Introduction
The doctrine of parliamentary sovereignty is the invariable starting point for UK constitutional theory. It is also the inevitable stumbling block when it comes to civil liberties and individual rights. Although the Human Rights Act 1998 (HRA) has had a significant impact on administrative law, by altering the balance of power between public bodies and the courts, it has not significantly strengthened the constitutional protection of rights and liberties in the UK. Its lack of legal entrenchment renders it vulnerable to future change. Moreover, the UK courts interpret neither the common law nor the European Convention on Human Rights (ECHR) to protect freedom of speech as ‘the matrix, the indispensable condition of nearly every other form of freedom’.Footnote 1 This reluctance creates a space for another concept, like democratic legitimacy, to take root.
Democracy is, of course, a multifarious concept. It ‘lacks a clear narrative line and conspicuously fails to carry its own meaning clearly on the surface’.Footnote 2 It also loses its value if it is reduced to its literal meaning. Conversely, the temptation to open the concept up to endless contestation or to overburden it with normative goals must also be resisted. The added value of democracy as a concept in constitutional discourse only becomes apparent if it can be presented as the conceptual starting point or the ‘indispensable condition’ on which the viability of all other constitutional concepts, including parliamentary sovereignty, depend. As Sartori notes:
Up until the 1940s people knew what democracy was and either liked it or rejected it; since then we all claim to like democracy but no longer know (understand, agree) what it is. We characteristically live, then, in an age of confused democracy. That ‘democracy’ obtains several meanings is something we can live with. But if ‘democracy’ can mean just anything, that is too much.Footnote 3
If democracy cannot mean anything, it must mean something. I will use this paper to explore three questions. First, in what situation does the absence of a concrete understanding of democracy become an inescapable problem for constitutional law? Secondly, to what extent are the existing constitutional models democratically deficient? Thirdly, what precisely must democracy prescribe as the indispensable condition for political legitimacy?
The inescapable problem
In this section I will identify three categories that illustrate the problems that stem from an absolutist conception of parliamentary sovereignty. The normative gap that a legal conception of democracy could close becomes apparent only in the final category. The first category consists of largely hypothetical and absurd examples. They imagine legislation that condemns all red-haired males to death,Footnote 4 requires the killing of ‘all blue eyed babies’,Footnote 5 or disenfranchises or discriminates against a particular group on arbitrary grounds.Footnote 6 Lord Hope summarily dismissed these examples as unhelpful: ‘Parliamentary sovereignty is an empty principle if legislation is passed which is so absurd or so unacceptable that the populace at large refuses to recognise it as law.’Footnote 7
Most scholars prefer to dismiss such examples as ‘unlikely, immoral or undesirable things which no one wishes it to do’.Footnote 8 In truth, a state that mandated the killing of blue-eyed babies would not be a legitimate state, let alone a democratic one. It would, in common parlance, be a rogue or a failed state. For that reason these examples do not even begin to address normative questions of democracy.
The second category involves violations of civil liberties with a national security or public safety dimension. Within this category are high-profile speech cases, such as Tisdall,Footnote 9 Ponting,Footnote 10 Spycatcher Footnote 11 and Shayler,Footnote 12 which involve criminal proceedings in the context of secrets, spies and whistle-blowers.Footnote 13 Other examples within this category involve indefinite detention, arbitrary arrests, broad police powers, data retention and surveillance powers. These examples challenge a state committed to civil liberties and the rule of law, and can be addressed with the technical administrative law language of rationality, proportionality and ECHR compliancy. They clearly also have implications for the democratic quality of a state. The House of Lords Constitution Committee, for instance, recognised that the erosion of privacy ‘weakens the constitutional foundations on which democracy and good governance have traditionally been based in this country’.Footnote 14 But these examples do not determine the democratic quality of a state: even a non-democratic state can desist from torture, respect due process and protect privacy.Footnote 15
The third and final category involves violations of civil liberties without a security or safety dimension. The first few years after the HRA came into force saw the cases of Percy,Footnote 16 Norwood Footnote 17 and Hammond.Footnote 18 These cases were picked up by civil liberties scholars because in each of these cases the individual's right to free expression had been restricted on ground of a pressing social need.Footnote 19 Hammond involved a street preacher in Bournemouth who held up a sign bearing the words ‘Stop Immorality’, ‘Stop Homosexuality’ and ‘Stop Lesbianism’, but who did not incite violence. Mr Hammond was arrested, charged and fined (upheld on appeal) for an offence under s 5 of the Public Order Act 1986. The court found that the link between homosexuality, lesbianism and immorality was capable of being insulting and of causing harassment, alarm or distress to a person standing nearby.
To be sure, the law has changed since the triplet of Percy, Norwood and Hammond were decided. The wording of s 5 at the time made it an offence to use ‘threatening, abusive or insulting words or behaviour, or disorderly behaviour’ or to display ‘any writing, sign or other visible representation which is threatening, abusive or insulting’ within the hearing or sight of a person ‘likely to be caused harassment, alarm or distress thereby’. After a high-profile campaign and a governmental defeat in the House of Lords, s 57 of the Crime and Courts Act 2013 removed the word ‘insulting’ in ss 5(1) and 6(4) of the Public Order Act 1986: the crime now requires ‘threatening or abusive’ words or behaviour. According to the Crown Prosecution Service (CPS), ‘the amendment is intended to enhance the protection of the right to freedom of expression under Art 10 of the European Convention on Human Rights (ECHR)’.Footnote 20 It might therefore be claimed that cases like Hammond would not be decided the same way today. However, three notes of caution need to be inserted.
First, both s 4(1) (fear or provocation of violence) and s 4A(1) (intentional harassment, alarm or distress) of the Public Order Act 1986 continue to proscribe ‘insulting’ behaviour. In other words, where the insulting words or behaviour are planned and malicious, a person could still be guilty of a criminal offence.
Secondly, a court can still punish identical conduct as ‘abusive’. After the government agreed to amend s 5, the then Home Secretary Theresa May told MPs:
Looking at past cases, the Director of Public Prosecutions could not identify any where the behaviour leading to a conviction could not be described as ‘abusive’ as well as ‘insulting’. He has stated that ‘the word “insulting” could safely be removed without the risk of undermining the ability of the CPS to bring prosecutions’.Footnote 21
To illustrate this point, in Abdul v DPP Footnote 22 the Crown Court had convicted the defendants of a Public Order Act 1986, s 5 offence. The protesters in question had shouted ‘burn in hell’, ‘murderers’ and ‘baby killers’ at soldiers returning home from Iraq. On appeal to the High Court, the question was whether their conviction for these utterances was compatible with Art 10 ECHR. It is of interest that the soldiers themselves were not bothered ‘one jot’ by the protesters,Footnote 23 and that the police made their arrests not during the demonstration but only months later after having reviewed film footage of the events. Nonetheless, the High Court found that ‘the words shouted by the defendants were both abusive and insulting’.Footnote 24 The first instance judge had, according to Gross LJ, carefully balanced freedom of expression against ‘a very clear threat to public order’,Footnote 25 and reached the right conclusion that prosecution was a proportionate response.Footnote 26 Davis J added that the freedoms of expression and assembly are subject to ‘duties and responsibilities’.Footnote 27 Both judges agree that the defendants’ actions had gone ‘well beyond legitimate expressions of protest’.Footnote 28
Thirdly, even without s 5 of the Public Order Act, prosecutors have access to ample legislative provisions to deal with ‘public order’ and ‘communication’ offences.Footnote 29 Section 127(1)(a) Communications Act 2003 creates an offence if three conditions are met: (i) a message; (ii) that is grossly offensive; and (iii) is sent by means of a public electronic communications network. Lord Bingham helpfully sets out the legislative history of the provision in DPP v Collins. Footnote 30 The purpose of the provision is to prohibit the sending of a grossly offensive communication via a public communications service. A letter that was personally delivered through a letterbox would, accordingly, not fall within this legislation – although it may be covered by s 1 Malicious Communications Act 1988.
The majority of cases dealing with bullies, trolls and stalkers on social media websites involve s 127.Footnote 31 In an attempt to stem the rise of s 127 prosecutions, the then Director of Public Prosecutions (DPP) Keir Starmer issued guidelines in 2012 which sought to establish a high threshold for launching criminal action against spontaneously written digital communications. Most notably, he made the following statement on the nature of offence:
The distinction [between offensive and grossly offensive] is an important one and not easily made. Context and circumstances are highly relevant and as the European Court of Human Rights observed in the case of Handyside v UK (1976), the right to freedom of expression includes the right to say things or express opinions ‘ … that offend, shock or disturb the state or any sector of the population’.Footnote 32
But mere guidelines do not resolve the issue, especially when it relates to the clarity, certainty and predictability of the fundamental right of freedom of expression. How strong is the protection of speech in the UK? Is it possible to identify
a profound national commitment to the principle that debate on public issues should be uninhibited, robust, and wide-open, and that it may well include vehement, caustic, and sometimes unpleasantly sharp attacks on government and public officials?Footnote 33
The notion that free speech is ‘bred in the bone of common law’,Footnote 34 or that ‘people are free to say and print what they like’ at common law,Footnote 35 is disputable. It clearly did not apply to Mr Lemon, who was prosecuted for blasphemy at English common law in the 1970s,Footnote 36 or Mr Hammond.Footnote 37 In any event, trial courts usually refer to Art 10 ECHR (freedom of expression). The potency of Art 10 ECHR, however, is destabilised by a number of qualifications, including interests of national security, territorial integrity, public safety, the prevention of disorder or crime, the protection of health or morals and the protection of the reputation or rights of others. Legal stability is further undermined when trial courts invert the operation of Art 10 ECHR, as happened in Hammond and Abdul. Instead of recognising the individual's right to expression, which ought to persist unless and until the state can justify a limitation, the courts began by asking whether the defendant's conduct was insulting and likely to harass, annoy or distress bystanders. If that was the case, the courts asked a second question, namely whether the conduct might nonetheless have been ‘reasonable’ or ‘legitimate’ under Art 10 ECHR. By erroneously privileging public peacefulness over freedom of expression, criminal convictions have been presented as a necessary limit on the right to speak under Art 10(2) ECHR. Davis J's casual remark in Abdul v DPP, that s 5 Public Order Act 1986 is ‘obviously … compatible with the Convention’,Footnote 38 is symptomatic of that approach. However, the more important question is whether the criminalisation of behaviour that is not of itself violent, but is threatening, abusive, insulting or disorderly, is necessary in a democratic society.
Freedom of expression is not only vital as an individual right, but as a constitutive element of democratic society. According to Weale, the values of democracy are ‘best understood in terms of the protection and promotion of common interests constrained by political equality and in conditions of human fallibility’.Footnote 39 In that context, the common interest commands possibilities for dissent and difference: ‘dissent does not undermine a society but underpins it.’Footnote 40 A legal concept democracy must, therefore, guarantee free speech – not as a constitutionally guaranteed individual right, but as a political or public right, and as a condition of democratic legitimacy.
Unfortunately, as we have seen, UK courts have tended not to approach freedom of expression of individual dissenters with the same degree of principle and conviction as, say, First Amendment jurisprudence in the USA. Domestic courts view deeply inflammatory speech as pernicious, and as promoting distrust, suspicion and violence between different social groups. They have indiscriminately constrained or criminalised the articulation of offensive and extremist ideas that fall short of criminal incitement. The cases discussed above are not isolated ones. Philip Johnston's opening chapter starts with the striking claim that ‘more people are being jailed or arrested in Britain today for what they think, believe and say than at any time since the eighteenth century’.Footnote 41 Brice Dickson claims that the attitude of the House of Lords and UK Supreme Court towards freedom of expression ‘cannot be regarded as particularly fervent’. He attributes this to the absence of constitutional guarantees as well as to ‘a strain of conservatism on the part of judges in our top court, most of whom do not seem to view free speech as deserving of extra-special protection’.Footnote 42 Eric Barendt finds no evidence that the HRA has radically altered the legal protection of free speech, and little evidence that UK courts treat freedom of expression ‘as the starting point; they have not always asked whether the restrictions on its exercise were necessary to safeguard public order or the other end for which they were imposed’.Footnote 43 James Weinstein concludes that the English right to free speech ‘is both too weak and too indeterminate to adequately protect the public expression of ideas that “offend, shock or disturb” dominant opinion’.Footnote 44 This state of affairs should be of concern not only to anyone with an interest in constitutional law, but also to anyone with an interest in democracy and legitimacy.
The next two sections will reveal the democracy deficit in the two rival constitutional models. The UK's constitutional traditions, concepts and vocabulary developed to describe a functioning, balanced and liberal constitution. Ideas of democracy and universal freedom were ‘invariably not forged afresh but rather tentatively grafted onto a pre-existing society that had been designed for the few’.Footnote 45 Although the origin of nineteenth-century concepts, such as ‘sovereignty’, ‘constitution’, ‘rule of law’, ‘judicial review’ and ‘separation of powers’, is pre-democratic, they complement our understanding of democracy today: an Act of Parliament is the highest source of law, and everyone is subject to the same laws that are interpreted by an independent judiciary. But that still leaves a gap in relation to the recognition of basic rights in terms of ‘democratic legitimacy’, the common good or ‘political’ and ‘public’ rights. The literature on UK constitutional law does not usually assess legitimacy with reference to such specifically democratic criteria. The two main models are both culpable for not adapting constitutional theory to fit the requirements of a modern democratic state.
Democratic deficiency I: The political constitution
The Westminster model of government encapsulates the so-called political constitution.Footnote 46 This model conceives of democracy as electoral representation, which is ordinarily deemed to legitimise all parliamentary legislation. The political constitution is rooted in the nineteenth-century concept of representative or parliamentary government. Its central presuppositions, such as the continuing sovereignty of Parliament, a procedural conception of the rule of law, and basic civil liberties predate the arrival of mass democracy. They reached their high point during the late Victorian constitution. Ultimate legal authority was still divided amongst the three estates of Parliament, and the franchise, although extended, was not yet universal. Although civil liberties expanded ratione personae, they remained fragile ratione materiae due to their lack of constitutional entrenchment.Footnote 47 ‘There are not, under English domestic law, any fundamental constitutional rights that are immune from legislative change.’Footnote 48
Anthony Bradley noted accurately in 2011 that the debates based on the model of the political constitution still revolve around a static and immutable understanding of parliamentary sovereignty and a subordinate role for the courts.Footnote 49 The conceptual starting point remains strikingly absolutist and gives rise to a legitimate concern about potentially oppressive and undemocratic outcomes: the legal principle of sovereignty by which parliament provides unconstrained legal authority for governmental policy choicesFootnote 50 is connected to the ‘political principle that in a democracy there should be no legal limit to the wishes of the people’.Footnote 51
This absolutist premise requires scrutiny. The legal principle of parliamentary sovereignty was born in the seventeenth century prior to the advent of mass democracy. As Bernard Manin notes, universal suffrage expanded the body of the electorate without transforming the undemocratic nature of the constitution: ‘ … there has been no significant change in the institutions regulating the selection of representatives and the influence of the popular will on their decisions once in office’.Footnote 52 The purpose of the Reform Acts of 1832, 1867 and 1884 was to resist the kinds of democracy sought by Radicals and Chartists by focusing reform on parliamentary procedure and electoral districts.Footnote 53 Ironically, parliamentary sovereignty received a boost by the growth in franchise, to the surprise of many observers who were sceptical as to its operation under universal suffrage.Footnote 54 Popular sovereignty was effectively absorbed by and channelled through Parliament. This had two consequences. First, the entire reform process from 1832 onwards occurred in the name of parliamentary government, which ‘fostered the conviction or delusion that the will of the nation could be expressed only through elected representatives’.Footnote 55 Secondly, with a more inclusive franchise, democracy became a convenient concept for the ruling elites who appropriated and contained the concept in order to ‘counteract the illusion that any random act was now possible’.Footnote 56 But it was too late. In the UK at least, parliamentary sovereignty was now so powerful that it became associated with a variant of absolute power that gave rise to concerns about ‘elective despotism’, a ‘plebiscitary dictatorship’Footnote 57 and Parliament ‘legibus solutum’.Footnote 58
The absence of normative benchmarks to constrain the omnipotent Parliament is also ‘unfortunately absolutely characteristic of English writing on constitutional law’.Footnote 59 Constitutional law, in spite of its common usage, is ‘not a technical phrase of English law’.Footnote 60 It is silent on governing principles underlying the constitution,Footnote 61 it does not address questions regarding the social foundations of law's legitimacyFootnote 62 and it lacks the higher law quality of modern constitutional documents. The core of the political constitution has not changed since De Lolme wrote that ‘the legislature can change the constitution, as God created the light’.Footnote 63 To this premise were added the nineteenth-century desiderata of representative and parliamentary government. This was followed by the twentieth-century view that a constitution should foster political opportunities for individuals on formally equal terms through the processes of representation and effective participation with the purpose of promoting sound governmental decisions.Footnote 64 The democratic process, narrowly conceived, allows the electorate ‘to turn out any government that it does not like’.Footnote 65 For Richard Bellamy, that ‘democratic process is the constitution’.Footnote 66 In other words, the constitutional view of democracy is purely procedural. But as Ewing acknowledges, the democratic process is not the same as a democratic constitution.Footnote 67
The political constitution and the Westminster model of government prioritise the source of the decision and presume the legitimacy of majoritarian decision making. Contrary to the central tenets of Western constitutional theory, nothing is protected from the ordinary legislative processes, and everything is ‘up for grabs’.Footnote 68 Jeffrey Goldsworthy looks favourably upon the unlimited nature of the political constitution: ‘[procedurally] democratic decision-making is facilitated, and reasonably just statutes are enacted.’Footnote 69 Bellamy takes the same view on democratic rule: ‘… the demos should be free to redefine the nature of their democracy whenever they want and not be tied to any given definition.’Footnote 70 Intricate questions about justice and human rights are, therefore, not to be resolved by recourse to first principles, but by the ‘opinion of a majority of the people of those elected to represent them’.Footnote 71 The principal remedy lies in ministerial responsibility to Parliament, which in the late Victorian age still conformed to the Burkean ideal of a lower house with independently minded MPs, but which today has been transformed by party discipline and the whip system.Footnote 72
Even Stephen Gardbaum's synthetic account of a ‘third model’ of constitutionalism ultimately rests on a purely procedural conception of democracy that assigns ultimate decision-making power to ordinary majority vote in the legislature.Footnote 73 He views as ‘compelling’ the argument that a legislative majority should ‘trump’ the views of a judicial majority. The political constitution regards the enactment of oppressive and undemocratic legislation as repugnant only to ‘moral principle’.Footnote 74 The legitimacy question has always been brushed off as a non-issue. Extreme cases have been kept at bay by ‘the sweet reasonableness of MPs and their constituents’,Footnote 75 ‘because Parliament has not been extreme’Footnote 76 and because Acts of Parliament ‘are rarely obviously and egregiously unjust’.Footnote 77 In short, the UK Constitution assumes that ‘the worst will not happen in the first place’.Footnote 78
The cases discussed in the third category above are clearly not ‘the worst’ that can happen. For reasons mentioned at the start I have chosen not to discuss the UK's attempts to allow the government to gag a newspaperFootnote 79 and a former spy;Footnote 80 to ban all homosexuals from serving in the military;Footnote 81 to indefinitely detain foreign terrorist suspects;Footnote 82 to criminalise speech that ‘encourages’ terrorism, even if the person making the statement does not intend to encourage terrorism;Footnote 83 to ban non-violent political organisations;Footnote 84 and to sanction some of the most sweeping surveillance powers in the Western world.Footnote 85 And even these examples stop short of a particular Rubicon that some of the most senior judges think Parliament should not cross, namely a legislative attempt to remove government action affecting individual rights from judicial scrutiny.Footnote 86 But Hammond and Abdul do reveal some home truths about the UK Constitution. Although free speech is routinely hailed as a hallmark of democratic society,Footnote 87 it is ‘residual’ to the common law,Footnote 88 and far from being recognised as an indispensable condition of legitimacy.
In the end, the political constitution explains nothing more than the Westminster way of doing politics.Footnote 89 John Austin's legal positivism, HLA Hart's analytical jurisprudence, AV Dicey's conservative normativism and John Griffith's functionalism all define law and legality in purely formal terms. Their normative blindness is self-serving. They perpetuate the political constitution by focusing only on the source of power. They favour representative and responsible government over substantive democratic government. They define arbitrariness narrowly as the absence of procedure. They equate legitimacy with legality. Their failure to grasp political questions as constitutional questions, and questions of legality as questions of legitimacy, baffles Continental commentators, yet is venerated domestically as reflecting the flexible and living constitution.
UK constitutional law has not produced a normative account of democracy beyond practical decision making. Without limits on popular sovereignty, without ground rules, with ‘everything up for grabs’, democracy becomes wholly incidental to the political constitution. Democracy collapses into the recapitulation of the status quo: whatever happens in a democracy is democratic. On this account, Lord Sumption's statement that ‘Parliament may do many things which undermine the democratic element of our constitution’,Footnote 90 and Lord Steyn's concern in Jackson regarding the enactment of ‘oppressive and wholly undemocratic legislation’,Footnote 91 are contradictions in terms. Such a relativistic account views the procedural conditions of periodic and competitive elections and effective, equal and universal participation as the necessary and sufficient conditions of democracy, which leaves a normative account of democracy under-determined, under-theorised and under-valued.
Democratic deficiency II: The common law constitution
The artifice of the common law constitution has been developed in opposition to the political constitution. It is animated by similar questions that form the backdrop of the present paper: what are the limits of legislative supremacy and what are the limits of the practice of judicial obedience to statute? In an attempt to counter orthodoxy with orthodoxy, this model rejects the doctrine of legislative supremacy in favour of the ancient common law tradition, which is re-imagined as a higher-order ersatz constitution.Footnote 92 The common law, it is claimed, contains a set of moral principles, which are transformed into law through the exercise of ‘artificial reason’.Footnote 93 Unformalised and undeclared, the moral principles are said to unite all common law jurisdictions and are said to be immune from judicial and parliamentary abrogation.Footnote 94 However, the invocation of mere morality to limit Parliament has meant that common lawyers since Coke and Hale have had to accept that Parliament retains the formal legal power to limit freedom and act oppressively. In the final analysis, the common law advances no more than a moral argument about what Parliament ought not to do, and not a constitutional argument about what Parliament cannot do.
Trevor Allan argues that an extensive conception of the rule of law acts as ‘a constitutional principle of real importance, capable of moderating the influence of majoritarian politics, especially in times of emergency or stress’.Footnote 95 The claims of the common law offer an alternative account of UK constitutional law. Formal equality is replaced with a substantive account that includes moral-turned-legal principles of the common law, such as equality, rationality, proportionality, fairness and basic rights.Footnote 96 Judges are elevated from their previously subordinate role to guardians of the common law constitution and of democracy, who assess the content of individual rights with reference to moral and ethical principles.Footnote 97 And the common law constitution creates a space for liberal rights and limits on legislative authority – in other words, the kind of constitutional debate which the finality of sovereignty in the political constitution forecloses by treating it as irrelevant, unnecessary and a waste of political resources.Footnote 98
However, the common law constitution continues to adhere to a pre-democratic and constitutional form that accepts Parliament's ability to limit freedom by legislation. Allan undoubtedly advances constitutional discourse by connecting parliamentary sovereignty, the rule of law and separation of powers to the world of politics and morality, which he regards as the pillars of liberty, justice, democracy and legality.Footnote 99 In so doing, he replaces the political constitution with a moral constitution.Footnote 100 The claim that the common law is the judicially recognised source of parliamentary sovereignty becomes not only factually correct (courts, public authorities and private individuals obey Acts of Parliament) but also morally desirable: sovereignty ought to reflect democratic notions of equality as well as a constitution ‘rooted in fundamental moral values’.Footnote 101
Drawing on Lon L Fuller's theory of moral law that respects the demands of human dignity, Allan coopts liberty and justice to infuse law with moral legitimacy and appeals to the reader's ‘moral and political judgement’.Footnote 102 The law does not command obedience because of its source, eg parliamentary sovereignty, but because of its substance, which derives from (i) the moral integrity of legislation and (ii) personal responsibility and judgment.Footnote 103 In shifting the rule of recognition from a source-based to a content-based conception, Allan concludes that ‘the sovereignty of Parliament is only a manifestation of the sovereignty of law’.Footnote 104 Statute law is a composition of a ‘present tense’ interpretationFootnote 105 of both the text and of the principles of the common law, which makes it in part ‘a product of our moral judgement’.Footnote 106
It is an unfortunate but inevitable consequence of analysing the constitution through the prism of analytical jurisprudence, and in particular the relationship between law and morality, that the key concepts of rule of law, legitimacy, democracy, liberty and free speech all remain underdeveloped. Although Allan asserts the rule of law as a ‘constitutional principle of real importance’ that preserves individual autonomy and independence, on closer reading it amounts to no more than a guarantee against arbitrary interference.Footnote 107 As a reviewer of the book notes, Allan's conception of ‘legitimacy is thicker than mere legality, but is nevertheless a thin legitimacy’.Footnote 108 The objections in the previous section notwithstanding, the idea of democracy is ‘clearly related’ to parliamentary sovereignty,Footnote 109 which aligns Allan with the dominant tradition. Three presuppositions contribute to the weakness of the common law model. First, Allan accepts the principle that ‘a political majority may legitimately impose its will on a dissenting minority’ – he questions only the extent.Footnote 110 Secondly, with the political constitutionalists, he accepts that parliamentary democracy and majoritarianism ‘are the appropriate means for settling the content of law’.Footnote 111 Thirdly, he defines liberty in negative terms as that which protects ‘a private sphere of thought, deliberation, and action consonant with his dignity as a free and independent citizen’.Footnote 112 This allows Allan to endorse ‘draconian restrictions on personal liberty’ on grounds of overriding interests of public safetyFootnote 113 and with the aid of ‘appropriately rigorous requirements of due process or procedural fairness’.Footnote 114 Even on a key question, whether Parliament could expressly curtail freedom of expression at election time, Allan offers only a weak response, saying it would be ‘a matter of opinion’.Footnote 115 Such subjectivism and relativism hardly amount to a principled defence of ‘uninhibited, robust, and wide-open’ speech as an essential feature of a liberal democratic state.Footnote 116
Allan concludes that egregious legislation (such as bills of attainder or statutes permitting the killing of all blue-eyed babies) would not qualify as ‘law’ due to the absence of ‘generality’. This response provides further evidence of Allan's thin conception of legitimacy: ‘generality’, ‘due process’ and ‘procedural fairness’ are criteria of the procedural rule of law conception. Allan invokes standard administrative law terminology when he declares that such legislation would be ‘wholly unreasonable or irrational [and] cannot qualify as a valid law’.Footnote 117 Contrary to Allan's argument, this technical objection does not suffice to ‘elucidate the features of what (in common with most of our fellow citizens) we take to be a legitimate scheme of government, worthy of our attention and loyalty’.Footnote 118
To the contrary, the fixation on providing legalistic responses to political problems inadvertently plays into the hands of, say, the US Supreme Court in its disreputable Dred Scott decision. In that case the Court ruled that the Missouri Compromise (1820), which had declared free all territories acquired after the creation of the USA, was unconstitutional. The US Supreme Court reached this conclusion on the ground that Congress's prohibition of slave-holding in the Western territories deprived Mr Sandford of his slave property without due process of law.Footnote 119 Constitutionally, the decision is problematic for the absence of judicial deference to Congress and the Supreme Court's failure to presume statutory validity. From a democratic standpoint, however, the decision stands out for its violations of substantive principles, such as liberty and equality.
The common lawyers’ equation of the legal constitution with a moral constitution is premised on an understanding of public law as ‘rationalist metaphysics’ that reveals itself as an ‘aesthetic preference, not an epistemological advance’.Footnote 120 This approach fails to understand the constitution as a site of political contestation. Instead of justifying claims about parliamentary sovereignty on grounds of history or principles of morality, they ought instead to be ‘rooted in an appreciation of the nature of the contemporary political condition’.Footnote 121 For Loughlin it means appreciating the autonomy of the political sphere and of droit politique. In the next section, I argue that it means appreciating the requirements of democracy.
Democracy as the legitimating condition
The existing constitutional models based on parliamentary sovereignty and the common law fail in their accounts of democracy. The political constitution simultaneously under-theorises democracy, by reducing it to an electoral condition or majoritarian procedure, and overextends it by treating popular and parliamentary sovereignty as legally limitless. Without democratic guarantees, freedom includes the freedom to change or abolish the social and legal order. The constitution does not contain predetermined rules, but simply facilitates the process of making new rules or keeping existing ones. Democracy's key index is fairness of procedure, which too is subordinated to Parliament's overriding legislative authority.
The common law constitution at least recognises that democracy contains biases in favour of individual freedoms and against arbitrary government and, insofar as its form is republican, against political domination. These biases potentially clash with the prevailing conceptions of sovereignty as absolute, of the constitution as political, and of democracy as purely procedural. The common law's account today is distinctly liberal and moral, but it is not distinctly democratic.
By contrast, I propose an assessment that enquires into the extent to which statutes, policy and case-law support an ideal of politically active citizenship. In a seminal speech marking the bicentennial of the US Constitution in 1987, Thurgood Marshall, the first African American member of the US Supreme Court, gave credit not to the ‘defective’ vision of the founding fathers, but ‘to those who refused to acquiesce in outdated notions of “liberty,” “justice,” and “equality,” and who strived to better them’.Footnote 122 In the same mould, UK constitutional scholars need to supplement their established criteria of legal validity with different conditions of political legitimacy. Those who seek to justify the decisions in Hammond, Abdul and similar cases by virtue of their legal consistency with the Public Order Act 1986 fail to notice that these decisions are unjust, wrong-headed and not just unnecessary in but incompatible with a democratic society.
Without the connection between formal legality and substantive legitimacy, the absolutist conceptions of parliamentary sovereignty and popular democracy become indistinguishable from Franz Neumann's definition of dictatorship as ‘the rule of a person or a group of persons who arrogate to themselves and monopolize power in the state, exercising it without restraint’.Footnote 123 The Westminster model ironically ends up sharing the same starting point of legally unlimited power as dictatorship. It is this coupling of absolute parliamentary sovereignty with a minimalist conception of democracy as electoral and majoritarian that gives rise to the tension between legality and legitimacy that underpins oppressive law and policy.
Three steps need to be made in order for the UK to embrace a concept of democracy that transcends its electoral-representative starting point. First, democracy must not be relativised as a ‘complex’ and ‘contested’ concept. The claim that democracy defies definition and is susceptible to wholly open-ended interpretations is a widely accepted truism that ought to be rejected as false.Footnote 124 The rule of law is a similarly complex and contested concept. Without agreement on its meaning, it is perfectly possible, and indeed is a current trend, for authoritarian regimes to proffer formal acceptance of the baselines of the rule of law (legal certainty and stability over economic rights) and basic democracy (elections, representation and majoritarianism) but to reject openly the requirements of civil rights and political freedoms.Footnote 125
In a second step, democracy needs to be distinguished from precepts held in common with liberalism, principally equality and freedom. First, can equality be shown to be distinctly democratic? Its negative and formal sense privileges a purely abstract and civic conception to the detriment of material (social and economic) interests. Material equality, however, could also be realised in non-democratic states. For Nadia Urbinati, therefore, formal political equality is the reference point in relation to which citizens assess the political process of democratic representation.Footnote 126 She distinguishes two kinds of formal equality, already present in Athens. Isonomia relates to strict equality before the law, which is satisfied through representation and the rule of law, and may also be found in non-democracies.Footnote 127 Isegoria, on the other hand, concerns an equal right to access the political assemblies and to support or oppose laws or policies. This conception concerns judgment and interpretation in complex and plural societies.Footnote 128 It relies on participation and active citizenship. As a result, ‘this equality is exquisitely political and democratic; in fact, it exists only in a democracy’.Footnote 129 For Urbinati, therefore, democratic equality is political equality, ie participation and active citizenship, which provides the link with freedom.
Secondly, is ‘freedom’ a distinctly democratic concept? The neo-Kantian ideal manifests itself, according to Hans Kelsen, in the historical ‘struggle for democracy’, which he interprets as a ‘struggle for political freedom; that is, for popular participation in the legislature and executive’.Footnote 130 Franz Neumann also places ‘the activist element of political freedom’ at the heart of the democratic political system, the essence of which does not lie in majoritarian decision making, ‘but in the making of politically responsible decisions’ and in ‘large-scale social changes maximising the freedom of man’.Footnote 131 According to Giovanni Sartori, freedom is the constitutive element of liberal democracy, although not necessarily of democracy as such.Footnote 132 In other words, these theorists collapse a democratic conception of freedom into a liberal conception. Unlike Urbinati, who privileges equality over liberty, Sartori claims that liberal democracies strive to achieve ‘equality through liberty’:Footnote 133 ‘From liberty we are free to go on to equality; from equality we are not free to get back to liberty.’Footnote 134 But both Urbinati and Sartori reduce democracy to the liberal values of political equality and individual freedom, neither of which are distinctly democratic.
Another liberal account of democracy stems from Ronald Dworkin, who argues that legitimate majority rule must necessarily require the existence of structural conditions above and beyond the principle of majority. A majority cannot, using democratic means, abolish future elections or disenfranchise a minority.Footnote 135 Dworkin endorses a constitutional conception of democracy that consists of three conditions: (i) ‘a majority or plurality of people’; (ii) ‘all citizens have the moral independence necessary to participate in the political decision as free moral agents’; and (iii) ‘the political process is such as to treat all citizens with equal concern’.Footnote 136 The problem is that these structural conditions correspond to basic liberal principles of popular sovereignty, political participation and citizen equality. None of the criteria is manifestly democratic. Moreover, Dworkin does not distinguish between different types of democracies (eg Western, transitional, authoritarian etc). Instead, Dworkin's ‘naïve universalism’ lies in inducing absolute principles of democracy from US constitutional principles.Footnote 137
The third and final step is to theorise democracy on its own terms. Jürgen Habermas rejects Dworkin's liberal-individual, and its rival republican-communitarian,Footnote 138 model of democracy in favour of a ‘procedural’ one. Sovereignty is generally conceived, with Jean-Jacques Rousseau, as harnessing the general will of a collective subject. However, in Habermas’ terms popular sovereignty becomes ‘desubstantialised’, anonymised and realised ‘through the communicative presuppositions and procedures of an institutionally differentiated opinion- and will-formation’.Footnote 139 By forcing popular sovereignty into ‘subjectless forms of communication’ and democratic procedures,Footnote 140 Habermas wrestles the concept from Rousseau and secures it as ‘the hinge between the system of rights and the construction of a constitutional democracy’.Footnote 141 ‘The democratic process bears the entire burden of legitimation’ by securing the private autonomy of legal subjects through individual rights and their public autonomy through communicative freedoms.Footnote 142
Habermas’ subjectless forms of communication have a liberal as well as a democratic component. On the one hand, they form the legal foundation for a free and open process of participation for constructing political will and opinion, which is indispensable for democratic freedom. On the other hand, they form the agreement on which to disagree by cutting short the infinite search for validity and justification criteria. The rights of communication determine the democratic character of the state and its laws and practices, but are themselves immune from democratic deliberation.
Both Dworkin and Habermas reject the simplistic maxim of the political constitution that any law properly enacted by the legislature is presumptively democratic. Instead of Dworkin's structural conditions, Habermas inserts his discourse principle according to which ‘the only law that counts as legitimate is one that could be rationally accepted by all citizens in a discursive process of opinion- and will-formation’.Footnote 143 This axiom bears a striking resemblance to Rawls’ ‘justice as fairness’, which ‘may be shared by citizens as a basis of a reasoned, informed, and willing political agreement’.Footnote 144 But where does this leave the expression of viewpoints that are not subject to rational acceptance or reasoned and informed political participation? What implications do these positions have, for instance, for the government's definition of extremism as ‘the vocal or active opposition to fundamental British values, including democracy, the rule of law, individual liberty and the mutual respect and tolerance of different faiths and beliefs’?Footnote 145
What makes these ‘British’ values fundamental is, presumably, that they are susceptible to rational acceptance by all (current and future) citizens as a basis for reasoned and informed political agreement, in accordance with Habermas and Rawls. But what about those advocates who are vocally and actively opposed to such values? Is there not a danger that, as one newspaper headline put it, ‘laws against “extremism” risk criminalising us all’?Footnote 146
Eric Heinze's claims are more targeted. He does not refer to all democracies, but only to longstanding, stable and prosperous democracies (LSPDs). He does not argue for an absolute freedom of speech, and he accepts many familiar limits on expression.Footnote 147 His argument is limited to the expression of ‘viewpoints’ aired within open, public discourse, including irrational, obnoxious and even dangerous world-views. Crucially, the expression of such views ought to be protected as the prerogative of every citizen, rather than as a liberal right or as an individual freedom. He proposes elements of citizenship, which, unlike standard corpuses of individual rights or freedoms, can never legitimately be subject to legislative or judicial balancing of interests on specifically democratic grounds. Governments may legitimately impose certain viewpoint-neutral restraints, for instance regarding commercial fraud, courtroom perjury or official secrets, and ‘time, manner, and place restrictions’.Footnote 148 However, Heinze's central thesis is that ‘the most distinctly democratic’ manifestation of free expression is viewpoint absolutism within public discourse,Footnote 149 and that ‘… no conceivable abridgement of that citizen prerogative could ever be deemed to promote democracy’.Footnote 150
If Heinze's argument is correct then the real threat to the specifically democratic element of a state – as opposed to threats to any state simply as a state, such as war, famine or environmental pollution – stems from laws seeking to gag the citizens’ prerogative to speak – which is precisely the issue in Hammond, Abdul and related cases. Similarly, from a civil liberties perspective, retaining the more standard constructs of higher-order rights, David Feldman rejects any suggestion that democracy assumes an unhindered power on the part of law makers to abridge such rights:
… it would be perverse to argue that there is anything undemocratic about a restriction on the capacity of decision-makers to interfere with the rights which are fundamental to democracy itself.Footnote 151
It does not follow from the above that any limitation of the expression of a viewpoint, for instance through hate speech bans, would wholly diminish the legitimacy of that democracy. Although bans are illegitimate within LSPDs, they delegitimise democracy only pro tanto: ‘The legitimacy of the entire democracy is not overcome by one defect. Nor, however, does a democracy's overall legitimacy suffice to overcome the defect.’Footnote 152
All three theories of democracy, by Dworkin, Habermas and Heinze, serve as a better vantage point for the discussion of UK constitutional law than the ‘political’ and ‘common law’ models. All three argue in unison that certain norms must be presupposed in any account of contemporary democratic authority and must lie beyond majoritarian politics. However, the three accounts ultimately differ. Dworkin's account of democracy is congruent with the liberal tradition that regards the constitution as a corset for those who wield power, and which conceives rights negatively, individually and instrumentally. More abstractly, Habermas rests democracy in the communicative conditions necessary for political participation and will-formation. More concretely, Heinze argues for the expression of all viewpoints as a prerogative of citizens in LSPDs. Neither Habermas nor Heinze view the constitutive norms of democracy as ‘corsets’ on a free political discourse. Instead, such norms substantiate ‘the necessary condition for having any discourse at all about how purposes are to be fulfilled in that society’.Footnote 153
Conclusion
Not all violations of constitutional principles or fundamental rights need to be assessed with reference to democracy. Some incursions in the area of human rights and civil liberties (blanket discrimination or indefinite detention, or deportation which compromises the absolute prohibition on torture, and broad police powers) may be analysed in relation to the rule of law and liberal constitutionalism. This paper cannot draw boundary lines that identify the precise conditions when specific issues such as electronic surveillance, extreme speech or indefinite detention cease to be compatible with the principle of democracy. In addition. compliance with devolution, and membership of the European Union and Council of Europe, may be scrutinised with respect to Elliott's useful concept of constitutionality.Footnote 154
However, the prevention or criminalisation of offensive, obnoxious or dangerous speech, without any security implications, raises separate and conspicuous issues for the UK as a contemporary democracy. The cases of Hammond and Abdul are paradigmatic. The defendants were arrested and fined for airing their personal views on same-sex relationships and the actions of UK soldiers in Iraq. The problem was not with how they said it; it was with what they said. If ‘there simply is no way that the speaker can express this core belief without risking such offence’, then the message is very clear ‘that society will not tolerate the public expression of his core beliefs’.Footnote 155 Viewpoint absolutism is, according to Heinze, the irreducible core of democratic legitimacy.Footnote 156 Criminalising the expression of viewpoints, therefore, resonates throughout the legal and political system as a general challenge to parliamentary sovereignty and the common law, and as a particular challenge to democratic legitimacy.
Asking what democracy requires in Hammond, Abdul and similar cases, is more principled than trading-off freedom of expression against the need to maintain public order within the framework of liberal constitutionalism. Criminalising non-violent free speech may not be against UK law. But it ought to contravene Art 10 ECHR, and it most certainly contravenes the criteria of democratic legitimacy. Hammond and Abdul illustrate not only damage done to individual rights and freedoms, but to the democratic credentials of the polis itself.Footnote 157
The legality of legislation will always depend on sovereign authority and formal procedure (highest might), but its legitimacy ought to derive from persuasion and social generality (highest right). The democratic legitimacy of laws stems from a complex constellation of requirements and conditions that no longer involves popular sovereignty alone, but also basic rights and liberties; not an overriding concern with public order, but an overriding concern with freedom; not just formal participatory rights, but an inclusive process of opinion- and will-formation; and not just the negative, individualist and liberal view of freedom as non-interference guaranteed by the rule of law, but the social and public conception of non-domination in a free, civic and democratic society. All of this is known. The untechnical phrase ‘constitutional law’ does not reflect it; Parliament's unqualified right to enact whatever law it thinks fit undermines it.
