The Icing on the Cake


Learie Constantine was a black Trinidadian cricketer, who went on to become Trinidad’s High Commissioner to the UK and the first black peer. In 1943 he arrived with his family at the Imperial Hotel in Russell Square. He was subsequently informed that he could only stay one night, because of complaints about his presence made by white American servicemen staying at the hotel. He sued for damages, and won.


Innkeepers are one of a group of trades called common callings (including farriers and attorneys) who are obliged to serve all those who present themselves. By refusing this service the Hotel wronged Constantine. Legally, the interesting point in the case is that damages of five guineas were payable despite the absence of proof of any consequential harm being suffered, as he had stayed at an alternative hotel.



Most trades are not “common callings” and there is no right to be served. So, a shop can, at common law, refuse to transact with me for no reason, or even a bad or discriminatory one. The Equality Act (Sexual Orientation) Regulations(Northern Ireland) 2006 (SI 2006/439) (“SOR”) and the Fair Employment and Treatment (Northern Ireland) Order 1998 (SI 1998/3162 (NI 21))(“FETO”) alter this common law position. SOR prohibits discrimination in the provision of goods, facilities or services on grounds of sexual orientation, FETO on the grounds of religious belief or political opinion. Outside of the specific cases regulated by discrimination law, shop owners remain free to refuse to serve on the basis of bad or irrational reasons.


Mr Lee, a gay man, ordered a cake from Ashers Bakers in Belfast. The icing design he ordered included the Sesame Street characters ‘Bert and Ernie’, a logo for the campaign group QueerSpace and the headline “Support Gay Marriage”. The owners, having initially taken the order, declined to produce the cake on the basis that to do so would be inconsistent with their Christian beliefs, in particular their opposition to gay marriage. The Northern Ireland Court of Appeal held that this was a case of discrimination on the basis of sexual orientation, this was appealed to the Supreme Court who concluded that no actionable discrimination had occurred.


Sexual Orientation

The court rejected the SOR claim on the basis that the objection of the bakers was not to who Mr Lee was, they had quite happily served him before and would not refuse a gay man a cake on that basis, but because of the message on the cake. If Mr Lee had been a straight campaigner for gay marriage the bakers would have refused to serve him on the same basis.


A slight change in the facts shows how narrow this ground is however. The cake in question had been ordered to mark the end of anti-homophobia week. What if a “Bert and Ernie” cake had been ordered by a gay couple to mark their wedding day? Could the bakers have refused to make it without falling foul of the SOR rule? There is no great demand among straight couples for wedding cakes themed in this way. How then could it have been said that bakers’ objection was merely “to the message, not the messenger?”


Political Beliefs

The route to ruling that there was no discrimination against Mr Lee on the basis of his political beliefs (here his support for gay marriage) was more difficult. Here the distinction between the political opinions of Mr Lee and the message he wished to convey was harder to draw. The only people who will be prejudiced by a refusal to produce cakes in support of gay marriage will be those who desire such cakes because of their support for gay marriage. Being prepared to offer them other kinds of cake is not really the point, anymore than the willingness of the proprietors of the Imperial Hotel to find Mr Constantine a different hotel.


Freedom of Speech

It was here that the Supreme Court relied upon s 3(1) of the Human Rights Act, which requires all legislation to be read so far as possible as compatible with the European Convention on Human Rights. Here the relevant rights were freedom of conscience and religion (article 9) and freedom of expression (article 10). Here the court found that requiring the bakers to express a message that they profoundly disagreed with would be a violation of the Convention, so that FETO should be read narrowly so as to not cover this case (although what the correct ‘reading’ of FETO is is somewhat obscure).


Here I have considerable sympathy with the NI Court of Appeal’s opinion that the baker was not being compelled to express anything at all. The message was that of Mr Lee. Nobody thinks that a baker endorses the messages placed on cakes. “I love you Mum” or “Manchester United Forever” appearing on cakes are not thought to be the statements of the baker himself. This is not like the case of the Muslim petty officer forced to doff his cap during Christian prayers. which involves an outward show of personal belief.


It may be objected that the baker was publishing the information, in much the same way as a printer publishes a newspaper. The printer may commit an actionable libel even where the author of the statement is another.


However. it may be doubted whether the concept of ‘freedom of expression’ is intended to be this broad. Its role is to allow a free market in ideas, however abhorrent. People should neither be prevented from expressing, or required to express, beliefs they do not hold. It may be doubted whether this is what was being required of the bakers, whom nobody considers to be the authors of the message on their cakes.



The intuitive problem with SOR and FETO, and why the court may be tempted to read them narrowly is that they require us to confer benefits upon other people. The law does not generally, and should not, require us to be good. If I wish to withhold a benefit from you, such as my particular skill in making a chocolate cake, I should be free to do so if I wish for good, bad or no reason at all. Or so the reasoning goes. The law should be reluctant to require you to confer a benefit upon me when this is in conflict with your sincerely held beliefs. Like Mr Constantine, beyond the sense of insult and affront, there is no harm to the individual consequent upon the discrimination because of the ability to obtain the goods or services elsewhere (monopoly suppliers requiring more regulation than those operating in a market with ready alternatives available).


Should shopkeepers be required to trade with those who support Brexit if they do not wish to do so, if we are not going to require them to trade with everyone who seeks their custom, as we do with innkeepers? The justification for doing so is a difficult one, and so it may be wise for a court to side on the basis of freedom outside of the clear cut case.



Law Books Everyone Should Read

The criteria I have used for inclusion on this list are as follows. First universalism. These are works people in all times and places would benefit from reading, ruling out works specific to, say, the English criminal justice system. Second I found each of them immensely helpful in learning how to be a lawyer. I have made no concessions for accessibility, some are easier to read than others. I wouldn’t recommend that a student about to begin on a first year undergraduate degree read all of them before starting.


All of the authors on this list are dead white men (with the possible exception of Gaius who may not have been white). That is, I think, an inevitable product of the exclusion of women and others from the discipline in the past.


 HLA Hart, The Concept of Law

Written as a book for students in the Clarendon law series, this is the single most important work of legal theory of the last century. My only quibble is with the title. It sets out an account of the conditions necessary for us to identify posited law, within any society. Other conceptions of the word “law” are in use, and are no less correct.


When I studied law several decades ago, the most important critic of Hart was Ronald Dworkin, who in a series of books and articles sought, amongst other things, to tear down the wall Hart built between the fact of the posited law as it is, and the separate question of what justice requires. My view, for what it is worth, is that Hart has proven the clear victor in this dispute. There is more to be gained from reading Lon Fuller’s The Morality of Law  as a response.


(Buy a cheap second hand copy here, there is no point buying the third edition instead of the second.)


The Institutes of Gaius

Strictly speaking, this is the only ‘law’ book on this list (conceived as an account of a body of posited rules).  It is an introductory law text for students written in the second century AD. Most things (art, science, philosophy etc) began with the Greeks. That is not true of law, where the Romans were the true originators.


Once upon a time, all students would have studied Roman law, but at least in England this practice has nearly died out. It is very useful for everyone to have some familiarity with a legal system different from their own. This enables us to break free from the rigid mindset that the law as it is is inevitable and just. For common lawyers, it is important to have some knowledge of civilian law (and vice versa) as the comparisons within the respective traditions are less fundamental.


Roman law is a useful comparator as it is (inevitably) less detailed than a modern legal system, enabling us to see its skeleton. Start with Gaius, and you will have a reasonable grasp of how any legal system has to be structured.


(Free to read in English here.)


JS Mill, On Liberty

A classic text on the proper limits of law. Although I was greatly influenced (and convinced) by Mill’s Harm Principle when younger, I am now certain that Mill’s arguments are often both weak, and contingent on facts about the world and human behaviour that are never proven. Regular readers of this blog will have realised that I am no utilitarian, and Mill’s work was, in my view, hamstrung by his commitment to this (ridiculous) intellectual tradition.


That said, Mill’s instincts were good, and there is benefit from considering why his arguments (largely) fail and what else might justify the kinds of conclusions he wished to reach.


(Available for free.)


I Kant, The Doctrine of Right

Which leads us to by far the worst written, most frustrating, and profound book on this list. Kant’s The Doctrine of Right is a difficult work. It is probably best to read the Mary Gregor translation, available here, rather than one of the creaky nineteenth century versions available online.  Part 1 is the section to read (and re-read, and re-re-read.) Most philosophers neither know much about law nor are interested in the kinds of questions lawyers need answering. Kant was an exception. (The Groundwork to the Metaphysics of Morals concerns quite separate issues.)


WN Hohfeld, Fundamental Legal Conceptions as Applied in Judicial Reasoning

Technically not a book but two articles in the Yale Law Journal from before the first World War (although subsequently collected as a book). Hohfeld died in the influenza epidemic of 1918 at the age of 39. His work was not wholly original, most of the distinctions he sought to draw had already been made by the New Zealander John Salmond. But, it is to say the least unusual for the work of any lawyer to be worth reading a century after his death, law books usually having the shelf life of a car manual. The distinctions he drew are basic and important, and avoid intellectual confusion.


If I had the choice of making the members of the UK Supreme Court read one work that would halp them to avoid error, it would be this.

The Supreme Court as Legislature

Marriage and Money

Last week the UK Supreme Court handed down two judgments on the same day, one of which was important.


The unimportant case was the wholly unsurprising decision in Owens v Owens The Supreme Court confirmed that the Matrimonial Causes Act 1973 does indeed say what it says, and that a divorce is not available simply on the basis that the marriage has broken down.


That no fault divorce is not available without a separation period of two years appears draconian and wrong, but is neither news nor a major social problem.  People such as the bloody-minded Mr Owens, who will contest a divorce to the bitter end, are fortunately few in number. Absent a dispute as to assets or child custody, and changing the rules on no fault divorce will not eliminate those, most divorcing couples will nowadays lie on the form, tick the box saying they have been separated for two or more years, and be able to go their separate ways. That the law requires this is silly, but not to the same degree as the pre-war world of requiring one party to be found in flagrante in a hotel in Brighton with a third party.


A far larger social problem is cost, not the silly rule. The divorce court fee is £550. This is in addition to the fees of solicitors (which in a non-contested case may be low if you shop around online.) This fee covers little more administration than the sending of a couple of letters. For the vast majority of people, and especially the poor, this is a far higher barrier to divorce. Many couples must choose to stay married for no better reason than that they cannot afford to pay not to be. As so often, in the real world it is the money that matters, not the rules.


The important ignored decision was Prudential Insurance Company Ltd v HMRCThis was significant for three reasons. First the decision was a surprise and will save the Revenue more than £4 billion (see paragraph 36 , this is probably a conservative figure because of the broad basis of the decision of the court). Second it is important within the technical and boring area of the common  law of restitution, changing the position from that which was previously understood. Third the Supreme Court overturned the decision of the House of Lords from ten years previously, Sempra Metals v IRC. This third aspect is the constitutionally significant one.



As is well known, up until 1966 the UK’s ultimate appellate court, until 2009 called the Judicial House of Lords, could not overturn its own decisions. So, however wrong one of its own decisions was, the House of Lords, and lower courts, was bound to follow it. The justification for this position was that it increased certainty, and the legislature could always intervene. However, this was never very satisfactory. The distracted legislature enacts very few laws (although nearly twice as many per annum thirty to forty years ago as now) and does not fulfill this corrective role in any meaningful way. This led to the oddity that is the Practice Statement of 1966. This was surprising not for what it did, it had long been thought by many that the House of Lords should be free to overrule itself, but because the change was made not through the decision in a case, where it could form part of its ratio and hence the law, but by the court simply announcing it. Since then our ultimate appellate court has overturned its own earlier decisions by relying on the Practice Statement on many occasions (over two dozen times on my count).


However, the Practice Statement makes it clear that the court will remain mindful of the need for certainty  and “the danger of disturbing retrospectively the basis on which contracts, settlement of property, and fiscal arrangements have been entered into and also the especial need for certainty as to the criminal law.” So, the House of Lords in Cambridge Water Co Ltd v East Counties Leather Ltd refused to overturn the ‘rule in Rylands v Fletcher’ not because it was correct, but because such a radical change would be, according to Lord Hoffmann, “inconsistent with the judicial function.” In Chartbrook v Persimmon the same judge refused to overturn the rule excluding pre-contractual negotiations for purposes of construction of a contract not on the basis that he thought the rule a good one, but rather because the Practice Statement was only supposed to be invoked in the small number of cases where the court was confident that the earlier decision was “impeding the proper development of the law or to have led to results which were unjust or contrary to public policy.”


Dangerour Radicals?

What is surprising about Prudential Insurance Company Ltd v HMRC is not that the court overturned a decision of relatively recent vintage, that has happened before. Rather they did so first without having been asked to do so by the successful party, second by adopting an argument that was not made to them (although it might be said to have been a logical extension of an argument that was made), and third without invoking the Practice Statement. (For those interested in the finer points of the dust dry law of restitution, the relevant arguments of counsel for the Revenue are available online temporarily here from 1 hr 44min and here  from 1 hr 29 min).


The court sought to justify their departure from the earlier authority on the basis that subsequent decisions had undermined it. So, it has now become clear that the CJEU was not as bothered about what the position was as a matter of UK domestic law as was once thought, it is now clear that the legislature is not going to intervene to correct the mess that is the English law of limitation, and some tangentially related subsequent cases don’t fit happily with the thrust of the reasoning in Sempra (see Prudential at [55]-[67]). However, the reason for overturning the earlier decision is, at root, that it was wrong and always was (see [68]-[80]).


At ultimate appellate court level there is a tension between acting as a law maker, and acting as a court so as to do procedural justice for the parties before them. The court’s role as law maker is constrained by the cases that come before it. It doesn’t matter how appalling and wrong an earlier decision may be, unless a case comes before them where it is challenged it will remain the positive law. So the (apparent) rule that a husband could not rape his wife was only authoritatively overturned in 1991, when a case finally came before the court. It is hard to believe the rule could have survived if ever challenged decades earlier. So, the temptation is to do the just thing, when the opportunity arises even if the litigants don’t ask you to be so bold.


In retrospect, the Revenue were too cautious in failing to challenge the earlier erroneous decision. This must by now have cost the UK a sum of money the mind cannot comfortably contemplate, enough to pay for all divorces for decades. In now overturning the earlier decision, the court has taken another step towards being a subordinate legislature, rather than an ultimate appellate court.

Legality, Justice and Syria part 2

I have already set out, as briefly and simply as I can, the positive claim of why I consider the military action against Syrian targets on 13/14 April to have been contrary to international law, but also why I do not consider that to be determinative of the moral question of whether it was the right thing to do. This post is negative in form. I shall consider a number of arguments against the position I have already set out in order to knock them down.


General Assembly Override?

In his in many ways excellent and clear advice to the Deputy Leader of the Labour Party, Professor Dapo Akande suggests that the collective action route is, and was, not closed off by deadlock in the Security Council.


In 1950 a deadlock had arisen because of the withdrawal of the Soviet Union from the Security Council over the refusal to recognise the People’s Republic of China as the legitimate representatives of the Chinese State. This prevented any UN military action over the Korean War. UN Resolution 377A, also known as the “Acheson Plan”, was a resolution of the General Assembly whereby if the permanent members of the Security Council were deadlocked, the General Assembly could in an emergency special session authorise the use of force. Although this route to authorise military force has never been used, could it be as Akande suggests?


No. That resolution did nothing to alter the terms of the Charter, which contains the rules which the signatory states have agreed to. Article 108 of the Charter allows for its own amendment where this is done by two thirds of the members of the General Assembly, and ratified by two thirds of the members of the United Nations, including all the permanent members of the Security Council. Resolution 377A did not meet this threshold (because it could not without Soviet agreement which it withheld). Resolution 377A does nothing at all therefore to the Charter. The Charter is what it purports to be: an exhaustive code of when military force is permitted. The power to block military action given to permanent members of the security council by the Charter has not been bypassed, and Article 108 is a form of double entrenchment of that power as those members can each block any change to the rules.


If an analogy is thought helpful, a resolution of the US Congress does not have precedence over the terms of the US Constitution with which it is inconsistent. The terms of the Charter take precedence over resolutions of the Assembly that are inconsistent with it, as those are the rules all parties agreed to.


Those who wish to rely upon strict legality should place no reliance at all upon Resolution 377A. If it had amended the Charter it would lower the bar for collective action. It has not.


Chemical Weapons

What of the Chemical Weapons Convention? Does Syria’s breach of that Convention justify military intervention?


As a matter of law, the answer is no.  One state breaking international law does not, alone, permit another state, or group of states, using military force against it.


Defence of the bright line rule against chemical weapon use may provide a moral argument as to why intervention is justified here, when it is not on other occasions, but it forms no part of the UK’s ‘humanitarian intervention’ legal justification for action. Theresa May in the Commons today understandably sought to conflate the two reasons, but they are quite separate.


Responsibility to Protect

What of the new international responsibility to protect doctrine (“R2P”), has that altered the position? The short answer is given by Professor Akande: it in no way altered the UN Charter rules on the use of force. This is clear from para 139 of the World Summit Document agreed by the Heads of State of the General Assembly. The responsibility to protect only arises within the confines already established by the Charter, it creates no new liberty to use force,


Article 2(4)

It has been argued that the wording of Article 2(4) of the Charter is not as restrictive in setting out a limit on the use of force as I have suggested. It provides:

All Members shall refrain in their international relations from the threat or use of force against the territorial integrity or political independence of any state, or in any other manner inconsistent with the Purposes of the United Nations.

So, it has been argued by Professor Anthony D’Amato (here at pp 4/5) that where the use of force does not threaten the territorial integrity of a state (as dropping a bomb on an airport does not) or its political independence it is not prohibited.


On this view, if the UK government decided to drop a bomb on Paris for a laugh that would not be prohibited. That is not a plausible reading of the provision. Dropping bombs on other states is the use of force against their political independence (states claiming a monopoly on when force is lawful within their territories).


[It may be noted in passing how badly drafted art 2(4) is. Something has gone wrong with the grammar. Should the last clause be read as “in any other manner [use force] inconsistent with the Purposes of….”?]


Customary International Law

The origin of the UK’s case for a right of humanitarian intervention is this article by Christopher Greenwood, pp 161- 171 of which should be read alongside Akande’s as the case for the government (pending the release of the advice that forms the basis of its published opinion.) The argument is that such a right has arisen by custom, through interventions in Western Africa, Kosovo, Iraq and elsewhere.


Akande has the better of the argument, but he obscures this with the order of his presentation.


Let it be assumed, arguendo, that a custom of humanitarian intervention has been established. Assume that countries like the UK, Russia, China and France have routinely, over a long period of time, used it as a ground for the use of force. Let us assume that there is widespread agreement amongst commentators that this has happened, and lots of reports and statements from a large majority of countries accepting that this has occurred. Does that suffice?


In the absence of the UN Charter, it would. Absent agreement, we must infer the positive law from customary usage. (And where that too is missing from natural law reasoning: It is just wrong to invade other countries just as it is wrong to murder people regardless of the absence of any custom or  positive law on the point.) Once however there has been an agreement entered into, this replaces the need to look to Convention. As in other areas, such as UK domestic law, Conventional rules are subordinate to, and rendered otiose by, posited legal norms. The rules applicable to Syria are determined by the agreement (the Charter) it has entered into. Conventions established by other states are neither here nor there.


It is possible to vary an agreement. This variation may be inferred from conduct, and not just express provision. However, the hurdle for proof of this is very high where the parties have, under their agreement, specified how variation is to occur (as they have in Article 108). Syria would also have to be a party to any such variation. Nobody seriously entertains that this could be established in this case.


So, before we get to the question of whether any such Convention can be established as a matter of fact (and I consider this doubtful) the preliminary question is would it matter if we could? The correct answer is that it would not.



Lawyers are prone to two kinds of romanticism, both of which are false and need to be resisted.


The first is that the law answers definitively the moral question of how to act and no further difficult and contentious moral enquiry needs to be entered into. Rules are important, and things will usually go better if we guide ourselves according to good rules set down in advance rather than using our own individual notions of what justice requires. Rules are not however the only things that matter. Trying to prevent the murder of innocents through chemical weapons, for example, may be an important consideration on the other side. The reasons for the ability of Permanent Members of the Security Council to block collective action through force are found in the political interests of Britain, the USA and the Soviet Union as negotiated at Yalta in 1945. It has little or nothing to do with considerations of what justice might require. We cannot assume that the rules are not abhorent, let alone that they are good.


The second is that we must assume that the positive law is just, and where it appears not to be interpret it, or add to it, or ignore it, in such a way that it is. It is this trap that those who argue that there is a right to humanitarian intervention are falling into. It seems wrong, as a matter of justice, that the rules should prioritise the sovereignty of nation states (which are legal constructs) over the interests of human beings. Surely there should be a rule permitting unilateral intervention in the case of genocide, therefore, (so the reasoning goes) there is. However, if we think like that, why bother with the positive law at all? Why not just cut to the chase and ask, in each case, what does justice require? The central point of the positive law is to give us guidance independent of what justice requires.


We should not be romantic. We should have the courage to look at the positive law and accept that it is unjust, as in some respects international law is.



The number of bad moral arguments that can be made is infinite, but here are two prominent ones.


What about Myanmar?

There is a large amount of injustice in the world. If it is morally required to intervene in this case, why not in all cases? Why no intervention to stop or deter the killing of Rohingya people in Myanmar?


This kind of “what about” argument is never very persuasive, but first it must be conceded that if intervention is morally justified in the case of Syria, it is morally justified in all morally identical cases. Like cases should be treated alike.


However, three further points may be made.


First the use of chemical weapons morally differentiates this case. The use of chemical weapons in World War I led to the 1925 Geneva Protocol banning the use of chemical weapons. This bright line rule has been respected for purposes of combat for nearly a century (chemicals have of course been used against human beings by states). Although someone left to die in agony by a cluster bomb may suffer as much as someone poisoned by chlorine gas, the absolute prohibition was an advance that is worth defending.


Second, ought implies can. The states intervening in Syria lack the practical ability to end or deter many injustices in the world. Without the capacity to act there is no duty to do so.


Third the moral argument in favour of intervention may only establish a liberty to intervene, not a duty to do so. Exercise of the liberty in one case does not require it in all others. We should not refrain from trying to stop one injustice on the basis that we cannot stop all others.


The Bad Motives of Leaders

What if, for the sake of argument, the leaders of the intervening powers had bad motives? Say President Trump’s motive was to distract attention from the Stormy Daniels affair, or Theresa May’s was to improve the local government election results. Does that make it immoral to follow them?


No. If our reason for approving of the bombing were the unthinking following of leaders, rather than preventing the murder of civilians by chemical weapons, then their bad motives may be imputed to us. The true justification for the state’s action (or lack of it) is independent of the subjective motives of the leaders of the country carrying out the act. We cannot assume that because it is carried out by Trump that it is, therefore, wrong.


A better argument might be the fear of escalation, that bad leaders with bad motives cannot be trusted not to go too far. That maybe so, but their bad motives alone cannot be a determining factor. The evil man may be prudent, whilst the good recklessly ambitious.

Legality, Justice and Syria


The law is the easy part.

Absent agreement between them, the law between states can be found in Conventions and, where even they are absent, reasons of justice. Since 1945 however we have had the United Nations Charter, that was created by agreement. In areas it covers, it provides us with the positive law, replacing what was there before. The central point of the positive law is that it gives us guidance independent of reasons of justice. It closes off appeals to them. If it didn’t, the positive law would have little point.


Article 2(4) of the UN Charter has a prohibition on the use of force. This allows of two exceptions. The first is in Chapter VII, where the Security Council has been given power in order to fulfill its responsibility for the maintenance of international peace and security. The second is the right to self defence in article 51. (I argued back in 2015 that the drone strikes against ISIS targets were justified on the basis of the defence of the state of Iraq.)


Humanitarian Intervention

Take an extreme example. Say a government of a nation state led by a brutal dictator was carrying out a campaign of genocide against a minority group. Let it be assumed that it can be proven beyond peradventure that dropping a bomb on an airfield, hurting nobody, could prevent this. One permanent member of the security council opposes this action. Can other countries, individually or collectively, drop the bomb saving millions of lives?


As a matter of the positive law, the answer is clearly no. There is no ‘humanitarian intervention’ exception in the text of the Charter. Maybe such a rule could have been justified absent the Charter, but it now covers the field. There is no way of interpreting the words as meaning anything other than what they say.


That the Charter allows for no such exception is apparently astonishing given the context of its drafting. However, in 1945 it was envisaged that such action would be taken collectively. There is no get out where one permanent member of the security council opposes intervention.



It follows from the two propositions that international law on the use of force is found in the Charter, and that the Charter by its words only permits the use of force in the two situations described, that I consider that the bombing in 1999 by Nato forces in order to (successfully) prevent the ethnic cleansing of thousands of Albanians as contrary to the positive law.


That does not, however, answer the moral question of whether it was wrong.



Governments have good reasons for complying with international law.


First, that is what they have agreed to do. Pacta sunt sevanda: agreements are to be kept. This is principle of justice that applies to states as much as to people. States agreed to the UN Charter, and if they do not like certain features of it then the correct course is to vary it through further agreement.


Second, breaking international law undermines respect for it. Why should other states comply with international law if we do not? Unlike domestic law, international law is (largely) free of any sanctions for breach.Without an enforcement mechanism, the only way the United Kingdom has of encouraging others to comply with their treaty obligations is for us to comply with ours. The absence of a strong regime of sanctions for breach means that the United Kingdom has a stronger, not weaker, moral reason for compliance.


Third the moral question of whether to intervene in cases such as Syria is often extremely fraught and difficult. Dropping bombs may deter the illegal use of chemical weapons, but such illegality by the Syrian state does not alone provide other states a legal privilege to drop bombs upon it.  On the other side, dropping bombs kills people, many innocent. There is a danger of escalation. It would be better if the question of when action is permitted and required were settled by rules drawn up dispassionately in advance, and that is what the UN Charter tries to do.



But although the positive law has great weight in our moral reasoning, it does not have infinite weight. Sometimes breaking the law is the right thing (morally) to do. Two facts lessen the weight international law has.


First, it is obviously deficient, as my extreme example of the single bomb to prevent genocide illustrates. The bar to collective action is set too high. One permanent member of the UN Security Council having power to block any collective action entails in practice and not just in theory that no legal force will be used in some cases to prevent even the most grotesque crimes committed within nation states.


The explanation, but not justification, for the high bar is realpolitik, not justice, Members of the security council wanted, and want, the power to block the lawful exercise of force against themselves or allied countries. There is no practical prospect of this being changed. In this respect, international law is in a worse state than it was before 1945 as a matter of justice.


Second, the question of whether to comply or not with international law is an example of the prisoners’ dilemma, All states will be better off if there is universal compliance, but if some states follow the rules, whilst others do not (by for example annexing part of a neighbouring state) that may place those who comply with the rules in a worse position than they otherwise would be.


UK Government’s Legal Position


The UK government sets out its legal position as follows


The legal basis for the use of force is humanitarian intervention, which requires three conditions to be met:

(i) there is convincing evidence, generally accepted by the international community as a whole, of extreme humanitarian distress on a large scale, requiring immediate and urgent relief;

(ii) it must be objectively clear that there is no practicable alternative to the use of force if lives are to be saved; and

(iii) the proposed use of force must be necessary and proportionate to the aim of relief of humanitarian suffering and must be strictly limited in time and in scope to this aim (i.e. the minimum necessary to achieve that end and for no other purpose).


There is no textual support for this within the UN Charter at all. It is implausible as a statement of the positive law.


It is however a good statement of what the law ought, as a matter of justice, to be. There is a danger of judging these things in hindsight. The bombing of the former Yugoslavia “worked” (whereas other interventions have not) but my judgement, for what that is worth, is that it was right but unlawful, regardless of the outcome.


The UK government has an interest in respect for the rule of law, both nationally and internationally. It cannot say “this action is right, albeit unlawful.” So, it portrays what is in fact an argument of justice as one of legality.


Whether the decision to bomb is, morally, the correct one cannot be definitively answered for us by the positive law, Each of us, in this regard, is on our own.

Why Punish?

Retribution is poor justice”

-Simon Jenkins


Politicians are relentlessly forward looking. What will make things go better in the future? Justice, by contrast, usually looks backwards. What do these people deserve, given what has happened? These perspectives are often in conflict with one another.


If we could prove, beyond peradventure, that locking a certain class of person up for long periods of time decreased their propensity to commit crimes, would it be ok to do so? Would it be ok to punish you if it made you a better person?


If we could prove, with absolute certainty, that locking up certain people would deter either them or other people from committing crimes, would it be ok to do so? Philip K Dick in his (brilliant) short story The Minority Report imagines a world where criminals could be arrested and detained for crimes before they’re committed because of “precogs” able to foresee the future. Would that be ok?


If an individual commits a grotesque crime, but immediately repents and becomes a thoroughly good person, who we know for certain would never offend again and whose punishment would deter nobody else from behaving similarly, should the state forgive and forget?


My answer to all these questions is an emphatic “no”. Instrumental, forward looking justifications for punishment, such as rehabilitation and deterrence, are insufficient. They do not suffice because it is immoral to use people as a mere means to an end. That is the point of Dick’s story, and the mockery by Voltaire of the British execution of Admiral Byng.



If I am under a duty to do (or not do) something, what happens to the duty once I have broken it? If I am under a duty not to murder you, and I do so, I can no longer comply with it. Does the duty then disappear into the ether? Is it now so much spilled milk? Should we all now look remorselessly forward, and decide how things would now go best, ignoring the past?


I would suggest not. The original reasons for the duty do not disappear once the duty has been broken. Those reasons persist. They now provide the explanation as to why the wrongdoer deserves to be punished.


If a metaphor is thought helpful, a criminal, by breaking his duty, has taken a freedom that he does not rightly have. The original justification for the duty now also provides the explanation for removing an equivalent freedom from him.


Other Factors

The above defence of retribution explains why we think it only acceptable to punish criminals, and not innocent people, even though the reforming or deterrence effects of punishment may be unrelated to whether the person subject to them is or is not a criminal.


However, the degree of punishment that is appropriate is underdetermined without more. We can accept that murder deserves more than a few hours of detention, and parking on a double yellow line does not deserve life imprisonment, but at the margin there is no right answer to the question of whether a violent robber deserves one, two, three years or more.


Within this margin of uncertainty therefore, other instrumental factors, such as deterrence or rehabiliation may be brought into play in determining a sentence. But we should not punish beyond the range that retribution justifies, that would be to use the wrongdoer as a means to an end. We should not have a system of outlawry, where criminals fall outside of the constraints on punishing people more than they deserve. A judge in sentencing should therefore start with the question: what is the range of punishment that retribution justifies? Within that range, instrumental reasons can then be brought into play in determining the sentence.



John Worboys was convicted of nineteen offences, including one count of rape.  He is however believed to have committed more than 100 rapes and sexual offences on women between 2002 and 2008. He has served over 10 years in custody, including a period on remand. After a parole board hearing in November his release was approved, subject to stringent licencing conditions. The trial judge had given Worboys a 16 year determinate sentence. Prisoners are released at the halfway stage of their sentence, hence eight years would be the ordinary point of release.


Imprisonment for Public Protection

Worboys was also sentenced to “imprisonment for public protection.” This enabled the detention of criminals indefinitely after the expiry of their tariff. This was introduced by David Blunekett in the Criminal Justice Act 2003, and abolished by Kenneth Clarke in 2012. This abolition was not however retrospective. Many hundreds are still detained on  IPP sentences for many multiples of their original tariff. Lord Brown, an under-appreciated member of the UK Supreme Court described this as a terrible scourge.  We are (still) living in Philip K Dick’s nightmare, but without the benefit of the pre-cogs able to tell with near certainty who will commit crimes. It is a disgrace.


It is government policy to release IPP prisoners where the risks they pose are manageable. The parole board have decided that this is the case with Worboys.


For Which Crimes?

It is only acceptable to punish people for the crimes they have been proven to have committed. We cannot punish Worboys for crimes he has not yet committed, nor for crimes for which he has not been convicted.


Why was Worboys tried for so few crimes, when it is thought that he committed many more?


I do not know the answer to this important question. There are two possibilities.


The first is that there was insufficient evidence to secure a conviction for other offences. A successful claim was brought under the Human Rights Act against the Metropolitan police for their failure to investigate complaints of serious sexual assaults committed by Worboys. Although this is currently subject to an appeal before the Supreme Court, this concerns the scope of the legal duty owed, there is no doubt as to the shambolic nature of the police investigationa as a matter of fact.


The second is that it was decided not to be in the public interest, presumably on the basis that Worboys was subject to an indefinite sentence in any event. If this was the basis of the decision, in retrospect it appears mistaken.


Deterrence, Rehabilitation

If then it is impermissible to punish save as justified by the crime committed, how else can we protect the public, by deterring crime and rehabilitating offenders? We primarily try to deter crime by having a police force. We seek to monitor and rehabilitate offenders after release through a publicly funded probation service. The last is only widely appreciated when it is absent,  after someone who has been released goes on to commit another offence. So much easier to lock people up.

European Union (Withdrawal) Bill

The EU (Withdrawal) Bill is a damp squib. Its effect is to repeal the European Comminities Act 1972 (clause 1), whilst leaving in place all of EU law as it was in force in the UK prior to the date of Brexit (clauses 2-6). Ministers are then given the power to make consequential regulations (clauses 7-9), as is inevitable given that some of the law currently in force in the UK is in form and substance predicated on our being members of the EU. It is hard to see what other form such legislation could take. Clause 1 is probably surplusage, as EU law will already cease to apply to the UK and hence within the UK by virtue of the ticking bomb that is Art 50. The Bill is one of re-enactment, not repeal.


One point of interest is that the EU Charter of Fundamental Rights is “not part of domestic law on or after exit day” (Cl 5(4)), and so is not to be brought into UK law after exit. This is potentially politically significant as the shadow Brexit secretary, Keir Starmer, has given the failure to incorporate the Charter into UK law as a reason for voting against the Bill. 


The Charter of Fundamental Rights

In order to understand the Charter, and its current position in UK law, it is useful to contrast the Human Rights Act. The Act implements Convention Rights into UK law in two ways. First legislation must be interpreted so as to be compatible with Convention rights.  Second public authorities must not act in a way that is incompatible with Convention rights.


The Charter is addressed to the institutions of the EU, and to the Member States when implementing EU law. After Brexit, its applicability to institutions of the EU will no longer be a UK concern. The Charter is not a source of any freestanding rights against member states or anyone else, other than the institutions of the EU. It does however apply to EU law. Its effect can therefore be seen as analogous to the interpretive duty in the Human Rights Act, but only applies specifically to EU law. It cannot therefore be relied upon directly, but as part of a claim or defence to expand or restrict a right conferred by EU law.


If the Charter were in the same terms as the ECHR no problems upon Brexit would arise. It is however, a more “modern” instrument, covering many goods and interests that the 1950 Convention does not (eg rights to personal data, integration of those with disabilities, cloning) and some social rights the Convention’s drafters balked at including (eg rights to health,  social security.)


One Charter right is deserving of special mention. Article 45 guarantees freedom of movement and residence for every citizen of the Union within the Member States. It must be doubtful whether Sir Keir’s new found enthusiasm for the Charter will extend to this central provision.


The British “Opt-out”

As the Charter applies to EU law and institutions, and not to UK law and institutions as such, the British (and Polish) “opt-out” from it, as negotiated by the then Labour government as part of the Lisbon Treaty that brought the Charter into effect, was peculiarly fatuous. It provides 

The Charter does not extend the ability of the Court of Justice of the European Union, or any court or tribunal of Poland or of the United Kingdom, to find that the laws, regulations or administrative provisions, practices or action of Poland or of the United Kingdom are inconsistent with the fundamental rights, freedoms and principles that it reaffirms.

As the Charter does not extend that ability with regard to any member state anyway, this opt-out meant nothing at all.


The Problem

The problem with the Charter is that it seems to have the vices of other human rights instruments, without the corresponding virtues. One of the central criticisms of human rights instruments is that they contravene the rule of law. Instead of detailed specifications of the duties we owe, what we find are goods or interests specified at a high level of generality that ground such duties. They not only permit but require judges to depart from the interpretation of other enactments that they would otherwise reach, based upon open textured values. This greatly increases the power (and obligation) of judges to make law.


It may be that we consider the resultant uncertainty and departure from democratic accountability to be justified where it forms part of an international legal order between signatory states. The European Convention is of this kind: a set of minimum standards that seek to ensure that all states meet certain standards. Incorporating such an open-textured set of values without more within a single system, such as the EU, and mandating judges to depart from the ordinary meaning of enactments based upon them, is hard to justify.


The Options

Upon Brexit, what are the options with respect to the Charter?


One option would be to apply it to all UK law and institutions. A new super-Human Rights Act. Nobody is suggesting that.


The second option would be to apply the Charter (presumably shorn of its embarrassing freedom of movement provision) to the re-enacted EU law. This seems anomalous. It would mean that one part of UK law was subject to a different human rights regime from the rest, based upon its historical source.


The third option is to do what the Labour government in 2009 tried, and failed to do, and opt-out of the Charter altogether. It is that option that the EU (Withdrawal) Bill seeks to achieve.

May’s Duty to Resign

In the New Statesman Stephen Bush, the brightest and best of the next generation of political commentators states:


Thanks to the Fixed Term Parliaments Act, “confidence votes” have been explicitly drawn to exclude votes on the Budget or the Queen’s Speech. A government only falls if it loses a vote of no confidence. It no longer falls if it loses a major vote, a Budget vote or even the Queen’s Speech.

This obviously increases the leverage of the DUP – and Labour’s ability to harry the government day-to-day. The DUP can hold the government up, by backing them in confidence votes. But they can also let them down by deserting them on essentially everything else to secure bigger concessions from the Conservative Party.


This repeats the error of thinking that the FtPA changes the duty of the Prime Minister to resign when she has lost the confidence of the Commons, and advise the sovereign to call on someone else who commands a majority.

The essential question is: What does the Fixed-term Parliaments Act fix? Does it fix Parliaments or governments? The correct answer is that it fixes Parliaments, and when a Prime Minister is under a duty to resign is wholly unaffected.


Parliaments and Governments

Parliaments begin when they are summoned by the sovereign. The next Parliament starts tomorrow on 13 June. They end when they are dissolved. We have had no MPs since the last dissolution.


Governments by contrast begin and end with Prime Ministers. So, last month the UK continued to have a government, and Theresa May continued to be Prime Minister, even though she like everyone else elected to the previous Parliament ceased to be an MP.  Blair’s government ran from 2 May 1997 to 27 June 2007, covering two and a bit Parliaments. Theresa May’s government started on 13 July 2016. This was a new government on that day, although she herself came from the same party as that of the previous government. However, in constitutional terms this is not significant.


May did not form a new government after the 8 June election. Her previous government continued. There was no constitutional reason whatsoever for her to pay a visit to the Queen as she was not resigning. This practice should be deprecated as its symbolism gives the indication that the matter of who can form a government is settled, when this is a matter for the Commons.



Conventions are rules that are not posited. Nowhere is it set down by a figure in authority that men should not wear a hat in church. In our constitution conventions constrain those who have wide discretionary powers as a matter of law. Looking at the posited law alone, the sovereign has the power to dismiss Theresa May and call on another to act as Prime Minister. It is convention that restrains her from doing so.


Before the FtPA, the sovereign had the power to dissolve Parliament. By Convention, the sovereign exercised this power at the Prime Minister’s request. This Convention has gone because the legal power to which it related has gone by virtue of the FtPA. This weakens the position of the Prime Minister.


Before the FtPA, a Prime Minister who lost an express vote of confidence in the Commons had two options. One option was to call for a dissolution, whilst remaining as Prime Minister pending the outcome of the election. This was the course Callaghan took in 1979. The other is to resign, and advise the sovereign to call on someone who can command the confidence of the Commons.


The last time the latter happened was in 1924. Baldwin was Prime Minister going into the election in 1923. The Conservative party had a plurality of seats but was well short of a majority. Upon reconvening, the House debated a vote of no confidence, which was passed, causing Baldwin to immediately resign, advising the King to call on MacDonald.


The FtPA has removed the option that Callaghan took. This means that a Prime Minister who no longer commands the confidence of the Commons must resign.


What is a Vote of Confidence?

The FtPA defines the kind of motion that must be passed for an early general election. It does not define the kinds of motion that indicate that the Commons has no confidence in the Prime Minister and must resign.


Clearly a motion may be expressed by its terms as one of confidence. Alternatively the Prime Minister may stipulate in advance that he will treat it as one of confidence and resign (or in the past ask the sovereign for a dissolution) if it is not carried.


Sometimes motions are implicitly ones of confidence. The two examples of this are votes on the Queen’s speech, and Budgets. If a government cannot pass a Budget it cannot govern, and the Prime Minister must resign and ask the sovereign to call on someone who can carry a majority. However, I think the loss of a Budget vote is merely indicative: it is possible to postulate examples where the Prime Minister maintains the confidence of the House for budget, just not the budget presented, as I explained here.



Other votes may similarly implicitly indicate a loss of confidence requiring resignation. In 1940 following the Norway debate. The government Chamberlain led actually won the vote, but with such a large defection of Conservatives to the opposition motion that it implicitly showed he had lost the confidence of the Commons.


What must a modern day Chamberlain do, after the FtPA?


The same thing. Resign. You have no choice. What does this mean for the DUP? It means that unless they wish to compel May to resign, they must support budgets.


Addendum: 15 October 2018

Brexit negotiations have made this post very topical, and so some updating is requried.


The DUP are threatening to oppose the Budget on 29 October if May agrees to keep Northern Ireland in the customs union with the EU, with the rest of the UK outside, as this would effectively mean a “border” in the Irish Sea.


The reason that Budgets are confidence votes is that they raise revenue for the year. Income tax is imposed annually and must be reimposed with a fresh Finance Act each year. A government cannot govern without one. Amendments are one thing, being unable to pass a Budget at all quite another.


So, May cannot agree to a Brexit deal on 16 October as she would almost certainly lose a vote on the Budget, and be forced to resign. (If forced to resign the sovereign is required to call on another capable of commanding the support of the Commons. That probably means someone else from the Conservative party to try to “negotiate” again with the EU, and not the leader of the opposition who certainly does not command majority support.)


The “meaningful” vote in January/February required by the EU Withdrawal Act will not, by contrast, be a confidence vote. If May agrees to keep Northern Ireland in the customs union, with the rest of the UK outside, she must hope to peel off sufficient Labout MPs to win a vote on her deal. (I’d expect the opposition leadership to whip to oppose any deal put forward by the government, their strategy being to oppose the government in everything it does, but not Brexit itself.)

Sources of Law

What is a source of law? To be a source is to be the originator. The source of a stream comes before the stream. The source of a leak is a necessary condition of a leak. The source of an idea is anterior to the idea.

An Act of Parliament is a source of law. An Act is constituted where a Bill passes both the House of Commons and the House of Lords and receives the Royal Assent from the sovereign. But how do we know this is so? What tells us that this is a source of law?

A beginner might naively think that there ought to be a yet further more foundational document that tells us what the sources of law in a State are. In the United States, Article 1 of the Constitution stipulates that legislative powers are vested in the Congress, which consists of the Senate and the House of Representatives. Copies of this document are on reverential display in the National Archive in Washington.

But why is the US Constitution a source of law? Is there a still yet higher document, kept in an even more secure glass box, that tells us that? If there were we would hit an infinite regress problem.

HLA Hart in The Concept of Law postulated that any system of positive law must have a Rule of Recognition at its base. This rule has as its primary function the provision of a test for the validity of a law. Examples of such rules are that a Bill that passes both Houses and receives Royal Assent is good law, or that the Constitution of the United States is valid. Such rules as they are a necessary condition of what constitutes a positive law cannot, as a matter of logic, themselves be posited. Where then do they come from?

Hart suggested, plausibly, that they arise by Convention among officials (usually judges) whereby they accept such rules. Conventions are a matter of fact and can change overtime. It is a Convention that men do not wear a hat in church. This Convention has not been posited anywhere. Although Hart was the greatest exponent of legal positivism, his account is dependent upon a rule that is not posited by authority at all.

This account works less well in relation to international law. Often there is no official or judge available to determine what international law is as no formal court has been set up to adjudicate on the matter. How did we know in, say, 1648 that the Treaty of Westphalia was legally binding? Because States by Convention treated, and treat, agreements entered into between them as binding.

Where there are multiple sources of law, the Rule of Recognition should (must?) also have rules for determining a hierarchy between them in cases of conflict. The common law, judge made law, is a source of law in the United Kingdom. It is not dependent for its validity upon any Act of Parliament. What happens when the common law says X and an Act of Parliament says not-X? The answer is that the legislation prevails. As a matter of logic, the legislature itself could not have alone validly stipulated that its Acts were to prevail. Further fine tuning is possible. So, the judges now seem to have recognised a special category of ‘Constitutional Statutes” that did not exist when I was young, which seem to have some kind of interpretive priority over other more humdrum Acts.

In the Miller decision the majority seek to meet the interpretive objections of the minority, by postulating that EU law is a source of law, and as such cannot be removed by ministerial action alone. This suggestion first came from Lord Sumption (see transcript at page 40). This seems plausible. After all, UK legislation is without doubt a source of law, and EU law has priority over it within our domestic legal order. EU law can be directly effective as soon as it is posited and is not dependent upon new legislation for its validity. It would be, to say the least, startling if ministers could, by their decisions, remove legislation or the common law as a source of law, so how could they do so with any other source?

Further, describing EU law as a source of law seems to circumvent Lord Reed’s nitpicking legalistic objections based upon the words of the European Communities Act. If EU law is a legal source it is anterior to the positive law itself, and is not determined by it.

Unfortunately, this characterisation of EU law is wrong. If the legislature passed an Act stipulating that the rules of the game Monopoly as they exist “from time to time” were henceforth to be legally enforceable, and to take priority over other rules, would that make those rules a ‘source of law’? They would not. The source would be the rule that Acts of Parliament are valid. The incorporation of the rules of the game would be wholly dependent on that rule.

Precisely the same is true of EU law within UK domestic law. It is only valid because of the European Communities Act. That Act in turn is dependent upon our Rule of Recognition in the UK. EU law is directly effective because the Act says so, and has priority over UK legislation because that is what the Act says and the judges have given effect to that. If EU law had its own freestanding validity it could not be removed even by an Act of Parliament overturning that Act.

At an international level, EU law is not dependent upon any UK legislation. But on that plane it takes its validity from the agreement between member states. It is not a new sui generis form of law.

Lord Sumption’s clever move of describing EU law as a ‘source of law’ has a strong rhetorical attraction. He was the best advocate of his generation. It is a sleight of hand.

The Vibe of the Constitution

One day I will provide a review of films for lawyers. One of the very best, that should be better known in the UK than it is, is the Australian movie, The Castle. The most famous scene is when a lawyer is asked to explain why his client should win. He cites “the Constitution.” When pressed on which particular provision of the Constitution he relies upon, he is unable to point to any specific section, but instead relies upon “the vibe of the thing”

In his dissent in the Miller decision, Lord Reed (at [177]) makes an argument that I have made repeatedly on this blog since last June

“the effect which Parliament has given to EU law in our domestic law, under the 1972 Act, is inherently conditional on the application of the EU treaties to the UK, and therefore on the UK’s membership of the EU. The Act imposes no requirement, and manifests no intention, in respect of the UK’s membership of the EU. It does not, therefore, affect the Crown’s exercise of prerogative powers in respect of UK membership… Further, since the effect of EU law in the UK is entirely dependent on the 1972 Act, no alteration in the fundamental rule governing the recognition of sources of law has resulted from membership of the EU, or will result from notification under article 50. It follows that Ministers are entitled to give notification under article 50, in the exercise of prerogative powers, without requiring authorisation by a further Act of Parliament.”

It is good that this argument, that was astonishingly not made before the High Court (see para 11) was at least put. It is based upon the express words of the 1972 Act. What is the response of the majority to it? It is on this that the result of the case turns. The key part of their reasoning is at paragraphs 78 to 81. What do they say?

The first claim is that there is a “fundamental difference” between variations in the content of EU law and withdrawal from the European Union, This is because the latter involves a unilateral action by the UK that changes UK constitutional law, whilst the former does not. If EU law were eliminated save for one directive on banana curvature, that would be fine, whilst its reduction to nothing at all would not.

This is very difficult to understand. It has no textual support as a distinction in the European Communities Act itself. The change in the application of EU law to the UK by withdrawal is mandated by EU law itself (article 50). There was and is nothing unilateral about that, it was agreed between the Member States. EU law changes in many different ways. Why is this change ‘fundamentally different’ from the others? Appeals to the long title and the side notes (para 88) of the Act might be justifiable if the words of the sections (which are the law) were ambiguous or could plausibly be said to have a meaning that they do not have on their face. They do not, and no reliance is placed upon the words of the Act. As Lord Reed states, no section in the Act states “the UK shall be a member of the European Union” or any equivalent.

Second it is said that EU law is a source of law, and not just law. It would be “inconsistent with long-standing and fundamental principle” for such a change to a source of law to be brought about by ministerial decision.

Again, this is difficult to understand. EU law has its status within UK law only because of the European Communities Act. It has no independent force in our domestic law. By contrast the common law, which is judge made, has no statutory basis. If EU law really were a source of law independent of an Act of Parliament, how could an Act of Parliament remove it from UK law? Legislation could not, for example, provide that future legislation ceased to be a source of law, nor could the legislature provide that what judges decide is no longer legally binding upon the parties without abolishing law within our legal system altogether. EU law within our domestic legal order is simply not like that. It is not an independent source of law, but is wholly dependent upon UK legislation. It is because it is not a source of law (ie anterior to posited law) that it can be removed by law.

The truth is that EU law is only effective within UK domestic law because the European Communities Act says so. It is, politely, surprising that the majority base their decision not upon the words of the Act, which in the key paragraphs they ignore, but on a rather more abstract enquiry into what they feel the right answer ought to be.

In the end, the case will probably be of little practical significance. We will leave the EU, and the dispute about what was once EU law’s constitutional status within the UK will become an historical curio. That said, and despite several attempts, I see no answer to Lord Reed’s dissent in the reasoning of the majority, and have consequently found it difficult to articulate what they are saying. Vague statements about ‘fundamental principles’ and basic ‘sources of law’ are little better than appeals to the vibe.