The CJEU Rules in Idziski: One Series, Two Media, Two Jurisdictional Regimes

This post was written by Giacomo Marola (PhD).

On 18 June 2026, following AG Rantos’ Opinion, discussed on this blog, the CJEU delivered its judgment in Case C-232/25, Idziski, concerning jurisdiction under Article 5(3) of the Brussels I Regulation, now Article 7(2) Brussels I bis, over personality rights infringements.

Facts of the Case

Idziski arises from proceedings brought in Poland by Z.R., a Polish resident and former member of a military organisation involved in the Polish resistance during World War II (“unit X”), and by Ś., an association whose purpose is to defend the dignity, reputation and memory of that unit and its members.

The dispute concerns a German television series broadcasted in several Member States, including Poland, and made available online. According to the claimants, the series depicted the soldiers of unit X as antisemitic, nationalistic and complicit in the Holocaust, thereby infringing their personality rights under Polish law, including dignity, national identity and an undistorted historical account.

Before the Polish courts, the claimants sought compensation for non-material damage and non-pecuniary remedies, including the display of specific information before each broadcast of the series and the publication of an apology, or appropriate statement, on television and online.

The Polish Supreme Court referred two main questions. First, it asked whether the Polish courts could exercise jurisdiction over the entire damage allegedly suffered, on the basis of the claimants’ centre of interests, although the series identified unit X rather than Z.R. or Ś. individually. Secondly, if full jurisdiction was unavailable, it asked whether the Polish courts could exercise territorial (“mosaic”) jurisdiction over the damage allegedly suffered in Poland, including for non-pecuniary remedies.

The Judgment

The Court starts by recalling its case law concerning jurisdiction based on the place of damage for personality rights infringements.

Since Shevill, in cases of press defamation, the victim may sue before the courts of each Member State where the publication was distributed and where the victim claims to have suffered injury to reputation, but only for the damage suffered in the forum State (“mosaic” approach).

Since eDate, for online personality rights infringements, the claimant may sue before the courts of each Member State where the content is accessible, with jurisdiction limited to the damage suffered in the forum State, as well as before the courts of the Member State in which their centre of interests is located, with jurisdiction covering all the damage suffered.

As clarified in Bolagsupplysningen and Gtflix, courts of each Member State where the online content is accessible have jurisdiction only for territorially divisible remedies, like compensation, but not territorially indivisible remedies, like rectification or removal of online content.

Since Mittelbayerischer, centre-of-interests jurisdiction is available only where the harmful online content contains objective and verifiable elements making it possible to identify the claimants, directly or indirectly, as individuals.

Against that background, the Court first clarifies that the eDate centre-of-interests jurisdiction, developed for online infringements, does not extend to television broadcasting. Unlike online content, television broadcasting is not instantly and globally accessible, but remains regionalised and limited to the geographical area in which the television signal is received. Even assuming that geoblocking or geolocation technologies may be relevant in determining whether the requirement of ubiquity is satisfied for online audiovisual content, they do not affect the conclusion that television broadcasting does not satisfy that condition.

Personality rights infringements resulting from television broadcasting therefore remain governed by Shevill: where audiovisual content is broadcast on television in several Member States, the claimant may sue before the courts of each Member State in which that content was broadcast and where the victim claims to have suffered injury to reputation, but each court has jurisdiction only in respect of the damage caused in its own Member State.

As regards online dissemination, relying on Mittelbayerischer, the Court denies centre-of-interests jurisdiction for the individual claimant, Z.R., but accepts it for the representative association, Ś.

For Z.R., the identification of unit X was not sufficient. The series depicted fictional members of that military unit, to which Z.R. had belonged, but did not identify him, either directly or indirectly. According to the Court, for reasons of predictability and legal certainty, indirect identification requires the person concerned to be identifiable “with certainty by reason of attributes which are specific to them and which distinguish them from all other individuals”.

By contrast, the Court accepts centre-of-interests jurisdiction for Ś. The series directly and unequivocally identified unit X, and Ś. is a legal person whose principal purpose is to defend the interests of that unit and its members. On that basis, the Court considers it predictable for the producers of the series to be sued before the courts of the association’s centre of interests.

Finally, since the Polish courts do not have centre-of-interests jurisdiction over the entirety of Z.R.’s claims, the Court considers the scope of their “mosaic” jurisdiction over the remedies sought.

For television broadcasting, the Polish courts may hear both the claim for compensation for the damage allegedly suffered in Poland and the non-pecuniary remedies sought by the claimants, provided that those remedies are limited, in their effects, to Polish territory.

The position is different for online dissemination. In line with its previous case law treating the rectification of online content as a single and indivisible remedy, the Court holds that the Polish courts’ “mosaic” jurisdiction is limited to compensation for the non-material damage allegedly suffered in Poland. It does not extend to a non-pecuniary remedy aimed at securing the rectification of information placed online.

A Judgment of Continuity and of Growing Complexity

Overall, Idziski is a judgment of continuity. Largely following AG Rantos’ Opinion, the Court preserves the main pillars of its existing case law: the coexistence of different jurisdictional regimes for regionalised media distribution and online dissemination under Shevill and eDate, the individual identification requirement developed in Mittelbayerischer, and the distinction between divisible and indivisible remedies under Bolagsupplysningen and Gtflix.

At the same time, by applying that framework to the television and online dissemination of the same audiovisual work, and to claims brought by both a natural person and a representative association, the judgment brings to the fore and arguably amplifies its underlying complexity and artificiality. As Tobias Lutzi has observed, Idziski may be read as a further indication that this area is ripe for legislative reform. Against that background, four aspects of the judgment call for further reflection: the territorialisation of online content; the application of different jurisdictional regimes to the same audiovisual work; individual identification, group-related harm and representative litigation; and the scope of “mosaic” jurisdiction for non-pecuniary remedies.

Comment

Preserving eDate while making its technological premise less absolute: ubiquity in the wake of geolocation technologies

Television has a more limited distribution capacity than Internet. Yet it is far less obvious that today’s Internet can still be described as genuinely “ubiquitous”.

As I have argued elsewhere, this idea no longer fully reflects the realities of the contemporary Internet. The assumption that online content is necessarily global overlooks the widespread use of geolocation technologies and the increasing territorialisation of online services, especially in the streaming sector, where dissemination is often structured around national catalogues, territorial licensing arrangements, geoblocking and country-specific availability.

In Idziski, however, the CJEU remains attached to the conceptual architecture developed in eDate. It distinguishes television from the Internet on the ground that television broadcasting “is not, in principle, available instantly and worldwide”, whereas online content is, in principle, intended to ensure ubiquity.

The Court’s adherence to eDate is understandable. The eDate framework is now a central component of the Court’s case law on Article 7(2) Brussels I bis, and any radical reconsideration could probably be better addressed by the EU legislature in the forthcoming reform of the Regulation.

Yet Idziski introduces a subtle but potentially important nuance. For the first time in this line of case law, the Court expressly acknowledges the existence of geoblocking and geolocation technologies. Although it does not treat them as sufficient to call into question the ubiquity of the Internet, the Court’s wording is carefully framed: “even assuming” that such developments may be relevant for determining whether the requirement of ubiquity of audiovisual content disseminated online is satisfied, they do not alter the conclusion that television broadcasting does not satisfy that condition.

The Court therefore appears to leave open the possibility that such technologies may be relevant in assessing whether the factual premise underlying eDate is actually satisfied in a given case. The result is ambivalent: Idziski preserves the traditional eDate framework, but may slightly soften the assumption that online content is always and necessarily ubiquitous. If ubiquity becomes a condition to be verified in individual cases, rather than a technological axiom, the future application of eDate may become more fact-sensitive than the original case law suggested.

Parallel Publications: One Content, Two Jurisdictional Regimes

One of the most significant aspects of Idziski concerns the treatment of “parallel publications”, where the same content is disseminated both online and through more traditional media, such as television.

The referring court had stressed that the distinction between television and online dissemination has become increasingly difficult to maintain. The same audiovisual work may reach the public through television broadcasting, online platforms, catch-up services or other digital channels. Applying different jurisdictional rules to that same work solely because of the channel through which it is accessed therefore risks fragmenting litigation and undermining the sound administration of justice.

The Court nevertheless confirms the application of different jurisdictional regimes to the same content. Echoing AG Rantos’ Opinion, it holds that Shevill governs alleged infringements resulting from television broadcasting, while eDate applies to those resulting from online dissemination.

The Court acknowledges that the concurrent application of Shevill and eDate may lead to parallel proceedings and to different courts ruling on different parts of the same dispute. Recalling Shevill, however, it observes that the claimant may still concentrate the action before a single court, either at the defendant’s domicile (forum rei) or at the place of the causal event (forum actus). Yet this offers only a limited response to the risk of fragmented litigation. In practice, claimants will normally sue before the courts of their own centre of interests or, at least, before the courts of the place where the damage occurred. In any event, concentration is left to the claimant’s procedural choice: if the claimant does not choose the forum rei or the forum actus, the same content may still give rise to fragmented proceedings across different media.

Moreover, for infringements resulting from television broadcasting, the judgment retains the traditional Shevill formula: jurisdiction based on the place of damage lies where the content is broadcast and where the victim claims to have suffered injury to reputation. This transposition is not entirely straightforward. Unlike Shevill, Idziski does not concern a classic defamation claim where the damage is framed as injury to the claimant’s reputation in the forum State. The alleged harm appears instead to derive primarily from the claimant’s membership of a particular group, making injury to reputation a less obvious connecting factor for locating the damage.

More fundamentally, the judgment accepts that the applicable jurisdictional regime depends on the technological medium through which the very same content is disseminated. The same audiovisual work may therefore fall simultaneously under the Shevill and eDate frameworks, not because the nature of the allegedly infringed right differs, nor because the harmful event changes, but because the content reaches the public through different communication channels.

The practical implications are significant. Claimants may have an incentive to frame different parts of the same dispute by reference to different communication channels, while defendants may contest jurisdiction by challenging the characterisation of the relevant medium. In a media environment where the same programme may be broadcast on television, streamed live online, made available on catch-up services, or distributed through on-demand platforms, jurisdiction may increasingly turn on unstable and artificial distinctions of media classification. This may also affect the remedies available and the subsequent circulation of judgments, especially where different courts rule on different territorial or media-specific aspects of the same dispute.

Individual Identification and the Centre of Interests of Representative Associations

Another important aspect of Idziski concerns the availability of centre-of-interests jurisdiction where the allegedly harmful content does not identify the claimant directly, but refers to a limited group to which the claimant is connected.

In this context, Idziski gives Mittelbayerischer a particularly strict reading. For reasons of predictability, the Court holds that indirect identification requires the person concerned to be identifiable “with certainty by reason of attributes which are specific to them and which distinguish them from all other individuals”. Here, the series depicted fictional members of unit X, not Z.R. himself.

This reasoning arguably goes further than Mittelbayerischer. The requirement of objective and verifiable elements making it possible for a person to be identified, directly or indirectly, becomes, in Idziski, a requirement of identification “with certainty” by reference to attributes specific to that person and distinguishing them from all others. This appears to narrow significantly the possibility of relying on indirect identification.

The Court’s position is different for the claimant association, Ś. Here, the relevant question is no longer whether the contested content identifies the association itself. It does not: the content concerns the members of the historical military unit, not Ś. as an organisation, which was established several decades after the events depicted in the series.

The Court focuses instead on the association’s principal purpose. Ś.’s principal purpose is to defend the dignity, reputation and memory of that group and its members. Since the group itself is directly and unequivocally identified in the contested content, the Court considers it predictable that proceedings may be brought before the courts of the centre of interests of the association entrusted with defending that group’s interests.

This is one of the most innovative aspects of the judgment. For natural persons, belonging to an identified group is not enough to trigger centre-of-interests jurisdiction. For the association, by contrast, the identification of that same group becomes sufficient because of its representative function and principal purpose. Jurisdiction therefore depends on a combination of two elements: the direct identification of a limited group in the contested content, and the existence of a legal person whose “principal purpose” is to defend that group’s interests.

This solution raises questions of predictability. To exclude centre-of-interests jurisdiction for natural persons, the Court stresses that the individual members of the relevant group may have their centres of interests in different Member States. Yet the same may also be true of associations: several associations established in different Member States may have as their principal purpose the defence of the same group. If each could rely on its own centre of interests, the result would again be a multiplication of fora. This makes it difficult to justify the distinction between natural persons and associations solely by reference to predictability.

The distinction appears instead to be driven, at least in part, by policy considerations. Where allegedly harmful content targets an identified group, the Court seems willing to ensure that a legal person whose principal purpose is to defend that group’s interests may sue before the courts of its own centre of interests. Until now, such entities would normally have had to rely on the forum rei or the forum actus in order to claim compensation for the entirety of the alleged damage.

This may be understandable from a policy perspective, but it also creates a certain asymmetry. Individual members of the group cannot rely on their own centres of interests unless they are identifiable as individuals. By contrast, an association may do so if the group is directly identified and the defence of its interests constitutes the association’s principal purpose. Future cases may need to clarify how close the link between the association and the identified group must be, and how the association’s “principal purpose” is to be assessed.

While Bolagsupplysningen and Gtflix concerned companies, in Idziski the Court extends centre-of-interests jurisdiction to an association defending collective interests. Yet it does not clearly explain how such an association’s centre of interests should be identified. It appears to assume that Ś.’s centre of interests is located in Poland, without clarifying whether this follows from its statutory seat, its activities, the group it represents, or the territorial significance of the allegedly harmful content.

Territorial Injunctions Under “Mosaic” Jurisdiction: Clarification or New Ambiguity?

The final point concerns the remedies available before courts exercising “mosaic” jurisdiction, where Idziski again confirms the distinction between television broadcasting and online dissemination.

For television broadcasting, the Court accepts that the courts of the Member State in which the content was broadcast may order injunctions, such as apologies or statements linked to the broadcast, provided that their effects are limited to the territory of that Member State. In other words, “mosaic” jurisdiction is not limited to compensation, but may also support territorially limited injunctions.

For online dissemination, by contrast, the Court confirms the approach taken in Bolagsupplysningen and Gtflix. Courts exercising “mosaic” jurisdiction may award compensation for the damage suffered in the forum State, but may not order remedies aimed at rectifying or removing online content. The judgment therefore treats the online remedy sought as falling, at least functionally, within the logic of rectification identified in Bolagsupplysningen and Gtflix.

This restraint is welcome. Had the Court accepted that online apologies or similar non-pecuniary remedies could be territorially limited through geoblocking or geolocation technologies, it would have considerably expanded “mosaic” jurisdiction beyond compensation and further fragmented personality rights litigation. It would also have made it more difficult to preserve the idea, central to Bolagsupplysningen, that online rectification and removal are, by their nature, single and indivisible remedies.

At the same time, the resulting distinction remains artificial. A court exercising “mosaic” jurisdiction may hear a claim seeking an apology in relation to television broadcasting, provided that the remedy is limited to the territory of the forum State. The same court may not hear a comparable claim in relation to online dissemination, because the remedy is treated as indivisible.

The underlying rationale again appears to be the alleged ubiquity of online content. This is particularly surprising because, elsewhere in the judgment, the Court acknowledges that geoblocking and geolocation technologies may be relevant in determining whether the requirement of ubiquity is satisfied for online content. Yet it draws no consequences from those same technologies when considering whether the effects of online remedies may be territorially restricted.

By treating substantially the same type of non-pecuniary remedy as territorially divisible or indivisible depending on the medium through which the contested content is disseminated, Idziski preserves the coherence of the Court’s previous case law, but also makes the distinction between divisible and indivisible remedies increasingly difficult to justify. This may generate further preliminary references concerning the territorial availability of non-pecuniary remedies under Article 7(2) Brussels I bis.