B. Trumping Internal Consensus

European consensus is not entirely what it says on the tin. Rather than describing a literal ‘consensus’ position on a particular question between all of the member States to the Convention, ‘European consensus’ as deployed by the Strasbourg Court might be more accurately described as an identifiable trend or commonality among at least some members of the Council of Europe in relation to the matter at hand.Footnote ¹¹ Once identified, European consensus is usually deployed to determine whether the Convention can be said to provide certain protection within the banner of Convention rights, where such protection is not expressly mentioned in the Convention itself (eg LGBT rights being given protection under the banner of Article 8 in spite of their unenumerated statusFootnote ¹²). Usually the question of European consensus arises only in situations where it is not clear whether or not the Convention protects individuals in the way that is claimed and, as a corollary, whether the State action impugned violates the Convention for failing to vindicate the right in question.Footnote ¹³ Moreover, the Court has tended not to apply European consensus if the Contracting States had commenced drafting a new Protocol to deal with the matter under consideration.Footnote ¹⁴ In the cases where European consensus has been deployed to date the Court has reached one of three conclusions: 1) that there is a lack of European consensus, usually meaning that the impugned law is Convention-compatibleFootnote ¹⁵ unless manifestly unreasonable, disproportionate or unnecessary in a democratic society; 2) that there is an identifiable European consensus of which the impugned law is part and, therefore, that the law is Convention-compatible;Footnote ¹⁶ or 3) that there is an identifiable European consensus with which the impugned law is incompatible, usually meaning that it is Convention-incompatible. Our concern is with the third of these scenarios.

Where a European consensus is identified and a respondent State's law falls outside of it without reasonable justification, the Court normally finds a violation of the Convention.Footnote ¹⁷ Whether or not divergence from the European consensus is justified is, of course the core question in such situations. The Court has identified the particular historical and social context extant in a respondent State as a justification for divergence, although the longer the divergence persists the more strongly it has to be justified, and in fact no case has yet been found to reach the threshold for justification under this ground.Footnote ¹⁸ Divergence from European consensus can also be justified on the basis of an incompatible international trend to which the impugned State action might be said to aligned, although that is rare indeed.Footnote ¹⁹ Finally—and of key concern in this article—divergence from the European consensus might be justified on the basis of an internal consensus that ‘trumps’ that found at a regional level. Internal consensus can be defined as a widespread (although not unanimous) attitude to a legal issue that is held by the majority of people within the respondent State. In net terms the internal consensus claim is twofold: first that in fact an internal consensus exists and can be verified; and second that the internal consensus is sufficient to justify divergence from (ie to trump) the identified European consensus and, as a result, that there is no incompatibility with the Convention.

It is rare for States to try to argue for trumping internal consensus, and in fact no such argument had succeeded prior to A, B & C. The internal consensus argument was deployed by the British Government in Tyrer v the United Kingdom. Footnote ²⁰Tyrer concerned the long-standing practice of sentencing people to ‘birching’ in criminal trials in the Isle of Man.Footnote ²¹ Tyrer—a juvenile who had been subjected to this punishment—claimed that it contravened Article 3 of the Convention, which protects against torture, inhuman and degrading treatment or punishment. The UK claimed, inter alia, that corporal punishment ‘did not outrage public opinion in the Island’Footnote ²² and ought to be saved from Convention incompatibility on that basis. The Court disagreed, holding:

Although the Court in Tyrer did not entirely rule out the prospect that an internal consensus might be relevant in analysing whether a particular State action contravenes the Convention, it clearly rejected any such argument in this case. This was based on both the nature of the right in question as an absolute right and on the European consensus as regards corporal punishment. In the latter connection, the Court held that it ‘cannot but be influenced by the developments and commonly accepted standards in the penal policy of the member States of the Council of Europe in this field’.Footnote ²⁴ Thus, even if the internal consensus existed, which the Court doubted, it seems that it was not an internal trumping consensus. Similar claims arose in the case of Dudgeon v the United Kingdom.Footnote ²⁵ Dudgeon claimed that the continuation in force of ‘the abominable crime of buggery’ as prohibited by sections 61 and 62 of the Offences against the Person Act 1861 was a violation of the Convention. The UK argued that continued criminalization of male homosexual sex in Northern Ireland did not contravene the Convention based on what it claimed was ‘the strength of feeling in Northern Ireland against the proposed change, and in particular the strength of the view that it would be seriously damaging to the moral fabric of Northern Irish society’ to decriminalize such acts.Footnote ²⁶ The UK further argued that Northern Ireland was a particularly conservative society that placed greater emphasis than most Convention States on religion, even as applied to heterosexual conduct.Footnote ²⁷ The thrust of this argument was that even if the European consensus was for the decriminalization of consensual homosexual sex between men, Northern Ireland contained an internal trumping consensus that ought to save the law in that jurisdiction from incompatibility. Although the Court accepted that these claims were relevant to the matter at bar, it did not agree that there existed any internal trumping consensus. Instead, the Court held that the Convention was violated, finding:

Neither Tyrer nor Dudgeon categorically ruled out the possibility that an internal trumping consensus could exist, but in neither case was the claim successful. This is significant, not least because in Dudgeon at least what was in question was a matter of significant moral debate and disagreement. Dudgeon was decided in 1981 when homosexual sex was unlawful in many countries and only one country in the world—Norway—had a law protecting people from discrimination on the basis of sexual orientation (and that itself was introduced in 1981).Footnote ²⁹ This suggests that even in an area of moral sensitivity an alleged internal trumping consensus required hefty supporting materials in order to succeed. The same proved not, however, to be the case in A, B & C v Ireland.Footnote ³⁰

C. Trumping Consensus in A, B & C v Ireland

A, B & C v Ireland concerned the extremely restrictive abortion regime in place in Ireland and claims of its incompatibility with the Convention. Although abortion is permitted under the Irish Constitution where the life of the mother is at riskFootnote ³¹ (including from suicideFootnote ³²), no legislation has ever been introduced regulating the mechanism by which medical professionals can determine whether or not a woman's life is endangered and certify that an abortion is constitutionally permissible.Footnote ³³ Furthermore, serious risk to the life (as opposed to the health) of the mother is the only situation in which an abortion can be legally acquired in Ireland, although travel to another jurisdiction in order to procure an abortion is permitted and information relating to abortion is freely available.Footnote ³⁴ The claimants in A, B & C challenged this abortion regime, claiming firstly that it violated the Convention because there was no clear and safe way of determining whether someone was entitled to an abortion within the existing constitutional limits, and secondly because it was claimed that the restrictiveness of the regime itself violated the Convention partially because it was out of step with the prevailing position across the Council of Europe States. In relation to the former, the Court reiterated its earlier case lawFootnote ³⁵ that where abortion is legally permitted there must be a mechanism to acquire it and found a violation on that basis.

In relation to the latter claim, the European Court of Human Rights confirmed its long-standing position that abortion was an area in relation to which member States enjoyed a significant margin of appreciation,Footnote ³⁶ but then proceeded to state that ‘the question remains whether this wide margin of appreciation is narrowed by the existence of a relevant consensus’.Footnote ³⁷ In other words, was the extreme restriction of abortion in Ireland to cases where the pregnancy and the woman's life were incompatible acceptable under the Convention, or did it require that abortion would be more broadly available including especially where the pregnancy endangered the health of the woman? The Court confirmed that there was a consensus amongst a substantial majority of the Contracting States of the Council of Europe towards allowing abortion on broader grounds than that permitted under Irish law,Footnote ³⁸ but held that this consensus did not decisively narrow the State's margin of appreciation.Footnote ³⁹ This was because the limited availability of abortion in Ireland was said to be based on the ‘profound moral views’Footnote ⁴⁰ of the Irish people, which constituted a trumping internal consensus. In this respect, internal consensus and European consensus were applied as competing variables determining the scope of the margin of appreciation afforded to the respondent State.

This conclusion was based on the argument proposed by the Irish government that the results of three constitutional referenda on abortion,Footnote ⁴¹ together with public demands for guarantees about abortion in the MaastrichtFootnote ⁴² and LisbonFootnote ⁴³ referenda,Footnote ⁴⁴ evidenced an internal consensus of a profound nature. As discussed further below, the methodology applied in reaching this decision is questionable, but notwithstanding that the Court concluded that ‘the impugned restrictions in the present case were based on profound moral values concerning the nature of life which were reflected in the stance of the majority of the Irish people against abortion during the 1983 referendum and which have not been demonstrated to have relevantly changed since then’.Footnote ⁴⁵ Although there was a European consensus inasmuch as Ireland is one of only a few member States to restrict abortion to situations where the life, as opposed to the health, of the mother is at risk,Footnote ⁴⁶ the Court found that—as previously held in Vo v France Footnote ⁴⁷—there is no legal or scientific consensus in Europe as to when life begins.Footnote ⁴⁸ This seems to have been perceived as a factor to weaken the probative nature of the European consensus, further highlighting the vulnerability of the European consensus to being trumped by internal consensus. The Court stated:

Here the Court is dancing around—although never precisely pinning down—the relationship between degree of abortion availability and views on the moment that life begins. A more precise analysis is found in the concurrence by Irish ad hoc judge, Judge Finlay-Geoghegan. She makes the connection inasmuch as she sees it between Vo and the decision in A, B & C more clear by holding that the European consensus was not one as to when life began and did not therefore go to the core public policy concern underpinning the Irish position. In other words, the two consensuses (European v internal) were, in her view, related to two different questions (how available abortion is (European) v when life begins (Irish)). As a result, it did not operate to narrow Ireland's margin of appreciation.Footnote ⁵⁰ Both this concurrence and the decision of the majority arguably conflate questions about when life begins (a question that fundamentally goes to whether abortion should be permitted at all and, if so, how far into pregnancy) and circumstances in which abortion should be permitted for reasons connected to maternal welfare (where life is at risk only, or where life or health is at risk).Footnote ⁵¹ Indeed, much the same point was made in the dissenting judgment.Footnote ⁵²

The application of a trumping internal consensus was criticized by Judges Rozakis, Tulkens, Fura, Hirvelä, Malinverni and Poalelungi dissenting, who wrote:

The danger of this departure is, we contend, largely bound up in the potential risk that it poses to the continuing development of a rights-based, constitutionalist public order throughout the European continent should the notion of trumping internal consensus not be left to perish on the rock of A, B & C.

D. Trumping Consensus and Frustrated Constitutionalism

Although internal consensus has only been allowed to ‘trump’ European consensus in the Strasbourg jurisprudence on one occasion, the phenomenon raises serious concerns. These concerns are deeply linked to the constitutionalist mission and identity of the Court. Even though the European Court of Human Rights is a not a supreme court for Europe, or even a constitutional court in the traditional sense, it does have an important constitutionalist function and identity.Footnote ⁵⁴ This is reflected in the fact that, at its heart, the Court's function is to ensure compliance with, and the development of, the Convention in a manner that contributes to the cultivation (together with the Court of Justice of the European Union) of an ordre publique for the continent in which basic minimum rights-based constitutionalist requirements are elaborated and safeguarded through domestic implementation and regional supervision.Footnote ⁵⁵ In order to succeed in its constitutionalist mission the Court needs the same resources as other constitutionalist courts, including legitimacy.Footnote ⁵⁶ Dzehtsiarou has previously argued that the concept and methodology of European consensus is itself a legitimacy-enhancing measure, enhancing its standing in the eyes of the member States, provided it is transparently and accurately constructed and consistently deployed.Footnote ⁵⁷ Furthermore, the impact that a judgment or series of judgments has on the lived experiences of those residing within the espace juridique Footnote ⁵⁸ of the Convention has implications for the extent to which the people covered by the Convention consider it and its court to be legitimate. Across each of these legitimacy planes (ie member States and protected individuals) trumping internal consensus has potentially troublesome and delegitimizing effects. Together, these effects undermine the constitutionalist impact and capacity of the Court.

In A, B & C trumping internal consensus allowed for the continuation of the status quo in Ireland, as desired by the Irish government itself. It is to be expected that this would in fact always be the result where a trumping internal consensus is found. In that respect it is likely that the respondent State in question would consider the use of trumping internal consensus to be a legitimizing factor for that particular judgment, taking into account the supposedly unique and deeply engrained national sentiment in favour of the position as it stands. Indeed, failure to take what is considered to be ‘proper’ account of the sentiment and tradition within a member State has, on occasion, been the basis for sharp criticism and allegations of illegitimacy levied against decisions emanating from Strasbourg in the past.Footnote ⁵⁹ Among the other member States, however, trumping internal consensus potentially undermines the legitimacy and effectiveness of the Court and the Convention more broadly.

First of all it may act to ‘fudge’ the question of what a particular provision of the Convention requires of states and of whether (and, if so, under what circumstances) a State can provide a level of protection lower than that apparently required but still remain within the bounds of the Convention. In other words, States may wonder what exactly it is that the Convention requires of them and whether they can themselves construct a compelling enough case for an internal consensus that would allow for them to ‘trump’ the general European position. Furthermore, a State that currently forms part of the European consensus as identified by the Court may in fact have in place a certain legal or regulatory regime precisely because it considered that to be required in order to be Convention compliant, even if it was not particularly desired. In other words, there may be situations where the Convention has had precisely the exogenous force on domestic legal and regulatory systems that was desired (ie it has resulted in domestic legal change ‘upwards’ even without a judgment against the State), but the identification of a trumping internal consensus in one State may either foster resentment in other States or result in other States recalibrating laws ‘downwards’ on the basis of a self-declared trumping internal consensus. That is arguably more likely to happen in relation to matters of particular moral and social controversy—such as abortion—where campaigners on a particular side of the debate might argue that in fact the Convention does not require as liberal a legal regime as exists and in fact that further restricting individual freedoms would be permissible based on the case in which a trumping internal consensus was established.

Second, the notion that European consensus might be subordinate to internal consensus within any particular State might bolster broader arguments that the Convention ought to be subordinate to decisions taken at the national level. This would likely undermine the effectiveness and harmonizing capacity of judgments emanating from the Strasbourg Court. At its core, an argument of trumping internal consensus is one that makes the nature and content of an international legal obligation subject to public sentiment and public acceptance of obligations within the contracting party, notwithstanding the fact that individuals are not full subjects of international law. This adds to the counterintuitive nature of the proposition that commonly accepted minimal human rights standards would be subordinate to public sentiment within a respondent State.

Thus, trumping internal consensus has important implications for the harmonization mission of the Convention and, by corollary, the Court itself. Indeed, the dissenting judges in A, B & C v Ireland recognized as much themselves when they held:

At a more basic level, trumping internal consensus as applied in A, B & C v Ireland raises serious methodological concerns. The occasional incoherence and incomplete reasoning in decisions of the Court have often been remarked upon.Footnote ⁶¹ Indeed, some decision-making mechanisms or principles of the Court, including the margin of appreciation and European consensus, have been especially singled out for criticism.Footnote ⁶² Even those who argue that European consensus has an important legitimizing role to play in the canon of the Convention have accepted that it is methodologically problematic.Footnote ⁶³ However, at least with European consensus the laws and policies of a number of States are identified and taken into account with some degree of transparency; where an internal consensus is identified the methodological concerns raised are arguably more acute. Unlike with comparative political and legal analysis—which is essentially the method used for European consensus—internal consensus tends to be identified through a crude analysis of sentiment. This was certainly the case in A, B & C v Ireland.

The trumping internal consensus in A, B & C was—as mentioned above—based on an allegedly profound moral value of the Irish people. This was established by reference to the outcomes of three hotly disputed constitutional referenda and demands made in the somewhat notoriously fraught context of EU-treaty referenda. The Court accepted that the net result in these referenda could be read as communicating accurately the position of the Irish people, ie that they were happy with the abortion regime as it stands in Ireland. However, a closer mining of the materials relied upon by the Court tells a somewhat more complex story. In 1983, the 8th Amendment to the Constitution (protecting the right to life of the unborn with equal regard to the right to life of the mother) was passed by a ‘yes’ vote of 66.45 per cent (66.9 per cent of the valid poll). Inasmuch as it concerned the availability of abortion in Ireland, the referendum in 1993 saw 62.27 per cent of voters who turned out (and 65.35 per cent of the valid poll) reject a proposal to restrict abortion to a greater degree than that determined by the Supreme Court in Attorney General v X (in which the risk of suicide was found to permit of abortion under the terms of the Constitution).Footnote ⁶⁴ The people had been asked to amend the Constitution to make it clear that a risk to the life of the mother emanating from the possibility of suicide would not be sufficient to make abortion constitutionally permissible; they were not asked whether they wanted to make abortion more liberal. Thus, the ‘no’ vote in that referendum was a simple refusal to further restrict abortion in Ireland. The same was true in the 2002 constitutional referendum, when 50.15 per cent of the people who voted (a mere 42.89 per cent turnout) refused once more to restrict abortion further (equating to 50.42 per cent of the valid poll).Footnote ⁶⁵ Rather than standing for the proposition that the Irish people have a deeply held or profound moral position that supports the status quo, as claimed by the Court, these referenda simply stand for the proposition that the Irish people have not been willing to support the amendments to the Constitution as presented to them in these referenda. The European Court of Human Rights also suggested that the demands for guarantees from the EU in relation to abortion in the Maastricht and Lisbon referenda reinforced the profound moral position advocated by the Irish government, however the most that could be said about these guarantees is that sufficient concern was expressed or perceived in these campaigns that the EU might force a change to Irish abortion law (in contrast to a domestically initiated change) as to make the government believe such guarantees were required.

It is perhaps too much to expect an international court—or perhaps even any court—to be able to take into full account the socio-political context that gives real meaning to referendum results, but if that is the case then perhaps the best course of action for the Court is not to read any meaning into these results at all outside of their role in determining the current legal position. Otherwise, as was the case in A, B & C v Ireland, the Court will pronounce a moral position that is suspect (at best) to the people of the country in question and who are alleged to hold this position.

If the European Court of Human Rights continues to allow internal consensus to trump European consensus it runs the risk of encountering unanticipated consequences within the respondent States themselves. This is fundamentally because, at least as deployed in A, B & C v Ireland, trumping internal consensus acts as a fudge allowing the Court to escape laying down the exact parameters of Convention rights in contentious areas. Rather than clearly stating that the Convention does or does not recognize a right to access abortion where a woman's health is in danger, the Court held that Ireland could determine the scope of its own abortion regime but that once such a decision had been taken appropriate systems for accessing the services had be put in place. Absent those systems there was a violation of the Convention. Rather than set down a clear minimum standard of rights protection, which is the core function of the Court and Convention, the Court thereby placed abortion firmly within the domestic sphere and left apparently complete discretion to the State. The immediate reaction in Ireland was for pro-choice campaigners to call for legislation regulating abortion in line with the limited constitutional exemptionFootnote ⁶⁶ and for pro-life campaigners to call for a new referendum to further restrict the availability of abortion.Footnote ⁶⁷ The Strasbourg Court's decision left both options open; without a constitutional change the decision does seem to require regulatory action, but it does not preclude closing abortion off even further, perhaps even introducing an absolute ban on abortion in Ireland, which (as it would have to be done by referendumFootnote ⁶⁸) could subsequently be represented as evidence of an internal consensus based on profound moral values. Any attempt to row back on the trumping internal consensus approach taken in A, B & C would be difficult to justify in such circumstances as—unlike in cases of evolutive jurisprudence using European consensus—the ‘consensus’ deemed most potent in the area of abortion now seems to be an internal, rather than an external, one.