Tom Hickman - London Reiew of Books

On​ 29 June, two days after the government announced it was shelving its Bill of Rights Bill, the Court of Appeal ruled that the government’s policy of sending asylum seekers to Rwanda breached the European Convention on Human Rights, reigniting calls on the right for the UK to withdraw from the convention. When it introduced the Bill of Rights in 2022, the government promised that the UK would continue to respect its obligations to the convention, but the Bill would have introduced various mechanisms to ensure that domestic courts interpreted those obligations more narrowly than the European Court of Human Rights in Strasbourg. This would have been to breach the convention: rulings of the ECHR are authoritative interpretations of the convention and its rulings against a state are binding on that state under international law. The UK cannot pick and choose which rulings it wishes to follow. Commitment to the convention, as in any rules-based system, entails taking the rough with the smooth.

The Court of Appeal’s Rwanda ruling is on the rough side of the ledger. By a majority, the court held that there are substantial grounds for believing that sending asylum seekers to Rwanda carries a real risk of their being sent back to their home countries to face death, torture, cruel, inhuman or degrading treatment, contrary to the ‘non-refoulement’ principle of international law. The court held that the issue was not the good faith of the Rwandan government’s assurance that it would properly process asylum seekers, but, in the words of Lord Justice Underhill, ‘its ability to deliver’.

Government ministers have refused to rule out withdrawing from the convention if this judgment is upheld by the Supreme Court, saying that they will do whatever is necessary to ensure the effectiveness of asylum policy. Withdrawal from the convention would have exceptionally serious consequences. The UK would have to leave the Council of Europe, the multinational organisation it helped found after the Second World War to protect democracy and human rights in Europe, and which has enabled a multinational approach in policy areas such as cybercrime, terrorism and the environment, and allowed tens of thousands of bilateral treaties and arrangements to be replaced. Withdrawal would also breach the Good Friday Agreement with Ireland and contravene the Trade and Co-operation Agreement with the European Union, resulting in the suspension of co-operation on police and security. Only the Greek junta in 1969 and Russia last year have left the convention – Russia went shortly before it was to be expelled for invading Ukraine. No democratic country has sought to do so.

There is also a more basic problem with withdrawal. The objection to the convention isn’t to its text, which even its opponents accept is unobjectionable, but to what is thought by some to be the expansive manner in which the European Court of Human Rights has interpreted it. That the court would develop principles through case law was understood when the convention was drafted, but sometimes it goes too far. One flashpoint concerned its ruling in 2005 in Hirst v. United Kingdom that denying prisoners a vote breached the convention, although the convention guarantees only the holding of free and fair elections and does not specify the make-up of the franchise. Even so, the court’s ruling was not as expansive as is sometimes claimed: it did not rule that all or even most prisoners must have a vote, merely that UK law and practice was unjustifiably sweeping. The issue was resolved after the UK accepted that since its policy was that those in custody should not have the vote it did not need to disenfranchise those released on licence or on home curfews, some of whom it already allowed to vote. The Committee of Ministers of the Council of Europe, which oversees compliance with court judgments, was satisfied with this response.

Another flashpoint concerned the ruling in Hatton v. United Kingdom, a case from 2003, that the noise caused by night flights into Heathrow breached residents’ rights to a private life. But the Strasbourg system provides governments with an opportunity to challenge rulings in a Grand Chamber, which reversed the decision. In other cases, the ECHR has modified its position to reflect criticism from domestic courts, notably in changing its position on the use of hearsay statements in criminal trials.

In the Rwanda case, the principle at issue is of a different kind, a fundamental part of international law: non-refoulement. The convention places duties on member states not to subject people in their territories to death, torture or cruel, inhuman and degrading treatment. It does not involve a surprising or expansive reading of the text for the court to hold that the convention prevents states handing people over to other states to face such treatment. This principle was first articulated by the court in 1989 in the case of Soering v. United Kingdom, a unanimous judgment of the court sitting in plenary session – with all judges present. They included Sir Vincent Evans, the UK representative, who before his appointment had been legal adviser to the Foreign Office.

The principle of non-refoulement is also enshrined in other international treaties that the UK has ratified and has no plans to withdraw from. The 1984 United Nations Convention against Torture, which has 173 state parties, provides in its third article that ‘No state party shall expel, return (“refouler”) or extradite a person to another state where there are substantial grounds for believing that he would be in danger of being subjected to torture.’ Article 33 of the Refugee Convention provides that no state shall ‘expel or return (“refouler”) a refugee in any manner whatsoever to the frontiers of territories where his life or freedom would be threatened on account of his race, religion, nationality, membership of a particular social group or political opinion’. The International Covenant on Civil and Political Rights, also with 173 parties, contains provisions similar to those in the European Convention on Human Rights and is interpreted in the same way. The UN Human Rights Committee, the oversight body that provides guidance on the covenant, recognises that membership ‘entails an obligation not to extradite, deport, expel or otherwise remove a person from their territory, where there are substantial grounds for believing that there is a real risk of irreparable harm … either in the country to which removal is to be effected or in any country to which the person may subsequently be removed’. In 1994 the UN high commissioner for refugees found the principle of non-refoulement to be part of customary international law, the universal law that binds all states irrespective of treaty obligations, observing that states regarded the principle as binding even if they had not ratified the Refugee Convention.

This is, in other words, a core principle of international law, and one that the UK has repeatedly signed up to. The principle is not derived from expansive jurisprudence of the ECHR and withdrawing from the convention would not free the UK from its obligation to observe it. The reaction to the Court of Appeal’s ruling therefore makes little sense. The true objection might not be to the principle itself but to its enforcement. But human rights are rules of law which must be observed and cannot be allowed to depend on a government’s assessment of its own compliance. One of the most deeply embedded principles of common law is that no man shall be judge in his own cause.

The UK accepted the jurisdiction of the European Court of Human Rights, and its system of adjudication is robust and fair. It is a far better system than in Canada, Australia and New Zealand, for example, which allow individual complaints to be made under the International Covenant on Civil and Political Rights to the UN Human Rights Committee, a quasi-judicial process without the protections, rigour or clarity of the convention process.

Furthermore, the great merit of the Human Rights Act 1998 is that it enabled international human rights law to be applied by our domestic courts. That is what the Court of Appeal was doing in the Rwanda case. British judges considered thousands of pages of evidence and legal submissions, and produced judgments explaining precisely why the non-refoulement principle had been breached (the dissenting judge, Lord Burnett, lord chief justice, explained why he took a different view).

Earlier in the Rwanda case, the ECHR attracted criticism for making an ‘interim measures’ ruling requiring the UK not to transfer anyone to Rwanda while the case was being considered. The court can request a state to refrain from taking action that may cause irreparable harm, an authority it has used sparingly, mainly to prevent deportation or extradition. Domestic courts had refused interim relief in the Rwanda case, partly on the basis that the memorandum of understanding with the Rwandan government provided that it would use its best endeavours to return asylum seekers to the UK if required by a UK court. The Strasbourg Court found this undertaking insufficient, since the memorandum is not legally enforceable and does not obviate the risk of return to a home state before the completion of the UK proceedings.

The intervention of the ECHR was controversial, in part because its process for granting such a ruling is far less thorough and transparent than a UK application for interim relief: in the Rwanda case there were judgments from the Divisional Court, the Court of Appeal and the Supreme Court. The interim measures were granted by a duty judge at the Strasbourg Court without a hearing. Criticism of the court for extending its jurisdiction also has some force here. Rule 39 allows the court to make a ‘request’ to a state about an interim measure a judge thinks should be taken. But there is no obligation to follow such measures in the convention and the court’s case law originally held that interim measures were not binding. However, in Mamatkulov and Askarov v. Turkey, a case from 2005, the court took a new approach, stating that a state would breach the convention if it failed to comply with an interim measure. The court reasoned that non-compliance would breach Article 34, by which states undertake not to hinder parties in exercising their right to individual petition. In that particular case there was no doubt that the Turkish government’s failure to comply with an interim measure had resulted in a breach of Article 34: the two applicants were extradited to Uzbekistan and lost contact with their lawyers, prejudicing their application. But non-compliance with an interim measure will not always have such an effect.

It is important, however, that the court has a means of ensuring that convention rights are not undermined by states taking action before an application has been determined. The government has accepted this. On 13 June, Lord Bellamy, the parliamentary undersecretary of state for justice, stated in response to a parliamentary question that the government ‘recognises that interim measures can be an important mechanism for securing an individual’s convention rights in exceptional circumstances’. He claimed that the government merely wants a ‘better balance’ of fairness, transparency and the administration of justice in the operation of the interim measures system. The government, he said, has had constructive discussions with the court about reform and the matter is being looked into as part of the court’s regular procedural review. Ideally, interim measures would be addressed in a separate protocol agreed by all member states. But this issue, which although important is at the margins of the court’s practice, shouldn’t be used to justify tearing up the UK’s fundamental international obligations – and its reputation for developing and upholding the rule of law.