EHRM, 21-10-2014, nr. 9540/07

The applicant was born in 1975 and lives in Ankara.

The facts of the case, as submitted by the parties and as they appear from the documents submitted by them, may be summarised as follows.

In the early hours of 28 April 2005 the applicant went to a primary school in the town of Sincan and poured paint on a statue of Atatürk1. which was situated in the school's garden. On the evening of the same day, he poured paint on a statue of Atatürk in the garden of another primary school.

On 6 May 2005 he did the same thing in the same two primary schools.

On 8 July 2005 the applicant poured paint on a statue of Atatürk in Sincan town centre.

On 12 September 2005 the applicant went to the same statue in Sincan town centre equipped with a tin of paint, paint thinner and a ladder. As he was about to open the tin of paint he was arrested by police officers and taken to a police station where he was questioned. In a statement taken from him on the same day the applicant was reported as having told the police officers that he had carried out the above-mentioned actions because he resented Atatürk and had expressed his resentment by pouring paint on the statues.

On the same day the applicant was brought before a prosecutor and then a judge, who ordered his detention on remand pending the opening of criminal proceedings against him. In his statement to the prosecutor the applicant maintained that he had carried out his actions to express his ‘lack of affection’ for Atatürk.

In his indictment of 15 September 2005, lodged with the Sincan Criminal Court of First Instance (hereinafter ‘the trial court’), the Sincan prosecutor charged the applicant with the offence of contravening the Law on Offences Committed Against Atatürk (Law no. 5816; see ‘Relevant Domestic Law and Practice’ below).

In the course of the trial the applicant admitted that he had poured paint on the statues. He told the trial court that he had completed his university studies and qualified as a teacher. However, he had been unemployed for a long time because his application to work as a teacher had not been accepted by the Ministry of Education. He had carried out his offences in order to protest against the Ministry's decision.

On 10 October 2005 the trial court found the applicant guilty as charged. Having regard to the fact that the offence was committed in a public place and on a number of occasions, the trial court sentenced him to three years' imprisonment instead of the minimum term of imprisonment applicable under Law no. 5816, which is one year. The fact that the offence had been committed in a public place also led the trial court to increase the sentence by half in accordance with section 2 of Law no. 5816. The trial court also considered that the applicant had committed the offence on five separate occasions, and decided to multiply the sentence by five. The applicant was thus sentenced to a total prison term of twenty-two years and six months for his above-mentioned actions.

The applicant appealed. In his appeal he argued that, according to the provisions of the Criminal Code, only one sentence should have been imposed on him because, regardless of the fact that he had poured paint on the statues on five occasions, he had in fact only committed one offence and not multiple offences. In support of his argument, he submitted that his five actions had been carried out within a short span of time.

The applicant also pointed out that, instead of imposing on him the minimum one-year prison sentence provided for in Law no. 5816 in respect of each offence, the trial court had handed down a three-year sentence because it had had regard to the number of times he had poured paint on the statues. The trial court had then gone on to rely on the frequency of his actions when multiplying the sentence by five.

The applicant also challenged the trial court's reliance on section 2 of Law no. 5816 when increasing his sentence by half because the offence had been committed in a public place. He drew the Court of Cassation's attention to the fact that, by their nature, statues are placed in public places.

The applicant added that he had carried out his actions in order to express his ‘lack of affection’ for Atatürk. As such, he had remained within the boundaries of his right to freedom of expression, which was guaranteed by Article 10 of the Convention. Thus, although it would have been reasonable to prosecute and punish him for damaging property, he had in fact been punished for expressing his opinions.

On 6 April 2006 the Court of Cassation rejected the applicant's argument that he had been expressing his opinion, but quashed the trial court's judgment on the ground of, inter alia, that court's failure to give adequate consideration to the possibility that the five separate incidents could form only one offence and not multiple offences. The Court of Cassation considered that the applicant had carried out his actions in order to protest against the Ministry of Education's decision not to appoint him as a teacher. The case file was sent back to the trial court.

In its decision of 5 July 2006 the trial court agreed with the Court of Cassation's conclusion, and held that the applicant's actions had amounted to a single offence and not five offences. However, having regard, inter alia, to the ‘contradictory reasons’ put forward by the applicant as justification for his actions, as well as ‘the effects of his actions on the public’, the trial court concluded that the applicant's actions had amounted to ‘insults’, and deemed it fit to sentence him to five years' imprisonment, which is the maximum allowed under Law no. 5816. The sentence was then increased by half because the acts had been committed in a public place. Furthermore, pursuant to Article 43 of the Criminal Code (see ‘Relevant Domestic Law and Practice’ below), the sentence was further increased by three quarters. The applicant was thus sentenced to a total of thirteen years, one month and fifteen days' imprisonment.

Furthermore, in its decision the trial court set out the restrictions under section 53 of the Criminal Code which were to be placed on the applicant on account of his conviction. Accordingly, until the execution of his sentence, the applicant was banned from, among other things, voting and taking part in elections, as well as from running associations, parties, trade unions and cooperatives (see ‘Relevant Domestic Law and Practice’).

The applicant appealed and repeated his arguments under various provisions of the Convention. He maintained, in particular, that he had carried out his actions in order to express his ‘lack of affection’ for Atatürk and had thus exercised his freedom of expression guaranteed in Article 10 of the Convention.

The appeal was dismissed by the Court of Cassation on 5 February 2007. No mention was made in the Court of Cassation's decision of the arguments raised by the applicant about his freedom of expression.

According to a document drawn up by the prosecutor on 16 April 2007 setting out the details of the applicant's prison sentence, the date of the applicant's release from prison was set as 22 October 2018, with a possibility of release on 7 June 2014 for good behaviour.

In the meantime, on 1 June 2005 the Law on the Execution of Prison Sentences and Other Security Measures (Law no. 5275) entered into force. This law sets out the circumstances in which prisoners can benefit from early release.

On 15 May 2007 the prosecutor responsible for the prison the applicant was serving his sentence in wrote to the trial court and asked for guidance in calculating the date of the applicant's possible early release. The prosecutor stated that, for offences committed before 1 June 2005, Law no. 647 was applicable and, for offences committed after that date, the new Law no. 5275 would be applicable. The applicant had carried out his actions both before and after that date.

On 16 May 2007 the trial court considered that the critical date was the date of the commission of the final act and thus the new law was applicable.

The applicant lodged an objection against that decision and argued that most of his actions had been carried out before 1 June 2005 and that therefore, when calculating his prison sentence, the old law should be taken into account. If his prison sentence were calculated in accordance with the new law, he would spend four more years in prison. That objection was rejected by the trial court on 18 June 2007 and the date of the applicant's possible release from prison was calculated in accordance with the document drawn up by the prosecutor on 16 April 2007 (see paragraph 24 above).

A request made by the applicant to the Ministry of Justice for his conviction to be quashed and another request to the Court of Cassation to rectify the judgment were rejected on 28 September 2007 and 28 December 2007 respectively.

On 11 June 2013 the applicant was released conditionally.

The Law on Offences Committed Against Atatürk (Law no. 5816, entry into force 31 July 1951) provides as follows:

Section 43 of the Criminal Code (Law no. 5237 of 2004), in so far as relevant, provides as follows:

The relevant provisions of section 53 of the Criminal Code (Law no. 5237 of 2004) provide as follows:

For more information concerning the legislation applicable to the issue of voting in Turkey, see Söyler v. Turkey (no. 29411/07, §§ 12–19, 17 September 2013).

A description of the relevant international materials and comparative law on the issue of voting can be found in Scoppola v. Italy (no. 3) [GC] (no. 126/05, §§ 40–60, 22 May 2012).

Relying on Article 10 of the Convention, the applicant complained that he had been punished for having expressed his opinions. He added that the punishment imposed on him had been excessive, disproportionate to the offence in question, and incompatible with Articles 17 and 18 of the Convention.

The Government contested the applicant's arguments.

The Court deems it appropriate to examine the complaint solely from the standpoint of Article 10 of the Convention, which reads as follows:

Relying on Article 3 of Protocol No. 1 to the Convention the applicant complained about the ban which had been imposed on him by the domestic courts and which prevents him from voting. Article 3 of Protocol No. 1 to the Convention reads as follows:

The Government contested that argument.

The applicant complained that, by imposing on him the maximum prison sentence applicable under domestic law and calculating his prison sentence on the basis of a new law (Law no. 5275), his rights under Articles 5, 6 and 7 of the Convention had been breached. The applicant further complained that Law no. 5816 was incompatible with Article 14 of the Convention because it gives the judge too wide a discretion to choose a prison sentence of between one year and five years. As a result, different courts handed down different sentences for the same offence. Finally, relying on Article 11 of the Convention, the applicant complained about the ban which was imposed on him by the domestic courts and which prevented him not only from voting and taking part in elections, but also from running associations, parties, trade unions and cooperatives.

Having regard to its conclusions under Article 10 of the Convention and Article 3 of Protocol No. 1 (see paragraphs 68 and 80 above), the Court considers it unnecessary to examine the admissibility and merits of these complaints.

Article 41 of the Convention provides:

What are the problems with a finding of a violation based on the excessive nature of the punishment? First, this Court, of all courts, cannot rely on a crude sense of justice (though all judicial decisions rendered in disregard of the sense of justice are open to criticism). This Court is concerned with the legitimacy of restrictions on human rights under the Convention and not with the appropriateness of sanctions measured on some mysterious scale. The Convention contains no prohibition on unusual punishment and we are not called upon to evaluate sentencing.

When judges and laymen talk about disproportionate punishment, they often compare the punishment imposed for a given crime with the punishment of another crime, or with the punishment of another person for a similar, comparable crime, or even with the moral seriousness of the crime in relation to the punishment4..

In the present case there is no specific reason given as to why the punishment is grossly disproportionate. Where judicial intuition determines that a matter does not deserve further clarification, those who are not privy to the intuition remain puzzled. Would one year be acceptable, for example, because the statue had to be cleaned or repaired? The Court does not even provide a comparable reference, a tertium comparationis; for example, the fact that thirteen years is a sentence that is ordinarily imposed on murderers. Under that reasoning, the present conviction treats the attack on memory as if were an attack on human life, thus attributing equal weight to life and to the honouring of a deceased person's memory (where the comparator is harm to individuals or harm to the community).

Because the dictates of the sense of justice are satisfied and the talismanic word ‘disproportionate’ is used, the judgment of the Court looks satisfactory. It is not. I share the feelings of my colleagues as to the gross inappropriateness of the sentence, but in an Article 10 case this is not the gist of the rights protection: the Court should look into the necessity of the interference in the light of its impact on the expression concerned.

The text of the Convention requires the Government to prove that an interference was necessary in a democratic society, and it is in the context of such necessity that the question of proportionality arises. The real issue in this case is not that an excessively severe punishment was imposed for an expressive act that did not cause serious damage, but that a whole class of expression (insults to Atatürk's memory) and related expressive acts are considered to be a crime for their content. The law that was applied singles out very specific content: all speech (including expressive action, as in the present case) that publicly insults the memory of Atatürk is punishable. The issue is not the protection of all public statues where harm to the statue has been caused by an expressive action. The issue, which is buried under the outrage of the excessive sentence, is the singling out of specific speech content for punishment. Law no. 5816 provides first and foremost that any ‘disrespect for Atatürk's memory’ is to be punished by a prison sentence of between one and three years, the use of paint on a monument (‘dirtying of a statue’) raising the sentence to five years; the applicant was then given an additional eight years of punishment for the aggravating circumstances.

Of course, eight additional years for degrading a statue is excessive in view of the degree of harm caused by the act, but this Court is ‘only’ called upon to see whether a limitation of freedom of expression is necessary in a democratic society.

I would argue that the problem can be better decided using a category-based analysis of the legislation, and even by an enhanced proportionality analysis of the means/end relationship of the legislation and the objective value of the intended aim, as is carried out, for example, in Canada and Germany. These approaches are superior to the Court's ‘standard’, often narrowly case-related analysis because they are more convincing and, above all, offer a better, broader, and more equivalent protection to free speech against governmental abuse.

The legislature's predominant concerns in Law no. 5816 are with the content of the speech as opposed to its secondary effects; it expresses the legislature's disagreement with the message the act conveys. In the category-based approach of the United States First Amendment law, known as the ‘categorical approach’5., this is plainly unconstitutional. So what is wrong with content discrimination? It is wrong because the Government disregard content-neutrality without compelling reasons. The requirement of content neutrality follows from the assumption that content-based restrictions (‘content-discrimination’) target specific messages, thus resulting in thought control, and ‘[such a restriction] raises the specter that the government may effectively drive certain ideas or viewpoints from the marketplace.6.’

The judgment operates within the straightjacket of the proportionality analysis; it is for this reason that the Court fails to make explicit the underlying (‘structural’) problem of Law no. 5816. I am aware of the advantages of the three mechanical prongs of the ‘standard’ proportionality analysis. They offer considerable legal certainty; the approach also offers the advantages of economies of scale. This kind of manufacturing certainty is understandably attractive where a court has thousands of cases and where a court is called upon to give advice to judges reading our judgments in forty-seven different member States.

However, even within the proportionality analysis there are other methods, slightly more complex in nature than the three-pronged approach used by the Court. One may add other levels of scrutiny.

Among others, when determining a measure's quality as a means to reach a (legitimate) end, the search must begin at the abstract level of the legislation. This search is particularly demanding (and therefore efficient) if and when a court enters into a substantive analysis of the veracity of the allegation that a regulatory measure actually serves a purported end. Moreover, the importance of the end itself may be subject to judicial analysis. Using this approach in the Articles 8–11 context, the Court would have to review how important and genuine the references are to one or another aim recognised in the Convention as a ground for restricting a Convention right. Is the end genuine? Or instead, is it a bluff couched in terms of public interest that pretends to be beyond the reach of judicial scrutiny in the name of democratic legitimation of the legislature?

Moreover, is the chosen means narrowly tailored? Is it not the case that the criminal provision is over-broad, even considering the need for sensitivity protection?

Where, as in the present case, the argument is made that the sensitivities and deep feelings of a population are to be protected, a court could and should take a long look at the relationship of this allegation to the ‘rights of others’. To accept that all interests ‘amount to rights of others’ and claim that all these alleged rights are of equal weight to that of Convention human rights is extremely dangerous for human rights: not all rights are created as equal. Is there a right to have one's feelings and deeply held convictions left undisturbed? Are feelings to be protected from potential inconvenience as a matter of right? Further, even assuming that all alleged interests constitute rights (a position that I find untenable), is this alleged right per se sufficient to justify certain forms of Convention-rights restriction (especially blanket bans, which used to be highly suspect even for the Court, at least until very recently, in the freedom of expression context)? This same analysis may also be appropriate when addressing the specific circumstances of the case at a later stage of the analysis; something that is often done in the form of balancing, as if Convention rights and other interests were of equal importance!

It may well be that certain measures simply do not serve the purported end or at least that they are not the least restrictive possible. One should ask the question: is mandatory imprisonment the only available means to protect political memory?

Of course, even if in the abstract the rights-restrictive means are acceptable and rationally connected to the legitimate and genuine end, their application in the specific context (the conduct of the applicant) may be disproportionate, because there are lesser rights-restrictive means to achieve the end in the circumstances of the case. In other instances it can be said (sometimes using the language of balancing) that the restriction on a right as a means to an end is excessive because it undermines the very right which one values more than the end. It should be added, in this logic, that Convention human rights are of a specific value (being singled out as superior values in an international convention).

Going beyond the above-mentioned, more demanding forms of scrutiny within the proportionality methodology, freedom of expression cases are sometimes (even regularly in the United States) resolved using a categorical approach7.. In principle, such an approach guarantees freedom of expression unequivocally and with more certainty than a case-by-case analysis, where the metrics of proportionality and balancing are not spelled out. The uncertainty that is inherent in the case-based proportionality analysis invites authorities to attempt to impose further restrictions. More importantly, it discourages speakers.

A court of human rights must go to the heart of this matter. In Turkey it is possible to imprison someone for an offence against the memory of Atatürk. I have no doubts that the Turkish nation has strong feelings of respect towards the founder of the modern Turkish State, and it is within the constitutional powers of the Turkish nation to express such feelings. I have full respect for these sentiments, but equally strong reservations as to the legal enforcement of sensitivities in matters of speech8.. I understand that the form of the expression is problematic here but, as the judgment demonstrates, it falls within expressive conduct; the pouring of paint is a form of expression, disputable though it may be9.. Destruction caused to a statue or other piece of art is an ordinary crime; to destroy Michelangelo's ‘Pieta’ would indeed be a serious crime. But in the present case it was the expressed content that was the ground for the conviction: the object of the crime is clearly ‘the memory of Atatürk’ and not the alleged vandalism, which of course might otherwise be subject to criminal sanctions. Moreover, I can envision the need for such a dramatic form of expression of political discontent in certain circumstances, a matter that did not have to be addressed in the present case. The Turkish courts never entered into a discussion of the appropriateness of the expressive act. In any event, all forms of expression of dislike of Atatürk and his memory, all the underlying discontent with the political system created by Atatürk and based on his political vision, are prohibited: this is the primary and fundamental issue.

I can envision situations where punishment for a similar offence is appropriate or even necessary in a democratic society, where insult to memory amounts to a call to violence or hatred against identifiable individuals, but that element is not required by the present law and no such danger is present in this case. It is the mere fact of the insult that is criminalised.

The limited analysis, resulting from the standard proportionality test, precludes the consideration of the law's impact on all speech acts. It is for this reason that the Court did not have the opportunity to look into the real problem. However, the Convention and even our own methodology calls us to consider the impact of the restriction on freedom of expression. ‘It is recalled that there is little scope under Article 10 § 2 for restrictions on debates on questions of public interest.10.’ The Court has always accepted that ‘there is little scope under Article 10 § 2 of the Convention for restrictions on political speech or on debate on questions of public interest.11.’ The expressive act of the applicant, being political speech, should have triggered strict scrutiny, and the Government certainly failed to provide justification based on compelling reasons why they had to criminalise insults to memory. Given that the law is content-discriminatory, we do not have to look into the effects of a content-neutral law such as the criminalisation of the destruction of statues.

Where disrespect for the memory of a political figure is punished, this has a chilling effect on all speakers. The State has not shown any compelling interest for this restriction. I cannot see the reasonable purpose of such a measure in a democratic society, given that no democratic society can exist without free expression on political matters12.. Even assuming that the deep feelings of the Turkish people will be hurt at the sight of the paint on the statue or on hearing disrespectful words, I cannot see how this can be a sufficient justification in a democratic society, where even disturbing political opinions are to be accepted.

This fundamental consideration is grievously absent in Turkish law when the mandatory sanction is one year in prison, let alone the thirteen years imposed on applicant. A law which enables, and even mandates, such interference is incompatible with the necessities of a democratic society. This Court should not shy away from considering the impermissibility of the alleged purpose of legislation that seemingly fits into one of the (over) broad categories of permissible restriction (‘rights of others’)13..

Given the chilling effect of the sanction in Law no. 5816, I would have used a categorical approach: the criminal law is never appropriate as a means to protect other people's political sensitivity, where the disrespect caused to a political figure does not amount to an actual (true) threat or call to violence. Such laws are simply not necessary in a democratic society (outside emergencies), being contrary to the fundamental assumptions of such a society based on free debate and exchange of ideas. The mere existence of content-prohibiting laws endangers and sometimes kills freedom of thought. It is fundamental for a democratic society that its citizens be treated as adults who accept, or learn to tolerate, even speech that they find offensive. This is the price to be paid for a free and democratic society.

A rather similar speech-protective result could have been achieved even within an enhanced proportionality analysis: the end, namely the protection of the alleged right of others, is such that it does not necessitate a prison sentence — not just in the present circumstances of a thirteen-year term, but also in general. In a proportionality analysis that looks first at the very law that is the source of an interference, one looks at the law as a means chosen and at the end served (the protection of alleged feelings). The means are excessive here in the light of the end, among other things because the end itself is problematic; the end in itself is simply not worth the inevitable sacrifice of freedom of expression resulting from the means chosen, but also from any less radical means. Alternatively, the present end is not legitimate; or, to the extent it might be legitimate for some, the means chosen are certainly not the least restrictive possible.

Following the ‘standard’ methodology I have signed on to many judgments where the severity of punishment was held to be an important or the decisive element of the disproportionality finding. The underlying message in those cases was clear: it is inappropriate in a democratic and free society at the level of civility and ‘civilisation’ that Europe hopes to have achieved to use sanctions, especially criminal sanctions, for thought crime (and criminal sanctions in cases of reputational harm)14.. But in those cases the Court did not find it appropriate to make express statements in this sense, probably as a result of its putative role related to Article 27 § 1 and Article 34 of the Convention, although pursuant to Article 19 the Court is called upon to ensure the observance of the engagements undertaken by the Parties; ‘engagements’ that are of a general and structural nature. The Law at issue constitutes a blanket ban on the expression of specific political content for the sake of public sensitivities elevated to the status of a ‘right’. In view of these engagements, content discrimination for the sake of the protection of the memory of a national hero by criminal law is incompatible with the Convention. In the present circumstances of extreme harshness, which will inevitably be repeated, this has to be made clear.

II.

The present judgment provides just satisfaction for the non-pecuniary damage suffered by the applicant. This is proportionate in the sense that it falls within the range of satisfaction provided in other similarly grave freedom of expression and disenfranchisement cases. (One may have doubts that such an amount is equitable in view of the seven years of unmerited suffering in prison). I accept that the amount follows our practice. But with all due respect, I cannot agree with my colleagues as regards pecuniary (material) damage, even if denial of an award on this ground is not uncommon in comparable cases. The applicant certainly suffered material damage (loss of income) because of his incarceration: there is a causal link with a loss of income. This loss is hard to quantify, but technical difficulties of calculation cannot negate the existence of a loss: the applicant was a qualified teacher, albeit unemployed before his conviction, who would have earned a living like any average person in his situation, had he not been incarcerated in violation of the Convention. The loss is thus quantifiable, either on the basis of the average income of a teacher in his position, or at least with regard to the minimum income of an employed person (using the unfair assumption that he could not have found a position in education). Moreover, because of the conviction, he will not be able to work again as a civil servant (it is even unlikely that, having been released on licence, he will find a position as a teacher in private education). To determine the loss of future income is not rocket science and courts do use estimates in such circumstances, taking life expectancy into consideration. I have had the opportunity to express my reservations regarding the Court's parsimonious approach in matters of pecuniary damage, concerned as it is with the risk of ‘speculative’ awards. The ‘gross injustice’ suffered by the applicant in the present case forces me to reach the sad conclusion that the Court has departed from those standards of remedy that national courts and international law find to be a matter of course; and a matter of reason15..

Finally, the Court should have applied the Gençel16. clause: the case should be reopened and the continuing effects of the applicant's conviction, in particular his release on licence, must be remedied.