Is it all part of the game? Victim differentiation and the normative protection of victims of online antagonism under the European Convention on Human Rights

The current debate on online antagonism – a snapshot from Sweden

The beginning of the general public debate in Sweden on different kinds of online antagonism (in Swedish generally called ‘Näthat’ ‘[tr. Online hate]) began in earnest in 2013 with a TV documentary by a Swedish public service channel highlighting how public media personalities and private individuals alike, primarily women, were bullied, harassed and threatened online for expressing their opinion, or simply doing their job (Chukri, 2013; Johansson, 2013). Through the documentary, the scope and scale of the issue became apparent, along with the failures of the legal system to handle complaints from victims of these types of crimes (Carp, 2013; Thurfjäll, 2013). This was followed by the commercial TV show ‘Trolljägarna’ [tr. ‘The Troll Hunters’] where individuals behind online antagonism were tracked down and confronted (Friman, 2014).

Meanwhile, online antagonism continues to claim victims in all social strata, and has led to both politicians and journalists leaving their jobs as the threats and hate aimed at them escalated (Bergström, 2013; TT Nyhetsbyrån, 2015). The scope of the problem does, however, not only concern adults or public officials, and a recent Swedish study has suggested that one third of all children aged 10–16 have been the victim of online harassment or bullying (Friends, 2015).

The response from the Swedish legal community on online antagonism and its victimisation has been varied. It has been suggested that the response to this sort of behaviour should be found in private procedures and torts (Schultz, 2013). Various forms of ‘online hate insurances’ have also arrived on the market, claiming to assist victims in handling the legal aspects of asserting their rights when subjected to online antagonism (Andersson, 2014). The recourse to civil law procedures (tort law, etc.), as well as the offered insurance, both focus on measures to be taken by the victims themselves. Such a focus placed on the victim and his or her own responsibility to deal with harassment and threats online may overshadow the normative responsibility of public authorities to – under the ECHR – prevent, investigate and prosecute this type of behaviour and thus protect the victims of online crime through the criminal justice system (Enarsson and Naarttijärvi, 2015).

In Sweden it has also proved difficult to obtain information and preserve evidence that may be traced to an identifiable person who violates others on the internet (Brottsförebyggande rådet, 2015: 6), and this is particularly problematic in relation to anonymous threats and insults on the internet. However, legislative changes have been made to develop stronger protection for those exposed to violations on the internet. Police and prosecutors can now gain access to subscription data from internet operators and telecommunication carriers in a broader range of cases when dealing with anonymous users online suspected of crime, which can enable and improve investigations of online antagonism (Government bill, 2012; The Electronic Communications Act (2003: 389), Chapter 6, Section 20).

Still, discussion on how to handle online antagonism is on-going in Sweden, and currently the Swedish government has set official inquiries in motion to determine whether or not current legislation is sufficient to handle these crimes (Ministry of Justice, 2014a, 2014b). Partly the inquiries focus on already existing legislation that was introduced with crimes committed online in mind, but that are still not working efficiently, namely the rules on contact with a child for sexual purposes (The Electronic Communications Act (2003: 389) Chapter 6, Section 20). This is when adults seek out and groom (or harass and threaten) children, online or offline, with intent to sexually abuse them. It has been questioned if the legislation targets the intended situations well enough after reports of very few cases being investigated and prosecuted in relation to those reported (Brottsförebyggande rådet, 2013: 14). This has led to an inquiry that is studying how to better handle such crimes (Ministry of Justice, 2014b).

However, one of the official inquiries set in motion targets much broader legislative issues and areas. Many types of victimisation and crimes that can be considered as online antagonism in Sweden today, such as defamation, threats or harassment, and their equivalent in the Swedish penal code, are being subjected to a general overview. This has been actualised with new technology in mind – which has provided new arenas for committing crime. One core issue to be investigated is if current Swedish legislation allows for a reasonable balance between the right to privacy and protection of private life for victims on the one hand, and freedom of expression on the other (Ministry of Justice, 2014a). A Government Bill on the matter is to be presented in 2016.

The implications for victims

The Swedish line of reasoning and the quest for balance between these two fundamental rights – privacy and freedom of expression – highlight an important core issue to handle within the ECtHR and the signatory states, which will be shown in the following analysis.

However, this balance is not only important from a legal perspective, but also from a victimological perspective. Finding this balance is what will enable efficient legislation that allows for individuals to have open discussions without violations online, which could surely benefit all who wish to partake in such discussion. The failure to provide a hate-free public arena may silence individuals, as Keats Citron (2010: 31) points out:

Not surprisingly, the abuse has a profound impact upon targeted individuals. It may intimidate them, chasing them offline. It may convince them to disguise their online identities. When individuals go offline or assume pseudonyms to avoid bigoted cyber attacks, they miss innumerable economic and social opportunities.

The impact of online antagonism can therefore affect not only many aspects of individuals’ lives, but also society as a whole. The effect of causing people to retreat from social media, from taking part in public debate or interacting with people online may inhibit free speech and democratic debate. To ensure that victims who turn to the legal system for help and redress are met with satisfying competence and help to counter such effects, a well-functioning legal process is necessary, as well as access to such a process.

As will be shown, the Convention provides not only normative guidelines for what legal protection must be provided for victims, but also for what kind of speech must be allowed, no matter how this is perceived by the person targeted. Thus, a more in-depth study of the ECHR and the framework it establishes for the legislation of signatory states regarding online antagonism is important, since it can have a large practical impact on the individual victim and his or her access to legal remedies when faced with online antagonism.

The concept of online antagonism under the Convention

Online antagonism may trigger a complex array of considerations relating to the ECHR. While different types of online harassment, threats or defamation can generally be seen as violations of the private life of targeted individuals, those same antagonistic actions may also be seen as expressions, although perhaps objectionable and potentially criminal ones. As will be shown, signatory states to the ECHR may consequently be held responsible under the Convention for actions taken by the state to counter online antagonism, as well as for failure to take action to counter it. The Convention thus has a Janus-faced relation to victims as it may provide both a minimum level of protection as well as – in relation to certain expressions – a maximum level of interference allowed with such expressions. Also, online antagonism cannot be identified as a coherent concept in the case law of the ECtHR as such. Instead, issues relating to online antagonism are treated within the framework of different articles of the Convention depending on the right that is primarily activated by the factual circumstances and the actions (or lack thereof) of the state in question. Also, to investigate ‘online’ issues one cannot necessarily separate them normatively from ‘offline’ issues, as the principles established by the Court in offline instances are, mutatis mutandis, applied online as well. Consequently, the case law relating to online antagonism must in many instances be found in cases dealing with offline situations. However, as the case law of the ECtHR is steadily evolving, the unique challenges of online situations are likely to occupy a larger swath of cases in front of the court, potentially affecting the balance between the two interests.

Antagonism and freedom of expression under article 10

As mentioned briefly above, online antagonism, as we have defined it here, generally and legally speaking consists of different forms of expression that victimise or otherwise target people in a negative way. This implies that expressions and statements that may be labelled as online antagonism may also in certain circumstances be protected under article 10 of the ECHR. In trying to counter online antagonism, governments must thus be wary of restricting the freedom of expression in a way that runs foul of their obligations under the Convention. Protection under article 10 is, however, not without exceptions and in practice the more severe forms of online antagonism may, as will be shown, be exempt from protection, or be of a nature that may legitimately be subject to restrictions or penalties. Article 10-2 of the Convention provides a possibility for states to apply formalities, conditions, restrictions or penalties provided that they are prescribed by law and necessary in a democratic society, in the interests of, inter alia, preventing crime or the protection of the reputation or rights of others.

Although article 10-2 shows that restrictions on freedom of expression are, in principle, possible, it is worth noting that simply being of an offensive nature does not preclude protection for an expression. In an early landmark case the ECtHR has held that article 10 is applicable:

…not only to ‘information’ or ‘ideas’ that are favourably received or regarded as inoffensive or as a matter of indifference, but also to those that offend, shock or disturb the State or any sector of the population. Such are the demands of that pluralism, tolerance and broadmindedness without which there is no ‘democratic society’. This means, amongst other things, that every ‘formality’, ‘condition’, ‘restriction’ or ‘penalty’ imposed in this sphere must be proportionate to the legitimate aim pursued ( Handyside v. United Kingdom, 1976: § 49).

The highly contextual nature of protection under article 10 will therefore also affect what protection different victims of online antagonism can expect under certain circumstances. Freedom of expression has generally been seen as wider by the ECtHR when taking place within the context of a journalistic enterprise or a political context, in order to protect the foundations of an open and democratic debate (Grabenwarter, 2014: 266; Rainey et al., 2014: 438). Consequently, political figures may be expected to endure harsher expressions, as the court has underlined in several cases ( Lindon, Otchakovski-Laurens and July v. France, 2007: § 46; Lingens v. Austria, 1986):

Furthermore, the limits of acceptable criticism are wider as regards a politician as such than as regards a private individual. Unlike the latter, the former inevitably and knowingly lays himself open to close scrutiny of his every word and deed by both journalists and the public at large, and he must consequently display a greater degree of tolerance ( Lindon, Otchakovski-Laurens and July v. France, 2007: § 46).

Consequently protection of statements aimed at politicians, and to some extent civil servants, is higher – although not absolute. To illustrate this, the Court has held that describing the leader of a far-right party as an ‘idiot’ in a journalistic context, and as a written reaction to polemical statements of the politician in question, was protected as it was considered a value judgement aimed at a public figure ( Oberschlick v. Austria (No.2), 1997). A description of another far-right political leader as ‘the chief of a gang of killers’ and a ‘vampire’ who ‘advocated the murder of an immigrant’, was held as going too far as it included allegations of fact and was aimed at stirring up violence and hatred, even though the statements were expressed within the context of a political struggle ( Lindon, Otchakovski-Laurens and July v. France, 2007).

Importantly for certain victim categories, the Court has emphasised the need for states to be able to intervene against hate speech and other expressions intended to breed contempt on account of, inter alia, race, colour, religion or sexual orientation – expressions that may, on account of their nature, be exempt from the protections of article 10 (see Gündüz v. Turkey, 2003; Jersild v. Denmark, 1994; Lehideux and Isorni v. France, 1998; Rainey et al., 2014: 440–441). On account of this disqualification of certain statements from protection under article 10, the resulting protection for victims will be higher, as such protection does not have to be tempered by balancing against opposing interest to the same extent.

As mentioned, several of the leading cases relating to article 10 do not, specifically, relate to the online sphere. The negative consequences for victims in this environment in relation to the spreading of statements and the possibility of anonymity have thus rarely been at the core of the case law. However, such factors have proved decisive in the recent case of Delfi AS v. Estonia. In this case the Court underscored these risks and found that (minor) economic penalties against an online news portal for failure to erase hateful expressions against public officials in the comment section of an article was acceptable under article 10, considering the anonymity of the perpetrators and the fact that the comments were exempt from protection under article 10 due to their hateful content ( Delfi AS v. Estonia, 2015). Also of importance was the Court’s insistence that the rights to privacy and freedom of expression should have equal importance ( Delfi AS v. Estonia, 2015: § 139).

This brief overview of Convention law relating to article 10 shows that, in principle, individuals expressing online antagonism may be subject to sanctions by states, provided such sanctions has been properly balanced against freedom of expression. The issue of the legitimacy of sanctions relating to article 10 must, however, be supplemented by investigating the potential obligations states may have to sanction such expressions under article 8.

Positive obligations in the context of article 8 and online antagonism

In terms of victims’ needs, the practical impact of the ECHR is in many ways seen most clearly through the positive obligations it entails. As noted above, the protection of privacy under article 8 is of fundamental importance for victims as it encompasses many different normative protective layers for both the physical and mental integrity of the individual. Positive obligations under the ECHR may in this context have an even greater impact in practice, as it provides an institutional layer of protection against the actions of other individuals. The concept of positive obligations has been developing steadily in the case law of the Court and must now be regarded as a general doctrine of Convention law (Mowbray, 2004; Xenos, 2012). It relates to the steps the state is expected to take to ensure the protection of Convention rights even in situations arising between individuals. Positive obligations may, for example, entail the responsibility of the state for criminalising and prosecuting violations of Convention rights on the horizontal level, but also a responsibility to ensure that there is a legislative and institutional framework in place to actually investigate violations (see Dodov v. Bulgaria, 2008; Kiliç v. Turkey, 2000). Positive obligations for signatory states are actualised in regard to article 8 of the Convention in particular. The nature of article 8 and the ‘right to respect for private and family life’ invites a wide interpretation of the responsibilities of the state. Also, the right to private life, including the development of personality and self-expression, and the conditions necessary for this development, are core aspects of the Convention when taken as a whole (Xenos, 2012). As mentioned above, positive obligations primarily arise through the requirement of states to secure protection of private and family life, home and correspondence through law. The ECtHR has, however, described the extent of this responsibility as unclear and subject to clarification on a case-by-case basis ( Abdulaziz, Cabales and Balkandali v. United Kingdom, 1985; Christine Goodwin v. United Kingdom, 2002). The application of positive obligations entails not only a normative balancing of the interests of the individual and the state, but also a consideration of available resources and necessary prioritisation, ensuring that the burden placed on the state is not out of proportion or impossible to live up to ( Appelby and others v. United Kingdom, 2003; Cossey v. United Kingdom, 1990; Özgür Gündem v. Turkey, 2000). As such, the so-called margin of appreciation of the state may therefore be deemed as wider in relation to positive obligations ( Christine Goodwin v. United Kingdom, 2002). In practice, the extent of these obligations will be dependent on factors such as if fundamental or essential aspects of private life are at stake, if the effect of a difference between social reality and the law becomes significant to the individual, or if the administrative practices of the state are coherent or not ( A, B and C v. Ireland, 2010).

Positive obligations have important implications for access to justice for victims. Even though the responsibilities of the state can sometimes be fulfilled by ensuring that civil law procedures are available for individuals to seek action against other individuals for violations of rights, the responsibility may at other times require that the criminal law measures are made available to ensure protection of private life. In the case of X and Y v. the Netherlands the Court found that a gap in Dutch law that prevented the rape of a mentally disabled 16 year old girl from being subject to criminal sanctions, leaving only civil law procedures as an available recourse, did not meet the requirements under the Convention:

This is a case where fundamental values and essential aspects of private life are at stake. Effective deterrence is indispensable in this area and it can be achieved only by criminal-law provisions; indeed, it is by such provisions that the matter is normally regulated ( X and Y v. the Netherlands, 1985: § 27).

A legal protection that is activated only through a civil procedure may be enough to fulfil the obligations of the state. However, the case law of the ECtHR implies that when the infringement of the right is more far-reaching and touches the core of the right to private life, or when the individual in question is particularly vulnerable (e.g. children), protection through criminal law may be necessary. In cases where functioning protection is available through the law, the positive obligations include a further responsibility of the state to ensure that the rights of the individual are secured in practice, through measures such as procedures and investigations of violations. This can be illustrated by current discussions in Sweden, where more forms of victimisation may now lead to public prosecution than before, due to recent changes in legislation (The Swedish Penal Code, 1962: 700, Chapter 5, Section 5). This is in situations when the online antagonism (or antagonism in real life) is alluding to a person’s race, the colour of their skin, nationality or ethnic origin, religious belief or sexual orientation. In such cases the prosecutor should be more inclined to carry through a public prosecution. This is also the case with defamation and insults targeting victims less than 18 years of age (Swedish Penal Code, 1962: 700, Chapter 5, Section 5). A majority of cases must all the same still be handled through private prosecution or a civil procedure. However, in 2014, as a result of the recent debate regarding online antagonism, a change was made in the legislation increasing the possibility of public prosecution. This change was made in the light of modern technology and the changes in victimisation such technology may bring. The risk of widespread defamation has, for example, increased in the age of the internet, and such exposure should, according to the Swedish government, also be taken into account (Government bill, 2013). With the exception of the view of public officials, the older as well as this newer line of reasoning from the Swedish legislature is consistent with the vulnerable groups mentioned above in relation to the considerations of the ECtHR.

The approach of the ECtHR to issues of positive obligations in the particular situation of online harassment, and its potential implication for victims, is illustrated by the case of K.U. v. Finland, where the victim also belongs to a particularly vulnerable group, namely, children. The case concerns an advertisement placed on an internet dating site, purportedly in the name of a 12-year-old boy, without his knowledge. The advertisement included identifying information and a link to the web page he had at the time, which showed his picture, as well as his telephone number, which was accurate save for one digit. In the advertisement, it was claimed that he was looking for an intimate relationship with a boy of his age or older ‘to show him the way’ ( K.U. v. Finland, 2008: § 7). The false ad constituted malicious misrepresentation under the Finnish penal code, but as the offence was not included in the acts that could warrant the obtaining of telecommunications data connecting the IP number used to access the site to an individual, the police had limited means through which to investigate the crime ( K.U. v. Finland, 2008: §§ 6–14). The ECtHR found that the lack of effective measures enabling the police to investigate the crime meant that the state did not fulfil its positive obligations under article 8. The ECtHR acknowledged that the interest in investigating the crime had to be subjected to balancing against the privacy in relation to identifying information of telecommunication data, but found that in this case the interest of upholding a functioning protection against this type of violations took precedence ( K.U. v. Finland, 2008: § 49). The Court found that the existence of an offence has limited deterrent effects if there is no means to identify the actual offender and to bring him (or her) to justice ( K.U. v. Finland, 2008: § 41 and 45). Also, the possibility to claim damages from the operator of the site where the ad was placed was not considered sufficient. Instead, the Court explicitly referred to the victimological and public interests of using criminal law sanctions.

It is plain that both the public interest and the protection of the interests of victims of crimes committed against their physical or psychological well-being require the availability of a remedy enabling the actual offender to be identified and brought to justice, in the instant case the person who placed the advertisement in the applicant’s name, and the victim to obtain financial reparation from him. ( K.U. v. Finland, 2008: § 47)

Prioritised victims and victimisation

From a victimological standpoint, article 8 of the ECHR and the interpretations of the ECtHR shed some light on what victims should be able to expect – as a minimum – in terms of legal protection and remedies when faced with online antagonism. According to the ECHR and the interpretations by the ECtHR, certain groups of victims and specific types of victimisation have stronger protection.

Children or other young victims are groups that can be seen to have stronger protection under the ECHR, as well as people who suffer especially severe invasions of their privacy, such as crimes targeting their sexual integrity ( K.U. v. Finland, 2008; Söderman v. Sweden, 2013; X and Y v. the Netherlands, 1985). With that said, people who, as a part of their profession, are involved in political and current debate have less protection, since they are considered to have voluntarily and knowingly entered into that arena and thus have to withstand more harshness online ( Janowski v. Poland, 1999; Lindon, Otchakovski-Laurens and July v. France, 2007).

Here, Sweden serves as an interesting contrast, since public officials have been given somewhat stronger protection against defamation and insults when the crime targets them in their professional capacity (Swedish Penal Code, 1962: 700, Chapter 5, Sections 1–3 and 5). The protection is stronger in the sense that the fact that a crime has targeted a public official exercising public authority is a guiding factor for the prosecutor indicating that the reported crime should lead to public prosecution (Swedish Penal Code, 1962: 700, Chapter 5, Section 5). In contrast, private prosecution (or a civil procedure) is under Swedish law – in the large majority of cases – the only way to legal redress in cases of defamation and insults (Government bill, 2013; Jareborg and Friberg, 2010; Swedish Penal Code, 1962: 700, Chapter 5, Section 5). Such crimes reported by public officials, however, do not automatically set public prosecution in motion, but are a factor to be taken into consideration. The main reason for this generally restrictive reasoning from the Swedish legislature with regard to defamation and insults is that the public interest in prosecuting such crimes has generally been regarded as too low (Government bill, 1962, section C, p.185; 2013). ³ Public officials also have somewhat stronger protection when targeted with threats, vengeful actions or violence when in the process of exercising or being about to exercise public authority (Swedish Penal Code, 1962: 700, Chapter 17, Sections 1, 2 and 4). Although physical violence may not be actualised online, other forms of threats and vengeance can be. Since such crimes in Sweden are viewed not only as targeting specific individuals but also as crimes targeting the state, particularly strong protection has been considered helpful to ensure correct and safe execution of public authority and power over the citizens (Statens offentliga utredningar, 1972: 191).

The special status of victims of hate speech seen in Swedish legislation is also worth mentioning. This special protection is an illuminating example of how certain kinds of victimisation or hateful expressions may be excluded from the protection of article 10 (see above, also Féret v. Belgium, 2009). When online antagonism consists of hate speech and hate crimes targeting inter alia race, ethnic origin or religion, it is no longer as relevant for states to consider article 10, since such actions do not necessarily gain protection under the Convention (see Norwood v. United Kingdom, 2004; Pavel Ivanov v. Russia, 2007; Witzsch v. Germany, 2005). As the ECtHR has put it, ‘abuse of freedom of expression is incompatible with democracy and human rights and infringes the rights of others’ ( Witzsch v. Germany, 2005). Accordingly, under the Convention all kinds of victimisation resulting from hate speech gain stronger protection since state action against such expressions does not have to be restricted to the same extent with regard to freedom of speech. This means that no matter what group or groups of victims an individual belong to – be it politicians or children – he or she should have a right to expect stronger legal protection and ways to legal redress when subjected to such crimes, as the positive obligations of the state do not necessarily have to be balanced against article 10.

Legal redress for victims

The ECHR and the interpretations by the ECtHR directly affect the possibility of access to justice and associated legal redress for individual victims, in Sweden and other signatory states, as shown by the cases of X and Y v. the Netherlands, K.U v. Finland and Söderman v. Sweden, discussed above. Not only can victims be compensated when states fail to provide proper protection of their rights, but states are also responsible for changing laws or institutions to ensure such protection in the future. This may have a substantial impact on victim needs in jurisdictions where the legislator does not prioritise those rights.

Signatory states may admittedly enact even more far-reaching laws to protect victims and their privacy, but such legislation cannot venture beyond what is reasonable in relation to freedom of expression under the ECHR. Consequently, the Convention in a sense establishes both a minimum and a maximum level of protection against online antagonism. Where the line is drawn will vary depending on the different kinds of victimisation presented above, as well as the individual victim and circumstances.

Legal redress for victims can be achieved in different ways, and civil proceedings with the possibility of claiming damages may be one way to achieve it, as is discussed in Söderman v. Sweden and X and Y v. the Netherlands, and by some Swedish legal scholars (Schultz, 2013). However, criminal law and criminal proceedings are clearly an important aspect in protecting the integrity of victims in relation to other interests, and the availability of the criminal justice system is therefore an important step towards protecting victims on the internet (cf. Burman, 2011: 289; Enarsson, 2014–2015: 879; Keats Citron, 2014: 121–125). The impact of online victimisation is still rather unexplored, but generally a legal process can be difficult to handle in the already stressful and exposed situation someone might find himself or herself in after having being threatened, harassed or had personal information spread about them. It could be even more so when not having access to the legal support provided by the criminal court and public prosecution (Boutellier, 2000; Herman, 2003: 161–162). As such, to focus on the responsibility of individual victims to protect themselves or claim redress or reparation may be problematic from a victim’s perspective. Boutellier (2000) also discusses the possible disadvantages for crime victims if they only have access to a civil court instead of a criminal court:

The replacement of the criminal court by a civil court […] would only seem to turn against the crime victim. Not every victim can be assumed to have enough time, money, energy and willpower to see to it that justice is done to him or her, for example via the civil court. (p.66)

Despite the fact that victims deal with victimisation in different ways and that victims have individual needs, some factors can still be considered more general in nature; such as victims’ need to be listened to and be informed about the judicial proceedings, and the need for various support and protection in the acute phase as well as later (Elliott et al., 2012; Goodey, 2005; Ten Boom and Kuijpers, 2012: 155–179). As shown above, states have a positive obligation to help satisfy some of these needs, by providing access to a legal process, and actual possibilities to investigate and potentially prosecute these crimes. This must then also, from a victimological perspective, be combined with other elements, such as respectful treatment of victims during that legal process to ensure a feeling of being heard and taken seriously, information and communication, and support and protection from judiciary and other relevant authorities (Enarsson, 2013; Laxminarayan, 2013; Shapland and Hall, 2007). In contacts between law enforcement and victims of online antagonism this could be crucial, since such crimes can be difficult to investigate and prosecute. As a result, other factors may have to play a large part in giving victims the experience of receiving some form of redress from the criminal procedure as a whole. Making sure that the judicial process and the contacts with the judicial system run smoothly and respectfully, can then help counter secondary victimisation, such as negative effects or societal responses in regard to the previous victimisation (Herman, 2003; Laxminarayan, 2012, 2013; Montada, 1994; Orth, 2002).

At a time when legislative changes in this area of online antagonism are being considered – as they are in Sweden – such considerations of secondary victimisation are also important to ensure that legislative changes are not made without any real possibilities of making a practical change. This can also be related to the case of K.U. v. Finland mentioned above ( K.U. v. Finland, 2008: § 49), and the discussion concerning whether states actually provide means to ensure investigation. If not, victims who report such crimes will nevertheless risk disappointment in a very early state, simply for having trusted legislation that lacked a realistic chance of providing redress.

Future developments: ECHR making a difference for victims?

There may certainly be remaining issues for victims when having gained access to a criminal procedure, such as the risk of secondary victimisation. However, it must still be seen as necessary, not only for the individual victim but also from an overall societal perspective, to provide access to legal remedies against online antagonism, preferably through criminal proceedings. If for nothing else, it is to make sure that the authorities show the severity of the problem, and through legislation and legal procedure prove that subjecting others to such hateful behaviour does not go unnoticed or is met with impunity. Otherwise, as highlighted by the ECtHR, the deterrent effect of criminal law will be limited. A retreat of the criminal law from an issue that is clearly affecting the lives of many individuals will undoubtedly also open the door to alternative approaches, such as the insurance policies against online antagonism mentioned at the beginning of this article. This insurance, however, still maintains focus on the responsibilities and participation of individual victims, rather than the state. As such, they are not likely help prevent further cases of online antagonism through a general preventive effect. Consequently, they might be regarded more as a symptom of a problem – the failure of the state to protect victims online – than a solution to that problem.

Another important consideration for legislatures is that in trying to enact measures they should take care not to over-reach. That is, in combating online antagonism, states should not, for example, wage war against anonymity on the web as a general concept, as that anonymity serves important purposes for public discourse as well as for victims, even though it is sometimes used for nefarious purposes (Akdeniz, 2002; see also Delfi AS v. Estonia, 2015). Nor should states enact measures that do not provide opportunity for individual assessment of relevant circumstances. Instead, legislatures should focus on providing functioning legal tools and resources to investigate individual complaints to ensure that victims that seek out the support of the criminal justice system may be given an actual, rather than simply a perfunctory, chance at redress.

However, as shown above not all victims will be able to gain access to a legal process when subjected to online antagonism, and even if they do gain access there is no guarantee that their case will be successful or that the legal process will lead to legal, or other kinds of, redress. Instead it has been deemed inappropriate by the ECtHR, in fact undemocratic, for some of these offences, when directed at individual politicians or civil servants to result in a criminal procedure. To do so would, according to the ECtHR, be too much of an intrusion on freedom of expression and the interest of a free and open public debate.

Consequently, there is an inherent tendency towards victim differentiation resulting from the case law of the Court. In most cases, it is likely that the positive obligations under article 8 and the negative obligations under article 10 will interact. That is, the positive obligations will be reduced as the negative obligations are increased. A harsh expression aimed at a public official in the context of political discourse does not necessarily activate the positive obligations to act under article 8; simultaneously, the state may have an obligation not to interfere under article 10. In contrast, hateful expressions aimed at a vulnerable ‘target’ such as a child simultaneously reduce the protection under article 10 and increase the obligation of the state to act to secure the rights of the victim under article 8.

In this context, it should again be emphasised that there is strong normative protection for certain other groups and types of victimisation under the ECtHR, although various aspects relating to other victims admittedly interact to require a more severe violation for protection to be realised. This differentiation may be seen as leaving some victims behind. However, awareness of this differentiation may aid legislators in their policy choices. As states push to develop legislation to further protect victims of online antagonism, it seems clear that a navigable avenue for legislative measures, without risking conflict with article 10, is to begin by securing stronger protection for certain types of vulnerability, such as children, victims of hate crimes or violations targeting the sexual integrity of the victim. This prioritisation may to some extent be necessary as a short-term solution, as the balancing needed in relation to victims such as elected officials will warrant more careful scrutiny by the legislator and courts. Meanwhile, as the ECtHR case law develops, it seems likely that the balancing between freedom of expression and privacy will develop with it, taking into consideration the unique challenges of the online sphere as evidenced by the reasoning in Delfi and K.U. discussed above – re-establishing some sense of equilibrium in the relationship between these two rights.

Finally, who you are will undoubtedly affect your individual possibility of legal redress when victimised online. This is of course important from a victimological perspective and on a micro level. However, considering these issues on a legal as well as victimological macro level, what considerations can and should be made between victims’ rights and freedom of expression? One answer can be found in the words of Salter and Bryden when discussing cyber harassment versus freedom of expression (2009: 114):

The authors consider that in balancing these two important competing rights, it is essential that sufficient weight is given to the potential for harm, emotional or physical, that cyber harassment may entail. In weighing the right of an anonymous individual to wilfully do acts that have the effect of causing harassment, alarm or distress against the right of any person not to be exposed to such acts, it is submitted that the scales must be tipped against freedom of expression in favour of protecting the individual. This is not to minimise or ignore the importance of the right to freedom of expression, nor to ride roughshod over established internet principles. However, such principles are not absolute and when held against the right of an individual not to suffer harassment, the latter must prevail.