1. Introduction

Drawing a line between one’s ‘right to offend’ and someone else’s ‘right not to be offended’ is a particularly arduous task when humor is involved. Due to its inherent ambiguity and elusiveness, humorous expression can be hard to decipher, and the difference between harmful behavior and an ‘innocent joke’ is often far from clear. A recent example is comedian Jo Brand’s joke on the milkshake attack against Brexit Party leader Nigel Farage in May 2019 (‘Why bother with a milkshake when you could get some battery acid?’), which resulted in a police investigation as well as in a heated public debate on the boundaries between dark humor and incitement to violence; the investigation was eventually dropped, due to the joke being uttered in a clearly signaled comedic context (Rawlinson and Siddique, 2019). The challenge is all the more evident in the case of predominantly visual humor, as best demonstrated through history by the genre of cartoons. Given their condensed format and their high degree of implicitness or ‘semiotic density’ (Pedrazzini and Scheuer, 2018), cartoons are a particularly vivid example of the link between humor and ambiguity, and of the ensuing difficulties in drawing a line between lawful and unlawful expression. Not by chance, cartoons have been at the center of several legal debates and actual court litigations in recent years. One of the most famous cases is the Muhammad cartoon controversy of 2005–2006, namely the first ‘transnational humor scandal’ (Kuipers, 2011), caused by the publication of twelve caricatures of Muhammad on the Danish newspaper Jyllands-Posten; on that occasion a trial never took place, but representatives of Muslim organizations did file a complaint with the Danish police thus leading to an investigation by the public prosecutor (Klausen, 2009). The same cartoons were then reprinted by Charlie Hebdo, which led instead to a trial at the Paris High Court (the magazine was eventually acquitted in 2007). Later instances include the lawsuits concerning Charlie Hebdo’s cartoon on the Amatrice earthquake (Griffen, 2016) and the one depicting far-right leader Marine Le Pen as a ‘steaming excrement’ (Chrisafis, 2019), as well as the allegedly antisemitic cartoon published by Brazilian newspaper O Dia in January 2019 (Harpin, 2019).

The juridical challenges posed by controversial humor are especially urgent in the digital age, where extreme speech and the reactions it may trigger are amplified by a growing set of ‘accelerants’ (Scharffs, 2017) including most notably the polarizing dynamics of social media. Yet, despite the topicality of the issue, there is still a striking gap in academic research on humor and the law. To be sure, an impressive amount of valuable scholarly work has been published on the legal regulation of freedom of expression, with satire often being discussed together with other modes of communication—recent landmark publications in this field, also featuring extensive bibliographies, include Kuhn (2019), Koltay and Temperman (2017), Garton Ash (2016), O’Reilly (2016), Waldron (2012), and Hare and Weinstein (2009). Nevertheless, nearly none of the available studies on the subject actively engage with the theoretical and methodological tools provided by humor studies; the only exception is represented by a small group of scholars focusing on the American context. More precisely, Parks and Heard (2010) combine federal case law with humor theory in their analysis of Sean Delonas’ infamous Obama cartoon published by the New York Post in February 2009, while Little (2019 and 2011) insightfully discusses First Amendment case law in light of incongruity theory and of the General Theory of Verbal Humor (Little 2011 also includes a comparative approach to Australian case law). Lastly, Todd (2016) draws on notions originating from literary studies (such as parody, intertextuality, and the distinction between Horatian and Juvenalian satire) to analyze a series of relevant cases. However, as pointed out by both Little and Todd, much work remains to be done in this sense; interdisciplinary dialogue between law and humor scholars is urgently required in order to provide courts and commentators with ‘an adequate terminology that is grounded in theory’, and thereby ‘clarify and rationalize the different outcomes’ reached in court (Todd, 2016: 69 and 61).

The present paper aims to mark a further step in the direction of what I propose to call forensic humor studies, i.e. the systematic application of insights from humor research to the juridical field. Here and throughout the paper, the word ‘humor’ is used in its broadest sense, encompassing all kinds of facetious expression—which also includes satire, commonly defined as a mode of communication where ridicule is used ‘to expose and criticize prevailing immorality or foolishness, esp. as a form of social or political commentary’ (Oxford English Dictionary). For the sake of feasibility, this pilot study will limit its scope to one specific genre of humorous expression (cartooning, with exceptions from neighboring forms of predominantly visual humor), and to the specific context of the European Court of Human Rights (ECtHR). The decision to focus on cartoons is due to their particular proneness to semiotic density and ambiguity, which makes them an ideal testing ground to investigate how humor is handled in court; however, many of the findings presented in this paper will hopefully be transferrable to other kinds of humorous texts as well (‘text’ is used in this paper in its broad semiotic sense, covering both verbal and non-verbal communication). The European Court of Human Rights, instead, was chosen because of its ongoing struggles in finding a consistent approach to the regulation of freedom of expression. Indeed, the ECtHR ‘has yet to define a single direction to be followed in this matter’ (Koltay, 2017: 245); it therefore stands out as a particularly compelling case study as well as an exemplary counterpart to the United States system, where freedom of expression is notoriously and consistently granted a privileged status in light of ‘the exceptional First Amendment’ (Schauer, 2005).

The following section will present the corpus of this study; Section 3 will provide a systematic overview of the criteria used by the ECtHR in the selected cases, thus setting the basis for a discussion of the problems underlying the current ECtHR approach (Section 4); Section 5 will demonstrate how relevant notions from humor studies can prove instrumental in addressing those problems, while the Conclusion will summarize the findings of the paper and outline possible directions for future research in the field. Before moving on to Section 2, one last remark is necessary: since this article is ideally addressed to both humor and legal scholars, a conscious effort was made to clarify all technical expressions from either side. A preliminary apology is therefore in order, should the phrasing of some passages be perceived as redundant by specialized readers in either discipline.

2. The Corpus

The European Court of Human Rights (ECtHR) was established in 1959; it deals with individual and State applications claiming violation of one or more rights established under the European Convention on Human Rights. The present paper focuses on cases revolving around the limits of freedom of expression, which is regulated by Article 10 of the Convention. The article reads as follows:

1. Everyone has the right to freedom of expression. This right shall include freedom to hold opinions and to receive and impart information and ideas without interference by public authority and regardless of frontiers. This article shall not prevent States from requiring the licensing of broadcasting, television or cinema enterprises.

2. The exercise of these freedoms, since it carries with it duties and responsibilities, may be subject to such formalities, conditions, restrictions or penalties as are prescribed by law and are necessary in a democratic society, in the interests of national security, territorial integrity or public safety, for the prevention of disorder or crime, for the protection of health or morals, for the protection of the reputation or rights of others, for preventing the disclosure of information received in confidence, or for masng the authority and impartiality of the judiciary.

Notably, the first paragraph of Article 10 is ordinarily applied ‘not only to “information” or “ideas” that are favorably received or regarded as inoffensive or as a matter of indifference, but also to those that offend, shock or disturb the State or any section of the population’, as first stated in Handyside v. United Kingdom (1976) and later reprised by the ECtHR in most Article 10 cases. Applications under the provision of Article 10 are typically presented by individuals or associations claiming that their right to freedom of expression was unlawfully interfered with by a ruling delivered in one of the contracting states; after the preliminary finding of admissibility, cases are tried by a chamber of seven judges or by a 17-member Grand Chamber. The outcome of the ECtHR decision may point to a violation of Article 10 on the part of the local court (thus protecting the applicant’s freedom of expression), or may back the decision made at a national level by not finding a violation of Article 10. The final decision may be shared unanimously by the Court, or include separate opinions (whether dissenting or concurring) by one or more judges.

Based on a systematic search through the comprehensive database of the ECtHR (HUDOC), it was possible to assemble a corpus of 10 cases regarding cartoons or other comparable forms of predominantly visual humor. The decision to focus on these examples is primarily due to the fact that cartoons stand out as a particularly compact and semantically dense form of communication, and therefore represent an ideal testbed for analyzing how judges deal with the ambiguity and elusiveness of humor. Since the selected cases will be the object of close comparative inspection in the following sections, and considering that the present study is meant for a broad interdisciplinary readership, concise lay summaries of each case are provided below:

• – Case of Vereinigung Bildender Künstler v. Austria (68354/01, final decision 25 April 2007): Between April and June 1998, the applicant (an association of artists called Vereinigung Bildender Künstler Wiener Secession) held an exhibition including the satirical painting ‘Apocalypse’ by Austrian painter Otto Mühl. The painting featured photomontages showing members of the conservative right-wing Austrian Freedom Party (as well as religious figures such as Mother Teresa and Austrian cardinal Hermann Groer) engaging in graphic sexual acts. One of the politicians portrayed in the painting, namely former Freedom Party secretary Walter Meischberger, filed a lawsuit against the association. Outcome at a national level: The artists were fined, and permanently banned from displaying the painting. ECtHR decision: Violation of Article 10 (four votes to three). (NB: ‘Violation of Article 10’ means that the ECtHR ruled in favor of the applicant’s freedom of expression; ‘No violation’ means that it ruled against the applicant’s freedom of expression, thus backing the previous decision taken at a national level.)

• – Case of Kuliś and Różycki v. Poland (27209/03, final decision 6 January 2010): On 16 May 2009, children’s magazine Angorka (represented by the applicants) published a cartoon satirizing the advertising campaign launched by food company Star Foods to promote its potato chips. The Star Foods campaign was based on a series of images that were deemed inappropriate for a product that was mostly meant for children, including one advertisement accusing Reksio (a popular cartoon character for Polish children) of being a murderer. Angorka’s cartoon showed a boy holding a packet of Star Foods potato chips, and saying to Reksio: ‘Don’t worry! I would be a murderer too if I ate this muck!’ The heading above the cartoon read ‘Polish children shocked by crisps advertisement, “Reksio is a murderer”’. Outcome at a national level: The applicants were convicted of defamation. ECtHR decision: Violation of Article 10 (unanimous).

• – Affaire Leroy c. France (36109/03, final decision 6 April 2009): On 13 September 2001, Basque magazine Ekaitza published a satirical cartoon by the applicant ostensibly celebrating the 9/11 attacks. The cartoon showed two planes crashing into the Twin Towers, accompanied by the caption ‘We all dreamed of it… Hamas did it’ (echoing the slogan of a Sony advertising campaign). Outcome at a national level: The applicant was convicted of apology of terrorism. ECtHR decision: No violation of Article 10 (unanimous).

• – Affaire Société de Conception de Presse et d’Édition et Ponson c. France (26935/05, final decision 5 June 2009): In June 2002, the widely read magazine Entrevue (represented by the applicant) published an article on the most paid athletes in the world, featuring a photograph of Michael Schumacher sporting the logo of a cigarette brand and highlighting that much of his income originated from his sponsor. The magazine also featured a satirical photomontage showing two packets of cigarettes of the same brand engaging in anal sex, accompanied by the caption ‘Warning: Smoking… gives you anal cancer’. Both the photograph and the photomontages were impugned for illegal advertising, and were discussed in separate paragraphs in the ECtHR decision. Due to its humorous component, only the photomontage is relevant for our analysis. Outcome at a national level: The applicant was convicted of illegal advertising. ECtHR decision: No violation of Article 10 (unanimous).

• – Case of Aguilera Jimenez and Others v. Spain (28389/06, 28955/06, 28957/06, 28959/06, 28961/06 and 28964/06, referred to the Grand Chamber on 10 May 2010): The applicants were employed as delivery men by the company P., against which they had filed a series of proceedings in employment tribunals. In 2001 they set up the trade union N.A.A. to defend the interests of the delivery staff. In April 2002, the trade union bulletin published a cartoon depicting two employees of the company waiting in line to perform oral sex to the company’s human resources manager; the two employees portrayed in the cartoon had testified against the trade union in the proceedings mentioned above. Upon being fired by the company as a result of the publication of the cartoon, the applicants challenged the company’s decision in court and evoked their right to freedom of expression. Outcome at a national level: The applicants’ claim was dismissed. ECtHR decision: No violation of Article 10 (six votes to one).

• – Case of Palomo Sánchez and Others v. Spain (28955/06, 28957/06, 28959/06 and 28964/06, final decision 12 September 2011): This is the Grand Chamber continuation of Aguilera Jimenez and Others, confirming the previous verdict. ECtHR decision: No violation of Article 10 (twelve votes to five).

• – Affaire Féret c. Belgique (15615/07, final decision 10 December 2009): Between 1999 and 2001, the right-wing populist party Front National (represented by the applicant) published a series of xenophobic cartoons and leaflets online and in print. One of the cartoons explicitly attributed the 9/11 attacks to the ‘couscous clan’, establishing a direct link between Islam and terrorism. Outcome at a national level: The applicant was convicted of incitement to hatred. ECtHR decision: No violation of Article 10 (four votes to three).

• – Case of Bohlen v. Germany (53495/09, final decision 19 May 2015): The applicant, German musician Dieter Bohlen, had published an autobiography titled Backstage in 2003; the book had to be heavily redacted because of a series of defamation proceedings which had been brought against the author. In October 2003, British American Tobacco GmbH launched a visual advertisement for Lucky Strike in Germany featuring two packets of cigarettes, a black marker, and the ostensibly redacted slogan ‘Look, dear Dieter, how easy it is to write super books’. Bohlen sued Lucky Strike for the allegedly unlawful and derogatory use of his forename, under the provision of Article 8 of the Convention (protecting personality rights); the company’s right to advertise its product, instead, was discussed by the ECtHR in the framework of Article 10. Outcome at a national level: The applicant’s claim was dismissed. ECtHR decision: No violation of Article 8 (six votes to one); the company’s freedom of expression is therefore protected.

• – Case of Ernst August von Hannover v. Germany (application no. 53649/09, final decision 19 May 2015): The case is formally identical to Dieter Bohlen. In 2000, a Lucky Strike visual advertisement was launched showing a crumpled pack of cigarettes accompanied by the slogan ‘Was it Ernst? Or was it August?’ The slogan alluded to Prince Ernst August of Hannover, who was involved in a series of scuffles and altercations between 1998 and 2000, thereby gaining a reputation as a violent and irascible person. Von Hannover sued British American Tobacco under the provision of Article 8; the company’s right to advertise the product was discussed by ECtHR in the framework of Article 10. Outcome at a national level: The applicant’s claim was dismissed. ECtHR decision: No violation of Article 8 (six votes to one); the company’s freedom of expression is therefore protected.

• – Case of Sekmadienis Ltd. v. Lithuania (69317/14, final decision 30 April 2018): In September and October 2012, the applicant company ran an advertising campaign for a clothing collection. The ads featured photographs of models whose looks hinted at Jesus and Mary, accompanied by the slogans ‘Jesus, what trousers!’, ‘Dear Mary, what a dress!’, ‘Jesus [and] Mary, what are you wearing!’ Outcome at a national level: The applicant company was convicted of ‘violating public morals’. ECtHR decision: Violation of Article 10 (unanimous, with one concurring opinion).

3. Identifying the Criteria

For each of the cases outlined above, the sections containing the arguments and decisions of ECtHR judges (Law, Judgement, and Dissenting or Concurring Opinions) were analyzed in order to tease out the criteria underlying the Court’s reasoning. On a general level, the selected cases—as is customary with Article 10 case law—adopt a three-part test for assessing restrictions on freedom of expression, which is articulated as follows: 1) the restriction must be prescribed by law; 2) the restriction must protect one of the interests listed in the second paragraph of Article 10; 3) the restriction must be ‘necessary in a democratic society’ and proportionate to the legitimate aim pursued. In relevant cases, the three-part test can be combined with the ‘abuse clause’ of Article 17, whereby freedom of expression should not imply any right to engage in the ‘destruction’ or ‘limitation’ of any of the rights and freedoms set forth in the Convention. Article 17 has been used by the ECtHR either as an interpretative aid within the three-part test, or as a ‘guillotine’ allowing for the categorical exclusion of certain forms of expression from the protection granted by Article 10;¹ within our corpus, however, Article 17 is only mentioned in one case (Féret), and its applicability is unanimously rejected by the Court.

In most Article 10 jurisprudence (including our corpus cases), the crucial element of the test is its last step, i.e. the ‘necessity test’. In order to determine whether restricting a certain form of expression is necessary in a democratic society, the Court may assess the application based on a variety of criteria, depending on the specific circumstances of the case. As far as our corpus is concerned, the following 10 criteria could be identified:

1. Humorous or satirical intent: The humorous or satirical intention of the text is marked by clear indicators, which allows for a higher degree of protection v. the text’s intention is not clearly indicated, and therefore there are no excuses for its violent and offensive tone. This criterion is used in all 10 cases. The first two examples refer to the first scenario (clear satirical intent), while the last two refer to the latter:

• ‘It was common ground in the understanding of the domestic courts at all levels that the painting obviously did not aim to reflect or even to suggest reality … The Court finds that such portrayal amounted to a caricature of the persons concerned using satirical elements. It notes that satire is a form of artistic expression and social commentary and, by its inherent features of exaggeration and distortion of reality, naturally aims to provoke and agitate. Accordingly, any interference with an artist’s right to such expression must be examined with particular care.’ (Vereinigung Bildender Künstler, §33)

• ‘As regards the cartoon on the newsletter’s cover, it is a caricature, which, while being vulgar and tasteless in nature, should be taken for what it is – a satirical representation. In other cases, the Court has recognised the satirical nature of an expression, publication or caricature.’ (Palomo Sanchez, Dissenting Opinion, §11)

• ‘An image will not become ‘satirical’ if the observer does not comprehend or detect any message in the form of a meaningful attack or criticism relating to a particular problem or a person’s conduct … [The painting] showed a number of unrelated personalities (some political, some religious) in a vulgar and grotesque presentation and context of senseless, disgusting images of erect and ejaculating penises and of naked figures adopting repulsive sexual poses, some even involving violence, with coloured and disproportionately large genitals or breasts.’ (Vereinigung Bildender Künstler, Dissenting Opinion of Judge Loucaides)

• ‘Les intentions du requérant … n’ont d’ailleurs été exprimées que postérieurement et n’étaient pas de nature, au vu du contexte, à effacer l’appréciation positive des effets d’un acte criminel.’ [The applicant’s intentions (i.e. satirizing US consumerism by echoing the Sony slogan) were only expressed ex post; and given the context, they were not enough to erase the positive assessment of a criminal act.] (Leroy, §43)

Detecting (or failing to detect) a humorous intent may orientate the Court’s assessment of whether the text constitutes a ‘defamatory statement of fact’ or merely a ‘value judgement’ (Kuliś and Różycki, §38); moreover, it often sets the basis for determining the aims of the text (gratuitous offense v. public interest, cf. criterion 8) and its possible effects (damage to dignity or threat to public peace, criteria 9–10).

2. Explicitness of the message: The potentially offensive message can only be deciphered by a relatively small audience (extenuating factor) v. the potentially offensive message is delivered in an explicit, assertive way (aggravating factor). Occurrences in the corpus: 7 cases.² The first two examples illustrate the former scenario, while the last one refers to the latter:

• ‘Only a limited number of people would have been able to make the connection between the advertisement and the applicant, namely those who had heard about the applicant’s scuffles, especially as the latter were not mentioned in the advertisement but were hinted at in a clever way.’ (Ernst August von Hannover, §54)

• ‘Even before Mr Meischberger instituted proceedings, the part of the painting showing him had been damaged so notably that the offensive painting of his body was completely covered by red paint. The Court considers that, at the very latest from this incident onwards, Mr Meischberger’s portrayal – even assuming that he was still recognisable … – was certainly diminished, if not totally eclipsed.’ (Vereinigung Bildender Künstler, §36)

• ‘The implication of the advertisement is clear enough. It insinuates that Mr Bohlen would not have committed the alleged mistakes in his writing if only he smoked Lucky Strike cigarettes. The lit cigarette on top of the cigarette box is a clear recommendation to this effect. The message is not even subliminal; it is assertive and suggestive.’ (Dieter Bohlen, Dissenting Opinion)

3. Context: The geographical, social or historical context of publication serves as a mitigating circumstance v. it serves as an aggravating circumstance. Occurrences in the corpus: 6 cases.³ The first two examples below cover the first scenario, while the third one refers to the latter:

• ‘The scene in which Mr Meischberger was portrayed could be understood to constitute some sort of counter-attack against the Austrian Freedom Party, whose members had strongly criticised the painter’s work.’ (Vereinigung Bildender Künstler, §34)

• ‘Est-il équitable d›attribuer un sens islamophobe qui serait apparu en septembre 2001 (la caricature du ‘couscous clan’) à des textes distribués en 1999?’ [Is it fair to attribute an Islamophobic sense that would only appear in September 2001 (the ‘couscous clan’ caricature) to texts that were distributed in 1999?] (Féret, Dissenting Opinion)

• ‘La caricature a pris une ampleur particulière dans les circonstances de l’espèce, que le requérant ne pouvait ignorer. Le jour des attentats, soit le 11 septembre 2001, il déposa son dessin et celui-ci fut publié le 13 septembre, alors que le monde entier était sous le choc de la nouvelle, sans que des précautions de langage ne soient prises de sa part. … De plus, l’impact d’un tel message dans une région politiquement sensible n’est pas à négliger.’ [The caricature took on a particular significance in the circumstances of the case, which the applicant could not ignore. On the day of the attacks, i.e. 11^th September 2001, he submitted his cartoon which was published on 13^th September, when the world was still in shock, without any linguistic precaution. Moreover, the impact of such a message in a politically sensitive region (= the Basque Country) is not to be overlooked.] (Leroy, §45)

4. Intertextuality: The meaning of the text is clarified (in an extenuating or aggravating sense) by means of intertextuality, i.e. by highlighting the relation to a previous text or common expression that the impugned work is echoing or reacting against (the notion of intertextuality is used here in the sense defined by Genette, 1997). Occurrences in the corpus: 3 cases.⁴ Examples:

• ‘The wording employed by the applicants had been exaggerated; however, they were reacting to slogans used in the plaintiff’s advertising campaign which also displayed a lack of sensitivity … The Court thus considers that the style of the applicants’ expression was motivated by the type of slogans to which they were reacting and, taking into account its context, did not overstep the boundaries permissible to a free press.’ (Kuliś and Różycki, §38)

• ‘The names of Jesus and Mary in the advertisements had been used not as religious references but as emotional interjections common in spoken Lithuanian, thereby creating a comic effect.’ (Sekmadienis Ltd., §79)

5. Lucid deliberation: The offensive message was uttered in the heat of the moment v. it was meant for publication online or in print, and is therefore the result of lucid deliberation. Occurrences in the corpus: 2 cases.⁵ Example: ‘The remarks did not constitute an instantaneous and ill-considered reaction, in the context of a rapid and spontaneous oral exchange, as is the case with verbal exaggeration. On the contrary, they were written assertions, published in a quite lucid manner and displayed publicly on the premises of the company P.’ (Palomo Sanchez, §73).

6. Reasonable avoidability: The text was only visible to relatively small groups of people who deliberately chose to be exposed to it (i.e. the text passes the ‘reasonable avoidability’ test, as defined in Feinberg, 1988: 32–33) v. the text was publicly displayed, and its view might have been imposed on an unaware audience. Occurrences in the corpus: 3 cases.⁶ Examples:

• ‘Les sites web se distinguent d’autres formes de distribution parce … les intéressés doivent rechercher eux-mêmes activement l’information. Autrement dit, les opinions ne sont pas ‘imposées’. [Websites differ from other forms of distribution because those who are interested have to actively look for the content. In other words, opinions are not ‘imposed’.] (Féret, Dissenting Opinion)

• ‘It might be useful to add that the large-sized painting in question was exhibited in an art gallery open to the general public so that even children could find themselves viewing it.’ (Vereinigung Bildender Künstler, Dissenting Opinion of Judge Loucaides)

• ‘[The cartoon was] displayed publicly on the premises of the company P.’ (Palomo Sanchez, §73)

7. Target: The text targets a public figure, which calls for a wider tolerance of criticism v. the text targets private individuals. Occurrences in the corpus: 5 cases.⁷ Examples:

• ‘The painting could hardly be understood to address details of Mr Meischberger’s private life, but rather related to Mr Meischberger’s public standing as a politician from the FPÖ. The Court notes that in this capacity Mr Meischberger has to display a wider tolerance in respect of criticism.’ (Vereinigung Bildender Künstler, §34)

• ‘The applicant belonged to the group of public figures who cannot claim protection of their right to respect for their private life in the same way as private individuals unknown to the public.’ (Ernst August von Hannover, §50)

• ‘[The cartoon is aimed] not directly against the company but against the two non-salaried delivery men and the human resources manager. The Court reiterates in this connection that the extent of acceptable criticism is narrower as regards private individuals than as regards politicians or civil servants.’ (Palomo Sanchez, §71)

8. Public interest: The text contributes to a public debate, or expresses an opinion on a subject of public interest v. the text is gratuitously offensive. This criterion is used in all 10 cases. The first three examples below refer to the former scenario (contribution to public debate), while the last two refer to the latter (gratuitous offense):

• ‘The Court considers that the domestic courts did not give sufficient attention to the applicants’ argument that the satirical cartoon had been a riposte to, in the applicants’ view, an unacceptable advertising campaign conducted by Star Foods and targeted at young children. The campaign used slogans referring not only to the Reksio character, but also to sexual and cultural behaviour, in a manner scarcely appropriate for children – the intended market segment. This clearly raises issues which are of interest and importance for the public.’ (Kuliś and Różycki, §37)

• ‘L’arrêt admet que les propos de M. Féret relèvent du ‘discours politique’.’ [The judgement acknowledges that Mr Féret’s statements pertained to political discourse.] (Féret, Dissenting Opinion)

• ‘As regards the existence of a debate of general interest, the Court notes that the German courts found that the impugned advertisement concerned a subject of public interest in so far as it referred humorously to the case of the applicant’s publication of a book, shortly after the event and in the context of the ensuing media debate on the subject.’ (Dieter Bohlen, §50)

• ‘The images depicted in this product of what is, to say the least, a strange imagination, convey no message; the “painting” is just a senseless, disgusting combination of lewd images whose only effect is to debase, insult and ridicule each and every person portrayed.’ (Vereinigung Bildender Künstler, Dissenting Opinion of Judge Loucaides)

• ‘The cartoon and some of the allegations contained in the articles from the offending bulletin constituted, by their gravity and tone, attacks of a personal, offensive, excessive and gratuitous nature that were certainly not necessary for the legitimate defense of the applicants’ interests.’ (Aguilera Jimenez, §34).

9. Damage to dignity: The text can significantly damage the dignity or reputation of (groups of) individuals v. the text is not likely to bring significant damage. Occurrences in the corpus: 8 cases.⁸ Examples:

• ‘The cartoon and the two impugned articles in the bulletin published and displayed by the applicants on the company’s premises had been offensive and likely to harm the reputation of others.’ (Aguilera Jimenez, §30)

• ‘To maintain that Mr von Hannover deserved the ‘particularly clever’ (pfiffig) negative publicity on account of his bellicose character … and simultaneously that because this was “only” a cigarette advertisement, it was not injurious to his personality rights, is going too far.’ (Ernst August von Hannover, Dissenting Opinion)

10. Threat to public peace or health: the text constitutes a threat to public peace or health v. the text is not likely to constitute a threat. Occurrences in the corpus: 4 cases.⁹ Examples:

• ‘Un tel discours est inévitablement de nature à susciter parmi le public … des sentiments de mépris, de rejet, voire, pour certains, de haine à l’égard des étrangers.’ [Such a discourse is inevitably bound to create in the public sentiments of contempt, rejection, and hate towards foreigners.] (Féret, §73)

• ‘Having viewed the advertisements for itself, the Court considers that at the outset they do not … incite hatred on the grounds of religious belief or attack a religion in an unwarranted or abusive manner.’ (Sekmadienis Ltd., §77)

It should be noted that only criterion 1 (humorous or satirical intention) is specifically related to humor and satire; the others also apply to non-humorous expression, and are in fact ordinarily used in Article 10 case law. However, many of these criteria are undoubtedly perturbed by the presence of irony, parody, facetious exaggeration and other humorous techniques. This is especially the case with the criteria that rely most closely on the direct interpretation of the text, with a view to reconstructing its meaning and intention: humor can make it particularly difficult to determine the intent and explicitness of the message (criteria 1-2), to decide whether the text is gratuitously offensive or addresses issues of public interest (criterion 8), and to draw a line between a joke and a direct threat to human dignity or public peace (9–10). In short, the specificity of humor-related cases in Article 10 jurisprudence does not lie in the choice of the criteria used for the necessity test (which tend to be same as in non-humor-related cases), but rather in how humor complicates the handling of many of these criteria. The specific challenges posed by the forensic assessment of humor will be further discussed in Sections 4 and 5.

4. Humor as a Challenge

The particularly problematic nature of humor-related cases is confirmed by the simple observation that, out of the 10 rulings under examination, seven feature separate opinions by one or more judges (six dissenting and one concurring); and although the corpus is not statistically representative in the broader context of Article 10 rulings, this quantitative finding is remarkable in itself, as it suggests a significantly higher frequency of separate opinions in humor-related cases compared to the ECtHR average of 53% (Franck, 2019; an advanced search via the HUDOC database in January 2020 yielded a similar result, with 2317 separate opinions out of 4129 cases). A close comparison between majority rulings and separate (especially dissenting) opinions highlights the centrality of criteria 1,2, 8, 9 and 10 from the list above—in other words, disagreements between majority and minority opinions usually revolve around the criteria that are most closely related to the direct interpretation of the text, namely the presence of a clear humorous intent and the reconstruction of the general aim or specific message of the impugned text. For example, with regard to criterion 1, the majority in Vereinigung Bildender Künstler states that Mühl’s painting ‘obviously’ had a satirical intent (§33), while Judge Loucaides objects that ‘in [his] view, the picture in question cannot, by any stretch of the imagination, be called satirical’ (Vereinigung Bildender Künstler, Dissenting Opinion of Judge Loucaides). Similarly, the dissenting opinion in Palomo Sánchez argues that the majority did not pay sufficient attention to the evident humor indicators offered by the text: ‘[The cartoon] is a caricature, which, while being vulgar and tasteless in nature, should be taken for what it is – a satirical representation. In other cases, the Court has recognised the satirical nature of an expression, publication or caricature’ (§11).

But even when judges do concur in recognizing humor markers, they might disagree when it comes to reconstructing the message hiding behind the humorous surface—a task which is essential to the implementation of criteria 8 (gratuitous offence v. public interest), 9 (damage to dignity) and 10 (threat to public peace). Ernst August von Hannover and Dieter Bohlen are a fine case in point: the majority claims that the Lucky Strike ads ‘had been devoid of any offensive or degrading content in relation to the applicant [and] had not been disparaging’, but rather ‘concerned a subject of public interest’ in so far as they referred to the applicants’ widely known public image (Hannover, §53 and Bohlen, §54); the dissenting judges, instead, maintain that both texts are clearly ‘mocking’ the applicants in a gratuitous way which is not justified by the aim of selling cigarettes (Hannover and Bohlen, Dissenting Opinions). In both cases, the interpretive disagreement also concerns the degree of explicitness of the message (criterion 2), with the dissenting opinions pointing out that the ads’ message ‘is not even subliminal; it is assertive and suggestive’, as opposed to the majority statement that the applicants’ deeds ‘were not mentioned in the advertisement’ but were only ‘hinted at in a clever way’ (Ernst August von Hannover, §54). The handling of criterion 8 is also an object of contention in Vereinigung Bildender Künstler, where the majority considers the painting as a piece of ‘social commentary’, while the dissenting opinion writes it off as gratuitous provocation: ‘the images depicted in this product of what is, to say the least, a strange imagination, convey no message; the “painting” is just a senseless, disgusting combination of lewd images whose only effect is to debase, insult and ridicule each and every person portrayed’ (Vereinigung Bildender Künstler, Dissenting Opinion of Judge Loucaides).

On a side note, it might be useful to remark that the distinction between ‘gratuitously offensive’ texts and contributions to ‘public debate’ (criterion 8) is widely criticized in recent scholarship. In particular, the very notion of ‘offense’ is highly problematic from a juridical standpoint, as it refers to the subjective emotional impact of a given text rather than to the objective damage created by said text. An effective solution to this problem was put forward by Philippe Yves Kuhn in an article on religious and racial hate speech (Kuhn, 2019), which follows Jeremy Waldron’s dignitarian approach in distinguishing harm from offense—the former is defined as ‘undermining a person’s dignity’ (i.e. ‘objective or social aspects of a person’s standing in society’), while the latter refers to ‘subjective aspects of feeling, including hurt, shock, and anger’ (Waldron cited in Kuhn, 2019: 129). It follows that harm is a better guide than offense when it comes to regulating freedom of expression, as it obviates the subjective pitfalls of the ‘gratuitously offensive’ test. Kuhn therefore proposes to discard offense as a relevant criterion and to only focus on the relatively more objective notion of ‘harm’, which he defines in terms of ‘seriously undermining the target’s assurance as to a status of equal worth in the community, having regard to the target’s knowledge, the speaker’s power and the forum of the expression, at the time it is made’ (2019: 120). Kuhn’s argument is compelling; and indeed, several cases in the corpus acknowledge the protection of human dignity and public peace as the fundamental reasons to restrict freedom of expression (cf. criteria 9 and 10). However, it should be stressed that the higher degree of objectivity inherent to the notion of harm is far from eliminating all traces of subjective interpretation, as implied by Kuhn when he points out that in order for the harm test to work, ‘it is important to recognise the subtextual meanings that hostile messages, signs or slurs can convey to their targets’ (2019: 128). The detection and interpretation of subtextual meanings is obviously bound to retain a certain degree of subjectivity, especially when dealing with humor and satire. This subjective component is abundantly illustrated in our corpus, as the conflict between majority and dissenting opinions often revolves around whether the text is harmful to the dignity of the target (Vereinigung Bildender Künstler, Aguilera Jimenez, Palomo Sánchez, Dieter Bohlen and Ernst August von Hannover), or more generally to public peace (Féret).

Lastly, provided that judges do agree on one interpretation of the text, one last interpretive issue may arise: to what extent should the ruling also account for different interpretations? Where does the author’s responsibility end, if the text might be (mis)interpreted in a dangerous, offensive or harmful way by some groups of readers? According to Kuhn, the harmful potential of a given expression should be ‘adjudged from the perspective of a reasonable member of the target group’ (2019: 120, my emphasis)—but reasonability is an extremely slippery and culturally determined concept, which leaves considerable leeway for interpretation (Moran, 2007). This issue is particularly relevant in Féret, where the majority decision takes into account the potential effect of the impugned texts on an ill-equipped or irrational audience: ‘Un tel discours est inévitablement de nature à susciter parmi le public, et particulièrement parmi le public le moins averti, des sentiments de mépris, de rejet, voire, pour certains, de haine à l’égard des étrangers’ [such a discourse is inevitably bound to create in the public, especially among the lesser-equipped, sentiments of contempt, rejection, and hate towards foreigners] (Féret, §73; my emphasis). Such an approach is heavily criticized in the dissenting opinion by Judges Sajó, Zagrebelsky and Tsotsoria, which reads as follows: ‘L’arrêt … juge des êtres humains et toute une couche sociale de “nigauds” incapables de répondre aux arguments et aux contre-arguments en raison de la pulsion irrésistible de leurs émotions irrationnelles’ [the judgement considers some human beings and a whole sector of society as ‘simpletons’ incapable of responding to arguments and counterarguments due to the irresistible pulsion of their irrational emotions]. In the dissenting judges’ view, the court should therefore only focus on how the text under examination might be interpreted by reasonable readers (however ‘reasonability’ is defined), rather than holding the author accountable for possible unreasonable interpretations.

In short, a cross-analysis of majority and minority opinions reveals major interpretive tensions on three levels: does the text clearly signal its humorous or satirical intent? What is the aim or message hiding behind the humorous surface? And to what extent should alternative interpretations (by a reasonable or unreasonable audience) be accounted for? To be sure, the recurrence of such questions in the corpus is not surprising or problematic as such; nevertheless, these hermeneutic conflicts are visibly multiplied and exacerbated by the lack of a shared vocabulary and a clear theoretical framework allowing judges to deal with humorous texts in a systematic and consistent way. Notably, however, these three questions can set the basis for streamlining the treatment of humor in Article 10 cases: while being compatible with the criteria presented in Section 3, they can provide judges with an efficient blueprint for the application of said criteria. As detailed in the following section, insights from humor studies can provide substantial help in this respect.

5. Towards Forensic Humor Studies

The present section will discuss how forensic humor studies can set the basis for a consistent treatment of the three questions outlined above: 1) does the text clearly signal its humorous or satirical intent? 2) What is the aim or message hiding behind the humorous surface? 3) To what extent should alternative interpretations (by a reasonable or unreasonable audience) be accounted for? Starting with Question 1, a substantial contribution can be derived from two studies focusing on how humor is processed by readers—namely Wayne Booth’s A Rhetoric of Irony (1974) and Paul Simpson’s On the Discourse of Satire (2003). Although both books only focus on specific humorous devices (irony, i.e. meaning something different from what is stated at surface level) or modes (satire), certain aspects of their respective models are highly transposable, and can pave the way for a forensic analysis of humor. With regard to humor indicators, a useful starting point is provided by Booth’s distinction between ‘overt’ and ‘covert’ irony: the former is characterized by direct assertions announcing the ironic intent (e.g. ‘It is ironic that…’), while the latter is intended to be reconstructed through the ‘secret work’ of the reader (Booth, 1974: 234). Despite referring to irony specifically, these labels can easily be transferred to humor at large: direct statements are indeed the first and most obvious indicator that a given expression is not meant to be taken seriously. While such cases are relatively rare, overt markers can still have juridical relevance, as was the case with the ‘battery acid’ joke mentioned in the Introduction: the fact that Jo Brand overtly indicated her humorous intention right after uttering the joke (‘I’m not going to do it, it’s purely a fantasy, but I think milkshakes are pathetic, I’m sorry’) played a key role in Scotland Yard’s decision to drop the investigation into the comedian (Rawlinson and Siddique, 2019). That being said, it is quite uncommon for humor to come with such an explicit label on it; it is therefore essential to develop a shared procedure for detecting and interpreting more ‘covert’ humor markers. In this respect, further inspiration comes from Booth’s four-step procedure for reconstructing irony in literary fiction, which is reproduced below in the concise summary provided by Liesbeth Korthals Altes:

(1) A reader rejects the surface meaning of an utterance, because she or he feels there is an incongruity “between the words and something else he knows” (10), especially between the beliefs she or he holds and those she or he thinks the author holds (73). This is where irony signals—incongruities or illogicalities, for instance—play their part.

(2) The reader then tries out alternatives, which come “flooding in” (11). These first two steps by themselves, however, cannot tell us that a statement is ironic; that requires

(3) The (re-)construction of the beliefs and intentions of the utterance’s ultimate source—in the case of literature, the implied author (11).

(4) Finally, the reader constructs a meaning in harmony with those hypotheses (10–12).

(Korthals Altes, 2014: 125; my emphasis)

‘Incongruities’ and ‘illogicalities’ are not only ‘irony signals’—they can serve, more generally, as covert (but nevertheless clear) indicators of any kind of humorous strategy. Also from a legal perspective, incongruity can effectively serve as a ‘testing rod’ for determining ‘whether a communication can be characterized as humorous’ (Little, 2011: 157). Humorous incongruity may present itself in several different forms, including among others non-sequiturs, anachronisms (Mary and Jesus wearing modern clothes and sporting tattoos in Sekmadienis Ltd.), exaggeration (the hyperbolic sexual scenes debated in Vereinigung Bildender Künstler or Palomo Sánchez), and the parodic echoing of previous texts or images (such as the Star Foods campaign in Kuliś and Różycki or the Sony slogan in Leroy). Parody, in turn, can be announced by a broad range of indicators, from the exaggeration of certain perceived traits (with its aggressive language, the Kuliś and Różycki cartoon reproduces and doubles up on the inappropriateness of the Star Foods campaign) to jarring decontextualization (the Leroy cartoon uses the Sony slogan in an entirely different context); a more fine-grained exploration of parodic mechanisms can be found in Genette (1997).

A valuable contribution towards a systematic classification of humorous incongruities is offered by Simpson’s analysis of satirical discourse, which too can be fruitfully extended to humor at large. In particular, Simpson distinguishes between ‘metonymic’ and ‘metaphoric’ mechanisms, respectively taking place within the same conceptual domain (as is the case with metonymy) or bridging different domains as metaphors do (2003: 125–149). The former category encompasses the following processes: 1) Saturation, such as caricature and other forms of exaggeration (e.g. the hyperbolic sexual content of Vereinigung Bildender Künstler); 2) Attenuation or ‘undercoding’, i.e. subtle allusion achieved by way of withholding information (e.g. the Lucky Strike ad hinting at Ernst August von Hannover’s scuffles without mentioning them explicitly); 3) Negation or reversal (for example, although this is never mentioned in the ECtHR judgement, the Lucky Strike campaigns represent a playful reversal of the ‘celebrity ad’ format, whereby the celebrity is not actually present in the ad). Metaphoric mechanisms, instead, are based on the decontextualization, merging or juxtaposition of elements coming from different conceptual domains, which are therefore at odds with each other: relevant examples include the combination of sexual imagery with prominent religious figures in Otto Mühl’s painting (Vereinigung Bildender Künstler), the use of a Sony slogan in a cartoon about 9/11 (Leroy), or the anachronisms involving Mary and Jesus in Sekmadienis Ltd.

As shown by this brief overview, Booth’s and Simpson’s models facilitate the systematic detection and discussion of humor indicators in a given text, thus allowing for a consistent handling of Question 1 above. Moving on to Question 2 (what is the aim or message hiding behind the humorous surface?), it becomes necessary to engage in what Booth designates as the ‘reconstruction’ of the text’s ultimate meaning. This stage is referred to by Simpson as the uptake phase, which follows the prime (taking the text at face value) and the dialectic phase (detecting the humor markers). In this final interpretive stage, the framing of the text in light of humor indicators is normally combined with contextual considerations; in Booth’s terminology, the ‘internal clues’ offered by the text tend to be read ‘in context’ in order to reconstruct the message (Booth, 1974: 76–86). But what does ‘context’ mean exactly? First of all, it may refer to the specific circumstances in which a text was originally published and circulated: for example, when discussing the aim and message of Otto Mühl’s painting, it is important to bear in mind that many of the figures portrayed in the painting were well-known exponents of a conservative right-wing party; likewise, the meaning of the Palomo Sánchez caricature is clarified by the circumstances under which it originated, namely in the midst of a trade-union dispute. But in addition to its primary meaning, the notion of context also encompasses the genre of a text, i.e. the general conventions and structuring principles guiding the understanding of a given specimen (e.g. the conventions of advertising in the cases of Ernst August von Hannover and Dieter Bohlen); as well as the intertexts that a given text is parodying or dialoguing with (such as the infamous advertising campaign echoed in the Kuliś and Różycki cartoon). Based on the insights coming from Booth and Simpson, as well as on contextual clues in the three senses that have just been presented, the discussion of the corpus texts can therefore be schematically reframed as shown in Table 1.

To be sure, the theory-grounded protocol presented in this section does not aim to become a method for objectively and infallibly determining the meaning of a given humorous text. Both the dialectic and the uptake phases are inevitably characterized by a degree of subjectivity; moreover, certain forms of humor can be extremely ‘unstable’, i.e. they make it particularly difficult to infer a stable meaning behind the humorous surface (Booth, 1974: 240). However, the categories derived from Booth and Simpson can mark a significant step towards a more consistent approach to humor in court, also with regard to Question 2 (reconstruction of aim and message).

The awareness of an inevitable margin of subjectivity lies at the basis of the third question listed at the beginning of this section: to what extent should alternative interpretations (by a reasonable or unreasonable audience) be accounted for? In order to tackle this question, judges need to move from the hermeneutic level (i.e. interpreting the text) to what Liesbeth Korthals Altes has defined as the meta-hermeneutic level, that is the systematic investigation of the interpretive pathways through which readers, viewers or listeners assign meanings and values to a text (Korthals Altes, 2014); in short, meta-hermeneutics is not interested in determining the single correct interpretation of a text, but rather in exploring the different interpretative options that are allowed for by the same text. But how can judges draw the boundaries of the author’s responsibility for the possible interpretations of an impugned text? As mentioned above, in Féret the court maintained that the author should be held accountable for how the material might be interpreted by unreasonable members of the majority, who might end up harming the targeted minority (‘such a discourse is inevitably bound to create in the public, especially among the lesser-equipped, sentiments of contempt, rejection, and hate towards foreigners’, Féret, §73). Others, instead, argue in favor of only assessing the author’s accountability from the perspective of a reasonable majority (as implied in the dissenting opinion to Féret) or from that of ‘a reasonable member of the target group’, i.e. of a vulnerable minority (Kuhn, 2019: 120). The dichotomy between majority and minority groups seems misleading and unnecessary, since judges should identify all potentially harmful interpretations of a given text—regardless of whether they are more likely to originate in any given sector of the audience. As for the hesitation between adopting the perspective of reasonable or unreasonable readers, this is exactly where interdisciplinary dialogue can be of particular help.

A valuable cue in this respect is provided once again by Wayne Booth, and more precisely by his notion of ‘implied reader’—namely the image of the recipient that the author presumably had in mind while writing, as it can be inferred based on textual and contextual indicators (Schmid, 2013 and Booth, 1983). This definition first originated in Booth’s work on narrative fiction at large, but can be particularly relevant in the analysis of humorous texts. In light of Booth’s definition, ‘implied reader’ can mean two different things: on the one hand, it designates the presumed addressee, i.e. the recipient ‘to whom the work is directed and whose linguistic codes, ideological norms, and aesthetic ideas must be taken into account if the work is to be understood’; on the other, it might refer to the ideal recipient, that is the person who ‘understands the work in a way that optimally matches its structure and adopts the interpretive position and aesthetic standpoint put forward by the work’ (Schmid, 2013). In short, the presumed addressee is the general public among which the author can rightfully expect her or his work to be circulated; the ideal recipient, instead, is the image of a reader who understands every nuance and agrees on every point with the author. The former category is obviously broader and more inclusive, and is therefore a safer guide when it comes to tackling the meta-hermeneutic issue of whose interpretations the author should be deemed accountable for. On such grounds, I argue that the question, to what extent should alternative interpretations (by a reasonable or unreasonable audience) be accounted for? should be reformulated as follows: to what extent can the text be interpreted in a harmful way by its presumed addressees? This latter formulation has the advantage of doing away with the shaky distinction between reasonable and unreasonable interpretations, as it focuses on the more feasible task of defining the presumed audience of the impugned work. For example, it is relatively easy to determine that (regardless of their final uptake) the presumed addressees of Otto Mühl’s painting must have been familiar with the political stances of the Austrian Freedom Party, or that those exposed to the Palomo Sánchez cartoon must have been able to connect it to an ongoing trade-union conflict; such considerations can prove instrumental in drawing a line between the possible (mis)interpretations that the author can be considered accountable for, and those that fall beyond the reach of the author’s responsibility.

6. Conclusion

The present study aimed to illustrate how legal scholars and professionals can benefit from a closer dialogue with humor studies; in order to do so, it provides the first systematic discussion of a corpus of humor-related cases from the European Court of Human Rights, with a special focus on cartoons and other comparable forms of predominantly visual humor. This paper has identified a set of 10 criteria used by ECtHR judges when dealing with humor or satire (Section 3), and discussed the specific challenges posed by humor to the implementation of such criteria (Section 4); most notably, the analysis of the corpus highlighted the lack of a shared vocabulary and of a consistent approach to the interpretation of humorous texts. Section 5 showed how interdisciplinary dialogue can contribute to streamlining and rationalizing those interpretive issues: first of all, the identification and classification of humor indicators is facilitated by Booth’s discussion of overt and covert irony markers, as well as by Simpson’s typology of metonymic and metaphoric satirical devices; secondly, after the identification of humor markers, Booth’s and Simpson’s models can also contribute to defining a consistent protocol for the reconstruction of the text’s aim and message; thirdly, Booth’s notion of implied reader (especially in the sense of ‘presumed addressee’) can help judges in defining the spectrum of alternative interpretations that the author of the impugned text can be deemed responsible for.

Needless to say, the findings presented in this paper can be further developed in several directions, by way of extending the focus to other forms of verbal or non-verbal humor, to other theoretical frameworks in addition to Booth and Simpson (e.g. Attardo and Raskin’s ‘General Theory of Verbal Humor’, as suggested in Little, 2011: 104), and to other institutional contexts within and beyond Europe—thereby favoring comparative work and interdisciplinary collaboration with legal scholars focusing on different areas, such the United States (Little, 2019 and Todd, 2016) or Latin America (Capelotti, 2016). For the time being, it is hoped that this preliminary exploration makes a valid case in favor of forensic humor studies as a promising approach to some of the most urgent challenges related to the juridical regulation of freedom of expression.

Competing Interests

The author has no competing interests to declare.