Written by Mayela Celis

Undoubtedly, Abortion is a hot topic. It is discussed in the news media and is the subject of heated political debate. Indeed, just when one thinks the matter is settled, it comes up again. In 2023, Elgar published the book entitled “Research Handbook on International Abortion Law”, ed. Mary Ziegler (Cheltenham: Edward Elgar Publishing Limited, 2023). For more information, click here. Although under a somewhat misleading name as it refers to international abortion law, this book provides a wonderful comparative overview of national abortion laws as regulated by States from all the four corners of the world and internal practices, as well as an analysis of human rights law.

This book does not deal with the conflict of laws that may arise under this topic. For a more detailed discussion, please refer to the post Singer on Conflict of Abortion Laws (in the U.S.) published on the blog of the European Association of Private International Law.

In this book review, I will briefly summarise 6 parts of this book (excluding the introduction) and will provide my views at the end.

This book is divided into 7 parts:

Part I – Introduction

Part II – Histories of liberalization

Part III – The promise and limits of decriminalization

Part IV – Abortion in popular politics

Part V – Movements against abortion

Part VI – Race, sex and religion

Part VII – The role of international human rights

Part II – Histories of Liberalization

Part II begins with a historical journey of the abortion reform in Sweden in the 1930s and 1940s. It highlights the limited legalization of abortion in Sweden in 1938 and the revised abortion law in 1946 introducing a “socialmedical” indication. In particular, it underscores how the voices of women were absent from the process.

It then moves on to a comparative study of the history of abortion in the USA and Canada from 1800 to 1970, that is before Roe (USA) and Morgentaler (Canada). It analyses the distinct approaches of Canada and the USA when dealing with abortion (legislative vs. court-based). Furthermore, it provides a very interesting historical account on how the right of abortion came about in both countries – it sets the stage for Roe v. Wade (pp. 50-52).

Finally, Part II examines the situation in South Africa by calling it “unfinished business”. In South Africa, Abortion is a right codified in law: The Choice on Termination of Pregnancy Act 92 of 1996. However, this article argues that the legislative response is not enough. Factors such as lack of enough health facilities that perform abortions, gender inequality etc. are an obstacle to making safe abortion a reality.

Part III – The promise and limits of decriminalization

This Part analyses several laws regarding abortion. First, it explores Malawi’s 160-year-old law that criminalises abortion based on a UK law, as well as the failed tentative attempt to adopt a new law in 2020. Interestingly, this article analyses CEDAW resolutions against the UK, which promptly complied with the resolution (pp. 92-93).

Secondly, it studies the recently adopted law in Thailand on 7 February 2021 that makes  abortion available up to 12 weeks’ gestation period. However, this article criticises that the law creates a loophole as the abortion must be performed by a physician or a registered medical facility and in compliance with the law, greatly medicalizing abortion.

Finally, this Part examines Australian laws and policy over the past 20 years and while acknowledging the significant advances in reproductive rights, it notes that a number of barriers to abortion still remain. This chapter is better read in conjunction with Chapter 10, also about Australia.

Part IV – Abortion in popular politics

This Part begins with an excellent comparative public policy study between France and the United States. In particular, it discusses the weaknesses of Roe v. Wade, underlining the role and analysis of the late justice Ruth Bader Ginsburg. It also puts into context the superiority of the French approach regarding abortion, which is proven with the reversal of Roe.

It then analyses abortion law in China, a State that has the most lenient abortion policies in the world. It discusses the Chinese one-child policy, which then changed to two and even three children-policy, as well as sex-selective abortions.

Subsequently, it recounts how South Australia became the last Australian jurisdiction to modernise its abortion laws and underlines the fact that laws in Australian jurisdictions on this topic are uneven and no two laws are the same.

Finally, it examines abortion history in Israel noting that apart from health reasons, abortions on no specific grounds are mainly intended for out-of-wedlock pregnancies. As a result, abortion is restricted to married women unless they claim adultery, a ground that must be reviewed by a Committee. Apparently, this leads married women to lie to get an abortion and go through the shameful process of getting approval by a Committee.

Part V – Movements against abortion

This Part begins with abortion politics in Brazil and the backlash that occurred with the government of former president Bolsonaro who, as is well known, is against abortion. It recounts a case where a priest filed an habeas corpus in favour of a foetus who had a severe birth defect. Although the case arrived at the Federal Supreme Court, it was not decided as the child died 7 minutes after being born (p. 232).

Secondly, a history scholar recounts the pro-life movement across continents and analyses what drives them (i.e. gender and religion).

Finally, it deals with abortion law in Poland and Hungary and the impact of illiberal courts. In particular, it discusses the trends against abortion and goes on to explain an interesting concept of “illiberal constitutionalism”. The authors argue that they do not see Poland and Hungary as authoritarian systems but as illiberal States, an undoubtedly interesting concept.

Part VI – Race, sex and religion

This Part begins examining the sex-selective abortions in India. In particular, the authors recommend an equality-based approach instead of anti-discriminatory approach in order to avoid recognising personhood to the foetus.

It then continues with an analysis of abortion law in the Arab world. The authors note that there is scant but emerging literature and that abortion laws in this region are – unsurprisingly – punitive or very restrictive. Interestingly, the position of Tunisia differs from other Arab States.

Finally, it discusses the struggles in Ecuador where a decision of the constitutional court of 2021 decriminalising abortion in cases of rape. It declared unconstitutional an article of the Ecuadorian Criminal Code, and in 2022 the legislature approved a bill based on this ruling. It also refers to teenage pregnancy and violence.

Part VII – The role of international human rights

For those interested in international human rights, this will be the most fascinating Part of the book. Part VII calls for the decriminalization of abortion in all circumstances and it supports this argument by making reference to several human rights documents such as those issued by the Human Rights Committee (in particular, General Comment No 36 –  Article 6: Right to life) and the Committee on the Elimination of Discrimination against Women (referring to a myriad of general comments and concluding observations).

Subsequently, this Part challenges the classification of European abortion law as fairly liberal and provides some convincing arguments (including the setbacks in Poland in this regard and other procedural or legal barriers to access abortion in more liberal States) and some surprising facts such as the practice in the Netherlands (see footnote 60). The authors -fortunately- dared to say that this chapter is drafted from a feminist perspective as opposed to the current “male norm” in legal doctrinal scholarship.

Finally, this Part explains the history of abortion laws including the fascinating recent developments in Argentina and Ireland (referred to as “small island”!) and the influence (or the lack thereof) of international human rights law. In particular, it makes reference to the Argentinian Law 27,610 of 2020 (now unfortunately in peril with the new government) and the repealing by referendum of the 8th Amendment in Ireland in 2018.

Below are a few personal thoughts and conclusions that particularly struck me from the book:

Starting from the beginning: the title of the book and the definitions.

In my view, and as I previously mentioned, the title of the book is somewhat misleading. Strictly speaking, there is no such thing as “international” abortion law but rather abortion prompts a discussion of international human rights, such as women’s rights and the right to life, and whether or not national laws are compliant with these rights or are coherent within their own national legal framework. This is in contrast to international child abduction / adoption laws where international treaties regulate those very topics.

While perhaps counterintuitive, the definition of a “woman” has been controversial; see for example the Australian versus the Thai approaches. The Australian approach deals with gender identification and the fact that persons who do not identify as a woman can become pregnant (p. 124, footnote 1). While the Thai approach defines a woman as those capable of bearing children (p. 112). Needless to say, the definition of a woman is essential when legislating on abortion and unavoidably reflects the cultural and political complexities of a particular society. A brief reference is made to men and gender non-conforming people and their access to abortion (p. 374, footnote 2).

A surprising fact is the pervasive sex-selective abortion in some countries (sadly against female foetuses), such as India and China, and which arguments are invoked by scholars to avoid them, without falling into the “trap” of recognising personhood to the foetus.

More importantly, this book shows that the abortion discussion is much more than the polarised “pro-life” and “pro-choice” movements. The history of abortion is complicated, full of intricacies. And what is frustrating to some, this area is rapidly evolving sometimes at the whim of political parties.

Most authors seem to agree that a legislative approach to abortion is more recommended than a court-based approach. Indeed, there is a preference for democratically elected lawmakers when it comes to dealing with abortion. This is evident from the recent setbacks that occurred in the USA.

Having said that, those expecting an in-depth analysis of the landmark US decision Dobbs v. Jackson Women’s Health Organization 597 U.S. 215 (2022), which overturned Roe v. Wade, will be disappointed (only referred to very briefly in the introduction and Chapters 8, 11 and 13 ). Instead, however, you will be able to immerse yourself into a multidisciplinary study of abortion law, including topics such as politics, sociology, constitutional law, health law and policy, history, etc. In addition, you will read unexpected facts such as the role of Pierre Trudeau (former Prime Minister (PM) of Canada and father of current Canadian PM, Justin Trudeau – p. 56 et seq.) in abortion law in Canada or the delivering of abortion pills via drones (p. 393).

Because of all the foregoing, and whatever one’s standpoint on abortion is, I fully recommend this book. But perhaps a cautionary note: people in favour of reproductive rights will be able to enjoy the book more fully.

I would like to end this book review with the words of the French writer and philosopher Simone de Beauvoir, which appear in her book entitled the Second Sex and which are also included in chapter 8 (p. 159) of this book:

“Never forget that a political, economic or religious crisis would suffice to call women’s rights into question”

Full citation:

“Rien n’est jamais définitivement acquis. Il suffira d’une crise politique, économique ou religieuse pour que les droits des femmes soient remis en question. Votre vie durant, vous devrez rester vigilantes.”

Written by Sandrine Brachotte who obtained a PhD. in Law at Sciences Po, Paris and is a Guest Lecturer at UCLouvain (Saint-Louis, Brussels).

1. PIL and (De)coloniality in Europe

This post follows Susanne Gössl’s blog post series on ‘Colonialism and German PIL’ (especially s. 3 of post (1)) and offers a French perspective of the issue of PIL and (de)coloniality – not especially focused on French PIL but based on a francophone article to be published soon in the law and anthropology journal Droit et Culture. This article, called ‘For a decolonisation of law in the global era: analysis of the application of postcolonial law in European states’, is addressed to non-PIL-specialist scholars but builds on a European debate about PIL and (de)coloniality that has been nourished by scholars like Ralf Michaels, Horatia Muir Watt, Veronica Ruiz Abou-Nigm, as well as by Maria Ochoa, Roxana Banu, and Nicole Štýbnarová, notably at the occasion of the 2022 Edinburgh conference (reported about on this blog, where I had the chance the share a panel with them in relation to my PhD dissertation (see a short presentation on the EAPIL blog)).

The PIL and (de)coloniality analysis proposed in this post is based on decolonial theory and postcolonial studies, which I will here call ‘decoloniality’. Given this framework (notably nicely presented here), I shall preliminarily stress that it requires acknowledging the limit of the contribution I can make to the debate on PIL and (de)coloniality as a Western jurist. Therefore, this post aims at encouraging non-Western and/or non-legal scholars to contribute to the discussion. It also urges the reader to consider that the non-West and non-legal scholarship about law and (de)coloniality is extremely rich and should not be missed by the Western PIL world.

2. For a Case-by-Case Approach

Against this background, the argument made here is that the decolonisation of Western PIL, if it is to happen (which decoloniality demands, based on the concept of global coloniality), should be based on a certain methodology (see eg the decolonial legal method elaborated by Tchepo Mosaka). Such methodology may require a case-by-case approach, to complement the study of the applicable legal framework. This seems at least necessary in the context, studied in the aforementioned article, where a postcolonial law is to be applied as foreign law by the Western forum (typically but not only in the context of migration), given that ‘postcolonial law’ hides a form of legal pluralism. It thus potentially covers not only state law, but also customary law and/or religious law.

To study this kind of situation, I argue, a case-by-case approach is needed because the legal pluralism of each postcolonial state is idiosyncratic. Notably, the postcolonial state law may refer to some religious or customary norms (which is a form of official legal pluralism); or these non-state norms may be followed by the population because the state institution is deficient or because a large part of the population simply does not follow the state legal standards (which is a form of de facto legal pluralism); or yet, certain state legal concepts or standards may reflect some custom or religious norms or practices.

More generally, the case-by-case approach allows a more nuanced (although also more complex) analysis of the (de)colonial character of current Western PIL standards. For PIL rules and judicial practices may appear colonial (ie, as imposing a Western ‘worldview’) or decolonial (ie, as granting space to ‘colonised’ worldviews) depending on the case, rule and/or judicial practice concerned. In addition, the case-by-case approach enables the consideration of the personal experience and possible vulnerable position of the parties – something that is also demanded by decoloniality. Therefore, the case-by-case approach seems appropriate to also study other questions than the application of postcolonial law discussed here, such as the limits of the Western definition of some important PIL concepts (like family and habitual residence, discussed in Susanne Gössl’s post (2), or party autonomy, of which I have shown a colonial aspect via a case study in my PhD dissertation (see here) and that is also discussed in Susanne Gössl’s post (4)).

3. The Example of X v Secretary of State for the Home Department ([2021] EWHC 355 (Fam))

To illustrate the argument, I choose a UK case that enters into a direct dialogue with Susanne Gössl’s reflection about the notion of habitual residence (see post (2)). In this case, X v Secretary of State for the Home Department ([2021] EWHC 355 (Fam)), the claimant demanded the recognition by the UK authorities of her child’s adoption in Nigeria. Under the applicable UK PIL rules, this adoption had to be recognised in the UK if it complied with the Nigerian law, ie Article 134(b) of the 2004 Child Rights Law. This article provides that the adopter and the adopted must have their residence in the same state. In the absence of any Nigerian caselaw interpreting the notion of residence under Article 134(b), the question came as to whether it had to be interpreted based on UK law or on local customary norms.

Pursuant to the relevant customary law, two circumstances should be considered that could lead to locate the claimant’s residence in Nigeria. On the one hand, the claimant had an ‘ancestral history and linkage’ with Nigeria. One the one hand, as she lived most of the time in the UK to work, she entrusted her adopted child to her mother but took full financial responsibility for the child and made all decisions relating to the child’s upbringing. Pursuant to UK law, more specifically Grace ([2009] EWCA Civ 1082), in case where someone lives in between several countries, the notion of residence had to be interpreted following a ‘flexible nuanced approach’ (para. 84(5)).

In February 2021, the UK judge recognised the adoption established in Nigeria, based on the interpretation of residence in UK law. To this end, the judge used the presumption, which is part of UK PIL, of similarity between foreign law and domestic law. Following Brownlie ([2021] UKSC 45), the judge applied the presumption because, like the UK, Nigeria is a common law system. Then, referring to Grace, the judge located the claimant’s residence in Nigeria. In this regard, she considered the claimant’s ‘close cultural and family ties’ with Nigeria, the fact that she maintained a home there for her mother and children, and the circumstance that ‘[h]er periods of time in [Nigeria] were not by chance, but regular, family focused and with a clear purpose to spent time with her children’ (para. 84(6)).

4. A PIL and Decoloniality Analysis: Opening the Floor

From a PIL and decoloniality perspective, several points can be made. Notably, from a strict legal point of view (lacking anthropological insights), the judge’s interpretation of the UK law notion of residence in this case seems flexible enough to include various, Western and non-Western, worldviews. Yet, one may question the application of the UK legal presumption. Because Nigerian state law is common law indeed, but it shares legality with customary laws and Sharia. Therefore, from a decolonial point of view, the judge could have usefully investigated the question as to whether, to interpret similar laws as the Child Rights Law, Nigerian courts consider customary law (and potentially, the judge did so (see para. 84(5)), but then it would have been welcome to mention it in the judgment). If so, she could have interpreted the notion of residence, not based on UK law, but based on the relevant local customary norms.

These case comments are made just to start a wider discussion – not only about this case but also about other cases. For, in my view, the PIL and (de)coloniality debate is a great occasion to have another, alternative, look at some rules and caselaw, and to open the floor to non-Western and/or non-PIL scholars.

Can an enforcement judgment issued by a foreign court be recognized and enforced in another jurisdiction? This is a fundamental question concerning the recognition and enforcement of foreign judgments. The answer appears to be relatively straightforward: “No”. Foreign enforcement judgments are not eligible to be recognized and enforced as they are not decisions on the merits (see in relation with the HCCH 2019 Convention, F Garcimartín and G Saumier, Explanatory Report (HCCH 2020) para. 95, p. 73;  W Hau “Judgments, Recognition, Enforcement” in M Weller et al. (eds.), The HCCH 2019 Judgments Convention: Cornerstones, Prospects, Outlooks (Hart 2023) 25). This is usually referred to as the “prohibition of double exequatur” or, following the French adage: “exequatur sur exequatur ne vaut”. This question was recently presented to the Dubai Supreme Court (DSC), and its decision in the Appeal No. 1556 of 16 January 2024 offers some useful insights into the status foreign enforcement (exequatur) decisions in the UAE.

I – Facts

In 2012, X (appellee) obtained a judgment of rehabilitation from the United States District Court for the Eastern District of New York ordering Y (appellant, residing and working in Dubai) to pay a certain amount of money. X later sought to enforce the American judgment in Canada (Ontario) via summary judgment procedures. In 2020, the Ontario court ordered enforcement of the American judgment, in addition to the payment of other fees and interests. The judgment was later amended by a judgment entered in 2021. X then sought enforcement of the Canadian judgment in Dubai by filing an application with the Execution Court of the Dubai Court of First Instance. The Enforcement Court issued an order declaring the Canadian judgment enforceable in Dubai. The enforcement order was later upheld on appeal. Y appealed to the DSC.

Before the DSC, Y argued that (1) the American judgment was criminal in nature, not civil; (2) the Canadian judgment was merely a summary order declaring the American judgment enforceable in Ontario; and (3) the Ontario judgment did not resolve any dispute between the parties, as it was a declaration that the American judgment was enforceable in Ontario.

II – Ruling

The DSC found merit in Y’s arguments. In particular, the DSC held that the Court of Appeal erred in allowing the enforcement of the Canadian judgment in Dubai despite Y’s arguments that the Canadian judgment was a summary judgment enforcing an American judgment. The Supreme Court reversed and remanded the appealed decision.

III – Comments

The case commented here is particularly interesting because, to the best of the author’s knowledge, it is the first case in which a UAE Supreme Court (it should be remembered that, there are four independent Supreme Courts in the UAE. For an overview, see here) has been called to rule on the issue of double exequatur. In this regard, it is remarkable that the issue of double exequatur is rarely discussed in the literature, both in the UAE and in the other Arab Middle Eastern jurisdictions. Nevertheless, it is widely accepted that a judgment a foreign court declaring enforceable a foreign judgment cannot be eligible to recognition and enforcement in other jurisdictions. (For some recent applications of this principle by some European courts, see eg. the Luxembourg Court of Appeal decision of 13 January 2021; the Court of Milan in a case rendered in February 2023. Comp. with the CJEU judgment of 7 April 2022, C?568/20, J v. H Limited. For a brief discussion on this issue in this blog, see here). This is because a judgment declaring enforceable a foreign judgment “is, by its own terms, self-limited to the issuing state’s territory, or: as a sovereign act it could not even purport to create effects in another sovereign’s territory” (Peter Hay, “Recognition of a Recognition Judgment within the European Union: “Double Exequatur” and the Public Policy Barrier” in Peter Hay et al. (eds.), Resolving International Conflicts – Liber Amicorum Tibor Várady (CEU Press, 2009) 144).

The present case highlights a possible lack of familiarity with this principle within the Dubai courts. Specifically, the lower courts overlooked the nature of the Canadian judgment and declare it enforceable in Dubai. In its appeal, the judgment debtor did not explicitly avail itself with the prohibition of double exequatur although it argued that that the Canadian judgment was “not a judgment on the merits”. The judgment debtor merely stated the Ontarion court’s judgment was a summary judgment declaring a foreign judgment of criminal rather than civil nature enforceable in Canada and not abroad .

While the Supreme Court acknowledged the merits of the judgment debtor’s arguments, its language also might suggest some hesitation or unfamiliarity with the legal issue involved. Indeed, although the Court did not dispute the judgment debtor’s assertions that the “Canadian judgment was a summary judgment declaring enforceability and an American reorganization judgment,” it reversed the appealed decision and remanded the case, stating that the judgment debtor’s arguments were likely – “if they appeared to be true” – to lead to different results.

In the author’s view, such a remand may have been unnecessary. The court could have simply declared the Ontario enforcement order unenforceable in Dubai on the basis of the “exequatur sur exequatur ne vaut” principle.

One might question the rationale behind the judgment creditor’s choice to seek the enforcement of the Canadian judgment rather than the original American judgment in this case. One might speculate that the judgment creditor sought to avoid enforcement of an order to pay a specific sum arising out of a criminal proceeding. However, it is recognized in the UAE that civil damages awarded in criminal proceedings are likely to be considered enforceable (see, eg., the Federal Supreme Court’s decision, Appeal No. 247 of November 6, 2012, regarding the enforcement of civil damages awarded by an Uzbek criminal court).

Another possible consideration is that the judgment creditor sought to increase the likelihood that its application would be granted, as Dubai courts have shown reluctance to enforce American judgments in the past (see eg., Dubai Court of Appeal, Appeal No. 717 of December 11, 2013, concerning a Nevada Court judgment; DSC, Appeal No. 517 of August 28, 2016, concerning a California court judgment). In both cases, enforcement of the American judgments was refused due to the lack of reciprocity with the United States (however, in the first case, on a later stage of the proceeding, the DSC treated the Nevada judgment as sufficient proof of the existence of the judgment creditor’s debt in a new action on the foreign judgment (DSC, Appeal No. 125/2017 of 27 April 2017). The first case is briefly introduced here).

The positive outcomes at both the first and second instance levels may lend credence to this hypothesis. In general, however, there is no inherent reason why a Canadian judgment would be treated differently in the absence of a relevant treaty between the UAE and Canada (on the challenges of enforcing foreign judgments in the UAE, particularly in Dubai, in the absence of a treaty, please see our previous posts here and here).

By Biset Sena Günes, Senior Research Fellow at the Max Planck Institute for Comparative and International Private Law, Hamburg

Recently, on 25 October 2023, the Austrian Supreme Court (‘OGH’) [2 Ob 179/23x, BeckRS 2023, 33709] ruled on whether a jurisdiction clause included in the terms of purchase (‘ToP’) was valid when a written contract made reference to the website containing the ToP but did not provide the corresponding internet link. The Court held that such a clause does not meet the formal requirements laid down under Article 25 of the Brussels I (recast) Regulation and, hence, is invalid. The judgment is undoubtedly of practical relevance for the conclusion of international commercial contracts that make reference to digitally available general terms and conditions (‘GTCs’), and it is an important follow-up to the decisions by the Court of Justice of the European Union (‘CJEU’) in the cases of El Majdoub (C-322/14, available here) and Tilman (C-358/21, available here).

Factual Background and Procedure

A German company and an Austrian company concluded a service agreement in which the German company (‘the service provider’) undertook to provide the engineering plans for a product to the Austrian party (‘the client’). The Austrian party sent its order to the service provider on a written form which stated (in translation): ‘we order in accordance with the terms of purchase known to you (available on our website) and expect your confirmation by email immediately’. The order specified the client’s place of business as the place of delivery. The German party subsequently signed and returned the same document, ticking its relevant parts and naming it as the ‘order confirmation’. This confirmation was also in written form. The ToP – which were not attached to the contract, but which were available on the client’s website – contained a jurisdiction clause conferring jurisdiction on the Austrian courts for the resolution of disputes arising from the parties’ contract. The clause also allowed the Austrian party to sue in another competent court and was thus asymmetric. The ToP additionally included a clause defining the place of performance for the delivery of goods or for the provision of services as the place specified by the client in the contract.

Upon a disagreement between the parties due to the allegedly defective performance of the service provider, the Austrian party brought proceedings against its contracting partner before the competent district court of Vienna, Austria, in reliance on the jurisdiction clause. The defendant successfully challenged the jurisdiction of the court by claiming that the clause did not meet the formal requirements of Article 25 of the Brussels I (recast) Regulation. Upon appeal, this issue was not addressed, but the judgment was nevertheless overturned as, in the court of appeals’ view, the first instance court was competent based on the parties’ agreement as to the place of performance. According to the court, the parties’ numerous references to the place of business of the client should be understood as an agreement on the place of performance within the meaning of Article 7 of the Brussels I (recast) Regulation, even though the defendant argued that the engineering plans were actually drafted at their place of business and not that of the client. The defendant appealed against the judgment before the Austrian Supreme Court.

The Issue at Stake and the Judgment of the Court

As could be easily identified from the facts and the parties’ dispute, the main question in this case is whether the formal requirements of the Brussels I (recast) Regulation, and in particular its demand of ‘written form’, could be satisfied by a simple reference to a website where the party’s ToP – including the jurisdiction clause – could (allegedly) be retrieved, hence allowing the court to conclude that parties indeed reached an agreement as to jurisdiction.

The Court answered the first question in the negative and found the jurisdiction clause invalid. This is because the ‘written form’ requirement under Article 25(1) (a) of the Brussels I (recast) Regulation is met only if the contract expressly refers to the GTCs containing a jurisdiction clause and if it can be proved that the other party actually received them. According to the Court’s reasoning, the mere reference to the website did not make the jurisdiction clause (or the ToP, in general) accessible to the other contracting party in a reproducible manner; this is unlike the case of a written contract providing a specific link (as in Tilman) or the case of ‘click-wrapping’ (as in El Majdoub), as those are contractual constellations sufficiently establishing that the parties had access to the terms of the agreement (paras 19–20 of the judgment).

General Assessment in Light of the Case Law of the CJEU

Choice-of-court agreements are undoubtedly an important part of today’s highly digitalised business environment, and it is to be expected that they will be found in digitally available GTCs. Yet in practice their validity is often challenged by one of the parties. The Court of Justice has indeed had to deal with such issues in the past, and the present case gives us cause to briefly revisit those rulings.

In El Majdoub (commented before on blogs, here and here), the CJEU had to decide on the question of whether a ‘click-wrap’ choice-of-court clause included in the GTCs provided a durable record which was to be considered as equivalent to a ‘writing’ under the then current Article 23(2) of the Brussels Regulation. In the El Majdoub case, a sales contract was concluded electronically between the parties by means of ‘click-wrapping’, i.e. in order to conclude the agreement, the buyer had to click on a box indicating acceptance of the seller’s GTCs. The GTCs – which containing the agreement as to jurisdiction – were available in that box via a separate hyperlink that stated ‘click here to open the conditions of delivery and payment in a new window’. Although this window did not open automatically upon registration to the website and upon every individual sale, the CJEU found that such a clause provided a durable record as required by Article 23(2) of the Brussels I Regulation since it gave the buyer the possibility of printing and saving the GTCs before conclusion of the contract. This holding should be welcomed as the CJEU gave its blessing to the already existing and much-used practice of ‘click-wrapping’ in the digital business environment, and the Court thus showed its support for the use of technology in contractual practices (in line with aims previously stated in the Commission Proposal (COM(1999) 348 Final)). The Court’s conclusion is, of course, limited in the sense that it only confirms that the ‘click-wrapping’ method provides a durable record of the agreement; there is no analysis as to the requirement of a ‘consensus’ on jurisdiction between the parties in the case of digital contracts. Since the buyer had to accept the terms before the purchase, the Court took this as a consent and did not address the issue (see, similarly, van Calster and Dickinson and Ungerer, LMCLQ 2016, 15, 18–19). It should, in this regard, be observed that establishing the existence of such an agreement is the purpose of the form requirements, a fact confirmed by the case law of the Court, see, e.g. Salotti, para 7 (C-24/76, available here). Still, one should admit that questions as to the existence of consent would probably not be much of an issue in the ‘click-wrapping’ context, especially in B2B cases, as the ‘click’ concludes the agreement – unless, of course, there are other circumstances (e.g. mistake) that affect the quality of consent (see, similarly, van Calster on Tilman).

In the later case of Tilman (previously commented on PIL blogs on a couple of occasions, see the comments by Pacula, by Ho-Dac, and by Van Calster, here and here), the situation was more complex. There was a written agreement between the parties in which the GTCs – which for their part contained an agreement as to jurisdiction in favour of English courts – were referred to by provision of the link to the website where they could be accessed. In other words, there was no ‘click-wrap’ type of agreement; rather, it was a written agreement specifying the link (i.e. the internet address) of the website on which the GTCs could be retrieved. The CJEU then had to deal with the question of whether this manner of incorporating a jurisdiction clause satisfies the conditions of Article 23(1) and (2) of the Lugano II Convention, which are identical to Article 23(1) and (2) of the Brussels I Regulation. The Court answered this question in the affirmative and expanded the possibility of making reference to GTCs by inclusion of the link in written contracts because, in the Court’s view, making those terms accessible to the other party via a link before the conclusion of the contract is sufficient to satisfy formal requirements, especially when the transaction involves commercial parties who can be expected to act diligently. There is no further requirement of actual receipt of those terms. This, again, is a modern and pragmatic approach that simplifies commercial contractual practice, and it is a ruling that should be welcomed. However, it is unfortunate that the Court did not address the technical details in the facts of the case; namely, the link did not open the GTCs directly and instead opened a page on which the GTCs could be searched for and downloaded (see, Summary of the Request for Preliminary Ruling, para 14, available here). This is a point which may give rise to questions as to the proper incorporation of GTCs into a contract (in this regard, see also Finkelmeier, NJW 2023, 33, 37; Capaul, GPR 2023, 222, 225) or as to the existence of consent (on further thoughts as regards the question of consent in both of the CJEU cases, see van Calster). The facts of the case also leave room for a different interpretation in other circumstances, such as when the link refers to a homepage, the link is broken, or the website has been updated (see, in this regard, Finkelmeier, 37; Capaul, 225, and also Krümmel, IWRZ, 131, 134).

In the present case before the Austrian Supreme Court, we encounter yet a different scenario in which there is definitely room for different interpretations. Again, there is a written contract which makes reference to GTCs and which states that they are available on the client’s website. But here, the client did not supply the service provider with the hyperlink address creating accessibility to the GTCs. And the Court rightly held that the CJEU’s conclusion in Tilman should not be understood as saying that a general reference to GTCs in the contract will always be sufficient to prove they have been made available. In the Austrian Court’s understanding, the mere reference to the existence of the GTCs was not sufficient so as to constitute their proper inclusion into the contract and to prove consensus between the parties in a clear and precise manner (paras 19–20 of the judgment). One could, of course, always argue in favour of a further relaxation of the form requirements, especially when the transaction involves commercial parties who should act diligently when entering into contracts. But it is obvious that in a case in which the written contract does not even provide the necessary link, it will be a burden for the counterparty to search the website and retrieve the actual version of the referenced GTCs before entering into the contract, whereas the other party would unduly benefit from being able to fulfil her/his obligation by making a mere reference to the existence of the GTCs. Hence, it is good that the Austrian court did not further extend Tilman’s already broad interpretation.

Conclusion

Despite being an important part of cross-border commercial practice, choice-of-court agreements often become the source of an additional dispute between the parties in terms of their existence and validity. In the vast majority of cases, these disputes are complex. This is probably even more the case with the increasing use of technology in contracting. All these cases are indeed good examples of such disputes. But they can only be seen as new and different additions to the jigsaw puzzle rather than the final pieces. More cases with even more complex scenarios will likely follow, as contracting practices continue to develop along with technological advancements.

Postscript: The Place of Performance

Having found the jurisdiction clause invalid, the Court would have had to determine the place of performance of the contract as another basis for special jurisdiction under the Regulation. A decision on this latter issue was deferred, however, since the Court had already referred a similar question on the determination of the place of performance to the CJEU in a different proceeding (OGH, decision of 13 July 2023, 1 Ob 73/23a) concerning a service contract.

Written by Eduardo Silva de Freitas (Erasmus University Rotterdam), member of the Vici project Affordable Access to Justice, financed by the Dutch Research Council (NWO), www.euciviljustice.eu.  

Introduction

In the final weeks leading up to Christmas in 2023, the District Court of Amsterdam referred a set of questions to the CJEU (DC Amsterdam, 20 December 2023, ECLI:NL:RBAMS:2023:8330; in Dutch). These questions, if comprehensively addressed, have the potential to bring clarity to longstanding debates regarding jurisdictional conflicts in collective actions. Despite being rooted in competition law with its unique intricacies, the issues surrounding the determination of online damage locations hold the promise of illuminating pertinent questions. Moreover, the forthcoming judgment is expected to provide insights into the centralization of jurisdiction in collective actions within a specific Member State, an aspect currently unclear. Recalling our previous discussion on the Dutch class action under the WAMCA in this blog, it is crucial to emphasize that, under the WAMCA, only one representative action can be allowed to proceed for the same event. In instances where multiple representative foundations seek to bring proceedings for the same event without reaching a settlement up to a certain point during the proceedings, the court will appoint an exclusive representative. This procedural detail adds an additional layer of complexity to the dynamics of collective actions under the WAMCA.

Following a brief overview of the case against Apple, we will delve into the rationale behind the court’s decision to refer the questions.

The claim against Apple

The claim revolves around Apple’s alleged anticompetitive behavior in the market for the distribution of apps and in-app products on iOS devices, such as iPhones, iPads, and iPod Touch. The foundations argue that Apple holds a monopoly in this market, as users are dependent on the App Store for downloading and using apps.

According to the foundations, Apple’s anticompetitive actions include controlling which apps are included in the App Store and imposing conditions for their inclusion. Furthermore, Apple is accused of having a monopoly on payment processing services for apps and digital in-app products, with the App Store payment system being the sole method for transactions.

The foundations argue that Apple charges an excessive commission of 30% for paid apps and digital in-app products, creating an unfair advantage and disrupting competition. They assert that Apple’s dominant position in the market and its behavior constitute an abuse of power. Users are said to be harmed by being forced to use the App Store and pay high commissions, leading to the claim that Apple has acted unlawfully. The legal bases of the claim are therefore abuse of economic dominance in the market (Article 102 TFEU) and prohibited vertical price fixing (Article 101 TFEU).

The jurisdictional conundrum

Apple Ireland functions as the subsidiary tasked with representing app suppliers within the EU. The international nature of the dispute stems from the users purportedly affected being located in the Netherlands, while the case is lodged against the subsidiary established in Ireland. The District Court of Amsterdam has opted to scrutinize the jurisdiction of Dutch courts under Article 7(2) Brussels I-bis Regulation. This provision grants jurisdiction to the courts of the place where the harmful event occurred or may occur, encompassing both prongs of the Bier paradigm. However, Apple contends that, within the Netherlands, the court would only possess jurisdiction under Article 7(2) Brussels I-bis Regulation with regard to users residing specifically in Amsterdam.

In the court’s view, the ascertainment of the Handlungsort should pertain only to allegations under Article 102 TFEU. In relation to Article 101 TFEU, the Netherlands was not considered the Handlungsort. This is due to the necessity of identifying a specific incident causing harm to ascertain the Handlungsort, and the absence of concrete facts renders it challenging to pinpoint such an event.

The court’s jurisdictional analysis commences with a reference to Case C?27/17 flyLAL-Lithuanian Airlines (ECLI:EU:C:2018:533), in which the CJEU established that the location of the harmful event in cases involving the abuse of a dominant position under Article 102 TFEU is closely linked to the actual implementation of such abuse. In the present case, the court observes that Apple’s actions, conducted through the Dutch storefront of the App Store tailored for the Dutch market, involve facilitating app and in-app product purchases. Acting as the exclusive distributor for third-party apps, Apple Ireland exerts control over the offered content.

Applying the criteria from flyLAL, the court concludes that the Handlungsort is situated in the Netherlands. However, the court agreed that the specific court within the Netherlands responsible for adjudicating the matter remains unspecified.

The court initiated its analysis of the Erfolgsort based on the established premise in CJEU case law which posits that there is no distinction between individual and collective actions when determining the location of the damage. The court clarified that the concept of the place where the damage occurs does not encompass any location where the consequences of the event may be felt; rather, only the damage directly resulting from the committed harm should be considered. Moreover, the court emphasized that when determining the Erfolgsort, there is no distinction based on whether the legal basis for the accusation of anticompetitive practices is grounded in Article 101 or Article 102 TFEU.

The court reiterated that the App Store with Dutch storefront is a targeted online sales platform for the Dutch market. Functioning as an exclusive distributor, Apple Ireland handles third-party apps and in-app products, contributing to an alleged influence of anticompetitive behavior in the Dutch market. It’s acknowledged that the majority of users making purchases reside in the Netherlands, paying through Dutch bank accounts, thus placing the Erfolgsort within the Netherlands for this user group. Nevertheless, the court reiterated that the particular court within the Netherlands tasked with adjudicating this case remains unspecified.

The questions referred

Despite the court having its perspective on establishing jurisdiction under Article 7(2) Brussels I-bis Regulation, it opted to seek clarification from the CJEU for the following reasons.

First, the court expresses reservations regarding the complete applicability of the flyLAL precedent to the current case. It emphasizes that the flyLAL case involved a precise location where the damage could be pinpointed. In contrast, the present case involves anticompetitive practices unfolding through an online platform accessible simultaneously in every location within a particular Member State and globally. The court is uncertain whether the nature of this online distribution makes a significant difference in this context, especially when considering whether the case involves a collective action.

Second, as mentioned above, the WAMCA stipulates that only a single representative action can be allowed to proceed for a given event. In situations where multiple representative foundations aim to commence legal proceedings for the same event without reaching a settlement by a specific stage in the proceedings, the court will designate an exclusive representative. In addition to that, Article 220 Dutch Code of Civil Procedure offers the opportunity to consolidate cases awaiting resolution before judges in various districts and involving identical subject matter and parties, allowing for a unified hearing of these cases.

Nevertheless, the court has reservations about the compatibility of relocating from the Erfolgsort within a Member State under the consolidation of proceedings, as Article 7(2) Brussels I-bis Regulation impacts the establishment of jurisdiction within that Member State. In questioning whether such relocation would run contrary to EU law, the court highlights the Brussels I-bis Regulation’s overarching objective of preventing parallel proceedings. This triggers a skepticism towards the interpretation that each District Court within the Netherlands would have competence to adjudicate a collective action pertaining to users situated in the specific Erfolgsort within their jurisdiction.

However, the court finds it necessary to refer these questions to the CJEU, considering that, in its assessment, the CJEU’s rationale in Case C?30/20 Volvo (ECLI:EU:C:2021:604) is not easily transposable to the current case. In Volvo, the CJEU permitted the concentration of proceedings in antitrust matters within a specialized court. This is not applicable here, as the consolidation of proceedings under the described framework arises from the efficiency in conducting the proceedings, not from specialization.

These are, in a nutshell, the reasons why the District Court of Amsterdam decided to refer the following questions to the CJEU:

 Question 1

1. What should be considered as the place of the damaging action in a case like this, where the alleged abuse of a dominant position within the meaning of Article 102 TFEU has been implemented in a Member State through sales via an online platform managed by Apple that is aimed at the entire Member State, with Apple Ireland acting as the exclusive distributor and as the developer’s commission agent and deducting commission on the purchase price, within the meaning of Article 7, point 2, Brussels I bis? Is it important that the online platform is in principle accessible worldwide?
2. Does it matter that in this case it concerns claims that have been instituted on the basis of Article 3:305a of the Dutch Civil Code by a legal entity whose purpose is to represent the collective interests of multiple users who have their seat in different jurisdictions (in the Netherlands: districts) within a Member State under its own right?
3. If on the basis of question 1a (and/or 1b) not only one but several internally competent judges in the relevant Member State are designated, does Article 7, point 2, Brussels I bis then oppose the application of national (procedural) law that allows referral to one court within that Member State?

 Question 2

1. Can in a case like this, where the alleged damage has occurred as a result of purchases of apps and digital in-app products via an online platform managed by Apple (the App Store) where Apple Ireland acts as the exclusive distributor and commission agent of the developers and deducts commission on the purchase price (and where both alleged abuse of a dominant position within the meaning of Article 102 TFEU has taken place and an alleged infringement of the cartel prohibition within the meaning of Article 101 TFEU), and where the place where these purchases have taken place cannot be determined, only the seat of the user serve as a reference point for the place where the damage has occurred within the meaning of Article 7, point 2, Brussels I bis? Or are there other points of connection in this situation to designate a competent judge?
2. Does it matter that in this case it concerns claims that have been instituted on the basis of Article 3:305a of the Dutch Civil Code by a legal entity whose purpose is to represent the collective interests of multiple users who have their seat in different jurisdictions (in the Netherlands: districts) within a Member State under its own right?
3. If on the basis of question 2a (and/or 2b) an internally competent judge in the relevant Member State is designated who is only competent for the claims on behalf of a part of the users in that Member State, while for the claims on behalf of another part of the users other judges in the same Member State are competent, does Article 7, point 2, Brussels I bis then oppose the application of national (procedural) law that allows referral to one court within that Member State?

 [Translation from Dutch by the author, with support of ChatGPT]

Discussion

The CJEU possesses case law that could be construed in a manner conducive to allowing the case to proceed in the Netherlands. Notably, Case C?251/20 Gtflix Tv (ECLI:EU:C:2021:1036) appears to be most closely aligned with this possibility, wherein the eDate rule was applied to a case involving French competition law, albeit the CJEU did not explicitly address this aspect (though AG Hogan did). Viewed from this angle, the Netherlands could be deemed the centre of interests for the affected users, making it a potential Erfolgsort.

Regarding the distinction between individual and collective proceedings, the CJEU, in Cases C-352/13 CDC (ECLI:EU:C:2015:335) and C-709/19 VEB v. BP (ECLI:EU:C:2021:377), declined to differentiate for the purpose of determining the locus of damage. We find no compelling reason for the CJEU to deviate from this precedent in the current case.

The truly intricate question centers on the feasibility of consolidating proceedings in a single court. In Case C-381/14 Sales Sinués (ECLI:EU:C:2016:252), the CJEU established that national law must not hinder consumers from pursuing individual claims under the Unfair Contract Terms Directive (UCTD – 93/13) by employing rules on the suspension of proceedings during the pendency of parallel collective actions. However, it is unclear whether this rationale can be extrapolated to parallel concurrent collective actions.

Conclusion

This referral arrives at a good time, coinciding with the recent coming into force of the Representative Actions Directive (RAD – 2020/1828) last summer. Seeking clarification on the feasibility of initiating collective actions within the jurisdictions of affected users for damages incurred in the online sphere holds significant added value. Notably, the inclusion of both the Digital Services Act and the Digital Markets Act within the purview of the RAD amplifies the pertinence of these questions.

Moreover, this case may offer insights into potential avenues for collective actions grounded in the GDPR. Such actions, permitted to proceed under Article 7(2) Brussels I-bis Regulation, as exemplified in our earlier analysis of the TikTok case in Amsterdam, share a parallel rationale. The convergence of these legal frameworks could yield valuable precedents and solutions in navigating the complex landscape of online damages and collective redress.

Marco Buzzoni, Doctoral Researcher at the Luxembourg Centre for European Law (LCEL) and PhD candidate at the Sorbonne Law School, offers a critical analysis of some recent rulings by the Court of Justice of the European Union in matters of data protection.

In a series of three preliminary rulings issued on 14^th December and 21^st December 2023, the Court of Justice of the European Union (‘CJEU’) was called upon again to rule on the interpretation of Article 82 of the General Data Protection Regulation (‘GDPR’). While these rulings provide some welcome clarifications regarding the civil liability of data controllers, their slightly inconsistent reasoning will most likely raise difficulties in future cases, especially those involving cross-border processing of personal data.

On the one hand, the judgments handed down in Cases C-456/22, Gemeinde Ummendorf, and C-340/21, Natsionalna agentsia za prihodite, explicitly held that three elements are sufficient to establish liability under Article 82 GDPR. In so doing, the Court built upon its previous case law by confirming that the right to compensation only requires proof of an infringement of the Regulation, some material or non-material damage, and a causal link between the two. On the other hand, however, the Court seemingly swayed away from this analysis in Case C-667/21, Krankenversicherung Nordrhein, by holding that a data controller can avoid liability if they prove that the damage occurred through no fault of their own.

In reaching this conclusion, the Court reasoned that imposing a strict liability regime upon data controllers would be incompatible with the goal of fostering legal certainty laid out in Recital 7 GDPR. By introducing a subjective element that finds no mention in the Regulation, the Court’s latest decision is nonetheless likely to raise difficulties in cross-border cases by introducing some degree of unpredictability with respect to the law applicable to data controllers’ duty of care. In time, this approach might lead to a departure from the autonomous and uniform reading of Article 82 that seemed to have prevailed in earlier cases.

The Court’s Rejection of Strict Liability for Data Controllers

According to the conceptual framework laid out by the CJEU in its own case law, compensation under Article 82 GDPR is subject to three cumulative conditions. These include an infringement of the Regulation, the presence of some material or non-material damage, and a causal link between the two (see Case C-300/21, UI v Österreichische Post AG, para 32). In the cases decided in December 2023, the Court was asked to delve deeper into each of these elements and offer some additional guidance on how data protection litigation should play out before national courts.

In case C-456/22, the CJEU was presented with a claim for compensation for non-material damage filed by an individual against a local government body. The plaintiff alleged that their data protection rights had been breached when the defendant intentionally published documents on the internet that displayed their unredacted full name and address without their consent. Noting that this information was only accessible on the local government’s website for a short time, the referring court asked the CJEU to clarify whether, in addition to the data subject’s mere short-term loss of control over their personal data, the concept of ‘non-material damage’ referred to in Article 82(1) of the GDPR required a significant disadvantage and an objectively comprehensible impairment of personal interests in order to qualify for compensation. Rather unsurprisingly, the Court (proceeding to judgment without an Opinion) answered this question in the negative and held that, while Article 82(1) GDPR requires proof of actual damage, it also precludes any national legislation or practice that would subject it to a “de minimis threshold” for compensation purposes.

In doing so, the Court followed the road map outlined in UI v Österreichische Post AG, which had already held that the concept of damage should receive an autonomous and uniform definition under the GDPR (Case C-456/22, para 15, quoting Case C-300/21, paras 30 and 44) and should not be limited to harm reaching a certain degree of seriousness. Arguably, however, the Court also went beyond its previous decision by stating that the presence of an infringement, material or non-material damage, and a link between the two were not only “cumulative” or “necessary” but also “sufficient” conditions for the application of Article 82(1) (Case C-456/22, para 14). Remarkably, the Court did not mention any other condition that could have excluded or limited the data subject’s right to compensation. Taken literally, this decision could thus have been understood as an implicit endorsement of a strict liability regime under the GDPR.

This impression was further strengthened by the judgment handed down in Case C-340/21, where the Court was asked to weigh in on the extent of a data controller’s liability in case of unauthorised access to and disclosure of personal data due to a “hacking attack”. In particular, one of the questions referred to the CJEU touched upon whether the data controller could be exempted from civil liability in the event of a personal data breach by a third party. Contrary to the Opinion delivered by AG Pitruzzella, who argued that the data controller might be exonerated by providing evidence that the damage occurred without negligence on their part (see Opinion, paras 62-66), the CJEU ignored once more the question of the data controller’s fault and rather ruled that the latter should establish “that there [was] no causal link between its possible breach of the data protection obligation and the damage suffered by the natural person” (Case C-340/21, para 72).

A few days later, however, the CJEU explicitly endorsed AG Pitruzzella’s reading of Article 82 GDPR in Case C-667/21. In a subtle yet significant shift from its previous reasoning, the Court there held that the liability of the data controller is subject to the existence of fault on their part, which is presumed unless the data controller can prove that they are in no way responsible for the event that caused the damage (Case C-667/21, holding). To reach this conclusion, The Court relied on certain linguistic discrepancies in Article 82 of the GDPR and held, contrary to the Opinion by AG Campos Sánchez-Bordona, that a contextual and teleological interpretation of the Regulation supported a liability regime based on presumed fault rather than a strict liability rule (Case C-667/21, paras 95-100). Formulated in very general terms, the holding in Case C-667/21 thus suggests that a controller could be released from liability not only if they prove that their conduct played no part in the causal chain leading to the damage but also — alternatively — that the breach of the data subject’s rights did not result from an intentional or negligent act on their part.

Lingering Issues Surrounding the Right to Compensation in Cross-Border Settings

According to the CJEU, only a liability regime based on a rebuttable presumption of fault is capable of guaranteeing a sufficient degree of legal certainty and a proper balance between the parties’ interests. Ironically, however, the Court’s approach in Case C-340/21 raises some significant methodological and procedural questions which might lead to unpredictable results and end up upsetting the parties’ expectations about their respective rights and obligations, especially in cases involving cross-border processing of personal data.

From a methodological perspective, the CJEU’s latest ruling does not fit squarely within the uniform reading of the GDPR that the Court had previously adopted with respect to the interpretation of Article 82 GDPR. In the earlier cases, in fact, the CJEU had consistently held that the civil liability requirements laid out in the Regulation, such as the notion of damage or the presence of an actual infringement of data protection laws, should be appreciated autonomously and without any reference to national law (on the latter, see in particular Case C-340/21, para 23). On the other hand, however, the Court has also made clear that if the GDPR remains silent on a specific issue, Member States should remain free to set their own rules, so long that they do not conflict with the principles of equivalence and effectiveness of EU law (on this point, see eg Case C-340/21, para 59).

Against this backdrop, the Court’s conclusion that the civil liability regime set up by the legislature implicitly includes the presence of some fault on the defendant’s part begs the question of whether this requirement should also receive a uniform interpretation throughout the European Union. In favour of this interpretation, one could argue that this condition should be subject to the same methodological approach applicable to the other substantive requirements laid out in Article 82 GDPR. Against this position, it could nonetheless be pointed out that in the absence of explicit indications in this Article, the defendant’s fault should be assessed by reference to national law unless another specific provision of the Regulation (such as Articles 24 or 32 of the GDPR) specifies the degree of care required of the data controller or processor. In the context of cross-border cases, the latter interpretation would thus allow each Member State to determine, based on their own conflict-of-laws rules, the law applicable to the defendant’s duty of care in cases of violations of data protection laws. If generalised, this approach might in time lead to considerable fragmentation across the Member States.

In addition to these methodological difficulties, the Court’s decision in Case C-340/21 also raises some doubts from a procedural point of view. In holding that the data controllers’ liability is subject to the existence of fault on their part, the CJEU calls into question the possible interaction between national court proceedings aimed at establishing civil liability under Article 82 GDPR and administrative decisions adopted by data protection authorities. With respect to the latter, the CJEU had in fact ruled in Case C-683/21, Nacionalinis visuomenés sveikatos centras, that Article 83 GDPR must be interpreted so that an administrative fine may be imposed pursuant to that provision “only where it is established that the controller has intentionally or negligently committed an infringement referred to in paragraphs 4 to 6 of that article” (Case C-683/21, holding). In other words, national supervisory authorities are also called upon to assess the existence of fault on the part of the data controller or processor before issuing fines for the violation of data protection laws.

At first glance, the CJEU’s decision in Case C-340/21 fosters some convergence between the private and public remedies set out in the GDPR. In reality, however, this interpretation might potentially create more hurdles than it solves. Indeed, future litigants will likely wonder what deference, if any, should be given to a supervisory authority’s determinations under Article 83 GDPR within the context of parallel court proceedings unfolding under Article 82. In a similar context, the Court has already held that the administrative remedies provided for in Article 77(1) and Article 78(1) GDPR may be exercised independently and concurrently with the right to an effective judicial remedy enshrined in Article 79 GDPR, provided that national procedural rules are able to ensure the effective, consistent and homogeneous application of the rights guaranteed by the Regulation (see Case C-132/21, Nemzeti Adatvédelmi és Információszabadság Hatóság v BE). Should the same principles apply to actions brought under Article 82 GDPR? If so, should the same rule also extend to conflicts between national court proceedings and decisions issued by foreign supervisory authorities (and vice-versa), even though each of them might have a different understanding of the degree of protection afforded by the Regulation?

Despite the CJEU’s laudable attempt to strike a balance between the interests of personal data controllers and those of the individuals whose data is processed, it is not certain that the Court has fully assessed all the consequences of its decision. Ultimately, in fact, the choice to reject a strict liability rule could lead not only to unequal protection of individual rights within the EU but also to major uncertainties for economic operators regarding the extent of their own liability under the GDPR.

The Max Planck Institute for Social Anthropology in Halle (Saale), Germany, is advertising two PhD positions in Private Law within the context of a research project on “Cultural Diversity in
Private Law” lead by Dr Mareike Schmidt.

More information can be found here.

This post is part of a series regarding Colonialism and the general structure of (German) Private International Law, based on a presentation I gave in spring 2023. See the introduction here.

As mentioned in the introduction, this series does not intent to automatically pass judgment on a norm or method influenced by colonialism as inherently negative. Instead, the aim is to reveal these influences and to initiate a first engagement with and awareness of this topic and to stimulate a discussion and reflection.

The first post (after the introduction) dealt with classic PIL and colonialism. This second considered structures and values inherent in German or European law, implicitly resonating within the PIL and, thus, expanding those values to people and cases from other parts of the world.  The third category discusses an imagined hierarchy between the Global North and Global South that is sometimes inherent in private international law thinking. The fourth and for the moment last (but not least) category deals with PIL rules that allow or at least contribute to the exploitation of a power asymmetry between parties from the Global North and the Global South. For example, this power and negotiation asymmetry, in conjunction with generous rules on party autonomy, can lead to arbitration and choice of law clauses being (ab)used to effectively undermine rights of land use under traditional tribal law.

After the first post, in the comment section a discussion evolved regarding the (non-)application of tribal law. One question asked for an example. This post can also (hopefully) serve as such an example.

1. Party Autonomy in German and EU PIL

One value inherent to the German and EU legal systems is that of private and party autonomy. It reflects and expresses the individualism of the Enlightenment and a neo-liberal social order and is recognised today, at least in part, as one of the “universal values” of PIL. However, the choice of law and, thus, party autonomy as a core connecting factor or method of PIL can lead to the exploitation of negotiation asymmetries in the relationship between companies in the Global North and states or companies in the Global South, particularly to the detriment of the population in the Global South, by avoiding state control and socially protective regulations.

2. “Land Grabbing” as an Example

“Land grabbing” refers to, among other things, the procedure used by foreign investors to acquire ownership to or rights to exploit territories in former colonies. The contract is concluded with the landowner, often the state, and includes an arbitration and choice of law clause, often within the framework of bilateral investment protection agreements. The use of the land can conflict with the collective, traditional use by certain local groups, which is based on customary and tribal law. Such rights of land use were often only fought for politically after the former colony gained independence, while the original colonial legal system overrode indigenous rights of use (see also former posts here and see the  discussion in the comment section of the post). These land use rights of indigenous groups often stem from public law and are conceived as protection rights of the indigenous population, who are thus authorised to live on their traditional land.

The arbitration agreement and the choice of law clause make it possible for legal disputes to be settled before a private arbitration tribunal. The tribes concerned, as they are not part of the treaty on the land and its use, can only become parties to the legal dispute with difficulty. Furthermore, they may not have knowledge of the treaty and the arbitration clause or the possibility to start a proceeding at the tribunal. In addition, a law applicable to the contract and its consequences may be chosen that does not recognise the right of land use based on tribal law. If the arbitrator, not knowing about the not applicable tribal law or the existence of the tribe, makes a decision based on the chosen law, the decision can subsequently become final and enforceable. This may force the tribes using the land having to vacate it as property disturbers without being able to take legal action against it.

3. Party Autonomy and Colonialism

This possibility of “land grabbing” is made possible by the fact that a state – often a former colony – has a high interest in attracting foreign investment. She, therefore, tries to organise its own legal system, and therefore also her conflict of laws, in an investment-friendly manner and accommodate the investor in the contract. The generous granting of party autonomy and individual negotiating power plays a key role here. A domino effect can be observed in former colonies, where a legal system follows that of neighbouring states once they have attracted foreign investment in order to be able to conclude corresponding agreements. The endeavours of states to introduce a liberal economy form, which is reflected in party autonomy in PIL, can therefore also express a structural hierarchy and form of neo-colonialism. It also indirectly revives the original behaviour of the colonial rulers towards the indigenous peoples with the support of the central state (see former post).

4. Assessment of “Land Grabbing”

If the aforementioned power asymmetry is not counter weighted, arbitration and choice of law clauses can lead to an avoidance of unwanted laws, such as those granting traditional land use rights to local tribes. From a German domestic perspective, the problem arises that the enforcement of (one’s own) local law is a matter for the foreign state. A case where local law will be addressed before German courts will be scarce, esp. in the case of an arbitration proceeding. German courts only come into contact with the legal dispute if an arbitration proceeding has already resulted in a legally binding award and this award is now to be enforced in Germany. In my opinion, this case has to be handled in the same procedure proposed in a former post for the integration of local, non-applicable law. If foreign tribal law is mandatory in the state in question, for example, because there is an obligation under international and domestic law, the arbitral tribunal should be presumed to also observe this obligation as an internationally mandatory norm, irrespective of which lex causae applies. When enforcing the arbitral award domestically, the declaration of enforceability should be prohibited on the grounds of a violation of public policy if the arbitral tribunal has not complied with this obligation.

Furthermore, the use of party autonomy could be more strictly controlled and restrictively authorised when special domestic values and interests of third parties are at stake, as can be the case in particular with the use of land. The lex rei sitae might be more appropriate without allowing for a choice of law.

Finally, restrictions on party autonomy in cases in which negotiation asymmetries are assumed are not unknown to German and European PIL. So, ideas from these rules could be taken up and consideration could be given to which negotiation asymmetries could arise in relation to non-European states. For example, certain types of contract that are particularly typical of power asymmetries could be provided with special protection mechanisms similar to consumer contracts under Art. 6 Rome I Regulation. But that is an international problem that should be discussed on the international level. Therefore, the international community could work towards an international consensus in arbitration proceedings that, for example, property law issues are subject to the lex rei sitae and are not open to a choice of law. Similarly, there could be a discussion whether safeguards should ensure that no choice of law can be made to the detriment of third parties and that, where applicable, participation rights must be examined in arbitration proceedings. Many legal systems already provide those saveguards, so this would not come as a huge novelty.

However, it would also be paternalistic and neo-colonialist if such considerations originated in the Global North without involving the countries to which they refer. It would therefore be desirable to have a stronger and more enhanced dialogue with countries from the Global South that also allows representatives of the local population and local communities to have their say, so that these interests and possibilities for exploiting negotiation asymmetries can be better taken into account.

5. Epilogue

This series has tried to start a debate about Colonialism and Private International Law from the point of view of German PIL. Posts from other jurisdictions might follow. It is a very complex topic and this series only scratched on its surface. As writen in the introduction, I welcome any comments, experiences and ideas from other countries and particularly from countries that are former colonies.

This post is part of a series regarding Colonialism and the general structure of (German) Private International Law, based on a presentation I gave in spring 2023. See the introduction here.

As mentioned in the introduction, this series does not intent to automatically pass judgment on a norm or method influenced by colonialism as inherently negative. Instead, the aim is to reveal these influences and to initiate a first engagement with and awareness of this topic and to stimulate a discussion and reflection.

The first post (after the introduction) dealt with classic PIL and colonialism and already sparked a vivid discussion in the comments section. This second considered structures and values inherent in German or European law, implicitly resonating within the PIL and, thus, expanding those values to people and cases from other parts of the world.  The third category discusses an imagined hierarchy between the Global North and Global South that is sometimes inherent in private international law thinking, for instances where courts or legislators abstractly or paternalistically apply the public policy to “protect” individuals from foreign legal norms. This is especially evident in areas like underage marriages and unilateral divorce practices found inter alia in Islamic law.

1. The public policy exception – abstract or concrete control?

The public policy exception is intended to prevent the application of foreign law by way of exception if the result of this application of law conflicts with fundamental domestic values. Such control is necessary for a legal system that is open to the application of foreign law and, in particular, foreign law of a completely different character. German law is typically very restrictive in its approach: The public policy control refers to a concrete control of the results of applying the provisions in question. In addition, the violation of fundamental domestic values must be obvious and there must be a sufficient domestic connection. In other countries, the approach is less restrictive. In particular, there are also courts that do not look at the result of the application of the law, but carry out an abstract review, i.e. assess the foreign legal system in the abstract. For a comparison of some EU Member States see this article.

2. Explicit paternalistic rules

Furthermore, there are some rules that exercise an abstract control of foreign law. Article 10 of the Rome III Regulation contains a provision that analyses foreign divorce law in the abstract to determine whether it contains gender inequality. According to this (prevailing, see e.g. conclusions of AG Saugmandsgaard Øe) interpretation, it is irrelevant whether the result of the application of the law actually leads to unequal treatment. This abstract assessment assumes – even more so than a review of the result – an over-under-ordering relationship between domestic and foreign law, as the former can assess the latter as “good” or “bad”.

Even beyond the ordre public control, there has recently been a tendency towards “paternalistic rules”, particularly triggered by the migration movements of the last decade. The legislator seems to assume that the persons concerned must be protected from the application of “their” foreign law, even if they may wish its application. In particular, the “Act to Combat Child Marriage” which was only partially deemed unconstitutional by the Federal Constitutional Court (see official press release and blog post), is one such example: the legislator considered the simple, restrictive ordre public provision to be insufficient. Therefore, it created additional, abstract regulations that block the application of foreign, “bad” law.

3. Assessment

In the described cases as a conceptual hierarchy can be identified: The impression arises that foreign legal systems, particularly from the “Global South”, are categorised in the abstract as “worse” than the German/EU legal system and that persons affected by it must be protected from it (“paternalistic norms”). As far as I can see there is a high consensus in the vast majority of German literature (but there are other voices) and also the majority of case law that the abstract ordre public approach should be rejected and that the aforementioned norms, i.e. in particular Art. 13 III EGBGB (against underage marriages) and Art. 10 Rome III-VO (different access to a divorce based on gender), should ideally be abolished. It would be desirable for the legislator to take greater account of the literature in this regard.

This interesting case comment has been kindly provided to the blog by Nicolás Zambrana-Tévar, LLM, PhD, KIMEP University

The United States Court of Appeals for the Ninth Circuit has found in favor of Spain as defendant in a property case spanning several decades. A panel of three judges has unanimously ruled that, applying California conflict of law rules, Spain has a stronger interest than the claimants in the application of its own domestic law, including its own rules on prescriptive acquisition of property and the statute of limitations, thus confirming the ownership of a stolen painting, now owned by a Spanish museum.

1. Background information

In 1939, Lilly Cassirer traded a Pissarro painting to the Nazis in exchange for her family´s safe passage out of Germany. In 1954, a tribunal set up by the Allied forces established that the Cassirer family were the rightful owners of the painting. However, believing that the painting had been lost during the war, the family accepted 13,000 US dollars in compensation from the German government, which would be the equivalent of 250,000 US dollars today.
After the painting was looted, it found its way into the United States and, in 1976, Baron Hans Heinrich Thyssen-Bornemisza bought it from the Hahn Gallery of New York, where the painting was publicly in display, allegedly ignoring its origin. The Museum Thyssen-Bornemisza purchased the painting from the Baron in 1993. Claude Cassirer – the grandson of Lilly Cassirer – found out that the painting was being exhibited in Madrid and commenced proceedings under the Foreign Sovereign Immunities Act (FSIA) in 2005. The Museum is the actual defendant in the suit but it is considered an instrumentality of the Kingdom of Spain.

2. Court decisions

In 2019, a US District Judge for the Central District of California, applying Spanish law, found that court filings did not demonstrate a “willful blindness” on the part of the Museum, when it added the painting to its collection. Moreover, the judge found that it could not force Spain or the Museum to comply with the “moral commitments” of international agreements concerning the return of works of art looted by the Nazis.
In 2020, the US Court of Appeals for the Ninth Circuit found in favor of Spain, again applying Spanish law. The court ruled that, regardless of the test applied by the district judge to determine the degree of care employed by the purchaser to determine the origin of the painting, both the Baron in 1976 and the Museum in 1993, lacked actual knowledge of the theft. It is important to note that both the district judge and the court of appeals determined the application of Spanish law because they were applying federal choice of law rules.
In 2022, the US Supreme Court ruled that this case did not involve any substantive federal law issues because it basically dealt with property law. Therefore, the choice of law rules that the district judge and the court of appeals should have applied were the conflict rules of the forum state, i.e. the conflict rules of California. The Supreme Court argued that Spanish law “made everything depend on whether, at the time of acquisition, the Foundation knew the painting was stolen”. On the other hand, the claimants argued that California conflict rules led to the application of California property law, in accordance with which “even a good-faith purchaser of stolen property cannot prevail against the rightful pre-theft owner.” Basically, the Supreme Court said that in an FSIA case, the foreign state defendant has to be treated like a private defendant and that if the Museum had been a purely private entity, it would have had to return the painting. The case was returned to the Court of Appeals.

3. Conflict-of-law analysis

On 9 January 2024, the US Court of Appeals ruled that, even applying California choice of law rules, Spanish law was applicable. The court came to this conclusion applying the “governmental interest approach”. In accordance with this approach, the court first had to ascertain that the two laws in conflict – Spain and California law – were different. They were because the Spanish law provision that the defendant was relying on was article 1955 of the Spanish Civil Code, which provides that “Ownership of movable goods prescribes by three years of uninterrupted bona fide possession. Ownership of movable goods also prescribes by six years of uninterrupted possession, without any other condition”. Therefore, in accordance with Spanish law “three years of uninterrupted possession in good faith” are enough for the acquisition of title whereas California law has not expressly adopted a doctrine of adverse possession for personal property – such as works of art – and, moreover, “thieves cannot pass good title to anyone, including a good faith purchaser”. Besides, California law extends to six years the statute of limitations for claims involving the return of stolen property and Cassirer brought the claim only five years after it discovered the painting hanging at the Museum in Madrid.
Having determined that the laws in conflict were different, the court of appeals then examined and agreed that both jurisdictions – Spain and California – “have a legitimate interest in applying their respective laws on ownership of stolen personal property”. “Spanish law assures Spanish residents that their title to personal property is protected after they have possessed the property in good faith for a set period of time, whereas California law seeks to deter theft, facilitate recovery for victims of theft, and create an expectation that a bona fide purchaser for value of movable property under a ‘chain of title traceable to the thief,’ … does not have title to that property.” Therefore, there was a true conflict of laws, as both jurisdictions had real and legitimate interests in applying their respective law. Additionally, the court had to determine which jurisdiction’s interest “would be more impaired if its policy were subordinated to the policy of the other state.” Otherwise said, “which jurisdiction should be allocated the predominating lawmaking power under the circumstances of the present case”.
To do this, the interests of each jurisdiction were to be measured based on “the circumstances of the particular dispute, not the jurisdiction’s general policy goals expressed in the laws implicated”. The factors to be taken into consideration in this analysis were the “current status of a statute… the location of the relevant transactions and conduct… and the extent to which one jurisdiction’s laws either impose similar duties to the other jurisdiction’s laws, or are accommodated by the other jurisdiction’s laws, such that the application of the other jurisdiction’s laws would only partially—rather than totally—impair the interests of the state whose law is not applied”.
With respect to the first factor, the court said that it was inappropriate to judge which law is better. Also, in reply to the alleged archaism of the Spanish rule, that says that property is acquired after six years of possession, regardless of the stolen nature of the asset, the court replied that the defendant was relying on the possession with good faith during three years.
With respect to the second factor, the court of appeals reasoned that, in accordance with several precedents from the Supreme Court of California, a “jurisdiction ordinarily has the predominant interest in regulating conduct that occurs within its borders”, i.e. on Spanish territory, whereas “where none of the relevant conduct occurs in California, a restrained view of California’s interest in facilitating recovery for one of its residents is warranted.” In the case at hand, “California’s sole contact to the dispute was the happenstance of the plaintiff’s residence there.” Similarly, “California’s governmental interest rests solely on the fortuity that Claude Cassirer moved to California in 1980, at a time when the Cassirer family believed the Painting had been lost or destroyed.” Therefore, “California’s interest in facilitating recovery for that resident was minimal and the extraterritorial reach of its laws was restrained.” Since “no relevant conduct with respect of the Painting occurred in California, the impairment of California’s interest that would result from applying Spanish law would be minimal.”
The court went on to say that, in contrast, “applying California law would significantly impair Spain’s interest in applying Article 1955 of the Spanish Civil Code. For one, because the relevant conduct [the purchase of the painting] occurred in Spain” so that “Spain has the “predominant interest in applying its laws to that conduct.” Furthermore, “applying California law would mean that Spain’s law would not apply to property possessed within Spain’s borders, so long as the initial owner (1) happened to be a California resident (a fact over which… the defendant has no way of knowing or controlling…, and (2) the California resident did not know where the property is located and who possessed it. Applying California law based only on Claude Cassirer’s decision to move to California would strike at the essence of a compelling Spanish law.”
With respect to the third factor and also in accordance with past precedents of the California Supreme Court, “the court should look to whether one jurisdiction’s laws accommodate the other jurisdiction’s interests or imposes duties the other jurisdiction already imposes… A state’s laws can more readily be discarded if the failure to apply its laws would only partially—rather than totally—impair the policy interests of the jurisdiction whose law is not applied…. Here, the failure to apply California’s laws would only partially undermine California’s interests in deterring theft and returning stolen art to victims of theft, which provides further support for limiting the extraterritorial reach of California’s laws to this dispute.
On the other hand, “applying Spanish law would only partially undermine California’s interests in facilitating recovery of stolen art for California residents. California law already contemplates that a person whose art—or other personal property—is stolen may eventually lose the ability to reclaim possession: namely, if the person fails to bring a lawsuit within six years after he discovers the whereabouts of the art… Similarly, Article 1955 of the Spanish Civil Code accommodates California’s interest in deterring theft. As we have explained, Spanish law makes it more difficult for title to vest in an “encubridor,” which includes, “an accessory after the fact,” or someone who “knowingly receives and benefits from stolen property…. If the possessor is proven to be an encubridor, Spanish law extends the period in which the property must be possessed before new prescriptive title is created.”

4. Concluding remarks

This complex and interesting case seems to be coming to an end. In brief, and despite the complexity of the application of the theory of interest analysis, it seems that the US court has given the same solution which a civil court would have given, applying the usual rule that the law applicable to property rights is the law of the place where the property is located at the time of the transfer. So far, it appears that the increasing sensitivity towards cultural property and towards unraveling war crimes has not fully displaced this conflicts rule.