The Corbyn Coup

A live legal question raised on the frontpage of today’s Times is whether the leader of the Labour party, Jeremy Corbyn, needs to obtain the nomination of 20% of the Parliamentary Labour Party in order to be a candidate for any new election for the leadership. It is claimed that legal advice has been sought, and that its substance is that any candidate, including Corbyn, would need to be nominated by that proportion of MPs (currently 38). Corbyn would be unlikely to achieve that number if it were required. The reasoning of the advice given is not disclosed.

 

This question has been considered by Jolyon Maugham QC.  Maugham is a tax barrister. He was responsible for Labour’s pledge in April 2015 to abolish the non-domicile tax status, Labour’s only clear win during the campaign. His blog is always well worth reading. He is no friend of Corbyn, but is of the clear view that the incumbent leader is not required to be nominated. I am no longer certain this is correct.

 

The Rules

 

The Labour party rule book is a contract between members. Its meaning is a matter of contract interpretation. Unsurprisingly the rules do not expressly deal with the situation where the leader has the support of less than 20% of the Parliamentary Labour Party, but wishes to stand.

 

The starting point must be the express words of the rule book. The strongest argument against the view that the incumbent must be nominated is the wording of Chapter 4 Clause II, 2 B ii which provides

 

Nominations

….

Where there is no vacancy, nominations may be sought by potential challengers each year prior to the annual session of party conference. In this case any nomination must be supported by 20 per cent of the Commons members of the PLP. Nominations not attaining this threshold shall be null and void.

A natural reading of this section is that it is only potential challengers who must cross the 20% threshold. However, the provision is not as clear as it might be. It does not say “a potential challenger must be supported by 20 per cent” but rather “any nomination must be supported by 20 per cent of the PLP.” It is silent as to whether the incumbent must be nominated.

 

There are indications in the rules that the election takes place between nominees. So, Clause II. 2 B v provides

Valid nominations shall be printed in the final agenda for Party conference, together with the names of the nominating organisations
and Commons members of the PLP supporting the nominations.

Clause 2 B vi states

Nominees who do not attend the relevant Party conference shall be deemed to have withdrawn their nominations, unless they
send to the General Secretary on or before
the day on which the conference opens

Clause 2 C x states

The votes cast for each nominee shall be recorded and published in a form to be determined by the NEC as soon as possible following any election.

These provisions seem to contemplate a competition between nominees, and this is reinforced by the heading of Clause 2 B: “Nomination”.

 

Maugham argues that the rules provide no provisions for a leader to be validly nominated. However Clause II, 2 B ii may be read as covering any who wish to be nominated, including the leader. It is true, as Maugham suggests, that the rules seem to use the words nominee and candidate interchangeably, but this supports the reading that the only contemplated candidates are nominees.

 

The Precedent

 

In 1988 after the 1987 General Election, Tony Benn challenged the incumbent Neil Kinnock for the leadership of the Labour party (an election the latter comfortably won). At that time both sought nominations from the parliamentary party . How can this impact upon the interpretation of the rules?

 

How a leadership election between an incumbent and a challenger has been run in the past cannot provide a binding precedent for how the rules are to be read. It may well be either that the rules were not understood at that time, or that Kinnock’s nominations were otiose and sought for political effect. Further, in interpreting a contract (in England), how the parties subsequently operate that contract is irrelevant to its interpretation. The words mean what they mean, not what the parties to the deal subsequently think they mean.

 

However, in interpreting words (as people, not just lawyers) we look at the context in which those words were used. The current rule book dates from 2014 (although many of the individual rules long pre-date that). That is the date at which the current agreement was entered into by members. Part of the background to that agreement is how the rules have been operated in the past.

 

Given that context, the way leadership contests have been run in the past may be used for purposes of determining whether any contest is to be between nominees only. My judgment (and I only have one) is that it is plainly arguable that it must be.

 

Update 1

 

In comments, David Boothroyd has kindly tweeted the provisions of the Labour rulebook in force in 1988. Back then the rules drew no distinction between the situation where there was and was not an incumbent, and required nominations of 5% of the PLP regardless. (This was subsequently raised to 20% to prevent Benn standing again). Under those rules it was quite clear that the incumbent had to be nominated.

 

The change to the current wording seems to me to have been made in order to require different levels of support where there is a vacancy and where there is not. Currently where there is a vacancy it is 15% and where there is not 20%. It is arguable that that was the only change that was intended, and not a change to the requirement that the incumbent be nominated.

 

Update 2

A rule change now includes MEPs within the numbers who can nominate and raise the threshold for nominations to 50. The rest of the analysis remains the same.

Unwilling or Unable

I have elsewhere explained why the UK acting against Da’esh target in Syria, at the request of the Iraqi government in the latter’s self defence is lawful. In acting against a non-state such as Da’esh the basis of such action is that the state from which they are attacking is ‘unable or unwilling’ to stop such attacks. Here I will seek to discuss some objections that might be made to this.

 

The UN Charter states

 

Article 51 Nothing in the present Charter shall impair the inherent right of individual or collective self-defence if an armed attack occurs against a Member of the United Nations, until the Security Council has taken measures necessary to maintain international peace and security. Measures taken by Members in the exercise of this right of self-defence shall be immediately reported to the Security Council and shall not in any way affect the authority and responsibility of the Security Council under the present Charter to take at any time such action as it deems necessary in order to maintain or restore international peace and security.

 

The first thing to note is that the right to self defence is not embodied in the UN Charter itself. Instead it states that nothing in the Charter shall impair the right of self defence (until the UNSC has taken the necessary steps to ensure peace). The defence is therefore recognised but not incorporated by article 51. Because it is not codified, the scope of self defence has to be determined by a combination of custom and ethical principle.

 

Does this right of self defence apply as against non-state actors? The answer to this question is emphatically (and perhaps surprisingly) no. The parties to international law are states: the right of self defence is one of three situations where it is lawful for one state to use force in another state’s territory (the others being a Security Council resolution under Chapter VII of the Charter, or with that other state’s consent). International law is much closer to a domestic law of contract than to a law of crime. The only parties to it are states, and their obligations are created by agreement which are either express or implied by custom.

 

Does that mean that any attack on Da’esh (which is not a state) is unlawful? Again, the answer is no. First, action within Iraq by the Iraqi government and those assisting it (such as the UK) is perfectly lawful because it is taking place within the territory of Iraq. Self defence is neither here nor there as Da’esh is not a state that could be wronged by military action taken against it within Iraq.

 

Second, action against Da’esh within Syria is potentially wrongful vis a vis the Syrian state (and not Da’esh). Self defence gives states a freedom to act vis a vis other states that they would not otherwise have. Da’esh (as a non-state) is invisible in the system of rights governing the relationship between states.

 

As Da’esh are not attacking Iraq under the direction or control of the Syrian government, does that mean that  Iraq (and its allies) have no freedom to use force within Syria? This would indeed be a surprising conclusion as it would mean that a non-state terror organisation such as Da’esh could attack states with impunity in a way that states could not.

 

The recognised answer is that where one state is “unable or unwilling” to suppress attacks coming from its territory, the attacked state may use force in the other’s territory to suppress it.

 

One objection to Iraq (and its allies) acting against Da’esh in Syria might be that the consent of the Syrian government has not been obtained. Should the UK ask for such consent before using military force in Syria?

 

Politically, there are good reasons why no UK government would in practice do so. It seems probable that in 2013 Assad used sarin gas in opposition controlled areas of Damascus, and other atrocities. Asking Assad for permission is not going to happen.

 

The point of seeking such consent would be to show that the host state was unwilling to tackle the non-state actor from within its borders. A refusal to consent, coupled with a failure to act, would demonstrate collusion by the harbouring state.

 

However, the refusal of consent would not demonstrate (or be relevant to) the question of whether the host state was unable to tackle the non-state actor. Since 2013 it has been proven beyond peradventure that the Syrian state is unable to suppress Da’esh.

 

If consent were sought from the Syrian government there would be two possible outcomes. Either consent would be given (in which case action against Da’esh in Syria would be lawful) or it would be refused (in which case action against Da’esh would be lawful on the basis that Syria was unwilling to take the steps necessary to suppress Da’esh). Such a formal step is not required as it is objectively demonstrable that Syria is unable to prevent attacks from its territory.

 

The United Kingdom has not been attacked by Da’esh, does that mean that we cannot act? The answer is, of course, no. If Belgium is attacked it is not just the state of Belgium that may act in its defence, but also those other states it authorises to act on its behalf. What is true of Belgium (or France) is true of Iraq. Collective self defence is expressly relied upon by the UK as grounding the legality of acting in Syria.

 

By resolution 2249 the UNSC called upon

 

Member States that have the capacity to do so to take all necessary measures, in compliance with international law, in particular with the United Nations Charter, as well as international human rights, refugee and humanitarian law, on the territory under the control of ISIL also known as Da’esh, in Syria and Iraq.

(emphasis added).

 

This was not a resolution under Chapter VII, and so does not itself give a mandate for military intervention in Syria by Member States. However, the phrase ‘all necessary measure’ encompasses military force. Implicitly the resolution endorses the proposition that it is compatible with international law for Member States to take these steps. If it were not so, the UNSC would be calling on Member States to behave unlawfully.

 

That acting against Da’esh in Syria is lawful does not, for good or ill, answer the ethical question of whether it is right to do so. That is a hard question that I shall seek to return to. What can be said however is that those who would claim that it is unlawful, or that yet further resolutions of the UNSC are required are trying to avoid giving an answer to that much more difficult question.

 

Apologies

“If he’s upset I apologise… But he can’t blame me.”

Ken Livingstone, 18/11/2015, World at One

 

If you wrong someone else you should apologise. We learn this as children. If you are a public figure who insults a class of people, although you cannot apologise to anyone in particular, you should take steps to express public regret and withdraw your words.

 

In law, when a person commits a wrong, it is very rare for courts to order an apology. The only partial exception to this I know of is the rule that an offer to make amends may be a defence to the wrong of defamation. Defamation is somewhat unusual however as a public retraction and apology can partially undo the injury that has been suffered, which is not the case for most wrongs.

 

The reason why the law does not track morality is obvious. An apology that you make because you have been compelled to do so merely has the outward form of an apology. It is a core part of what it is to apologise to be sincere. A court ordered apology would lack a component part of what it means to apologise. That does not mean that they would be pointless. They are equivalent to placing the wrongdoer in the stocks for ritual humiliation. That may give the victim some emotional satisfaction, but they are not the real thing.

 

Ken Livingstone

Labour’s shadow defence minister Kevan Jones was critical  today of Ken Livingstone’s appointment as a co-chair of Labour’s defence review. Mr Livingstone responded by saying 

 

I think [Jones] might need some psychiatric help. He’s obviously very depressed and disturbed. He should pop off and see his GP before he makes these offensive comments.

Mr Jones has suffered from depression, something he has given  widely publicised speeches in the Commons about. Initially Mr Livingstone refused to apologise, but the leader of his party then stated that he must do so. Mr Livingstone then gave the apology that appears at the top of this post.

 

When is an apology not an apology?

Livingstone’s words fall short of being an apology in three ways.

 

First he expresses regret for any adverse consequence (“if he is upset”) but not his own action (“he can’t blame me”). This is a denial of responsibility, not an acceptance of it. Any of us may express regret for a bad thing (“World War One was most regrettable”), but it is essential to an apology that you accept responsibility for what you did.

 

Second the gist of what he did wrong was nothing to do with whether or not Mr Jones was upset (I suspect he wasn’t). For a public figure like Mr Livingstone to make disparaging remarks about mental illness is a public wrong, not a wrong to Jones. His duty to apologise is nothing to do with any upset caused to anyone.

 

Third, just as with a court ordered apology, an insincere apology may be worse than none at all, and indeed compound the wrong. Mr Livingstone might have avoided the impression that he didn’t mean it by giving a fulsome retraction quickly. By “apologising” only after being told to do so, and in terms that are at best ambiguous, he insulted all of us.

 

Update

Mr Livingstone’s words on ITV were even further way from an apology than on the BBC.

 

His twitter feed has now carried this apology which seems better

 

I unreservedly apologise to Kevan Jones for my comments. They should not have been made at all, let alone in this context.I also make this apology because Jeremy is right to insist on a more civil politics and as a party we should take this seriously

This again mistakes who has been wronged, and indicates that the motive for the apology is not sincere regret but the insistence of Jeremy Corbyn on people being polite to one another.

 

In the evening Mr Livingstone was first interviewed on channel 4 with Mr Jones and then on his own on Newsnight. Both interviews are  watchable in a gruesome kind of way, but also give the flavour of the sincerity of the twitter apology under Mr Livingstone’s name.

 

 

 

 

Zoe Williams on Human Rights

Today there is a trivial polemic by Zoe Williams on the government’s Bill of Rights plans. It is riddled with mistakes. It is a good example of how columnists should not tackle topics they don’t understand.

In order, the errors I can spot are

The treatment of prisoners and the Human Rights Act are only partially related to one another: the most prominent intersection between the two was the European ruling during the last parliament that prisoners should get the vote

The decision establishing that the blanket ban on prisoners exercising the right to vote contravened the Convention was Hirst v UK (No 2)It was decided in 2005. Labour had five years in government to comply with it, and did not.

Under a new British bill of rights, soldiers and journalists are explicitly offered more protection from damages claims.

The source of this error is the Sunday Times report. You cannot bring claims against individual soldiers or journalists under the Act

In the first case, offences committed abroad will not be prosecutable, which represents a huge saving, given that £85m has so far been spent on claims from Iraq and Afghanistan alone.

This confuses the criminal and civil law. The Human Rights Act gives rise to no criminal offences which are ‘prosecutable’.

The Human Rights Act doesn’t even have compensation in its armoury

Yes it does, see section 9.

The Human Rights Act already says that judges should have regard to human rights violations yet not be bound by them.

think  this is a reference to the duty to take into account decisions of the European Court of Human Rights under s 2(1)(a). On its face however, it makes no sense.

the real threat is to Article 8 of the act

Article 8 is in the Convention. Acts are capitalised. It is wholly obscure why it is thought that Article 8 is peculiarly threatened.

The Extra-Territorial Impact of the Human Rights Act

In today’s Sunday Times we have a leak of the proposals to amend the Human Rights Act. Most of what is proposed seems to be empty nonsense, such as enabling the UK courts to take into account rulings from other Commonwealth jurisdictions. One significant proposal is that “human rights laws would in future apply only in Britain.”  One reason for opting for this reform may well be that it avoids any problems relating to devolution.

This is an issue the courts have repeatedly had to confront over the years, most recently in the UK in relation to the UK’s duties to its own soldiers and to civilians within the sphere of factual control of the British military in carrying out operations in Iraq.  There is an excellent summary of the history of this question by Lord Dyson, which should be read as he reaches the opposite conclusion from the one I do here. I shall not, as he does, seek to take the reader through the twists and turns of each case.

One of the allegations made by Professor John Finnis was, it will be recalled, that the European Court of Human Rights has adopted an overly expansive interpretation of the Convention that is inconsistent with the intentions of those drafting it. In this case, I agree.

Jurisdiction

Article 1 of the European Convention on Human Rights provides:

“The High Contracting Parties shall secure to everyone within their jurisdiction the rights and freedoms defined in Section I of this Convention”.

What does ‘jurisdiction’ mean? Does it mean ‘territorial jurisdiction’ (ie within Britain) or whenever the State exercises authority over another (wherever they happen to be)?

I have argued before that where courts interpret provisions that control a democratic state they should do so conservatively. This applies with even more force to an international court interpreting an open textured agreement between states.

There are at least three indications that the word ‘jurisdiction’ means ‘territorial jurisdiction’.

First, in interpreting the meaning of words we must look to the context in which they are used. The context of the convention includes earlier drafts and the travaux préparatoires. The initial version of article 1 had applied the Convention to those “residing within their territories”. This would have excluded the Convention’s operation from those who were within the UK but not resident (eg newly arrived refugees). The travaux préparatoires stated that the reason for the change of wording was that:

“It seemed to the Committee that the term ‘residing’ might be considered too restrictive. It was felt that there were good grounds for extending the benefits of the Convention to all persons in the territories of the signatory States, even those who could not be considered as residing there in the legal sense of the word.”

This justification for the change indicates that ‘jurisdiction’ means ‘territorial jurisdiction’. The point of the change was to bring within the Convention people who were non-residents but within the territory.

Second, the word ‘jurisdiction’ has a technical meaning in public international law. It meant then what it means now. To be within a state’s jurisdiction is to be subject to its legal sovereignty. That is where its laws applies. If the Convention uses a word with an established technical legal meaning in international law, it would be expected that the word has that meaning.

Third, the structure of the Convention indicates that all the rights come together as a package. You are either within the Convention, in which case you get all the rights, or outside of it, in which case you get none.  If a Convention State’s troops are in occupation of part of a foreign country, where the laws of that foreign country continue to apply, is the occupying state expected to secure the right to privacy or freedom of association? Clearly not. The primary way in which Convention states are expected to secure Convention rights is through their domestic law: which will not apply outside their own territorial jurisdiction.

The Court

The European Court of Human Rights decisively abandoned the interpretation of jurisdiction to mean ‘territorial jurisdiction’ in the Al-Skeini decision. Instead they interpreted the Convention to mean “whenever the state through its agents exercises control and authority over an individual.” This entails that British soldiers operating outside of the United Kingdom are within the Convention, as are non-UK nationals in areas under UK occupation.

This decision was, in my view, not only plainly wrong as an interpretation of the words of the Convention, but inconsistent with the Court’s own earlier jurisprudence. Earlier cases had extended ‘jurisdiction’ to cover embassies and prisons abroad, but this merely involved a consideration of how far the concept of ‘territory’ went. The decision in Loizidou v Turkey as interpreted in Banković v Belgium also indicated such a narrow interpretation. In the former case one Convention state (Turkey) excluded another (Greece) from part of its territory (northern Cyprus). As Greece was unable to secure the Convention rights of the persons situated there, in order to avoid there being a gap in protection, Turkey, the signatory state doing the excluding, was required to do so. If this were not so neither Greece nor Turkey would have had any duty to secure the rights of people within the territory of the Convention’s application.

The departure from the earlier authority in Al-Skeini cannot be justified on the basis of any kind of ‘living tree’ interpretation of the Convention. If we see the values set out in the Convention as embodying conventional as opposed to true morality, then their meaning can justifiably change overtime. What is an acceptable infringement of privacy in 1950, may not be in 2050.

The word ‘jurisdiction’ in the Convention is not of this kind. It is not a value of conventional morality, but a technical meaning fixed at the time of use. Unless the earlier decisions could be shown to be wrong (and they weren’t) they should have been followed.

The House of Lords (here referring to the previous label given our ultimate appellate court) and the Supreme Court had before Al-Skeini repeatedly and correctly concluded that the Convention had no extra-territorial application. The Supreme Court were then faced with the question of whether to follow the European Court of Human Rights, or depart from it. The court in Smith followed it.

Is the Government right then?

If, as I have argued, the European Court of Human Rights was plainly wrong in Al-Skeini does that mean that the UK government is right to cut back the scope of the Human Rights Act, and reverse Smith?

If we want other states to comply with the European Convention on Human Rights the only way the UK has of applying pressure to do so is to do so ourselves. We will not be sending troops into Russia to make sure they keep up to the mark. That entails that we comply with decisions of the ECtHR which are wrong, as I think Al-Skeini was. No court anywhere is right in every case.

United Kingdom domestic legislation cannot cut back the jurisdictional scope of the Convention. Changing our law will not reverse the ECtHR’s decision in Al-Skeini. Instead, the UK will have no remedy under our domestic law for breaches of the Convention outside of the UK’s territorial jurisdiction. That will place the United Kingdom in breach of its duties under Article 13 of the Convention to ensure adequate protection under its domestic law. That the ECtHR was wrong is not a sufficient reason for placing ourselves in breach of international law.

Against Human Rights (Part 2)

Here I turn to some arguments made by Professor John Finnis against the Human Rights Act. His recent lecture is centrally about judicial responsibility, but in the second half he returns to arguments he made 30 years ago against the incorporation of the European Convention on Human Rights.

There have already been two academic replies, by Professor Elliott and Professor Webber. Being critical, I would suggest that one of the strengths of Finnis’ lecture is that he deals, sometimes at length, with examples taken from the law, whereas the responses have been at a high level of generality.

The ad hominem issue

To non-specialists, Professor Finnis is probably best known as a Catholic who defends ethical positions on abortion, contraception and homosexuality that are out of line with our current liberal consensus (a consensus that I share). This conservatism can lead some to fail to engage with what he says. This is a mistake.

Professor Finnis is also one of the two or three most important legal philosophers alive in the English speaking world. He is also an acute lawyer (being a legal philosopher, perhaps surprisingly, doesn’t necessarily entail any knowledge of law).

Politics v Principle

One argument in favour of Bills of Rights in general, and the Human Rights Act in particular, is that the legislature is the natural forum for resolving questions of the pursuit of the common good, the courts for the protection of individuals against the majority.

It is true that courts are peculiarly ill-suited to resolve what may be broadly called ‘political’ questions of the general welfare. Judges merely have before them the parties to individual disputes and the arguments they choose to present. Judges are generally not very good at determining which result will turn out to be economically optimal. They lack the democratic mandate to legislate for all of us. Legislatures, for all their imperfections, seem to be far better as forums for the resolution of issues of general welfare.

As Finnis argues, however, it does not follow that judges are better placed to protect or determine the rights individuals should have to restrain the untrammelled pursuit of the majority’s view of general welfare. Legislators should, and observably do, consider this question. Judges achieve appointment and promotion because of their skill in the law. There is no necessary reason to believe that that gives them special insights into the rights individuals should have against the state to, say, the provision of healthcare. Does the Convention require them to fulfil this legislative function?

Human Rights Contravene the Rule of Law

The most significant criticism of the Convention and the HRA made by Finnis is that they contravene the rule of law.

The first problem he identifies is technical, and unjustified. To understand it, Article 8 of the Convention is illustrative

 Article 8 – Right to respect for private and family life

  1. Everyone has the right to respect for his private and family life, his home and his correspondence.
  2. There shall be no interference by a public authority with the exercise of this right except such as is in accordance with the law and is necessary in a democratic society in the interests of national security, public safety or the economic well-being of the country, for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others.

In language we use ‘rights’ in different senses. Usually when lawyers speak of rights, they are talking about specified conclusions. If I say “I have a right that you do not punch me on the nose” or “I have a right that you do not walk across my field” or “I have a right that you pay me $100 for the books I delivered to you” I am making a claim about specific duties that you are under to me to do certain things. Each of these is a specified (claim-)right. I am identifying a duty another is under to me to do a certain act.

The rights (or is it ‘rights’?) in the Convention are not of this kind. They do not specify who must do what. Rather they are values or interests that provide reasons at a high level of generality that may ground such duties. They are not the end of the analysis, but the beginning.

Under the Convention, the “right” to private life may be interfered with where justified by, for example, the protection of morals. Finnis states that this is inept because a right properly so-called is a conclusion that one reaches only after taking into account all the reasons for and against its existence. As a matter of language this is unfair. We have for centuries used rights in the Convention’s sense of being grounding reasons at a high level of generality (“life, liberty, and the pursuit of happiness”). It is language which is at fault in using one word (“rights”) to capture more than one idea.

The Human Rights Act is peculiar, however, as UK legislation otherwise takes the form of setting out rules telling people in a specific way how they ought to behave. The rule of law problem is that Convention rights are not rules at all. Instead what we are presented with are reasons for and against the State being under (unspecified) duties. And that is it. The job of specifying what the rules are (the claim-rights in the lawyer’s sense) is left entirely open. Legislation does not ordinarily take this form.

Now, of course judges have always been in the business of creating law. If, to take a standard example, legislation states “it is an offence to use vehicles in public parks” a problem may arise as to what “vehicles” means. Are skateboards included, or not? Before a case comes before a judge there is no rule one way or the other. In common law jurisdictions with their doctrine of binding precedent, once it has been determined by a judge that vehicles does (or does not) include skateboards a new rule has been created. How judges can create positive law is constrained, but that they do so is not doubted.

These problems of uncertainty are ineradicable. The problem with the Convention is not just that it is peculiarly uncertain when compared to, say, UK legislation, but rather that it specifies no rules at all. It isn’t, in an important sense, law. It asks the judge to act as the legislator.

The problem is partially hidden within Article 8 by the fact that in determining what the rules are a judge may only allow the interest of privacy to be outweighed by the other countervailing reasons where ‘necessary’. As Finnis notes, however, that word cannot mean what it says.

“We can always imagine getting by without any restriction (and just submitting to accepting the loss and damage), or think up some restriction different from the one under challenge and apt for the same purpose. So no particular restrictive rule is itself necessary.“

What the Article actually requires, at least in a case of first impression, is a, more or less, free form weighing up of the reasons for and against any rule, and no further constraint on the judge in reaching a conclusion one way or the other.

The problem does not disappear in relation to other Articles with apparently more tightly circumscribed exceptions. So, Article 2 states

Article 2 – Right to Life

  1. Everyone’s right to life shall be protected by law. No one shall be deprived of his life intentionally save in the execution of a sentence of a court following his conviction of a crime for which this penalty is provided by law.
  2. Deprivation of life shall not be regarded as inflicted in contravention of this article when it results from the use of force which is no more than absolutely necessary (a) in defence of any person from unlawful violence; (b) in order to effect a lawful arrest or to prevent the escape of a person lawfully detained; (c) in action lawfully taken for the purpose of quelling a riot or insurrection.

Article 2 has not been interpreted as giving rise to a wholly negative duty on the state to refrain from actions that interfere with the relevant interest, but also a positive duty on the state to secure it. So, for example, the police are under a duty to protect us from being the victims of crime.

That positive duty cannot be (nearly) absolute in the way that Article 2 seems to contemplate (which is a good reason for thinking it was to misinterpret the words of the Convention to conclude that it gave rise to such positive duties). The State could, in theory, devote all its resources to the eradication of crime. Nobody would consider that sensible. As a result the duty to protect is not, nor could be, absolute, but is also subject to other (unspecified) countervailing considerations.

Response

There are two responses possible to the rule of law objection to the HRA, neither of which is free from difficulty.

First the party subject to the duties which the Convention gives rise to is the United Kingdom, and under the Human Rights Act it is the constituent parts of the UK State. It is a serious contravention of the rule of law if individuals are unable to determine in advance what their duties are. It is much less problematic if it is the State itself which is subject to this uncertainty.

Unfortunately, this is not the only way in which the Human Rights Act operates. It can have an indirect impact on the duties of individuals inter se (how it does so is a topic for another day). Further the State  can only act through real world agents. It is, to say the least, unacceptable if social workers, teachers, police officers and so on are unable to determine in advance what their duties are.

A second response is that we have now had nearly 50 years of European Court of Human Rights rulings setting out the duties of signatory states, and fifteen years of UK courts interpreting the relevant rights in a domestic context. If we go back in time far enough the entire common law was an unspecified open question. Judges were, many centuries ago, originally appointed by the King simply to do justice. The common law is entirely judge made, and is now at a high level of specificity. What is objectionable about going through this process again, and have we not now done enough to know what the rules are?

The answer to that is that we should only return to the conditions of 1154 if we have really good reasons for doing so. As to where we are, this can only be assessed by looking in detail at the rules set down and the many judgments delivered.

The Intentions of the Drafters

How legislation, treaties, contracts, wills or any other rights creating legal instrument should be interpreted is a large and difficult topic that I cannot tackle at length here. Professor Finnis argues that courts (in particular the ECtHR, ECJ, SCOTUS and the SCC) have arrogated to themselves an essentially legislative role by departing from the intentions of those responsible for the enactment of the words they are tasked with interpreting.

The meaning of legislation, or a treaty. is not controlled by the subjective intentions of those enacting or drafting it. We should not care what Maxwell Fyfe (or anyone else) privately thought the scope of any of the articles was for the purpose of interpreting the Convention today. The words mean what they mean, not what those drafting it intended.

Further, if the facts change the application of rules may change. A ‘right to freedom of speech’ could not have entailed access to a broadband internet connection in 1950, when it may in 2015.

However, the meaning of words is controlled by the context in which they are used. Words from a friend that “you have a nice car there”, would mean something quite different coming from a gang of boys when you park in a remote carpark.

The context of the European Convention is the world of 1950. If a ‘living tree’ interpretation is intended to allow a court to ignore the context of the words when used it is illegitimate. That does not mean we are slaves to history, but it does not mean we can ignore it either. Whether the ECtHR adopts an interpretation of the Convention that unfairly ignores the context in which it was adopted, as Finnis claims, is difficult to assess: partly because of the open-textured nature of the Convention itself.

Cases

There is currently a campaign to save the Human Rights Act. One form this takes are billboards featuring people who have successfully brought claims under it.

As any lawyer familiar with a case that has been reported in the national press would tell you, brief accounts very rarely give a fair feel for the issues raised, and billboard adverts suffer from this problem to an even higher degree. (The case summaries at the rightsinfo campaign are far superior to those used by ActfortheAct). Professor Finnis deals with some problematic cases from the ECtHR in detail, but of course he, like they, has been selective in his choices.

As I have tried to show in this and my previous post, it is not easy to successfully defend the Human Rights Act in the abstract. What needs to be shown is that in actual real world cases it leads to greater justice. This can only be done by carefully looking at examples. This is a task I shall seek to return to.

Should International Law be followed?

I had thought that my defence of the change to the ministerial code had been robustly expressed, until I read Professor Finnis on the same topic. Finnis makes many of the same points that I do in arguing that public international law, as such, is not binding on ministers. However, in two ways he seems to go further. I shall reject both. They are controversial and related.

The first claim is that treaty obligations cannot alter the legal duties of ministers, which are entirely to be found within our domestic legal order.

It is entirely true, as I also argued, that international treaty obligations are binding upon the states entering into them, and not upon individuals within those states however eminent. It is true that such agreed rules may be incorporated by Parliament into UK domestic law, as was most (in)famously done by the Human Rights Act and the European Convention on Human Rights. However, it is vitally important to remember that that must be done before such rules can be applicable within our domestic legal order. It is ironic that those attacking the change to the ministerial code are much the same people who are loudest in calling for the retention of the Human Rights Act. The latter was necessary because the ministerial code’s assumptions about international law were wrong.

However, Finnis claims that the duty on ministers to seek to ensure that the UK complies with international law is not a legal one at all, but

“morally grounded UK policy of complying with its treaty obligations, of encouraging other states to do likewise, and of supporting the due extension of customary international law and appropriate international institutions.”

I do not think that is correct. That is because I think ministers are under a legal duty to act in the best interests of the United Kingdom. When ministers exercise the various powers that are conferred upon them because of their status they are not free to exercise them in any way that they see fit. They must exercise this power for a proper purpose, and, at a high level of generality, in the best interests of the United Kingdom.

If ministers do fail to act properly they are not necessarily subject to sanctions. However, as we shall see, the absence of a sanction for breach is not proof of the non-existence of a duty. Rather, the more usual consequence of their failure to act as they ought is that their purported exercise of power is invalid.

In Australia it has become popular to describe this duty to act in the state’s best interests as being a public ‘fiduciary duty’, by analogy with the same idea as it occurs in private law with regard to, for example the duties of a director to his company. This seems correct, and to capture the idea that the duty of ministers to act in the interests of the United Kingdom are recognised by law.

Once we accept that ministers are under a legal duty to act in the United Kingdom’s interests, and that it is in the United Kingdom’s interests to (almost always) comply with international law, then the United Kingdom’s entering into treaty obligations does alter how ministers ought to behave, as a matter of law.

Second, Finnis argues that “international law remains, like it or not, a defective example of law.” In order to understand this claim it is important to know his views (which I find persuasive) as to the nature of law. Finnis is not here claiming that international law is not law at all, nor is he claiming that signatory states are not obliged to comply with it. Rather he is arguing that international law lacks some of the features that core or central examples of law possess. In particular, and most obviously, they (often or usually) lack either any sanction or enforcement measure where the rules are broken, and it is (common) for there to be no adequate adjudication system for determining their meaning. Just as lex iniusta non est lex, is properly understood as a claim that an unjust law lacks a feature that law ought to possess, but if it does not may still be capable of being described as ‘law’, a law without adequate enforcement mechanism is still law.

This is true, but it is an argument that points to the opposite conclusion from the one Finnis seeks to draw. 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. Again, ironically, the absence of a strong regime of sanctions for breach means that the United Kingdom has a stronger, not weaker, reason for compliance.

We might by analogy look at the law of gaming as it was in the United Kingdom (and still is elsewhere). Gaming contracts were unenforceable. This meant that in principle if a punter won an enormous long odds bet, there was no legal mechanism by which the bookmaker could be induced to pay out.

Why then did bookmakers routinely cough up? The answer is that it was in their interests to do so. Any bookmaker who failed to pay out would immediately have his commercial reputation destroyed, ensuring that no business was ever done with him again. This meant that bookmakers were extremely careful to (over) comply with their obligations. If we compare their trade with that of insurance companies, who are essentially in the same business but whose contracts have always been legally enforceable, we find that the absence of a legal sanction for breach gave rise to higher compliance.

If the system of international law is not to be undermined, the United Kingdom must comply with its obligations. We have to pay out, even on losing deals.

What Finnis is most obviously right about is the well-meaning intellectual sloppiness of the critics of change. He delivers similar blows in relation to human rights, and I shall seek to return to the arguments that those who wish to argue against him should be making, but largely are not.