The Cake Bill

The flaws in the UK government’s two-faced Bill of Rights Bill

The UK government has introduced its Bill of Rights Bill: a long, if not exactly eagerly, awaited replacement for the Human Rights Act 1998, which gives effect to the European Convention on Human Rights in UK law. The Bill will limit the ability of the UK courts to enforce rights protections in the UK in various ways, some of them arguably defensible ― at least in the abstract ― and many not defensible at all. In this, I offer my initial thoughts on some of the Bill’s most salient aspects. My overarching theme will be that the government is trying to have its cake ― or rather, several different cakes ― and eat it ― or them ― too.

It may be worth briefly noting where I’m coming from on this. I think that I am more sympathetic to the concerns with judicial overreach in the implementation of the Convention and the Human Rights Act 1998 than many, perhaps most UK public law academics. Moreover, I have no particular attachment to the Convention and especially the European Court of Human Rights, whose judgments consistently strike me as unimpressive or worse. At the same time, as readers of this blog will know, I do strongly favour protections for individual rights vigorously enforced by an independent judiciary. So if the point of human rights law reform were for the UK to go its own way, and even leave the Convention so as to reject the Strasbourg Court’s mistakes, while making robust arrangements to secure rights, I would be quite happy.

But that is not at all what it is proposed. It would be an exaggeration to say that the Bill embraces the worst of both worlds ― the Convention/Strasbourg world and that of UK parliamentary sovereignty ― but it blends them in a way that strikes me as remarkably inelegant and unattractive.


For all the talk of a “British bill of rights” over the years, the Bill of Rights Bill remains closely tethered to the Convention. It (largely) eschews any definition of rights, and in clause 2 tamely incorporates by reference the substantive provisions of the Convention (which are also set out in a Schedule), just as the Human Rights Act had done. It also refers to various other definitions and provisions of the Convention. Perhaps this was the path of least resistance, but if the idea was to produce a statement of the UK’s own commitment to rights, this is a missed opportunity. Perhaps, on the contrary, the government wanted to signal that rights are simply alien to the UK’s legal system. That would be a deplorable distortion of the (admittedly complex) historical and constitutional truth. Either way, this is an example of the government trying to have it both ways: both distancing the UK legal system from that of the Convention and the Strasbourg court, but also remaining bound to it.

The main apparent exception to this refusal to articulate a distinct list of rights concerns clause 4 of the Bill, which refers to “the right to freedom of speech”. The Convention itself refers, instead, to the freedom of expression. But this distinction is mostly for show. Subclause 2 clarifies that “‘the right to freedom of speech’ means the Convention right set out in Article 10 of the Convention (freedom of expression) so far as it consists of a right to impart ideas, opinions or information by means of speech, writing or images (including in electronic form).” Again, the Bill is acting like Very Grownup child who will not stray out of mommy’s sight.

More importantly, clause 4 is mostly just for show substantively. Its first subclause says that “a court must give great weight to the importance of
protecting” free speech. Put to one side the question of what this even means, and whether courts now fail to “give great weight” to the freedom of speech. This hardly matters, because subclause 3 excludes most conceivable use cases from the scope of clause 4’s application. Freedom of speech is not to be given great weight in deciding “any question [regarding] a provision of primary or subordinate legislation that creates a criminal offence”, or questions about contractual or professional duties of confidentiality, or immigration, citizenship, and national security cases. Just that! What’s left? So far as I can tell, defamation and privacy issues (and note that clause 22 of the Bill puts a thumb on the scale against pre-trial restraints on publication ― though it does not prevent them entirely). It’s not nothing, I suppose, but a provision that grandly announces the importance of an English-sounding freedom of speech (rather than the dastardly Latinate “expression”) only to clarify that it applies only to fairly narrow categories of cases is another example of the Bill’s two-facedness.

I turn now to a different aspect of the Bill, the one to which I have at least a modicum of sympathy: its interpretive provision, clause 3. The Bill does away with one of the contentious elements of the Human Rights Act, section 3 (coincidentally), which provided that “[s]o far as it is possible to do so … legislation must be read and given effect in a way which is compatible with the Convention rights”. Courts took that pretty far, holding at one point that even unnatural readings of statutory provisions were “possible”, provided they did not mess with the main thrust of the legislation at issue. Where primary legislation was concerned, such re-interpretation was the only remedy that could do an applicant some tangible good, and moreover it avoided the need to declare legislation incompatible with convention rights. But by my own lights it was inappropriate nonetheless, and I am not sorry to see it go. I wish the UK allowed the courts to disapply legislation incompatible with rights, but I don’t think that judicial re-writing is an appropriate substitute for such a remedy (see e.g. here).

I also appreciate the Bill’s gesture at textualism and perhaps even an originalism of sorts with its requirement, in clause 3(2)(a) that courts interpreting a Convention right “must have particular regard to [its] text … and
in interpreting the text may have regard to the preparatory work of the Convention”. As an abstract matter, this is the right approach to interpretation. More on whether it makes sense in the context of UK human rights law presently. First, let me note that the Bill doesn’t actually embrace originalism, because it also allows the court to “have regard to the development under the common law of any right that is similar to the Convention right”. Contrast this with the Supreme Court of Canada’s rightful scepticism of jurisprudential developments post-dating the framing of the Charter in Quebec (Attorney General) v 9147-0732 Québec inc, 2020 SCC 32 (on which see here).

Anyway, the trouble is that this provision is another show of rigour and independence that will do no one much good. To the extent that the courts will follow it and adopt readings of Convention rights that are tethered to the text and “that diverg[e] from Strasbourg jurisprudence” as contemplated by clause 3(3)(b), they simply ensure that the Strasbourg court will find that the UK has violated its Convention obligations as interpreted by Strasbourg itself. It will be a pain in the neck for claimants, and it might allow the government to rage at those unconscionable European judges ― indeed, it is hard not to wonder whether this, as much as anything else, is really the point ― but that’s about it. The UK cannot unilaterally change the way the Convention is interpreted, even if its proposed interpretive methodology is better than the one endorsed by the European Court of Human Rights, and it cannot escape its Convention obligations by proclaiming that Strasbourg jurisprudence is no part of UK law.

Other interpretive provisions aren’t even well-intentioned. Clause 3(3)(a) makes adjudication of Convention rights into a one-way-ratchet by providing that courts “may not adopt an interpretation of [a] right that expands the protection conferred by the right unless the court has no reasonable doubt that” Strasbourg would do the same. While I understand discomfort with the idea that rights can be ― seemingly ― forever expanding by judicial fiat, this is unambiguously bad, though not unambiguously much else. The Bill doesn’t explain what it means by “expand” ― notably, what is the baseline? The existing Strasbourg jurisprudence? The original meaning? The original expected applications? Just what is “the protection” that must not be expanded? Does a new factual scenario count? And, fundamentally, whatever this all means, why is that (by implication) restricting the scope of a right is permitted but expanding it is not? If rights are in some sense fixed, they must be fixed against restriction as well as expansion; indeed, this is an important argument for originalism (see e.g. here), though not the most important one.

Another largely arbitrary limitation on the way rights are to be interpreted and applied is clause 5, which prohibits interpretations of Convention rights that would impose “positive obligation[s]” on public authorities ― i.e. simply require them “to do any act”. (The prohibition is categorical for the future cases, while existing interpretations that would fall afoul of it can only be retained on some stringent conditions.) Now, here too, I have some sympathy for the underlying motivations: so far as I can tell, the Strasbourg court can be fairly cavalier with demands that authorities do this or that, and its conception of the limits of the judicial role is different from that which you will find in common law jurisdiction. The Convention itself protects primarily what are known as negative rights ― that is, “freedoms from” rather than “rights to”. But understandable motivations aren’t enough.

The lines drawn by the Bill are too rigid. While it can be a useful guideline, the distinction between positive and negative rights is not nearly as clear-cut as the Bill’s drafters seem to assume. Sometimes, this is a textual evidence. Take Article 3 of the First Protocol to the Convention, by which the UK “undertake to hold free elections at reasonable intervals by secret ballot”. This is manifestly a commitment to “do acts”, lots and lots of them, and if the UK should fail to live up to it, I don’t understand how a court ― let alone a court having “particular regard to the text” can decline to order the government to get on with it. Once again, Strasbourg, here we come. But this is only the most obvious example. Even a seemingly purely “negative” right, say to be free from a random arrest by a rogue police officer, can have a positive corollary ― namely, to be promptly released if so arrested. Does the government really think a UK court should not be able to infer such a right (assuming it has not already been inferred ― sorry, I am far from being fully caught up on Convention jurisprudence) from Article 5 of the Convention? Meanwhile, the Bill doesn’t address what might actually a more disturbing aspect of Strasbourg’s positive obligations jurisprudence: their indirect imposition on private parties, who are thus burdened with duties the Convention quite clearly didn’t intend to impose on them.

I finally turn to the last issue I want to discuss at some length: the Bill’s attempt to force courts to defer to Parliament. Specifically, clause 7 provides that when determining whether a statutory provision is incompatible with a Convention right and, in the course of doing so, “decid[ing] whether the effect of the provision … strikes an appropriate balance between different policy aims [or] different Convention rights, or … the Convention rights of different persons … [t]he court must regard Parliament as having decided … that the Act” does strike such a balance. The Court is, further, to “give the greatest possible weight to the principle that, in a Parliamentary democracy, decisions about how such a balance should be struck are properly made by Parliament”. One problem with this is that this is all quite vague. Indeed, perhaps all this bluster means nothing at all. A court may well stipulate that Parliament decided that its law was fine and dandy and conclude that the greatest possible weight to give to this decision is precisely zero. On its face, the clause doesn’t actually preclude that.

But of course that’s not the interpretation the government will be hoping for. So let’s try taking this clause more seriously. So taken, clause 7(2)(a), which deems Parliament to have appropriately balanced all the rights and policy considerations involved is reminiscent of the late and unlamented “presumption of expertise” in Canadian administrative law, whereby courts were required (albeit by judicial precedent, not an Act of Parliament) to pretend that administrative decision-makers were experts regardless of whether the decision-maker in question had demonstrated any expertise bearing on the issue or could be plausibly expected ever to do so. I have called this “post-truth jurisprudence“, and I regard clause 7(2)(a) as a specimen of similarly post-truth legislation. It demands that the courts accept for a fact something that will by no means always be true. Many rights issues are unanticipated ― indeed, they arise precisely because they were not thought of when the legislation was being drafted. To the extent that, as the Bill’s drafters want us to believe, Parliament does take rights seriously, it will usually redress the issues it can anticipate before enacting legislation. It is no calumny against Parliament, however, to say that it cannot foresee all the problems that can arise. If anything, the calumny is to insist that whatever problems do occur, Parliament must have intended them to.

And then, there’s the matter of the assertion in Clause 7(2)(b) that decisions about balancing rights, or rights and policies, “are properly made by Parliament” “in a parliamentary democracy”. The “parliamentary democracy” bit is either a red herring or a misnomer. There are parliamentary democracies with robust judicial review of legislation ― Germany and India come to mind. What the Bill really means, but doesn’t quite want to say, is something like “a constitution based on parliamentary sovereignty”. Indeed, clause 7(2)(b) is reminiscent of the language in the preamble of Québec’s anti-religious dress code statute, which proclaims that “in accordance with the principle of parliamentary sovereignty, it is incumbent on the Parliament of Québec to determine the principles according to which and manner in which relations between the State and religions are to be governed in Québec”, by way of foreshadowing exclusion of judicial supervision of this law’s compliance with constitutional rights. I cannot help but suspect that the UK government is deliberately less forthright than its Québec counterpart because, yet again, it is trying to have its cake and eat it too. It wants to make courts to rubber-stamp parliamentary legislation instead of passing their own judgment on its compliance with rights, but it doesn’t want to admit that it is undermining the (already weak-form, and often quite deferential!) judicial review that UK courts have been engaging in. It might even be hoping to trade on the respect the European Court of Human Rights has developed for UK courts over the years to persuade the Strasbourg judges that legislation they rubber-stamped was really alright. I doubt it will work very well.


There would be a lot more to say. Much ― really, a shocking part ― of the Bill is devoted to nipping various claims in the immigration and refugee context in the bud. Some ― though less ― also tries to stick it to prisoners. I don’t like that one bit. As the most intelligent and principled opponent of judicial review of legislation, Jeremy Waldron, has come to recognise, if anyone has a claim to the assistance of the courts in order to defend their rights, it is precisely these groups, often unpopular and politically voiceless. Instead of being granted special solicitude, they are disgracefully singled out for special burdens. That said, in various smaller ways the Bill gets in the way of other rights claimants too.

But this is already a long post, and it should be clear enough that, in its present form, the Bill is not much good. To repeat, I’m no great fan of the Human Rights Act that it is meant to replace. That law’s weaknesses are mostly baked in for as long as the UK remains party to the Convention, but perhaps some of them could have been ameliorated. Instead of trying to do that, the government came up with a set of proposals that will, if enacted, make everything worse. Quite radically worse for some people, and less radically, but just enough to be noticeable, for everyone else. And for what? Chest-thumping now, and lost cases at Strasbourg later. Even a sovereign legislature in a parliamentary democracy can only ever say that it will have its cake and eat it too; it cannot actually do it.

Case Not Made

Unconvincing arguments against judicial enforcement of rights under the UK’s Human Rights Act 1998

Policy Exchange has recently posted a paper by Richard Ekins and John Larkin QC on “How and Why to Amend the Human Rights Act 1998“. Lord Sumption has written the foreword, picking up on themes explored in his Reith Lectures, which I have summarised and commented on here. There is much to disagree with in the paper, as well as some interesting ideas. Time permitting I might do a short series of posts on it. For now, I want to focus on one idea raised by Lord Sumption and addressed in a rather different way in the paper. The idea in question is that the UK’s Human Rights Act 1998, and so presumably any constitutional or statutory enactment that grants judges the authority to verify whether legislation and administrative decision-making complies with a set of enumerated individual rights, results in judges making decisions that are political rather than properly judicial.

Lord Sumption writes that the Act “treats broad areas of public policy as questions of law, and not as proper matters for political debate or democratic input”. (5) One example that seems to exercise him ― and that has exercised the UK’s political leaders for years ― is that of the franchise. He denounces the European Court of Human Rights for having rejected prisoner disenfranchisement despite its approbation by legislatures on the basis that “it was a question of law and not a matter for Parliament or any other forum for democratic input”. (5) For Lord Sumption, “the suggestion that the electoral franchise is not a matter in which the representatives of the general body of citizens have any say, seems startling”. (5)

But, more broadly, Lord Sumption argues that cases involving balancing between public policy objectives and individual rights ― which is a great many under the European Convention on Human Rights and the Human Rights Act and, in theory, all of them under the Canadian Charter of Rights and Freedoms ― are not fit for judicial resolution. Since policy-making means “a choice between competing considerations, and sometimes compromise between them … [i]t is necessarily a political question.” (6) Treating such choice “as a question of legal proportionality, requiring judges rather than elected representatives to assess the relative importance of the various values engaged before deciding which should prevail” (6) is, in his view, a fatal mistake.

As I have previously argued here in response to another distinguished, if less famous, judge, this argument is misconceived. Similarly to Lord Sumption, Chief Justice Joyal of the Manitoba Court of Queen’s Bench has expressed the worry that

judicial incursion into subject areas and issues of profound political, moral and social complexity[] has the potential effect of removing these issues from the civic and political realms where ongoing and evolving debate and discussion may have taken place.

In response, I pointed out that, taken all the way, this leads one to Jeremy Waldron’s rejection of judicial review of legislation. Chief Justice Joyal did not, ostensibly, want to go all the way. Lord Sumption might ― indeed, he may well want to go beyond Professor Waldron, who specifically objects to strong-form judicial review, where courts can actually refuse to apply legislation, not so much the weak-form arrangement that the Human Rights Act 1998 put in place. But strongly argued though it is, this position is not all that compelling. As I wrote in response to Chief Justice Joyal,

The frontiers between law’s empire and that of politics are not immutable. There is no reason to believe that the position that every social issue is by default subject to politics is entitled to be treated as a baseline against which a polity’s constitutional arrangements ought to be measured, and any departure from it justified and limited. It is the position of some political cultures … But these political cultures have no automatic claim to superiority or to permanence. They are liable to be supplanted, just as they supplanted their predecessors.

Issues cannot be declared political ― or non-political, for that matter ― by stipulation. For instance, the extent of the franchise can meaningfully be addressed in the courts, as it has been under the European Convention and in Canada. It takes more than a bald assertion that this is truly a political matter, or the existence of public controversy, or the involvement of moral considerations, to show that courts should keep out of it or defer to political judgments that are, as often as not, driven by prejudice or self-interest. (As to the point about morality: courts make judgments influenced by morality when applying concepts such as reasonableness, negligence, or unconscionability. One can certainly be sceptical of the resulting jurisprudence, but it’s not plausible to claim that morality is something courts should always stay away from.)

Rather, for any given right that the designers of a constitutional order might consider, they should ask themselves whether, given their respective strengths and weaknesses, a given institution would do a better job of protecting ― better, that is, all things considered, including the downsides of allocating the task to this institution instead of a different one. Institutional considerations have to be front and centre in this analysis. Issues cannot be declared to be political or legal apart from a consideration of actual political and legal institutions that would be dealing with them. Lord Sumption only gestures at institutional factors, claiming that “judges lack the information, experience and democratic legitimacy to make … choices” involved in the proportionality analysis. Even here, the appeal to democratic legitimacy is largely question-begging. It’s not obvious that these choices need to be made democratically, as is evident from the fact that, in the absence of the Human Rights Act, many of them would be made by bureaucrats rather than Parliament.


Professor Ekins and Mr Larkin engage with the institutional issues to a greater extent. To be sure, they too assert that proportionality analysis

requires judges … to answer a series of political questions, about the legitimacy of the legislative objective, the suitability of the means adopted to that objective, and, especially, about the fairness of the balance to be struck between attaining that objective and the claimant’s interest. [33]

But they also say that these “are not questions that a court is well-placed by training or ethos to answer”. [33] They worry, too, “that courts will be drawn into political controversy, with litigation a rational means to enjoin the court to lend its authority to one’s cause”. [34] They also claim that the outcome of rights litigation often depends on the subjective and personal beliefs of the judges hearing the case (and hence on who happens to be on the relevant court and panel).

What should we make of this? To start, it’s important to note that, although Professor Eakins and Mr Larkin have very little to say about Parliament and the executive, deciding which institution should be given the role to uphold rights is necessarily a comparative exercise. It is not enough to point to the shortcomings of the courts, even if these are real enough. It is necessary to show that courts are worse than legislatures, ministers, and bureaucrats, either on a specific dimension where it is possible to compare them directly or on due to some concerns unique to them. With this in mind, I don’t think that Professor Ekins and Mr Larkin make a convincing case at all.

It is of course true that judges lack the “training” that might be helpful to answer the sort of questions that arise in the course of proportionality analysis. But what training have members of Parliament? What about Ministers? Are they trained to weigh up rights when they make policy? They are not, of course. As for ethos: for the high-minded rhetoric of the defenders of legislative articulation of rights, it is very far from obvious to me that politicians care about rights on a regular basis. They do sometimes, of course, especially if the rights of their constituents may be at issue. But their record is patchy at best, and does not suggest an ethos of weighing up rights and social needs in a rigorous fashion.

The most that Professor Ekins and Mr Larkin say on this is that, when it comes to delegated legislation, “Parliamentary scrutiny, including anticipation of political controversy, is an important discipline on ministers, even if secondary legislation is almost never rejected outright”. [48] We are, I suppose, to take this claim on faith. Meanwhile, Professor Ekins and Mr Larkin also note that there are “limits on parliamentary time” which, they say, combine with “scarcity of political capital” to “make[] it relatively difficult … for Parliament to legislate to correct judicial lawmaking” in relation to rights. [40] To their mind, this is a sign that “judicial lawmaking” needs to be curbed. But one can just as easily argue that limits on Parliament’s time and reluctance (or indeed inability) to spend political capital on decisions that will be unpopular even if right are a key reason for wanting judges to make decisions about rights, especially about the rights “discrete and insular minorities”, in the American parlance, and of especially unpopular groups such as criminal suspects and prisoners (a concern that Professor Waldron, for example, has come to acknowledge).

The concern about courts being drawn into politics is legitimate though it is all too often self-fulfilling, in the sense that it is commentators and politicians who share Professor Ekins’s and Mr Larkin’s views who generate much of the controversy. Still, it is fair to worry about the authority of the courts being undermined by their having to make decisions that are bound to be politically controversial. Then again, would the authority of the judiciary not be negatively affected by its having to blindly apply laws that disregard human rights? Besides, occasional flair-ups of criticism notwithstanding, in countries like the United States in Canada, where courts have been given the mandate to make decisions about rights long before the United Kingdom, their standing in the public opinion is much higher than that of legislatures. Indeed, there is an element of self-contradition in the arguments advanced by Professor Ekins and Mr Larkin: if the courts were really suffering from a legitimacy crisis due to all those controversial decisions the Human Rights Act foisted on them, why would Parliament need to expend scarce political capital on disagreeing with them? The authority of the courts, then, may benefit rather than suffer from their having jurisdiction over rights issues.

As for the alleged subjectivity of judicial decisions regarding rights: I think this too may be an issue. It may be more of an issue in the United Kingdom, where the Supreme Court (almost) never sits en banc, than in the United States and in Canada, whose supreme courts do (respectively always and, these days, usually). Then again, if this is acceptable in other cases, which can also divide the bench, sometimes closely, perhaps this is no more concerning where rights are involved. More importantly, though, the criticism of the courts, in the abstract, does not tell us much. In what sense is decision-making by Parliament, by ministers, or by officials not subjective? When it comes to Parliament and ministers, their inclinations and decisions will fluctuate depending on which party is in power. Precedent and legal doctrine constrain judicial decisions based on rights imperfectly. But if constraint and principle are valuable in such decision-making, then courts still do better than the other branches of government.


So neither Lord Sumption nor Professor Ekins and Mr Larkin have advanced particularly convincing arguments against having judges enforce individual rights. Rights issues are not inherently incapable of judicial enforcement, and the institutional arguments against having the judges deal with them are far from obvious. None of this fully addresses an argument along Waldronian lines, one that is purely about ineradicable disagreement and the fairness of resolving it via democratic procedures. But that argument only goes so far ― and, in particular, as Professor Waldron recognised, I think, it does not obviously apply to prevent courts from overriding decisions by the executive branch, which is what Professor Ekins and Mr Larkin want to do.