Constitutional Conventions as a Model

The examples drawn from the UK and the Netherlands mainly serve to teach us two things. First, the absence of either an eternity clause or a judicial ‘basic structure doctrine’ in both systems does not imply that amendment of the constitution cannot, in practice, be contested.

We think this model may be found in the concept of constitutional conventions. Rather than a clear rule, whether written or judge-made, that a particular consti­tutional concept is unamendable, covert unamendability triggers the emergence of a particular convention that the legislature will not abolish or significantly alter a given constitutional arrangement. The enforcement of such a convention—if it may be called so—is, as we have seen, a matter of disobedience on the part of the other branches of government, notably the courts. In order to gain a deeper understanding of this kind of convention, let us first turn to an inquiry in the nature of constitu­tional conventions in general.

6.1 The Nature of Constitutional Conventions

In politically enforced constitutions, the final word about the application of a particular constitutional provision is not given by the courts, but rather by the political branches, i.e. the legislature and the executive. The power of these bran­ches are primarily curbed in political sense. That is not to say that they will enact any legislation whatever. On the contrary, a political constitution only works well if the legislature adheres to certain self-imposed moral and political constraints, of which many of them are regarded as constitutional conventions.^[496] They originate from the constitutional doctrine of the UK, in which context Dicey has referred to them as ‘the conventions, understandings, habits or practices’ regulating the conduct of the several members of the sovereign power and complementing the legal rules of the constitution.^[497] The existence of a convention is usually deter­mined by the application of the Jennings test.^[498] According to this test, three questions have to be answered:

First, what are the precedents; secondly, did the actors in the precedents believe that they were bound by a rule; and thirdly, is there a reason for the rule?^[499]

Constitutional conventions are thus not mere precedents, nor mere rules, but a combination of both. A mere practice of a particular office holder can develop into a convention from the moment that others increasingly expect him to act in con­formity with this practice.^[500] The expectation that an institution will continue to follow a practice may have different causes, varying inter alia from a settlement about the interpretation of a constitutional principle which two rivals in the con­stitution may have reached, to a simple need for efficiency in the political process. In this sense, conventions can be compared to the lithograph ‘Drawing hands’ (1948) by the Dutch artist M.C. Escher, which depicts two hands that draw each other since the precedents simultaneously create the reason for the convention and vice versa.^[501] The coexistence of the precedents and the reasons subsequently create the belief among the responsible institutions are bound by the convention, which is the final component of these social rules.

Constitutional conventions issue a moral dimension to politically enforced constitutions by subjecting the legal supremacy of the political actors to extra-legal (i.e. moral and political) constraints.^[503] It is by means of convention, for instance, that the UK Parliament will never pass the ‘Abolition of the Democracy Act’ or the ‘Murder of Blue Babies Act’.^[504] Political constitutions can thus only be properly understood if both the legal rules and the applicable conventions surrounding the constitution are also taken into account.

The conventions of the constitution give voice to the way in which institutions deploy their power in practice. Constitutions attribute different tasks and objectives to the different institutions of the state, which are likely to result in tension or disagreement.^[505] In this constellation, every institution aims to maintain the authority which it has acquired or may even compete for more power with other institutions.^[506] The relationship within and between institutions can, therefore, be seen as a ‘domain of continuing tensions’, which sometimes may slowly evolve in ‘a process of mutual accommodation’.^[507] Conventions are the reflection of the settlements which one or more institutions may have reached. The ongoing com­petition within and between institutions is not regulated by fixed rules. Neither has a pre-eminent body been appointed to settle disputes between institutions. Disputes on the basis of constitutional conventions thus have to be settled in practice. The resulting settlements will necessarily be only temporary since institutions can call each resolution for a constitutional problem continuously into question.^[508]

As opposed to law, conventions appear in a great variety of normative standards with a similarly diverging binding force.

In default of a body that could settle disputes between institutions, conventions are merely ‘self-policing’. The binding force of a convention particularly depends on the inherent values of a convention, which are inherently linked to the ideals and values of a particular constitutional tradition.^[511] As was seen before, conventions are based on often conflicting principles as democracy, the rule of law or the independence of the judiciary.^[512] The breach of a convention will thus raise a question of principle. Whether the breach of a convention will be regarded as unconstitutional depends on the actual content of the convention involved and on the constitutional value at stake. A breach that can be considered unconstitutional will usually be uttered in the language of ‘wrong-doing, of moving the goal posts, of not playing the game, of failing to respect established values, and so on’.^[513] Breaches of conventions which do not have to be couched in this terminology may be allowed, for instance, because they preserve another constitutional value, or may point at the end of the operation of a convention.^[514] Institutions may be willing to respect those values or not.

It is not always simple to determine whether a convention exists.^[517] This raises the question whether how much consensus is needed in order to determine that a convention actually exists.^[518] Even the (constitutional) legislator does not have a decisive voice about how disagreement between institutions in the political context should be resolved.^[519] Admittedly, a statute or a Constitution might stipulate a sort of settlement for a conflict within an institution or between institutions. However, the actual effect of such legislation remains ultimately dependent on its application by the institutions involved. Moreover, conventions are not dependent on legal structures. They can thus be adjusted after a change with regard to the political and/ or normative views of the political institutions involved. The only requirement for change is thus, as opposed to legal or constitutional change, the consent of the institutions involved.

6.2 Constitutional Conventions and Covert Constitutional Unamendability

What does all this teach us with regard to the existence of a possible constitutional convention of unamendability? First that any principle of unamendability in the politically enforced constitutions such as we have identified is not of a binding nature.

As we have seen, ignoring a convention by one institution may induce another institution to ignore some other convention. In the case of court-legislature dia­logue, this will usually be the conventions that regulate the proper judicial role in the framework of the separation of powers. An unconventional role by the legis­lature thus provokes an unconventional judicial role. However, this role is—again —subject to powerful political constraints. The courts, even if they step outside their traditional role, cannot do anything. Thus, the end result of a court-legislature dialogue provoked by the violation of a convention of unamendability will usually end in a new settlement, somewhere in the middle. The flexible nature of con­ventions then enables all three branches of government to maintain that the new settlement was part of the original convention all along.

Last but not least, the nature of conventions as political and flexible norms on the functioning of institutions, renders the decision to enforce (or not to enforce) a convention of unamendability essentially of a political nature. It is ultimately a judicial assessment of the political gains and losses institutionally, that determines whether the courts will interfere with the legislative decision to amend. In a way, the same kind of balancing will probably take place in the ‘regular types’ of unamendability review. The Indian or Pakistani Supreme Courts will, no doubt, have had their thoughts on the prudence, in terms of institutional politics, of the decision whether not to interfere with the constitutional amendments they were confronted with. What makes the convention-type of unamendability stand apart though is that this kind of assessment is not a by-product of the judicial decision, but a matter of principle connected to the idea that the convention (and its enforcement) is a more political rather than a legal rule.

7