ABSTRACT The positive obligation to prevent private acts of violence is still developing in international human rights law. The article discusses what preventive measures States are obliged to take against such violence under the Council of Europe Convention on Preventing and Combating Violence against Women and Domestic Violence (‘the Istanbul Convention’). It examines the content of the obligation to prevent as well as what triggers it. This is exemplified by analysing the obligation under the Istanbul Convention to prevent honour-related violence. The article concludes that the triggers of the obligations to prevent and to protect differ and that the Istanbul Convention has the potential to influence the understanding in particular of the former obligation. However, first the Istanbul Convention’s monitoring body will have to clarify the extent to which primary prevention measures are required in response to individual risks of violence. 1. INTRODUCTION The Council of Europe Convention on Preventing and Combating Violence against Women and Domestic Violence (‘Istanbul Convention’)1 entered into force on 1 August 2014. As the monitoring of the Istanbul Convention is only in its initial stages, this article focuses on what potential it holds for the prevention of violence. The analysis is made in the light of the wording of the Istanbul Convention, its preparatory materials and its Explanatory Report.2 Where relevant, reference is also made to the case law of the European Court of Human Rights (ECtHR), which will guide the application of the Istanbul Convention.3 There exists other research on the potential of the Istanbul Convention in the elimination of violence against women,4 but, so far, no study has analysed its unpublished preparatory materials to assist in the interpretation of the treaty.5 The unpublished materials referred to are Compilation 1, which consists of the comments of the delegations on the Second Draft Convention, and Compilation 2 of reservations and comments of the delegations on the Third Draft Convention, as well as the four draft Conventions.6 When assessing the interpretative value of these unpublished materials, one should keep in mind the limitation that they do not constitute traditional travaux préparatoires that would reflect the entire discussions held, only the specific comments on the draft conventions by the delegations. Nonetheless, they are of interest as they reflect the drafters’ opinions regarding particular legal points and thereby aid understanding of the context in which a particular wording came about. The potential of the Istanbul Convention in the prevention of violence will be exemplified by studying the preventive obligations applicable to persons threatened by honour-related violence. These particular obligations have not been in focus in earlier research on the Istanbul Convention. Honour-related violence is a multi-faceted and deeply rooted phenomenon.7 However, it is preventable, even more so than certain other forms of gender-based violence, due to its tendency to generally be preceded by threats of violence and premeditated.8 It has already been established that prevention is essential to any strategy to end one particular form of violence which can be undertaken in the name of honour, namely female genital mutilation (FGM), and that it needs to complement legislation and other measures in order to effectively eliminate this practice.9 Meanwhile, legal discussions on other forms of honour-related violence still primarily centre on prosecution and, to a smaller extent, protection measures after violence has taken place, at the expense of preventing violence from occurring in the first place. This article looks at the nature of the different preventive obligations in order to identity whether the Convention also foresees primary prevention or whether it is confined to reacting to abuses after these have taken place.10 It finds that once the scope of the obligation of states to undertake primary prevention has been clarified, the Istanbul Convention could have a remarkable impact in terms of driving ratifying States to seek effective ways to assist those many girls and women in Europe who are faced with the threat of violence in the name of honour in their everyday life. This article deals with two broad topics, honour-related violence and prevention. Other authors have made useful analyses of honour-related violence as a phenomenon, noting that it takes different forms in different communities and also in different families within the same community and that these forms change over time.11 Also, prevention is a wide concept, the many aspects of which this article cannot cover. It confines itself to throwing light on the implications of the confusion between prevention and protection. Thus, we do not elaborate on, for example, the effectiveness of specific preventive measures, including the deterrent effect of criminal law measures. The article departs from an explanation of how the Istanbul Convention will benefit the prevention of violence. It discusses the limitations imposed by the scope of the Convention as well as whether specific preventive obligations apply to honour-related violence. The Istanbul Convention refers explicitly to honour-related violence only in the context of the prohibition of justifying acts of violence with honour. It notably needs to be clarified whether under the Convention honour-related violence constitutes domestic violence. Certain provisions apply specifically to domestic violence but do not mention honour-related violence. The author argues that those preventive obligations are applicable also to honour-related violence. Nonetheless, it would clearly be flawed to simply subsume honour-related violence under the term domestic violence. Like all forms of gender-based violence, honour-related violence has distinct characteristics, which affect the detailed positive obligations of the state. We then proceed to discuss what triggers the obligation to prevent violence under the Istanbul Convention. We propose that the Convention could assist in refining the understanding of prevention by clarifying the difference between protection and prevention. Finally, we take a closer look at the preventive obligations arising under the Istanbul Convention, both those of a general nature and those pertaining to prevention of violence in individual cases. 2. APPLICATION OF THE ISTANBUL CONVENTION TO SPECIFIC FORMS OF VIOLENCE A. The Scope of the Istanbul Convention The Istanbul Convention builds on the case law of the ECtHR.12 It also constitutes a development of the standards elaborated under the Convention on the Elimination of All Forms of Discrimination against Women (CEDAW).13 The ultimate aim of the Istanbul Convention is the prevention of all forms of violence covered by its scope.14 It expressly requires that States exercise due diligence to prevent private acts of violence.15 However, it also foresees measures to protect individuals already subjected to violence from further acts of violence as well as support and assistance to overcome the consequences of violence and rebuild victims’ lives.16 It is thus a potentially powerful tool in the elimination of violence against women. During the drafting process, there were attempts to weaken the protection offered by the Istanbul Convention, but these were largely defeated in the end. Notably, certain delegates questioned the application of the notion of due diligence to the prevention of violence against women.17 They felt that it was not clear what due diligence means in this context. Some suggested that the notion of ‘duty of care’ also used by the ECtHR would be more suitable.18 The principle of due diligence is taken for granted by most human rights advocates today, and the fact that it was put into question in relation to prevention of violence against women demonstrates the strong need for the Convention.19 The Istanbul Convention defines violence widely. It covers both domestic violence and violence against women, but distinguishes the two. It regards domestic violence as a form of violence against women and notes that it affects women disproportionately.20 It is defined in Article 3(b) as ‘all acts of physical, sexual, psychological or economic violence that occur within the family or domestic unit or between former or current spouses or partners, whether or not the perpetrator shares or has shared the same residence with the victim’. Meanwhile, violence against women is defined as ‘all acts of gender-based violence that result in, or are likely to result in, physical, sexual, psychological or economic harm or suffering to women, including threats of such acts, coercion or arbitrary deprivation of liberty, whether occurring in public or private life’.21 The Convention does not define honour-related violence, nor is there any universally accepted definition of it. The author adheres to the following definition, which sums up the main elements of the complex phenomenon: violations of physical or psychological integrity or personal autonomy by means of threat or use of violence committed by family members in the name of honour, in which a central element is the requirement to adhere to the sexual and social conduct that is the norm within the community.22 The perpetrators are not always immediate family but can also be other relatives. They are generally men. Most victims of honour-related violence are women, but the social norms on honour also set rigid codes of conduct for men, usually including heterosexuality and the duty to control the behaviour of female family members. In the case of domestic violence and violence against women, at worst the surrounding community may silently approve of the violence. When it comes to honour-related violence, some (parts of) communities strongly demand that violence is used to uphold social norms. These social norms can have a very strong standing, as a consequence of which criminalizing legislation and general awareness-raising measures only have a limited preventive effect.23 The limitations as to the personal scope of the Istanbul Convention were much debated during the drafting process. In the end, the drafters decided that the Convention applies to all girls and women.24 Ratifying States have the option of applying the Convention to all victims of domestic violence (also men and boys), but are not obliged to do so.25 Meanwhile, it was clear from the outset that girls would be included,26 although the explicit reference to children as victims of domestic violence was introduced into the Preamble only in the Third Draft Convention. During the drafting, it was repeatedly suggested that the text should specify that the Convention deals with male violence against women, but others stressed that women also commit forms of violence falling within the Convention and the suggestion was rejected. Here, the example of FGM was advanced.27 It was also proposed during the drafting process that the specific human rights protected by the Istanbul Convention should be listed, but this suggestion was not taken up.28 It would have been beneficial to reassert that violence in the home can implicate a number of human rights norms, ranging from norms such as the right to life and prohibition of torture and inhuman or degrading treatment to the right to private life. The Istanbul Convention’s monitoring body will nonetheless still be able to draw upon the case law of the ECtHR in relation to those rights. Most of the provisions on prevention are generally formulated, and it appears that the drafters decided that as each form of violence covered by the Convention requires measures that take into account its characteristics, there is no need to specify measures particularly suited for the prevention of a specific form of violence. This approach is not consistent, though, as there are some rather detailed requirements for specific measures against certain forms of violence. This article argues that it would have been helpful also to single out preventive measures against violence that is incited by the community.29 Although honour-related violence does not figure prominently in the Convention text, there was some discussion of it during the drafting process. In the Second Draft Convention, there was a separate provision requiring criminalisation of crimes committed in the name of honour but it was later removed.30 The desire to avoid ‘othering’ and stigmatising communities with an immigrant background by dealing with honour-related violence separately from violence against women is understandable.31 Also, honour-related violence consists of physical and psychological acts of violence that are similar to other forms of violence covered by the Convention, so a separate mention of its criminalization would indeed have been unnecessary. However, it would have been useful to contemplate its prevention. As the drafters decided that the Convention would only include a mention of honour being an unacceptable justification for crimes, there was little discussion of what forms honour-related violence takes and, above all, how they can most effectively be prevented.32 The final adopted text of the Istanbul Convention can be criticized for watering down some rights included during the drafting process. For example, the present Article 42 originally contained a requirement to criminalize incitement to commit suicide, a not uncommon form of honour-related violence,33 but this part of the provision was later removed.34 At the same time, it should be kept in mind that in some respects the adopted text is stronger than in the early drafts. For example, the provision in Article 31 demanding that incidents of violence are taken into account in the determination of custody or visitation rights of children originally referred only to significant incidents of violence, a formulation which could have been interpreted as leaving States with the option to disregard anything but very severe forms of violence against children.35 A strong asset of the Convention is the timely reminder of the requirement to provide non-discriminatory protection against violence. It protects all women and girls and it lists prohibited grounds of discrimination when States provide protection. These expressly include disability but not, for example, legal capacity, although during the drafting process delegates suggested including this discrimination ground.36 The list is, however, not exhaustive. It would arguably be contrary to this provision to limit, for example, access to specialist support services so as to exclude asylum seekers or undocumented migrants.37 B. Applicability to Honour-Related Violence While there is no doubt that the Istanbul Convention as such encompasses honour-related violence, the applicability of individual provisions of the Convention to honour-related violence merits clarification. This particular form of violence is only explicitly mentioned in the context of the obligation to ensure that culture, custom, religion, tradition or honour are not considered a justification for any acts of violence. The positive obligation to ensure this in a penal context is contained in Article 42(1), while Article 12(5) also applies to the civil law context. The Explanatory Report clarifies that the prohibition of any of the acts of violence covered by the Convention in accordance with Article 42(1) cannot be invoked as a restriction of the perpetrator’s cultural or religious rights or freedoms.38 Furthermore, the article in question precludes any official statements, reports or proclamations that condone violence based on culture, custom, religion, tradition or honour, thus also containing an essentially negative obligation.39 The Convention separately mentions forced marriage and FGM, both of which can be undertaken for reasons of honour, but only to require the criminalization of both crimes in domestic law, determine the civil consequences of forced marriages and regulate the residence status of victims of forced marriage. Other forms of honour-related violence are not expressly mentioned, perhaps for the simple reason that most of the drafters were mainly aware of these two forms. Certain articles of the Istanbul Convention refer specifically to domestic violence, which centres on the family or domestic unit. The Convention notably contains an obligation in Article 16(1) to set up or support programmes aimed at teaching perpetrators to adopt non-violent behaviour in interpersonal relationships, but this obligation relates only to domestic violence, sexual assault and rape. Such measures would be highly beneficial also to the prevention of honour-related violence. The question is whether all acts of honour-related violence as such fall within the definition of domestic violence. The term ‘family’ is not defined in the Convention or the Explanatory Report, leaving open the possibility that acts of honour-related violence perpetrated by more distant relatives might not be included.40 However, such an interpretation would not be compatible with the object and purpose of the Convention to protect women against all forms of violence. We, therefore, suggest that Article 16(1) should be applied to all honour-related violence against girls and women.41 The definition of what constitutes a family will no doubt be clarified as the Group of Experts on Action against Violence against Women and Domestic Violence (GREVIO) established to monitor its implementation has the opportunity to comment on State reports as well as issue general comments. Logically, it would do so in a way which includes acts of violence committed by any relative within the concept of domestic violence. There is also the issue of which forms of honour-related violence are covered by the Convention, as the only forms specifically mentioned are forced marriages and FGM. GREVIO monitors the implementation of the Istanbul Convention through State reports, which will reflect the conception of States regarding the forms of honour-related violence to which the Convention applies. GREVIO has prepared a set of questions to be used as a basis for the first State reports.42 As in the Convention text, the document refers to honour-related violence only in the context of the requirement for measures to ensure that culture, custom, religion, tradition or so-called honour are not considered as justification for any acts of violence. The first State reports to GREVIO were submitted by Austria, Monaco, Albania and Denmark. Only the Danish report deals with the prevention of honour-related violence other than forced marriage and FGM.43 There is thus an obvious need for clarification by GREVIO as to the specific forms of violence regarding which States should report their preventive measures. Those cannot be limited to forced marriage and FGM. As the Istanbul Convention requires States to collect and disaggregate data by forms of violence, it is conceivable that GREVIO will introduce a requirement to collect and report data also on other forms of honour-related violence.44 Examples of other forms of honour-related violence which have occurred relatively recently in Europe and may in the future need to be reflected in reports to GREVIO include instances of acid attacks and so-called honour killings. Reporting separately on such crimes would require States to demonstrate how their preventive methods have taken into account the motive of the acts. 3. TRIGGERS OF THE OBLIGATION TO PREVENT This section seeks to clarify the confusion between the obligation to prevent violence and the obligation to protect against violence. The determination of the scope of the State’s obligation to prevent and protect individuals against private acts of violence has been described as problematic.45 The Istanbul Convention has the potential to help clarify not only what States must do to prevent violence by private individuals but also in what situations they are required to take action. In Section 4, we discuss what kinds of measures are required. Here we focus on what triggers the duty to prevent violence. International human rights bodies do not always distinguish clearly between protective measures and preventive measures. In the general usage of the terms, protective measures are initiated after violence has already occurred. Preventive measures, on the other hand, aim to preclude violence from occurring in the first place. In the literature, the concepts are to some extent used interchangeably. There can be said to exist three stages of prevention: primary prevention, secondary prevention and tertiary prevention. Primary prevention aims to forestall violence before it occurs, secondary prevention aims to detect violence in time or to terminate it at the earliest possible point, and tertiary prevention aims to prevent a renewed outbreak of violence or to lessen its impact.46 Here, we are concerned with all three stages of prevention. Secondary and tertiary prevention involve measures, which in the Istanbul Convention and ECtHR case law are called protection measures. There is general support for the argument that a State is required to take measures of a general nature when there is a pattern of violence in that State.47 Obviously, it is not efficient for a State to initiate random preventive measures on the general level just for the sake of it. The Istanbul Convention requires States to collect data and support research on all forms of violence covered by its scope.48 Measures to abolish practices that discriminate against women should be taken without delay, so the collection of data and support to research should be initiated swiftly if not already available.49 This information will constitute the basis for assessing when there is a pattern of violence that will trigger the obligation to undertake general measures. It should be kept in mind that the trigger for the obligation to prevent violence in individual cases might differ depending on whether the person under threat is a child or an adult. Under the Convention on the Rights of the Child (CRC)50 the authorities must take action to protect children against all violence. There is thus no minimum level of severity of ill-treatment in relation to honour-related violence against children under the CRC.51 Any violence against a child triggers the obligation to intervene to prevent further violence. As for the Convention for the Protection of Human Rights and Fundamental Freedoms (ECHR),52 there are indications that the ECtHR may also not require a minimum level of severity in some cases concerning violence against children. Where children have been victims of violence, the ECtHR has in some more recent cases left out the minimum level of severity test in the context of Article 8. In, for example, Eremia v Moldova,53 the authorities were held to have failed to protect the right to private life of two children who were subjected to verbal abuse and had to watch the ill-treatment of their mother. The authorities had issued a protection order prohibiting the perpetrator from contacting, insulting or ill-treating the children, but had not followed-up when he violated the order. There was no discussion of the level of severity of the treatment to which the children were subjected.54 GREVIO may also require that States intervene against any violence against children. In the case of adults, the positive obligation to intervene to prevent violence is somewhat more muddled. As the case law of the ECtHR will guide the application of the Istanbul Convention, one needs to recall the principles laid down by the Court to guide States in responding to risks of violence. The ECtHR generally assesses if the authorities knew, or ought to have known at the time, of the existence of a real and immediate risk to the life or limb of a person and if they failed to take measures within the scope of their powers which, judged reasonably, might have been expected to avoid that risk.55 The case law indicates that under the ECHR, States must intervene to protect adult victims of violence only when the effects of the violence reach a certain level of severity. This is a context-specific evaluation,56 implying that there should be a lower threshold for intervention where physical or psychological violence is directed at persons in a vulnerable position, as persons threatened by honour-related violence often are. However, the requirement of a minimum level of severity relates specifically to assessing the response of authorities to situations where violence has already occurred, and we will argue below that one should not apply it as such to primary prevention measures. The ECtHR case law also includes the requirement that the risk of violence must be immediate to trigger an obligation to take action in an individual case. It is apparent that the threshold relating to immediacy of violence is unsuitable for violence within the home, and its application to domestic violence has rightly been criticised.57 In cases of honour-related violence, reserving intervention only for when there clearly is an immediate risk similarly places victims in a very perilous situation. In practice, the ECtHR has not always demanded the existence of an immediate risk, including in cases involving an assessment of whether national authorities have done enough to protect victims of domestic violence against further violence.58 Ebert and Sijniensky indeed find that the Court has applied the requirement for an immediate risk in a rather flexible manner depending on the context of the case.59 The threshold for when States must intervene in individual cases may be lower under the Istanbul Convention. Its only reference to immediate risk as a precondition for action concerns the emergency barring orders foreseen in Article 52. The Istanbul Convention also does not contain any reference to a minimum level of severity of violence triggering the State obligation to intervene, with the exception of psychological violence, which must seriously impair psychological integrity in order to fall under Article 33. Disconnecting the level of severity from the obligation to intervene to prevent violence is indeed crucial to a coherent approach to prevention. Before violence has materialized, it is logically often difficult to know how serious it will be. Therefore, a risk of violence that is not trivial, should as such be sufficient to require States to initiate preventive measures under the Istanbul Convention. This author submits that we need to start perceiving the duty to prevent violence as an obligation that is distinct from the obligation to protect against the effects of violence and further acts of violence. Above all, we need to severe the link made between prevention and the minimum level of severity. Renzulli has made a similar observation.60 A human rights court will normally scrutinize individual cases only after violence has taken place, and then will assess factors such as the severity of violence in determining whether the measures taken by the State were sufficient. This does not imply that national authorities should have to or be entitled to second-guess what specific acts of violence the person at risk will be subjected to before initiating preventive action. That assessment is relevant only in relation to decisions on protection measures (restraining orders, etc). Naturally, a situation where a family member brandishes a firearm and describes how she or he will use it needs to be designated as a threat of serious violence, but this does not imply that a very dangerous situation cannot arise in the absence of any weapon.61 Professional tools for assessing the risk of violence exist in abundance.62 4. PREVENTIVE OBLIGATIONS UNDER THE ISTANBUL CONVENTION Above, we have established what triggers the positive obligation to intervene to prevent violence. We will now turn to scrutinize what measures can be required of a State that has ratified the Istanbul Convention. International case law indicates that there is an obligation to carry out both general and individual preventive measures. Most of the case law still focuses mainly on individual measures. For this reason, GREVIO’s interpretation of the general preventive obligations under the Istanbul Convention are likely to be influential on legal argumentation. The Convention’s obligations relating to investigation, prosecution and punishment are intrinsically linked to the obligations to prevent honour-related violence outside the criminal law setting. They, therefore, need to be seen as a whole. However, for reasons of clarity of analysis, they are treated separately below. A. The Requirement for General Measures The Istanbul Convention builds on four components, namely prevention, protection against further violence, prosecution of perpetrators and integrated policies.63 The legal obligations relevant to the prevention of violence can be found within all four areas, not only under the heading prevention. The understanding that States should not just prohibit, punish and remedy violence in individual cases, but prevent it through systemic measures aimed at changing attitudes, permeates the entire Istanbul Convention. Measures of a general nature are a precondition to the State being able to take effective measures against individual acts of violence. The Istanbul Convention recognises that genuine prevention requires a change of attitudes, and this cannot be accomplished without measures of an awareness-raising or educational character.64 Article 12(1) lays down an obligation to take the necessary measures to promote changes in social and cultural patterns of behaviour in order to eradicate prejudices, customs, traditions and practices based on the idea of the inferiority of women or on stereotyped roles for women and men. Furthermore, Article 5(2) provides: States shall take the necessary legislative and other measures to exercise due diligence to prevent, investigate, punish and provide reparations for acts of violence covered by the scope of this Convention that are perpetrated by non-State actors. The reference to the necessary measures indicates a duty of due diligence, as explicitly provided for in Article 5(2). The principle of due diligence requires States to organise their response to violence in a manner which allows relevant authorities to diligently prevent such acts.65 Put differently, this is a requirement to provide for and apply measures that are designed to be effective. All measures to prevent violence contrary to the Istanbul Convention should form part of policies, which according to Article 7 must be effective. The requirement for effective measures also forms a central part of the case law of the ECtHR on prevention of violence.66 Where the State deems it appropriate, it should in accordance with Article 14 include teaching on issues such as the right to personal integrity and non-violent conflict resolution in the formal curricula. It is hard to imagine a setting where such teaching would not be necessary. Article 13 adds the obligation to undertake awareness-raising activities expressly aimed at the general public. The Explanatory Report recommends that these activities include dissemination of information on non-violent conflict resolution in interpersonal relationships.67 There is no mention of awareness-raising specifically designed to target communities, such as those in which some families adhere to the practice of honour-related violence. There was also no discussion of this issue during the drafting process.68 However, a requirement to design such specific awareness-raising measures is arguably implicit in Articles 12 and 13. In families adhering to the practice of honour-related violence, general awareness-raising measures on violence against women are unlikely to be effective, as the real or perceived pressure by the larger family and the community to resort to honour-related violence is strong. In order to be genuinely effective, the awareness-raising needs to target relevant communities and individual families, based on their conception of honour.69 The Explanatory Report also makes little reference to the particular efforts that will be required from States in order to prevent community- and family-sanctioned violence. Meanwhile, a paper on the implementation of Article 13 commissioned by the Council of Europe explains that in order to reach all parts of the general public, a range of different measures and different messages will be needed to target different audiences effectively.70 It is to be hoped that this wider, and in the view of this author quite correct, interpretation of the obligation to undertake awareness-raising is adopted by ratifying States. This would be in line with the requirements for effective measures and due diligence. In general, there is yet insufficient evidence regarding the effectiveness of specific preventive measures against gender-based violence. The choice of measures is, therefore, largely left up to States.71 When it comes to FGM, there are studies showing that multisector, sustained and community-led awareness-raising effectively prevents it.72 A parallel can be made to other traditional practices such as honour-related violence that share the same characteristics, notably of being incited by the community. This is also the approach of another paper on prevention of violence commissioned by the Council of Europe. Similarly to this author, it finds that changing the social norms that underpin practices such as honour-related violence is an integral part of the prevention requirements of the Istanbul Convention.73 For its part, the ECtHR has so far interpreted the standard of due diligence as demanding that States adopt adequate legislation and procedures and organize structures in a way which enable them to prevent violence.74 It has said very little in terms of other forms of prevention, such as measures aimed at attitude change among the population. However, there is potential to regard also such measures as being essential to fulfilling the requirement for effective measures read into the ECHR by the Court. It is probable that the ECtHR’s interpretation of due diligence will continue to influence what preventive measures GREVIO demands from States, but the above conclusion on the necessity of supportive measures targeting communities indicates that GREVIO may have to go beyond the interpretation of the ECtHR. The great advantage of GREVIO is of course that while the ECtHR is confined to pronouncing on the situation in a specific case, GREVIO is able to comment on preventive requirements in abstract. It is, therefore, foreseeable that it will provide further specification regarding the positive obligations to undertake preventive measures aimed at attitude change and perhaps thereby in turn influence the case law of the ECtHR. Article 4(2) of the Istanbul Convention requires that the national legal and policy framework enable States to take measures to abolish practices that discriminate against women without delay. It also entails certain specifications regarding how this should be done. Ratifying States need to appoint a co-ordinating body to ensure inter alia that prevention measures are timely, relevant and effective.75 Importantly, the Convention also requires ratifying States to allocate appropriate financial and human resources to the prevention of violence, including supporting relevant activities carried out by non-governmental organisations (NGOs) and civil society. The wording originally used was ‘sufficient’ resources, but the drafters preferred the word ‘appropriate’.76 While the latter term certainly leaves room for national discretion in allocating resources, it does away with the possibility to set aside no funding at all when there are patterns of violence.77 The Istanbul Convention furthermore requires setting up a sufficient amount of shelters and round-the-clock telephone helplines.78 Information about their existence must be widely disseminated.79 The Convention encourages the establishment of specialised shelters, stressing that each type of violence requires a different kind of support and protection.80 Relevant professionals must in addition be trained on prevention and detection of violence, and provided with clear protocols and guidelines, setting the standards that they are expected to follow in their respective work.81 Moreover, Articles 18(2) of the Convention demands that there be appropriate mechanisms to ensure effective cooperation between relevant public officials, such as judges, prosecutors, police, local and regional authorities and NGOs, in protecting and supporting victims of violence. Interagency cooperation is quite rightly seen as so fundamental to preventing violence that it is listed among the purposes of the Convention in Article 1(1)(e). It is also mentioned in Articles 7(2), 15(2), 18(2), 51(1) and it is implicit in Articles 9 and 13. Furthermore, States must undertake data collection and research, an obligation which is of remarkable importance to the understanding of where and how to allocate preventive resources.82 In many States in Europe, there is no reliable data on the forms and scope of honour-related violence, which leads to the consequence that its prevention is either largely ignored or subsumed under the prevention of domestic violence, which, as discussed, may be insufficient to effectively prevent honour-related violence.83 The Convention also takes into account the international dimension of gender-based violence, including honour-related violence. In addition to an entire chapter on international cooperation, it notably requires States to make arrangements so that they can afford consular protection to persons who have been subjected to or risk violence. At present, only one reservation has been expressed to the Article in question.84 This is fortunate as, in particular in cases of forced marriage abroad, consular support to the victim or potential victim can be crucial. B. Prevention of Violence in Individual Cases A great merit of the Convention are the obligations designed to prevent violence from continuing or recurring in individual cases. However, one looks in vain to find any concrete primary prevention obligations in individual cases. (i) The criminal law setting The Istanbul Convention requires that physical violence, forced marriage and FGM be criminalised. The requirement for criminalization of forced marriage and FGM does not imply that specific provisions on criminalization of these particular forms of violence is mandatory; a general criminal law provision applicable to them is sufficient.85 States are also under the obligation to criminalize psychological violence, but this is limited to conduct that is intentional, seriously impairs psychological integrity and takes place through coercion or threats.86 The Explanatory Report clarifies that psychological violence refers to a course of conduct rather than a single event and is intended to capture the criminal nature of an abusive pattern of behaviour occurring over time, within or outside the family.87 The use of psychological violence preceding or in addition to physical violence is typical of honour-related violence. States may opt to apply non-criminal sanctions against psychological violence (but not the other forms of honour-related violence mentioned above) by introducing a reservation to this effect.88 It is important in the context of honour-related violence that intentionally aiding or abetting the commission of FGM, forced marriage, psychological violence and physical violence must be made an offence.89 Similarly, inciting a minor to commit an offence shall not diminish the criminal liability of the inciting person for the acts committed.90 These provisions address the problem that persons planning acts of honour-related violence at times have persuaded or forced minors to commit the crime in order to escape criminal punishment.91 Looking at the deterrent effect of criminalisation on the individual level, this is likely to be insufficient in the case of community-incited violence.92 There has also been debate among those working with prevention of forced marriage whether criminalization is desirable, or whether civil remedies should suffice, notably because the prospect of criminal punishment for family members could deter some girls and women from seeking help. The option of giving States the choice of civil or criminal measures in cases of forced marriage was proposed during the drafting of the Convention, but did not make it into the final text.93 While forced marriage must be criminalised, there is no corresponding provision regarding hindering a woman from marrying the person of her choice, a not infrequent form of honour-related violence. It can, however, be read into the provision on psychological violence when it fulfils the criteria in Article 33. The above criminal law provisions must also be effectively applied. States are required by Article 49, read in conjunction with Article 5(2), to demonstrate due diligence in prosecuting perpetrators of honour-related violence. Prosecution may have a deterrent effect on some perpetrators, but above all, it signals clearly to perpetrators and victims alike that the authorities do not approve of such violence and are committed to eliminating it. This can encourage victims to seek help from the authorities. There are also other provisions intended to facilitate the prosecution of acts of violence. Article 55(1) of the Convention requires that the prosecution of certain crimes is not ‘wholly dependent’ upon the report or complaint filed by the victim and that proactive investigations ensure that the legal proceedings may continue even if the victim withdraws her (or his) statement or complaint.94 These crimes are physical violence, sexual violence, forced marriage, FGM, forced abortion and sterilization. While the wording of the provision could be read to imply that the victim must at least initiate an investigation, the Explanatory Report dispels any such misunderstanding.95Ex officio prosecution is not mandatory in cases of psychological violence, stalking and sexual harassment. States may also choose to issue a reservation excluding public prosecution of physical violence in case of minor offences, a concept to be defined by the State.96 It should be stressed that this does not imply that the State does not have an obligation to prevent to also less serious forms of physical violence.97 It may not always be possible to establish whether a certain crime has been committed due to conceptions of honour or for other reasons. In such cases, it is important to note that the Istanbul Convention does not only prohibit that culture, tradition, religion, custom or honour is regarded as a justification for any act of violence covered by the Convention. It has been suggested that Article 42 also implicitly excludes the possibility of permitting loss of self-control (due to jealousy, substance abuse or similar circumstances) to be used as justification for crimes.98 Furthermore, a number of elements typical of honour-related violence can be treated as aggravating circumstances in sentencing. These include that the crime was committed against a family member or a person made vulnerable by the circumstances or by several persons acting together. The decision whether to apply these aggravating circumstances in a specific case is up to the judge, but they should be listed among the aggravating circumstances to be taken into consideration.99 Importantly, criminalisation and prosecution of honour-related violence are not the only nor the most efficient ways of preventing its occurrence. We will now turn to preventive measures outside the criminal law context. (ii) Preventive obligations outside the criminal law setting Under the Istanbul Convention, a key requirement is that national law enforcement agencies take the (unspecified) preventive operational measures foreseen in Article 50 when faced with individuals at risk of violence.100 This provision reflects and should be interpreted in light of ECtHR case law. The measures can be described as secondary prevention. The Court has provided a fair amount of case law on what the principle of due diligence implies in relation to prevention of private acts of violence, including demanding preventive operational measures where the (potential) victim’s life or limb is at risk.101 The Court has also occasionally indicated other forms of measures, including that social authorities should work with individual perpetrators to reduce the risk of violence in the future.102 The Istanbul Convention goes further than this by introducing an obligation of primary prevention, namely a requirement to change attitudes. The requirements as regards awareness-raising on the general level have been discussed above in Section 4.A. There is also an obligation to set up a system for changing attitudes in specific individual cases. While as a rule it is up to the State to choose by which preventive measures it addresses an individual risk of violence, Article 16 of the Convention requires certain secondary prevention measures specifically against domestic violence, sexual assault and rape. These entail programmes aimed at influencing perpetrators of violence so that they adopt a non-violent behaviour in interpersonal relationships. States need not make participation of perpetrators in these programmes obligatory; it can be voluntary or court-ordered.103 We have discussed the applicability of this provision to honour-related violence above.104 The obligation to undertake preventive interventions should arguably not be interpreted as being limited to measures aimed at perpetrators stricto sensu. When the whole family is complicit in honour-related violence, measures must be aimed at all of its members.105 Article 12(3) requires positive action to ensure that any preventive measures specifically address and take into account the needs of persons made vulnerable by particular circumstances. While the non-exhaustive list of vulnerable persons in the Explanatory Report does not include persons living in communities that incite the use of a specific form of violence, these individuals may arguably be in a particularly vulnerable position if they lack outside support and information about available protection measures.106 The list does include migrants, but far from all persons subjected to honour-related violence in the ratifying States are migrants; nor is this necessarily what makes those who are migrants vulnerable. The ECtHR has held that in situations, in which the physical well-being of individuals is dependent, to a decisive extent, on the actions by the authorities, these are required to take the measures within the scope of their powers that are necessary to avoid the risk of damage to life or limb caused by private individuals.107 It is clearly established that authorities are obliged, when any doubts to this effect arises, to assess whether parents provide children with proper care and protection.108 Numerous cases on domestic violence confirm that the obligation to intervene to prevent violence goes beyond the protection of children. Authorities have a positive obligation to intervene to protect vulnerable adults against violence by family members as well.109 Furthermore, the obligation of due diligence entails the demand that, if an adopted measure fails, other more effective measures must be sought.110 The CAHVIO Committee drafting the Istanbul Convention agreed at the outset that violations of civil law measures for the protection of victims (for example, protection or restraining orders) should trigger criminal sanctions.111 However, this is not the only obligation to ensure the effectiveness of measures. The Convention uses strong language in introducing in Article 18(1) a general obligation for States to take the necessary legislative or other measures to protect all victims from any further acts of violence. Moreover, Article 18(3) requires that all protective and supportive measures take the relationship between victims, perpetrators and children as well as their wider social environment into account. GREVIO will surely apply a similar approach to preventive measures in individual situations, as it is implicit in the requirement for effective measures. The Explanatory Report notes that this requirement is designed to avoid the risk of addressing the needs of victims and others involved in isolation and without acknowledging their social reality.112 The importance of this demand cannot be overestimated in the often complex situations facing victims of honour-related violence. Consequently, placement of a woman in a shelter where she cannot stay longer than a very limited time cannot as such be considered a sufficient measure of protection in situations of honour-related violence, as she will often eventually be compelled to return to the family home, where the threat of violence is in no way reduced. Nor does the power to order a perpetrator of honour-related violence to leave the residence of the victim and to bar him or her from returning or contacting the victim for a limited period constitute any long-term solution.113 In order to reduce the risk of violence in the long term, such protection measures need to be accompanied by preventive awareness-raising measures aimed at the rest of the family, for example, by trained social workers.114 Here, States could be guided by the ECtHR’s constructive standpoint in Đorđević v Croatia,115 fn which concerned the abuse and harassment of a person with a disability by a group of children. The Court suggested that social services should have been involved in the case and that relevant authorities should have worked with the violent children concerned in order to combat the underlying problem, namely their attitudes. The wording of the Convention’s preventive obligations indicate that the focus is on preventing any further acts of violence (thus secondary or tertiary prevention) rather than on primary prevention in individual cases. This notably applies to preventive intervention programmes (Article 16), general obligations of protection and support, some of which would be essential also for primary prevention (Article 18), specialist support services,116 including shelters (Article 22) and reporting of professionals (Article 28). This author finds the choice of a wording that ignores the possibility to prevent violence from occurring in the first place unfortunate. The Istanbul Convention expressly recognises the possibility of primary prevention only in relation to prevention on the general level, not as a response to risks and threats in individual situations. There is one exception, namely the provision on reporting by private persons (Article 27). Regrettably, the corresponding Article on reporting by professionals is limited to acts that have already been committed and expectance of further serious acts of violence.117 Meanwhile, other provisions leave the door open for the interpretation that primary prevention is required (notably Articles 50(2) and 51(1)). There is no reflection in the preparatory material of any discussion of the word ‘further’. During the drafting process, some delegations stressed that the terminology used should reflect that there is both an obligation of primary prevention and an obligation to respond comprehensively to violence that has already taken place, including other than legal or judicial measures.118 It was also proposed that prevention should be defined as primary prevention.119 Neither the Convention text nor the Explanatory Report takes up this suggestion. It is certainly often challenging to show that there is a risk of violence in an individual case, and thus that an obligation to take preventive measures arises, without first demonstrating that acts of physical violence have already taken place. However, it is not impossible and we should not dismiss the possibility. This applies in particular to premeditated violence where pronounced threats are an integral part of the violence, as is the case with, for example, honour-related violence. If the relevant authorities act upon receiving information of a clear threat, they may be able to avoid an escalation of the situation into acts of physical violence. Simultaneously, threats of honour-related violence are at times of such a character that they can be regarded as psychological violence seriously impairing the person’s integrity, and as such oblige States to take action to protect her (or him) against further acts of violence. Even in situations where this is not the case, States are required to take ‘the necessary measures’ to prevent violence. This means that the wording regarding preventive intervention programmes (Article 16) and reporting of risks by professionals (Article 28) should not be understood to imply that identical obligations do not arise in situations where violence has not yet taken place. This author submits that such an interpretation would run counter to the object and purpose of the treaty, which is to prevent violence from occurring. 5. CONCLUSION The Istanbul Convention reflects the understanding that if we genuinely want to prevent violence, exacting punishment is necessary but not sufficient. Operational measures (such as restraining orders) putting a distance between potential perpetrators and victims are necessary to alleviate immediate risks of violence but are not a long-term solution. In addition to these measures, we need to influence perpetrators and potential perpetrators so that they do not resort to violence. This requires measures aimed at changing attitudes upholding violence. The impact of the Istanbul Convention on preventive efforts will conceivably be, above all, to broaden the scope of what prevention means to include measures aimed at changing attitudes towards the use of violence and gender roles, not only on the societal level but also on the individual and community level. Under the ECHR, the choice of methods to prevent violence has mainly been left to the discretion of the State, although the ECtHR has provided some clarification as to minimum standards. The preventive obligations in the Istanbul Convention are more detailed, but this article demonstrates that their content also needs to be spelled out. Singling out honour-related violence for scrutiny serves to demonstrate that every form of gender-based violence needs to be tackled in a manner that takes into account its specific characteristics. The Istanbul Convention reflects the strong role of the community in the perpetration of honour-related violence mainly in the criminal law setting.120 There is a requirement to criminalize incitement to commit a crime and to include factors central to honour-related violence among the aggravating factors to take into account in determining punishment for a crime. Meanwhile, when it comes to primary prevention measures other than legislation, there is no direct reflection of the need to consider the communal character of the violence. Although rather detailed, the preventive interventions expressly foreseen omit certain measures essential to the prevention of acts of honour-related violence, namely attitude change within communities and family counselling. Instead, the wording focuses on the general public and individual perpetrators. In light of the object and purpose of the Convention and its preparatory materials, a requirement to undertake these measures can, however, be read into existing provisions. Pursuing a prevention perspective under the Istanbul Convention rather than, for example, the ECHR has some definite advantages. The obligation to undertake general measures is one of the features that makes the Convention a strong tool for primary prevention. In addition, it also encompasses forms of violence that are more difficult to address under the ECHR. Basing one’s argument for a legal obligation to prevent honour-related violence on individual human rights norms under the ECHR requires that the violence in question fulfil the minimum level of severity and other criteria for the rights in question. Under the Istanbul Convention, there is an obligation to undertake general preventive measures against honour-related violence once there is information that there is a pattern of such violence in the State in question. This means that the measures, if effective, can prevent what are often intrinsic components of honour-related violence, namely excessive control and strict limitations on the social life of girls and women. Such control makes everyday life very harsh for many individuals but it may be challenging to establish that it violates any specific provision in the ECHR. Moreover, we suggest that individual measures of a primary prevention character must be undertaken when there is a risk of violence that is not negligible. Decisions on primary prevention measures should thus be distanced from the assessments of severity of violence and immediacy of risk that apply to protection measures. GREVIO will have to clarify the extent of the obligation to take primary prevention measures in response to individual risks of violence. It might well be that it takes the stand that the obligation of primary prevention in individual cases notably concerns persons in a vulnerable position. This would appear to be in line with ECtHR jurisprudence.121 In individual cases, the Istanbul Convention notably entails an obligation to devise measures aimed at changing the attitudes of the (potential) perpetrators, an obligation which may be emerging but is not very pronounced under the ECHR. We argue that measures similar to those foreseen in Articles 16 and 28 are required to stop violence from happening in the first place. Without professionals alerting relevant authorities when there is a risk of violence, many vulnerable girls and women at risk will be subjected to violence that could have been avoided. In situations where there is a known threat of honour-related violence but no physical violence has yet occurred, relevant authorities (such as social services or child protection services) must be tasked with offering the possibility of family counselling or similar services to the entire family. A valuable aspect of the Istanbul Convention is the requirement to take the possible vulnerability of the potential victim and the wider social environment into account in adopting protection and, arguably, prevention measures. In individual cases, this places an obligation on national authorities to look into what will diminish the threat of violence in that very case. Simply resorting to a standard set of measures will arguably not fulfil the level of due diligence required. Authorities also need to look at the context of the individuals (victim and perpetrator(s)) concerned. This, in turn, demands training of relevant professionals on the characteristics of the different types of violence and cooperation between different authorities, which is strongly encouraged by the Convention. The article concludes that the Istanbul Convention could help clarify the difference between the obligation to prevent and the obligation to protect against violence. It also establishes that the triggers of the two obligations are not identical. The interpretation of the precise trigger for the obligation to undertake protective measures against further violence is likely to be guided by ECtHR case law, although the wording of the Istanbul Convention sets no minimum level of severity, except for psychological violence, and contains no requirement for immediacy of risk. We argue that these criteria should not be applied when national authorities decide whether to initiate measures of a primary prevention character. Reserving family intervention for situations where there is an immediate risk of very severe violence would essentially deprive primary prevention of meaning. The Istanbul Convention has the potential to help promote this understanding. Footnotes 1 2011, ETS 210. 2 Council of Europe, Explanatory Report to the Council of Europe Convention on Preventing and Combating Violence against Women and Domestic Violence (2011) (‘Explanatory Report’). 3 Ibid. at para 29. The Preamble of the first three drafts of the Convention expressly mentioned the growing body of case law of the ECtHR on violence against women as setting important standards. The drafters then moved this wording to the Explanatory Report. 4 See, for example, Šimonović, ‘Global and Regional Standards on Violence Against Women: The Evolution and Synergy of the CEDAW and Istanbul Conventions’ (2014) 36 Human Rights Quarterly 590; Acar and Popa, ‘From Feminist Legal Project to Groundbreaking Regional Treaty: The Making of the Council of Europe Convention on Preventing and Combating Violence against Women and Domestic Violence’ (2016) 3 European Journal of Human Rights 287. 5 According to Article 32 of the Vienna Convention on the Law of Treaties 1969, 1155 UNTS 33, the preparatory work of a treaty constitutes a supplementary means of interpretation. Its actual role remains debated, see, for example, Mortenson, ‘The Travaux of Travaux: Is the Vienna Convention Hostile to Drafting History?’ (2013) 107 American Journal of International Law 780. 6 The Compilation of Comments on the Draft Convention on Preventing and Combating Violence against Women and Domestic Violence, 5 April 2010, CAHVIO (2010) 5 is cited as ‘Compilation 1’, and Compilation of reservations and comments, 28 October 2010, CAHVIO (2010) 19 as ‘Compilation 2’. The Draft Convention on preventing and combating violence against women and domestic violence, 15 October 2009, CAHVIO (2009) 32 prov; Second Draft Convention on preventing and combating violence against women and domestic violence, 18 May 2010, CAHVIO (2009) 32 rev; Third Draft Convention on preventing and combating violence against women and domestic violence, 28 October 2010, CAHVIO (2010) 17 rev and Draft Convention on preventing and combating violence against women and domestic violence, 16 December 2010, CAHVIO (2010) 26 FIN are cited as ‘Draft Convention’, ‘Second Draft Convention’, ‘Third Draft Convention’ and ‘Fourth Draft Convention’, respectively. 7 Its definition is discussed below in Section 2.A. 8 For example, Begikhani, Gill and Hague, Honour-Based Violence: Experiences and Counter-Strategies in Iraqi Kurdistan and the UK Kurdish Dispora (2016) 4. Their examples of different forms of honour-related violence include physical violence, assaults, maiming, enforced self-immolation, forced suicide, starvation, poisoning, curtailment of liberty and education and various forms of coercion such as removal of children and forced hymen repair. 9 United Nations Commission on the Status of Women, Ending Female Genital Mutilation, Report of the Secretary General, E/CN.6/2012/8, 5 December 2011, at para 24. FGM can be undertaken for reasons of honour, and also on other grounds. 10 The concepts of primary, secondary and tertiary prevention are discussed in the text accompanying infra n 48. 11 For example, Gill, ‘Introduction: “Honour” and “Honour-Based” Violence: Challenging Common Assumptions’ in Gill, Strange and Roberts (eds), ‘Honour’ Killing and Violence: Theory, Policy and Practice (2014) 1 at 8. 12 See supra n 3. 13 1979, 1249 UNTS 13. See Šimonović, supra n 4 at 605. 14 Explanatory Report, supra n 2 at para 110. 15 Article 5(2) Istanbul Convention. 16 Explanatory Report, supra n 2 at para 110. 17 See Acar and Popa, supra n 4 at 287. 18 Compilation 1 at 44, referring to the judgments in Opuz v Turkey Application No 33401/02, Merits and Just Satisfaction, 9 June 2009 at paras 128–129 and Osman v United Kingdom Application No 23452/94, Merits and Just Satisfaction, 28 October 1998. As a result of the accompanying request to the Secretariat to clarify the concept of due diligence, a study was commissioned by Christine Chinkin, see CHAVIO, The Duty of Due Diligence (CHAVIO, 2010) 7. 19 Chinkin, ibid. Also the draft update of CEDAW General Comment No 19 stresses that failure by the State to take all appropriate measures to prevent acts of gender-based violence of which authorities are or should be aware constitute a human rights violation: see General Recommendation No 19: Violence against Women. Addendum, CEDAW/C/GC/19/Add.1, 28 July 2016, at para 13(b)(ii). 20 Article 2(1) Istanbul Convention. 21 Article 3(a) Istanbul Convention. 22 Grans, ‘A Right Not to Be Left Alone – Utilising the Right to Private Life to Prevent Honour-related Violence’ (2016) 85 Nordic Journal of International Law 169 at 170. A further discussion of the definition of honour-related violence and whether it should simply be regarded as violence against women can be found in Gill, supra n 11 at 4 and Grans, ’The State Obligation to Prevent Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment: The Case of Honour-Related Violence’ (2015) 15 Human Rights Law Review 695 at 697. 23 On the strong standing of norms regarding honour, see, for example, Begikhani, Gill and Hague, supra n 8 at 28. 24 Article 3(f) Istanbul Convention clarifies that girls under the age of 18 years are included in the term ‘women’. 25 Explanatory Report, supra n 2 at para 37. Meanwhile, the Second Draft Convention originally encompassed all victims of domestic violence, thus including boys and men. 26 This intention was expressed by several delegations: see Compilation 1 at 8, 15, 25, 28, 35. 27 Compilation 1 at 25. 28 Ibid. at 8, 43. 29 Here, the term ‘incited’ is not used in a criminal law sense but to designate that violence is demanded (or perceived by the perpetrators as being demanded) by the community. 30 Compilation 1 at 143–144. See also infra n 34 . 31 Compare Peroni, ‘Violence Against Migrant Women: The Istanbul Convention Through a Postcolonial Feminist Lens’ (2016) 24 Feminist Legal Studies 49 at 50, who finds it positive that honour-related violence is not singled out for separate attention. 32 There was, however, a suggestion to include a definition of honour-related violence as follows: ‘Acts of violence against women including threats and acts of violence that lead to suicide, which are committed in the name of honour [sic], understood as acts of violence committed in order to prevent or punish a woman’s suspected, perceived or actual transgression of cultural, religious, social or traditional norms or customs of appropriate behaviour.’ See Compilation 2 at 55. 33 See, for example, Begikhani, Gill and Hague, supra n 8 at 56. 34 This requirement was introduced in the Second Draft Convention as Article 30 bis and remained in the Third Draft Convention, but was thereafter removed. One could argue that incitement to suicide must be prevented as a form of serious psychological violence. There remained doubts during the drafting process whether national criminal legislation should cover such acts: see Compilation 2 at 54. 35 This formulation remained until the Third Draft Convention but thereafter the word ’significant’ was removed. It should be kept in mind that Article 19 of the CRC prohibits all forms of violence against children. 36 Compilation 1 at 37. The prohibition of discrimination is contained in Article 4(3) Istanbul Convention. 37 Ibid. at 93 reflects that there was awareness of this issue among the delegates. 38 Explanatory Report, supra n 2 at para 89. 39 Ibid. 40 See Council of Europe, Elements for discussion, CAHVIO (2009) 3 at para 33, which lists forms of domestic violence separately from forms of honour-related violence. This listing is, however, not entirely consistent as it also mentions traditional practices harmful to women separately from honour-related violence and FGM. The issue of the need for a common definition of family members was raised also during the drafting process but it was not included in any version of the draft convention. See Compilation 1 at 27, 29. 41 The purposes of the Convention are listed in Article 1(1). States may opt to include violence against boys and men as well. 42 Council of Europe, Questionnaire on legislative and other measures giving effect to the provisions of the Council of Europe Convention on Preventing and Combating Violence against Women and Domestic Violence (Istanbul Convention), GREVIO/Inf(2016)1 (11 March 2016). 43 GREVIO/Inf(2016)2 (Austria) particularly at 26–7; GREVIO/Inf(2016)3 (Monaco); GREVIO/Inf(2017)1 (Albania); and GREVIO/Inf(2017)2 (Denmark). 44 Compare Walby, Ensuring Data Collection and Research on Violence against Women and Domestic Violence: Article 11 of the Istanbul Convention, A collection of papers on the Council of Europe Convention on preventing and combating violence against women and domestic violence (2016) at 8, which also mentions reporting on crimes committed in the name of honour. In the first two Evaluation Reports published, GREVIO did not yet demand that reports show a fuller picture of the prevalence and prevention of honour-related violence by including other forms of honour-related violence than FGM and forced marriages: see GREVIO/Inf (2017) 3 (Monaco) and GREVIO/Inf (2017) 4 (Austria). 45 Valiulienė v Lithuania Application No 33234/07, Merits, 26 March 2013 at Concurring Opinion of Judge Pinto de Albuquerque, para 30. Compare the majority judgment in Đorđević v Croatia Application No 41526/10, Merits, 24 July 2012 at para 139. 46 Women against Violence Europe (WAVE), ‘WAVE Report 2015 on the Role of Specialist Women’s Support Services in Europe’ (WAVE 2015) at 65, available at: wave-network.org [last accessed 10 December 2017]. 47 See, for example, Cook and Cusack, Gender Stereotyping: Transnational Legal Perspectives (2009) at 73. 48 Article 1 Istanbul Convention. 49 Article 4(2) Istanbul Convention. 50 1989, 1577 UNTS 3. 51 See CRC General Comment No 13: The right of the child to freedom from all forms of violence (2011), CRC/C/GC/13 at para 17. 52 1950, ETS 5. 53 Application No 3564/11, Merits, 28 May 2013. 54 The Court stated (at para 74) that the children’s psychological well-being had been adversely affected, without discussing severity. However, M and M v Croatia Application No 10161/13, Merits, 3 September 2015 at para 135, demonstrates that the Court applies an assessment of severity of ill-treatment in respect of Article 3 ECHR, but of course the age of the victim is relevant to this assessment. 55 This test originates from the Osman case, and originally concerned a risk to the life of an identified individual: see Osman v United Kingdom, supra n 18 at para 116. It was thereafter introduced by the Court also in cases focusing on Articles 3 and 8 ECHR. The expression ‘risk of damage to life or limb’ was adopted in Denis Vasilyev v Russia Application No 32704/04, Merits, 17 December 2009 at para 115. 56 Costello-Roberts v United Kingdom Application No 13134/87, Merits, 25 March 1993 at para 36. 57 Valiulienė v Lithuania, supra n 45 at Concurring Opinion of Judge Pinto de Albuquerque. 58 For example, A v Croatia Application No 55164/08, Merits, 14 October 2010 at paras 75–80; Hajduová v Slovakia Application No 2660/03, Merits, 30 November 2010 at para 50 (where the danger of future violence and threats was sufficient to engage the positive obligations of the State). But in, for example, Opuz, supra n 18 at para 129, the Court stated that for a positive obligation to arise, a real and immediate risk must exist. 59 Ebert and Sijniensky, ‘Preventing Violations of the Right to Life in the European and the Inter-American Human Rights Systems: From the Osman Test to a Coherent Doctrine on Risk Prevention?’ (2015) 15 Human Rights Law Review 343 at 360. They find that there are also other limitations to the application of the Osman Test, including in relation to risks caused by social structures. 60 Renzulli, ‘A Critical Reflection on the Conceptual and Legal Foundations of the Duty to Prevent Torture’ (2016) 20 International Journal of Human Rights 1244 at 1245–6. 61 Article 51 Istanbul Convention. The fact that assessment of the seriousness of the situation is mentioned separately from the lethality risk implies that this is a separate assessment. Similarly, see Compilation 1 at 169–70. 62 A common tool to identify individuals at high risk is the Multi-Agency Risk Assessment Conference (MARAC) checklist: see Assessment of Risk for Honour Based Violence (PATRIARCH) has been developed particularly for honour-related violence. 63 Explanatory Report, supra n 2 at para 63. 64 The main international case law supporting the existence of an obligation to undertake similar measures is found in the Inter-American human rights system: see, for example, Case 12.051, Maria da Penha v Brazil Commission Report No 54/01 (2001) at para 56. 65 Explanatory Report, supra n 2 at para 59. 66 See, for example, Bevacqua and S v Bulgaria Application No 71127/01, Merits, 12 June 2008 at para 64. For a discussion of the ECtHR’s effectiveness requirement, see Rietiker, ‘The Principle of “Effectiveness” in the Recent Case Law of the European Court of Human Rights: Its Different Dimensions and Its Consistency with Public International Law – No Need for the Concept of Treaty Sui Generis’ (2010) 79 Nordic Journal of Human Rights 245. 67 Explanatory Report, supra n 2 at para 91. 68 There is no reflection of such a discussion in Compilation 1 or Compilation 2. 69 The drafters did discuss the need for particular prevention measures for women subjected to violence which is based on specific social norms within their own community, referring to Parliamentary Assembly Recommendation 1891 (20 November 2009): see Compilation 1 at 60–1. While there is no reflection of this in the text of the Convention or the Explanatory Report, this discussion of individual preventive measures is important to keep in mind in interpreting the scope of the preventive obligations of States. 70 Heisecke, Raising Awareness of Violence against Women: Article 13 of the Istanbul Convention, A Collection of Papers on the Council of Europe Convention on Preventing and Combating Violence against Women and Domestic Violence (2014) at 20. 71 Hester and Lilley, Preventing violence against women: Article 12 of the Istanbul Convention. A Collection of Papers on the Council of Europe Convention on Preventing and Combating Violence against Women and Domestic Violence (2014) at 13. The document suggests factors that domestic authorities should take into account in designing preventive measures. 72 World Health Organisation, Eliminating Female Genital Mutilation: An Interagency Statement, OHCHR, UNAIDS, UNDP, UNECA, UNESCO, UNFPA, UNHCR, UNICEF, UNIFEM, WHO (2008) at 13. 73 Hester and Lilley, supra n 71 at 28. 74 In terms of prevention, the ECtHR has mostly focused on effective deterrence, see, for example, Opuz, supra n 18 at paras 168-169, 172 and 176. Human Rights Committee Draft General Comment No 36, Article 6: Right to Life, CCPR/C/GC/R.36/Rev.2 (2015), First Reading in July 2017, at 30, see: www.ohchr.org/EN/HRBodies/CCPR/Pages/GC36Article6Righttolife.aspx [last accessed 8 December 2017], seems to go further than this. It notes that the duty to protect life implies an obligation to take appropriate measures to address the general conditions in society that may eventually give rise to direct threats to life or prevent individuals from enjoying their right to life with dignity. It requires, for example, detailed plans to promote education on non-violence and for campaigns to raise awareness of domestic violence. 75 Article 10(1) Istanbul Convention. 76 Compilation 1 at 51–54. 77 A similar obligation to allocate appropriate funding to prevention is also included in the draft update of CEDAW General Recommendation No 19, supra n 19 at 12. 78 Articles 23 and 24 Istanbul Convention. 79 Article 13(2) Istanbul Convention. 80 See Explanatory Report, supra n 2 at para 135. 81 Ibid. at para 99. 82 Promotion of the collection of data and research on violence against women and domestic violence was even mentioned among the purposes of the Convention in Article 1 in the First Draft Convention. The importance of ensuring that preventive measures are effective through basing them on reliable data was noted, for example, in Compilation 1 at 64. 83 The draft update of CEDAW General Comment No 19, supra n 19 at 12, similarly requires the collection, analysis and publication of data on gender-based violence disaggregated by type of violence. 84 Article 18(5) Istanbul Convention. In its reservation, Poland inter alia states that it does not grant consular protection to nationals of the host State. Of course, nowadays, international law does not bar a State from offering protection to a dual national who is also a national of the host State, when the former is the State of effective nationality: see, for example, Forcese, ‘The Capacity to Protect: Diplomatic Protection of Dual Nationals in the “War on Terror”’ (2006) 17 European Journal of International Law 369 at 389. 85 Explanatory Report, supra n 2 at para 155. The preparatory materials indicate that the view was expressed that child marriages as such would fall under forced marriages: see Compilation 1 at 135; 138; 139 and Compilation 2 at 51. Meanwhile, Article 37 requires only that forcing a child or an adult into marriage be criminalized. Some States allow minors to marry with a special permission, an arrangement the Convention does not seem to exclude. Another problematic issue, namely that of differentiating between FGM of adults and esthetical genital surgery, was also raised during the drafting process. Some of the drafters appeared to be in favour of excluding at least the latter form of operations from the scope of the Convention, see Compilation 1 at 139–143. The Explanatory Report does not discuss the distinction. 86 Article 33 Istanbul Convention. 87 Explanatory Report, supra n 2 at para 181. 88 Article 78(3) Istanbul Convention. 89 Article 41(1) Istanbul Convention. 90 Article 42(2) Istanbul Convention. 91 Explanatory Report, supra n 2 at para 218. 92 For a similar conclusion on gender-based violence in general, see Krizsan and Pap, Implementing a Comprehensive and Co-ordinated approach: An Assessment of Poland’s Response To Prevent and Combat Gender-based Violence (2016) at 10. 93 Compilation 2 at 51. 94 Explanatory Report, supra n 2 at para 280. 95 Ibid. at para 279. This is also clear from the preparatory materials: see Compilation 1 at 177–8. 96 Article 78(2) Istanbul Convention and Explanatory Report, supra n 2 at para 281. 97 Several delegations would have preferred that only physical violence committed against the person’s will was criminalised: see Compilation 2 at 49. However, in some circumstances this would already follow from the case law of the ECtHR, see, for example, KA and AD v Belgium Applications Nos 42758/98 and 45558/99, Merits and Just Satisfaction, 17 February 2005 at para 87. See also the Explanatory Report, supra n 2 at para 190. 98 Usal Kanzler, Legislative review of Bulgarian criminal law in light of the standards established by the Council of Europe Convention on Preventing and Combating Violence against Women and Domestic Violence (2016) at 34. See also Explanatory Report, supra n 2 at para 216 (‘other form of personal reason’). 99 Article 46 of the Istanbul Convention; Explanatory Report, supra n 2 at para 235. 100 Some examples of such measures are listed in the Explanatory Report, ibid. at para 258. 101 For example, Opuz v Turkey, supra n 18 at paras 128 and 148, where the Court suggested that the authorities could have issued an injunction banning the person posing a threat from contacting, communicating with or approaching the person under threat. 102 Notably Đorđević v Croatia, supra n 45 at para 148, discussed further below. 103 Explanatory Report, supra n 2 at para 104. 104 See Section 2.B. above. 105 Support to families is often framed as ensuring the right to family life of minor children and their parents. A discussion of whether the right to family life should be interpreted as encompassing also the relations between adults is outside the scope of the present article. 106 Compare Compilation 1 at 94, where a delegation pointed out that some groups require specific strategies to ensure that they have access to and use specialist support services. 107 Denis Vasilyev v Russia, supra n 55 at para 115. The case concerned a violent robbery in a public place, but the Court’s statement could arguably describe the due diligence obligation also in cases of violence within the home. 108 See for example, Dolhamre v Sweden Application No 67/04, Merits, 8 June 2010 at para 116. 109 For example, M and M v Croatia, supra n 54 at para 136; Bevacqua and S v Bulgaria, supra n 66 at paras 64–65. 110 See supra text accompanying n 65. 111 Council of Europe Directorate General of Human and Legal Affairs, Ad Hoc Committee on Preventing and Combating Violence against Women and Domestic Violence (CAHVIO) - Elements of Discussion, CAHVIO (2009) 3, 16 March 2009, at para 18. 112 Explanatory Report, supra n 2 at para 116. In accordance with its wording, Article 18(3) applies to measures undertaken under chapter IV, thus not preventive measures under chapter III, but this author argues that a similar requirement to take the wider social environment into account could be regarded as being implicit in the duty laid down in Article 12(3) to take into account the specific needs of persons made vulnerable by particular circumstances. 113 Such emergency barring orders are foreseen by Article 52 Istanbul Convention. See also Explanatory Report, supra n 2 at para 264. 114 During the drafting of the Istanbul Convention, some delegations recognised the need for discussions with individual families on the use of violence, although this is not reflected in the Istanbul Convention or the Explanatory Report: see Compilation 1 at 76, 83. 115 See supra n 45 at para 148. 116 Article 9(3)(b) Directive 2012/29/EU of the European Parliament and of the Council of 25 October 2012 establishing minimum standards on the rights, support and protection of victims of crime, OJ EU L 315/57, similarly recognises that victims of gender-based violence and their children often require special support and protection because of the high risk of secondary and repeat victimisation, of intimidation and of retaliation connected with such violence. 117 In drafting Article 28, the drafters were concerned that there should not be a duty for professionals to report violence without the consent of persons who are of adult age or legally capable, but perhaps for children and persons who are legally incapable. There was a proposal to add a paragraph to this effect, which was not taken up in the Draft Convention: see Compilation 1 at 97–8. The requirement to report violence against children and legally incapable person may, however, be implicit in the general obligation to prevent violence with due diligence. The Explanatory Report, supra n 2 at para 148, only adds that the term ’under appropriate conditions’ would enable States to limit the application of Article 28 to cases where the victim consents, unless the victim is a minor or is unable to protect her or himself due to physical or mental disabilities. 118 Compilation 1 at 7–8, 21. 119 Ibid. at 7–8, 35. 120 However, also in relation to protection measures, the fact that each type of violence requires a different kind of support and protection is recognised: see supra n 80. 121 See cases cited supra n 109. © The Author(s) . Published by Oxford University Press. All rights reserved. For Permissions, please email: email@example.com
Human Rights Law Review – Oxford University Press
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.
All for just $49/month
Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.
Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.
It’s easy to organize your research with our built-in tools.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera