Civil Liability under the Revised CSDDD: Why Private International Law Matters More Than Ever

, ,

This post was written by Prof. Dr. Giesela Rühl, LL.M. (Berkeley), Humboldt University of Berlin.


The European Corporate Sustainability Due Diligence Directive (CSDDD) has had a turbulent start. Adopted in 2024 after heated debates, it was substantially scaled back in early 2026 before the deadline for its transposition had even expired. Against the backdrop of growing opposition to European sustainability legislation and out of concern for the competitiveness of European companies the European legislature reduced the number of companies covered and relaxed some of the Directive’s due diligence requirements. Most importantly, however, it removed Article 29(1) and (7) CSDDD, which had required Member States to introduce a civil liability regime and to ensure that national liability provisions would apply irrespective of the law otherwise applicable under the Rome II Regulation.

The deletion of both Article 29(1) and Article 29(7) CSDDD raises (at least) two questions. First, are Member States (still) required to introduce a civil liability regime where none currently exists under national law – or is the decision whether to provide for one left to their discretion? Second, if Member States decide to establish a civil liability regime, how can they ensure that the respective provisions actually apply?

Both questions have significant practical implications. Their answers determine whether victims of human rights violations will be able to recover compensation for harm they have suffered. The following post argues that Member States should establish a civil liability regime even in the absence of a clear legal obligation to do so. In addition, they should classify these provisions as overriding mandatory norms and take additional steps to ensure that victims can actually pursue damages claims before European courts.

Trying to Understand what the European Legislature had in Mind

The question of whether companies should incur liability for breaches of human rights due diligence obligations has been the subject of intense debate since negotiations on the Corporate Sustainability Due Diligence Directive began in 2020. When the Directive was eventually adopted in 2024, the debate seemed to have been settled in favour of civil liability. However, in 2025 the European legislature decided to backtrack and remove the corresponding provision from the Directive. This seems to suggest that Member States are no longer required to provide for civil liability in case companies breach their due diligence obligations and thereby cause harm.

Closer scrutiny, however, casts doubt on this interpretation: The first sentence of Recital 49 of the revised CSDDD explains the decision to delete Article 29(1) CSDDD by reference to the principle of subsidiarity and speaks only of abandoning a “specific Union-wide liability regime”. In addition, the third sentence of Recital 49 states that Member States should ensure that, where a company does not comply with the Directive’s due diligence requirements and where such failure causes damage, victims are able to receive full compensation. It has, therefore, been argued that the European legislature did not want to remove the obligation to establish a civil liability as such but merely intended to leave the details of its design to the Member States. This interpretation is supported by various provisions of the Directive that refer to civil liability under or in accordance with Article 29 (cf. Article 6(1) Sentence 2, Article 10(6), Article 11(7), Article 14(7), Article 25(4) and (9) CSDDD) as well as by Article 29(2) and (3) CSDDD which continue to regulate various aspects of damages claims, notably limitation periods, litigation costs and the disclosure of evidence. All these provisions would be meaningless if Member States were under no obligation to establish a civil liability regime.

There are, however, also very good arguments for a different reading. To begin with, there is no denying the fact that the European legislature deleted Article 29(1) CSDDD altogether. It did not modify it. And it did not adjust it, even though it would have been easy to require Member States to provide for civil liability in accordance with their national law. It is therefore difficult to see why the complete removal of Article 29(1) CSDDD should not also be understood as removing the obligation to establish a civil liability regime for violations of the Directive. Admittedly, the European legislature introduced Recital 49 which calls on Member States to ensure that victims may claim compensation if companies violate their obligations under the Directive. A recital, however, does not compensate for the deletion of a provision in the operative part of the Directive. In fact, according to settled CJEU case law, recitals cannot establish obligations that have no basis in the operative provisions of a legislative instrument (CJEU, C-136-04, DeutschesMilch-Kontor, EU:C:2005:716, para. 32; CJEU, C-418/18 Puppinck, ECLI:EU:C:2019:1113, para 75). Specifically, they cannot establish obligations that were deliberately deleted from the instrument itself (cf. CJEU, C-397/16 and C-435/16, Acacia and D’Amato, ECLI:EU:C:2017:992, paras 33 ff.). In a similar vein, Article 29(2) and (3) cannot be understood as creating an obligation to establish a civil liability regime. While it may seem unconvincing to regulate the details of liability claims without requiring Member States to introduce a civil liability regime, Article 29(2) and (3) as well as the many other provisions of the Directive that make reference to civil liability under Article 29 are not without any meaning if Member States are under no such obligation. Specifically, they can be understood as regulating civil liability where it already exists under national law (as in France), where Member States decide to introduce it voluntarily or where liability arises from foreign law.

Establishing Civil Liability even in the Absence of a Clear Obligation to do so

Against this background, a fair reading of the Directive does not clearly establish whether Member States are required to introduce a civil liability regime or not. How, then, should they proceed when implementing the Directive? Three possible approaches can be distinguished:

First, Member States could take the position that, in the absence of a clear obligation under EU law, there is no need to take any action. It would then remain to be seen whether the European Commission would initiate infringement proceedings or whether the CJEU, following a reference for a preliminary ruling by a national court, would find the lack of civil liability to be incompatible with the Directive.

Second, Member States could choose to err on the side of caution and introduce civil liability for breaches of the Directive’s human rights due diligence obligations. Since the Directive undoubtedly permits Member States to introduce such liability voluntarily, doing so would eliminate the risk of violating the Directive and subsequently being required to establish a civil liability regime.

Third, Member States could set aside these considerations and ask whether civil liability is desirable in its own right. They would then realize that there are compelling reasons to supplement the public enforcement scheme envisaged by the Directive with a private enforcement mechanism: Civil liability puts enforcement in the hands of victims who have both strong incentives to take action and access to information about the harmful event. Moreover, civil liability does not depend on government resources or taxpayers’ money to be effective. Since the cost of civil litigation has to be borne by the parties, civil liability is a relatively inexpensive enforcement option (at least from the perspective of the state). Admittedly, arguments that can also be raised against civil liability. Most importantly, it can be said that exposing companies to damages claims creates financial risks and, thus, places an additionally burden on companies already burdened by the introduction of human rights due diligence obligations. However, it should be borne in mind that the European legislature consciously decided to impose due diligence obligations on (certain) companies and upheld this decision in the 2026 reform of the Directive. Member States should therefore not undermine this legislative choice but ensure that the obligations of the Directive are enforced effectively.

Taken together, the above considerations suggest that Member States should introduce a civil liability regime when implementing the Directive. While refraining from doing so would leave them exposed to the risk of violating EU law, providing for civil liability would not only eliminate that risk but also strengthen the enforcement of the Directive’s due diligence obligations. As regards the design of such a regime, Member States have two options: They can either follow the French model (cf. Article L. 225-102-5 Code de commerce, Article 1240, 1241 Code civil) and allow damages claims for violations of the Directive’s due diligence obligations under the general rules of national tort law. Or they can create a new, stand-alone civil liability regime specifically tailored to violations of the Directive’s human rights due diligence obligations. Which option is preferable will ultimately depend on the legal and institutional context of each Member State.

Introducing civil liability, however, can only be the first step. Given the inherently international nature of supply chain cases, Member States should also consider the private international law implications and address the many issues that may, in practice, prevent civil liability from creating incentives to comply with the Directive. Three of these deserve to be highlighted.

Making National Liability Rules Applicable across Borders

The first issue concerns the applicable law. As is well known, Article 4(1) Rome II Regulation generally subjects tort claims to the law of the place of injury. In supply chain cases, this will ordinarily be the law of a third country. Consequently, national liability provisions introduced to enforce the CSDDD would usually not apply. The analysis may change in cases involving environmental harm because Article 7 Rome II Regulation allows the claimant to rely on either the law of the place of injury under Article 4(1) or on the law of the place where the event giving rise to the damage occurred, i.e. the law of the place of action. Opinions, however, differ on whether the breach of diligence obligation by European companies at their European headquarters counts as an event giving rise to the damage leading to Europe being the place of action in the meaning of Article 7 Rome II Regulation. Some commentators answer this question in the affirmative, as did the District Court of The Hague in its landmark first instance Shell judgment (Rechtbank Den Haag, 26 May 2021, Case Az. C/09/571932, para. 4.3.6). Others favour a narrower understanding, locating the event giving rise to the damage exclusively at the site of the activity directly responsible for the environmental harm, such as the operation of the emitting facility. On this latter approach, the place of the event giving rise to the damage will frequently coincide with the place where the damage occurs. Failing a decision by the Court of Justice of the European Union, it remains unclear whether Article 7 Rome II Regulation can actually lead to the application of European civil liability provisions. In any event, application of the law of a Member State would be limited to environmental harm cases.

Against this background, the only way to ensure the systematic application of any civil liability regime that Member States might introduce when implementing the Directive is Article 16 Rome II Regulation. The provision allows Member States to apply overriding mandatory provisions of the forum irrespective of the otherwise applicable law. Member States could, therefore, say that their national civil liability provisions have overriding mandatory effect– and courts in that Member State would be required to apply them.

The revised Directive leaves Member States free to pursue this solution. Although Article 29(7) CSDDD no longer requires national liability provisions to be given overriding mandatory effect, Recital 49 expressly preserves the possibility of doing so. The deletion of Article 29(7) therefore removes an obligation imposed by EU law without excluding a corresponding choice at national level.

Using Article 16 Rome II Regulation would also be consistent with the general requirements established by the CJEU in Unamar for classifying national provisions as overriding mandatory in nature. In that decision the Court held that Member States enjoy considerable discretion in determining which provisions of national law are given overriding mandatory effect – as long as they comply with their obligations under EU primary and secondary law (CJEU, Case C-184/12, Unamar, ECLI:EU:C:2013:663, para. 46). Conferring overriding mandatory status on civil liability rules designed to enforce the Directive’s human rights due diligence obligations would certainly meet that bar. Specifically, using Article 16 Rome II Regulation to ensure their application would not violate Member States’ obligations under the Directive. On the contrary, as Recital 49 indicates, it would be perfectly in line with the Directive and contribute to its effectiveness.

Application of Article 16 Rome II Regulation would also not be barred by the CJEU’s recent judgment in HUK-COBURG. Admittedly, the Court stressed that the concept of overriding mandatory provisions must be “interpreted restrictively” and requires (1) that the legal situation in question has sufficiently close links with the forum and (2) that a “detailed analysis of the wording, general scheme, objectives and the context” shows that the provision “pursues … an essential public interest” that cannot be achieved by the application of the law designated pursuant to Article 4 Rome II Regulation. (CJEU, Case C-86/23, HUK-COBURG, ECLI:EU:C:2024:689, paras. 30 ff, 57). However, HUK-COBURG concerned a provision whose overriding mandatory character had not been expressly determined by the national legislature. The Court, therefore, had to establish whether the provision qualified as an overriding provision through interpretation. Whether the Court would apply and require the same degree of scrutiny where Member States expressly designated a provision as overriding mandatory is unclear. However, even if the requirements developed by the Court in HUK-COBURG had to be applied, civil liability rules enforcing the Directive would have a strong claim to meeting them: To begin with, the relevant provisions of national law would display a sufficiently close connection with the forum state because they would only apply to companies seated or active in the Member State that enacted the provision. In addition, the relevant national provisions would also serve to protect essential public interests. While they would help victims to claim compensation and, hence, protect private rights, human rights due diligence obligations and the corresponding civil liability rules would also promote responsible corporate conduct throughout global value chains and the integrity of the European single market. Finally, applying the law of the place of injury would, in all likelihood, not adequately protect these public interests.

Against this background, Member States could ensure application of their national civil liability provision via Article 16 Rome II Regulation. Likewise, existing liability provisions – such as the liability provisions to be found in the French Duty of Vigilance Law – could also be classified as overriding mandatory. Although this requires the HUK-COBURG criteria to bemet, one may be confident that this will usually be the case. A recent French judgment shares this view: On 12 March 2026, the Tribunal judiciaire de Paris held that the liability provisions of the French Duty of Vigilance Law qualify as overriding mandatory under Article 16 Rome II (Tribunal judiciaire de Paris, 12 March 2026, Yves Rocher, paras 48 et seq.). The court reached this conclusion – after careful consideration of HUK Coburg – even though the French law relies on the general rules of tort liability in Articles 1240 and 1241 of the Civil Code. The decision, which was discussed on this blog by Olivera Boskovic, illustrates the potential of Article 16 Rome II Regulation in supply chain cases.

Bringing Supply Chain Disputes before European Courts

Yet, classifying national liability rules  as overriding mandatory provisions will not in itself be sufficient to ensure their application in practice. Since non-European courts will not apply overriding mandatory provisions of Member States, victims will need to have access to European courts to benefit from the protection they afford. The second issue that Member States should address to ensure effective enforcement of the Directive is, therefore, international jurisdiction.

At first sight, international jurisdiction does not seem to pose many difficulties in supply chain cases: Under Article 4 Brussels Ia Regulation, companies may generally be sued in the Member State of their domicile. Victims can therefore usually bring damages claims against EU-based parent companies or buyers before EU courts. Difficulties, however, arise where victims seek to recover damages from an EU-based parent or buyer company alongside a subsidiary or supplier domiciled in a third country. Since the Brussels Ia Regulation generally does not govern jurisdiction over defendants domiciled outside the EU, it does not itself provide a basis for bringing a non-EU company before a European court. Instead, jurisdiction depends on the domestic jurisdictional rules of each Member State which differ considerably in their willingness to accommodate joint proceedings against both domestic and foreign defendants. Dutch law, for example, is fairly open, allowing joint claims where this is justified for reasons of efficiency (cf. Article 7 of the Dutch Code of Civil Procedure). German law, in contrast, takes a rather restrictive approach. As a result, victims may be unable to pursue related claims against parent companies and subsidiaries, or buyers and suppliers, in a single proceeding. This is unfortunate, because separate proceedings may be both costly and inefficient and may also result in conflicting judgments. In the absence of reform of the Brussels Ia Regulation, notably an extension of Article 8 to non-EU companies, Member States should therefore consider whether their existing rules provide sufficient scope for hearing closely connected supply chain claims together and, where necessary, adapt them accordingly.

In addition, Member States should consider amending their national jurisdictional rules to better cover stand-alone claims against non-EU companies falling within the scope of the CSDDD. These companies are subject to EU due diligence obligations because they generate substantial net turnover in the EU (cf. Article 2(2) CSDDD). Yet, the Brussels Ia Regulation provides no basis for jurisdiction. As a consequence, companies may be subject to the due diligence obligations of the Directive without necessarily being amenable to civil proceedings before an EU court. Failing an amendment of the Brussels Ia Regulation Member States should therefore consider closing this gap by adopting provisions that allows victims to sue non-EU companies under the same conditions under which consumers can sue non-EU companies under Article 17(1)(c) Brussels Ia Regulation. Specifically, they should allow victims of human rights violations to sue non-EU companies covered by the Directive if they pursue commercial activities in that Member State or target their activities towards that Member State. This would help close the gap between the territorial reach of the Directive’s substantive obligations and the jurisdiction of European courts to adjudicate civil liability claims arising from their alleged breach.

Making Cross-Border Litigation Work in Practice

The third and last issue that Member States should address when implementing the Directive relates to the procedural conditions under which civil liability claims can be enforced. In this context, Member States must, first and foremost, ensure that their national laws comply with the requirements set out in Article 29(3) CSDDD. They must, therefore, ensure that limitation periods are at least five years, are no more restrictive than those under general national civil liability rules, and do not begin to run before the infringement has ceased and the claimant has, or can reasonably be expected to have, knowledge of the infringement, the resulting harm and the identity of the infringer (lit. a). Furthermore, Member States must ensure that the costs of proceedings are not prohibitively expensive (lit. b), that claimants can seek injunctive measures (lit. c), and that courts can order the disclosure of relevant evidence under a company’s control where the claimant has sufficiently substantiated the plausibility of the claim (lit. e).

However, even if Member States do all of the above, it seems unlikely that this will make it sufficiently easy to bring damages claims in Europe. Since victims of human rights violations often come from socially and economically disadvantaged communities in the Global South, they most likely need more support than envisaged by the Directive. Specifically, they need additional forms of financial assistance, such as legal aid, third-party litigation funding or contingency fee arrangements, as well as practical support in initiating and conducting proceedings including, for example, from non-governmental organisations. Article 29(3) CSDDD, however, does not mandate any of these measures. Even the limited support with a view to NGO-led litigation originally envisaged by Article 29(3) lit. (d) CSDDD was deleted in the 2026 reform of the Directive. Member States should therefore not confine their implementation efforts to the implementation of Article 29(3) CSDDD. Rather, they should consider introducing additional procedural measures capable of addressing the financial and other barriers that continue to hinder victims’ effective access to justice.

Summary and Outlook

The deletion of Article 29(1) and (7) CSDDD marks a significant retreat from the European legislature’s original ambition to enforce human rights due diligence obligations through civil liability. It does not, however, diminish the case for private enforcement. Nor does it relieve Member States of the responsibility to ensure that the Directive is effectively enforced. When deciding whether to establish a civil liability regime, Member States should therefore concentrate on the merits of private enforcement and put the more fundamental debate about the desirability of human rights due diligence obligations aside. While the latter undoubtedly impose costs and burdens on companies, the European legislature not only choose to impose them in 2024 but also retained them in 2026. The priority should now be to ensure that the Directive’s obligations are effectively enforced.

Effective implementation, however, requires more than the introduction of a civil liability regime. Given the inherently cross-border nature of global supply chains, private international law will largely determine whether such rules can fulfil their intended purpose. Member States should therefore take three steps: First, they should ensure that national liability provisions apply in cross-border cases by giving them overriding mandatory effect under Article 16 Rome II Regulation. Second, they should improve access to European courts, particularly in cases involving non-EU subsidiaries, suppliers or companies subject to the Directive. And third, they should remove procedural and financial obstacles that prevent victims from effectively pursuing damages claims in practice.

The central point is simple: Civil liability can only contribute to the protection of human rights in global supply chains if victims can actually invoke it. Private international law is therefore not merely a technical consideration to be addressed after a liability regime has been established. It is essential to its effectiveness and should be at the heart of Member States’ efforts to implement the revised CSDDD.

0 replies

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Join the discussion

This site uses Akismet to reduce spam. Learn how your comment data is processed.