Free Speech versus Religious Feelings, the Sequel: Defamation of the Prophet Muhammad in E.S. v Austria

In E.S. v Austria the European Court of Human Rights (Strasbourg Court or Court) ruled, not for the first time, ¹ that Austria can legitimately curb free speech to protect the religious feelings of believers. ² The applicant in E.S. intimated, during seminars she gave on Islam, that the Prophet Muhammad was a paedophile. The Austrian courts subsequently found her guilty of the criminal offence of publicly disparaging, in a manner capable of arousing justified indignation, a person (Muhammad) who is an object of veneration. ³ In E.S., the Strasbourg Court endorsed the Austrian courts’ reasoning and ruled that the applicant’s freedom of expression had not been violated.

The Strasbourg Court has received impassioned criticism in some media and the blogosphere for its judgment in E.S. ⁴ Marko Milanovic, in particular, has castigated the Court for giving ‘its blessing to the criminalization of blasphemy, in all but name’. ⁵ Milanovic even warns that E.S. ‘[leads] us not to a slippery slope of a further erosion of free speech, but to a cliff’. ⁶ The harsh nature of this criticism, verging on shock and dismay, can be understood in the context of a secularised Europe in which blasphemy laws are generally considered unacceptable impediments to free speech. Most European countries, especially in Western Europe, have moved away from intolerance of blasphemy (i.e. enforcement of criminal blasphemy laws), often after a period of tolerance for blasphemous speech (i.e. deliberate non-enforcement of blasphemy laws), but eventually to full acceptance of speech that is critical of or ridicules religion (i.e. abolition of blasphemy laws). ⁷

At the same time, however, there remains a lingering consciousness of the (potential) fragility of religious tolerance in Europe. Although many blasphemy laws have been repealed, other laws and doctrines have taken their place. In most European countries, religious tolerance is no longer enforced by protecting religious texts, doctrines, and figures as such. Instead, it is ensured – at least in some contexts – by protecting the religious feelings of believers about speech that offends those very same religious texts, doctrines, and figures. ⁸ The Strasbourg Court’s case law reflects this shift from blasphemy bans to the protection of religious feelings. On the one hand, the Court is adamant that ‘[religious persons] must tolerate and accept the denial by others of their religious beliefs and even the propagation by others of doctrines hostile to their faith’. ⁹ On the other, however, the Court insists that ‘respect for the religious feelings of believers [imposes] a duty to avoid as far as possible an expression that is, in regard to objects of veneration, gratuitously offensive to [religious believers]’. ¹⁰

Yet this shift to the protection of religious feelings is open to criticism because the distinction from the blasphemy laws of old is not always easy to discern. The Strasbourg Court, in particular, has been charged with effectively endorsing blasphemy bans in repackaged form, something Milanovic refers to as ‘blasphemy plus’. ¹¹ In this note, I argue that the Court’s reasoning in E.S. is especially vulnerable to this charge. In E.S., the Court conflates the principles of its case law on the protection of religious feelings with standards from its defamation case law. This conflation results in a judgment whose reasoning seems more inclined towards addressing defamation of the Prophet Muhammad, than gratuitous offence to the religious feelings of Muslims. I will elaborate this point in due course.

I begin by discussing the original Strasbourg Court judgment on insult to religious feelings: Otto-Preminger-Institut v Austria (henceforth Otto-Preminger). This prelude is important for at least two reasons. First, because much – but not all – of what the Court says in E.S. is drawn directly from Otto-Preminger. Second, since much of the scholarly lamentation on E.S. echoes earlier criticism of Otto-Preminger. Having introduced the legal context in which E.S. was decided, I go on to summarise the facts, domestic judgments and Strasbourg Court ruling in E.S. itself. Afterwards, I critique the Court’s reasoning in E.S. for conflating two lines of case law. This conflation generates a number of serious problems, which have provoked justified criticism from Milanovic and others. Before concluding, I explore the possibility of a less contentious reading of E.S., cast in the language of tolerance.

The ‘origin story’: Otto-Preminger-Institut v Austria

Considering the outrage over E.S. in the blogosphere and media, those less familiar with Strasbourg case law might have guessed differently, but E.S. is not an isolated or ‘fluke’ judgment. ¹² Instead, it fits rather snugly into a line of Strasbourg case law on ‘offence to religious feelings’ that is over two decades old. The origins of this line of case law can be traced back to another judgment against Austria, the 1994 case Otto-Preminger, ¹³ in which the Court ruled for the first time that European states could legitimately curb free speech to protect the religious feelings of believers. In Otto-Preminger, the Austrian courts had invoked Article 188 of the Criminal Code to justify the seizure of the film Das Liebeskonzil, for insulting the religious feelings of Roman Catholics. Das Liebeskonzil shows a performance of the play by the same name and reconstructs the 1895 blasphemy trial of its author, Oskar Panizza. Play and film portray God as a senile old man, the Virgin Mary in erotic tension with the Devil, and the adult Jesus Christ as a low-grade mental defective. ¹⁴ In reviewing the seizure of the film, the Strasbourg Court found no violation of freedom of expression.

In Otto-Preminger, the Court introduced a series of important novel principles to its case law. First and foremost, the Court interpreted article 9 of the Convention, which protects freedom of religion, to include a right to respect for religious feelings. ¹⁵ As a result, Otto-Preminger entailed a conflict between human rights. ¹⁶ The Court held that in order to resolve the conflict between freedom of expression and freedom of religion, both rights had to be balanced against each other. ¹⁷ But rather than conduct the balancing exercise itself, the Court deferred to the rulings of the Austrian courts, since ‘[i]t is in the first place for the national authorities, who are better placed than the international judge, to assess the need for [measures to ensure religious peace]’. ¹⁸ The Court did, however, give some guidance on how the balancing exercise ought to be conducted, chiefly by laying down the following standard:

‘[freedom of expression includes an] obligation to avoid as far as possible expressions that are gratuitously offensive to others and thus an infringement of their rights, and which therefore do not contribute to any form of public debate capable of furthering progress in human affairs’. ¹⁹

Otto-Preminger is notorious for having provoked widespread and harsh criticism. Of particular note is the near unanimity of critique, even decades later, among both opponents and proponents of proportionality analysis. ²⁰ In his scorching 2009 critique of proportionality analysis, Stavros Tsakyrakis deployed Otto-Preminger to show that ‘the balancing approach fails, spectacularly, to deliver what it promises’. ²¹ Tsakyrakis argued, in particular, that the Court’s judgment reveals ‘an embarrassing implication of the balancing method’:

‘if there is no such thing as a right to have one’s religious feelings protected, then it makes no sense to speak of balancing in the first place, since we would be lacking that against which we are supposed to balance freedom of speech’. ²²

George Letsas expressed similar criticism in 2012, arguing that ‘the European Court’s balancing methodology [in Otto-Preminger] obscures an important matter of principle’, to wit the normative claim that – according to Letsas – there is no right not to be offended in one’s religious feelings. ²³

But Otto-Preminger not only met with the ire of critics of balancing; the Court’s judgment also fared poorly among proponents of proportionality. In their 2012 book on proportionality, Matthias Klatt and Moritz Meister presented Otto-Preminger as their single case study to illustrate everything that can go wrong when applying the proportionality principle. ²⁴ In contrast to Tsakyrakis and Letsas, Klatt and Meister did not negate the basic premise of the Court, agreeing that the case called for a balancing exercise between the competing rights. Rather, Klatt and Meister argued that the Strasbourg Court had simply struck the wrong balance. They instead aligned themselves with the balance struck by the dissenting judges in Otto-Preminger. In the opinion of those three judges, the seizure of the film had been disproportionate because it was highly unlikely that anyone who might be offended by the film could have seen it unwittingly (since, among other considerations, the cinema at which the film was to be screened catered to a relatively small public and because the film had been clearly advertised, meaning that potential viewers were forewarned about the film’s content). ²⁵

Knowing the ‘origin story’ of E.S. is important for at least two reasons. First, to understand that much – but not all – of what the Strasbourg Court says in E.S. is a direct application of Otto-Preminger. Second, to appreciate that much of the criticism of E.S. echoes the earlier critiques of Otto-Preminger. In his comments on E.S., Emmanouil Bougiakiotis laments, for instance, that the Strasbourg Court ‘has never convincingly explained why the feelings of religious believers are protected under Article 9 of the Convention’. ²⁶ And Milanovic claims that E.S. ‘can tell us something about the dangers of proportionality as an analytical framework when it is applied without sufficient doctrinal rigor’. ²⁷

The sequel: E.S. v Austria

As already intimated, E.S. revolved around a criminal conviction for publicly disparaging, in a manner capable of arousing justified indignation, the Prophet Muhammad. Ms E.S., the applicant, gave seminars on Islam at the Education Institute of the far-right Freedom Party of Austria. ²⁸ The seminars were advertised on the Institute’s website and via leaflets targeted at young voters. ²⁹ In the end, only two seminars were held. Both were attended by around 30 participants. ³⁰

In the course of the seminars, E.S. made two statements for which she would later be convicted. She first made the general claim that Muslims treat the Prophet Muhammad as an example of how to live their lives, which E.S. found problematic because she considered Muhammad’s behaviour incompatible with contemporary standards and values. To buttress this point, she made a specific allegation. Referring to religious scripture, she intimated – without stating in so many words – that Muhammad was a paedophile, since he had married one of his wives, Aisha, when she was six years old and consummated the marriage when she was nine. ³¹

Based on these statements, the Vienna Regional Criminal Court convicted E.S. under Article 188 of the Criminal Code for publicly disparaging the Prophet Muhammad in a manner capable of arousing justified indignation. ³² The court imposed a fine of €480. The conviction was upheld by the Vienna Court of Appeal and the Supreme Court. ³³ In their reasoning, the Austrian courts combined – for reasons unknown – two strands of the case law of the Strasbourg Court. ³⁴ Much of the Austrian courts’ reasoning was a direct application of the Strasbourg Court’s case law on offence to religious feelings, Otto-Preminger in particular. Nothing out of the ordinary, in that respect. Curiously, however, the Austrian courts also went beyond that case law by deploying concepts drawn from the Strasbourg Court’s defamation case law. In particular, they labelled the statements made by E.S. as derogatory value judgments that did not contribute to a debate of general interest. ³⁵ The courts held that the statements were instead intended to defame Muhammad by implying that he was not a worthy subject of worship because of his alleged paedophilia. ³⁶ They subsequently dedicated a sizeable portion of their reasoning to demonstrate that the allegation lacked the requisite proof. ³⁷ The courts emphasised that E.S. had disregarded the fact that Muhammad’s marriage continued until his death, at which point Aisha was no longer a minor. They also indicated that no documentary evidence existed to suggest that Muhammad’s other wives or concubines had been similarly young. The Austrian courts concluded that the applicant’s statements lacked a factual basis and constituted a malicious violation of the spirit of tolerance. The statements did not, therefore, enjoy the protection of freedom of expression. ³⁸

In reviewing the Austrian courts’ rulings, the Strasbourg Court made several predictable statements in direct reference to Otto-Preminger. The Court reiterated that ‘a religious group must tolerate the denial by others of their religious beliefs and even the propagation by others of doctrines hostile to their faith, as long as the statements at issue do not incite hatred or religious intolerance’. ³⁹ It also referred to the ‘duty to avoid as far as possible an expression that is, in regard to objects of veneration, gratuitously offensive to others and profane’. ⁴⁰ It further described the case as involving a conflict of rights, which called for a balancing exercise. ⁴¹ And it held that the Austrian authorities had a wide margin of appreciation in finding the right balance because they were ‘in a better position to evaluate which statements were likely to disturb the religious peace in their country’. ⁴² The Strasbourg Court finally ‘endorsed’ the assessment of the Austrian courts, noting that they ‘extensively explained why they considered that the applicant’s statements had been capable of arousing justified indignation’, since the statements ‘could only be understood as having been aimed at demonstrating that Muhammad was not a worthy subject of worship’. ⁴³ So far, this is nothing more than a relatively straightforward application of Otto-Preminger. Had the Court only said the above, E.S. would not have made for a particularly remarkable – yet, of course, still contestable – judgment. But the Court did not stop there. It also stated that it ‘agrees with the domestic courts that the impugned statements can be classified as value judgments without sufficient factual basis’. ⁴⁴ In doing so, the Strasbourg Court perpetuated, rather than corrected, the Austrian courts’ conflation of the two strands of case law. This conflation caused a few serious problems.

Problematic conflation of two strands of case law in E.S.

It is important to note that the distinction between value judgments and statements of fact, as well as the concomitant standards of proof, is prevalent in the Strasbourg Court’s case law. These notions are, however, most commonly deployed in the Court’s defamation case law. ⁴⁵ Before E.S., they had never featured explicitly in its case law on offence to religious feelings. ⁴⁶

Now, however, they are an overt part of that line of case law. In E.S., the Court recalled that ‘[i]n its case law [it] has distinguished between statements of fact and value judgments’ and repeated that ‘where a statement amounts to a value judgment, the proportionality of an interference [depends] on whether there exists a sufficient factual basis for the impugned statement’. ⁴⁷ For both propositions, the Court cited leading judgments from its defamation case law. ⁴⁸ The Court went on to apply these standards to the case at hand. In doing so, it stated that it ‘agrees with the domestic courts that the applicant must have been aware that her statements were partly based on untrue facts’. ⁴⁹ The Court concluded that ‘the impugned statements [by E.S.] were not phrased in a neutral manner aimed at being an objective contribution to a public debate […] but amounted to a generalisation without factual basis’. ⁵⁰

In E.S., the Strasbourg Court thus conflated (or, less pejoratively, combined) principles from its case law on offence to religious feelings with standards from its defamation case law. The source of this conflation is easy enough to identify. Under Otto-Preminger, the Austrian courts enjoyed a wide margin of appreciation. Under a more recent series of judgments, the Strasbourg Court would, moreover, ‘require strong reasons to substitute its view for that of the domestic courts’ when ‘the balancing exercise [between conflicting Convention rights] has been undertaken by the national authorities in conformity with the criteria laid down in the Court’s case-law’. ⁵¹ By applying both principles in E.S., the Strasbourg Court deferred to the reasoning of the Austrian courts. Whether deliberately or without reflection, this deference extended to the Austrian courts’ ‘borrowing’ of notions from defamation law.

Yet the ensuing conflation of the two strands of case law is deeply problematic. The Court effectively reduced a complex case involving the difficulty of balancing free speech and the preservation of religious tolerance to a single factual question: does having sex with one child 1,400 years ago merit being labelled a paedophile today? This narrow take on the case generated at least two serious problems.

First, it put the Strasbourg Court in the awkward position – after having been led there in the footsteps of the Austrian courts – of being compelled to investigate the truth value of religious scripture involving a person who lived 1,400 years ago. While religious scripture surely holds truth value to the adherents of the religion in question, it should be fairly obvious that this is not the kind of truth a court of law looks for in determining the factual basis for a value judgment. In defamation cases, the question before the Strasbourg Court is usually whether sources like government reports, confidential letters or previous newspaper articles provide a sufficient factual basis for the statements at issue. There is, in contrast, something distinctly alienating about the Court posing a similar question in relation to religious scripture involving disputed ⁵² events that are alleged to have taken place some 1,400 years ago. Notions such as ‘truth’ and ‘factual accuracy’ hardly seem appropriate here. By nonetheless investigating the ‘factual basis’ of E.S.’s statements, the Court came dangerously close to – at least implicitly – interpreting religious scripture and doctrine. It thereby took a stand against the broadly held scholarly position that this is not the role of secular courts. ⁵³ Although the Strasbourg Court has moreover agreed with this position in its broader freedom of religion case law, ⁵⁴ it arguably failed to live up to its own standards in E.S.

Second, by reducing E.S. to a single factual question – and deploying standards from defamation law to answer it – the Strasbourg Court effectively treated the case as entailing defamation of the Prophet Muhammad. As Milanovic and Bougiakiotis rightly point out, although the Court referred in E.S. to the ‘gratuitously offensive’ standard from Otto-Preminger, it did not actually apply this test to the case at hand. ⁵⁵ Instead, the Court joined the Austrian courts in resorting to the application of defamation law standards. As indicated above, this was not necessarily the result of a deliberate or conscious decision. The Strasbourg Court ‘simply’ endorsed the entirety of the Austrian courts’ reasoning, since it ‘would require strong reasons to substitute its view for that of the domestic courts’. ⁵⁶ The immediate result, however, is a judgment that is open to an entirely reasonable reading by which the central question is not one of gratuitous offence to the religious feelings of Muslims, but of defamation of Muhammad. Unsurprisingly, this invites the criticism that in E.S., the Court has given ‘its blessing to the criminalization of blasphemy, in all but name’. ⁵⁷

Like all other Chamber judgments, E.S. could, in theory, be referred to the Grand Chamber. But even if a referral request were to be submitted, ⁵⁸ it seems unlikely that the panel of five judges that reviews such requests would refer this particular case to the Grand Chamber, given that its outcome was entirely in line with long-established case law. Moreover, as I explain in the conclusion, E.S. is hardly the right case to overrule Otto-Preminger-Institut. Before concluding, however, I will argue that the Chamber judgment in E.S. is open to an alternative and arguably less contentious reading. On this alternative reading, the notion of tolerance – rather than value judgments – plays the leading role.

An alternative reading of E.S.: the procedural role of tolerance

It is well known that the arguments for religious tolerance made by John Locke, Pierre Bayle and Baruch de Spinoza were central to bringing about a (relatively) peaceful state of religious co-existence in early modern Europe. Some contemporary political philosophers insist that tolerance remains vital to ensuring peaceful co-existence in (European) societies marked by religious, cultural and ethnic diversity. ⁵⁹ At the same time, tolerance’s salience to the law – and the law on religious freedom in particular – remains underexplored. ⁶⁰

The Strasbourg Court routinely recalls that ‘[p]luralism, tolerance and broadmindedness are hallmarks of a “democratic society”’. ⁶¹ Yet the Court does not employ the notion of tolerance to resolve cases. In this, it does not, of course, stand alone. Since tolerance is usually understood as a moral virtue and/or political principle, courts tend to only reference it as ‘background scenery’ when adjudicating religious disputes. ⁶² A notable exception is the Supreme Court of Israel. In its case law, the Israeli Supreme Court has developed a ‘threshold of tolerance’ test, which it uses, inter alia, to adjudicate cases in which free speech collides with religious feelings. ⁶³ The ‘threshold of tolerance’ test is, however, extremely stringent: free speech prevails ‘unless the violation of religious feelings is nearly certain and their violation is real and severe’. ⁶⁴ The reason for the test’s stringency is that, unlike the Strasbourg Court, the Supreme Court of Israel has excluded religious feelings from the scope of freedom of religion. ⁶⁵ Instead, it treats the protection of religious feelings as a public interest. ⁶⁶ And, as a public interest, the protection of religious feelings occupies – within the conception of proportionality employed by the Israeli Supreme Court – a prima facie subordinate position to constitutional rights such as freedom of expression. ⁶⁷

I do not mean to suggest here that the Strasbourg Court ought to elaborate the ‘gratuitously offensive’ standard of Otto-Preminger into a full-fledged ‘spirit of tolerance’ test akin to the Israeli ‘threshold of tolerance’ test. Although that would give the notion of tolerance more substantive bite in the Court’s case law, it would also imply an increase in the protection given to free speech when it clashes with religious feelings. Much to the dismay of critics like Milanovic, the Strasbourg Court is clearly unwilling to go down this road.

Tolerance could, however, play a more robust procedural role in the resolution of cases involving offence to religious feelings. ⁶⁸ The Court’s judgment in E.S. is arguably open to a reading – provided we momentarily ignore its misconceived reliance on defamation law – by which tolerance plays precisely this role. In E.S., the Court first held that when ‘expressions go beyond the limits of a critical denial of other people’s religious beliefs and are likely to incite religious intolerance’, states may (not must) consider such expressions incompatible with freedom of religion. ⁶⁹ The Court went on to note that ‘Article 188 of the [Austrian] Criminal Code [aims] at the protection of religious peace and tolerance’. ⁷⁰ The Court then ruled that the Austrian courts enjoyed a wide margin of appreciation since ‘they were in a better position to evaluate which statements were likely to disturb the religious peace [and tolerance] in their country’. ⁷¹ Finally, the Court stated that it ‘endorses’ – on a purely procedural argument, ‘defers to’ would be preferable – the Austrian courts’ conclusion that the statements ‘could be conceived as a malicious violation of the spirit of tolerance’. ⁷² Here we have the broad outlines of a procedural reading of E.S. that avoids problematic arguments on the truth value of the statements at issue.

Admittedly, this procedural reading of E.S., which is centred on the notion of tolerance, is unlikely to convince the Court’s critics. ⁷³ Perhaps for good reason; one of the great difficulties with cases like E.S. remains that the curtailment of free speech is motivated by the authorities’ assumptions about offence to religious feelings, rather than evidence of actual offence. And regardless of whether the Strasbourg Court engages in either substantive or procedural review, assumptions hardly provide a solid basis for any (drastic) curtailment of the freedom of expression.

Conclusion

Irrespective of how one feels (no pun intended) about E.S., it is evident that the Strasbourg Court has backed itself into a corner with its case law on religious feelings. Otto-Preminger is generally considered bad case law. Yet the Court appears destined to repeat and even reinforce it. In the 2005 case I.A. v Turkey, which also involved disparaging statements about the Prophet Muhammad, there appeared to be an opening for the eventual overruling of Otto-Preminger. The Court decided against free speech in I.A., but only by the barest of majorities (4-3 judges). The three judges in dissent moreover argued that ‘the time has perhaps come to “revisit” [Otto-Preminger]’. ⁷⁴

More than a decade later, however, the Strasbourg Court has again confirmed Otto-Preminger, and this time unanimously. The persistence of Otto-Preminger is partly due to the wide margin of appreciation enjoyed by member states in this area of law. But the close alignment between Otto-Preminger and E.S. also played an important role. Arguably, the only major difference between both cases was the ‘targeted’ group. Otto-Preminger involved alleged offence to the religious feelings of the Christian majority. ⁷⁵ In E.S., by contrast, the supposedly offended believers were Muslims, a religious minority in an Austria that is now (though not at the time of the facts) led by a right-wing populist government that targets Islam. Given this constellation, distinguishing E.S. or using it as an occasion to overrule Otto-Preminger would surely have opened the Court up to renewed claims of prejudice against Islam. ⁷⁶ This eventuality may well have been on the judges’ minds when deciding E.S. This, combined with its consistent earlier case law and the workings of the (wide) margin of appreciation, makes it difficult to imagine how the Strasbourg Court could have ruled differently. Although this might serve to explain the outcome in E.S., it is hardly an excuse for poor reasoning.