Little over a year ago, I wrote a blog post discussing the Advisory Opinion No. 1 issued under Protocol No. 16, with the title “The mountain gave birth to a mouse”. Last week, the Grand Chamber of the European Court of Human Rights give birth to Advisory Opinion No. 2 “concerning the use of the ‘blanket reference’ or ‘legislation by reference’ technique in the definition of an offence and the standards of comparison between the criminal law in force at the time of the offence and the amended criminal law”, requested by the Armenian Constitutional Court. After a brief examination, my preliminary conclusion is that, this time, we’re dealing with a slightly bigger rodent – perhaps a rabbit but certainly not a capybara. In this contribution, I will discuss the new Advisory Opinion, focusing in particular on what further lessons can be drawn from it. The key takeaway from this blog post, compared to its predecessor, is that it provides a somewhat more meaningful contribution to the development of the case law. However, the messages given by the Court, regarding the kinds of questions it wants to respond to, are unlikely to spark a lot of enthusiasm among domestic courts to make use of the Advisory Opinion mechanism, potentially thwarting Protocol No. 16’s objective of encouraging judicial dialogue.
But first a short recap of the mechanism. Protocol No. 16, which entered into force in August 2018, has so far been ratified by fifteen States. It allows the ‘highest domestic courts’ to request the ECtHR to give an Advisory Opinion on ‘questions of principle relating to the interpretation or application of the rights and freedoms defined in the Convention or the protocols thereto’. The objective of the Protocol is to ‘further enhance the interaction between the Court and national authorities and thereby reinforce implementation of the Convention, in accordance with the principle of subsidiarity’.
Now what is the case about? Former Armenian President Robert Kocharyan stood trial for the offence of overthrowing the constitutional order (Article 300.1 § 1 of the 2009 Criminal Code) for facts related to the declaration of a state of emergency and the violent squashing of post-election protests in 2008, which had resulted in the death of eight civilians and two law-enforcement officers. In the context of this trial, the First-Instance Court of General Jurisdiction of Yerevan had asked preliminary questions to the Armenian Constitutional Court on the compatibility of the offence in question with the nullum crimen sine lege principle and the principle of the non-retroactivity of the criminal law. With its request for an Advisory Opinion, the Armenian Constitutional Court basically asked guidance from the European Court of Human Rights on how to assess the compatibility of this offence with Article 7 ECHR.
Advisory Opinion No. 2 starts off with a number of preliminary considerations. Firstly, while the Court in Advisory Opinion No. 1 had slightly modified the questions, it now explicitly considered itself to have ‘the power to reformulate questions asked by the requesting Court having regard to the specific factual and legal circumstances at issue in the domestic proceedings’ and ‘that, similarly, it may also combine certain questions asked by the requesting court’ (§ 45). Secondly, it considers that the mere fact that the Grand Chamber panel decided to accept a request for an advisory opinion does not ‘preclude the Grand Chamber from assessing whether each of the questions composing the request fulfils the requirements of Article 1 of Protocol No. 16’, including in light of the requirement that an Advisory Opinion ‘must be confined to the points that are directly connected to the proceedings pending at domestic level’ (§ 47):
‘Should it come to the conclusion, taking due account of the factual and legal context of the case, that certain questions do not fulfil these requirements, it shall not examine these questions and will make a statement to this effect in its advisory opinion.’
Thirdly, the Grand Chamber acknowledges the specific difficulties in answering a question in a ‘double referral’ procedural setting – a situation which may arise in any State that provides for a preliminary constitutionality question procedure. In the case under consideration, this implies that the ECtHR must ‘proceed on the basis of the facts as provided by the Constitutional Court, albeit those facts may be subject to subsequent review by the first instance court’ (§ 49). This is to be contrasted with Advisory Opinion No. 1, where ‘information as to the precise factual circumstances underlying the legal questions raised in the advisory opinion request was available to the Court’ (§ 49). The higher degree of factual uncertainty underlying Advisory Opinion No. 2 however inevitably turns the Court’s task into a somewhat more abstract exercise, as the concrete impact of the Advisory Opinion on the domestic proceedings is more hypothetical. Based on the guidance given by the Court in its Advisory Opinion, it will then be for the Armenian Constitutional Court to address the question regarding the constitutionality of Article 300.1 of the 2009 Criminal Code, the answer to which will then ultimately have to be applied by the First-Instance Court in the context of the underlying criminal proceedings (§ 50).
On questions answered
Before turning to a discussion of the questions the Court declined to answer (questions 1 and 2), and which raise broader issues regarding the purpose of the Protocol No. 16 mechanism, I will discuss the questions it did answer (questions 3 and 4). The third question was worded as follows:
‘3) Does the criminal law that defines a crime and contains a reference to certain legal provisions of a legal act with supreme legal force and higher level of abstraction meet the requirements of certainty, accessibility, foreseeability and stability?’
This basically concerns the question as to the Article 7 ECHR compatibility of the so-called ‘blanket reference’ or ‘legislation by reference’ technique – defined by the Court as ‘the legislative technique where substantive provisions of criminal law, when setting out the constituents elements of criminal offences, refer to legal provisions outside criminal law’ (§ 31). In the present context, the latter provisions are those provisions enshrined in the Armenian Constitution, which set out a number of fundamental rule of law and separation of power principles. In the cases of Kuolelis and Others v. Lithuania and Haarde v. Iceland, the Court had implicitly accepted that the ‘legislation by reference’ technique may be compatible with the requirement that criminal provisions must be sufficiently clear and foreseeable in their application to be compatible with Article 7. In its Advisory Opinion, the Court explicitly affirms this position, which it moreover supports by comparative-law material highlighting the wide use of the technique by Council of Europe Member States, in particular as regards criminal offences against the constitutional order of their country (§ 70-71 and 74). The Court however emphasizes that the criminal law defining an offence using the ‘blanket reference’ technique must comply with the “quality of law” requirements, i.e. it must be sufficiently precise, accessible and foreseeable in its application:
‘Given that the referenced provision becomes part of the definition of the offence, both norms (the referencing and the referenced provision) taken together must enable the persons concerned to foresee, if need be with the help of appropriate legal advice, what conduct may make them criminally liable.’
Quite ambiguously, the Court then concludes that it
‘considers that the most effective way of ensuring clarity and foreseeability is for the reference to be explicit, and for the referencing provision to set out the constituent elements of the offence. Moreover, the referenced provisions may not extend the scope of criminalisation as set out by the referencing provision. In any event, it is up to the court applying both the referencing provision and the referenced provision to assess whether criminal liability was foreseeable in the circumstances of the case.’ (§ 73, also see § 74)
One is left to wonder what the added value is of the Court emphasizing what it considers to be ‘the most effective way of ensuring clarity and foreseeability’, which suggests that there may be other ways as well, without however specifying what conditions would have to be met in the latter case. Basically, the Court sends the message to the domestic court that it should figure this out for itself.
Next, the Court addresses the fourth question asked by the Armenian Constitutional Court, which reads as follows:
‘In the light of the principle of non-retroactivity of criminal law (Article 7 § 1 of the Convention), what standards are established for comparing the criminal law in force at the time of committal of the crime and the amended criminal law, in order to identify their contextual (essential) similarities or differences?’
This question basically boils down to the question regarding how to determine whether a criminal law is more lenient when applying the principle of the retrospective application of the more lenient criminal law (recognized for the first time in Scoppola v. Italy (no. 2)). At the domestic level, prosecution took place on the basis of the new Criminal Code from 2009, rather than on the basis of Article 300 (usurpation of State power) of the former Criminal Code, which had been in force at the time of the events in question. At the outset, the ECtHR admits that its case law ‘does not offer a comprehensive set of criteria for comparing the criminal law in force at the time of commission of the offence and the amended criminal law’ (§ 86). Based on the principle that the foreseeability of the law has to be assessed having regard ‘to the domestic’s court’s case law, if any exists, elucidating the notions used in the law in force at the time’ (§ 86), the Court holds that:
‘the comparison between the criminal law in force at the time of the commission of the offence and the amended criminal law has to be carried out by the competent court, not by comparing the definitions of the offence in abstracto, but having regard to the specific circumstances of the case.’ (§ 88)
According to the Court, this ‘principle of concretization’ (§ 92) implies that the domestic courts should assess in concreto 1) ‘whether all constitutive elements of the offence and other conditions for criminality were fulfilled under the provisions of the Criminal Code in the version in force at the time of the impugned events’ and 2) whether application of the new criminal law provision ‘would attract more serious consequences for the accused’ than the application of the old one (§ 91). If the former question is answered in the negative or the latter in the positive, the new provision may not be applied in the case. As with the third question, the Court thus gives the message to the domestic courts to figure out for themselves how to answer these questions, but at least now it provides more guidance as to how these courts should make this assessment.
On questions discarded
Now let’s turn to questions 1 and 2:
‘1) Does the concept of ‘law’ under Article 7 of the Convention and referred to in other Articles of the Convention, for instance, in Articles 8-11, have the same degree of qualitative requirements (certainty, accessibility, foreseeability and stability)?
2) If not, what are the standards of delineation?’
Taking into account the legal context of the domestic proceedings, the Court ‘finds it difficult to see which questions the Constitutional Court wishes to determine with the help of the Court’s opinion’. Therefore, it considers that:
‘The Court’s answer to the Constitutional Court’s first and second questions would be of an abstract and general nature, thus going beyond the scope of an advisory opinion as envisaged by Protocol No. 16. In particular, it does not appear possible to reformulate the questions so as to allow the Court to confine its advisory opinion to “points that are directly connected to the proceedings pending at domestic level”’.
The Court does make the caveat that, in so far as the questions concern the issues of legal certainty and foreseeability under Article 7, ‘these can be addressed sufficiently in the Court’s answer to the third question’ (§ 55). The Court therefore decides not to answer these questions because it considers that doing so would not be in accordance with the purpose of Protocol No. 16 (§ 56).
One may, however, wonder whether rejecting questions for being formulated at a too high level abstraction does not go against the objective of Protocol No. 16 to encourage judicial dialogue in a spirit of subsidiarity. From such a perspective, it would be more logical to leave it up to the requesting court to assess the relevance of the Court’s answer for the pending domestic proceedings. Relatively ‘abstract and general’ questions of this type, pertaining to the overall interpretation of key ECHR principles, could certainly contribute to providing guidance to domestic courts. Moreover, there is nothing in the text of the Protocol nor in the Explanatory Report thereto that prevents the Court from answering such questions. All Article 1, § 2 of Protocol No. 16 requires is that an opinion is sought ‘in the context of a case pending before it’. There is no mention on whether the Court should consider those questions to be related to ‘points that are directly connected to the proceedings pending at domestic level’. In relation to this provision, the Explanatory Report only mentions that ‘[t]he procedure is not intended, for example, to allow for abstract review of legislation which is not to be applied in that pending case’. Clearly, abstract review of domestic legislation is something wholly different than abstract guidance on the interpretation of the ECHR, which is what the Armenian Constitutional Court was seeking in this case.
All of this is even more remarkable given the fact that the Court, as it indicated itself, could have easily reformulated the questions to discuss them together with the third question if it didn’t feel like addressing them at a more abstract level. Thus, it rather seems that the Court deliberately intended to set a precedent with a view to communicating to domestic courts that it is not interested in receiving overly abstract questions on the interpretation of the ECHR. At the same time, it however also emphasizes that ‘[t]he aim of the procedure is not to transfer the dispute to the Court’ (§ 43), excluding overly concrete questions as well. What is left is some residual intermediary category that is abstract yet concrete or concrete yet abstract, of which the boundaries remain far from clear. The lack of clarity on the self-defined scope on the abstractness/concreteness scale of the Court’s advisory jurisdiction in combination with its willingness to second-guess the domestic courts’ assessment of the relevance of particular questions, is unlikely to encourage domestic courts to engage with the Protocol No. 16 mechanism. This holds even more true given the time-consuming character of the procedure – in this regard, Ad Hoc Judge Sarvarian in his Concurring Opinion deplores the fact that the Court does not go further than it did in its opinion, being ‘especially cognizant of the fact that these advisory proceedings have necessarily occasioned delay of some ten months to the proceedings before the Constitutional Court and by extension to the underlying trial’.
Arguably, the Court’s answers to the third and fourth questions provide more of a contribution to the development of its jurisprudence than Advisory Opinion No. 1 did regarding the question of surrogacy. However, given the fact that the Court’s response ultimately boils down to the fact that the domestic courts should figure out the relevant Convention issues for themselves, I cannot but come to the same conclusion regarding Advisory Opinion No. 2 than I did in my blog post on Advisory Opinion No. 1: one may wonder whether it will encourage the highest domestic courts to seek this kind of guidance of the Strasbourg Court, which is particularly regrettable given the fact that procedure demands a lot of the Court’s time and resources. Let’s just hope the next Advisory Opinion will turn out to be more of a capybara instead.