Written and Unwritten Constitutions: The Modality of Change

Paul Craig

1 Introduction

Written constitutions are the norm, unwritten constitutions the rare exception. There is much

that can be said as to the commonalities and differences between them. There are, nonetheless,

two principal variables that underscore the respective forms of constitution, which in turn

impact on the modality of change: the extent to which the constitution is committed to writing,

and the extent which the constitution is perceived to be separate from, and superior to, ordinary

law.

These twin variables are commonly perceived to be linked. There is empirical warrant

for this assumption. The paradigmatic written constitution, which contains the principal

constitutional norms in a single document, will commonly also enshrine, expressly or

impliedly, constitutional review through which courts subject ordinary statute to review for

constitutional compliance; the constitutional norms of the paradigmatic unwritten constitution

will normally be found in a plethora of sources, and ordinary statute will customarily have

pride of place in the constitutional schema.

There is, however, no a priori normative reason why these twin variables should be

linked in this manner. The embodiment of constitutional norms in a canonical document does

not predetermine the content of the constitution, nor does it impel the conclusion that

constitutional review exercised by a court must be part of the resulting schema. This is

exemplified, as will be seen below, by the UK, where some suggestions for the shift to a written


Emeritus Professor of English Law, St John’s College, Oxford.

Electronic copy available at: https://ssrn.com/abstract=3897906

constitution are premised on the assumption that parliamentary sovereignty remains a

cornerstone of the constitutional ordering. It is, by parity of reason, equally possible in principle

to think of an unwritten constitution, the central features of which are markedly similar to those

of the paradigmatic written constitution, such that it is accepted by all relevant players that

ordinary legislative power is bounded by certain structural and rights-based precepts that are

enforced, inter alia, by the courts. The manner through which this might occur will be

considered more fully below.

It follows from the foregoing that while the extent to which constitutional norms are

committed to writing is of significance in distinguishing the two forms of constitution, the

second variable is more important in normative terms, and shapes the way in which change is

effectuated. The contrast in this respect is sharp. The foundational starting point of the standard

written constitution is the distinction between law-making through ordinary politics and the

constitution; the latter circumscribes the former, such that the terrain of ordinary politics is

bounded by the constitution. The foundational starting point of the UK unwritten constitution

is that parliamentary sovereignty is the principal constitutional principle; the terrain of ordinary

politics is boundless and parliamentary sovereignty instantiates and elevates ordinary statute

as the vehicle for its delivery.

This chapter in honour of Bruce Harris explores the foregoing ideas and their

implication for the modality of constitutional change. Issues of constitutional change within

the unwritten constitution have been one of the main themes in Harris’ scholarship.1

The discussion of written constitutions begins with certain preliminaries as to the

origins of the modern constitutional form. It is followed by consideration of constitutional

1
See, eg, BV Harris, ‘The Constitutional Future of New Zealand’ [2004] NZLR 269; BV Harris, ‘Constitutional

Change’, in R Miller (ed) New Zealand Government and Politics (4th ed, Melbourne, Oxford University Press,

2006) 115.

Electronic copy available at: https://ssrn.com/abstract=3897906

content, distinguishing between the horizontal, vertical and structural features generally found

in written constitutions, and includes discussion of over- and under-constitutionalization. The

analysis concludes with brief reflection on the manner of constitutional change.

The focus then turns to the UK unwritten constitution. It begins with certain

preliminaries as to historical origins. The ensuing discussion considers the horizontal, vertical

and structural provisions of the UK unwritten constitution and the way in which they are shaped

by parliamentary sovereignty. The analysis includes the emergence of parliamentary

sovereignty in the UK constitutional order, the parallel historical discourse concerning

fundamental law, the resultant centrality of parliamentary sovereignty and the ways in which

this was justified over time. The inquiry concludes with the modality and frame for

constitutional change in the UK.

2 Written Constitutions

(a) Preliminaries

Constitutions have existed for a very long time. This is a trite proposition. The relative

provenance of the ‘modern’ written constitution is, however, contested, as are the criteria that

should count in this respect, and the cause of its emergence. This is not the place for exegesis

on this issue, which could well occupy an article or book. Some sense of the argumentation is,

nonetheless, helpful for the ensuing analysis. Dieter Grimm’s account is helpful in this regard.2

He argues that the modern written constitution should be dated from the American and

French revolutions, and the subsequent constitutional settlements. The term ‘constitution’ can

be found earlier, but it generally denoted ‘the condition or situation of a state’, as shaped by

2
D Grimm, Constitutionalism, Past, Present and Future (Oxford University Press, 2016).

Electronic copy available at: https://ssrn.com/abstract=3897906

‘historical development, natural features and legal order’.3 In this sense ‘every state was in a

certain “constitution”, and where no constitution could be identified, no state existed’.4 This

older conception of the constitution was thus principally empirical, by way of comparison to

the modern constitution that was normative, insofar as it prescribed how state power should be

established. The modern constitution thus ‘constituted’ rule, it was ‘comprehensive’ in the

sense of regulating the entirety of rule, and also ‘universal’ in that it benefited all persons

subject to rule.5 For Grimm, the distinguishing feature of the American and French

revolutionaries was that their discontent was ‘not limited to the person of the ruler, but

encompassed the system of rule’.6 The forces that shaped the modern constitution appear from

the following passage.7

Historically, constitutions are a product of the conflict between the liberal

bourgeoisie and the absolute monarchy. The bourgeoisie was seeking to
emancipate the production and commercial sphere from hierarchical
direction and to guarantee the predictability of state power. To this end, it
demanded the limitation of princely competence to questions of security and
order as well as participation in state decision-making concerning issues of
freedom and property. These demands were justified on the basis of natural
law and were to be secured through positive law. The means was the
constitution, which regardless of how it came to be in each individual case
had a contractual character. It typically contained basic rights to re-define the
social sphere of freedom, on the one hand, and rules regulating the holders
and modalities of the exercise of state power on the other. In particular, it
provided for representational bodies to communicate civic interests to the
state, a guarantee that the state complied with the decisions of these bodies,
together with the establishment of independent courts as a complementary
safeguard.

3
Ibid 44.
4
Ibid 44.
5
Ibid 43.
6
Ibid 6.
7
Ibid 128-129.

Electronic copy available at: https://ssrn.com/abstract=3897906

 For Dieter Grimm, the single biggest challenge to this conception of the constitution

has been the changing nature of state obligations, principally, albeit not exclusively, through

the emerging conception of the social state, with the consequence that the ‘constitution can no

longer restrict itself to organizing the state apparatus and limiting state power’, but must ‘order

not only the state, but society as well’.8

(b) Content: Horizontal, Vertical and Structural

Constitutions vary significantly in terms of the topics covered, and the depth of the coverage.

This is an issue that will be addressed below when considering over-constitutionalization. This

heterogeneity should not, however, conceal the commonality of subject-matter as between

written constitutions, and this is so notwithstanding the obvious fact that different polities may

choose to address such issues in different ways. The common core consists of constitutional

provisions that are horizontal, vertical and structural in nature.

The horizontal dimension speaks to the procedural and substantive rules that establish

and regulate the principal organs of government, their constitution and powers. These rules

classically relate to the legislature, executive and judiciary. They will identify the respective

organs, and specify their powers. Procedural norms will, for example, specify how legislation

is to be passed, whether special majorities are required and the like. Substantive norms will,

for example, denominate the legislature and the nature of its power, the same being true for the

executive and the judiciary. There can be difficulties in defining the institutions that exercise

certain types of power, and there can also be difficulties in deciding on the precise powers that

a certain institution should have. This is especially so in relation to executive power.

8
Ibid 135.

Electronic copy available at: https://ssrn.com/abstract=3897906

Constitutions are, in general, not good at defining the executive, or the scope of its powers.9

This is so for a concatenation of reasons, a principal cause being the duality in the meaning of

the term executive. It can mean the body charged with developing policy and the legislative

agenda; it can also connote the body charged with implementing legislation. The increased use

of agencies that exercise executive powers, and operate outside the strict confines of

departmental structures, has exacerbated the difficulties in identification of the executive and

the delineation of its powers. Constitutional uncertainty concerning the location and scope of

executive power then generates rival theories that give diverse answers to these issues, which

play out not just in law review exchanges, but in high-level politics.

The vertical dimension of the constitution traditionally captures constitutional rules that

regulate the interrelationship between citizen and state. The paradigm situation is rights-based

constraints on government action, through a Bill of Rights enshrined in the Constitution. This

may form part of the initial constitution, as is commonly the case with constitutions created

after 1945. It may be added later through constitutional amendment, as with the US

Constitution. There will then be a range of second order issues that fall to be resolved, most

prominently the rights to be included in the constitutional document. There is significant choice

in this respect, ranging from the traditional focus on civil and political rights, to more expansive

documents that also include social and economic rights, such as the EU Charter of Rights. The

other central issue concerns the reach of the Bill of Rights, whether it operates only on a vertical

plane, or whether it also has a horizontal impact as between private parties. This choice may

be made explicitly in the Bill of Rights, but where there is uncertainty it falls to the courts to

make the interpretive choice.

9
P Craig and A Tomkins (eds), The Executive and Public Law: Power and Accountability in Comparative

Perspective (Oxford University Press, 2005).

Electronic copy available at: https://ssrn.com/abstract=3897906

 The structural dimension of the constitution in nation states that are federal is concerned

with the division of power between the federal and state level. A written constitution may also

contain analogous provisions that demarcate the boundaries between central power and that of

regions where there is a measure of devolution. The constitutional criteria used for such

division or demarcation of power are varied, including specification of subject matter and

impact across intra-state territorial boundaries. The criteria often lack pristine clarity, thereby

sowing the seeds for contestation as to the boundaries of power that fall to be resolved in the

courts. This uncertainty can impact diverse issues, including the allocation of regulatory

authority as between the federal and state level, and the federal constraints on state rules

concerning who can vote in federal elections.

There will, however, always be an unwritten element in all written constitutions, even

those that are more detailed. There may be issues that were never addressed in the constitution,

because they were never thought about by those who drafted the document. There may be

certain constitutional provisions, whether horizontal or vertical, which are set at a high level of

abstraction, thereby requiring further specification through legislation, judicial interpretation

or administrative practice, or an admixture of all three.

(c) Content: Over and under-constitutionalization

Constitutions vary significantly when judged by a plethora of criteria. A significant factor in

this respect is the constitution’s relative breadth or narrowness, as judged by the range of its

coverage and the detail thereof. To put the point more starkly, constitutions can be relatively

fat or thin. There is no a priori reason to prefer one over the other. The choice betwixt the two

can be the result of conscious constitutional design, or adventitious historical circumstance

flowing from the founding constituent moment. A range of considerations may well play into

the option thus selected. The choice will perforce have consequences. Thus, other things being

Electronic copy available at: https://ssrn.com/abstract=3897906

equal, the greater the constitutional coverage, the more issues are taken off the agenda of

normal politics, or the normal political process is circumscribed. This in turn prompts inquiry

into the problem of over-constitutionalization.

This can be exemplified by discourse in the EU, most notably by Dieter Grimm.10 His

central thesis is that the EU Treaties are over-constitutionalized, with the consequence that

issues are thereby taken off the agenda of normal politics, notwithstanding the fact that they

might be regarded as within the province of ordinary law in Member States: ‘in the EU the

crucial difference between the rules for political decisions and the decisions themselves is to a

large extent levelled’.11 It is inherent in the nature of constitutions that they function as the

framework for political decisions, with the consequence that elections ‘do not matter as far as

constitutional law extends’.12 There may be too little constitutionalism, but there may also be

too much, with the consequence that the democratic process is fettered. 13 While there are no

universally applicable principles for determining the content of a constitution, the ‘function of

constitutions is to legitimize and to limit political power, not to replace it’,14 with the

consequence that constitutions are a ‘framework for politics, not the blueprint for all political

decisions’.15

For Grimm, the EU Treaties fulfil many functions of national constitutions, specifying

matters such as the inter-institutional distribution of power, the mode of law-making, and the

respective competence of the EU and Member States. They also go significantly beyond the

10
D Grimm, ‘The Democratic Costs of Constitutionalization: The European Case’ (2015) 21 ELJ 460.
11
Ibid 470.
12
Ibid 463.
13
Ibid 464.
14
Ibid 465.
15
Ibid 465.

Electronic copy available at: https://ssrn.com/abstract=3897906

remit of national constitutions, with the consequence that a wide range of matters becomes

constitutionalized and taken off the agenda of normal politics. The effect of this is further

enhanced by the constitutional doctrines of direct effect and supremacy, which transformed the

four economic freedoms from ‘objective principles for legislation into subjective rights of the

market participants who could claim them against the Member States before the national

courts’.16

This then meant that there were two modes of EU integration. The Treaty precepts could

be advanced through legislation enacted by the EU institutions, or they could be taken forward

through judicial decisions, which were imbued with considerable force through direct effect

and supremacy.17 Member States had limited influence over the latter, and this was particularly

important since the lack of differentiation between the constitutional law level and the ordinary

law level meant that the ‘constitutionalization of the treaties immunizes the Commission and

particularly the ECJ against any attempt by the democratically responsible institutions of the

EU to react to the Court’s jurisprudence by changing the law’. 18 For Grimm, the remedy was

to limit the EU Treaties to their truly constitutional elements and downgrade other Treaty

provisions that were not constitutional nature to the status of secondary law.

(d) Change: Modality and Frame

This brief excursus into written constitutions ends with consideration of the modality and frame

though which constitutional change occurs. There are four principal modes in this respect.

16
Ibid 467.
17
J Weiler, ‘The Community System: The Dual Character of Supranationalism’ (1981) 1 YBEL 267, and ‘The

Transformation of Europe’ (1991) 100 Yale LJ 2403, 2412–31; P Craig, ‘Once Upon a Time in the West: Direct

Effect and the Federalization of EEC Law’ (1992) 12 OJLS 453.

18
Grimm (n 10) 471.

Electronic copy available at: https://ssrn.com/abstract=3897906

 The most obvious is perforce formal constitutional amendment,19 the catalyst for which

varies. 20 In some instances, it is the realization that the initial constitutional design is imperfect;

in others it is the result of new political configurations that wish to re-shape the country’s

institutional and constitutional architecture; in yet other instances, the motives may be darker,

such that current majorities seek to reinforce their power through constitutional amendment

designed, for example, to limit the power of the judiciary.21

The second impulse for constitutional change is judicial interpretation. This language

is, however, inherently tendentious, since it is based on assumptions as to how courts do and

should interpret the constitution. The very idea that the constitution has or has not been changed

depends, inter alia, on the theory of constitutional adjudication accepted by that legal order, or

espoused by academic commentary. This is the site of well-known battles, including that

between constitutional originalists, and those who subscribe to a more Dworkinian view of

constitutional interpretation. Matters are rendered more complex by the plethora of very

different theories of what constitutional originalism means, and by similar variation in the

latitude afforded by non-originalist theories.22 This is quite apart from the small matter as to

19
R Albert, Constitutional Amendments, Making, Breaking and Changing Constitutions (Oxford University Press,

2019); Y Roznai, Unconstitutional Amendments, the Limits of Amendment Powers (Oxford University Press,

2019).
20
Compilation of Venice Commission Opinions concerning Constitutional Provisions for Amending the

Constitution, CDL-PI(2015)023; Venice Commission, Report on Constitutional Amendment, CDL-

AD(2010)001.
21
Kim Lane Scheppele, ‘Autocracy under Cover of the Transnational Legal Order’ in G Shaffer, T Ginsburg and

T Halliday (eds), Constitution-Making and Transnational Legal Order (Cambridge University Press, 2019) Ch 7.
22
See, eg, J Goldsworthy, ‘Constitutional Interpretation: Originalism’ (2009) 4 Philosophy Compass 682; K

Whittington, ‘Originalism: A Critical Introduction’ (2013) 82 Fordham Law Review 375; Symposium on

10

Electronic copy available at: https://ssrn.com/abstract=3897906

whether court decisions over time can be said to subscribe to any consistent view within a

particular jurisdiction.

The third driver of constitutional change is constitutional legislation, which is

somewhat less obvious, but equally important. It is normatively and pragmatically mistaken to

draw a rigid line between the constitution and legislation that is designed to fill out the meaning

of its provisions, and it is misguided to elide the existence of the same provision in many

constitutions with equivalence of application. Constitutional legislation is in many respects the

lifeblood of the constitution. Other things being equal, the thinner the constitution and the more

abstract its provisions, the greater the role for constitutional legislation and adjudication to

imbue the terms with more concrete meaning. Such legislation is nonetheless important in all

constitutions, although its incidence may vary. The constitution may contain noble provisions

concerning free and fair elections, but the reality can only be tested by examination of the

detailed electoral legislation that puts flesh on these constitutional bones. It is powerfully

attested to by recent legislation in some states of the United States, which has placed increased

barriers on people’s ability to vote in state and federal elections. There may be laudable

constitutional guarantees for political parties, only for these to be heavily circumscribed by

legislation that defines the qualifications for a political party to appear on the ballot. The force

of constitutional protections for free speech may be qualified by legislative rules related to

association. The need to be mindful of the link between constitutional terms and consequent

legislation is important in relation to many other issues commonly found in constitutions, such

as principles concerning equality, the judiciary and the like, which are then articulated in more

detail in, or affected by, legislation.

Originalism in (2019) 37(3) Law and History Review; JH Ely, Democracy and Distrust: A Theory of Judicial

Review (Harvard University Press, 1980); R Dworkin, Taking Rights Seriously (Harvard University Press, 1978).

11

Electronic copy available at: https://ssrn.com/abstract=3897906

 The final factor in this respect is constitutional practice, which may take various forms.

There may be institutional conventions, or perceptions of the constitution by the people. There

may be practice by the relevant players, which either undermines the formal constitutional

norms, or imbues them with greater force than they would otherwise have as judged by their

formal text. There may be events that test the ambit of formal constitutional provisions, with

the resolution of the contestation casting the constitutional norms in a somewhat different light

than hitherto.

3 Unwritten Constitutions

(a) Preliminaries

Most current constitutions are written. Unwritten constitutions are the exception. This, like the

proposition that began the previous section, is trite. However, in common with that discussion,

matters rapidly become more complex and contestable when we move beyond this simple

terrain. This is more especially so, if the inquiry is as to why a particular polity chose to buck

the trend, and to stick with an unwritten constitution. The answer will perforce depend, inter

alia, on an admixture of historical, social and economic circumstance. These issues are beyond

the remit of the present chapter and would require an article or book to do justice to the topic.

Nor is the intent to engage in the debate as to whether the UK should have a written

constitution.23

23
The principal proposals for a written constitution can be found in J Murkens, ‘A Written Constitution: A Case

not Made’ (2021) 41 OJLS 1. See also, V Bogdanor and S Vogenauer, ‘Enacting a British Constitution: Some

Problems’ [2008] PL 38; Nick Barber, ‘Against a Written Constitution’ [2008] PL 11; Political and Constitutional

Reform Committee, A New Magna Carta? (HC 2014-15, 463); Vernon Bogdanor, Beyond Brexit: Towards a

British Constitution (Hart 2019).

12

Electronic copy available at: https://ssrn.com/abstract=3897906

 This preliminary foray is limited to making a simple conceptual point, which will be

developed hereafter. The content of unwritten constitutions can, in principle, vary significantly.

There is, moreover, no a priori reason why a state might not choose to stick with an unwritten

constitution for an admixture of historical, social and economic circumstance, and yet still

replicate, in unwritten form, many features commonly found in a written constitution. It might

limit the remit of ordinary statute in accord with constraints typically found in written

constitutions, and these rules might be accepted as obligations by the key political and legal

players. The net effect might be a constitutional order that is similar to, and equally efficacious

as, its written constitutional cousin.

There can, however, also be an unwritten constitution, such as that in the UK, where

the animating constitutional principle is that of parliamentary sovereignty. This constitutional

choice is premised on normative assumptions that are very different from those that underpin

written constitutions, and the alternative unwritten constitution sketched above. This in turn

affects the modality and scope for constitutional change. It should, moreover, be recognized

that if a polity decides to shift from an unwritten to a written constitution it has considerable

choice as to its content. Thus, an interesting consideration in the context of the present

discussion is that many of the proposals for a written constitution in the UK are predicated on

preservation of parliamentary sovereignty, or something closely akin thereto.24

(b) Content: Parliamentary Sovereignty, the Horizontal, Vertical and

Structural

There can, as noted above, be difficulties in determining the more particular content of written

constitutions. Difficulties are, however, relative by nature, and they are greater when trying to

24
Murkens (n 23) 6-8.

13

Electronic copy available at: https://ssrn.com/abstract=3897906

determine the content of an unwritten constitution, such as the UK. We can, nonetheless, make

progress in this respect. The basic taxonomy of horizontal, vertical and structural constitutional

provisions is equally pertinent to an unwritten constitution as to the more common written

version.

There is no doubt that the UK unwritten constitution contains rules that speak to the

horizontal, vertical and structural features commonly found in written constitutions. Thus, there

are a number of statutes dealing with the horizontal dimension, insofar as this connotes the

powers vested in the legislative, executive and judicial branches of government, including,

inter alia, the Bill of Rights 1689, Act of Settlement 1700, Parliament Acts 1911 and 1949,

Ministers of the Crown Act 1975, and hitherto the European Communities Act 1972. The

statutes that address the vertical features of a constitution include the Bill of Rights 1689 and

the Human Rights Act 1998. The structural constitutional dimension is exemplified by the

legislation relating to devolution, the modern incarnation of which began with the Scotland Act

1998 and the Government of Wales Act 1998. This legislation has been held by the courts to

have constitutional status, which has, as will be seen below, implications for the manner of

change.25

There are, in addition, important constitutional precepts such as the rule of law. This is

not the place for exegesis on the meaning of the rule of law, and its significance within the UK

unwritten constitutional order. Suffice it to say that it is of considerable importance, as manifest

in principles of judicial interpretation whereby statutes that limit access to courts will be

narrowly construed, as will legislation that limits human rights. The rule of law also shapes the

legislative process, through the importance attached to ensuring that statutory provisions are

clear and prospective, and it underpins judicial review, such that power accorded to ministers

25
R (HS2 Action Alliance Ltd) v Secretary of State for Transport [2014] UKSC 3.

14

Electronic copy available at: https://ssrn.com/abstract=3897906

and agencies is held to account. There are, nonetheless, abiding difficulties concerning the

scope and source of executive power, which is a powerful theme in Bruce Harris’ work.26

It is, nonetheless, necessary to confront the centrality of parliamentary sovereignty in

the UK unwritten constitution. It is the dominant occupant of this conceptual terrain. The

orthodox reading of parliamentary sovereignty is that Parliament is omnipotent, and thus can

be subject to no substantive or procedural limits. This precept is subject to one exception, which

is that Parliament cannot bind its successors. The thesis is thus one of continuing parliamentary

sovereignty, such that each successive Parliament is invested with the same omnipotence as its

forbears, thereby reflecting the normative claim that each Parliament is entitled to the same

authority flowing from the fact that it has equal democratic pedigree to the Parliaments that

preceded it.

The possible twin qualifications to this proposition are well-known and do not warrant

detailed explication here. They relate to the possibility of procedural constraints on

sovereignty, whereby Parliament would be bound by manner and form provisions contained in

earlier legislation, unless and until they are altered; and to the possibility that the courts might,

in extreme circumstances, refuse to apply legislation that is deeply problematic in terms of

substance.

These possible qualifications should not mask the centrality of parliamentary

sovereignty to the UK unwritten constitution, or the contrast with the basic precepts of written

constitutions. The foundational starting point of written constitutions is the distinction between

ordinary politics and the constitution; the latter circumscribes the former, such that the terrain

of ordinary politics is bounded by the constitution. The foundational starting point of the UK

26
See, eg, BV Harris, ‘The “Third Source” of Authority for Government Action’ (1992) 108 LQR 626; ‘The

“Third Source” of Authority for Government Action Revisited’ (2007) 123 LQR 225; ‘Government “Third

Source” Action and Common Law Constitutionalism’ (2010) 126 LQR 373.

15

Electronic copy available at: https://ssrn.com/abstract=3897906

unwritten constitution is that parliamentary sovereignty is the principal constitutional principle;

the terrain of ordinary politics is boundless and parliamentary sovereignty instantiates and

elevates ordinary statute as the vehicle for its delivery.

This in turn leads to the following irony. Written constitutions have an inherent

weakness, insofar as they do not have the same enforcement mechanism as ordinary statute.

There are to be sure methods of enforcement through constitutional courts and the like, but this

does not dispel the preceding concern. An unwritten constitution grounded on parliamentary

sovereignty obviates this problem, by the very fact that ordinary statute, cloaked with the

mantle of parliamentary sovereignty, is constitutional centre-stage, and partakes of

enforcement mechanisms attendant on normal legislation.

This naturally invites inquiry as to why the UK opted for an unwritten constitution with

parliamentary sovereignty as its central component, given that the latter is not a pre-ordained

component of the former. The reasons are assuredly eclectic, and linked.27 They reside in part

in historical circumstance.28 The Bill of Rights 1689 was only a partial constitutional

settlement. It did, however, circumscribe monarchical power, and bolster that of Parliament.

The nobility and the middle class did not feel the imperative for a formal written constitution

to safeguard their rights, and the flipside of this same coin was that they did not wish for limits

on the powers of a Parliament over which they had control. The emergence of parliamentary

sovereignty was, moreover, influenced at various junctures by Parliament’s status as the

27
C Hill, The World Turned Upside Down, Radical ideas During the English Revolution (Penguin, 1972) Ch 7;

D Veall, The Popular Movement for Law Reform, 1640-1660 (Oxford University Press, 1970); S Sedley and L

Kaplan (eds), A Spark in the Ashes, The Pamphlets of John Warr (Verso, 1992).
28
See, J Goldsworthy, Parliamentary Sovereignty: Contemporary Debates (Cambridge University Press, 2010)

for detailed discussion of contending views as to the nature of parliamentary sovereignty in the seventeenth

century.

16

Electronic copy available at: https://ssrn.com/abstract=3897906

highest court, as well as the legislature.29 It was also affected by the writings of Hobbes30 and

Filmer,31 who lent credence to the idea that there had to be a locus of undiminished power. The

dispatch of the Stuarts and the post-1689 settlement laid to rest the remnants of previous debate

about the possessor of sovereign power. Sovereignty resided in the King-in-Parliament, in the

sense that it was the King acting in conjunction with the Commons and Lords which possessed

all-embracing power.

The gradual emergence of parliamentary sovereignty in something akin to its modern

form by the end of the 17th century should not, however, lead to neglect of the concept of

fundamental law, or the relationship between fundamental law and the constitution, to which

reference was made throughout the century and beyond. English scholars of the Tudor age used

Aristotelian ideas of ‘politeia’, but did not speak of an English constitution.32 Recourse to the

term ‘constitution’ in connection with forms of government began in the 17th century. Its roots

lay, in part, in analogies between nature and politics, as epitomised in the ideal of a healthy

body and a healthy body politic; and in part in development from the Latin term ‘constitutio’,

which was used in Roman law to connote imperial decrees, and by the mid-17th century was

employed to capture the fundamental constitution of the kingdom.33 It is, nonetheless, clear

that the concept of fundamental law bore various meanings, but did not generally connote the

idea of a constitution policed through courts and judicial review,34 notwithstanding that writers

29
JW Gough, Fundamental Law in English Constitutional History (Clarendon Press, 1955), 82-84, 171-173.
30
T Hobbes, Leviathan (WG Pogson Smith, Clarendon Press, 1909).
31
P Laslett (ed), Patriarcha and other works of Sir Robert Filmer (Garland, 1984).
32
G Stourzh, Fundamental Laws and Individual Rights in the 18 th Century Constitution (Claremont Institute,

Bicentennial Essay Number 5, 1984) 2-3.

33
Ibid 3-6.
34
Gough (n 29) 2-3, 206-207.

17

Electronic copy available at: https://ssrn.com/abstract=3897906

drew on the work of Coke in support of such developments in the United States.35 In England,

it was used to capture a plethora of ideas, including the social contractarian nature of the bond

between the people and their representatives; the existence of certain liberties and rights that

should be protected; that politics was subordinate to moral obligation; and that there should be

constraints against arbitrary power, inter alia, by ensuring that there were checks and balances

within government broadly conceived. There was resort to fundamental law by all sides in the

17th century conflicts, and unsurprisingly the meaning accorded to it shifted depending on the

position of the particular party.36

Recourse to ideas of fundamental law waned as conceptions of parliamentary

sovereignty gained traction. However, as John Gough duly notes, this coincided with more

detailed attempts to justify the emerging constitutional order, to show that the omnipotence

was defensible and would not jeopardize rights and liberties valued by the people.37 This was

readily apparent from discourse in the 18th, 19th and 20th centuries, wherein different arguments

were advanced for the constitutional centrality of parliamentary sovereignty and ordinary law.

Blackstone furnished the principal argument in the 18th century. He followed Coke and

affirmed that the power and jurisdiction of parliament was so transcendent and absolute, that it

could not be confined, either for causes or persons, within any bounds, nor could it be

controlled by an external body.38 The existence of this power was, however, premised on a

35
See, e.g., ES Corwin, ‘The “Higher Law” Background of American Constitutional Law’ 42 Harv Law Rev 149,

365 (1928-29).
36
Gough (n 29) 162-163.
37
Ibid 173; Stourzh (n 32) 8-9.
38
Sir W Blackstone, Commentaries on the Law of England (Cadell and Butterworth, 16th edn, 1825), Vol I, Bk

2, 160-61.

18

Electronic copy available at: https://ssrn.com/abstract=3897906

political theory of balanced constitutionalism expounded by Blackstone, whose reasoning

echoed earlier discourse, most notably that of Harrington and the republican tradition.39

The famous Blackstonian quotation appeared at the end of a long chapter, in which he

described how Parliament functioned. Tyrannical governments were, said Blackstone, those in

which the power of making and enforcing laws was vested in the same person. Things were

different where the legislative and executive authority were in different hands, since ‘the

former will take care not to entrust the latter with so large a power as may tend to the subversion

of its own independence, and therewith the liberty of the subject’.40 In England power was thus

divided. The legislative branch was the Parliament, which consisted of the King, Lords and

Commons; the executive branch consisted of the King alone.41 The relationship between them

was vital.42

It is highly necessary for preserving the balance of the constitution, that the
executive power should be a branch, but not the whole, of the legislative. The
total union of them, we have seen, would be productive of tyranny; the total
disjunction of them, for the present, would in the end produce the same
effects, by causing that union against which it seems to provide. The
legislature would soon become tyrannical, by making continual
encroachments, and gradually assuming to itself the rights of the executive
power.

For Blackstone the ‘true excellence’ of English government was that all parts operated

as a mutual check upon each other. In the legislature, the people were a check on the nobility,

39
J Harrington, ‘The Commonwealth of Oceana’, The Political Works of James Harrington (JGA Pocock ed,

Cambridge University Press, 1977); J de Lolme, The Constitution of England, or an Account of the English

Government (Robinson, 1796 edn.). See generally, JGA Pocock, The Machiavellian Moment: Florentine Political

Thought and the Atlantic Republican Tradition (Princeton, 1975); C Weston, English Constitutional Theory and

the House of Lords 1556-1832 (Routledge and Kegan Paul, 1965).

40
Blackstone (n 38) 146.
41
Ibid 146-7.
42
Ibid 153-4.

19

Electronic copy available at: https://ssrn.com/abstract=3897906

and the nobility a check upon the people, through ‘the mutual privilege of rejecting what the

other has resolved’.43 The King was a check upon both, which preserved the executive power

from encroachments. The executive power itself was checked and kept within due bounds by

the two houses, through the privilege of inquiring into, impeaching, and punishing the conduct,

not of the King or Queen since this would destroy their constitutional independence, but of

‘evil and pernicious counsellors’.44 The two houses of Parliament naturally represented

different interests, and the Crown yet another. They mutually kept each other from exceeding

their proper limits. It is only after this discourse that the quotation about parliamentary

omnipotence appears, and the body ‘thus united’ is the body functioning in the manner

described above.

Dicey furnished the leading reaffirmation of parliamentary sovereignty in the 19th

century, drawing on Coke and Blackstone. However, the normative justification for this

ascription of parliamentary power differed.45 He recognized that he could not simply draw on

the Blackstonian normative argument, since it would not have been applicable, without

substantial modifications, to the changed conditions of the nineteenth century. Dicey, therefore,

constructed an alternative normative argument to justify the Parliamentary sovereignty that he

had empirically described.

The essence of the argument was that a Parliament, duly elected on the extended

franchise, represented the most authoritative expression of the will of the nation. The

Parliament thereby elected should therefore be able to carry out any action. Dicey believed that

Parliament would control the executive and that MPs would not pass legislation contrary to the

43
Ibid 154-5.
44
Ibid 155.
45
P Craig, Public Law and Democracy in the United Kingdom and the United States of America (Clarendon,

1990) Ch 2.

20

Electronic copy available at: https://ssrn.com/abstract=3897906

interests of those who elected them. Constitutional protections against the exercise of

parliamentary power were not therefore required, since, ‘the permanent wishes of the

representative portion of Parliament can hardly in the long run differ from the wishes of the

English people, or at any rate of the electors; that which the majority of the House of Commons

command, the majority of English people usually desire’.46 The normative argument was,

therefore, apparently simple. UK democracy was founded upon a channel of authority flowing

from the bottom upwards. The expanded electorate chose representatives. The Parliament thus

chosen had legitimacy because of the extended franchise and should therefore have all

embracing powers. The elected MPs articulated the views of those who had chosen them, and

they controlled the executive. Legislation which was constitutionally questionable would not,

therefore, be passed, or would be repealed expeditiously. Parliamentary sovereignty would not

place the rights of individual citizens in jeopardy. The defects in this mode of argument have

been examined elsewhere.47 Suffice it to say for the present that our system of democracy never

really operated in this self-correcting way, and this vision of the relationship between electors,

Parliament and the executive certainly does not accord with present reality.

The 20th century normative argument for parliamentary sovereignty as the cornerstone

of the UK constitutional order differs yet again from the Blackstonian and Diceyan antecedents,

in part at least because the ideals of balanced constitution and self-correcting democracy no

longer have purchase. The 20th century claim is therefore commonly grounded on two

complementary arguments. There is the contention that a Parliament duly elected on a full

democratic franchise has legitimacy and thus should have such sovereign authority. This is

complemented by the argument that it should not be subject to limits such as constitutional

review, whereby the court can invalidate legislation. This reasoning underpins much political

46
AV Dicey, An Introduction to the Study of the Law of the Constitution (Macmillan, 10th edn, 1967) 83.
47
Craig (n 45) Ch 2.

21

Electronic copy available at: https://ssrn.com/abstract=3897906

constitutionalist thought. Waldron48 and Bellamy49 are the leading opponents of rights-based

constitutional review of statute.50 Waldron is a court-sceptic, but not a rights-sceptic. Bellamy’s

position is more complex. He is certainly a court-sceptic, and is also more sceptical about

rights, insofar as he believes that citizenship should not be equated with a narrow concept of

individuals being rights-holders against the state, but comprises a ‘continuously reflexive

process, with citizens reinterpreting the basis of their collective life in new ways that

correspond to their evolving needs and ideals’.51 There is, nonetheless, much common ground

in the reasons for their court-scepticism. Thus, the central premise to Waldron and Bellamy’s

argument is the prevalence of disagreement concerning the rights that should be included in a

Bill of Rights and their interpretation. For both writers such disagreement pervades the very

foundational ideas of justice on which society is grounded. They maintain therefore that

whether viewed in terms of process, or in terms of outcome, it is preferable for such matters to

be decided ultimately by the political and not the legal process.

(c) Content: Over and under-constitutionalization

48
J Waldron, Law and Disagreement (Oxford University Press, 1999) and ‘The Core Case against Judicial

Review’ (2006) 115 Yale LJ 1346.

49
R Bellamy, Political Constitutionalism, A Republican Defence of the Constitutionality of Democracy

(Cambridge University Press, 2007).

50
See also, M Tushnet, Taking the Constitution away from the Courts (Princeton University Press, 1999); L

Kramer, The People Themselves: Popular Constitutionalism and Judicial Review (Oxford University Press,

2004).
51
R Bellamy, ‘Constitutive Citizenship versus Constitutional Rights: Republican Reflections on the EU Charter

and the Human Rights Act’, in T Campbell, KD Ewing and A Tomkins (eds), Sceptical Essays on Human Rights

(Oxford University Press, 2001), 15.

22

Electronic copy available at: https://ssrn.com/abstract=3897906

The preceding analysis considered the issue of over and under-constitutionalization in the

context of written constitutions. The issue is also relevant for unwritten constitutions, and the

answer will depend on the relative breadth and depth of its coverage. It is possible in theory for

an unwritten constitution to be relatively dense in terms of the mesh of rules that speak to

constitutional issues. The empirical reality tends to the contrary, such that the constitutional

frame is more skeletal in this respect. This is more especially so where the dominant

constitutional principle is parliamentary sovereignty, which, as seen above, weakens the

binding force of other constitutional constraints. There is then a greater danger of under-

constitutionalization. Whether this would be alleviated by a written constitution depends in

large part on the content thereof, and its animating principles. Thus, a UK written constitution

crafted on the bedrock of parliamentary sovereignty could still be under-constitutionalized.

Having said this, it should be acknowledged that the very ascription of the label under-

constitutionalized to such a putative written constitution is predicated on certain normative

assumptions, the principal one being that Parliament should not have such legal omnipotence.

This assumption is normatively contestable, but so too is the contrary assumption. The under-

constitutionalized nature of the UK unwritten constitution can, nonetheless, be exemplified by

experience with Brexit.

Take by way of example the way in which the devolved regions were treated in the

Brexit process. They were denied effective voice in the shaping of the Brexit process, ever

more so the longer it went on; and their substantive preferences received scant attention.52

Whether matters would have been different if we had a written constitution would depend very

markedly on its terms. It is possible to imagine a UK written constitution that embodied the

legal status quo, in which case the position of the devolved regions would be no better than at

52
P Craig, ‘Brexit, A Drama: The Interregnum’ [2017] Yearbook of European Law 1.

23

Electronic copy available at: https://ssrn.com/abstract=3897906

present. It is, by way of contrast, possible to imagine a very different form of constitutional

ordering, which gave greater procedural protection to regional voice, invested the Sewel

convention, namely that the Westminster Parliament cannot normally legislate on a matter

within devolved competence without the consent of the devolved legislatures, 53 with binding

constitutional force, and made constitutional provision for protection of the regions’

substantive political preferences, within the areas that fell within their scope of authority.

Take by way of another example the balance between prime ministerial, executive, and

legislative power revealed by Brexit. The executive, and increasingly the Prime Minister,

dominated the Brexit process. Whether a written constitution would have made any difference

would depend on its terms. This is more especially so, since constitutions do not have a good

track record at identifying the nature of executive power and the boundaries thereof.54 Nor are

they particularly good at specifying the more granular detail of relations between the executive

and the legislature. However, if the UK had a written constitution then it is likely that UK

membership of the EU would have been enshrined therein. It would have had constitutional

status, the corollary being that it could only be changed by the process for constitutional

amendment. Whatsoever such requirements for change might be, the reality is that a

referendum, followed by an ordinary statute, would almost certainly not suffice to change the

constitution.

(d) Change: Modality and Frame

There can assuredly be change in unwritten constitutions. The ability to adapt is indeed said to

be a virtue of this constitutional form. There is some truth in this. The claim is nonetheless

53
Scotland Act 2016, s 2.
54
Craig and Tomkins (n 9).

24

Electronic copy available at: https://ssrn.com/abstract=3897906

commonly overplayed, insofar as it fails to distinguish between the ability of a system to adapt

existing constitutional rules to novel circumstance, and its capacity to meet change because

there are no constitutional rules that govern the matter. The focal point of the ensuing

discussion is, however, somewhat different. It concerns two related issues: the way in which

constitutional change can occur in a system premised on parliamentary sovereignty; and the

way in which parliamentary sovereignty frames this process. They will be considered in turn.

(i) Modality

The key to an understanding of the way in which constitutional change can occur is that the

orthodox proposition concerning parliamentary sovereignty is empirical, not a priori. It is

dynamic and not static and hence can alter over time. The rule concerning parliamentary

sovereignty, as set out above, can be regarded as the rule of recognition of the UK system, in

the sense that it is the ultimate rule of the legal system, to which the validity of all other legal

rules can be traced. Its validity is not, however, dependent on the rules that constitute its content

at any point in time, since that would be bootstrapping. It is rather dependent upon social

acceptance by those within the polity.55

The proposition that statutes duly enacted have the force of law, and that subject to this

so too do judicial decisions, constitutes the core of the rule of recognition because it is accepted

by those within the system, including the legislature and the courts. The enactment of a statute

attesting to this fact could not itself invest statute with this status. A statute cannot, therefore,

in itself, alter the rule of recognition, but it can act as the catalyst for a shift in the social

acceptance on which the rule of recognition depends.

55
HLA Hart, The Concept of Law (Oxford University Press, 1961) 97-107.

25

Electronic copy available at: https://ssrn.com/abstract=3897906

 It follows that the rule of recognition is inherently dynamic, not static. The version we

currently operate with, that Parliament can do whatsoever it likes through statute duly enacted

by simple majority, has not always been so. It was certainly not the ultimate test of legal validity

prior to the latter part of the seventeenth century, and it is over-simplistic to imagine that

Parliament’s status thereafter was secured in the manner that it is commonly regarded now.

The dynamic nature of the concept means, moreover, that the rule might change and having

done so it might revert in the future to the status quo ante, as the result of developments that

trigger shifts in the social acceptance that underpin the concept.56 The initial catalyst for change

might come from Parliament, the executive or the courts. Change in the rule of recognition

will, however, normally require acceptance from all such key players.

This can be exemplified by the accommodation of EU law prior to Brexit, as evidenced

by the leading UK decisions.57 The UK courts held that the relationship between EU law and

national law in terms of supremacy was to be decided by the UK courts as a matter of UK

constitutional law, taking account of any statutes enacted by Parliament. They also held that

the concept of implied repeal, or implied disapplication, under which inconsistencies between

later and earlier norms were resolved in favour of the former, would generally not apply to

clashes concerning EU and national law, with the consequence that if Parliament wished to

derogate from its EU obligations then it would have to do so expressly and unequivocally. The

reaction of our national courts to such an eventuality was never tested. The courts might have

chosen to follow the latest will of Parliament, or they might have decided that it was not open

56
Ibid 144-150.
57
R v Secretary of State for Transport, ex p Factortame Ltd (No 2) [1991] 1 AC 603; R v Secretary of State for

Employment, ex p Equal Opportunities Commission [1995] 1 AC 1; Thoburn v Sunderland City Council [2003]

QB 151; R (HS2 Action Alliance Ltd) v Secretary of State for Transport [2014] UKSC 3.

26

Electronic copy available at: https://ssrn.com/abstract=3897906

to the legislature to pick and choose which obligations to subscribe to while still remaining

within the EU.

This still left open the way in which this change should be conceptualized. There were

suggestions that this could be achieved through statutory construction. This was, however,

problematic. While all would agree that if a statute could be reconciled with an EU norm

through construing the statutory words without unduly distorting them then this should be done.

However, the species of statutory construction considered here was more far-reaching, because

the rule of ‘construction’ was more akin to a ‘priority’ rule, whereby inconsistencies would be

resolved in favour of the EU law, unless Parliament indicated clearly and ambiguously that it

intended to derogate from EU law. The degree of linguistic inconsistency between the statute

and the EU norm was not the essential point of the inquiry.

An alternative explanation could, however, be cast in terms of change to the rule of

recognition, through a technical legal revolution. On this view, the case law considered above,

combined with acceptance by the leading political players, brought about modification of the

rule of recognition. An accurate rendition of the ultimate legal principle during the pendency

of UK membership of the EU would then have been as follows: Parliament can do anything

that it wishes through statute enacted by simple majority, including implied repeal or

amendment, except in the areas covered by EU law where Parliament would have to legislate

expressly and unequivocally in order to derogate from EU law. This modification of the top

rule of the system was, as noted above, not immutable. The rule of recognition could have been

recast, so as not include the preceding qualification. This might have occurred if the UK courts

had changed their legal position concerning the relationship between UK and EU law while the

UK was still a Member State, and the other organs of government accepted the change. The

fact of Brexit, ex hypothesi, did recast the rule of recognition, since the very rationale for

accommodating EU law had been removed. However, the legal Brexit arrangements attest to

27

Electronic copy available at: https://ssrn.com/abstract=3897906

the continuing force of the preceding reasoning, since Parliament, pursuant to its sovereignty,

decided to continue to accord authority to EU law in certain respects, which has been accepted

by the courts.58

The modality of change in the rule of recognition, and the limits thereto, can be further

exemplified by devolution. The enactment of devolution legislation by the Labour government

in the late 1990s, the Scotland Act 1998 and the Government of Wales Act 1998, was

significant in political terms, through the very grant of autonomy over the devolved issues.

This was further enhanced through subsequent legislation, notably the Government of Wales

Act 2006, the Wales Act 2014 and the Scotland Act 2016, which further augmented their

respective powers.

There is little doubt that the enactment of this legislation might have had implications

for the rule of recognition. It would not, for the reasons given above, have effectuated this

change in itself, but could have been the catalyst for alteration in the social acceptance that

underpins the rule of recognition. The reality is that any such change has been relatively minor.

This is in part because of the terms of the enabling legislation, which make clear that the

Westminster Parliament can still make laws for Scotland and Wales.59 It is in part because of

political considerations. The UK government is opposed to the Scottish National Party’s

separatist ambitions, and opposed also to the holding of a referendum to determine the wishes

of the Scottish people. This has played out in constitutional terms, with the UK government

emphasizing the sovereignty of the Westminster Parliament and the fact that its consent is a

condition precedent to the holding of a referendum. The limits to any change in the rule of

recognition also flow in part from judicial decisions. Courts interpret legislation and not

infrequently make choices as to the consequences that should flow from statutory amendments.

58
European Union (Withdrawal) Act 2018, ss 5-6.
59
Scotland Act 1998, s 28(7).

28

Electronic copy available at: https://ssrn.com/abstract=3897906

This is readily apparent from the judicial reaction to the change in the Sewel Convention, which

established that the Westminster Parliament would not legislate on devolved matters without

the consent of the Scottish Parliament. The Convention was embodied in the Scotland Act

199860 as a result of the Scotland Act 2016. However, in Miller the Supreme Court held that

this legislative amendment did not render the Sewel Convention legally enforceable. 61 Courts

of law ‘cannot enforce a political convention’.62 Judges were ‘neither the parents nor the

guardians of political conventions; they are merely observers’.63 The fact that the Sewel

Convention had been enshrined in legislation by section 2 of the Scotland Act 2016 did not

thereby render a political convention legally enforceable.

The devolution legislation has, therefore, not been the trigger for significant change in

the rule of recognition so far as Scotland, Wales and Northern Ireland are concerned. It would,

however, be misguided to assume that it has had no such effect. Consider in this respect the

constitutional reaction to wholesale repeal of this devolution legislation, such that all matters

without exception were run from the Westminster Parliament, with the consequence that the

principal devolved political institutions were dissolved. Whatsoever the judicial response to

such legislation might be, it would almost certainly be regarded as a constitutional revolution.

This reaction would be premised on the assumption that post-1998 the constitutional settlement

in the UK was one in which Scotland, Wales and Northern Ireland should have some

considerable authority over their own affairs. This, in turn, would be reflected in the rule of

recognition, which would, in essence, have read that while Parliament could do anything that

it wished by simple majority, it would not generally do so in the devolved areas without

60
Ibid s 28(8).
61
R (Miller) v Secretary of State for Exiting the European Union [2017] UKSC 5, [136]-[137].
62
Ibid [141].
63
Ibid [146].

29

Electronic copy available at: https://ssrn.com/abstract=3897906

securing their assent through their respective political institutions. Repeal of the devolution

legislation would thus be regarded as a re-orientation of the rule of recognition.

(ii) Frame

Change in the rule of recognition is, therefore, perfectly possible. Parliamentary sovereignty,

conceptualized in terms of the absence of substantive and procedural limits as to what can be

done through ordinary legislation, nonetheless frames this process. This is manifest in at least

four ways.

First, there is the flipside of the point in the previous section, which is also salient here.

Thus, while limits on parliamentary sovereignty might develop through change in the social

acceptance that constitutes the rule of recognition, the assumed starting point is that there are

no substantive and procedural limits as to what can be done through ordinary legislation, the

corollary of which is that social acceptance as manifest by the key institutional players might

revert back to this position, even if there has been movement therefrom.

Secondly, elevation of the traditional conception of parliamentary sovereignty as the

key constitutional principle has implications for any limits placed thereon. Consider in this

respect the way in which we conceptualize constitutional statutes.64 The recognition afforded

to this concept is to be welcomed. It is, nonetheless, acknowledged that the ascription of this

status is limited in terms of consequences. It is still open to Parliament, in accord with its

sovereignty, to repeal or amend any statutes commonly included on this list, subject to the

64
Thoburn [2003] QB 151, [59]-[63]; McWhirter v Secretary of State for Foreign and Commonwealth Affairs

[2003] EWCA Civ 384, [14]; H v Lord Advocate [2012] UKSC 24; HS2 (n 25); D Feldman, ‘The Nature and

Significance of “Constitutional” Legislation’ (2013) 129 LQR 343; P Craig, ‘Constitutionalizing Constitutional

Law: HS2’ [2014] PL 373; F Ahmed and A Perry, ‘Constitutional Statutes’ (2016) 37 OJLS 461.

30

Electronic copy available at: https://ssrn.com/abstract=3897906

injunction that it must do so expressly and unequivocally, with only very limited, if any, room

for implied change. Consider further in this respect the way in which we think of the principle

of legality. It captures the idea that Parliament can do anything it likes by statute duly enacted,

subject to the interpretive precept that the courts will not read general statutory words so as to

interfere with fundamental rights. Parliament must, as Lord Hoffmann stated,65 pay the political

cost of its action, which meant that if it wished to limit such rights it would have to do so clearly

and unequivocally through language that revealed that Parliament understood what it was

doing. It is, however, still open to Parliament to take such action and any challenge based on

the Human Rights Act 1998 is bounded by section 4, which limits the court to a declaration of

incompatibility, while leaving the validity of the legislation unaffected.

Thirdly, the traditional conception of parliamentary sovereignty frames government

thinking and reform initiatives in the public law context. This is exemplified by Ministry of

Justice initiatives concerning judicial review,66 and the HRA.67 These initiatives have been

fuelled prominently, albeit not exclusively, by claims of judicial overreach, the allegation being

that the courts have trespassed too far on terrain that is properly the preserve of the legislature

and/or the executive. The terms of reference for the HRA review are motivated by similar

considerations, as evident in questions as to whether the courts have been too intrusive in their

interpretation of section 3 of the HRA. The claims of systemic judicial overreach do not

withstand examination,68 and the argument that the courts have overstepped proper bounds in

65
R v Secretary of State for the Home Department, ex p Simms [2000] 2 AC 115.
66
Independent Review of Administrative Law, CP 407, 2021.
67
Independent Human Rights Act Review, IHRAR, 2021.
68
P Craig, ‘Judicial Review, Methodology and Reform’, forthcoming Public Law January 2022.

31

Electronic copy available at: https://ssrn.com/abstract=3897906

the context of the HRA has been rejected by the Joint Committee on Human Rights. 69 The

apposite point for present purposes is that a particular conception of parliamentary sovereignty

frames these inquiries and the current executive’s approach to them.

Finally, there is a second order dimension to the way in which parliamentary

sovereignty frames UK constitutional discourse. This connotes the way in which the concept

underpins, or influences, other constitutional doctrine. Consider, by way of example, the law

relating to parliamentary privilege, and judicial strictures that courts should not look beyond

the parliamentary roll. There are assuredly multiple purposes served by doctrines of

parliamentary privilege, and they are not logically dependent on parliamentary sovereignty.

This can be accepted. There is, nonetheless, a sense in which the overall contours of such

doctrine are shaped by parliamentary sovereignty. The primacy accorded to ordinary statute

grounded in Parliament’s electoral legitimacy is external in its orientation. The privilege

afforded to Parliament in the conduct of its own proceedings is an internal manifestation of

Parliament’s sovereign position, a recognition that it should be master of its internal domain.

4 Conclusion

There are, to draw on a common metaphor, many views of the cathedral. This very fact is

enriching, since it can thereby furnish fresh perspectives on oft-discussed issues. This chapter

does not claim such novelty in this respect. It has, nonetheless, sought to highlight the

significance of the difference between the normative precepts that underpin the standard

written constitution and the unwritten constitution that has parliamentary sovereignty as its

centrepiece. The normative rationale for the ascription of parliamentary sovereignty has, as we

69
Joint Committee on Human Rights, The Government’s Independent Review of the Human Rights Act (HC 89,

HL Paper 31, 8 July 2021).

32

Electronic copy available at: https://ssrn.com/abstract=3897906

have seen, altered over time and it continues to shape the modality and frame for constitutional

change in the UK.

33

Electronic copy available at: https://ssrn.com/abstract=3897906