July 29, 2025
***
To celebrate the 15th anniversary of the Strasbourg Observers Blog, we organised an in-person symposium with scholars, practitioners, and members of the ECtHR on 8–9 May 2025 in Ghent. Connecting in person with so many regular contributors was a wonderful experience and led to engaging dialogue with current and former judges of the Court. To allow our online community to participate in the thought-provoking discussions that took place, we will be publishing selected position papers on the blog over the next two summer months. We wish all of our readers a well-deserved break with a little peek on Strasbourg Observers here and there.
Causation plays a central role in assigning responsibility within national legal systems and is more broadly seen as a fundamental component of legal reasoning. It underpins assessments of responsibility/liability and the outcomes that follow once responsibility is established. Yet, defining causation remains conceptually challenging – especially when the issue concerns omissions rather than actions. This challenge has long posed difficulties for national tort law (see e.g. Steel).
As Plakokefalos and Lanovoy have noted, causation has received relatively little attention in international law. In the codification of state responsibility under international law, the issue of causation was largely set aside, with the conclusion that it falls to the primary rules themselves – that is, the primary substantive obligations – to define the applicable causation standards (see Yearbook of the International Law Commission, 2001, vol. II, Part Two: Commentary to Art 31, para 10 (‘the requirement of a causal link is not necessarily the same in relation to every breach of an international obligation’).
Human rights law has largely overlooked the test of causation. The works of Vladislava Stoyanova (open access), David McGrogan and Laurens Lavrysen are notable exceptions. The question as to what causal links are applied in human rights law between state conduct (that can be in the form of acts or omissions) and harm for the purpose of establishing state responsibility, has remained underexplored. No causation standard/test has ever been developed. Any causal inquiries performed by adjudicative bodies, such as the European Court of Human Rights (ECtHR), have been chaotic and inconsistent. One can also say quite unprincipled since different terms that appear to reflect the notion of causation, are used in different judgments (Stoyanova). The test of causation required or any evidentiary standards for proving causation have barely been discussed in the ECtHR’s reasoning.
Recently with the climate-change litigation before the Court, the importance of addressing causation in human rights law, has more clearly emerged. KlimaSeniorinnen is the first judgment where the ECtHR devoted whole sections to the question of causation (for detailed analysis see Stoyanova). In KlimaSeniorinnen, ECtHR addressed causation upfront. The analytical gymnastics that the Court performed in its causation inquiry are yet to be fully understood. The confusion as to what is meant by causation and the links between which things are actually addressed, is pervasive in the reasoning. Overall, it is fair to say that the reasoning regarding causation is confusing and that it is not clear how specifically the ‘real prospect’ test is applied for finding a breach for the claimed omissions in KlimaSeniorinnen (see here).
Such a confusion is not helpful for any national legislator or national court that has to address the problem of climate change and environmental harm from a human rights law perspective. Such a confusion is not helpful since it creates uncertainty at domestic level as to the role of human right law more generally.
Conceptually causation is a key element in the reasoning not only in cases where there might be multiple causes and complex cause-effect relationships, such as climate-change related cases. For example, in every case where an applicant formulates an argument that the State should be found responsible under the ECHR for omissions, a link needs to be established between the claimed omissions and the harm. For example, in the case of Biba v. Albania, where a child was harmed by another child during their break at the school premises, the ECtHR was clear to the effect that such a link might not be established for the purposes of national civil proceedings (tort law); yet, the link might be found existing for the purposes of establishing responsibility under the European Convention on Human Rights (see Patrick Leisure). This could be explained by the different normative underpinnings of human rights law (see Stoyanova and Stoyanova ‘Causation and Breach of Positive Obligations under the European Convention on Human Rights: Correcting the Past, Preventing the Future’ in V Stoyanova and M Andenæs (eds) The Role of the Causal Inquiry under the European Convention on Human Rights (forthcoming). All of this implies that while national tort law might be relatively more restrictive in finding causal links, human rights law appears to be more generous (possibly to the point of being simply unconcerned about the question of causation) (see Sophie Treacy ‘Causation, Reasonableness and Positive Obligations’ in V. Stoyanova, & D. McGrogan (Eds.), From Protection to Coercion: the Limits of Positive Obligations in Human Rights Law Hart/Bloomsbury forthcoming). This discrepancy itself can raise a whole set of separate questions regarding, for example, the requirement for exhaustion of domestic remedies.
My objective in this blog post is to explain causation and the establishing of breach of positive obligations under the ECHR. In other words, the objective is to explain the link (the connection) between, on the one hand, omissions (failures by the State) and, on the other, harm to important interests as protected by the ECHR, in the system of state responsibility under the Convention. This connection is crucial for attributing responsibility for harm, as it rests on the assumption that, had the State acted differently – i.e., taken appropriate measures – the harm might have been avoided. In this sense, the relationship is best understood in terms of preventability.
I will make three points that I think are important for understanding the link: (1) reframing of the harm as risk of harm, (2) omissions and their counterparts (i.e. the measures that the State should have arguably taken at the relevant point in time in the past arguably to fulfil its positive obligations) can be framed at different levels of specificity, and (3) the mixture of legal and factual causation.
In the reasoning about omissions as a basis for responsibility, often the harm is reframed as risk of harm. In this way, it is much easier to advance an argument that the harm (already reframed as risk of harm) was knowable and an omission to prevent it is identifiable. For example in para 390 in Cannavacciuolo and Others v. Italy, the following was noted: ‘ […] the Court accepts the existence of a “sufficiently serious, genuine and ascertainable” risk to life to engage Article 2 of the Convention and trigger a duty to act on the authorities’ part.’ The Court’s reasoning often shifts between framing the issue as actual harm to protected interests and as the risk of such harm, creating confusion about whether the State was expected to prevent the harm itself or merely the risk of it at the relevant time when action was arguably required.
Omissions and therefore any positive obligations upon the State to act can be framed with different levels of specificity (for detailed analysis see Stoyanova). Different levels of abstraction can therefore be applied (see also Stoyanova, ‘Positive Obligations as Coercive “Rights” and Compulsory Vaccination under the European Convention on Human Rights’ International Journal of Constitutional Law forthcoming). In its reasoning, the Court can choose the level of abstraction. The higher the level of abstraction used to frame the omission(s) (and accordingly any possible counterparts to these omissions), the easier it is to conceptually link them to the harm (even easier if the harm is reframed as risk of harm). For example, as I have explained in detail here, the GC in Kurt v. Austria chose to frame the omissions in a narrow way, which made the establishment of no violation easier.
L.F. and Others v Italy can be used as an illustration for a relatively much more abstract framing of the omissions. At para 157, it was noted that ‘ […] the Court considers that the complaints raised by the applicants should not be analysed from the standpoint of the alleged absence of an adequate legal framework, but from that of the protective measures the authorities put in place in the specific circumstances of the case.’ At para 162 it was stated that ‘In the light of the foregoing, the Court finds that, after allowing residential development of the area surrounding the foundry, the national authorities did not take all the measures necessary to ensure the effective protectionof the right to respect for private life of the people concerned, at least for the period from 2008 to 2016 [emphasis added].’ The phrase “all necessary measures” is very abstract and raises an important question: does the State’s failure to take even one measure deemed necessary amount to a violation of its positive obligations under Article 8 of the Convention?
The distinction between factual and legal causation has long been the subject of theoretical analysis and debates. Legal systems seek to differentiate among the many factually relevant causes to identify those that are legally significant in attributing harm. In theory, this causal inquiry is often understood as involving two successive steps (see e.g. Stapleton). The first one is factual causation that seeks to establish the factual link between an omission and the harmful outcome. This is meant to reflect the historical picture of the events leading to the outcome. This picture can be very wide, which can make the inquiry overly inclusive. The inquiry may be also hampered by scientific complexity and uncertainty.
Once the first step of factual causation is performed, the second step follows. This second step has been framed as legal causation. Other possible framings that have been used are remoteness, ‘proximate cause’ assessment or scope of responsibility. This second step seeks to determine for which effects (harms) of the wrongful act, the defendant is to be held legally responsible and is to accordingly ensure remedies (see Plakokefalos and more recently André Nollkaemper, Causation Puzzles in International Climate Litigation)
Legal responsibility is therefore determined by certain normative (value-related) considerations that are not related to any factual causal links. In the case law of the Court, such normative considerations are often captured by the standard of reasonableness. As I have explained here (open access book), the Court consistently refers to reasonableness (and/or disproportionate burden – see Verein Klimaseniorinnen Schweiz and Others v Switzerland [GC] para 538 for further references) to determine breach of positive obligations. What functions does the invocation of the standard of reasonableness serve in the Court’s reasoning? Corten has noted that ‘the notion of reasonableness enables judges to present their decisions and motivations, often of their own creation, as perfectly in line with the intention of States [emphasis added].’ More recently Ilias Plakokefalos in his book chapter ‘The Standard of Reasonableness in Human Rights Positive Obligations’ in V Stoyanova and D McGrogan (eds) From Protection to Coercion: the Limits of Positive Obligations in Human Rights Law (Hart Publishing, forthcoming), has also examined how reasonableness is used in the Court’s reasoning.
Yet, despite the usage of the standard of reasonableness that can both support and undermine the normative proposition that there is a link between harm and State omission(s), the Court also refers to science and factual scientific studies. In this sense, factual and legal causation are mixed in the reasoning. Such references to facts (as supported by science) are often made with the well-justified acknowledgement that scientific studies can be inconclusive and selective. Yes, such references to science (that can support factual causal links) are made. What purpose do they serve in the reasoning?
Because omissions lack clearly defined counterparts – particularly given the discretion States have in how to ‘ensure’ ECHR rights (Stoyanova) – the specific content of positive obligations is not predetermined. As a result, when the Court determines a breach after the fact, it must rely on forms of reasoning that give the appearance that the obligations in question were not created retroactively. The goal is to convey that, in the case at hand, obligations with a certain degree of specificity already existed at the time the harm (or risk thereof) occurred. This helps avoid the perception that the Court’s analysis rests solely on normative or policy-based judgments, or on the retroactive creation of standards that States never explicitly accepted. One strategy to maintain this appearance is to appeal to rationality, often through the inclusion of factual causation as supported by scientific studies. By referencing scientific studies – even when these are inconclusive or selectively cited – the Court can reinforce the legitimacy of its reasoning and obscure the extent to which it is, in effect, crafting new obligations (for further elaboration see Stoyanova, ‘Causation: Why and How for Establishing Breach of Positive Obligations under the European Convention on Human Rights?’ in V Stoyanova and D. McGrogran (eds) From Protection to Coercion: the Limits of Positive Obligations in Human Rights Law (Hart Publishing, forthcoming).
In this blog post, I have outlined three key clarifications that, in my view, are essential for understanding how the Court approaches causal reasoning. First, when harm is reframed as a risk of harm, establishing causation becomes easier. Second, omissions can be framed at varying levels of abstraction, and the more abstract the framing, the easier it is to argue that the omission caused the harm. Third, the Court conflates legal and factual causation in its reasoning.
Strasbourg Observers is an academic blog that discusses recent developments at the European Court of Human Rights. The editorial team of Strasbourg Observers is based at the Human Rights Centre of Ghent University and the Centre for Government and Law of Hasselt University.
Strasbourg Observers would not be possible without the valued work of countless guest bloggers. Visit the “Contact” page on our website to find out how to become one!
