The law and democracy: Cherish both, but keep them distinct

– Dr Frederick Wilmot-Smith, barrister and academic, All Souls College, University of Oxford

In his Discourses on Livy, Machiavelli claimed that “[w]here a thing works well on its own without the law, the law is not necessary; but when some good custom is lacking, at once the law is necessary.”[ref]Machiavelli, N (trans Mansfield, H, & Tarcov, N). 1531 [1996]. Discourses on Livy, Book 1, Chapter 3, p.15.[/ref] What about where the ‘thing’ is democracy? Can a democracy work well without the law? And if a democracy is not working well, can the law help?

A popular view holds that law and democracy are entwined. Lord Bingham, for example, the senior Lord of Appeal in Ordinary[ref]The formal name of the Law Lords, who carried out the judicial work of the House of Lords in an Appellate Committee. In October 2009, the Appellate Committee was replaced by the Supreme Court as the highest appeal court in the United Kingdom.[/ref] from 2000 until 2008, contended that democracy was a precondition of the rule of law.[ref]Lord Bingham. 2007. The Rule of Law, 66 CLJ 67, 84.[/ref] In an October 2024 lecture in Bingham’s honour, the Attorney General, Richard Hermer KC, reversed the order of priority. He claimed that “the rule of law – both domestically and internationally – is the necessary precursor to … democratic values”.[ref]Attorney General’s Office and The Rt Hon Lord Hermer KC. 2024. Attorney General’s 2024 Bingham Lecture on the rule of law. Available from: https://www.gov.uk/government/speeches/attorney-generals-2024-bingham-lecture-on-the-rule-of-law.[/ref] These claims suggest a way of approaching the questions I have posed. Before we can assess them, we need to know more about the constituent parts: the rule of law and democracy.

A singular virtue

The rule of law is a virtue of a legal system. In an account owed to the legal philosopher Lon Fuller, for a regime to achieve that virtue its laws must be general, publicly promulgated, non-retroactive, sufficiently clear, consistent, possible to comply with, and relatively constant through time.[ref]Fuller, L. 1964. The Morality of Law.[/ref] Yet even these features would do little good if officials disregarded the law. A further important condition emerges: officials’ actions should be congruent with the law.

Compliance with the rule of law can be valuable in various ways. A society which complies with the rule of law can, for example, secure a kind of freedom for citizens which no other regime could give: it is only if we have legal rights, protected by a system which complies with the rule of law, that we can be free from arbitrary and dominating influence. For the same reason, the rule of law may ensure a kind of equal standing between citizens. And the ideal may also contribute to good government: it reduces arbitrariness and may lead to better decisions. So there can be a lot to celebrate in a system which complies with the rule of law, and a lot to fear about its loss.

I have argued elsewhere that there are reasons to be concerned, on rule of law grounds, about the changes made to the legal system in recent years and decades.[ref]Wilmot-Smith, F. 2014. Necessity or Ideology? London Review of Books 36 (available from: https://www.lrb.co.uk/the-paper/v36/n21/frederick-wilmot-smith/necessity-or-ideology); —. 2015. Court Cuts. London Review of Books 37 (available from: https://www.lrb.co.uk/the-paper/v37/n15/frederick-wilmot-smith/court-cuts); —. 2019. Justice eBay style. London Review of Books 41 (available from: https://www.lrb.co.uk/the-paper/v41/n18/frederick-wilmot-smith/justice-ebay-style).[/ref] Two threads are particularly apparent. The first is the policy movement, increasing in prominence since perhaps the 1980s, which characterises litigation as a failure and proposes settlement of claims (especially through mediation) as a better solution. This movement stems from a deep but ancient problem with legal systems: they are very expensive to use. But it reacts to that problem with a solution antithetical to the rule of law: mediation itself does not seek to resolve cases justly according to law; it tries to get parties to negotiate a compromise. A vivid recent acceptance of this is the statement of a judge, when ordering parties to mediate, that the process leads people to “recognise the desirability of settling for less than their strict legal rights”.[ref]DKH Retail Limited & Ors v City Football Group Limited [2024] EWHC 3231 (Ch) at [39] (Bright J).[/ref]

The second thread is the attendant downgrading of legal disputes with a lower value, and prioritisation of those where more money is at stake. Consider, for example, the Ministry of Justice’s report from December 2023 that the average time for a claim for less than £10,000 to come to trial is now over a year. That is almost 20 weeks longer than the same process took in 2019. These legal processes do not need massive investment, but they are not a priority, it seems, because (unlike the high-value claims) they do not make money for the state. Such claims can be steadfastly denied right up until the eve of trial, and no sanction will be incurred; the unscrupulous can therefore delay the day of reckoning, or avoid it altogether if people give up or are bought off along the way. This is a rule of law concern because facts other than the legally relevant ones – who is legally obliged to whom, and for how much – become material.

Tempting tangles

Are these rule of law concerns also democratic concerns? I have written elsewhere that there is a danger in assuming that those two ideals march hand in hand: not every rule of law problem is a democratic problem, and vice versa.[ref]Wilmot-Smith, F. 2024. Good Vibrations. London Review of Books 46. Available from: https://www.lrb.co.uk/the-paper/v46/n17/frederick-wilmot-smith/good-vibrations.[/ref] Hermer’s account gets this wrong. He claims that we should endorse a “substantive” account of the rule of law pursuant to which “the law must afford adequate protection of fundamental human rights”; and he condemns accounts like Fuller’s as the “purely procedural and formal conception that populists and authoritarians can themselves so often use as a cloak of legitimacy”. There are a few errors here.

The first mistake is to think that it is profitable to distinguish between ‘formal’ and ‘substantive’ accounts of the rule of law.[ref]The best discussion is Gardner, J. 2012. ‘The Supposed Formality of the Rule of Law’ in Law as a Leap of Faith: Essays on Law in General.[/ref] Accounts like the one I have presented place substantive constraints upon the content of a polity’s laws, and there is no helpful sense in which they are described as merely ‘formal’. The next mistake is larger. The claim that the protection of human rights is part-constitutive of the rule of law reveals an attempt – pervasive among lawyers – to try to stuff all the political values of a society into the concept of the rule of law. This makes it harder to keep separate the different political values a community should promote and is therefore an intellectual mistake.[ref]As well as my piece at note 5, see Tasioulas, J. 2021. The inflation of concepts. Aeon. Available from: https://aeon.co/essays/conceptual-overreach-threatens-the-quality-of-public-reason.[/ref]

Even so, some stains on the rule of law are also stains on democracy. To see which, and to know when a rule of law problem is also a problem for a democracy, we need to know more about what democracy is. A democratic regime is, roughly, one where the governed are themselves the governors; it is a regime of self-rule.[ref]It is a difficult question in democratic theory, which I here sidestep, how the relevant class of persons within ‘the governed’ is to be delineated.[/ref] A representative democracy is one where the rules which determine who is entitled to govern incorporate the voice of the governed. Yet to keep the link with the basic regime of self-rule, it is not enough that the governed are merely consulted: they must be able to influence the outcome (that is, who is to govern), and the opportunity to influence must (across at least some dimensions) be equal. The customary way of exercising democratic influence is through a vote, where each vote counts equally.

Hermer talks, though, of democratic values, not democracy itself. ‘Democratic value’ is most naturally understood as referring to the intrinsic goods a democratic political arrangement constitutes or promotes, or to the norms that support democratic political arrangements. Whether there are such goods is a massive and urgent question in a contemporary society with remarkable discontent with democratic methods of governance.

At least one important promise of democracy is that it avoids the ills of oligarchy or authoritarianism. Many such ills are instrumental: oligarchic regimes tend to be corrupt, and corrupt regimes govern poorly. Democracy promises a better form of government (though it is ultimately an empirical question whether it keeps its promise). Another vice of oligarchy or authoritarianism, which democracy promises to cure, is that it is inherently inegalitarian: the governed do not stand in relations of equality to one another because there is an asymmetric power between the governed and the governors.

Why think that compliance with the rule of law is a necessary precondition of the promotion or constitution of that kind of equality? This is rarely spelled out explicitly. But here is a tempting argument: certain types of legal rules are necessary foundations of a democracy (and therefore of democratic value); and adherence to the rule of law in respect of those laws is therefore necessary for democratic value. Let me develop that argument – and explain why even this attempt to sustain a necessary connection between law and democracy fails.

The rules and who rules

Any political system must have rules to determine who governs. The distinctive feature of such rules in a democracy, as I have said, is that they must provide a mechanism for the voice of the governed to be expressed – and to allow that expression to influence political outcomes. Let us, therefore, consider how the rule of law might promote (or undermine) democratic values in respect of two sets of rules that are necessary for a democracy to function: the rules which determine the distribution of power to would-be rulers; and the rules which govern who is entitled to participate in that distribution, and on what terms.

To illustrate my first category, consider the distinction between presidential or parliamentary systems. Both such regimes must have systems for the counting of votes, but votes are counted (and outcomes therefore determined) differently depending on the rules setting up the system. Similarly, any system with multiple voting districts must establish how those districts are delineated. All such rules constitute the ‘rules of the game’, defining how power is allocated to rulers.

Democracy is not possible if such rules are not enforced. This is a manifestation of the last condition of the rule of law, the requirement that laws be followed by officials. The events in Washington DC on 6 January 2021 illustrate the point. Donald J. Trump and his supporters sought to have the vice president of the time refuse to count electoral votes certified and submitted by several states. Success would have entailed the seizure of power by those unauthorised under existing laws. This would have been an undemocratic event. Concerns that votes be counted are not ones we have had to contend with in this country. But perhaps few in the United States had thought, prior to January 6^th, that it remained a concern in that country.

This shows a contingently necessary connection between the rule of law and democracy. That phrase might look paradoxical. How can something be contingent and necessary? My point is that compliance with the rule of law is necessary to realise democratic value only if the underlying rules promote democracy. Consider, to illustrate, gerrymandering (manipulating electoral boundaries in an attempt to influence the outcome of elections) in the United States.

Sophisticated computer models can draw electoral districts designed to perpetuate the ruling party’s success. An election held pursuant to grossly gerrymandered districts results in a less democratic outcome, and less democratic value, because the basic equality that democratic systems aim to constitute is not achieved. Yet elections held under such districts may adhere scrupulously to the rule of law. And adherence to the rule of law does not guarantee the promotion of democratic values when the laws themselves are undemocratic.

The second category of rules includes who is entitled to vote, and the conditions under which they are entitled to do so; it also includes matters such as the funding of political campaigns. Rules like these are necessary for a democracy: one cannot have a system of voting without some rules on who is entitled to vote.

When such rules are not enforced – when, for example, eligible voters are denied access to the polling station – that is undemocratic. But the connection, again, is not a necessary one. Imagine a society where women are not entitled to vote, or which requires voters to present an identification card inaccessible to most citizens if the voter’s ballot is to be counted. Those rules might be laid down in advance, very clear and properly enforced. The system might comply with the rule of law. But there would still be a profound democratic deficit. If the votes of women were – contrary to law – counted, that would be a democratic advance (but a deficit in terms of the rule of law).

Arguments about campaign finance exemplify the challenges faced in practice. In the United States, the Supreme Court has held: the expenditure of money is speech, restrictions on campaign finance are therefore an interference with freedom of speech, and therefore that certain kinds of campaign finance laws are unconstitutional.[ref]Citizens United v Federal Election Commission 558 US 310 (2010).[/ref] It is a matter of dispute whether that is a democratic advance (though the answer to that question is, by the Supreme Court’s fiat, not a matter of democratic debate). Absent some restriction on the financing of electoral campaigns, there is a risk that money (not votes) buy influence; and this worry is especially acute where there are drastic inequalities of wealth. These concerns are not local to the United States. Elon Musk’s flirtation with Reform UK – rather belatedly – aroused political interest in the risks posed by the lax regulations on foreign business donations to political parties.

If such rules are to succeed in their aims, the system to enforce the rules must also be just; otherwise the rules can be ignored with impunity. This requires a court system with adequate resources and staffed with willing and able judges; it also requires that there be a system to enable litigants to enforce the rules in question.

These reflections reveal a few points. First, certain legal rules are conceptually indispensable for democracy. Second, compliance with the rule of law in respect of those rules does not ensure democratic value. The democratic value promoted by compliance with the rule of law depends on the content of the laws it upholds.[ref]This second point is sometimes hard to see because compliance with the rule of law does ensure a kind of liberty between citizen and the state. I regard this as a separate value to democratic value: a regime without voting rights could grant that form of liberty. Pursuing this point in detail, however, would require a separate essay.[/ref]

A third important point is that a focus on laws that are conceptually necessary for democracy risks neglecting other laws that contribute to democracy’s functioning and flourishing. For example, freedoms of expression and assembly, though not conceptually necessary for democracy, are crucial for its effectiveness and for its value to be realised. There must be protections, for example, to ensure that voices are not silenced; it may also occasionally be possible to justify limits on expression in the name of democracy. Restrictions on exit polls, for example, may justifiably be imposed to prevent voter behaviour from being influenced during elections by arbitrary facts.

Laws are conceptually necessary to establish democracy, but the connection between the rule of law and democratic value is limited. For this reason, the philosopher Jeremy Waldron aptly described the rule of law as “but one star in a constellation of political ideals”, a constellation that also includes respect for human rights and democracy.[ref]Waldron, J. 2023. Thoughtfulness and the Rule of Law, p.2.[/ref] Each star burns its own light and we should therefore see each light separately. One reason that clarity is important is that these different values can conflict. Adherence to the rule of law might promote democracy – but it might also impede it. Political values, including the rule of law, may sometimes need to be balanced or even traded off with one another. The need to do so is one reason why democracy is necessary: to decide, in a manner which experts cannot, how to balance such values.

I want next to consider how good custom relates to democratic value, and how the intrusion of law can promote (or undermine) that value.

Over-ruled?

In 2019, Prime Minister Boris Johnson advised the Queen to prorogue Parliament for five weeks. This was the longest prorogation since 1930 following the longest session of Parliament since 1653. It was widely believed that Mr Johnson sought an unusually long period of prorogation to negotiate an agreement with the European Union without the scrutiny of Parliament. The Supreme Court held in the case of R (Miller) v Prime Minister that his advice to the Queen was unlawful and that the prorogation was, as a result, “unlawful, null and of no effect”.[ref]R (Miller) v Prime Minister [2019] UKSC 41 at [69].[/ref]

The prerogative to prorogue Parliament was, historically, insulated from judicial review. Its proper use depended on conventions requiring prime ministers to act responsibly. Mr Johnson had resolved to use the prerogative for an improper purpose – to help him to escape political scrutiny. It is possible that the Supreme Court felt obliged to intervene for that reason, to prevent Mr Johnson’s abuse of his own discretionary power. If that is the right diagnosis, it is a regrettable turn of events. What was previously a matter of political convention became governed by legal rules. And there is an anti-democratic concern with any society where ‘the law rules’. Since the law cannot rule without humans to interpret and apply it – in this case, the eleven Supreme Court judges who heard the case – the ‘rule of law’ can mean the rule of those officials charged with enforcing the law (rather than those with democratic legitimacy).[ref]This worry can be traced to Aristotle but is most famously expressed by Thomas Hobbes in his Leviathan. See Malcolm, N (ed). 1968 [2012]. The Clarendon Edition of the Works of Thomas Hobbes, vol 4: Leviathan, p.429.[/ref] That is in obvious conflict with the egalitarian nature of democratic value.

Some might respond that this criticism is a travesty, the stuff of tabloid hyperbole. The court held that the advice to prorogue was unlawful because it had the “the effect of frustrating or preventing, without reasonable justification, the ability of Parliament to carry out its constitutional functions as a legislature and as the body responsible for the supervision of the executive”.[ref]R (Miller) v Prime Minister [2019] UKSC 41 at [50].[/ref] The judges were, the objection might run, acting to safeguard democracy.

Any such retort is wrong on the facts. The relevant order in council, approved at the Privy Council meeting held by the Queen on 28 August 2019, stated that Parliament would be prorogued “no earlier than Monday the 9th day of September”.[ref]Privy Council. 2019. Orders approved at the Privy Council held by the Queen at Balmoral on 28th August 2019.[/ref] Parliament returned from summer recess on 3 September 2019. It follows that Parliament could have intervened to reverse the prorogation. It chose not to. It chose, instead, to use its legislative time to pass a different Act, the European Union (Withdrawal) (No.2) Act 2019 (the so-called anti-no-deal Act). As Timothy Endicott explained, the Supreme Court “nullified a prorogation that Parliament did not choose to nullify”.[ref]Endicott, T. 2020. Making constitutional principles into laws. Law Quarterly Review 136. 175–181, p.176.[/ref] Rather than rescuing parliamentary democracy from executive fiat, the court sought to rescue parliamentary democracy from itself, a different proposition.

Machiavelli’s observations about law and custom suggest a tension at the heart of democracy. Laws are indispensable to the establishment of democracy, and certain laws are necessary for the flourishing of democratic value. Yet democratic value may at times be promoted by the absence of law, and the advance of legal governance can undermine democracy and democratic value. A proper balance is difficult to sustain, requiring as it does that political leaders act only for certain reasons and purposes – including where that is not in their immediate interests. It can seem an impossible ask.

Our legal system has many defects and needs our attention and support. Done well, that may strengthen the justice system and improve the rule of law. Whether it would strengthen our democratic institutions is another matter. We cannot save our democracy through the accretion of rules. And the rule of law is no substitute for the cultural norms that allow democratic systems to thrive. It is important to defend and promote the rule of law, yet the increased prominence of that value in our societies, and the expansive claims sometimes made about it, should be treated with regret and caution.

Optimism is hard. But in a world of increasing democratic discontent, it is democracy, not law, which most urgently needs our support.

The Nuffield Foundation has commissioned this project, but the views expressed are those of the authors and not necessarily the Foundation.