Violence and Discrimination before the Strasbourg Court

Yonko Grozev, Bulgarian Helsinki Committee

There is a distinction between discrimination proper—i.e. differential access to services and employment—and discriminatory use of violence. In this presentation I will address the second: violence motivated or driven by race or ethnicity, and the type of remedies we have to challenge this type of discrimination. Violence is used most often by the state, but is by no means limited to agents of the state. My specific experience has been particularly in challenging widespread and systematic police abuse, targeting primarily Roma, but what I will be saying is by no means limited to Roma only. It is relevant to any form of racially motivated violence. And I will also discuss the question of how to organise litigation work in challenging police violence.

One of the first questions of course is—why target particularly police violence, when Roma are discriminated against in Bulgaria in many other ways—in the job market, in education, in all kinds of municipal services, and in many other aspects of their daily life? One answer is because it is quite difficult to single out one specific event—some government or private action which is the sole source of discrimination—when it comes to education, or municipal services or even employment. Quite often what we face is many layers of smaller or larger injustices that take place over the years and add up to an extremely unjust overall result. That, unfortunately, is hardly the stuff of likely court victories—where specific instances of violence might be.

And second, when faced with a domestic judiciary that was not raised on a daily diet on anti-discrimination law, to put it mildly, one has to include international litigation as part of the long-term strategy. The major international forum we have in Bulgaria is the European Court of Human Rights. The European Convention on Human Rights (ECHR) does not provide good protection against non-violent forms of discrimination. However, it provides very good protection in cases of unlawful violence by law enforcement agents. In interpreting Arts 2 and 3 of the ECHR, the right to life and the right to freedom from torture or cruel and unusual punishment respectively, the Court of Human Rights has established a significant body of case law providing good protection in cases of police brutality.

The sheer extent of the problem with police brutality against Roma in Bulgaria and the gravity of the consequent human rights violations—as well as the good protection afforded by international mechanisms elsewhere—made this a good target for litigation. Still, in the early 90s when human rights activists were setting the priorities back home, it was not so obvious that this was the way to go. Nor was it so simple. It is often hard to get victims to step forward and testify. People who are in police custody are often there for a reason. That makes them vulnerable, and if they complain against the police they fear it will come back against them later on. Also they are often mistrustful of the system and suspect that they will be taking big risks by complaining, without any guarantees that the system will work for them and produce a just result. Some of these difficulties don’t apply in cases of loss of life in custody and as a result we have taken on quite a few of these. And in other cases, people who have not committed a crime but were detained and beaten up are more willing to come forward and complain.

It has also been our experience that it is very important for lawyers doing such cases to work together with human rights groups. First, they are good at motivating individuals to take up their case and bring it to court. Secondly, they are good at digging up cases which will set precedents. The victims of the worst human rights abuses, and these are the best cases for litigation, are usually socially disadvantaged—to the point that they are unlikely to complain or look for a lawyer. Third, human rights groups are there for the long haul. Litigation may take years and years. In many of our cases it has taken up to ten years already, and an organization is much better placed to provide the necessary moral and other support to the victims.

Assenov v. Bulgaria

After having worked through the domestic system on police violence cases for years, generally unsuccessfully, Bulgarian groups started filing complaints with the European Court of Human Rights under Articles 2 and 3. One of the first cases of police violence against Roma was Assenov v. Bulgaria. This concerned a fourteen year-old boy, arrested on the street for illegal gambling. He was taken to his parents by the police, following which an argument ensued between the police and the parents. Then he was taken to the police station. At this point, the stories diverge. According to the boys and his parents, the police beat up the kid with a truncheon—and there is no question that he was beaten up, as there was supporting medical evidence. According to the police, however, during the argument between the parents and the police, the father said in front of the police that he would punish his son himself, and the police claim he beat him with a piece of wood. An investigation ensued, but it lacked vigor and was inconclusive, and was finally closed because of lack of evidence that the police had beaten up the boy.

After the domestic investigation was closed, a complaint was filed with the European Court of Human Rights. And here the main issue was whether there was a violation of Article 3, the prohibition of degrading treatment. The court looked into the evidence supporting a claim of ill-treatment. Whenever a credible claim has been made of police violence while the person is in the hands of the police, the burden of proof shifts to the government to give a plausible account of how harm was sustained. This is a standard that was developed earlier in police brutality cases by the Strasbourg court. In this particular case, the court said, however, that the evidence was not sufficient to prove the alleged beating and held that there was no violation of the substantive aspect of Article 3.

This conclusion is not beyond dispute, given the medical evidence in the case. There was a long list of injuries—including marks clearly inflicted by an object shaped like a truncheon and at places where the police did not claim the father hit his son. I would dare to say that on the same facts today, with the far vaster experience of the court in such cases, even the substantive result might have been different. What was more important however, in that case, was the finding of a violation of Article 3 on a completely different ground. In this case, for the first time, the Court ruled that besides the substantive duty not to treat individuals in violation of Article 3, governments have an additional duty, to investigate thoroughly complaints about police ill-treatment. This is the procedural aspect of Article 3. Such an obligation was at that time already ruled to be part of the right to life, Article 2, but in the Assenov judgment the Court ruled for the first time that also in Art. 3 cases there is a burden on the police to carry out a full investigation. As the investigation that was carried out in Assenov was delayed and far from complete there was a violation of the Convention.

Another question that was asked here was what should we do facing a hostile government that puts pressure on victims not to pursue their rights. This was also an issue in the Assenov case. The police put pressure on the applicant to withdraw his complaints. They also put pressure on Assenov’s parents. At some point the local police even made the parents sign a declaration that they never filed a complaint, and sent the declaration to Strasbourg. As they clearly had filed a complaint with Strasbourg, and the court had their signatures on the complaint, the declaration was taken to be evidence of unlawful pressure on the applicants to withdraw their complaint. The court held this to be a separate violation of the Convention, Article 25. This was quite an important finding, as there have been no cases since where the Bulgarian government has tried to put pressure on applicants to withdraw their applications, where before there had been many such cases.

In this case, the first Bulgarian case of police violence against Roma, there was no complaint of discrimination, and as the court was not asked to rule on Art 14, it did not. The reason why there was no complaint seemed like a good one at the time—the standard of proof under Article 14 is very high. Discrimination has to be proved “beyond reasonable doubt”, and there was little evidence to show a racist motive on the part of the police officer who allegedly beat up Assenov, nor the investigators who later did not properly investigate.

In later cases, however, the strategy was changed and a separate complaint of a violation of Article 14 was made in similar cases. These are Velikova v Bulgaria and Anguelova v Bulgaria.

Velikova v. Bulgaria

Velikova concerned a Roma man who died in custody. He died from blood loss after being severely beaten up. Police did not find a credible reason for his death and made no real effort to investigate what happened. They did not give any explanation. Much later the government argued in front of the European Court that he had stumbled and fallen down the stairs. But this was not credible, as the injuries were so severe, that they could not have been possibly caused in that way. As the standard under Article 2 of the convention is that once a person was detained in good health, the police should provide a plausible explanation of how he or she suffered their injuries, the Bulgarian government clearly did not meet that standard. So, the Art 2 claims were clear violations and this is also what the court said. But what about Art 14?

The evidence was as follows:
1) The police knew the victim was Roma;
2) The following was noted in the investigation report, describing the body of the victim when he was found dead in the police station: “Because of the dark colour of the skin there were no other visible injuries on the body”;
3) There were credible reports by numerous human rights groups of persistent and widespread police violence against Roma.

The fact that the police knew the victim was Roma was beyond doubt. He was arrested in the Roma neighbourhood, close to where he lived, and one of the police officers referred to him as “the Gypsy”. As to the investigation report, well, it is Roma who have dark colour of the skin. What was more important, there were quite a few other injuries on the victim’s body, besides those described by the investigator. So, in our view this note was a clear sign of anti-Roma bias, even worse, such wording effectively sent a signal down the investigation line that “actually this is a Roma and there is no need to be over-zealous with the investigation.” We did not have, however, figures on the overall number of Roma detained, ill-treated in custody, who may or may not have complained and had no thorough investigation if they did. Ideally, we would have had that information and the same information on non-Roma, so that it could be compared and a conclusion drawn that Roma are far more likely than non-Roma to be mistreated while in custody, and consequently to have no investigation. But the police would not keep reliable statistics on police violence and the statistics that are there do not take account of the ethnicity of the victim. So in the absence of such statistics, the best we could come up with were the numerous reports by human rights groups, describing individual cases of gross violation of the rights of Roma in police custody. The sheer number of those cases it was argued reveals a clear pattern, proving discriminatory treatment.

All that evidence, however, was not considered by the court as establishing beyond reasonable doubt that the victim has been discriminated against and the court did not find a violation of Art 14. In the words of the court “[t]he material before it does not enable the Court to conclude beyond reasonable doubt that Mr Tsonchev's death and the lack of a meaningful investigation into it were motivated by racial prejudice, as claimed by the applicant.” Still, it held, that “the applicant's complaint under Article 14 is grounded on a number of serious arguments.”

Anguelova v. Bulgaria

Anguelova concerned a 17-year old Roma boy who also died in police custody. He died of a brain hemorrhage, caused by a skull fracture from a blow to the face. The time of the injury was not established with sufficient precision by the investigation—it could have been before he arrived in police custody or after. Still, the court ruled there was a violation both under the substantive ground of Art 2—that is, the government had failed to provide a plausible explanation as to how the injury took place—and under the procedural ground, namely that the investigation was not thorough and effective.

More interesting from our perspective, however, was the complaint under Article 14. The applicant again cited evidence from human rights groups. Numerous human rights organisations have established patterns of systematic police violence against Roma. There was all this evidence out there and all these cases of violence against Roma—it can’t be accidental, it was argued. But again, the court held that the complaint under Art 14 was not proved beyond reasonable doubt, and the Court did not find a violation. One judge dissented however—Judge Bonello—and his dissent deserves to be quoted at length:

“[T]he Court, in over fifty years of pertinacious judicial scrutiny, has not, to date, found one single instance of violation of the right to life (Article 2) or the right not to be subjected to torture or other degrading or inhuman treatment or punishment (Article 3) induced by the race, colour or place of origin of the victim. […] An uninformed observer would be justified to conclude that, for over fifty years democratic Europe has been exempted from any suspicion of racism, intolerance or xenophobia. The Europe projected by the Court's case-law is that of an exemplary haven of ethnic fraternity, in which peoples of the most diverse origin coalesce without distress, prejudice or recrimination. The present case energises that delusion.

Frequently and regularly the Court acknowledges that members of vulnerable minorities are deprived of life or subjected to appalling treatment in violation of Article 3; but not once has the Court found that this happens to be linked to their ethnicity. Kurds, coloureds, Islamics, Roma and others are again and again killed, tortured or maimed, but the Court is not persuaded that their race, colour, nationality or place of origin has anything to do with it. Misfortunes punctually visit disadvantaged minority groups, but only as the result of well-disposed coincidence.

At the root of this injurious escape from reality lies the evidentiary rule which the Court has inflicted on itself: “The Court recalls that the standard of proof required under the Convention is 'proof beyond reasonable doubt'”. […]

Nowhere does the Convention mandate the ‘proof-beyond-reasonable- doubt’ standard today required of the victim to convince the Court that death or ill-treatment were induced by ethnic prejudice. Article 32, on the contrary, gives the Court the widest possible discretion as to the interpretation and the application of the Convention. What the Convention does mandate is quite the opposite: that its provisions should be given thorough implementation. Any exercise in interpreting the Convention must be geared to “securing the universal and effective recognition and observation” of the guarantees enumerated, unless it is to turn into a betrayal of the spirit and the letter of its momentous preamble.

The Convention has to be applied by the Court in such a way as to guarantee “not rights that are theoretical or illusory, but rights that are practical and effective”[See Artico v. Italy]. No more effective tool could be devised to ensure that the protection against racial discrimination becomes illusory and inoperative than requiring from the victim a standard of proof that, in other civil-law disputes, is required of no one else. […]

So long as the Court persists in requiring in human rights disputes a standard of proof that fifty years experience has shown to be as unreal as it is unrealistic and unrealisable, it will, in effect, only continue to pay lip service to the guarantees it then makes impossible to uphold.

Judge Bonello went on to recommend that the court follow instead the “altogether more reasonable” example of the IACHR and the US Supreme Court, and noted a number of techniques at the court’s disposal, including reversal of the burden of proof in cases of racial discrimination. This was something that the European Roma Rights Center (ERRC) had argued for in an amicus brief in a similar case. So far, this is just one case of dissent, but if we look from a historical perspective, it has been through dissents that basic rights jurisprudence has developed, and that gives dissent a particular importance.

Challenging the standard of proof

On a more technical note, I wanted to emphasize that in none of these cases was a complaint of discrimination made before the domestic investigation authorities. Still, the court took the position that an investigation into the main facts is sufficient for the purposes of exhaustion of domestic remedies and did not reject the claim under Art 14 as inadmissible, but instead considered it on its merits.

The Velikova and Anguelova judgments demonstrate that a challenge of the standard of proof under the convention is possible. As the ERRC had suggested in its amicus curiae there are a number of possible approaches. The “beyond reasonable doubt” standard could be replaced by a lower burden of proof, like a balance of probability. Should it fail to lower the standard for proof, the court could instead adopt a similar standard to that adopted regarding Arts 2 and 3—developing some procedural guarantees. The government will have violated Art 14 if it does not conduct a conclusive investigation into the possibility that discrimination has taken place where there is some evidence supporting a claim of discrimination.

And challenging the existing standard is vital as it is currently extremely difficult to prove discrimination. Good examples are the Kurdish cases against Turkey, where the Court delivers dozens of judgments every year finding violations of Arts 2 and 3 but still no violation of Art 14. Another example would be several cases against the UK involving Irish terrorists, McKerr v UK, Shanaghan v UK, Hugh Jordan v UK and Kelly v UK. These are all cases raising complaints of unlawful use of firearms by the British security forces in Northern Ireland where the victims died as a result. The lawyers presented evidence that the number of Catholics killed by security forces over the years far outnumbers (by a factor of 7 or so) the number of Protestants killed by security forces. Clear evidence of discrimination on the part of the security forces, said the plaintiff. The court, however, rejected this logic, saying that statistical evidence comparing the nationality of individuals against whom firearms are used is not sufficient for a finding of violation of Art 14. In the words of the court:

“[w]here a general policy or measure has disproportionately prejudicial effects on a particular group, it is not excluded that this may be considered as discriminatory notwithstanding that it is not specifically aimed or directed at that group. However, even though statistically it appears that the majority of people shot by the security forces were from the Catholic or nationalist community, the court does not consider that statistics can in themselves disclose a practice which could be classified as discriminatory within the meaning of Article 14. There is no evidence before the court, which would entitle it to conclude that any of those killings, save the four which resulted in convictions, involved the unlawful or excessive use of force by members of the security forces.” (McKerr v UK judgment of 4 May 2001 para 165.)

And one final note on an issue that was brought up earlier, sexual violence against women. Surely domestic remedies would be decisive for a long-term solution. Still, development of domestic remedies might be speeded up by international litigation, which I would argue for. It is possible to bring rape cases to the European Court. One example would be a recent Bulgarian case, MC v Bulgaria, in which the national authorities decided that as there was no hard evidence that the girl physically resisted intercourse, therefore it could not have been rape. The case was considered admissible by the court last December, where violations of Art 3 and 8 are argued.

Q&A

Yonko Grozev's talk generated much interest from conference participants, a number of whom commented on lessons Russia could draw from the experiences presented. Some examples follow.

Q: Running through Yonko’s presentation is an idea that the rulings of the court are fluid. On the day before 19 October 1998, there was no requirement to investigate under Art 3—the day after there was. The European Convention had changed—and it did so because the Bulgarian lawyers structured the case in such a way as to allow the Court to reach this conclusions.

Q: There are numerous criminal cases against Roma going ahead in Russia—but not one of them is going forward fairly and without discrimination. In one case, the pre-trial investigation went to all sorts of trouble to have the case dismissed. And there was no substantive reason—it was a tribunal case. In all sorts of cases directly related to negligence, we find crimes committed by the law enforcement officers of Russia. Law enforcement authorities cover up crimes committed by their subordinates and do everything possible to prevent defense attorneys doing their job. There were ten substantive points in that case. Yet the composition of the court was not such that you could believe there was a fair trial. Defence lawyers are helpless in these situations, especially when it the victims are Roma. This legal outrage shows no signs of changing.

Q: As to there being no investigation—that is now a separate violation under the Convention. Whether the police officer will go to prison as a result of a finding of a violation of the Convention is of course a different question. They surely will not. But a finding of an international court of a violation of international law works in many different ways. It forces governments to change practice and it does force a change of mind and heart.

Q: Strasbourg case law does not seem to have affected practice in Russian courts at all. Also, a great number of cases filed from Russia to the European Court of Human Rights have been refused admissability without any reasons being given.

YG: We had the same problem that the authorities in Bulgaria didn’t seem to pay attention to judgments of the court after the Convention was ratified. The authorities thought judgments against other countries were not something they should worry about. But once a number of judgments against Bulgaria started coming out, they started to pay attention.

Q: The ECHR is not perfect. It is underfunded and has numerous problems. But it is an existing mechanism that has achieved some effective remedy and made some important findings of law. Regarding the time it takes to get a change in law—in the U.S. it took 30 years before the Supreme Court ruled against segregation.

We welcome your comments