This case involved Italy, not Britain – but nonetheless today’s judgment of the Grand Chamber of the European Court of Human Rights, about prisoner’s rights to vote, represents a small but significant victory for the British government. The Court has rowed back considerably from the reasoning in the Chamber judgment, which found Italy’s automatic ban on voting for prisoners in breach of Article 3 of Protocol 1 to the European Convention on Human Rights, and gives significantly more flexibility to the UK in the way it brings in prisoners’ votes.

It does not solve the government’s dilemma on prisoners’ votes – far from it. The UK must still allow at least some prisoners the vote, as required by the judgments in Hirst v UK, and Greens & MT v UK. But it does ease the difficulty. It vindicates at least to some extent the strategy of “dialogue” pursued in particular by the Attorney General Dominic Grieve, in which the government has sought to influence the court to recognise a greater margin of appreciation for national policymakers, and especially to respect elected legislatures like Parliament.

The Chamber’s reasoning in January 2011 amounted to a serious limitation of state’s room for policy maneouvre. Applying the earlier case of Frodl v Austria it said (para. 43 of the Chamber judgment; the translation from French is mine):

it is essential that the decision on disenfranchisement be taken by a judge and be appropriately reasoned

The breach of the Convention arose because (para 49)

the criterion laid down in the law was merely a temporal one in this case, as the applicant was deprived of the right to vote because of the length of his custodial sentence, irrespective of the offence committed or of any examination by the trial court of the nature and gravity of the offence (see Frodl v. Austria, cited above, §§ 34 and 35)

The big question today was whether the Grand Chamber would affirm the Frodl approach adopted by the Chamber, or whether it would ease the problems of the UK government by returning to the more flexible approach the Court took in its “pilot judgment” in Greens & MT v UK, in which it had stressed the room for policy choice available to the UK in removing what the Court has called the “blanket” ban on prisoner voting. This judgment is a clear return to the Greens & MT approach.

The Grand Chamber’s legal reasoning is at paragraphs 93-110. It criticises the reasoning in Frodl (para. 99):

That reasoning takes a broad view of the principles set out in Hirst, which the Grand Chamber does not fully share. The Grand Chamber points out that the Hirst judgment makes no explicit mention of the intervention of a judge among the essential criteria for determining the proportionality of a disenfranchisement measure. The relevant criteria relate solely to whether the measure is applicable generally, automatically and indiscriminately within the meaning indicated by the Court (see paragraphs 85, 86 and 96 above). While the intervention of a judge is in principle likely to guarantee the proportionality of restrictions on prisoners’ voting rights, such restrictions will not necessarily be automatic, general and indiscriminate simply because they were not ordered by a judge. Indeed, the circumstances in which the right to vote is forfeited may be detailed in the law, making its application conditional on such factors as the nature or the gravity of the offence committed.

and stresses at para. 102 the room allowed for national policymaking:

the Contracting States may decide either to leave it to the courts to determine the proportionality of a measure restricting convicted prisoners’ voting rights, or to incorporate provisions into their laws defining the circumstances in which such a measure should be applied. In this latter case, it will be for the legislature itself to balance the competing interests in order to avoid any general, automatic and indiscriminate restriction. It will then be the role of the Court to examine whether, in a given case, this result was achieved and whether the wording of the law, or the judicial decision, was in compliance with Article 3 of Protocol No. 1.

Interestingly the court notes at 105 an important fact – that the Italian system in question in this case is in one respect harsher than the British system condemned in Hirst. The Italian ban may only bite on those sentenced to three years, but it lasts even beyond the term of their sentence:

prisoners sentenced by the courts to three years’ imprisonment or more forfeit the right to vote temporarily, for five years, while those sentenced to five years or more, or to life imprisonment, permanently forfeit the right to vote

The Grand Chamber makes clear it realises that the Italian system does not apply automatically to all prisoners, and that the length of the ban is related to the length of imprisonment, and, therefore to the seriousness of the offence for which the prisoner is sentenced. This is an important point for the UK (para. 106):

In the Court’s opinion the legal provisions in Italy defining the circumstances in which individuals may be deprived of the right to vote show the legislature’s concern to adjust the application of the measure to the particular circumstances of the case in hand, taking into account such factors as the gravity of the offence committed and the conduct of the offender. It is applied only in connection with certain offences against the State or the judicial system, or with offences which the courts consider to warrant a particularly harsh sentence, regard being had to the criteria listed in Articles 132 and 133 of the Criminal Code (see paragraph 37 above), including the offender’s personal situation, and also to the mitigating and aggravating circumstances. The measure is not applied, therefore, to all individuals sentenced to a term of imprisonment but only to those sentenced to a prison term of three years or more. Italian law also adjusts the duration of the measure to the sentence imposed and thus, by the same token, to the gravity of the offence: the disenfranchisement is for five years for sentences of three to five years and permanent for sentences of five years or more.

These factors lead the court to distinguish the case from Hirst (para. 108):

In the circumstances the Court cannot conclude that the Italian system has the general, automatic and indiscriminate character that led it, in the Hirst (no. 2) case, to find a violation of Article 3 of Protocol No. 1. In Italy there is no disenfranchisement in connection with minor offences or those which, although more serious in principle, do not attract sentences of three years’ imprisonment or more

And the ultimate possibility of restoration of the right to vote also makes a difference, in the court’s view (para. 109):

Furthermore, the Court cannot underestimate the fact that under Italian law it is possible for a convicted person who has been permanently deprived of the right to vote to recover that right. Three years after having finished serving his sentence, he can apply for rehabilitation

This is another important point for the UK since British prisoners automatically get back the right to vote not just at the expiry of their sentence, but on release – which is earlier.

The Grand Chamber’s conclusion is at para. 110:

Taking the above considerations into account, the Court finds that, in the circumstances of the present case, the restrictions imposed on the applicant’s right to vote did not “thwart the free expression of the people in the choice of the legislature”, and maintained “the integrity and effectiveness of an electoral procedure aimed at identifying the will of the people through universal suffrage” (see Hirst (no. 2) [GC], cited above, § 62). The margin of appreciation afforded to the respondent Government in this sphere has therefore not been overstepped.

There is a dissenting judgment from Judge Björgvinsson, whose main criticism of the majority is its inconsistency with Hirst. He rightly points out that the Italian legislation is in some ways harsher that British law:

The main difference between the two is that the Italian legislation deprives of voting rights only those who are sentenced to three years or more in prison, while the United Kingdom’s legislation deprives all persons sentenced to imprisonment, for the duration of their time in prison. While the Italian legislation may seem for this reason to be more lenient in comparison with that of the United Kingdom, it is stricter in the sense that it deprives prisoners of their right to vote beyond the duration of their prison sentence and, for a large group of prisoners, for life. Therefore, unlike the majority, I find that these differences are not sufficient to reach a different conclusion …

… I find the distinction made in this judgment between these two cases as a ground for justifying different conclusions to be unsatisfactory. The present judgment offers a very narrow interpretation of the Hirst judgment and in fact a retreat from the main arguments advanced therein. Regrettably the judgment in the present case has now stripped the Hirst judgment of all its bite as a landmark precedent for the protection of prisoners’ voting rights in Europe.

I don’t agree with Björgvinsson’s legal conclusion, but I do think his dissent rightly points up the intellectual inconsistency of the majority. The only basis on which the Italian system can be seen as more liberal in its treatment of prisoners is that it does not take the vote from most prisoners serving under 3 years. Logically, all that remains of Hirst is that automatically banning prisoners from voting breaches the Convention if they are sentenced under 3 years. The majority ought either to have departed from Hirst, as the Attorney General invited them to; or to have made clear that its true rationale must now be understood as being that that the margin of appreciation does not permit states to ban shorter-term prisoners from voting. The majority’s failure to do this is a political fudge.

So: what does this mean for the government? It’s inconceivable that there’ll be any more significant movement from the court, so Hirst must now be implemented within six months of today’s judgment (i.e. by 23 November). Ministers will need to bring this issue back to Parliament.

But what’s also now clear is that their original compromise plan to give the vote to all prisoners serving less than four year sentences, and to give trial judges the power to ban even them from voting in appropriate individual cases, goes further than they need to in order to satisfy Strasbourg. Ministers can confidently propose to restate their plan based on a three-year cut off. In fact, I’d go further: because of the ways in which our own legislation is more generous than that of Italy, there’s a good chance that a voting ban taking effect on prisoners sentenced to two years will, if challenged, be upheld by the European Court.

A pretty good day for Dominic Grieve personally, then, and for the government. They can go back to Parliament arguing that they have persuaded the court to back down to a significant degree, and that the way is open to an acceptable, if not palatable, compromise.

The vital question now is what Parliament will make of that. Some MPs will want to continue to ignore the Hirst judgment and to defy the court. They’ll undoubtedly have their way if they command a majority, leaving the government with a diplomatic headache.

But the government could just conceivably get support for a solution if it at the same time it throws some red meat to its Conservative Eurosceptic backbenchers – which this judgment may give it some room to do. A bill giving the vote to prisoners serving under three years but extending the ban for those serving 3 years or more for a period beyond the expiry of their sentence – as the court has today ruled is permissible – would in a sense take from the court as well as giving to it. Parliament would be saying “Okay: you’ve ruled out our in many ways relatively liberal voting ban; fine. We’ll do just as you say, and bring in the in other ways more draconian ban you prefer.”

Could MPs be persuaded to vote for that? I wonder.

The BBC reported yesterday that there’s “doubt” about the deportation of Abu Qatada, following his arrest on Tuesday and now his appeal to the European Court of Human Rights – which the Home Secretary Theresa May says is out of time. So: is she right? Is the appeal out of time? How has the Home Office got into this apparent mess? And what if any difference does this appeal make?

The European Court’s judgment in Abu Qatada’s case was dated January 17th 2012. Of that there’s no doubt; and it’s irrelevant whether the government or anyone else was given notice of the judgment before, or received it later.

Article 43.1 of the European Convention on Human Rights says

Within a period of three months from the date of the judgment of the Chamber, any party to the case may, in exceptional cases, request that the case be referred to the Grand Chamber.

It’s this reference to the Grand Chamber that’s the “appeal” Abu Qatada has now lodged.

Article 44.2(b) says that the judgment becomes final

three months after the date of the judgment, if reference of the case to the Grand Chamber has not been requested

If you’re interested in the French text (and you should be – the meaning of the Convention depends on both texts), it says (Article 43.1)

Dans un délai de trois mois à compter de la date de l’arrêt d’une chambre, toute partie à l’affaire peut, dans des cas exceptionnels, demander le renvoi de l’affaire devant la Grande Chambre

and that judgment becomes final (Article 44.2(b)

trois mois après la date de l’arrêt, si le renvoi de l’affaire devant la Grande Chambre n’a pas été demandé

On the face of it, the English texts are ambiguous: it would have been clearer had article 43.1 had said either that reference could be made within three months beginning on and including the date of judgment, or else within three months after that date. You might think the French text is more clearly helpful to the government, since it emphasises that counting begins from the date of judgment – but even that’s not completely clear. Yes, you clearly count from the 17th January; but the legal issue is whether, when you then point to 18th January on the calendar, you mentally count that as day 1 (the clock having notionally started at day zero on the 17th), or day 2 (“judgment day” having been day one).

The first point to make about this ambiguity is that it illustrates the importance of precision in legal texts. People are too often tempted to think that legal exactitude in drafting amounts to mere verbiage. Not so. The second is that it’s a bit surprising that the Convention is unclear – but not amazing. It’s not easy to achieve complete clarity in texts negotiated internationally, as these texts were in Protocol 11 to the ECHR. Whatever the government achieves at Brighton, I expect plenty of argument about what any new Protocol means.

How, then, to resolve this ambiguity? The first principle legally speaking must be to interpret these provisions of Convention in the light of their object and purpose – in accordance with Article 31 of the Vienna Convention on the Law of Treaties. In my view the purpose of the provisions is to give a clear and finite period for “appeal”, and their purpose is best served by reading them firstly, as allowing as clear a period as possible, and secondly, by reading them at the margin favourably to applicants. To me, that argues in favour of seeing applicants as having a full, clear three calendar months in which to apply, i.e. three months after the date of the judgment. Such a period would begin the day after judgment and would last three full months expiring at midnight (which midnight, by the way? Strasbourg midnight, presumably) at the end of Tuesday 17th April. The appeal would be in time, on this reckoning.

Adam Wagner was typically quick and right on Twitter yesterday to draw our attention to the relevant cases: Praha v Czech Republic, and Otto v Germany, neither of which is specifically about this three-month period but rather the six-month time limit for bringing your claim to the ECtHR in the first place. I read both cases as supporting Abu Qatada’s reading, rather than the Home Secretary’s.

In Otto, the Court said

the day on which the final domestic decision is pronounced is not counted in the six-month period referred to in Article 35 § 1 of the Convention. Time starts to run on the date following the date on which the final decision has been pronounced

and in Praha that

the six-month period begins to run on the day after the date on which the final domestic decision was pronounced

both of which support Abu Qatada’s contention that the three-month period began on 18th January – the day after its judgment was pronounced – and so expired at the end of the 17th April.

Adam also drew our attention to the Court’s “General Practice” note on Article 43 applications (page 14):

the period of three months within which referral may be requested runs starts to run on the date of the delivery of the judgment, irrespective of whether the party concerned may have learned of it at a later stage.

That doesn’t really help any further because as in the text of the Convention itself, it’s not clear whether the count starts with “judgment day” as day zero or day one. I think the answer must be found in purposive interpretation, supported by Otto and Praha.

How could the Home Office get mixed up about this, then? Some commentators on the media last night, with the complacent superiority that I know comes naturally to us commentators, expressed disbelief that the government didn’t simply “phone the court”. Things are not as simple as the commentator’s imagination might suggest, though. I’ve no doubt government lawyers were on the phone to Strasbourg more than once.

There may well have been crossed wires here between Foreign Office Lawyers – who deal with the European Court of Human Rights and who formally conduct litigation in it – and Home Office lawyers and civil servants. The prime concern in Whitehall’s mind must have been how long the government itself had to appeal the Abu Qatada judgment, which after all went his way in a practical sense (although a key element of its legal reasoning suited the government, as Theresa May agreed in Parliament on Tuesday). It’s not at all surprising that, if it were asked how long the government had to appeal (and what other purpose could the UK have had to ask?) Court staff should cautiously have cited the 16th of April. If I’d been an FCO lawyer asked to advise when the UK needed to appeal by, I’d have said 16th April, to be safe.

To be fair to government lawyers (of whom I’ve been one), these time issues can be tricky. It often amazes non-lawyers that there can be confusion about questions like this, but one of the surprising things you learn at law school is that it’s not obvious how you calculate time. Indeed, a whole section of the massive and brilliant law encyclopedia Halsbury’s Laws is devoted to the law of time.

In government you have the added difficulty that your ministerial clients seem obsessed with time, and the room it gives to delay decisions, in contexts where giving precise advice on time is difficult. I’ve advised on many EU law cases where the time-limit for a UK response depended in part on complex rules involving additional days allowed to government that varied according to how long the post was assumed to take between Luxembourg and the national capital. It frustrated me enormously that the only legal advice ministers seemed to be interested in was how I calculated the precise date, and that no one seemed anxious to listen to whether I thought we could win, or in getting on with deciding what if any arguments we’d make. A week or two would go by, then at the next meeting the only question would again be: “Till when do we have?”

What I learned was that, in order to focus minds, I had to be precise, cautious and unambiguous about time limits. The worst thing that could happen would be to give a date that turned out to be too late. I also learned in government that it’s best to avoid complexity in your answers, if possible: it’s generally unwanted, and often interpreted as proof that you’re a typical civil service ditherer, whose view can be ignored. That’s why, as I’ve already said, if I’d been advising ministers in a meeting in a context where we ourselves were considering appealing, I’d have given Monday 16th April as the final day for appealing. No other answer would have been so clear and safe, or have commanded confidence. I’d have kept the hidden complexity to myself unless asked for written advice, or unless asked specifically about how long the other side might have to appeal.

What was really needed here was good, old-fashioned lawyers’ advice. Regardless of when the time-limit technically ran out, the government would have been well advised to wait until at least midnight at the end of April 17th before treating the judgment as final. Theresa May’s statement should have taken place after Prime Minister’s Questions yesterday, rather than on Tuesday. But often in modern government, advice in that broad sense seems unwelcome: if they offer it, lawyers can be told sharply that all ministers want is a simple technical answer to a simple technical question such as “how long do we have to appeal”. In this case, especially if as I suspect a government appeal was a live issue, then the only good answer to give was Monday 16th April. I wonder whether this impatient, advice-averse culture and the “garbage in, garbage out” sort of lawyering it can spawn might be partly to blame for the apparent muck-up here.

Finally, some people will wonder why on earth Abu Qatada’s lawyers would have waited till the last moment – and possibly too late – to enter their appeal. Well, first, I’ve already said I think their view of the time-limit is correct. They’re in time, which is all that matters. But there may be a practical explanation for why the application came so late.

It’s worth remembering that Abu Qatada in effect won on 17th January: the European Court of Human Rights decided he couldn’t be deported because of the possible use in Jordan of evidence obtained by torture. It’s not obvious why he’d want to appeal that ruling, or whether they have serious grounds to do so. But what his lawyers certainly want is to find some procedural means of preventing deportation now, if they can. There’s a possibility of a final appeal to the Special Immigration Appeals Commission against deportation – in which case, deportation itself will be suspended pending the outcome. But given that SIAC and the higher UK appeal courts have already ruled his deportation lawful, even before the most recent agreement with Jordan, Abu Qatada must know those proceedings offer him little hope, and may not be drawn out all that long.

He must have hoped that the government – which after all lost on 17th January and has much better grounds on which to contest it – would appeal the ECtHR’s judgment, therefore keeping proceedings in that court alive and enabling him if need be, once deportation action was revived, to ask Strasbourg again for interim measures to temporarily block it. Once it became clear from Theresa May’s Commons statement that the government was not appealing the judgment, but believed it could effect a deportation compliant with it, Abu Qatada’s lawyers finally realised the only way the existing interim measures could be kept in existence or further interim measures applied for would be their own last-minute appeal, however odd that might seem. No doubt the papers had already been prepared in advance – his lawyers are certainly a capable lot – and so were ready to be sent to the court on the day or his arrest, after a few final hours work.

Oddly what was forgotten in yesterday’s media flurry about “confusion” and “doubt” was that this appeal makes little real difference, in the scheme of things. Theresa May told the Commons on Tuesday that she expected deportation could take “many months” because of appeals – this just proves her right. And deciding not to arrest Qatada till Wednesday would not actually have prevented an appeal being made on Tuesday.

I don’t think there’s any question of Tuesday’s arrest being unlawful, by the way. What the interim measures prevent is deportation itself, rather than arrest or detention with a view to deportation.

I expect Abu Qatada to land in Jordan eventually; but not before Julian Assange lands in Sweden.