Anurag Deb and Lewis Graham
Introduction
There are many well-tuned arguments both for and against the liberalisation of the UK’s strict euthanasia laws, some more helpful than others. This piece is not concerned with weighing up the policy arguments for or against such a move, nor does it consider which “side” of the argument is ultimately more convincing. Indeed, the authors do not necessarily agree with one another on the discrete question of whether Kim Leadbeater MP’s Bill should be supported.
But one curious argument has recently emerged which is of serious concern to both authors: the argument that liberalising euthanasia laws, in line with the proposed changes in Leadbeater’s Bill, should be resisted, as doing so would be to contravene the rights under the European Convention on Human Rights (ECHR). In this post, we seek to debunk this argument.
There are, we think, two main variants of ECHR-based arguments to this effect: one invoking Article 14 (freedom from discrimination) and the second relying on Article 2 (right to life). Neither is convincing.
The Article 14 argument
The more sophisticated argument, put forward by Philip Murray in a post on the UKCLA Blog, goes like this. Allowing some groups, such as terminally ill people, to access euthanasia treats two comparable groups differently; this difference in treatment constitutes discrimination under the Convention; this discrimination is unlikely to be justifiable; therefore to bring about the changes Leadbeater proposes would put the UK in breach of Article 14 ECHR.
There are four stages to a discrimination claim under Article 14. Each must be satisfied in order to establish a breach of the Convention. The first step is the only one which is easily satisfied in this case. First, the difference in treatment complained of must come within the ambit of a Convention right. Euthanasia laws engage, at minimum, Article 2 of the ECHR (and likely Article 8 of the ECHR, too).
Next, the difference in treatment must be based on the “status” of the parties. Whilst the notion of “status” is wide, the concept is not open-ended and “is clearly not intended to cover differential treatment on any ground whatsoever” (A and B v Criminal Injuries Compensation Authority [2021] UKSC 27 at [40]). Further, status cannot be defined solely on the basis of being treated differently from another, or else it would be indistinguishable from justification, and this part of the case law would be circular and become moribund: R (SC) v Secretary of State for Work and Pensions [2021] UKSC 26 at [69]-[71]. Nevertheless, we concede that it is likely that “status” could be defined, for example, according to whether an individual is terminally ill or not.
Here, too, we can concede that despite the unclear and sometimes contradictory case law on this aspect of Article 14, it is conceivable that any difference in treatment between, say, terminally ill people and non-terminally ill people will be based on their illness, which qualifies as a “status”. The courts have, for example, accepted that both having a disability, and having a particularly serious disability, constitutes a relevant status for the purposes of Article 14 (SM and SDJ v London Borough of Hackney [2021] EWHC 3294 (Admin)). The reasoning in these cases could easily be read across to the terminal illness context.
The third step in the Article 14 exercise involves determining whether there is a relevant difference in treatment between the claimant and another comparator group. This group must be in a sufficiently similar position to the claimant in order to qualify – this does not require absolute parity, but neither does it permit spurious comparisons: the courts have used the language that the two groups must be in a “relatively similar” situation in order to be considered analogous: SC at [47]. We suggest that the claim that Article 14 applies to the Leadbeater Bill and the euthanasia context may run into trouble at this stage. The courts have recently rejected the argument, for example, that financially dependent carers can be fairly compared with non-dependent carers (R (ZX) v Criminal Injuries Compensation Authority [2024] EWHC 1703 (Admin) at [142]). The same was found to be the case with regard to pensioners in receipt of old-age pension compared to those who were not (Milivojevic v Serbia, App no 11944/16, decision of 5 July 2022). The position of terminally ill people may not be considered “relatively similar” to those who are not in that position, especially if compared to those who are seriously, but not terminally, unwell.
It is, however, the fourth and final requirement to Article 14 which is fatal to the claim that limited euthanasia laws would necessarily breach it: any difference in treatment must be shown to be unjustified. Murray argues that “the UK would need the clearest justification for terminally ill adults being able to end their lives but not those experiencing equal or greater suffering because of physical disability, acute mental illness, etc.” But this is to ignore how both Strasbourg and domestic authorities review the question of justification. At the European level, there is plenty of authority – take Mortier v Belgium (App No 78017/17), Judgment of 4 October 2022 – for the proposition that a significant margin of appreciation will apply to issues such as euthanasia which raises “complex legal, social, moral and ethical issues” (Mortier, para 142). This is especially so when there is a lack of European consensus on matters – where “the legal opinions and responses among the States Parties to the Convention vary greatly, and there is no consensus as to the right of an individual to decide in which way and at what time his or her life should end” (Mortier, para 142). This all translates, against Murray’s suggestion to the contrary, that a particularly significant margin of appreciation will apply in this context (Mortier, para 143).
On the domestic front, any challenger will have to additionally contend with the statements of Lord Reed in SC that a significant amount of deference will be afforded to democratic determinations of whether differences in treatment can be justified, especially in knotty political areas: “domestic courts have to respect the separation of powers between the judiciary and the elected branches of government” in such circumstances ([144]) and they must respect “the boundaries between legality and the political process” ([162]). It goes without saying that the balancing of competing interests when it comes to drawing the boundaries of state-sanctioned euthanasia is exactly the kind of question which Lord Reed believes should fall to be answered by political, rather than legal actors. The legislative determination as to how far euthanasia is to be permitted, if at all, is the paradigm case for Reed-style political deference (on which, see Lord Reed’s dissenting view in Nicklinson). Domestic courts will not have the appetite to revisit the seven-judge panel in SC nor the nine-judge panel in Nicklinson.
There is an additional point when considering the ambit of Article 14 issues with respect to a State’s attempt to positively legislate an entitlement. Leadbeater’s Bill is titled the Terminally Ill Adults (End of Life) Bill. At the point of writing, the Bill itself remains unpublished. The title, however, would suggest that that the entitlement to euthanasia is restricted to the only cohort of people (terminally ill adults) with whom the Bill is concerned. Put another way, the Bill may not define terms such as ‘suffering’ in a general way before restricting the entitlement to end it to only one cohort of sufferers. This matters. Consider, for example, that under a general definition of the term ‘suffering’, which can capture the experience of those who are terminally ill and those who are not, whether adults or minors, the restriction of an entitlement to assisted dying only to one of these cohorts can be difficult to justify. The parallel here is with Canada. When the Canadian Supreme Court in Carter v Attorney General of Canada 2015 SCC 5 invalidated the blanket ban on assisted dying under the Criminal Code, holding that it should be available for ‘competent adults who seek such assistance as a result of a grievous and irremediable medical condition that causes enduring and intolerable suffering’, the Canadian Parliament’s legislated restriction of the availability of that assistance to only those whose natural death had become ‘reasonably foreseeable’ was invalidated 3 years later by the Superior Court of Quebec in Truchon v Attorney General of Canada 2019 QCCS 3792, in part because the Supreme Court’s decision had identified a general cohort of those who were suffering from a ‘grievous and irremediable’ condition, which the Parliament had sought to narrow ([495]):
La Cour suprême ne limite pas ni ne restreint la possibilité de recevoir l’aide médicale à mourir aux seules personnes dont la mort naturelle est raisonnablement prévisible ou qui sont en fin de vie et ce ni explicitement ni implicitement.
By contrast, a legislative measure which both creates and prescriptively defines a right by reference only to a specific cohort of people does not narrow the right only to that cohort. This leads back to the point we make earlier – that those who are terminally ill and those who aren’t may not be relevantly similar for the purposes of Article 14, but this leads us to a much bigger point about how both the Convention and the Human Rights Act 1998 contemplate remedying breaches by positive law (covered further below).
The Article 2 argument
An alternative, much more blunt argument, put forward by Rajiv Shah on the UK Human Rights Blog, foregoes reliance on Article 14 and argues straightforwardly that allowing euthanasia would per se constitute a breach of Article 2 of the ECHR. The argument goes like this: the right to life under Article 2 is absolute and allows no exception; there is a negative obligation upon the UK not to take life; any euthanasia laws would necessarily involve the taking of life; therefore euthanasia laws would breach Article 2.
The problem is that the case law does not support this argument. Consider again Mortier v Belgium. Whilst a decision ultimately on the “particular circumstances” of the case at hand (para 140) it was “the first case in which the [European] Court ha[d] been called upon to examine the compliance with the Convention of an act of euthanasia” (para 115). There was before it an ample opportunity to “clarify the nature and scope of a State’s obligations under Article 2 of the Convention” in the context of euthanasia (para 115). The court set about “determining whether [euthanasia] can, in certain circumstances, be performed without contravening Article 2 of the Convention” (para 125). The Court determined that it could.
Shah contends that the European Court has “not had occasion to rule that the State itself doing it is in compliance with the negative obligation”, says Shah. This is not true. This was canvassed, and rejected, in Mortier. The Court had the opportunity to do what Judge Serghides did in his dissent: confirm that “no form of euthanasia or legal framework regulating such practice… can safeguard the right to life under Article 2 of the Convention” (Dissent, 5). It did not. The Court turning down an opportunity is not the same thing as it not having had an opportunity.
Shah has another string in his bow. We should be attracted to his argument – the same argument Judge Serghides made, and which was rejected by the majority of judges, in Mortier – because the existing case law, Mortier included, was concerned with positive obligations under the Convention. If Leadbeater’s Bill were to become law, this would require state agents to administer euthanasia directly, resulting in a breach of negative obligations under the Convention. The European Court has not ruled on whether euthanasia is compatible with a state’s negative obligations under the Convention. Whilst positive obligations are “nebulous”, negative obligations are construed by the Court in a much stricter fashion, and there is “no scope for reasonable disagreement” as to how negative obligations are framed (Shah).
Let us put to one side for a moment the fact that Shah suggests, without evidence, that euthanasia in most countries is “privately delivered”. Belgium’s healthcare system is complicated, and cannot be straightforwardly dismissed as a private healthcare system, to which the negative obligations under the Convention could not conceivably apply. In any case, the Court in Mortier scrutinised and upheld legislation which applies to “doctors” generally (see para 51) without trouble. Its choice of examining the complaint via the lens of positive obligations was not because negative obligations were inapplicable to that case. More fundamentally, even if we discard the Belgian state as some kind of libertarian outlier, there are many other countries subject to the jurisdiction of the ECHR which have fully, or partially, public healthcare systems, and which administer state-sanctioned euthanasia in at least some circumstances, such as Spain and Luxembourg. Moreover, the Court’s reference in Mortier to “doctors” in the general sense aligns with much of the Court’s jurisprudence around obligations on healthcare professionals in the context of a State duty (such as in cases of clinical negligence, see Lopes de Sousa Fernandes v Portugal, App No. 56080/13, judgment of 19 December 2017) without any sort of distinction based on whether the healthcare system in question is privately or publicly funded.
But let us suppose that the UK really is different when it comes to administering euthanasia. Does this mean, as Shah contends, that there is “no scope for reasonable disagreement” as to how negative obligations in relation to euthanasia are framed? That is nonsense. To use the example of another issue which engages controversial issues under Article 2, the Court has frequently deferred to states on the question of when life begins, and how, if at all, abortion services should be regulated: see A, B and C v Ireland (App No 25579/05), Judgment of 16 December 2010. The political and ethical difficulties involved, as well as the apparent lack of European consensus on such matters, ultimately led the court to defer this question to individual states (including, crucially, in circumstances where state agents would be involved in the abortion process, potentially engaging a state’s negative obligation). If the question of when and how life begins can be left to states to determine, so can the question of when and how life ends.
Remedying breaches by positive law
Under the HRA, a declaration of incompatibility can be remedied by way of a specific kind of secondary legislation enabled under section 10. Although this is a classic Henry VIII power – in the sense that it is explicitly permitted to amend primary legislation – the process around it is inherently dialogic, consuming both time and effort. Schedule 2 of the HRA sets out a number of conditions for making remedial orders: the draft of such an order has to be approved by resolution of each House of Parliament within a period of 60 days from when the draft is laid before the relevant House, with any representations on that draft made within the 60-day period also required to be laid in summary form, with any resultant changes on the draft also laid for Parliament’s consideration. The only circumvention of this whole process is when a remedial order is declared urgent. In this case, the person making the order is still required to lay the order before Parliament, together with any representations and resultant changes to the order within a 60-day period following the making of the order. Moreover, urgent remedial orders must be confirmed by resolution of each House of Parliament within 120 days of the order having been made, otherwise these urgent orders cease to have any effect. In his amended post in the UKCLA, Murray issues the following warning in relation to remedial orders:
…the government could amend the law and extend it beyond its limits without requiring Parliament to pass a new Act of Parliament (although the government’s remedial order would be subject to approval by resolution of both Houses, in accordance with the process set out in Schedule 2). Sliding down the slippery slope is perfectly possible as a matter of law.
This description, with respect, barely covers the detail of making remedial orders set out above, to say nothing of the reality of such detail. Let us consider the example of the remedial order process involving the extension of eligibility to bereavement payments under the Fatal Accidents Act 1976. The draft order, laid (non-urgently) in May 2019, was accompanied by reasons as required under the HRA. The Joint Committee on Human Rights accepted the Government’s reasons for making the remedial order (rather than proceeding by way of a Bill) but not before criticising how long it took for the Government to move on an issue flagged by the Law Commission two decades prior, and on which the Court of Appeal had made a declaration two years before the draft order had been laid. If the level of parliamentary scrutiny over the remedial order process for an issue which is, on any view, significantly less politically controversial than assisted dying is any indication, any remedial order made to extend the entitlement to assisted dying will certainly be pored over by the JCHR.
This is even before we come to the question of whether the Government would (or could) easily make such a remedial order, purporting to extend the scope of primary legislation beyond itself. The HRA is silent on the issue, except permitting remedial orders to ‘have the same extent as the legislation which it affects’, though this may apply to territorial extent (as the term ‘extent’, as a statutory subheading, often refers to territorial extent) rather than substantive extent. A remedial order which amends primary legislation by extending its substantive reach may well be possible as a matter strictly of legal theory, but it would be fraught with political consequences to which Parliament is especially alive.
All of this is to question the premise of Murray’s ‘slippery slope’. The identification and mitigation of risk may be a necessary part of lawmaking, especially where people’s lives are concerned. But that risk should have a basis in reality. If, as the above exploration shows, obtaining a declaration of incompatibility is just as much of an uphill struggle as making the remedial order, in what reality do we ‘slide’ down the slippery slope?
Conclusion
The legal debate on euthanasia is heated and complicated. Politicians, campaigners, and interested members of the public may well be swayed by arguments on either side. They may agree with the “anti” camp, whose strongest arguments relate to the sanctity of life and potential scope for abuse. They may align themselves with the “pro” camp, whose strongest arguments relate to the dignity of the person and the alleviation of unnecessary suffering. But nobody should be persuaded by the argument that we should refrain from liberalising our euthanasia laws because doing so would put us in breach of the ECHR. It would not.
Anurag Deb is a PhD candidate at Queen’s University Belfast and a paralegal at KRW LAW LLP. Lewis Graham is a fellow in law at Christ’s College, Cambridge. The authors would like to thank Stevie Martin for comments on a draft.
The post No, legislating to allow euthanasia would not breach the European Convention on Human Rights appeared first on UK Human Rights Blog.