ECtHR Really Applies Less Restrictive Alternative: Saint-Paul Luxembourg S.A. v. Luxembourg

The structured proportionality test, as utilised by the German Constitutional Court (among others) and championed by Robert Alexy and his followers, subjects limitations of fundamental rights to a three-pronged test. The test is intended to examine – step by step – a measure’s (i) suitability, (ii) necessity and (iii) proportionality stricto sensu. Correct application of the test demands, according to Alexy and his followers, the examination of each of these three elements in order. As soon as a measure fails one of the steps, it is unconstitutional and there is thus no need to examine the next step(s).

Scholars who advocate this structured version of the proportionality test often lament its poor or wrongful application by courts, including the European Court of Human Rights. The ECtHR has, by and large, indeed not developed nor applied such a strictly ordered version of the proportionality test. It is particularly rare for the Court to apply the necessity test as separate from the proportionality stricto sensu test (also known as balancing). Instead, the Court has generally looked at the existence of less intrusive measures as an element to consider in the balance or it has continued to examine a measure’s proportionality in the strict sense after having indicated that less restrictive measures were available.

Not so in the recent case of Saint-Paul Luxembourg S.A. v. Luxembourg. In its judgment in that case, the ECtHR has – to my knowledge in a very exceptional move – applied the less restrictive alternative test as Alexy intended it to function, much to the satisfaction of the proponents of a structured proportionality test, I imagine.


Saint-Paul Luxembourg S.A. v. Luxembourg concerned a search and seizure operation conducted at the offices of the Portuguese language newspaper “Contacto Semanário”, of which the applicant company is the publisher. The newspaper had published an article describing the situation of families in which the parents had lost custody over their children due to measures taken by the social services. The article featured and named two adolescents and the social worker in charge of their case file. It was signed “Domingos Martins”. The social worker complained of defamation and the domestic authorities commenced an investigation against the author of the article, both for defamation and for breach of legislation protecting minors. Because the name “Domingos Martins” did not feature on the official list of recognised journalists, the investigating judge issued a search and seizure warrant against the applicant company, aimed at obtaining the identity of the author of the contested article. The applicant claimed that the search and seizure operation that was subsequently conducted at the premises of “Contacto Semanário” had violated its right to respect for its “home” and its right to freedom of expression.

Less Restrictive Alternative Test under Article 8 ECHR

The Court’s legal reasoning under article 8 will be of particular interest to the proponents of a structured proportionality test, Alexy-style. The Court first determined, not without contestation, that the search and seizure operation had interfered with the applicant company’s right to respect for its “home” (qua business premises). In his partly dissenting opinion, Judge Jäderblom vehemently disagreed. Judge Jäderblom emphasised the distinction between the journalist (an individual person) and the applicant company (a professional entity) to hold that the measure at issue was not sufficiently important to denote an interference with the “private life” of the applicant company. He was thus of the opinion that the claim under article 8 should have been declared manifestly ill-founded. The majority, however, did find an interference with article 8. It went on to hold that this interference was prescribed by law and pursued a number of legitimate aims, namely the defense of public order, the prevention of crime and the protection of the rights of others (the privacy/reputation rights of the social assistant and the minors who had been named in the article).

The Court then actually skipped the “suitability test”, but it did engage substantially with the necessity of the interference. The Court pointed out that, even if the name “Domingos Martins” did not appear on the official list of journalists, a similar name – “De Araujo Martins Domingos Alberto” – did feature on the list. The journalist with that name was moreover indicated as working for the newspaper at issue (“Contacto Semanário”). Given that information, the Court concluded that the investigating judge could have – as a first option – taken a less restrictive measure to confirm the identity of the author of the article, rather than issuing a search and seizure measure. The search and seizure had thus not been necessary.

While proportionality scholars have gotten used to the Court not ending its enquiry there, instead conducting an “unnecessary” balancing exercise, the Court did not do so in this case. Instead, it only wrote two more lines after having ruled that the measure was not necessary. One to conclude that the measure was not proportionate to the legitimate aim pursued and another to declare that article 8 had been violated. Certainly, this is an interesting example of how the structured, step-by-step proportionality test advocated by Alexy can feature in the Court’s case law.

Not all Judges agreed, though. Not only did Judge Jäderblom not even find an interference, Judge Lemmens explicitly devoted part of his concurring opinion to express a different view on the necessity test. Judge Lemmens would apparently prefer the necessity test to function as a strictly procedural test, rather than a substantive one. In his concurring opinion, he expressed doubts about the Court’s use of the expression “the investigating judge could have taken a less restrictive measure”.[1] Judge Lemmens was not convinced that it is the role of the Court to draw such specific conclusions. Instead, he would have preferred “une formulation moins affirmative”. According to Judge Lemmens, “the Court could have for instance asked itself whether the investigating judge had not had the possibility to take less restrictive measures”.[2] The Court could also have found that the government had not attempted to demonstrate that it was impossible to obtain the journalist’s identity without the search and seizure measure. Both are distinctly procedural questions and findings, which would avoid having the Court substantively examining the availability of less restrictive alternatives.

No Less Restrictive Alternative Test under Article 10 ECHR

Somewhat surprisingly, at least at first glance, the Court did not apply the necessity test in a ‘definitive’ manner in its examination of the article 10 claim in Saint-Paul Luxembourg S.A. v. Luxembourg. Given the ‘strict’ application of the test under article 8, one could have expected the Court to apply it in a similar manner under article 10. But it did not. Instead, it focused its attention on the broad formulation of the search and seizure warrant to rule it disproportionate, because it could have led to abuse by the police officials conducting the search and seizure operation. Although one could perhaps interpret the Court’s reasoning as de facto applying a necessity test, it is nevertheless striking that the Court does not use any less intrusive means language in its article 10 reasoning, while it had so heavily relied on that very language in its article 8 reasoning.

One can only speculate as to the reason for the difference. The most relevant explanation appears to lie in the different (potential) consequences of the measure. Under article 8, the search and seizure operation was only linked to the intention to obtain the identity of the journalist who wrote the article. However, there were clearly less intrusive measures available to achieve that aim. Under article 10, conversely, the Court was also concerned about the potential impact of the search and seizure on the protection of journalistic sources. It held in particular that the broadly worded search and seizure order could have led to the discovery – among the journalist’s files – of additional sources who had not been named in the article. In that sense, it was the disproportionality – in the sense of overbreadth – of the search and seizure order that was at issue. Application of the less restrictive alternative test would perhaps not have been very helpful in this regard, since it was never the authorities’ intention to obtain the identity of potential additional journalistic sources. A failure to adequately protect their secrecy was instead an unwarranted side-effect of the warrant.

[1] My translation; in French “le juge d’instruction aurait pu prendre”.

[2] My translation; in French “la Cour aurait pu se demander si le juge d’instruction n’avait pas eu la possibilité…”.

9 thoughts on “ECtHR Really Applies Less Restrictive Alternative: Saint-Paul Luxembourg S.A. v. Luxembourg

  1. I’m confused: your proof that the Court engaged in a full Alexy analysis was that “it only wrote two more lines after having ruled that the measure was not necessary”? 2 more lines?

    Saying that 2 lines of cursory analysis is sufficient to excite proportionality advocates seems to overstate the matter considerably!

    • Well, yes.

      It is true that the Court’s reasoning is not entirely in line with an Alexy-style – or German Constitutional Court style – proportionality analysis, because the Court skipped the “suitability” step (at least in its explicit legal reasoning reflected in the judgment). But the Court’s reasoning is in line with Alexy-style proportionality in another sense: as soon as the measure had been found to not be necessary – i.e. because less restrictive alternatives were available – there was no need to say anything more (hence the only two more lines part). The measure violated the Convention because it did not pass the “necessity” test. In that sense, the Court’s reasoning is actually in line with the structured proportionality test as explicated by Alexy, which maintains that the analysis should only move to the next step if the measure passed the one before it. If the measure fails at one of the steps, there is no need to examine the following one(s).

      A frequent complaint of proportionality enthousiasts who advocate the structured version of the test is that the Court mixes all steps in one open ended balancing test instead of keeping them – suitability, necessity, proportionality stricto sensu – strictly separated. For that reason, I do imagine that they will find it interesting to note a judgment in which the Court did follow the order (if still not entirely, because it skipped the “suitability” step) by finding a violation as soon as it noted that less restrictive measures were available, thereby stopping its analysis after the examination of the factually possible (“necessity”, in this case) without embarking on the – in light of its findings under the “necessity” step not required – analysis of the legally possible (balancing).

  2. I am a bit confused: while I have great respect for Alexy’s work I have my doubts as to your proposition that he “invented” the structured proportionality test. Rather, I believe, long before Alexy started his career the structured proportionality test was already well established case-law of the German Federal Constitutional Court.
    I would also like to point out that you will more often than usual in the Court’s case-law find a pretty well structured proportionality test in cases against Germany, e.g. Uzun ./. Germany, 2 September 2010, No. 35623/05, §§ 77 ff.

    • You’re absolutely right of course, Christiane. Thank you for your comment! Alexy indeed distilled / described the structured proportionality test as employed by the German Constitutional Court and built his theory of principles as optimisation requirements around it (with the proportionality test logically following, according to Alexy, from the character of principles as optimisation requirements). I’ll actually make some changes to the text to address any lack of clarity on that part. It was certainly not my intention to make it sound as if Alexy “invented” the proportionality test.

      • Great, thanks for that!
        By the way, I do agree with you that the Strasbourg court very often tends to be a bit, well, sloppy when it comes to the proportionality test…

      • Yes, I think it may be. Having a German Judge on the bench may have an influence. Another reason, however, could be that the Court will usually have a judgment of the German Constitutional Court at its disposal in German cases (if the case went all the way to the Constitutional Court during the domestic proceedings or if there is established case law on the subject). Especially if the ECtHR’s check is on the procedural side, it would make sense that it ends up reproducing or reformulating the reasoning employed by the Constitutional Court (i.e. a well structured proportionality test).

        N.B.: in case it is not clear, these are reactions to Christiane’s “I would also like to point out that you will more often than usual in the Court’s case-law find a pretty well structured proportionality test in cases against Germany.”

      • Regarding Th’s comment I would like to add that it will of course help having a German Judge on the bench; probabply especially so if the Judge is a former Constitutional Court Judge as was the case with Renate Jaeger, Judge elected in respect of Germany to the ECtHR from 2004 – 2010.
        While Stijn’s comment also makes sense, I need to point out that not all (actually rather few) of the decisions of the Constitutional Court reaching the ECtHR contain (lengthy) reasoning; even in respect of applications leading to a judgment of the ECtHR. Going into the reasons for that would, however, go beyond the scope of this comment.
        Finally, what could also be a reason for more structured proportionality tests in cases against Germany is that those decisions and judgements are (usually) drafted by German educated Registry lawyers who have “grown up” with the structured proportionality test and hence have more or less internalized it.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s