Horatia Muir Watt is Professor of Law at Sciences Po Law School
I. Framing the child-return issue. Several recent cases handed down by the two European Courts appear to be opening new vistas for conflicts of laws, in which human rights play a large part. The cases are well-known (ECJ/CJUE Aguirre v Pelz 2010; ECtHR Sneersone & Sneersone & Kampanella v. Italy 2011, Povse v. Austria 2013). They concern cross-border child abduction, and, more specifically, “fast-track” orders for the return of the abducted child, made by the (national) court of the child’s pre-abduction residence under article 11 (8) of Regulation Brussels II bis. This provision was designed to avoid the effect of delaying tactics by the abducting parent, which were progressively becoming systematic by virtue of article 13(b) of the 1980 Hague Convention (allowing the authorities of the country to which the child has been abducted, to refuse exceptionally to order the return if to do so would be to expose that child to a serious risk of harm). To this end, the fast-track return order is immediately enforceable, notwithstanding the resistance of that local court (hereafter, the court of the “country of refuge”). The difficulty, addressed partially by each of the cases above, concerns potential collision between the “notwithstanding” provision of article 11 (8) and with both procedural (6-1 ECHR, including the right of the child to be heard; article 24 EU Charter) and substantive (article 8 ECHR) human rights requirements.
This situation is particularly complex because it involves the articulation, in an identical dispute arising out of the same set of facts, of the two European legal orders. While both guarantee fundamental rights on the basis of constitutional provisions (EU Charter and ECHR), among which the rights of the child are accorded the utmost supremacy, they may not share a methodology in the assessment of the existence of a violation, nor give exactly the same weight to the various factors which weigh into the process. This is the context in which the “Bosphorus presumption” (ECtHR Bosphorus v. Ireland 2005), which allows an overlapping consensus between the two universes, is now brought into the equation (Povse). Meanwhile, back down among the national courts, local judges – sometimes “siding” with the parent who is the national or domiciliary of their jurisdiction and who prefers to litigate to the bitter end rather than let the other win on the (theoretically) preliminary issue of where the merits of the custody dispute is to be decided – have to decide this mega-conflict between two supra-national regimes which both purport to promote the interests of the child! The child is often the prime victim of all this. To my mind, the real problem may well lie with the whole design of the cross-border child-return system, which focuses on the restitution of the abducted child before the custody dispute can be decided on the merits. While a highly creative idea at the outset, its undoubted potential to absorb tension when the parents are cooperative is as great as the risk of amplification of conflict it carries with it when they are not. See the sheer length and number of procedural incidents in the Povse case (which led to a first preliminary ruling under Brussels II bis by the ECJ before the case was lodged with the ECtHR).
However, although Gilles Cuniberti mentions the Povse case in his opening lines to this symposium, the question for debate is framed in more general terms as concerning the abolition of exequatur (within the EU) and human rights. Therefore, beyond child return issues, it can be understood to be about the primacy either of the new, highly efficient, nuclear missile which has emerged progressively in recent EU secondary legislation (direct cross-border enforceability of a court order without intermediary enforcement proceedings), or of the ultimate joker of fundamental rights (which will be invoked in the very forum that has been by-passed by direct cross-border enforcement). So I’ll start with the larger picture, which, in addition to Brussels II bis, extends to Brussels I recast, and various other instruments that have abolished the formality of exequatur or enforcement proceedings (alimentary obligations, TEE, small claims…). Thoughts on the circulation of debt may be helpful for reflecting upon the more sensitive issues relating to children.
II. The wider picture. Much of the literature on the abolition of exequatur within the European Union under, or in anticipation of, Brussels I recast, turns on whether or not it implies a significant reduction in the protection due to the fundamental rights (particularly procedural rights, which will therefore be the focus of the remarks below), of defendants. In other words, in re-establishing the balance in favor of the creditors of the internal market, who have traditionally suffered from the partitioning of national spheres of enforcement (including the costs of bringing even informal enforcement proceedings), have the tables turned too far in the opposite direction, in diminishing the guarantees due to henceforth vulnerable defendants? According to many accounts, abolishing the intermediate procedural filter of exequatur creates a significant risk of free-wheeling misfit-judgments, of which, when the floodgates are opened in 2015, the defects will be amplified by their cross-border effects.
A first observation is that in assessing this risk, the strength of assertions on either side contrast with the scarcity of empirical findings, as to its extent. We have, for instance, the Commission’s own statistics for the (small) number of effective appeals against enforcement orders (under the existing provisions of Brussels I), according to which it made sense to abolish the remaining procedural formalities (article 38 s. Brussels I). On the other hand, we also have an idea of the very large number of cases in which Member States have been called for account for procedural faults, either in Strasbourg, in Luxembourg, or in the shadow of either in domestic cases in national courts. In the specifically transnational sphere, many of the usual suspects are various forms of transnational injunctive relief, which have met with the disapproval of the ECJ itself (Krombach 2000, Gambazzi 2007…). But such cases can be used to demonstrate either the escalation of vitiated judgments with transnational effects, or the inevitable cultural determination of core standards of fairness. That is not to say that there will not always be (more or less) occasional duds among the number of judicial decisions produced by any legal system; that is precisely indeed why fair process requires allowing an appeal. However, the question here is specifically whether the risk of being subject to misjudgments from another country is greater with or without exequatur.
The political terms of the debate are also complex. For instance, while France has produced its highly predictable strain of critique against any European Union initiative, which though probably accurate in some instances would be more credible if it were not so frequently histrionic or indeed couched in the language of fantasized or quaint accounts of parliamentary democracy, the detractors of Brussels I are now calling for more human rights protection, which of course leads them from Scylla to Charybdis, to the extent that the latter are usually denounced, in private international law and beyond, as a worse methodological sin than the former. Interestingly, the focus of the new ire is no longer a defense of the idiosyncratic play of national public policy, but the safeguard of the due process requirements of the ECHR. Allez savoir!
Moreover, many of the historical and contextual arguments voiced in this context can be unhelpful. The main theoretical support for exequatur appears to be that free movement of judgments assumes their interchangeability, as does a market for non-judicial products; in a world composed of legal systems of very variable quality or content, producing equally heterogeneous judgments, exequatur thus fulfills the leveling function of a lock. However, such a function was constructed at a time when there was no supervisory device ensuring procedural (and indeed substantive) guarantees “from above” (that is, based on the ECHR or, where applicable, the EU Charter), nor indeed any common standard as to their content; a horizontal filter of incoming decisions supplied by exequatur or enforcement proceedings was therefore, naturally, put into place in each national forum, on the basis of highly variable conceptions of procedural and substantive fairness. The origins of the whole Brussels jurisdiction and judgment system are to be found in the supposed costs that this variation created for those supplying credit in the internal market (at a time when Member States also used purely jurisdictional criteria as part of the filter). In retaining exequatur, if only as a formality, the existing Brussels I Regulation still adheres to a similar logic.
The shift wrought by the new regime in Brussels I recast is therefore a form of trade-off, made possible by the fact that each domestic court is deemed accountable within its own legal system in respect of the content of fair trial resulting from article 6-1° ECHR. Every court of origin, in handing down a judgment, is committed to respect ex ante the very same guarantees that can at present (under the existing Brussels I) be invoked additionally ex post in exequatur proceedings (or more accurately in appeal therefrom). Thus, the question is: does the reshuffling of the places of control, which under the new regime means that any challenge to the procedural fairness of a judgment or public act is to take place ex ante in the country of origin, and not ex post in the courts of the place of enforcement, potentially reduce fundamental procedural rights protection?
At this stage it is also worth pointing out that the emergence of a common core of procedural standards under article 6-1° ECHR put an end to the traditionally “attenuated” form of public policy control which had hitherto been associated (as such, or as an expression of Inlandbeziehung) with the recognition and enforcement of foreign judgments, at least as far as procedural guarantees are concerned. In other words, the enforcing state is bound by exactly the same standards (of which, however, the open-endedness subtly precludes absolute identity of procedural rules) as the state of origin. These are indeed applicable in full to judgments from third states (see ECJ Pellegrini 2001). Within the European Union, the question is once again how far maintaining only one set of controls, ex ante in the state of judgment (rather than two sets, of which one in the enforcing state under identical standards), implies a reduction of the level of protection for potential debtor-defendants. In other words, how far is the second control ex post actually useful as a human rights safeguard, and to what extent is it parasitical in terms of costs to (both) parties?
The statistics upon which the Commission acted seem to indicate that it is not indeed indispensable, since exequatur orders give rise to appeals infrequently. But the debate continues. Thus, even if the statistics hold true across the board (are they really significant beyond small or uncontested claims?), there may be additional advantages attached to the existence of an intermediary procedure. One of these might be an important element of inter-systemic judicial dialogue which works to boost human rights protection (“outsiders’ insights”, to use the phrase of Basil Markesinis): look, after all, what it took in Krombach to challenge the civil effects of contumace in French (criminal) procedure. It may be, on the other hand, that given the large corpus of common standards which have developed since 2000 in the case-law of the ECtHR on the basis of article 6-1° ECHR, such an argument is becoming increasingly irrelevant; after all, lawyers are far more accustomed now to invoking such case-law within domestic settings, so that the time may have come to dispense with an external source of challenge and concentrate on efficiency.
But what if (exceptionally?), nevertheless, a vitiated judgment slips though the net? Part of the answer lies with the power of the court at the place of enforcement to refuse to give it effect. In the case of Brussels I recast, articles 46 et seq allow both preventive and remedial opposition to mis-judged foreign judgments, thereby transferring to the enforcing judge the control exercised until now in the course of (on appeal from) exequatur proceedings. The grounds for opposition (article 45) are indeed the same and allow for refusal of enforcement for both (exceptional) substantive (a) and procedural (b) reasons. What was the point of so much ado over the “recast”, then, one might ask? Certainly, in the end, the burden of initiating the unforceability proceedings shifts to the defendant. Nevertheless, under the existing system, it is also the defendant who shoulders the (lesser?) weight and cost of the appeal against the exequatur. The result is probably similar, therefore, no better no worse, than within the previous framework.
However, whether or not in the latter context, there is always a possibility (arguably – though not necessarily convincingly – amplified by this shift), that the requirements of article 6-1° may not be satisfied nevertheless, following an unsuccessful attempt to oppose such enforcement before the local court. At first glance this might give rise to a risk of the type encountered in the child abduction case Sneersone & Kampanella cited above, where insufficient regard to the fundamental rights of the abducting parent or child by the original pre-abduction home court, ordering an immediately enforceable return, created not only a cause of refusal but also a jurisdictional-procedural incident unprovided-for by Brussels II bis’ fast-track procedure. However, the analogy may not be as clear-cut as it might seem at first glance since, in the latter context, the whole point of the fast-track is that it is intended to eliminate all obstacles to the enforcement of the initial cross-border return order along the way, in the name of the superior interests of the child. Whereas, in the context of Brussels I recast (as far I can see), the local enforcement procedure would appear to make all the difference, by providing an opportunity to resist a foreign judgment on fundamental rights grounds (at least those covered by article 45), as a last resort. Much, therefore, turns on this local enforcement procedure; the cases in which no such procedure exists (alimentary obligations, TEE..) may be more dicey. Be that as it may, in the context of Brussels I recast, I’m not convinced that in terms of loss of protection of defendants‘ fundamental rights, the change is as big a deal as is sometimes made out (although of course – no sooner said than done – practice will probably come up with a morally inacceptable cross-border small claims case…).
III. Now for the real difficulty. By contrast, article 11 (8) Brussels II bis provides for a return order by the pre-abduction home court, notwithstanding a judgment of non-return by the court at the place of enforcement; in other words, the fast-track is designed to by-pass resistance in the country of refuge, where the abducting parent seeks to keep the child (by virtue of article 13b 1980 Hague Convention). This provision takes the speediness of return to be of the essence, in the name of the best interests of the child, whatever the risk invoked under article 13b. The stakes are (merely) jurisdictional here: ultimately, it is for the court of the child’s pre-abduction home to decide, where appropriate, on the substantive custody issue. However, the need for speed, and the (merely) restitutionary nature of the return, are no apology for sloppy process. Because the nuclear weapon inscribed in article 11(8) suffers no further procedural delay before the child is effectively returned home, it is counterbalanced by the particular duty of the home court under article 42 Brussels IIbis to ensure, before ordering the child’s return notwithstanding the refusal of the court of the country of refuge, that the reasons for such refusal have been properly considered (at stake in Sneersone & Kampanella) and the child heard, unless inappropriate (at stake in Aguirre). If the home court does not do so, or does so unsatisfactorily, it is open to the applicant to challenge the order – including through an individual application to the ECtHR (as indicated in Povse).
But can the human rights joker still be played, as a last resort, at the place of enforcement (in the country of refuge)? Or is such a possibility, which has obvious implications for the allocation of jurisdiction, excluded by the very architecture of the fast-track, in the name of the child’s own best interests? The answer, taking account of the positions of both European courts, is a bit of both, in a subtle dosage of which national courts will now have to take account. What is particularly complex is that the human rights complaint (typically for violation of article 8 ECHR) may involve an issue of access to relief in the country of refuge, that is, a question of international jurisdiction, which is one and the same as that of the procedural (or indeed substantive) guarantees due to the child and/or the abducting parent.
In Aguirre (as indeed in its own preliminary ruling in Povse), the ECJ/CJUE allows no exception to the concentration of jurisdiction at the child’s pre-abduction home – including for the purposes of human rights protection, deemed explicitly to be effective here (§69) by reason of locally available remedies despite the fact that the child and abducting parent are precisely elsewhere. On the other hand, in Sneersone & Kampanella, the ECtHR allows the human rights joker (article 8 ECHR) to be raised at the place of enforcement (country of refuge). Then, however, in Povse, the Bosphorus presumption of “equivalent protection” weighs into the equation. This presumption is conceded by the ECtHR in the name of inter-judicial comity “so as to reduce the intensity of its supervisory role” and avoid putting national courts in the distressful situation of having to choose between competing international obligations. In Povse, it was held that nothing justified a rebuttal of the presumption in the case of the applicants’ claim (article 8 ECHR) within the framework of Brussels II bis. How does all this fit together? It is probably clearer if one distinguishes two different, successive, issues.
(1) The first is whether the lack of recourse per se (abolition of exequatur), as a structural feature of the fast-track procedure, deprives the child of adequate protection (as claimed for instance by the applicant in Povse).
– In Aguirre (as in the Povse preliminary ruling), the ECJ judges that the fact that challenges to the return order are all to be raised exclusively in the country of origin does not run counter to article 24 of the Charter, in the light of which article 42 Brussels II bis has to be read.
– While the ECtHR endorses this result (in Povse), it is by virtue of a line of reasoning in two steps.
(i) Firstly, the “Bosphorus presumption” is applicable because under article 11(8) Brussels IIbis, the court of the country of refuge, having no choice but to order the return of the child, exercises no discretion (see ECtHR MSS 2011). Moreover, the ECJ/CJEU had already considered (as would have to be the case under ECtHR Michaud v France 2012, §114 et s.) the specific issue of the compatibility between article 11 (8) Brussels II bis and the article 8 Convention right to a family relationship (it having judged in its own preliminary ruling in the Povse case that the availability of an appeal on the basis of article 8 before the courts of the pre-abduction home country was sufficient protection: see on the CJUE’s position, ECtHR Povse, §85). Given these two factors (no discretion and prior decision of the CJUE), the protection accorded to the right claimed under the ECHR is deemed by the ECtHR to be equivalent, under the Bosphorus presumption, to the protection afforded by Brussels II bis; the jurisdiction of the home court remains exclusive.
(ii) Secondly, there is no showing here, in the specific context of the Povse case, that the presumption should be rebutted. The decisive reason seems to be that the applicants did not even attempt to avail themselves here of the opportunity of challenging the order in the court of origin (ultimately, if necessary, by lodging an application with the ECtHR if such an attempt were to fail). This circumstance is clearly salient precisely because the availability of an appeal on the basis of article 8 ECHR in the home country is taken to be the reason for which the Michaud requirement (relating to the CJEU’s own confirmation of adequate protection in respect of the right invoked) is fulfilled here (see above). Implicitly, according to the Bosphorus line of reasoning, there is an exhaustion-of-local-remedies condition, that does not – of course – preclude a challenge to the return order at the place of enforcement, if all else fails.
(2) Considering, then, that the presumption is rebuttable (even if not rebutted in Povse), would it still be possible to raise a human rights joker before the courts of the country of refuge (as in Sneersone & Kampanella, decided before Bosphorus was brought into the equation) if, in a particular case, the (pre-abduction home) court ordering the return did not deal, or dealt inadequately, with the human rights challenge? Under Bosphorus, the rebuttal of the presumption of equivalent protection would have to meet a particularly rigorous standard of proof of the violation (§156 : a “manifest deficiency” of protection) in a particular case in order to justify that the constitutional values of the ECHR prevail over the interests of international cooperation. In principle, however, if it could be shown that despite exhaustion of all available remedies in the pre-abduction home country, the protection of child’s (or a parent’s) right has nevertheless been severely hampered, this would then still seem to imply, as in Sneersone & Kampanella, that there would be a right of access to the court of the place of refuge, and grounds for a refusal of enforcement of the notwithstanding order by such court. However, since the exhaustion of remedies in the home country would include (again, as indicated in Povse) an application to the ECtHR itself, it would only be if for some reason the access to such remedy proved to be impossible that the access argument could be made effectively in the courts of the country of refuge. Of course, it also appears from Sneersone & Kampanella and Povse combined, that in most (all?) cases, had the return order been effectively challenged locally and had the courts of the pre-abduction home country (on appeal) carried out their obligations under article 42 Brussels II bis (and the Charter), there would be no need – and indeed, by the same token, no right – to call for help from the courts of the country of refuge under the ECHR.
In the meantime, the policy problem is whether the current child-return system, designed to ensure against (assumedly) opportunistic forum shopping by the abducting parent, really works to further the best interests of the child. It may be that the current litigation inflation is transitional and that, once stabilized, the system will work more satisfactorily, with less collateral damage. Arguably, however, the multi-level jurisdictional scheme may have become too unwieldy, and whether or not it now weighs too heavily in favor of the non-abducting or stay-at-home parent (see Kampanella), such violent and probably costly legal battles can only be detrimental to the child. While on the one hand Brussels II bis supports speedy return in the name of the child’s interest in abstracto, on the other, the circumstances of particular children in individual cases, to which the ECtHR directs its attention, often point in a different direction. These two opposite viewpoints, which also correspond to two competing epistemological schemes in the two European courts’ patterns of reasoning, may indeed be at the very heart of the new mega-conflict-of-laws.