DISSENTING OPINION OF JUDGE NICOLAOU

In Hirst v. the United Kingdom (no. 2) [GC], no. 74025/01, ECHR 2005 ‑ IX, the Court was prepared to accept, albeit rather reluctantly (see § 75), that section 3 of the Representation of the People Act 1983, which did not allow convicted prisoners to vote for as long as they remained in custody, pursued the legitimate aim of discouraging crime while “enhancing civic responsibility and respect for the rule of law”, as the respondent State had suggested. Then, ruling on the proportionality of the measure, the Court found that the measure was too broad, since it covered the whole spectrum of those convicted, and expressed the view that it was a blunt instrument, embodying past notions which remained untested by “any substantive debate ... in light of modern-day penal policy and of current human rights standards” (see § 79). However, it would seem from the tenor of the judgment that such a measure, with a single all-embracing restriction, could not have been saved even if it had been preceded by such debate. The Court said, at § 82, that

“... although the situation was somewhat improved by the 2000 Act which for the first time granted the vote to persons detained on remand, section 3 of the 1983 Act remains a blunt instrument. It strips of their Convention right to vote a significant category of persons and it does so in a way which is indiscriminate. The provision imposes a blanket restriction on all convicted prisoners in prison. It applies automatically to such prisoners, irrespective of the length of their sentence and irrespective of the nature or gravity of their offence and their individual circumstances. Such a general, automatic and indiscriminate restriction on a vitally important Convention right must be seen as falling outside any acceptable margin of appreciation, however wide that margin might be, and as being incompatible with Article 3 of Protocol No. 1”.

The Court also pointed out that disenfranchisement, in order to be justified, required “a discernible and sufficient link between the sanction and the conduct and circumstances of the individual concerned” (see § 71). In this regard the Court cited with approval a recommendation of the Venice Commission that “... the withdrawal of political rights ... may only be imposed by express decision of a court of law” and noted, at § 77, that “...when sentencing, the criminal courts in England and Wales make no reference to disenfranchisement and it is not apparent, beyond the fact that a court considered it appropriate to impose a sentence of imprisonment, that there is any direct link between the facts of any individual case and the removal of the right to vote.” That paragraph ended with the statement that “[a]s in other contexts, an independent court, applying an adversarial procedure, provides a strong safeguard against arbitrariness.”

In light of those views, the Court subsequently held in Frodl v. Austria (no. 20201/10, §§ 34-35, 8 April 2010), that “the decision on disenfranchisement should be taken by a judge” and must contain reasoning as to “... why in the circumstances of the specific case disenfranchisement was necessary”. However, in Scoppola v. Italy (no. 3) ([GC], no. 126/05, 22 May 2012), where the applicant had been convicted of murder in aggravating circumstances and was ultimately sentenced to thirty years’ imprisonment, it was said that the “reasoning in Frodl takes a broad view of the principles set out in Hirst (no. 2), which the Grand Chamber does not fully share” and that disenfranchisement may take effect on the basis of legislative provisions, without specifically being ordered by judicial decision. This was summed up in the following statement, at § 99:

“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”.

The Court then went on to find, in Scoppola (no. 3), that the general measure which Italy had adopted was Convention compliant. This was, essentially, because

“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”.

Further clarifications made by the Court are not directly relevant to what is now in issue.

What I consider to be critical is that in Scoppola (no. 3) the Court expressed its approval of a specific level of protection above which, therefore, States are not obliged to go, while it did not exclude that a lower level of protection may also suffice.

It will be recalled that in Hirst (no. 2) the United Kingdom had argued that, even if a much less restrictive measure had been in place, the applicant, with a sentence of life imprisonment, could not reasonably have expected to benefit. He thus lacked victim status and his application was actually an actio popularis . The Chamber answered this, in its judgment, by saying, at § 51, that

“The Court cannot speculate as to whether the applicant would still have been deprived of the vote even if a more limited restriction on the right of prisoners to vote had been imposed, which was such as to comply with the requirements of Article 3 of Protocol No.1.”

The Grand Chamber endorsed that statement but with some shift in emphasis. It said this:

“[The applicant] was directly and immediately affected by the legislative provision of which he complained, and in these circumstances the Chamber was justified in examining the compatibility with the Convention of such a measure, without regard to the question whether, had the measure been drafted differently and in a way which was compatible with the Convention, the applicant might still have been deprived of the vote.”

As to what restrictions, if any, would be compatible with Article 3 of Protocol No. 1 the Grand Chamber declined to offer guidance. It was not prepared, owing to the nature of the matter, to take the initiative of saying how and where to draw the line. It explained, at § 84, that

“In a case such as the present one, where Contracting States have adopted a number of different ways of addressing the question of the right of convicted prisoners to vote, the Court must confine itself to determining whether the restriction affecting all convicted prisoners in custody exceeds any acceptable margin of appreciation, leaving it to the legislature to decide on the choice of means for securing the rights guaranteed by Article 3 of Protocol No. 1.”

Now, however, on the basis of the judgment in Scoppola (no. 3) it can be said with certainty, first, that disenfranchisement can take place without a specific judicial decision to that effect if the legislation of the State so provides; secondly, that the law may provide for disenfranchisement where a convicted person is sentenced to imprisonment for five years or more; and, thirdly, a lower threshold for disenfranchisement is not excluded. The conclusion that no specific judicial decision is required for disenfranchisement has obviously been the result of further reflection, while the recognition that disenfranchisement is permissible, where imprisonment exceeds a certain threshold, has been made possible by examining a concrete legislative measure that provided limits. The finding of a violation in Hirst (no. 2) must, I think, be attributed to the fact that the Court was not then prepared to accept these propositions without more.

In my opinion, it would not in principle be right to read Hirst (no.2) as meaning that if a general legislative measure regulating pre-defined situations does not meet with Strasbourg’s approval, an applicant must necessarily succeed in his claim. A general measure transcends the individual facts of any particular case and may indeed work hardship in some. Yet, where its necessity and proportionality are demonstrated by relevant and sufficient reasons, the measure will be upheld irrespective of how the balance lies in the individual case (see Animal Defenders International v. the United Kingdom [GC], no. 48876/08, §§ 106-24, 22 April 2013) and the concurring opinion of Judge Bratza, at point 4. Where, on the other hand, a general measure does not meet these requirements the Court has to consider the matter outside the context of the measure and, the legitimate aim of the restriction having been established, to examine the proportionality of the interference on the facts of the individual case seen in the light of the case-law.

The applicants have given no details of the respective dates of their conviction and length of sentence, that is, whether a life sentence, a long term of imprisonment or a relatively short one had been imposed. Neither have they informed the Court of whether they are still in detention, almost five years after their applications were introduced. It was incumbent on them to do so (see Dunn v. the United Kingdom, (dec.), no. 566/10 and 130 other applications, §§ 16 and 17, and McLean and Cole v. the United Kingdom (dec.), nos. 12626/13 and 2522/12, §§ 5-12, 11 June 2013).

In these circumstances I am unable to accept the majority view that the applicants’ complaints are admissible or, having been held so to be, that they have been substantiated. I conclude, therefore, that there has been no violation of Article 3 of Protocol No. 1.