Tuesday 19 January 2016

Miranda and Legality

David Miranda (Credit: The Guardian)
Many of you will have been paying close attention to a controversial use of powers under the Terrorism Act 2000, which followed-on from the Snowdon leaks of classified NSA files. A man called David Miranda, partner of journalist Glenn Greenwald, had obtained on a USB stick encrypted files in connection with the leak, and was detained at Heathrow airport under what are known as the "Schedule 7 powers" of the Terrorism Act 2000. The factual background of the case is quite neatly summarised by Lord Justice Laws in the Divisional Court, and repeated by the Court of Appeal, whose judgment came out today.

I do not intend in this blogpost to go through the particulars of all of the five grounds of appeal. Others have done this better than I can do justice to. However, I do wish to comment on what has been described as an "inconsistency" in Lord Dyson's judgment handed down today, raised by Carl Gardner in his piece here.

The Judgment

The main finding against the government was based on the conclusion that, for want of adequate safeguards to prevent arbitrary use, Schedule 7's detention and search powers fail the test that must be satisfied to comply with the European Convention on Human Rights. As such a "declarator of incompatibility" under s4 of the Human Rights Act was issued. In particular, s4(2) provides that:
"If the court is satisfied that the provision is incompatible with a Convention right, it may make a declaration of that incompatibility."
The effect of a declarator is not to invalidate, discontinue or limit the enforcement of primary legislation (s4(6)(a)). On the contrary, it has full force and effect.

It is not unlawful in a specific instance for a public body to act incompatibly with the ECHR where they are "acting so as to give effect to or enforce" "one or more provisions of, or made under, primary legislation which cannot be read or given effect in a way which is compatible with the Convention rights". This is very clearly set-out in s6(2)(b) of the Human Rights Act.

This means that a piece of legislation can violate the ECHR, but it is not competent for a British court to make a finding of illegality in respect of a public body acting incompatibly, where that illegality results, and only results, from the incompatibility of the primary legislation that it is enforcing. Obviously if, for example, secondary legislation was incompatible with the Convention for additional reasons and/or a public body supplementarily acted in a manner which was incompatible with the Convention, then a British court could strike down that secondary legislation and find the actions of the public body to be unlawful. The appropriate remedy in that instance would not have to be a declarator of incompatibility.

Acting compatibly with the ECHR

There are three criteria that must be satisfied if an interference with a qualified Convention right is to be considered to be lawful, and therefore escape any risk of this declarator being made. An interference must be (a) in pursuit of a legitimate aim (b) prescribed by law and (c) necessary in a democratic society.

Prescribed by Law

The importance for our purposes is the relationship between (b) prescribed by law and (c) necessary in a democratic society. The former is a standard that requires, as Carl puts it:
"any law such as Schedule 7 must be accessible and its operation sufficiently foreseeable so that people subject to it can regulate their conduct; and it must contain sufficient safeguards to avoid its arbitrary exercise"
Necessary in a Democratic Society

The latter constitutes a proportionality test, the particulars of which have been expressed on several occasions by the UK Supreme Court. One such exposition is given in the Bank Mellat case, where Lord Reed sets out a 4-stage test:
"It is necessary to determine
(1) whether the objective of the measure is sufficiently important to justify the limitation of a protected right,
(2) whether the measure is rationally connected to the objective,
(3) whether a less intrusive measure could have been used without unacceptably compromising the achievement of the objective, and
(4) whether, balancing the severity of the measure’s effects on the rights of the persons to whom it applies against the importance of the objective, to the extent that the measure will contribute to its achievement, the former outweighs the latter…. In essence, the question at step four is whether the impact of the rights infringement is disproportionate to the likely benefit of the impugned measure."

Third Ground of Appeal

The third ground of appeal in the Miranda case maintained that [59]:
"the use of Schedule 7 power against Mr Miranda was an unjustified and disproportionate interference with his right to freedom of expression guaranteed by article 10 of the Convention"
This ground is solely concerned with whether the interference with David Miranda's Article 10 right was proportionate. It is not concerned with whether it was in the pursuit of a legitimate aim or prescribed by law. It is, moreover, concerned with the particular. What the court was being asked to determine there was whether the test set-out by Lord Reed in Bank Mellat had been satisfied. Lord Dyson concludes [84]:
"The compelling national security interests clearly outweighed Mr Miranda’s article 10 rights on the facts of this case. In reaching this conclusion, I also bear in mind the considerable deference that the court should accord to a decision to invoke the Schedule 7 power in a case of this kind. It follows that, subject to the point raised by the fourth ground of appeal, the decision to exercise the power was proportionate on the facts of this case"
Clearly these remarks are confined to the question of proportionality, and not to whether it is prescribed by law.

Fifth Ground of Appeal

The fifth ground of appeal, which was ultimately successful, maintained that Schedule 7 of the Terrorism Act 2000 was "incompatible" with Art 10 of the ECHR. The conclusion reached was that Schedule 7 [119]:
"Is not subject to adequate safeguards against its arbitrary exercise"
And is therefore not prescribed by law.

Carl's Argument

Carl argues that it is inconsistent, on the one hand, for Lord Dyson to make his finding in respect of the fifth ground of appeal, but not to have found that the third ground of appeal also succeeded. As he puts it (emphasis added):
"If Lord Dyson is right that judicial review is inadequate, and cannot sufficiently protect journalistic material, then it cannot protect it in this case; Lord Dyson’s own ruling that questioning Miranda was lawful must be inadequate to satisfy human rights law. Without prior independent authorisation Miranda’s questioning can’t have been “prescribed by law”, and so must have breached the article 10 right to free expression. Lord Dyson could not possibly have “cured” that breach today."
This is a perfectly correct statement of law from the perspective of the Strasbourg court. They couldn't care less if the failure to be prescribed by law arises from the primary legislation itself or from the particular nature of the action purporting to be taken under it. The applicant is suing the state as a whole, not a particular branch of it. If this goes to Strasbourg, and they agree with Lord Dyson that Schedule 7 is incompatible with Article 10(2) for want of being legally prescribed, they will make a declaration to that effect and award damages in just satisfaction.

The Problem

But for a UK court, the distinction is extremely important. Indeed, it is precisely the reason that the third ground of appeal was framed as it was, in terms of proportionality alone, rather than in terms of compatibility with the Convention. The third ground of appeal concerned, in reality, not just whether the interference was proportionate, but also, by inference, whether the public body exercising the Schedule 7 powers could avail itself of s6(2) of the Human Rights Act to escape illegality completely.

This is important from Miranda's perspective. The declarator does not entitle him to any additional form of redress. A finding that the specific application of the law, but not the law itself, was disproportionate, would likely entitle him to some sort of pecuniary redress. Paradoxically, were he to prove that Schedule 7 was in and of itself disproportionate, he would be enabling the government to grant a broader immunity from compliance with the Human Rights Act to those exercising powers under Schedule 7, as it would, much like the prescribed by law failing, fall under the terms of a declarator of incompatibility.

If the specific instance of the exercise of this power was disproportionate (in the Bank Mellat sense), it would not have mattered whether or not the provision was prescribed by law. The public body would in any case have acted illegally in the specific instance. It would have been no "defence" to say that they were "acting so as to give effect to or enforce" "one or more provisions of, or made under, primary legislation which cannot be read or given effect in a way which is compatible with the Convention rights". The primary legislation would not have authorised them to act in that way. With respect to proportionality, the provisions can be, and in Miranda's case were, exercised in a manner which is compatible with the Convention rights. Any body which exercises that power in a way that is disproportionate is going beyond what Parliament intended.

Carl's Response

In response to this, I'm sure Carl would still respond with a combination of these sentiments:
"Legality is a logically prior requirement: only if a power is “prescribed by law” can its use be proportionate and so rights-compliant."
"The concept under the Human Rights Act 1998 that legislation may in itself be incompatible with a Convention right is a radical one: it means that the legislation necessarily and systematically causes breaches of human rights. Conversely, if not every use of a piece of legislation breaches human rights, then it is compatible with those rights."
"The Human Rights Act forces judges to make a choice. Either a piece of legislation is in principle compatible with human rights, in which case its use in particular cases may still breach rights (and so be unlawful); or it’s incompatible with those rights in principle, in which case its use in every case will breach rights (and, in an apparent paradox, will therefore be lawful). You can’t have it both ways. You can’t hold the questioning of David Miranda compatible with human rights if you also think the legislation it was done under is not."
The problem here is that the first of these three quotes is not actually correct. In many instances, legality may operate in that way, and there may well be some overlap on the facts between the threshold to be "prescribed by law" and the arguments made in relation to parts 3 and 4 of the Bank Mellat proportionality test. But it does not follow that all exercises of a power are disproportionate if none of them are prescribed by law. Some of them may be proportionate. This may not save you in front of the Strasbourg court, but it does save you under a British court enforcing the Human Rights Act, provided that you can show s6(2)(b) applies.

As such, even if a British court concludes that "the [primary] legislation necessarily and systemically causes breaches of human rights" that does not mean that it can find a wholly derivative and otherwise lawful action to be illegal. Where the second and third quote breaks down here is that Lord Dyson did not say anything that contradicts Carl's "converse" inference. The specific use of Schedule 7 against Miranda was not prescribed by law, but British courts are compelled not to do anything about it where it is the direct, exclusive and necessary consequence of enforcing an incompatible primary statute that was incompatible by, and only by, reason of not prescribing the interference by law.


Whether you like it or not, this is how the Human Rights Act was formulated, and this is specifically how declarators of incompatibility are supposed to work. I suspect Carl and I disagree quite drastically as to the trajectory our domestic human rights framework should follow were we to move away from this muddy compromise.


  1. I'm not persuaded, I'm afraid, Graeme. Let's see if I can explain why not.

    I find myself agreeing, in isolation, with quite a few of the things you say in your post. But I also think you may be misunderstanding my approach. You say

    Carl argues that it is inconsistent, on the one hand, for Lord Dyson to make his finding in respect of the fifth ground of appeal, but not to have found that the third ground of appeal also succeeded.

    But I'm not arguing that the 3rd ground of appeal should have succeeded. It couldn't, if the 5th ground succeeded, since the result of allowing the appeal on ground 5 is that the stop had to be lawful under the provision you've identified: section 6(2)(b) of the Human Rights Act.

    My argument is that Lord Dyson's *reasons* for dismissing the 3rd ground of appeal are invalidated by his allowing ground 5. He could have dismissed ground 3 on the basis of section 6(2)(b). Had he done that, today’s appeal judgment would have made sense. I say so in my post.

    But once you've found that the stop power is not prescribed by law, then it's simply impossible to hold that any use of it is prescribed by law. Yet Dyson does exactly this, since he says not merely that Laws LJ was not in error on proportionality — but that the stop *was proportionate and so rights-compliant*. The lawfulness finding is not based on section 6(2)(b) and incompatibility, but on straightforward human rights compliance by the police. Which there cannot be, if the interference is not "prescribed by law". That, surely, is an elementary point arising from the text of article 10(2). It doesn't matter whether a UK court or Strasbourg is deciding.

    You say (and I agree) that the absence of "prescription by law" does not make an interference with a right disproportionate. But again that wasn't quite my point (and I chose my words in my post carefully in an attempt to be clear on this). Sure, proportionality is separate from "prescription by law", and Dyson would have been entitled simply to say that Laws LJ did not err, looking at proportionality in isolation. But he went further than that: he found the stop *proportionate and so rights-compliant*.

    My point is that only if a power is “prescribed by law” however can its use be also *found proportionate and so rights-compliant*. Finding an interference proportionate while not "prescribed by law" is purely academic, since it cannot possibly be compliant with rights. Yet Dyson found the stop proportionate *and so rights-compliant*. The reasoning, therefore, by which Dyson reached his conclusion that the stop was lawful cannot be sound (if his reasoning on ground 5 is sound).

    I think the only way one can rationalise the conflicting approaches to prescription by law here is to make a purely formal procedural argument that, since ground 3 was formulated only in terms of proportionality, Dyson had to proceed on the basis that the stop satisfied every other requirement for rights-compliance, and could not properly make a ruling on "prescription by law". I find that unconvincing for two reasons.

    One: Dyson could just as easily have dished ground 3 on the basis of section 6(2)(b), as I think you (and I) have argued. He wasn't actually forced into his ultimate analysis.

    Two: I just don't think the courts consider themselves quite so procedurally straitjacketed that they're compelled to base parts of a ruling on legal assumptions they find in other parts of the same ruling (as Dyson does in his approach to ground 5) to be wrong.

  2. The issue here is that "proportionate and so rights-compliant" is your construction and not Lord Dyson's. From Twitter and your main post, I'm gathering this construction is based on what he says at para 84 and para 118.

    At para 84 he says, and only says, that it is "proportionate". He does not say that it is "rights compliant".

    At para 118 he says the exercise of the power is "lawful". At UK law, a specific, proportionate, exercise of a primary statute is always lawful, regardless of whether or not it is prescribed by law. At Convention law level, it isn't lawful, it's incompatible if it is not prescribed by law.

    This is what the declarator of incompatibility is all about. It renders lawful derivative legislation and acts which would otherwise and only otherwise, be unlawful by reason of incompatibility with the Convention, that incompatibility being caused and only caused by the primary legislation from which they derive their power to enforce.

    1. I agree that "proportionate and so-rights compliant" is my phrase, not Lord Dyson's.

      But I chose it as a way of summarising Lord Dyson's legal reasoning on ground 3, and I think it's an accurate summary. He addresses that ground, assesses the stop as proportionate and for that reason (see para. 118) concludes it was lawful. I don't think it's possible to dispute that my formulation is a good summary of his reasoning on ground 3.

      I think you're arguing that Dyson's conclusion on the lawfulness of the stop is unrelated to his conclusion on its proportionality, and that he concludes the stop was lawful because of section 6(2)(b) of the HRA. That seems to me unsustainable, reading the judgment. If you were right, you'd need to explain why Dyson went through the exercise of assessing proportionality at all.

      And yes, I think you have an answer to that: you were arguing last night on Twitter that his discussion of proportionality was actually about a kind of "supplemtary proportionality": that in spite of section 6(2)(b), acts done under incompatible legislation can still be unlawful if disproportionate, and this is what Dyson was assessing.

      That approaches faces two problems, I think.

      First, it'd mean a radical new understanding of the HRA, in which it does not (as I think most lawyers agree) protect Parliament's ability to legislate contrary to rights. On your analysis, even if Parliament does knowingly legislate in breach of rights, its application of those laws can still be challenged on grounds of proportionality.

      That would mean the HRA failed to deliver what politicans though, and most lawyers think, it did deliver. It would also mean (I think) there can now be a new round of litigation on prison votes, in which prisoners can seek damages arguing that the UK's denial of the vote is disproportionate in their individual case, and that the incompatibility of the legislation itself is no defence. To be fair, I think you rejected this argument by saying this kind of thing is only possible where the legislation is incompatible because *not prescribed by law*, and not where it is incompatible *because disproportionate*. I don't understand why that distinction should be made, though.

      The second problem is that, had Lord Dyson been doing what you claim, why would he make that so opaque? It would have been far clearer forst to deal with the incompatibility of the legislation, explain that the stop was therefore lawful subject to your "supplementary" issue of proportionality, and then deal with that issue.

      These are complex issues, of course, and both our arguments are complex. But I think your is more complex than mine; less orthodox in terms of the HRA; and harder to reconcile with what Lord Dyson wrote.

    2. Carl is overlooking a very important part of paragraph 33 in Beghal. The final sentence where Lord Hughes says "It [legality] does not, however, subsume the issue of proportionality, whether the issue is the proportionality of the measure as a whole or the proportionality of its application in a particular case."

    3. Parliament will always be able to legislate contrary to rights. The issue is always whether the statute can be read down on its application. If the statute is explicit in its contravening ECHR, there will be no way to read it down. If a court manages to find a way of reading it down then Parliament wasn't clear enough.

    4. Matt (Flay),

      I'm, not overlooking the last part of what Lord Hughes said at all. I agree with it.

      I absolutely agree that the "prescription by law" issue does not subsume the "proportionality" issue. If an action or statute is "prescribed by law" then it may still be disproportionate, and so breach rights. An example of such a statute, according to Strasbourg, is UK legislation barring prisoners from voting.

      Nor does failure to pass the "prescription by law" condition mean an action or statute is proportionate.

      However, if an action or a statute is not "prescriped by law", then it cannot comply with article 10. You can if you like enter into the purely academic exercise of assessing whether it pursues a legitimate aim and is proportionate to that aim. But even if it is, it still breaches rights, as it was not "prescribed by law". This is what Lord Hughes means when he says it's a "prior test".

    5. On twitter I asked:

      Me: "Okay, so let me get this straight. Judicial review is not sufficient protection against arbitrary exercise..."

      Carl: "So Lord Dyson ruled, yes."

      Me: "...of power. And for that reason alone, a PA is now free to ignore ECHR in all respects when exercising the power"

      You: "Yes, because section 6(2)(b) displaces section 6(1) whenever it gives effect to the power."

      If you think that these two outcomes are in any way rationally connected, then there is no point in any further discussion. They are so far from rationality that it requires a worm hole of the mind to get there. I would bet everything I have that you are wrong. A judge decides that there is insufficient protection against arbitrary power and in so doing he ensures that the power can be exercised as arbitrary as domestic law makes possible. I'd pay good money to see you try to argue that in the Royal Courts of Justice.

    6. Graeme,

      Carl seems to be referring mainly to the proportionality in principle of a law that sets down a power to interfere with a convention right. That a law could prescriptively grant a PA the power to behave in a way that would otherwise be seen as disproportionate to the aim of the legislation. Is this an important distinction? Is it even possible? Would proportionality in this sense not actually become a part of the reasoning against legality? A law that blesses a disproportionate exercise of power could be seen as arbitrary. Although there is no excuse for ignorance of the law, the ordinary person cannot be expected to be aware of every single law and the manner in which it is designed to be applied. If a law goes against common sense and values, then surely most people would be in danger of falling afoul of it and it cannot be prescribed.

  3. Part 1 (4096 char limit):
    I am not a lawyer, but I am someone interested in language and philosophy.

    I find it odd that this Lord Reed with pt.4 had used of this word 'achievement', because it seems to me to be wildly inappropriate and intellectually fraudulent the way this word is used.

    (4) whether, balancing the severity of the measure’s effects on the rights of the persons to whom it applies against the importance of the objective, to the extent that the measure will contribute to its achievement, the former outweighs the latter…. In essence, the question at step four is whether the impact of the rights infringement is disproportionate to the likely benefit of the impugned measure."

    There is with pt. 4 obviously a wish to make a comparison between rights of persons and the "will" of the state so to speak, and then maybe do an adjustment or make a change, but only as long as a measure is still effective so to speak, for the state. Therein lies is the fraud. What is argued for here is not the right of the state, but rather the will or perhaps 'the power' or even 'the effect' or even 'the influence' of the state. The notion of "achievement" here is in a word the embodiment of the will of the state, but as a metaphor only, because otherwise 'an achievement' would come to mean the same thing as 'an objective' (specific goals, that can be achieved specifically), and here achievements as specific goals is obviously not a part of the argument, nor is it a point having been made.

    Unless the meaning of the word 'achievement' is really meant to be self reflexive, changing with a change of context, somehow referring back to itself as some conceptual metaphor (in itself a terrible ting, the same way you probably wouldn't accept circular logic), then I'd claim that the following sentence doesn't make any sense: "(...)to the extent that the measure will contribute to its achievement"(...)".

    If the sentence is meant for the phrase "to the extent that the measure will contribute to its achievement" to be self-reflexive, the argument is imo intellectually fraudulent, not only because the phrase "an achievement" will probably be given meaning retroactively (an 'achievement' becoming a metaphor, not something real), but also because this sentence is sort of a qualifier for this entire argument to be real (entire pt.4).

  4. Part 2 (4096 char limit):

    At the end, one can read:
    "In essence, the question at step four is whether the impact of the rights infringement is disproportionate to the likely benefit of the impugned measure."

    This too seem basically meaningless to me, because the phrase "the likely benefit" obviously is come to mean that any idea of a 'benefit' is to be SOLELY determined retroactively, same with determining the "impact of the rights infringement", so with the phrase "the measure will contribute to its achievement", the reality of the 'achievement' part is detatched from the reality of 'the measure' part and "the achievement" part have no meaning other than being something for being determined retroactively, and thus in this context, a measure can have no innate achievement, which again doesn't make any sense.

    Another thing, the word "balancing" was used, but the use of this word there is probably meant to convey the meaning of making a change or adjustment, with the word "balancing" having no deeper meaning beyond making a change.

    Note, I am no expert on legal matters, but the crux of the matter here seem to be: a major point is to make a determination if something is proportional when the police want to arrest someone, or do something, BUT not afterwards, and especially, not afterwards after that again, as if needing to make REevaluations of 'proportionality'.

    Weirdly enough, pt.2 makes a point about the importance of determining if a measure is rationally connected to the objective. Obviously, the 'objective' is not the same thing as the 'achievement'. Presumably an 'objective' is also self reflexive, a conceptual metaphor, only existing for the purpose of being self referential.