By Lewis Graham
In 2005, the Grand Chamber of the European Court of Human Rights handed down its landmark decision in Hirst v the United Kingdom, finding that the effect of section 3 of the Representation of the People Act 1983, bringing into effect a blanket ban on the ability of prisoners in the UK to vote in elections, constituted a breach of Article 3 of Protocol 1 of the Convention (the right to free elections).
To say the case was controversial is an understatement, with the judgment becoming something of a bête noire for Strasbourg sceptics. Murray suggests that the judgment was pivotal in the “monstering” of the European Court. It is often presented as a case which epitomises Strasbourg overreach, taking the number 1 spot in the Judicial Power Project’s buffet of unfavourable, “problematic” legal cases. David Cameron, of course, famously remarked that the idea of complying with the judgment and giving (some) prisoners the vote made him feel “physically sick”.
In 2018, after some 13 years of waiting, the Strasbourg Committee of Ministers adopted a resolution (“the 2018 Resolution”) concluding that the UK had done enough to remedy the breach of A3P1 – it had promised, at the time, to instigate reforms to allow some prisoners to vote. Non-legislative changes were also made around that time, allowing prisoners released on temporary licence to vote.
Now, in 2025, the European Court of Human Rights has ruled, a full 20 years after Hirst, that an applicant prisoner’s inability to vote in a UK General Election did not breach the Convention. The case is Hora v the United Kingdom. As the Court put it ([132]-[135]):
The present applicant was convicted of rape and sexual assault in 2007. There can be no doubt as to the gravity of these offences, which constituted a serious attack on the values of society and on social order… it cannot be said that the disenfranchisement of the present applicant, on account of the seriousness of his offending, his conduct, the risk he was found to pose to the public and the resulting imposition of a harsh sentence of indeterminate detention, was disproportionate to the legitimate aims pursued by restrictions of the franchise applied to convicted prisoners … there has accordingly been no violation of Article 3 of Protocol No. 1 to the Convention on account of the ineligibility of the applicant to vote in the general election…
So how did we get here? The same legislation which the Strasbourg Court found incompatible with the Convention in Hirst – section 3 of the Representation of the People Act 1983 – remains in full force. In Hirst this was enough for the Court to find a breach of A3P1; in Hora it was not. What changed?
Three factors may be pertinent. Firstly, the Hora Court “[took] into account” the 2018 Resolution that the incompatibility identified in Hirst had been remedied by the UK’s promise to introduce exceptions to the blanket ban. But in 2019 the Court, in Miller and Others v the United Kingdom, found a breach of A3P1, on the same basis as the Court in Hirst, notwithstanding the 2018 Resolution. So this cannot be a decisive reason for the change.
Secondly, the Hora Court acknowledged that the case law on disenfranchisement had been “develop[ed] and refine[d]” since Hirst ([123]). This, presumably, is a reference to the Grand Chamber’s decision in Scoppola v Italy (No 3). But important decisions against the UK were handed down after the Scoppola decision, including the aforementioned Miller decision, confirming the Hirst finding that the UK’s blanket ban did not comply with the Convention. So this, too, cannot be a decisive reason for the change.
The third and most important reason that the Court in Hora diverted from the Hirst judgment is the very important fact that the Hora Court did not examine the compatibility of the legislation which prevents prisoners from voting. It held that it was “not justified to examine the Convention compatibility of… the Act… instead, [the Court would] examine [the situation of] this specific applicant, in his particular circumstances” ([130]).
Of course, the Court did examine the compatibility of the Act as a whole in Hirst (the Hora Court accepts as much at [115]). The 1983 Act as a whole was the reason for the breach in that case. The Hora Court doesn’t say that the Hirst Court was wrong to do so explicitly – but does imply that this approach was out of step with the court’s overall task ([114]).
Assessing each applicant’s situation (the Hora approach), rather than the legislation as a whole (the Hirst approach) means that, as the Court puts it, the application of a general measure might be proportionate for one applicant but not another ([125]). This is important. Whilst the applicant’s disenfranchisement was considered proportionate due to his serious rape conviction ([132] to [135]), this might not be the case for other prisoners.
This is how Hirst can be reconciled with Hora. The latter does not overrule the former. In particular, the door remains open for a prisoner serving a shorter sentence for a less serious offence to challenge the application of the 1983 Act to their particular case. Following Hora, it cannot be said that their challenge would fail.
Dr Lewis Graham is a Lecturer in Human Rights Law at the University of Manchester.
