particular amendment threshold. This leaves to political actors the task of choosing which amendment threshold is most appropriate when they pursue an amendment to an unwritten constitutional principle. Acting in their own self-interest, political actors will likely choose the least onerous amendment threshold as the one that applies to the constitutional change they wish to bring to a given constitutional principle.

This is understandable. Political actors would gain little and risk a lot by assigning the highest amendment threshold to their amendment of an unwritten constitutional principle. It is the better political strategy to take a formalist view of the constitution by invoking the lowest or default amendment threshold as the one that applies to amending an unwritten constitutional principle than to take the magnanimous, constitutionally correct, and risky route of making the amendment susceptible to defeat under the highest amendment threshold.

The Government’s Strategy

The government in Trinidad appears to have taken the strategically expedient but constitutionally problematic route. It has taken the view that the introduction of prime ministerial term limits is by default made subject to the lowest amendment threshold because it is not otherwise specified as amendable by a higher threshold.

This raises a problem because introducing prime ministerial term limits implicates fundamental constitutional principles that are not entrenched in the constitutional text. It is therefore not possible to point to a particular formal amendment threshold to which the principles are assigned. They are not written in the Trinidadian Constitution. But they nonetheless form part of it. These principles are inherited from the Westminster traditions of responsible government, Royal prerogative (which now rests with the president), and executive discretion. Prime ministerial term limits amend each of these unwritten constitutional principles.

Introducing prime ministerial term limits alters responsible government in Trinidad by narrowing the leadership options available to the majority or governing party. It moreover prevents the president from exercising the prerogative to appoint a person who may be best positioned to form a government, precisely by denying the president the choice of the term-limited prime minister. This in turn removes from the president the discretion that the executive head of state exercises by convention in Commonwealth democracies.

There is nothing inherently problematic with prime ministerial term limits. But if they are to be introduced in a constitutionally sound manner in Trinidad, political actors should legitimate the amendment only with democratic procedures that reflect its gravity. A change so fundamental to the unwritten constitutional principles of government in Trinidad should arguably be made with more than a simple majority.

Amendment and Revision

Some scholars distinguish in this respect between an amendment and a revision.[11] An amendment affects a change to the constitution within the existing framework of government whereas a revision entails a more fundamental shift from the presuppositions of the existing regime. On this theoretical distinction, revisions require more participatory processes than amendments in order to highlight for citizens the importance of the change being proposed and the likely significant transformation in political culture the change will visit upon the constitutional regime.

In Trinidad, the introduction of prime ministerial term limits amends in a material way at least three fundamental constitutional principles. The amendment will have the kind of impact on constitutional politics that is commonly associated with a revision, not a mere amendment (though, admittedly, a “mere” amendment can have a significant impact, as we know from the American or German experience, for example). Yet the importance of these principles is not reflected in the way the Constitution (Amendment) Bill, 2014 proposes to amend them—by a simple majority.

The onus is on the Trinidadian government to introduce and ratify the prime ministerial term limits amendment with a higher-than-normal threshold. The problem is of course that the government has no incentive to risk defeat on the amendment. Why should it call for a higher supermajority threshold when it can pass the amendment easily by simple majority with its existing parliamentary majority?

Political expediency counsels against that choice. What is more, the distinction between amendment and revision—not to mention the idea of amending an unwritten constitutional principle—matters much less to political actors engaged in real constitutional politics than to scholars who trade mostly in theory.

At the very least, however, the potential unconstitutionality of prime ministerial term limits should give reason for pause, even more so given that the governing party in Trinidad seems intent on entrenching this constitutional amendment as rapidly as possible.

Three Additional Thoughts on the Amendment Bill

Three additional observations seem relevant by way of conclusion.

First, the proposals in the Constitution (Amendment) Bill, 2014 are not entirely new, nor should local political actors and observers feel blindsided by the government’s decision to introduce them in this bill. All of the proposals in some way derive from the broad objectives or the specific proposals outlined in the government-sponsored Constitution Reform Commission’s December 2013 recommendations for modernizing the Constitution.

Second, each of the three proposed changes may prove transformative of Trinidadian political culture if the bill ultimately becomes law. But it is hard to project the long-term consequences of these changes today. The eventual effect of this bill is therefore like throwing up a “jump ball reform,” a term used by John Whyte, the great Canadian constitutional theorist, to describe recent efforts to amend the Senate.

Third, these proposed changes are fascinating inasmuch as they show that conventional features of presidentialism may be fused innovatively with conventional features of parliamentarism to create new hybrid forms of governance that may (or may not) improve democratic outcomes. I have discussed in much greater detail elsewhere how modern constitutional democracies have increasingly fused presidentialism and parliamentarism in interesting ways that have forced us to reconsider whether the rigid constitutional structures of presidentialism and parliamentarism are as different as each other, and as incompatible with each other, as scholars have long tended to presuppose. These recent proposals in Trinidad show that the trend persists, for better or worse.

Suggested Citation: Richard Albert, An Unconstitutional Amendment in Trinidad & Tobago?, Int’l J. Const. L. Blog, Aug. 14, 2014, available at:

[1] Constitution (Amendment) Bill, 2014, ats. 8, available at:

[2] Id. at ss. 4-7.

[3] Id. at s. 9.

[4] Id. at s. 8.

[5] Id. at s. 6.

[6] Id.

[7] Id. at s. 9.

[8] Constitution of the Republic of Trinidad, Ch. 4, Pt. II, s. 54(3).

[9] Id. at s. 54(2).

[10] Id. at ss. 53, 54(1).

[11] See, e.g., John Rawls, Political Liberalism 238 (2d ed. 1996); Thomas M. Cooley, The Power to Amend the Federal Constitution, 2 Mich. L.J. 109, 118 (1893).


9 responses to “An Unconstitutional Constitutional Amendment in Trinidad & Tobago?”

  1. Andrew Arato Avatar
    Andrew Arato

    Dear Richard:

    I think there is a serious flaw in your argument, in your otherwise highly interesting and sophisticated blog entry.

    What I called a multi-track amendment structure [Constellations v. 18 # 3: #), what you call an “escalating” one, does not suffer from the difficulty you mention, namely that principles and conventions that are unwritten necessarily fall outside this structure. Therefore you claim political actors will choose the lowest level (in Trinidad simple majorities) to amend what may be important and even most important.

    That second point may be right, but your conclusion does not follow. Legislatures or governments after all often try to amend a constitution by passing simple statutes. Nothing new in that, and the phenomenon is not caused in any way by a formal multi track structure.. But, a multi track structure, by definition establishes the possibility of the constitutional review of amendments. After all who is to ever say that the right level of difficulty was used in a given case? A court under such a structure can certainly argue that constitutional principles belong to the most difficult level, and indeed to a level higher, namely what has been called the basic structure of the constitution. By your lights, the Indian Supreme court, by the way under a multi track amendment structure, never could have articulated the basic structure doctrine. But it did, and was right to do so, as subsequent history showed.

    This is the first time I have read the Trinidad-Tobago rule. But even with this liability, and the second one of not being a lawyer, I can see that the phrase within the first, ¾ rule, namely “changes to important provisions such as [italics mine] the Constitution’s formal amendment rules, specified presidential powers…etc.” arguably includes all important constitutional principles. A Court could certainly argue that, and I hope it would. I also think that the phrase under the same rule, namely “certain parliamentary powers” clearly can include if a court wishes, term limits, that obviously touches on the powers of parliament to elect a prime minister.

    Whether or not Trinidad’s supreme court ever invalidates statutes, and especially constitutional amendments I do not happen to know. It is perhaps time that it should begin to do so. Certainly the multi track or “escalating” amendment structure logically requires such a review.

    I will now read your longer, interesting sounding study, on the Expressive Function… and apologize if I misrepresented or misunderstood the point of that study. But it is you who seem to be drawing on it in making the argument that I find incorrect.

  2. Gerald Ackermann Avatar
    Gerald Ackermann

    Great blog entry! Richard, as someone currently on the ground in Trinidad and Tobago, the main outcry to the Constitution Amendment Bill is really with the lack of consultation on the run-off provision. Also, the potential within those 15 days for political manipulation by the incumbent government is a cause of concern.

    Just for your information Andrew, the High Court of T&T does in fact have judicial review powers similar to that of the U.S. Supreme Court’s powers, and can strike down legislation which is inconsistent with the Constitution of Trinidad and Tobago.

  3. Richard Albert Avatar

    Andrew, thanks for taking the time to make this important comment. Let me just clarify the point to which you’ve kindly responded. I don’t think the clarification will bring us into agreement, but I would like to leave less doubt about my observation.

    First, though, I should note that I think escalating amendment rules are an excellent design for constitutional democracies, as I argued years ago in a paper on “Constitutional Handcuffs.” (Were I to write this paper today, I would have changed the tone and approach, but I would have reached the very same conclusion.)

    In my post above, I observed that one difficulty with escalating amendment thresholds is that they sometimes leave the applicable amendment rule open to contestation and potentially exploitation: they will expressly require certain constitutional provisions to be amended by specific amendment thresholds but they cannot expressly assign unwritten constitutional norms to a specified amendment threshold–precisely because those unwritten norms are not textually entrenched.

    So, for example, in a constitutional text with 90 provisions and three escalating amendment rules (A, B, C), provisions 1-30 might be expressly designated as amendable by only A, 31-60 by only B, and 61-90 by only C. (This is a very crude illustration.)

    But unwritten norms like responsible government or the expectation that a prime minister will hold an elected seat in the lower house will not be expressly designated as amendable by either A, B or C because those norms are conventional.

    From this observation, I posited in my post that self-interested political actors constrained by escalating amendment rules would take a formalist approach to constitutional amendment, and try to amend these unwritten norms by simple statute on the theory that the non-entrenchment of these unwritten norms means that these norms are not subject to the rules of formal amendment and therefore are within the power of the legislature to change.

    I do not believe that the escalating structure of formal amendment is incompatible with simple legislative change. Nor do I believe that the escalating structure of formal amendment is incompatible with judicial review.

    I do believe, however, that the escalating structure of formal amendment does not generally specify a formal amendment rule for principles, though it does so quite effectively for provisions. (Escalating amendment rules *could* specify amendment rules for principles, as some escalating amendment thresholds do in constitutional democracies but certainly not most.) And this leads to serious problems for predictability, stability and clarity, all three important values underpinning the rule of law: which amendment rule should apply?

    Canada’s current debate about Senate reform illustrates this particular problem with escalating amendment rules.

  4. Richard Albert Avatar

    Gerald, thank you for reading and for your comment. I agree with you that the run-off election raises questions. I’ll give this more thought. In the meantime, please feel free to post relevant web links in this comment section.

  5. Gerald Ackermann Avatar
    Gerald Ackermann

    Professor Albert, you are most welcome! Just to provide a bit more context on this controversy, the provisions dealing with term limits and right to recall have been very much the subject of consultation by the Constitution Reform Commission. I should qualify my statement above by noting that the Opposition has opposed the Amendment in its entirety. However, consultation was not held by the CRC on the provisions dealing with the run-off election:


    I totally agree with your statement that such far reaching amendment should not be done by simple majority. However, as you rightly pointed out such amendment was prescribed by the Republican Constitution to be done by simple majority.

    See below commentary by two of our leading constitutional practitioners:

  6. Richard Albert Avatar

    Gerald, thank you for these weblinks!

    One question: I see that there has been a legal challenge to the constitutional amendment. In your view, what are the odds for success here?

  7. Tracy Robinson Avatar
    Tracy Robinson

    Thanks for bringing these important issues of constitutional reform in the Caribbean to the blog. I want to take up your suggestion that the royal prerogative is ‘unwritten constitutional principle’ in Caribbean constitutions. I accept the concern about unwritten constitutional norms and the amendment process, but would see the difficulties more in relation to norms like separation of powers, judicial independence and the rule of law, which have been treated as strong implied norms in Caribbean constitutions. I am less persuaded about the royal prerogative.

    Historically Caribbean constitutions have codified aspects of the prerogative power in the constitutions along with constitutional conventions associated with them (See Margaret Demerieux, ‘The Codification of Constitutional Conventions in the Commonwealth Caribbean Constitutions, (1982) 31 ICLQ 263). In the same way that legislation codifying a prerogative power governs the scope of the power, it is generally accepted that where Caribbean constitutions codify what was a prerogative power, it is the constitution that determines the extent of the power. It may well be that in understanding the nature of the power and its background, its original source is relevant but not decisive.

    So to the question, what is the extent of the power residing in the president to appoint a prime minister, I would say the answer is in the text of the Constitution (s 76) and not in an unwritten constitutional principle of royal prerogative. In fact I would say prerogatives are more of a power, recognised at common law and in constitutions, than a principle.

    That said, I think prerogative powers still exist beyond those codified by the constitutional text. The Trinidad and Tobago Constitution contemplates this. Here I would argue that they exist as part of the common law, to the extent they don’t conflict with the provisions of the constitutions. Such prerogatives could include the right of the executive to make treaties or the power of the head of state to confer honours (see Sanatan Dharma Maha Sabha v AG). I don’t think there is any reason why such prerogative powers, as exist at common law, cannot be circumscribed by an ordinary law, and indeed many have.

    In sum, I don’t see the royal prerogative as a fundamental unwritten constitutional principle that is entrenched by implication in the Trinidad and Tobago Constitution. As a source of the constitution it is either captured and limited by the explicit provisions of the constitution, as in section 76 here, or contemplated as part of the common law and open to change by ordinary legislation. Since not all principles and powers that have constitutional significance are ‘entrenched’.

  8. Richard Albert Avatar

    Tracy, thank you for this helpful critique. I have been giving this some thought over the past couple of days. I am persuaded by your point that textual entrenchment alters the form and perhaps as a consequence the scope of the prerogative, in addition to the appropriate methods by which we can bring changes to the prerogative (or, more accurately, to what was originally an unwritten prerogative power, but has since been textually entrenched).

    But I wonder whether we might not instead construe the entrenchment of (what was then) the prerogative as a kind of placeholder, that is to say, as a marker of a power or a family of powers held by a political actor–powers whose full range cannot be described exhaustively in the text but for which we must instead turn to inherited traditions or prior practice to understand what those powers entail.

Leave a Reply

Your email address will not be published. Required fields are marked *