'Justice is What the Judge Had for Breakfast': When Rights Matter and When They Don't

On what immigration, Brexit and lockdowns don't have in common

‘Of the many things which have been said of the mystery of the judicial process, the most salient is that decision is reached after an emotive experience in which principles and logic play a secondary part.’

-Hessel E Yntema, 1928

The UK Supreme Court (UKSC) last week handed down judgement in the case of R (on the application of AAA (Syria) and others) v Secretary of State for the Home Department and various other conjoined appeals. For overseas readers, this requires a little bit of unpacking but, in essence, the dispute was over the government’s ‘Rwanda scheme’, through which asylum seekers coming to the UK would be sent to Rwanda to have their applications processed, in order ultimately to be granted asylum in that country instead. (What Rwanda gets in return being a big fat cheque.)

This tortuous approach was adopted with the explicit aim of deterring asylum seekers from entering the UK, the idea being that if everyone knows that when you go to the UK to claim asylum you will end up on a flight to Kigali, then few people will do so. It is basically the same idea as the Australian process which ships asylum seekers to Nauru and Papua New Guinea, and various other schemes that have begun in other European countries (and which also involve sending asylum seekers to Rwanda).

The snag for the UK government was, and is, the Human Rights Act 1998, which incorporates the European Convention on Human Rights (ECHR) into UK law. Section 6 of the Human Rights Act requires public authorities, including the government, to act consistently with the ECHR, and the claimants in this case (all of whom were asylum seekers who were due to take the first flight to Rwanda) challenged the legality of the scheme on the basis that, since there was a risk that asylum claims would not be ‘properly determined’ in Rwanda, there was also a risk of ‘refoulement’ (the lawyerly term for the return of a refugee to the country from which they were fleeing, where they will face ill-treatment). This could ultimately thereby indirectly result in a violation of Article 3 of the Convention, which prohibits torture and inhumane or degrading punishment.

In the end, the Supreme Court determined that the scheme was not unlawful in principle, but that currently there was too much of a risk that asylum seekers (from, for example, Syria) who were sent to Rwanda would end up being shipped back to their home country to face torture or similar. The UK government had simply not done enough to ensure that would not happen.

There is a lot that could be said about the judgment, but, at the level of what is nowadays called ‘optics’, it is fully in keeping with what the UK tells itself about its legal system, which is that courts exist to forensically police government excesses so as to ensure respect for the rule of law. And, setting aside the political debate about the Rwanda scheme (which is something of a can of worms), I basically accept the court’s own framing of its role here, which was to apply what in the olden days was called ‘anxious scrutiny’ to a policy which would have important consequences with regard to fundamental rights.

The problem, though, is also at the level of optics, and it is this: ‘anxious scrutiny’ of the kind on display in AAA v Secretary of State for the Home Department has become something of a movable feast in recent years - so much so, indeed, that it is has become impossible to determine when courts will concern themselves with the effects of policy on the protection of fundamental rights, and when they won’t - except through speculation about the political preferences of judges. That is to say, our judiciary has managed to be consistently inconsistent about the approach it takes to these issues, but it has done so in such a way that demonstrates a marked consistency of regard for the broadly centre-left opinions of the chattering classes whose company judges nowadays keep. To be blunt, it seems to be the case that fundamental rights are taken very seriously by courts when that would produce an outcome that is roughly in line with the views of the average Liberal Democrat MP or Guardian journalist. But otherwise fundamental rights don’t seem to matter all that much. One doesn’t wish to think this way about the judiciary, a profession which I generally hold in high esteem. But it is becoming harder and harder to ignore.

To demonstrate this, we need to get into the weeds of some cases from recent years, beginning with the 2015 ‘black spider memos’ (non)scandal, which involved what ultimately transpired to be very anodyne letters - revealing among other things an obsession with rare birds - which (then) Prince Charles had sent to various government departments over the years. Rob Evans, a journalist at The Guardian, had heard about these letters, thought there was something improper about the heir to the throne apparently seeking to influence government policy, and submitted a series of Freedom of Information (FOI) requests to the departments in question, asking for the letters to be disclosed.¹ The departments refused on the grounds that the letters were exempt from disclosure on the basis that they concerned communications with the heir to the throne (pursuant to s. 37 (1) of the Freedom of Information Act 2000), that they contained personal information (pursuant to s. 40 of the Act) and so on.

Evans, as was his right under the Act, appealed this refusal to the Information Commissioner, who upheld the decision not to disclose. Evans then appealed this decision to what ended up being the Upper Tribunal (essentially a division of the High Court), which found in his favour. But the Attorney General, Dominic Grieve, then issued a certificate under s. 53 (2) of the Act to prevent this Upper Tribunal decision from having effect; he did not give a reason in the certificate, and was not required to, but he later said more or less that he didn’t want Prince Charles to be embarrassed. (You can read the certificate in question, for what it is worth, here.) Evans then issued judicial review proceedings to quash that certificate, and that dispute ultimately made its way up to the Supreme Court.

The case did not hinge on an ECHR claim. But the Supreme Court’s decision is nonetheless instructive in setting the scene. In R (on the application of Evans) v Attorney General [2015] (the judgment of which is available here), it referred to, and considerably advanced, a nebulous concept known as the ‘principle of legality’, which essentially means that very clear wording must be used in order to legislate contrary to fundamental rights. The idea here is that Parliament, being sovereign, can always legislate against fundamental rights, but if it does so it must do so absolutely explicitly (and hence face the political cost). It cannot do it accidentally, by implication, or by the back door. This means that if there is any ambiguity in a statutory provision that may be contrary to fundamental rights, it must be read in such a way that it is not.

The statutory provision which the Attorney General was relying on to issue his ‘executive override’ certificate, s. 53 (2) of the FOI Act 2000, gave him the power to issue such a certificate and prevent disclosure of information where he had ‘reasonable grounds’ for doing so. Grieve’s certificate made clear that in his view he had such grounds, and used that phrasing expressly. The question, in the UKSC’s view, was whether the principle of legality had anything to say about the matter.

And they decided that it did. Rob Evans, like everybody else in the UK, had a fundamental right to a final decision from a court that could not be overruled by the executive. And therefore the phrase ‘reasonable grounds’ in s. 53 (2) had to be interpreted restrictively. It did not mean that the Attorney General had the power issue a certificate overruling a notice from the Upper Tribunal enforcing disclosure of information because he happened not to agree with the decision to issue the notice and felt himself to have reasonable grounds for doing so. It could only mean that he had the power to issue such a certificate in some extreme circumstance such as when an error had been made.

This was tantamount to saying that the plain meaning of a statute, made by Parliament, is subject to a freestanding power that the judiciary has to declare itself interested in the ‘principle of legality’, and on that basis decide that words mean something else to what they appear to say. No sensible person could read s. 53 (2) of the Freedom of Information Act 2000 and decide it implies that ‘reasonable grounds’ means ‘only in extreme circumstances’. (And it is probably worth mentioning that in UK public law it is normally the case that ‘reasonable’ means anything short of complete irrationality.) But that was what the UKSC in effect determined, on the basis that when fundamental rights were at stake the judiciary ought to bend over backwards to, where possible, subject the executive to oversight.

The approach in Evans was continued through to the first of the big Brexit decisions not long afterwards, R (on the application of Miller) v Secretary of State for Exiting the European Union [2017] (often referred to as ‘Miller One’ to distinguish it from the the 2019 ‘Miller Two’ case, which had the same applicant, and which will come to shortly). Here - and UK-based readers must forgive me for dragging the ‘Brexit years’ up once more from the memory dungeon to which they have been consigned - the question essentially revolved around the use of what is called the ‘royal prerogative’. These are the executive powers, technically still vested in the Crown, which the government exercises, one of those being the power to make and unmake treaties.

After the Brexit referendum vote of 2016, Theresa May’s government had intended to use this power to simply notify the EU of the UK’s withdrawal from the different EU treaties. But judicial review was brought by a wealthy Remain-backing activist called Gina Miller, whose argument was that since leaving the EU would mean depriving UK citizens of various rights deriving from the EU treaties (which they held as EU citizens), the use of the royal prerogative power to unmake treaties would be tantamount to the executive depriving citizens of rights. This would be executive overreach, and therefore notice to withdrawal was subject to Parliamentary oversight - there had to be legislation first to allow it to happen.

This was transparently an attempt to thwart Brexit, and it paved the way for all manner of Parliamentary chicanery which was to occupy the House of Commons, and the news media, for the next two years. But, in any event, the Supreme Court agreed with Miller, insisting all the while that this had nothing to do with considerations of politics, but the purely legal principle that when fundamental rights were at stake, a restrictive approach had to be taken to the prerogative powers - which are normally within the government’s gift to deploy as it sees fit.

The Supreme Court, then, in the period from the mid 2010s, was clearly taking an activist, non-deferential approach to its role, under the stewardship first of Lord Neuberger and, later, Lady Hale. I well remember the mood of that period amongst academic lawyers and public lawyers in particular; they were almost giddy with the notion that what they called ‘common law constitutionalism’ was coming of age. The idea here was that judges had discovered all kinds of unwritten principles in English law which just happened to give judges sweeping power to arrange the relationship between the government and Parliament, and between government and the people, in accordance with their ideals about ‘thick’ democracy (i.e., democracy that involves the voters as little as possible).

This all culminated in the famous decision in R (on the application of Miller) v The Prime Minister [2019], ‘Miller Two’, in which Lady Hale - wearing a bizarre spider-shaped brooch - declared Boris Johnson’s five-week prorogation (meaning the bringing to an end of a session) of Parliament to have been unlawful on the grounds that it was designed primarily to prevent MPs in the Commons from acting to prevent the UK from leaving the EU (as they had been consistently doing in the period 2017-2019). Though that case did not pertain to individual rights, the message it sent was clear. As Lady Hale herself put it in her judgment:

[T]he courts have the responsibility of upholding the values and principles of our constitution and making them effective. It is their particular responsibility to determine the legal limits of the powers conferred on each branch of government, and to decide whether any exercise of power has transgressed those limits. The courts cannot shirk that responsibility merely on the ground that the question raised is political in tone or context.

In other words, whatever the subject, and whatever the circumstances, it was the job of the courts to make sure the government kept in line with cherished principles of the UK constitution, and the judiciary, it seemed, was going to get much more ‘in your face’ about doing so.

The stage then was set for the biggest and most pervasive series of executive overreaches in history, which took place in March 2020 and thereafter. The lockdown era saw the greatest and most pervasive interference with civil liberties certainly since the Second World War. And what to a lawyer was most striking about the period was the extraordinary extent to which we became governed more or less by executive fiat. The original lockdown, begun on March 23rd 2020, for example, had absolutely no legal basis whatsoever - it was simply announced by the Prime Minister, Boris Johnson, on TV. It wasn’t until three days later, March 26th, that the necessary regulations, which for example prohibited the public from leaving their homes without ‘reasonable excuse’ were actually made.

And this cavalierness - with the government simply making up ‘rules’ and announcing them on the telly - was characteristic of the years that followed. During the entire original lockdown period, for example, stretching from March 18th to roughly June, schools were closed by order of the Prime Minister (again, via television rather than Parliament) without any legal instrument ever being created to say so. Like Genghis Khan, he spoke, headteachers listened, and that was that. We became used to government ministers appearing at evening press conferences to tell us that we should only be going out once per day, or only going out for an hour a day, when the rules themselves said no such thing - with this being taken as gospel by the innocent public and credulous journalists who were supposed to be informing them about the news. Later, we were treated to the bizarre spectacle of government ministers taking it upon themselves, rather than judges, to interpret legal provisions - in order to answer important questions such as whether a scotch egg qualified as a ‘substantial meal’ or what ‘mingling’ meant. We even witnessed the Secretary of State for Health and Social Care interfering in individual criminal cases.

In short, the entire lockdown era was a litany of flagrant breaches of the most rudimentary elements of the separation of powers, which saw the executive subverting the role of legislature by making up phoney rules on the fly and announcing them on TV before they were enacted or even not bothering to enact them in law at all; subverting the role of the judiciary by opining publicly and loudly about what the law meant; and even pronouncing on criminal sentencing. And when technically playing by the rules the government worked at every turn to undermine Parliament, at times submitting for scrutiny new tweaks to regulations literally minutes before midnight so that they could come into effect at 00:01am. We were to all intents and purposes ruled by an absolute monarch, albeit one with three heads (Boris Johnson, Matt Hancock, and Michael Gove).

One would therefore have expected the judiciary to have taken all of this very seriously, given all of the highhanded rhetoric from previous years about ‘upholding the values and principles of our constitution and making them effective’. But when push came to shove all of that rhetoric disappeared, and our courts became suddenly bashful and coy.

The most clear example of this was the root-and-branch challenge to the lockdown regulations brought by Simon Dolan in R (on the application of Dolan) v Secretary of State for Health and Social Care [2020]. Here, the original lockdown regulations of 26th March 2020, the Health Protection (Coronavirus, Restrictions) (England) Regulations 2020, was challenged on a number of grounds, including those concerning the rights to liberty, to privacy and family life, to freedom of religion (since places of worship were closed) and so on. It was also argued that the Regulations themselves, having been made by the Secretary of State for Health and Social Care pursuant to powers under s. 45 of the Public Health At 1984, were unlawful; s. 45 of the Act gives the Secretary of State powers to make ‘a special restriction or requirement’ of a ‘general nature’, but it is quite clear from the text that such special restrictions or requirements are only supposed to be equivalent to those that can be imposed by a magistrate under the Act. Since a magistrate obviously cannot announce a national lockdown, the claimant’s argument was that the Secretary of State was acting ultra vires or in ‘excess of powers’, meaning unlawfully going beyond what statute authorised him to do.

But the courts gave all of this remarkably short shrift. The idea that the interference with rights had been disproportionate was ‘unarguable’; this was ‘clear’ and any division of expert opinion on the advisability of lockdowns was ‘neither here nor there’. The Secretary of State was being advised by experts and balancing this alongside ‘other matters’ and was entitled to do so. The Regulations clearly had a legitimate aim and in any case ‘Government and Parliament’ (the two things had suddenly become elided) had to be given a ‘wide margin’ of judgment. And unlike in Evans, in which, you will remember, the principle of legality meant that the words ‘reasonable grounds’ had to be reinterpreted restrictively to mean ‘in extreme circumstances’ because fundamental rights were at stake, in Dolan the words ‘special restriction or requirement’ and ‘of a general nature’ were held not to be ambiguous at all, and that anyway the Secretary of State had the power to impose all the restrictions he had ever imposed ‘by necessary implication’ of the Public Health Act.

Mr Dolan got all the way to the Court of Appeal, only to have his appeal dismissed and to have his bottom smacked for good measure for the ‘length and complexity’ of his challenge. And his request for permission to appeal this decision, to the Supreme Court, was refused out of hand in a one-sentence email.

The contrast to the previous period was stark. And Dolan was only one of many appeals for judicial review which were contemptuously swatted aside by the judiciary during the lockdown years; a recent article in the Edinburgh Law Review found after a comprehensive survey that there was only one successful legal challenge to a Covid-related regulation (in Scotland) during the entire period, and that this solitary exception was a case with ‘no concrete implications’. And this dismissive attitude concerning the most serious rights-related crisis since the War (some incidents in Northern Ireland during the Troubles notwithstanding) has continued long after the so-called emergency passed; I have written previously in this substack about the case of Adil v GMC [2023], in which a surgeon who said a few controversial things about Covid-19 and Covid vaccines appealed the General Medical Council’s decision to suspend him to the Administrative Court, on freedom of speech grounds, and was told by the judge that he had gone ‘far beyond helpful legitimate comment into the realms of scaremongering conspiracy theories’ and that was that.

No doubt members of the judiciary, and academics working on public law, would be able to draw important legal distinctions between cases on the one hand that are like Evans, Millers One and Two, and AAA, and cases on the other that are like Dolan and Adil. But, to return to the matter of ‘optics’, to members of the public it looks rather like something else: namely, that when it comes to matters that are dear to the hearts of the ‘new elite’ - embarrassing the monarchy, protecting immigrants from deportation, and preventing Brexit - judges seem to take constitutional principle and fundamental rights very seriously. But when it comes to matters like lockdown- and vaccine-scepticism, about which the ‘new elite’ has rather a different view, suddenly judges come over all deferential and give the government and public authorities like the GMC ‘wide margin’ to do whatever they like. This may not be an accurate picture of the underlying reality, but it is the picture that a significant chunk of the public now sees.

All of this brings to mind the work of those American legal scholars - most of them associated with the early 20th century progressive movement and, later, the New Deal - who called themselves ‘realists’. These men (the most well-remembered nowadays being Oliver Wendell Holmes Jr., Karl Llwellyn, and Jerome Frank) argued that judges do not apply the law formalistically in the sense of neutrally applying clear rules. Rather, since rules themselves are often ambiguous, judges tend to find ways to give effect to their pre-existing ideas and then find ways to justify their judgments in reference to rules later on. Oversimplifying the argument, the essential idea is that a judge already knows (consciously or unconsciously) what he wants the outcome of a case to be; the trial is simply the means by which he determines how the rules will be interpreted to give this outcome effect.

The most readable and interesting of the legal realists, Jerome Frank (about whom I will write a proper post in due course) is said to have remarked (though this may be apocryphal) that ‘justice is what the judge had for breakfast’. What he meant by this was that the justice given effect by judges is not legal justice in the sense of the application of neutral and objective rules, but rather in the sense of the ideas and values that the judge had already digested in advance. For Frank (and the other realists), this was not a criticism - it was an inevitability - and it was therefore in fact to be embraced. Since judges just give effect to values, it was better to acknowledge this and do it openly, so as to push the New Deal agenda.

Reasonable people can disagree about the advisability of this (I know where I stand), but we can probably all concur with Frank that, if judges are indeed going to promote certain values in their judgements and the way they apply the rules, they should be open about doing so. (Aside from anything else, this then makes space for democratic discussion and political accountability.) But the UK judiciary’s conduct over the past ten or twenty years has been anything but open, insisting that it is simply holding the government and other public bodies to constitutional principles where fundamental rights are at stake, when in fact it appears to be picking and choosing the circumstances in which it does this on political grounds. Indeed, looking at the cases and judgments I have described here, the observer is led to one of three different possible conclusions, none of which paints the judiciary as a whole in a favourable light.

The first is that judges genuinely do deliberately give effect to the liberal values of the ‘new elite’ from whom they draw membership, and do so covertly. For what it is worth, I don’t think that this is true across the piece, although it may be true of certain judges at certain times. (There is one such figure in particular who I have in mind, here, but they can wait for another day.)

The second is that judges give effect to these values unconsciously, simply because they are not exposed to other, competing sources of value - or see those other sources of value as necessarily inferior to their own. The judgment in Dolan is a classic example of this; it is evident that it both Lewis J in the High Court and the panel in the Court of Appeal had simply imbibed the official line on Covid amongst their class, which was that the ‘spread’ had to be stopped by government fiat, that this was ‘the science’, that the only problem indeed with government policy was that the lockdowns had not been harsh or early enough, and that there were basically no drawbacks to lockdowns except for the piffling matter of the economy. It makes a lot of sense that they would allow their decision-making to be influenced in this way, given what we know about human psychology: judges simply think that their values, and the values of the people they go to dinner parties with, are objective reality, and therefore don’t really think of them as values in the first place. They are simply self-evident truths.

And the third is that judges are much more political animals than we in the UK tend to give them credit for. The Supreme Court in its rulings in Miller One and Miller Two was pushing the envelope in its willingness to take a confrontational line towards the government. And after the general election victory of December 2019, the new government declared one of its main aims to be putting the Supreme Court back in its box and undoing the New Labor constitutional settlement (based on various Policy Exchange proposals).

It is hard not to read developments in the case law in light of this. In the two Miller cases, the Court’s decisions chimed with the views of a clear majority of MPs in the House of Commons (across all parties) during a period of weak government, and when opinion among the chattering classes had it that the Tories would not be in power for very much longer. These were, in other words, the perfect conditions in which the Court could go off piste. By the time Dolan came around, however, the upper echelons of the judiciary found themselves in almost precisely the opposite set of circumstances, being on notice that the government (with a strong majority) had its heart set on reform. It was therefore natural for the courts to be minding their P’s and Q’s again - and to tow as deferential a line as possible.² When seen in this light, the recent decision in AAA makes perfect sense, and can be understood to be simply a return to a more activist stance based on the (probably entirely justifiable) prediction that the current government won’t see out the next 12 months and that the constitutional reform agenda is as dead as a dodo. They are off the hook, in other words, and have gone demob happy as a result.

None of these conclusions is at all flattering for the judiciary. But it is the result of a basic failure on the part of judges to see the wood for the trees. I would prefer a deferential judiciary, all things considered, as that is more consonant with our constitutional settlement, in which Parliament is sovereign because it represents voters. But I can understand the rationale for a more avowedly activist one, even if I disagree with it. What we have currently, however, is the worst of both worlds, in which it is difficult to predict whether a deferential or activist line will be taken without resort to speculation about the political agendas and values of judges and how those things relate to the subject matter of the judgment. In the long term, this helps nobody, and will only result in a decay in the public’s trust in the court system of a kind that has set in with regard to many other of our public institutions.

1. He also submitted some requests for information under the Environmental Information Regulations 2004. For simplicity’s sake, I will not speak to that issue in this post.

2. It is probably also worth mentioning here the case of R (on the application of Begum v Secretary of State for the Home Department [2021] (available here), in which an unusually deferential line was taken by the Supreme Court - again, presumably keeping its head down in a radically different political climate to the one that had prevailed when Miller Two (or indeed AAA) was decided.