Human Rights and Rape

Admiral Byng was a royal navy officer who failed to relieve a besieged British garrison in Minorca during the Seven Years War. As a result, he was sentenced to death and shot by firing squad. In Candide, Voltaire satirised the justification for the execution of the almost certainly blameless Byng on the basis that “it is good to kill an admiral from time to time, pour encourager les autres.”

The point of this story is that it is immoral for the state to use individuals as a means to an end, even if that end, the defeat of the French, is a good one. If human rights mean anything at all, it is that. This is so even if the good end one seeks is itself more justice in the world.

Legislators, from wherever on the political spectrum, tend to have a different perspective from judges in this regard. A legislator tends to think in terms of the common good. They are concerned with how things will turn out best overall. A judge, by contrast, tends to be concerned to ensure justice for the individual(s) before her. As Admiral Byng demonstrated, these perspectives may be in tension one with another. The frustrations in government of both left and right with the constraints of the Human Rights Act comes from the same source.

Ched Evans

The acquittal of the footballer Ched Evans at his retrial for rape has proven controversial, with some claiming that it sets a dangerous new precedent in cases of sexual violence. The case is  insightfully discussed by legal bloggers here and here, and those unfamiliar with the circumstances of the case are advised to start there.

The result has been a letter to the Attorney General by a number of Labour MPs seeking to change the law. Here I wish to explain why their proposals are misguided.

Sub-section 41(1) of the Youth Justice and Criminal Evidence Act 1999 provides.

If at a trial a person is charged with a sexual offence, then, except with the leave of the court—

(a)no evidence may be adduced, and

(b)no question may be asked in cross-examination,

by or on behalf of any accused at the trial, about any sexual behaviour of the complainant.

The purpose of this section was to restrict the evidence going to the jury because of the persistent myths that “unchaste women are more likely to consent to intercourse, and are less worthy of belief.”

 The section does allow of some narrow exceptions, importantly for our purposes sub-section 41(3)(c), where

it is an issue of consent and the sexual behaviour of the complainant to which the evidence or question relates is alleged to have been, in any respect, so similar

(i)to any sexual behaviour of the complainant which (according to evidence adduced or to be adduced by or on behalf of the accused) took place as part of the event which is the subject matter of the charge against the accused, or

(ii)to any other sexual behaviour of the complainant which (according to such evidence) took place at or about the same time as that event,

that the similarity cannot reasonably be explained as a coincidence.

But such evidence is not to be adduced to impugn the credibility of the complainant as a witness.

On its face this provision does not permit the admission of evidence of a prolonged period of cohabitation between complainant and accused for years beforehand (“at or about the same time”). If the natural reading of the words were given effect, they would have the effect of violating the defendant’s right to a fair trial under the Human Rights Act. When the issue of this conflict came before the Judicial Committee of the House of Lords (the forerunner to our current Supreme Court) Lord Steyn stated that whilst section 41

pursued desirable goals, the methods adopted amounted to legislative overkill.

The court took the “bold” step of reading section 41 as compatible with the Human Rights Convention by inserting an ” implied provision that evidence or questioning which is required to ensure a fair trial under article 6 of the Convention should not be treated as inadmissible.” (As an aside, I do not think this was a “reading” of the legislation that was permissible within the possible range of meanings of the words in the Act. What the judges have in fact done is re-written the Youth Justice and Criminal Evidence Act and made it subordinate to the earlier Human Rights Act. The court should instead have made a declaration of incompatibility.)

In the Ched Evans case, two witnesses came forward each of whom said they had had sex with the complainant in circumstances similar to those Evans had described to the police when interviewed in 2011.  This, the defence said, was relevant to whether she in fact consented on this occasion. The difficulty arises in relation to the meaning of “similar behaviour”. All that the section requires is that it “go beyond the realm of what could reasonably be explained as a coincidence.” It does not need to be unusual, rare or bizarre. It is a question of judgement whether the evidence on this occasion was sufficient, and the circumstances of its being obtained sufficiently trustworthy, so as to be of probative value. That is a matter for the finder of fact: the jury.

Intentions of Parliament

In their letter proposing a change to the law, the MPs argue that the Court of Appeal have failed to give effect to the intentions of Parliament. It is stated that only evidence that was “unusual or bizarre” was ever intended to be admissible. In support of this claim a statement by Lord Williams of Mostyn, then the Attorney General, in Parliament in promoting the Bill is relied upon, where he said the section

“might include something….so unusual that the jury should know that the complainant had consented in strikingly similar situations in the past.”

Is what Lord Williams said in 1999 relevant, as is claimed?

Nowadays, we do sometimes allow statements of minsters to be admitted for purposes of construing a statute under the rule in Pepper v Hart. It is very important to understand the limited role such evidence has. What we are concerned to discover is the meaning of the words of the Act. The subjective intentions of those Parliamentarians enacting the Act are neither here nor there. We don’t care what Lord Williams himself thought the words meant. We are concerned with what they mean. The meaning of words is however always determined by its context. “Warning! Duck!” may mean that you need to lower your head on a ship, or alert to avian aggression in a bird sanctuary: it depends on the context. In legislation, part of that context includes the words of ministers in promoting a Bill.

Is Lord Williams’ statement enough to add in a “bizarre or unusual” limitation to the words where they do not appear, as is claimed? Clearly not. He merely gave this as a non-exhaustive example of what the words covered, and nothing he said sheds such startling new light on the context of the words so as to alter their natural meaning.

Should the law be changed?

Should we change the law so as to only permit “bizarre or unusual” evidence? If this were done, the law of evidence would clearly violate the Human Rights Act. After all, long term sexual relationships are (now) admissible under s 41(1)(c)(ii) and these are hardly “bizarre or unusual.”

Over several decades successive governments have changed our rules of evidence, for example relating to hearsay or bad character, in a way that favours the prosecution. We now tend to favour allowing the jury to make up their own minds. Here the argument goes the other way.

Parliamentarians tend to be interested in justice in the aggregate, in particular securing more convictions of criminals. So, we are told that “Parliament must send a clear message to victims of sexual violence that no matter their history or background the law in the UK (sic) is there to protect them from harm.”(The criminal law is a devolved matter.) No doubt it should, but it should not do so in a way that would violate the rights of the accused to a fair trial. Politics of both left and right is still unfortunately dominated by a Benthamite utilitarian calculus, with the dispute being about how the ledger on either side is to be calculated. The individual being used disappears from view.

We can think of hypotheticals where we may be prepared to sacrifice the rights of an individual for a much larger greater good (the death of an innocent to save the world). However the case for doing so must be supported by clear and overwhelming evidence, which here is missing. We should be sceptical of those who suggest change only after the result of a trial they dislike, and not after the decision several months earlier of the Court of Appeal as to the admissibility of evidence. Scaremongering as to the ease of use of this evidence is, at a minimum, counter-productive.

Ched Evans is not a very appealing character, but human rights are not just for the virtuous.

Advertisements

Why “Turing’s Law” is Wrong

The government has announced that thousands of gay and bisexual men who were convicted under laws criminalising homosexual acts are to be pardoned.

 

This is wrong, indeed offensive.

 

To pardon  someone is to forgive a wrong. So, a monarch may pardon a criminal, relieving them from sanction. If we “forgive those who trespass against us” we relieve another from the wrong they have committed. There may be good reasons for doing so, but we can only pardon if there is a wrong to forgive. A necessary implication of “I pardon you” is to implicitly say that a wrong has been suffered.

 

The laws in place in the United Kingdom criminalising homosexual acts were wrong and barbaric. Such consensual acts between people of the same sex wronged nobody. There were no good public policy reasons for criminalising their actions. Those who were convicted were themselves the victims of a wrong. There is nothing for the state to forgive or pardon, indeed to say that there is is to (literally) add insult to injury.

 

How then should the government respond to this past injustice?

 

What it should do is retrospectively decriminalise these offences.  In England, consensual homosexual acts between men were only decriminalised in 1967 (shockingly, it remained criminal in Scotland until 1980, and in Northern Ireland until 1982). This legislation was prospective only. Such behaviour after the legislation was passed ceased to constitute a crime. What the legislature should now do is retrospectively repeal what should never have been criminal at all.
Why does the government not wish to do that? Because if it did so, those who were wrongfully treated would have a claim from the date of repeal for their wrongful treatment (time would only start to run against them from the moment of repeal, as that would be the first time they could sue for their wrongful imprisonment and other forms of mistreatment). If the view is taken that compensation should not be payable, because too much time has now passed, that should be incorporated into any Act retrospectively repealing these offences.

 

But a pardon, by the UK government? The only people who have the capacity to forgive are the victims of this injustice: the men wrongfully convicted.

 

 

 

Reversing Article 50

Once the UK gives notice under Article 50 to withdraw from the European Union, can it unilaterally withdraw such notice or is it locked into inevitable Brexit? Article 50 (1-3) states

1. Any Member State may decide to withdraw from the Union in accordance with its own constitutional requirements.

2. A Member State which decides to withdraw shall notify the European Council of its intention. In the light of the guidelines provided by the European Council, the Union shall negotiate and conclude an agreement with that State, setting out the arrangements for its withdrawal, taking account of the framework for its future relationship with the Union. That agreement shall be negotiated in accordance with Article 218(3) of the Treaty on the Functioning of the European Union. It shall be concluded on behalf of the Union by the Council, acting by a qualified majority, after obtaining the consent of the European Parliament.

3. The Treaties shall cease to apply to the State in question from the date of entry into force of the withdrawal agreement or, failing that, two years after the notification referred to in paragraph 2, unless the European Council, in agreement with the Member State concerned, unanimously decides to extend this period.

A natural interpretation of those words indicates that once notification is given, Brexit is inevitable and the Member State cannot unilaterally withdraw its notification. Four reasons may be given. First, and most importantly. the words of the provision make no mention of the Member State having any such power of withdrawal. Second, the provision does allow for a delay in execution, but requires unanimous agreement of Member States to do so. This may be read as excluding both  the stopping of execution, and a  Member State having the unilateral power of choice. Third, as the institutions of the European Union, and other Member States, will have begun the process of adjusting to such withdrawal, such a unilateral power in the Member State may be thought unfair. The clear tenor of Article 50 is to give the balance of power to the Union, and not the withdrawing Member State. Fourth, the alternative interpretation leads to ridiculous results. Could a member state give notification, withdraw it one day before the expiry of the two year limit, and re-start the entire process on the following day?

The final word on the meaning of EU law is that of the Court of Justice of the European Union. It has been known to take what our civilian friends call a “teleological” approach to interpretation. Politely this means that it will take a purposive approach and may sometimes depart from the plain meaning of words. Impolitely, it sometimes cheats. If it were faced with the question of maintaining the integrity of a the Union if a repentant UK somewhat improbably sought to withdraw its notification, it is not impossible that it could read the words in a way to permit it to do so. Donald Tusk has suggested that that is indeed the position.

(For different views, that don’t seem to me to focus on the words of Article 50, see Auriel Sari and Professor Paul Craig  (page 35 of latter’s paper).)

One point to note about the above is the dissonance between the law and predicting what the CJEU will do. My firm view, despite the eminence of those I disagree with, is that the words of Article 50, and hence the law itself, are quite clear. The uncertainty concerns whether the CJEU will depart from the meaning of the words. Oliver Wendell Holmes In The Path of Law has been read, perhaps unfairly, as advocating the prediction theory of law: law is just our prediction of what the judges will decide. I, by contrast, consider the law, and our predictions as to what judges will do, to be different questions. The law is clear, what the judges may in the end do is not.

Do we currently care? 

Jolyon Maugham QC, who is fast becoming our most important informed commentator on tax matters, is also a passionate Remainer (as, somewhat to my surprise, I have discovered that I am). Currently before the courts is litigation concerning whether the government can give notice of withdrawal under Article 50 without Parliamentary approval. He argues that it is possible that this could turn upon the reversibility of such notification. This is important because if it does, a reference to the Court of Justice will be necessary to authoritatively determine this question.

The Court of Justice has many qualities. Swiftness is not among them. If such a referral were made it could take many months to obtain a ruling, delaying the giving of notice and postponing Brexit. Who knows whether, as the main opposition to Brexit comes from the irresistible economic gravity of the market, this could provide room for a change of course?

For good or ill, this seems very unlikely. It will be recalled that the claimant’s argument is that the government cannot give notice without Parliament as to do so would be to overturn various statutory rights. Can the correctness of this argument turn upon the reversibility of notice?

If, as the government claims, no statutory rights are overturned by the Article 50 process, then the issue of reversibility is irrelevant. As nothing is affected, it doesn’t matter. It is for this reason, I would suggest, that the government has not address the issue of reversibility. It cannot arise as a question if they are right.

If statutory rights would be overturned by Brexit, does this mean that the reversibility of notice becomes a live issue? As any notice would, inevitably, be given by the government, its ability to reverse this decision would still leave the choice whether to overturn such statutory rights in its power. If the claimant’s argument is correct then we would still require statutory authorisation even if the government could reverse itself. Reversibility cannot alter the answer.

So, reversibility cannot be determinative, and so no referral to the Court of Justice is required.

More Brexit in the Courts

 

“Just so I follow, are you saying then, that if on 2 January 1973 the government had changed its mind about the desirability of the EC treaties and their full scope in that time and sought to go and negotiate with the other member states and said well, let’s remove the right of establishment from the treaties, and they might say yes, and that would modify what the EC obligations were; that the government would in fact by reason of section 2 of the European communities act have been disbarred from seeking to exercise the prerogative in that way, ie by approaching other states to see if this he would agree to a change in the treatise because it would be such a fundamental change within domestic law”

-Sales LJ to Lord Pannick in argument

“I am troubled by this, Lord Pannick”

-Thomas LCJ to Lord Pannick in argument

It is sometimes foolish to read too much into the reaction of the bench to the submissions of counsel. Many lawyers will have stories of cases where they are convinced that the judge is for (or against) them, only to have the decision handed down give the opposite result.

However, the above exchange (at pp 108-) towards the end of Lord Pannick’s submissions in the Article 50 Brexit litigation on my reading indicates that the government is going to win.

As will be recalled, section 2 of the European Communites Act incorporates into UK domestic law all “rights, powers, liabilities, obligations and restrictions from time to time created or arising by or under the Treaties”. The point of Sales LJ’s question is what happens if some of those rights are eliminated under EU law by agreement between member states. Does that require further legislation to be given effect?

The answer is clearly no.

So what if those rights etc were reduced to almost nothing (leaving, say, just a directive on banana curvature in force)?

Again the answer is no fresh legislation is required.

And is it any different if those rights etc are reduced to nothing, as would be the result of the government invoking article 50, which is itself part of EU law?

No.

Derogating from the Convention

 

 

“Much of the litigation we face comes from the extension of the European Convention on Human Rights to the battlefield. This is damaging our troops, undermining military operations, and costing taxpayers’ millions. So I can announce today that in future conflicts we intend to derogate from the Convention. That would protect our Armed Forces from many of the industrial scale claims we have seen post Iraq and Afghanistan. “

 

Michael Fallon, Secretary of State for Defence, 4 October 2016

 

This is nonsense.

 

To understand why, it is necessary to understand why it is now the case that claims can be brought under the Human Rights Act for actions taken by British troops serving abroad. At one time it was thought that the jurisdictional scope of the European Convention was limited to the territorial jurisdiction of signatory states. In the ECtHR’s decision in Al-Skeini this was decisively abandoned and the test became “wherever a state through its agents exercised control and authority over an individual.”

 

Now, my view is that Al-Skeini  was plainly wrongly decided, both according to the text of the Convention and the ECtHR’s earlier jurisprudence. I can, therefore, well understand the government’s frustration at losing. The United Kingdom Supreme Court followed Al-Skeini in determining the scope of the Human Rights Act in Smith v MoD

 

Can this result be bypassed by making a derogation from the Convention? Article 15 provides

 

 

1. In time of war or other public emergency threatening the life of the nation any High Contracting Party may take measures derogating from its obligations under this Convention to the extent strictly required by the exigencies of the situation, provided that such measures are not inconsistent with its other obligations under international law.

 

This provision does not permit a derogation from all the obligations under Convention whenever a signatory state is at war. If it did that would render the result in Al-Skeini redundant. Where the court finds, as it did with respect to the United Kingdom after the September 11th attacks, that there are grounds for making a derogation, this does not provide the state with a blank cheque. Derogating measures must be such as are strictly required by the exigences of the situation. The kinds of claims brought against British troops involve actions that, on no view, are necessary for the conduct of a war.

 

So, whilst it is possible for the United Kingdom under its own domestic law to restrict the application of the Human Rights Act (but in a way that will place us in breach of the Convention as we will no longer be providing an effective remedy for Convention violations as required by article 13), and certainly possible to extend combat immunity as it applies to claims at common law, we cannot simply create an immunity from suit for our armed forces that is compliant with the Convention.