“Constitutional conventions are rules of political practice which are regarded as binding by those to whom they apply, but which are not laws as they are not enforced by the Courts or by the House of Parliament.”

Constitutional conventions

“Constitutional conventions are rules of political practice which are regarded as binding by those to whom they apply, but which are not laws as they are not enforced by the Courts or by the House of Parliament.”


Constitutional conventions are suggests a radical re-alignment in favour of the regions, local communities and the individual.[1] Constitutions are about resolving the struggle of rival contenders of power. Constitutional law is about the rules and arrangements a country makes concerning the nature, structure, and powers of its organs of government and the relationship between the citizen and the state. In the narrow sense, a Constitution consists of a document or set of documents which are intentionally drafted to form the fundamental law of the state. A constitution is a set of fundamental principles or established precedents according to which a state or other organization is governed. Constitutional conventions are not, and cannot be, enforced by courts of law. To these proposed measures must be added proposals for the radical reform of local government.[2]

This definition concentrates on what conventions are supposed to achieve. However, this view is not entirely accurate and it is important that conventions are distinguished from habits and practices. Conventions are different from habits or practices in that these concepts do not prescribe or dictate what ought to happen but are merely descriptive of what in fact does happen. A further definition of the purpose of conventions was given by Sir Ivor Jennings as, “The short explanation of the constitutional conventions is that they provide the flesh that clothes the dry bones of the law; they make the legal constitution work; they keep it in touch with the growth of ideas.”


A common definition of constitutional conventions is: “By convention is meant a binding rule, a rule of behaviour accepted as obligatory by those concerned in the working of the onstitution.”[3]A Further definition of the purpose of conventions was given by Sir Ivor Jennings as:

“The short explanation of the constitutional conventions is that they provide the flesh that clothes the dry bones of the law; they make the legal constitution work; they keep it in touch with the growth of ideas.”The definition offered by Sir Ivor Jennings is widely accepted.[4] He stated that the fulfilment of three tests isnecessary to identify a convention.[5] First, are there any precedents; secondly, do those operating the constitution believe that they are bound by a rule? thirdly, is there a constitutional reason for the convention? This has been accepted by the Canadian courts.[6]Constitutional conventions arise when the exercise of a certain type of power, which is not prohibited by law, arouses such opposition that it becomes impossible, on future occasions, to engage in further exercises of this power.Constitutions may be classified as written or unwritten, flexible or rigid, monarchical or republican, federal or unitary, supreme or subordinate to the legislature, based on the separation of powers.The British Constitution is unwritten, flexible, monarchical (at least in its legal form), unitary, subordinate to the legislature, based on a partial or limited degree of separation of powers. To investigate the proposition, ‘there is no clear agreement about what amounts to a constitutional convention’, firstly, will require a look at the different definitions of what amounts to a constitutional convention, and to discuss their function or purpose. Constitutional conventions will be possible to discuss or not it would be possible or even useful, to change constitutional conventions into legally binding rules. Most constitutions also attempt to define the relationship between individuals and the state, and to establish the broad rights of individual citizens. The term was first used by British legal scholar A. V. Dicey in his 1883 book, Introduction to the Study of the Law of the Constitution. Dicey wrote that in Britain, the actions of political actors and institutions are governed by two parallel and complementary sets of rules:

‘The one set of rules are in the strictest sense “laws”, since they are rules which are enforced by the courts…’[7]

Relation between Law and Constitutional Conventions

Law and constitutional conventions are closely related. Constitutional conventions presuppose the existence of a legal framework, and do not exist in a legal vacuum. For example, in the UK, the constitutional conventions of forming a Cabinet presuppose the laws relating to the Queen’s royal prerogative, the office and powers of Ministers, and the composition of Parliament.[8] A difference between law and constitutional conventions is that laws are enforceable by the courts, but constitutional conventions are not enforced by the courts. If there is a conflict between constitutional conventions and law, the courts must enforce the law. In some countries, such as the United Kingdom (UK) and Canada, the courts acknowledge the existence of constitutional conventions as aids to judicial interpretation.[9] Academics consider that such acknowledgement may at times appear to be similar to enforcement.[10]

Functions of Constitutional Conventions

First it is necessary to acknowledge the notorious absence of agreement amongst writers of authority about what constitutes a convention.[11] Conventions, as distinguished from mere habits, practices or usages, are generally accepted as prescriptive and mandate certain behaviour as either constitutionally desirable or necessary.

Every political community, and thus every state, has a constitution, at least insofar as it operates its important institutions according to some fundamental body of rules. By this conception of the term, the only conceivable alternative to a constitution is a condition of anarchy. Nevertheless, the form a constitution may take varies considerably. Constitutions may be written or unwritten, codified or uncodified, and complex or simple, and they may provide for vastly different patterns of governance.

Conventions serve a number of purposes not all of which are easy companions of democratic and representative government. At one level it is possible to assert that conventions rather than laws have provided an important evolutionary mechanism for the progress from a monarchical to a democratic constitution without the need for a series of divisive statutes repealing the sovereign’s legal powers.[12]Others, including Horwitz, for instance, have argued that conventions were developed as undemocratic devices to reassure the ruling class that constitutional fundamentals would continue to be developed within government largely beyond the influence of the rising middleclasses following rapid extension of the franchise after the Reform Act 1867.[13] Conventions, according to this interpretation, fortify a class barricade against the triumph of democracy.

If the purpose of a constitution is to impose external controls on government, then conventions which are generated within government must invite concern. As we have seen, conventions may be established, abandoned or altered at will by the Executive, often without reference to Parliament. Even those conventions which have the fundamental role of establishing key parts of the constitution of responsible government are not immune from subtle redefinition. Concerns are aroused when such a re-formulation dilutes the obligations of those advocating the change.

There is a problem that conventions obscure the constitution. This raises three issues. The first of these is that conventions offer a conception of the constitution that assumes that change and flux are continuous and permanent. At any one time it would be impossible for the scholar to identify which practices have constitutional status as some are falling into desuetude whilst others are emerging.[14] This means that there is always a debate about which practices are constitutional.

Second, where fundamental conventional rules are acknowledged, their scope and application is subject to debate. This raises the pragmatic problem that in the fog and smoke of a dynamic system the curious surveyor cannot fully delineate the squares and avenues of the constitution, and that this can only properly be the task entrusted to the retrospective gaze of the historian.[15]

The third pragmatic reason why the mystery of the constitution will be impenetrable to contemporary study. This relates to the enduring endemic secrecy of government and its mistrust of open debate. It is impossible for the citizen to know what the constitution is, for the answer lies in a file of precedents held in the private office of the Cabinet Secretary.[16] Presumably only the initiates of the civil service and ministers are admitted to behold the contents of the file, and the application of the precedents it contains lies within their exclusive judgment. It is understood that it is not made available to Parliament.

The “Rule of Law” Administrative State

Civil servants are entrusted with public authority in order to implement the policy agenda of the political executive. It is a postulate of our constitution, however, that all public authority must adhere to the rule of law.[17] This would be a straightforward constraint on government action but for the fact that the rule of law is a deeply contested notion.[18] While it has been recognized as an underlying constitutional principle,[19] and as the animating principle for the judicial review of administrative action,[20] and is mentioned alongside the supremacy of God in the preamble to the Charter of Rights, the rule of law remains surprisingly unexplored as a constitutional norm by courts in Canada. In the Secession Reference, the Supreme Court described the importance of the rule of law in terms of subjecting executive authority to legal accountability and protecting citizens from arbitrary state action:

The principles of constitutionalism and the rule of law lie at the root of our system of government. The rule of law, as observed in Roncarelli v. Duplessis, [1959] S.C.R. 121, at p. 142, is “a fundamental postulate of our constitutional structure”. As we noted in the Patriation Reference, supra, at pp. 805?6, “[t]he ‘rule of law’ is a highly textured expression, importing many things which are beyond the need of these reasons to explore but conveying, for example, a sense of orderliness, of subjection to known legal rules and of executive accountability to legal authority”. At its most basic level, the rule of law vouchsafes to the citizens and residents of the country a stable, predictable and ordered society in which to conduct their affairs. It provides a shield for individuals from arbitrary state action.[21]

The executive accountability to legal authority referred to in this passage is accomplished by another constitutional postulate – all executive authority is subject to judicial review on the grounds that the rule of law has been contravened, even if a statutory privative or preclusive clause purports to immunize an administrative official from review completely.[22]


Conventions are prescriptive of constitutional behaviour, but there is neither a body to declare their existence, nor to interpret them nor to elucidate their meaning, nor to impose a sanction for their breach. They are created by political insiders and operated within the exigencies and limitations of adversarial politics. Conventions may be adapted to suit the interests of the government of the day. They are the “handmaiden” of the party in power[23] yet they are the means by which an unsatisfactory and hopelessly archaic formal constitution has been adapted to modern political values. They can be re-cast by those whom they are supposed to control, deleted when they become inconvenient, and ignored when they are embarrassing. They are but chaff before the heavy armour of a government with a large parliamentary majority. But any attempt comprehensively to juridify them would ultimately fail, and attempts at codification, even if practicable, would tend to fossilise a system in which fluid mutability is its chief virtue as much as it is its frailty. And crystallisation must be a limited venture within current conceptions of the role of the judiciary within the separation of powers. The buttressing in the nineteenth century of the undemocratic control of a major lever of constitutional change as a conservative reaction to the extension of the franchise requires review. This is particularly timely now that Labour appears to place an increased emphasis upon addressing the democratic deficit in the UK constitution. The UK has just crossed the threshold of major constitutional re-alignment following devolution. This will provide a major accelerant to the growth of new conventions to regulate and relationship between Westminster and the devolved institutions. Accordingly it has been argued that within a system of accountable government the democratic supervision of conventions could be achieved by Parliament. There should be a more reflective, inclusive and transparent deliberation on the evolution and application of conventions. It has been proposed that a new and distinct select committee be established which will scrutinise the development, interpretation and abandonment of conventions. It will be advised by a panel of experts and will make an annual report to Parliament. This will provide an authoritative body whose work will have the effect of recognising and declaring conventions and exposing those who breach them to public obloquy.

The merits of this proposal are that parliamentary scrutiny is established without sacrificing the flexibility of the constitution and thus its ability to accommodate change. The Committee will have no power to impose a sanction but whilst it recognises that conventions are prescriptive it also meets the objection that they must be capable of evolution. At its heart is the principle that constitutionally significant behaviour must be justified to Parliament. This would be achieved in a more systematic and reflective manner than at present. If this were achieved it would significantly contribute to New Labour’s “People’s Constitution”.

[1] Labour first committed itself to this policy in Meet the Challenge: Make the Change, Labour Party, 1989.

[2] White Paper, Modern Local Government: In Touch with the People Cm 4014.

[3] Kenneth Wheare, Modern Constitutions, 1951, p.179, quoted in Marshall, Constitutional Conventions, p. 7.

[4] But has been subject to criticism, see e.g., J. Jaconelli, The Nature of Constitutional Conventions

(1999) 19 Legal Studies 24.

[5] Law and the Constitution, 5th edition, 1959, p.136

[6] Reference re Amendment of the Constitution of Canada (Nos 1, 2 and 3) (1982) 125 DLR (3d) 1.

[7] AV Dicey, Introduction to the Study of the Law of the Constitution, 10th edition, pp. 23-24.

[8] For a detailed discussion on these legal matters, please refer to Chau Pak-kwan and Cheung Wailam, Process of Appointment of Senior Members of Government in Selected Countries, HK: Legislative Council Secretariat, 2002, Part 2.

[9] For relevant cases, see Att.-Gen. v Jonathan Cape Ltd [1976] Q. B. 752 and RE Amendment of the Constitution of Canada [1983] A. C. 394, HL.

[10] For an example, see John F. McEldowney, Public Law, 2nd ed., London: Sweet & Maxwell, 1998, p. 113.

[11] See generally Munro, Studies in Constitutional Law, Butterworths 1987, and Alder, Constitutional

and Administrative Law 3rd edition, Macmillan, 1999.

[12] A W Bradley and K D Ewing, Constitutional Law, 12th edition, (Longman: 1997) at p. 24.

[13] Morton J Horwitz, Why is Anglo-American Jurisprudence Unhistorical? [1997] OJLS 551.

[14] Baldwin quoted in Ivor Jennings, Cabinet Government, and Cambridge University Press 1936, p.12.

[15] This may be so, but it is equally an issue where some laws are concerned. However, the courts

Provide a mechanism for the interpretation of legal rules. Parliament may deliberately enact laws in

which the meaning of statutory words is unclear. This relies on the courts to supply meaning. For

example, many questions surrounding the possible meaning of working time in the Working Time

Regulations 1998 SI 1998/1833 has deliberately been left to judiciary to determine. The absence of

such an institution means that conventions are liable to be interpreted by those who are bound by the

rule itself. If a sufficient consensus exists an inconvenient rule can simply be ignored and the

Constitutional obligation itself destroyed.

[16] Hennessy, loc.cit. at p.45-6.

[17] Roncarelli v. Duplessis, [1959] S.C.R. 121 at 142. In that case, Rand J. stated that “there is always a perspective within which a statute is intended to operate” (at 140). In other words, every grant of statutory authority has an implied limitation which restricts its exercise to proper and not improper purposes, in good faith and not in bad faith, and based on reasoned and not arbitrary or discriminatory factors. For discussion, see D. J. M. Brown and J. M. Evans, Judicial Review of Administrative Action in Canada (loose-leaf ed.), at para. 13:1221.

[18] For recent appraisals, see A. Hutchinson, “The Rule of Law Revisited: Democracy and the Courts” in Recrafting the Rule of Law, D. Dyzenhaus, ed. (London: Oxford University Press, 1999), pp. 196-226 J. Jowell, “Beyond the Rule of Law: Towards Constitutional Judicial Review” [2000] Public Law 671; and T.R.S. Allan, “The Rule of Law as the Rule of Reason: Consent and Constitutionalism” (1999) 115 L.Q.R. 221.

[19] D. Mullan, “The Role of Underlying Constitutional Principles in a Bill of Rights World” (2003) NZLR (forthcoming)

[20] See Baker at paras. 53, 56.

[21] Secession Reference, supra note   , at para. 70.

[22] Crevier v. Attorney General of Quebec, [1981] 2 S.C.R. 220. See also M. Elliot, The Constitutional Foundations of Judicial Review (Oxford: Hart Publishing, 2001).

[23] This follows a fortiori from Brazier, loc. cit. who uses this term in relation to the Constitution