2.1. Harm-based and Wrong-based Approaches to Victimhood

Victimisation represents the victim-centred perception of gross violations of international human rights and humanitarian law.Footnote ³ In other words, it reflects the commission of an international crime from the perspective of victims and, as such, it rearranges the dominant approach to the crime phenomenon, which consists of picturing the crime through the lens of its commission,Footnote ⁴ and a subsequent attribution of such criminal (in)actions to an offender.Footnote ⁵ The concept of victimhood, in turn, reflects the status of being a (crime) victim, and takes either a naturalistic (harm-based) or a normative (wrong-based) form.Footnote ⁶ Depending on which model of victimhood is applied, not only the contours of direct victimhood but also the normative content and boundaries of indirect victimhood will change accordingly. However, before presenting the use of these models in relation to gross violations of international human rights and humanitarian law, it is useful to start with an explanation of why they should not be treated as simply different methods of expressing the same normative content of victimhood.

The harm-based model of victimhood assumes, as its name suggests, that what matters is whether anyone has been harmed by an action or inaction of a prima facie criminal character. This model thus requires there to be a nexus between such (in)actions and negative effects (results or consequences – see below) that stem from them. It remains immaterial, however, which interests or goods become violated by the said criminal behaviour. As long as there is a causal connection between the presumed source of victimisation and its various effects (harm), a person who has sustained the latter will be recognised as a victim. The harm-based model accepts virtually every crime to be a source of diverse types of harm that may be sustained by various people even if they are not immediate victims against whose goods specific criminal acts have been directed. One may thus be a direct victim of a crime which does not violate their individual, specifically protected interests, but merely exerts some negative effects (harm) onto them.

An approach that revolves around interests, goods and rights is, in turn, depicted by the wrong-based model of victimhood, which probes into the protective sphere of every type of crime. For example, one becomes a direct victim of the crime of murder only when one is unlawfully killed in so far as the crime of murder is prohibited in order to protect human beings’ right to life. An analogous reasoning can be employed in examining other crimes (theft, pillage, rape), although some offences are more difficult to pinpoint from that victim-centred perspective (offences against the state). In terms of victimhood, what the wrong-based model therefore submits is that it is only those subjects against whose protected interests an act has been directed who can be recognised as direct victims (property rights of an owner – theft). Others, in turn, even if they have also sustained some harm (XYZ's loss of income as a result of the theft of ABC's car) cannot be treated in this way as their violated goods or interests are not protected by the definition of the offence in question (theft – loss of income by a third party).

The wrong-based model of victimhood thus creates a strong normative link between each unlawful act and its direct victims, and allows one to identify them even before an offence is committed through analysis of its definition.Footnote ⁷ One thing that remains debatable in relation to the wrong-based model, however, is whether it pursues what may be called a constrained or an extended model of victimisation. The former assumes that one may be considered a victim only if the crime has been committed and one's interests have been effectively infringed (right to life – death). By contrast, the latter model does not deny this dimension but supplements it with an additional protective sphere against acts which do not violate but merely endanger certain protected interests. Accordingly, one may become a victim even if the violation (result – death) does not occur. It is sufficient that one's protected interests have been put at risk by a criminal act (victim of an attempted murder).Footnote ⁸

A comparative analysis of both models of victimhood justifies the conclusion that while they overlap to some extent, each focuses on different aspects of victimisation. The wrong-based model considers only normative criteria, which are either expressed explicitly (death as a normative result) or not (the right to sexual autonomy protected under the laws prohibiting sexual violence).Footnote ⁹ It is not concerned with more remote consequences. The harm-based model, in turn, is premised on both immediate and distant effects (types of harm) which may often be numerous and not easily identifiable.

Some conceptual assistance in this respect, however, may be found in the distinction between conduct and result crimes. The latter category, as its name suggests, points to specific results in definitions of crime. Result crimes, such as murder, may thus be committed only when the requisite result (death) actually occurs. By contrast, conduct crimes do not introduce such conditions, which means that they may be committed regardless of their negative effects.Footnote ¹⁰

However, results (harm 1) do not exhaust the whole category of negative effects of victimisation; for, apart from results, crimes may also bring consequences (harm 2). According to the adopted terminology, effects of victimisation may thus take the form of results or consequences. What distinguishes consequences from results is that only the latter category takes a normative form by being mentioned explicitly in definitions of specific offences (death as a result of murder). Results, therefore, characterise only result crimes. Consequences, on the other hand, may arise in relation to both result and conduct crimes in that they are not defined in their respective definitions but may arise following their commission.

This distinction, arguably, is useful for when the concept of victimhood is wrong based, consequences (such as psychological harm resulting from rape) will remain immaterial for determining who may be a direct victim.Footnote ¹¹ Results, on the other hand, will always have to be considered by a court of law as they must be proven for the crime to have occurred and, given that they specify the protective sphere of each crime (murder – death – right to life), are necessary for the determination of direct victimhood.Footnote ¹² By contrast, and as regards the harm-based model of victimhood, there is little difference between results and consequences as they both provide sufficient grounds for the recognition of one's direct and indirect victimhood. The harm-based model of victimhood is therefore broader in this respect.

It does not follow, however, that the harm-based model is broader in every case. For, as mentioned above, the wrong-based model may also recognise an endangerment of certain protected interests as the basis of direct victimhood. In such a case, neither results nor consequences need to be corroborated for an applicant to be accepted as a direct victim. It suffices that their interests have been endangered by the criminal behaviour of an offender. Although it cannot be excluded that the victim applicant will be able to prove to have suffered, for instance, psychological harm as a result of an attempted crime,Footnote ¹³ in practice this seems to be less probable mainly for evidentiary reasons. The wrong-based model thus appears to be more effective in safeguarding such interests by creating an additional sphere of protection around the core protected value (life – (1) attempted murder – (2) murder).

Furthermore, it should also be stated that neither of the two models of victimhood is flawless. The harm-based approach suffers from a legal under-specification by leaving broad discretionary powers to the judges as far as identification of relevant types of harm is concerned (see Section 6.3). It thus moves the whole burden of determining the scope of victimhood to evidentiary standards and causal requirements. Moreover, the harm-based model makes it difficult to distinguish between immediate and remote victims pursuant to the causal distance between criminal acts and the consequences sustained by them. As regards the wrong-based model, its main problem lies in identifying and framing interests and goods protected by specific offences (for instance, the crime of aggression). At least in some configurations, this exercise may even provide arguments for questioning whether any abstract (individual) victimhood can be derived from such offences at all.

2.2. Harm-based and Wrong-based Models of Victimhood in International Law

According to the 1985 Declaration of Basic Principles of Justice for Victims of Crime and Abuse of Power,Footnote ¹⁴ victims are:

Apart from direct victims, the Declaration recognises the concept of indirect victimhood, stipulating that ‘the term “victim” also includes, where appropriate, the immediate family or dependants of the direct victim and persons who have suffered harm in intervening to assist victims in distress or to prevent victimization’.

Twenty years later, the Basic Principles and Guidelines on the Right to a Remedy and Reparation for Victims of Gross Violations of International Human Rights Law and Serious Violations of International Humanitarian Law were adopted.Footnote ¹⁵ Although they feature a similar definition of victimhood in their provisions and refer in a similar fashion to family members as potential victims, in comparison with the Declaration they focus on a more specific source of victimisation, namely an infringement of international human rights law and international humanitarian law. Accordingly, victims are defined as:

These broad and internally varied definitions absorb both types of harm (injury, suffering, loss) and wrongs (impairment of rights) as grounds of victimhood.Footnote ¹⁶ As such, they combine two previously discussed theoretical models of victimhood. Moreover, they are not confined to direct injuries but provide that those individuals who were not directly targeted may also fall within the scope of legally recognised victimhood. Since such victims do not have to be in any way related to a direct victim (immediate family or dependants), the personal scope of victimhood derived from these documents is undoubtedly broad.

2.3. Punitive and Reparative Justice Approaches to Victimhood

International punitive justice advances predominantly retributivist and deterrent ideals.Footnote ¹⁷ Together with other, less relevant sentencing objectives, these two philosophical currents form bases for punishing guilty offenders. They evidence that punishment should be both backward- and forward-looking. From the victim-centred perspective, this signifies that punishment is intended to address both the wrongness and harmfulness of acts already committed (for instance, against specific victims – ‘real victimhood’) as well as to prevent other similar violations from taking place in the future (against potential victims – ‘future victimhood’). Even if the spectrum of retributivism and deterrence is not constrained solely to that victim-centred perspective but includes other actors (an offender, society and others), real and future victimhood are arguably present in both of these theories.

As demonstrated by the jurisprudence of IHCTs, just sentence is determined, among others, by reference to the gravity of crimes and aggravating circumstances, which may also comprise the consequences of international crimes such as harm suffered by direct and indirect victims.Footnote ¹⁸ With regard to the gravity criterion, it has been reasoned that some international crimes are inherently more serious than others.Footnote ¹⁹ In Al Mahdi, for instance, the ICC clarified that ‘not all crimes forming the grounds for a criminal conviction are necessarily of equivalent gravity and the Chamber has the duty to weigh each by distinguishing, for example, between those against persons and those targeting property’.Footnote ²⁰ The reason for this differentiation lies in an abstract victimhood that underpins most international crimes. Certain values, such as life (murder) or sexual autonomy (rape), are therefore more significant than other goods such as property (pillage).Footnote ²¹ Consequently, crimes directed at vital interests such as life will be considered particularly grave – both in abstracto and in concreto – justifying a more severe punitive response to their violation or endangerment.Footnote ²²

Indirect victimhood, however, is not expressed in definitions of international crimes. As such, it does not constitute another principle justifying international criminalisationFootnote ²³ and cannot be translated directly into an assessment of the in abstracto gravity of an offence. On the other hand, once confirmed with regard to the circumstances of a particular case, indirect victimhood may also modify the penalty imposed on a responsible actor. For instance, according to Rule 145(1)(c) of the Rules of Procedure and Evidence of the ICC (RPE ICC),Footnote ²⁴ in determining the sentence the Court shall ‘give consideration, inter alia, to the extent of the damage caused, in particular the harm caused to the victims and their families’.Footnote ²⁵ Similarly, the ICTY has held that the gravity of the offences is assessed, among others, through the prism of ‘the effect of the crimes on the broader targeted group, and the suffering inflicted on the victims’.Footnote ²⁶

Reparative justice is a more recent addition to the field of international criminal law.Footnote ²⁷ Its philosophical foundations have been displayed most explicitly in provisions and the case law of the ECCC and the ICC. Rule 23(1)quinquies of the Internal Rules of the ECCC (IR ECCC)Footnote ²⁸ stipulates, for instance, that reparations awarded to Civil Parties ‘acknowledge the harm suffered by Civil Parties as a result of the commission of the crimes for which an Accused’ has been convicted and ‘provide benefits to the Civil Parties’. In the legal regime of the ICC reparations are intended to fulfil several functions as well. To start with ‘they oblige those responsible for serious crimes to repair the harm they have caused and enable the Court to ensure that offenders account for their acts’.Footnote ²⁹ Moreover, they aim to ‘relieve the suffering caused by serious crimes, afford justice to the victims by addressing the consequences of the wrongful acts committed by the convicted person, deter future violations, and enable the victims to recover their dignity’.Footnote ³⁰ However, even if specific forms and modalities of affording reparative justice to victims will differ given the circumstances of specific cases, it has been explicitly acknowledged that ‘[a] victim-centred approach shall guide the reparations proceedings, and the Court should strive to ensure that reparations are prompt and meaningful to the victims’.Footnote ³¹ By its nature, reparative justice is therefore focused predominantly on countering negative effects of victimisation. Its function, however, is not limited to direct victims and their harm as – at least in theory – it follows an ambitious vision that every type of harm stemming from international crimes and sustained by both direct and indirect victims should be recognised as reparable (critically, Section 6.3).

Despite their general importance, the above definitions of victimhood (Section 2.2), developed within the United Nations system, were not translated directly into the pluralistic system of international criminal law. The ensuing discrepancy spans not only definitional but also practical aspects of victim participation before IHCTs:Footnote ³² both direct and indirect victims are no longer merely anonymous and distant figures whose presence in international criminal justice remains limited to an indirect form (judicial reasoning in the field of punitive and reparative justice) or to the traditional role of witnesses in criminal proceedings.Footnote ³³ Some IHCTs have also afforded victims a separate standing of direct participant in a criminal trialFootnote ³⁴ and granted them specific rights (such as the right to reparationsFootnote ³⁵). To whom the victim label will be afforded before these tribunals is therefore not just of symbolic importance as it may also bring significant practical consequences for the system of international criminal justice.

In Sections 3, 4 and 5 the article presents three approaches to victimhood that can be derived from documents and case law of IHCTs. Moreover, it demonstrates how the concept of indirect victimhood premised on both family and situational links has been employed in the field of punitive and reparative justice.

3.1. Family Relations, Indirect Victimhood and Reparative Justice before the ICTY and the ICTR

The first model of victimhood was adopted by the ICTY and the ICTR (ad hoc tribunals). Although victims possessed a markedly weak position in the proceedings before these tribunals,Footnote ³⁶ the Rules of Procedure and Evidence (RPE ICTY/ICTR)Footnote ³⁷ did provide for a definition of victimhood. Accordingly, a victim was defined as ‘[a] person against whom a crime over which the Tribunal has jurisdiction has allegedly been committed’ (Rule 2 RPE ICTY/ICTR). This concise definition explicitly displayed a narrow conception of victimhood premised on wrongs as factors that delimit its scope (the wrong-based model). As such, it hinged on a normative nexus between specific international crimes and their victims. This, in turn, seems to have precluded any notion of indirect victimhood in so far as indirect victims are not persons against whom a criminal act is being committed. This feature characterises only direct victims.

The definition adopted by ad hoc tribunals thus followed the model of direct victimisation and victimhood. It was premised on the idea that for a person to be recognised as a victim, an offender must act against their protected interests. To give just two, rather obvious, examples: in the case of rape, the raped person would be considered the sole direct victim of this crime; in the case of murder, only a killed person would be a direct victim. Accordingly, for the recognition of victimhood it would not be sufficient that certain harm has stemmed from a crime if the said crime was not committed against an alleged victim.

The definition of victimhood applicable before the ICTY and the ICTR could not be translated directly into the sphere of reparative justice as neither of the two ad hoc tribunals was empowered to grant reparations to victims of international crimes. Instead, their legal frameworks provided for practically inoperative mechanisms of restitution and compensation for victims.Footnote ³⁸ Article 24(3) of the ICTY StatuteFootnote ³⁹ stipulated, for instance, that ‘[i]n addition to imprisonment, the Trial Chambers may order the return of any property and proceeds acquired by criminal conduct, including by means of duress, to their rightful owners’.Footnote ⁴⁰

More detailed regulations were included in the rules of ad hoc tribunals. Rule 105(D) RPE ICTY provided that ‘[s]hould the Trial Chamber be able to determine the rightful owner on the balance of probabilities, it shall order the restitution either of the property or the proceeds or make such other order as it may deem appropriate’. This provision, however, did not feature any explicit reference to victims as such. Nevertheless, one could argue that ‘rightful owners’ who have lost their property are in fact direct victims given that it is to their detriment that the crime attributed to an offender was committed.Footnote ⁴¹ On the other hand, restitution of this kind could not be granted to indirect victims as the only harm from which they could suffer would be secondary by its nature – that is, an injury sustained from the loss of property.

Apart from substantiating the said provision in relation to the restitution of property, the Rules also contained a separate provision entitled ‘Compensation to Victims’, which provided that ‘[t]he Registrar shall transmit to the competent authorities of the States concerned the judgment finding the accused guilty of a crime which has caused injury to a victim’ (Rule 106 RPE ICTY). Accordingly, victims could be compensated on the basis that they have suffered harm (injury) stemming from a crime attributed to an offender only if domestic courts issued such decisions. In spite of defining ‘victims’ in Rule 2 RPE ICTY/ICTR, however, the framework established for ad hoc tribunals did not specify whether indirect victims could also pursue their own compensatory actions before domestic courts on these bases.

3.2. Family Relations, Indirect Victimhood and Punitive Justice before the ICTY and the ICTR

Ad hoc tribunals, as already noted above, were equipped with the definition of victimhood premised on the wrong-based model of victimhood applicable solely to direct victims. This definition, however, was not considered explicitly by these tribunals in their sentencing judgments, which do not mention Rule 2 RPE ICTY/ICTR despite featuring numerous references to victims who could well be considered indirect under virtually all definitions of indirect victimhood. Thus, although it enriches the conceptual landscape of victimhood in international criminal law, the utility of the wrong-based definition of direct victimhood turned out to be minimal in the operation of ad hoc tribunals. In what follows, I demonstrate that the wrong-based definition of victimhood was, in fact, omitted in sentencing practices of ad hoc tribunals, which opted to adopt the harm-based model of victimhood. This allowed them to consider indirect victimhood in the process of determining penalties for international crimes.

The ICTY and the ICTR did not limit their sentencing reasoning to direct victims. Many cases completed before these tribunals confirm that harm sustained by people other than those directly victimised may also be relevant for sentencing purposes. In the case of Blaskić, for instance, the ICTY reasoned that ‘the effects of the crime on relatives of the immediate victims may be considered as relevant to the culpability of the offender and in determining a sentence’.Footnote ⁴² In Mrksić et al., in the course of assessing the reasoning of the Trial Chamber (TC), the Appeals Chamber contended that ‘in light of the murders of the prisoners of war, close family members had been left without their loved ones, and that in almost all cases the anguish and hurt of such tragedy had been aggravated by uncertainty about the fate which befell these victims’.Footnote ⁴³ In the same vein, in the more recent case of Mladić, the ICTY underlined that the gravity of the offence may be assessed by reference to ‘the overall impact of the crimes upon the victims and their families’.Footnote ⁴⁴

This initial compilation shows that the consequences of international crimes often are not limited to immediate direct victims and may also involve the suffering of remote direct and indirect victims.Footnote ⁴⁵ Accordingly, in the case of Jelisić the ICTY recognised harm suffered not only by immediate victims, but also by their families as well as witnesses to the crime who found themselves in a situation similar to those targeted – that is, ‘at the mercy’ of the accused.Footnote ⁴⁶ Related findings were entered in the cases of Mladić and Kayishema et al. before the ICTY and the ICTR respectively.Footnote ⁴⁷ In Galić, the Chamber referred to ‘the effect of the crimes on the broader targeted group’.Footnote ⁴⁸ Similarly, in Obrenović the Tribunal stated that crimes following the fall of Srebrenica had an impact ‘not only upon specific individuals, but also upon the entire Bosnian Muslim community’.Footnote ⁴⁹ In Krajišnik, the Chamber noted that collective victimisation to which the defendant contributed will be felt ‘in Bosnia-Herzegovina for decades, affecting hundreds of thousands of people’.Footnote ⁵⁰

On the other hand, some sentencing decisions provide more conflicting visions as far as the recognition of remote effects of victimisation is concerned. In Krnojelac, for instance, the TC rejected the Prosecutor's submission to be guided by the effects that the offence in question brought ‘on the relatives of immediate victims’.Footnote ⁵¹ Such an approach, it was contended, ‘would be unfair’ for determining the sentence.Footnote ⁵² Under the TC's approach, what should be seen as relevant in relation to sentencing is consideration of the consequences that a crime has caused only to those directly affected by it.Footnote ⁵³ For, as explained in the previously decided case of Kunarac et al., ‘consideration of the consequences of a crime upon the victim who is directly injured by it is … always relevant to the sentencing of the offender’.Footnote ⁵⁴ By contrast, harm suffered by indirect victims should not be normatively connected to the offender's culpability. It would be substantively unfair to attribute this to the convicted person and punish them more severely on such grounds.Footnote ⁵⁵

This reasoning raised controversies and was subsequently challenged before the Appeals Chamber of the ICTY. The Appeals Chamber adopted a critical assessment of the TC's restrictive approach, and explained that ‘even where no blood relationships have been established, a trier of fact would be right to presume that the accused knew that his victim did not live cut off from the world but had established bonds with others’.Footnote ⁵⁶ In other words, by inflicting such suffering on direct victims, offenders are deemed to be aware that they do not harm only those immediate victims, but exert an impact on other people's lives as well. This explanation has been followed by the Tribunal in subsequent cases.Footnote ⁵⁷ In Brdanin, for instance, the ICTY underscored that ‘the impact of the crimes … affected not only specific individuals but almost the entire Bosnian Muslim and Bosnian Croat communities in the ARK that ended up forcibly displaced’.Footnote ⁵⁸

4.1. Family Relations, Indirect Victimhood and Reparative Justice before the ICC

A novel definition of victimhood was introduced to international criminal law together with the codification of the RPE ICC.Footnote ⁵⁹ For Rule 85 RPE ICC, which codes a binary definition of victimhood, is premised on the notion of harm. According to Rule 85(a) RPE ICC, victims are ‘natural persons who have suffered harm as a result of the commission of any crime within the jurisdiction of the Court’. The second prong of this definition is expressed in Rule 85(b) RPE ICC, which stipulates that ‘[v]ictims may include organizations or institutions that have sustained direct harm to any of their property which is dedicated to religion, education, art or science or charitable purposes, and to their historic monuments, hospitals and other places and objects for humanitarian purposes’.

Considering that the victimhood of legal persons is constrained by the directness of harm, it does not provide for the recognition of indirect victimhood suffered by institutions and organisations other than those directly harmed by international crimes.Footnote ⁶⁰ Although this additional qualification of directness is mentioned in relation to harm, Rule 85(b) RPE ICC clearly pursues the wrong-based model of victimhood. The consequence of this legislative decision is that legal persons cannot be indirect victims of international crimes in proceedings before the ICC, and their victimhood will be recognised only in relation to some types of international crime (war crimes) which protect specific property interests.

Natural persons, on the other hand, may suffer from both direct and indirect (physical, psychological, monetary and other) types of harm as victims pursuant to Rule 85(a) RPE ICC. Negative effects of the offender's actions are therefore not constrained in the substantive sense. As long as they are capable of proving the nexus between the crime falling within the jurisdiction of the Court and their personal harm, they may be considered victims even if the said crime (source of their alleged victimisation) was not designed in such a way as to protect interests consequentially violated by the crime in question (for instance, psychological harm resulting from crimes against property). Thus, while Rule 85(a) RPE ICC follows the harm-based model (direct and indirect victimhood), Rule 85(b) RPE ICC espouses the wrong-based model of victimhood (direct victimhood). Viewed in its entirety, the ICC definition of victimhood is effectively premised on the mixed model of victimhood.

The link between family relations and victimhood has been examined extensively by the ICC. In the first case against Thomas Lubanga Dyilo, the Court openly admitted that:Footnote ⁶¹

Indirect victims, therefore, are those who suffer harm as a result of the harm suffered by direct victims. In the 2009 Decision on Indirect Victims, the ICC clarified the criteria that are applicable to the evaluation of an applicant with regard to victim status.Footnote ⁶² The question at stake concerned the status of those allegedly victimised by Thomas Lubanga's crimes.Footnote ⁶³ The Trial Chamber was asked to determine ‘whether applicants who are the victims of crimes committed by persons conscripted or enlisted whilst under the age of fifteen, or used to participate actively in hostilities, can be considered “indirect victims”’.Footnote ⁶⁴ The Court answered this question in the negative by reasoning that the concept of indirect victimhood is ‘restricted to those whose harm is linked to the harm of the affected children when the confirmed offences were committed, not those whose harm is linked to any subsequent conduct by the children, criminal or otherwise’.Footnote ⁶⁵ Accordingly, ‘the person attacked by a child soldier is not an indirect victim … because his or her loss is not linked to the harm inflicted on the child when the offence was committed’.Footnote ⁶⁶

A few years later, in the first decision to establish the principles of reparations, the ICC contended that indirect victims must possess ‘a close personal relationship’ with a direct victim, ‘for instance as exists between a child soldier and his or her parents’.Footnote ⁶⁷ Apart from that, and following its previous considerations, the TC added that those who have ‘suffered harm when helping or intervening on behalf of direct victims’ may also be considered indirect victims.Footnote ⁶⁸ In this particular case, it may be contended that the nexus between a direct and an indirect victim took a strictly situational form in so far as the connection between a direct and an indirect victim stemmed from the latter being involved in assisting a direct victim. This is not a requirement that is associated with a paradigmatic form of indirect victimhood premised on family relations (for example, parents learning about their child's victimisation ex post facto).

However, the scope of indirect victimhood and its bases were not set in stone by the TC in the case of Lubanga. On the contrary, they were further expounded by the Appeals Chamber, which identified four different types of harm that indirect victims may suffer, and which are remediable under Article 75 of the ICC Statute. The said list comprised (i) ‘psychological suffering experienced as a result of the sudden loss of a family member’; (ii) ‘material deprivation stemming from the loss of the family members’ contributions’; (iii) ‘loss, injury or damage suffered by the intervening person in attempting to prevent the child from being further harmed as a result of a relevant crime’; and (iv) ‘psychological and/or material sufferings as a result of aggressiveness on the part of former child soldiers relocated to their families and communities’.Footnote ⁶⁹ In the most recent decision in what has already become ‘the Lubanga saga’, the Court underlined that ‘the situation of indirect victims … must be addressed in an appropriate manner, again appreciating the difference in needs that such victims have, as they most likely require reparations that differ from those required for direct victims’ (here, child soldiers).Footnote ⁷⁰

The case of Lubanga was followed by the cases of Germain Katanga and Ahmad Al Mahdi, which also contained references to various dimensions of indirect victimhood. In Al Mahdi, some victim applicants claimed that as a result of attacks against the protected objects in Timbuktu (Mali), harm was sustained either by them personally or by their families.Footnote ⁷¹ It was also submitted that ‘moral harms’ resulting from deaths connected with the destruction of these buildings were suffered by other victims.Footnote ⁷² These applications were eventually rejected by the Court on account that such types of harm did not play ‘any part in the criminal plan’ of Ahmad Al Mahdi, and therefore they were not ‘sufficiently foreseeable’ to meet the requisite standard of normative causation.Footnote ⁷³

In Al Mahdi, the Court also addressed a rather distinctive form of victimisation of ‘those whose ancestors’ burial sites were damaged in the attacks (such as ‘descendants of the saints’)’, given that they have had ‘a different kind of emotional connection to the destroyed sites than the rest of the Timbuktu population’.Footnote ⁷⁴ It was confirmed that in order to properly address the mental pain and anguish of those victims, it would be justifiable to afford them individual reparations.Footnote ⁷⁵ The main reason for the recognition of those applicants’ suffering arguably lay in the specific connection that they had had with their relatives: it lay, in other words, in their family relations.Footnote ⁷⁶ It was because of this connection that the destruction of burial sites led to their legally relevant victimisation as opposed to the rest of the population of Timbuktu. This example attests to the fact that in the domain of reparations, it is not just the crime of murder but also other criminal acts directed at the deceased that may provide grounds for recognition of indirect victimhood.

The case of Katanga contains a more internally diverse analysis of the contours of victimhood in the sphere of reparative justice. To begin with, some victim-applicants claimed to have suffered harm ensuing from the destruction and pillaging of family property which belonged to their ancestors.Footnote ⁷⁷ The ICC TC did not consider itself competent to grant those reparations, however, because of the difficulty in deciding the terms of succession, as well as related evidentiary obstacles.Footnote ⁷⁸

The most explicit and thorough picture of indirect victimhood in the context of reparations was presented by the Court with regard to psychological harm. This is not surprising, of course, for even intuitively one may contend that psychological harm will constitute the most common denominator of indirect victimhood. Accordingly, in Katanga, some applicants submitted that they suffered harm as a result of the deaths of their relatives.Footnote ⁷⁹ In response to this submission, the TC explained that ‘[t]he harm caused to an indirect victim may include psychological suffering experienced as a result of the sudden loss of a family member … To that end, the applicant must show that he or she had a close personal relationship with the direct victim’.Footnote ⁸⁰

After establishing that the deaths of the applicants’ relatives resulted from the crimes of which Germain Katanga was convicted and ascertaining the veracity of the documents notifying the deaths of direct victims, the Chamber identified close personal relations between direct and alleged indirect victims. In a rather liberal fashion, it further decided that ‘it is of little consequence whether the relative was near or distant’.Footnote ⁸¹ A detailed analysis presented by the Court concluded with the statement that the psychological harm resulting from the death of a relative had been properly established.Footnote ⁸² Accordingly, indirect victims were granted the right to receive individual reparations for their harm.Footnote ⁸³ In its concluding findings the Court confirmed that 201 and 284 instances of psychological harm resulted from the death of, respectively, a near or a distant relative.Footnote ⁸⁴ Considering that the total number of applications received by the Court in these proceedings amounted to 341, the final number of recognised indirect victims showcases that the concept of indirect victimhood played a significant role in determining the scope of Germain Katanga's liability.Footnote ⁸⁵

Further, psychological harm was addressed by the Court in the context of sexual violence. The Chamber reasoned that the gravity of acts amounting to rape and sexual slavery is premised on two foundations: first, the gravity of these crimes stems from the nature of the acts themselves; second, it comprises psychological consequences for the victim which result from such criminal acts.Footnote ⁸⁶ The gravity, in other words, is premised on the wrongness of such conduct as well as on its harmfulness. More importantly for this analysis, however, the Court subsequently contended that physical pain compounded by psychological suffering ‘has repercussions on the life of the victim, but also on the lives of the victim's nearest and dearest’.Footnote ⁸⁷

Yet another form of indirect victimhood that was addressed by the ICC in the case of Katanga took the form of what the Court dubbed a sui generis harm – that is, among others, a loss of standard of living,Footnote ⁸⁸ which constitutes one form of damages that indirect victims may sustain according to some domestic legal systems.Footnote ⁸⁹ Victim-applicants put forward an allegation that:Footnote ⁹⁰

Because of insufficient evidentiary support, however, the Chamber decided not to award compensation on these bases, but instead underlined that the category of psychological harm connected with the attack on Bogoro, already recognised by the Court, also encompasses such sui generis harm.Footnote ⁹¹

The approach of the ICC TC towards indirect victimhood in Katanga was subsequently challenged by the defence counsel in the appellate proceedings. Interpreting the TC judgment, the Appeals Chamber (AC) assessed that the TC ‘presumed psychological harm in respect of all family members of direct victims of the attack on Bogoro’, which meant that the applicant ‘did not have to specifically prove the existence of the claimed psychological harm, even if that family member was distant’.Footnote ⁹² The presumption adopted by the TC was therefore straightforward: relatives, by definition, will suffer harm because of the loss of their next of kin. The AC further observed that what Article 75 of the ICC Statute and Rule 85 RPE ICC emphasise is ‘the requirement of the existence of harm rather than whether the indirect victim was a close or distant family member of the direct victim’.Footnote ⁹³ By rejecting Germain Katanga's submission as to the limitation of the personal scope of indirect victimhood, the AC seems to have accepted the robust and extensive interpretation of this concept.

Following the orders for reparations in the cases of Lubanga, Al Mahdi and Katanga, the latest decision pursuant to Article 75 ICC Statute was issued by the ICC in the case of Bosco Ntaganda, who was previously convicted of five counts of crimes against humanity and thirteen counts of war crimes.Footnote ⁹⁴ In terms of conditions that must be met by victims applying for reparations, the Court confirmed that ‘[n]atural persons may be direct or indirect victims, provided they suffered a personal, but not necessarily direct, harm’.Footnote ⁹⁵ In contrast to direct victims ‘whose harm is the result of the commission of a crime’,Footnote ⁹⁶ ‘[i]ndirect victims are those who suffer harm as a result of the harm suffered by the direct victims’.Footnote ⁹⁷ Following the ‘case of Lubanga, in Ntaganda the ICC identified four categories of indirect victims, namely (i) the family members of direct victims;’ (ii) ‘anyone who attempted to prevent the commission of one or more of the crimes under consideration’; (iii) ‘individuals who suffered harm when helping or intervening on behalf of direct victims;’ (iv) ‘other persons who suffered personal harm as a result of these offences’.Footnote ⁹⁸

The four groups of indirect victims enumerated by the Court have a general character. Consequently, the personal scope of indirect victimhood before the ICC remains largely unspecified. The fourth group shows that virtually every person capable of proving the nexus between their harm and that sustained by a direct victim may be recognised as an indirect victim. Moreover, the Chamber's reasoning in Ntaganda confirms that one person may have a dual status by being both a direct and an indirect victim. Such victims will thus be allowed to act on two separate grounds when seeking reparations for the harm suffered.Footnote ⁹⁹

Apart from direct and indirect victims, however, there are other people who become affected in different ways by international crimes. The Court admitted that ‘the inherent features of the crimes under the jurisdiction of the Court also result in mass victimization, affecting victims as members of families and entire communities’.Footnote ¹⁰⁰ This broader vision of victimhood is noticeable not only in the Court's reasoning; it is also reflected in reparation programmes that enable direct victims as well as ‘their families, and their communities to benefit from measures over an extended period of time’.Footnote ¹⁰¹

In the process of considering the scope of Bosco Ntaganda's liability, the Court accepted that indirect victims sustained various types of harm as a result of Ntaganda's crimes. The TC thus reaffirmed determinations previously entered by the AC in the case of Katanga, and emphasised that ‘the concept of indirect victims shall not lead to the discrimination of individuals on the basis of birth or marital status’.Footnote ¹⁰² The Court decided that children of female victims of rape or sexual slavery, persons who did not have a close personal relationship with a direct victim but who remained important in their lives, as well as those who witnessed the crime when it was committed may also be recognised as indirect victims and receive reparations.Footnote ¹⁰³ This determination attests to the fact that the concept of indirect victimhood in the ICC system is not constrained by strict formal criteria apart from the requirement, repeatedly underlined by the Court in its case law, that every indirect victim must be able to demonstrate having suffered harm as a result of the commission of a crime against a direct victim.

In specifying the harm sustained by indirect victims in Ntaganda, the Chamber identified six types of harm of such victims:Footnote ¹⁰⁴

What they demonstrate is an internal variety of victimisation of indirect victims that is not constrained to family members only. Moreover, they point to the practical difficulty that may arise in identification of specific negative consequences on the part of alleged indirect victims. An approach adopted by the Court to this end will obviously have a direct bearing on the number of people recognised as indirect victims. Tellingly, in the case of Ntaganda, the Chamber underscored that the Appointed Experts were not able to ascertain the exact number of indirect victims eligible for reparations in contrast to the number of direct victims, estimated at 3,500 victims at least.Footnote ¹⁰⁵ This is yet another indicator of the inherently imprecise nature of direct and indirect victimhood.

4.2. Family Relations, Indirect Victimhood and Punitive Justice before the ICC

The practice of affording certain weight to harm suffered by family members and other people affected by gross violations of international human rights and humanitarian law in sentencing decisions characterises not only the reasoning of ad hoc tribunals (Section 3.2), for already in the first case of Lubanga, the ICC – following Rule 145(1)(c) RPE ICC – made a similar finding by emphasising that crimes attributed to an offender were serious as they left an impact not just on direct victims (child soldiers) but also on other subjects (the international community, family).Footnote ¹⁰⁶ It was indicated in the case of Bemba that in so far as ‘murder deprives the direct victim of life’, it is ‘the ultimate harm’.Footnote ¹⁰⁷ However, harmful consequences do not end with a direct victim given that ‘[r]elatives and dependents left behind are not only deprived of the direct victim, an impact that cannot be underestimated, but may also be directly injured – physically and/or psychologically – as a result of the murder’.Footnote ¹⁰⁸ To support this determination, the TC referred in that case to the testimony of a witness who was present when his sister was murdered. In his words, ‘I saw the brain of my sister. I saw that as if an animal's skull had been hit’.Footnote ¹⁰⁹

Psychological harm of relatives of direct victims is probably the most common source of indirect victimhood. In Bemba, however, the Chamber noted that negative consequences may also take other forms. Jean-Pierre Bemba's (alleged) crimesFootnote ¹¹⁰ ‘rippled through the relevant communities. Owing to the prevailing chaotic and traumatic circumstances, family members of, and others with special bonds of affection to, some murder victims were deprived of the comforts that funeral services and burial rituals may provide in periods of grief’.Footnote ¹¹¹ For some victims ‘the impact of the murders was chronic and severe’.Footnote ¹¹² All of this led the ICC to rule that the crime of murder was of serious gravity in so far as ‘[t]he direct victims lost their lives. The indirect victims, in particular, family members, also suffered severe and lasting harm’.Footnote ¹¹³

Occasionally, however, less precise expressions employed by IHCTs make it difficult to pinpoint the genuine nature of victimhood (direct versus indirect) and its overall impact on sentencing. In the case of Al Mahdi, for instance, the ICC explained that the attacks on the protected objects appeared ‘to be of particular gravity as their destruction does not only affect the direct victims of the crimes, namely the faithful and inhabitants of Timbuktu, but also people throughout Mali and the international community’.Footnote ¹¹⁴ At first sight this conclusion may seem to suggest that crimes attributed to Ahmad Al Mahdi led to both direct and indirect victimhood. However, once one considers the nature of the destruction of cultural heritage, it becomes clear that all three groups (local population, people of Mali, international community) were, in fact, viewed by the Court as direct victims of this crime, although their victimisation differed considerably in terms of its character and distance.Footnote ¹¹⁵ For some it was proximate (local population); for others it was not (international community).

The Chamber's reasoning in Al Mahdi not only denoted a consequential way of thinking about victimisation, but also involved, even if only implicitly, an element of harm gradation. Harm, after all, may not only take different forms; it may also be of divergent intensity in abstracto and in concreto. Importantly, and without introducing any distinction between direct and indirect victims, the Court clarified that:Footnote ¹¹⁶

After Lubanga, Katanga and Al Mahdi, the idea that harmful effects of indirect victimisation should be examined for sentencing purposes was followed by the ICC in two most recent cases that resulted in the imposition of severe sentences of 30 (Bosco Ntaganda) and 25 (Dominic Ongwen) years’ imprisonment.Footnote ¹¹⁷

The Court confirmed in Ntaganda that, in the case of murder, ‘relatives and dependants left behind are deprived of a family member, and thereby of love and care, and, depending on the situation, of support, be it financial, physical, emotional, psychological, moral, or otherwise’.Footnote ¹¹⁸ Apart from close relatives of those killed, the Court acknowledged that other people may also be indirectly made victims by such offences. For instance, ‘some individuals who survived or witnessed these crimes still bear permanent scars’.Footnote ¹¹⁹ These findings led the Court to conclude that ‘[t]he murders … irreversibly impacted not only the direct victims but also those who witnessed them, and the direct victims’ family members and relatives left behind’.Footnote ¹²⁰ However, crimes against human life and security are not the only ones that may result in indirect victimisation. In relation to particularly gruesome sexual offences, the Court made note of the fact that ‘[s]ome of the effects were also experienced by the victims’ family members and communities’.Footnote ¹²¹

On the other hand, a noticeable absence of references to non-direct victims characterises the sentencing decision in the case of Ongwen. Indeed, only two crimes attributed to Dominic Ongwen were considered by the Chamber as sources not just of direct, but also of indirect victimisation. First, the ICC underscored that ‘it is important to pay sufficient attention also to the psychological harm done to the victims and their family members’ by the crime of enslavement.Footnote ¹²² Second, with regard to the crime of conscription of children under the age of 15, the Court remained ‘attentive to the emotional suffering that the abduction and integration into the LRA brought upon the direct victims, as well as their families, who were separated from their children and left without any information about them for a long time, if not indefinitely’.Footnote ¹²³ Similar findings, however, were not made by the Court in determining the sentence for other offences previously attributed to Dominic Ongwen. It could be argued, therefore, that the case of Ongwen departs from an approach adopted by the ICC in earlier decided cases in terms of the weight assigned to harm suffered by indirect victims for sentencing purposes.

5.1. Family Relations, Indirect Victimhood and Reparative Justice before the ECCC

The third way of understanding victimhood in the context of gross violations of international human rights and humanitarian law is represented by the framework established for the ECCC. It employs two different concepts of victims and victims participating in the capacity of civil parties to describe injured parties in the proceedings.Footnote ¹²⁴ Acting as an appellate court, the Supreme Court of Cambodia (SC/ECCC) explained that the definition of a civil party is not premised on a ‘formal designation of a specific class of persons, but the substantive criterion of an actual injury resulting as a direct consequence of the crime’.Footnote ¹²⁵ By adopting this broader perspective, the SC/ECCC pointed to the fact that:Footnote ¹²⁶

However, this model is not applicable before the ECCC, the framework of which operates on the basis of the distinction between victims and civil parties: although every civil party is most likely to be the crime victim given that a crime violates his or her rights, not every victim will also be a civil party.

In contrast to the ICC (Rule 85(a) RPE ICC), the SC/ECCC defines victims (but not civil parties) in substantive terms by drawing a normative nexus between specific protected interests and the crime committed. Accordingly, the word ‘victim’ refers in this legal regime to ‘a person whose rights were the object of the criminal attack in the act charged, in other words, “against whom the crimes were committed”’.Footnote ¹²⁷ As a result of this criterion, this definition resembles the definition applicable before ad hoc tribunals. To be treated as a victim by the ECCC, an applicant must prove that they have been wronged (but not necessarily harmfully wronged) by a criminal act. To acquire the status of a civil party, on the other hand, it is neither sufficient nor legally required that a crime has been committed against an applicant's protected interests.Footnote ¹²⁸ It is an injury sustained by an applicant as a result of an international crime that determines who may or may not be a civil party according to the above definition.Footnote ¹²⁹

All the same, it would be a mistake to claim that an injury is totally detached from the normative sphere. The SC/ECCC noted in this respect that an‘[i]njury is contingent, or more likely to come into being, where there is a violation of a right’.Footnote ¹³⁰ That said, the Court further added that ‘a violation of a right does not in itself always presuppose or produce injury’.Footnote ¹³¹ In effect, while civil parties will normally encompass crime victims, this will be possible only when victims are capable of proving that they continue to suffer harm (‘injury’) resulting from the crime charged at the time when the case is brought to trial. The approach of the SC/ECCC thus reflected the following correlation: while one may be a victim even if no harm can (any longer) be detected provided that there has been a violation of one's protected interests, for civil parties seeking compensation for the harm suffered it is essential that they are capable of demonstrating not the violation of their rights and interests as such, but injuries sustained as a direct consequence of the charged crime.Footnote ¹³² It may even be possible in some cases that no individual right has been violated, but an applicant will still be recognised as a civil party because they can prove an injury linked to the charged crime.Footnote ¹³³ Both terms – the ‘civil party’ and the ‘victim’ – may therefore overlap and coincide, but they should not be considered identical in every case.Footnote ¹³⁴

In relation to the direct/indirect victimhood distinction, the SC/ECCC further explained that while alleged direct victims ‘are only required to demonstrate the violation or endangerment of their rights’,Footnote ¹³⁵ the definition of indirect victimhood is more strictly tied to the concept of an injury.Footnote ¹³⁶ As indirect victims cannot claim that a crime has been committed against them (they are not victimised in that sense), their status, in practice, seems to resemble that of a civil party. The SC/ECCC adopted this idea by noting that ‘indirect victims encompass persons who actually suffered … injury, for example, as a result of the injury, whether temporary or permanent, of their loved ones’.Footnote ¹³⁷ It also further explained that ‘an indirect victim would need no less to have suffered injury as a direct consequence of the crimes committed against the direct victim(s)’.Footnote ¹³⁸ It begs the question, however, whether by mentioning ‘injuries of direct victims’, the SC/ECCC opted to restrict the scope of indirect victimhood only to those potential applicants whose own personal suffering comes from injuries of direct victims (harm of indirect victims – harmful wrongs of direct victims), or whether it would be sufficient if direct victims were wronged as long as indirect victims have sustained an injury because of that (harm of indirect victims – wrongs of direct victims).

The Supreme Court's reasoning presented above remains unclear in this respect for it seems to cover both correlations. However, given that in its further reasoning the Court reiterated that civil parties ‘also include indirect victims who personally suffered injury as a direct result of the crime committed against the direct victim’, it may be argued that the wrong-based correlation is the one applicable in proceedings before the SC/ECCC (harm of indirect victims – wrongs of direct victims).Footnote ¹³⁹ Therefore, one can be an indirect victim even if a direct victim is not afforded the status of civil party during an ongoing trial before the ECCC. What matters, in turn, is whether a crime has been committed against protected interests of a direct victim, and whether an indirect victim-civil party can prove to have suffered some injury because of that.

Arguably the most far-reaching part of the SC/ECCC's reasoning concerned the personal scope of indirect victimhood. In this respect the SC/ECCC made a decisive interpretative move by severing the link of family relations as a requisite nexus for establishing indirect victimhood. The Supreme Court stated:Footnote ¹⁴⁰

Nonetheless, the above determination should not be interpreted as a total rejection of the family-based notion of indirect victimhood in the ECCC system. Notably, in its subsequent remarks relating to various presumptions of an injury suffered by alleged indirect victims, the SC/ECCC noted that ‘bonds of affection and dependence are dynamics that usually exist among close family members’, and therefore it is correct to ‘relieve the class of immediate family from discharging the burden of proof of injury’.Footnote ¹⁴¹ Thus, while the Court decided not to preclude in advance any person from being recognised as an indirect victim, it nevertheless adopted the presumption of victimisation, which narrows the scope of indirect victimhood in relation to non-family members. This evidentiary differentiation provides grounds for a binary structure of indirect victimhood in terms of evidentiary conditions (family members versus others). Needless to say, those who do not belong to the immediate family of a direct victim will be required to meet more stringent criteria and a much higher threshold of proof than applicants who happen to be family members of a direct victim. The plausibility and persuasiveness of this approach is disputable (Section 6.3).

Further, in Case 002/01, the Chambers followed the testimonies of victims and concluded that ‘as a consequence of the crimes of which the Accused have been convicted, the Civil Parties and a very large number of additional victims, have suffered … psychological trauma and grief arising from the loss of family members or close relations’.Footnote ¹⁴² The Chambers thus reaffirmed that the scope of indirect victimhood may also include non-family relations. In Case 002/02, the ECCC found in turn that ‘families were divided and children separated from their parents’.Footnote ¹⁴³ This resulted in extreme physical, emotional and psychological harm sustained by both direct and indirect victims. The Chambers contended that ‘[t]ens of thousands of direct victims lost their lives. When victims’ families learned of the deaths of their loved ones, it caused them deep suffering’.Footnote ¹⁴⁴ Similarly, arrests and disappearances of direct victims compounded ‘a sense of loss for their families with the uncertainty of their loved one's fate’.Footnote ¹⁴⁵ The Chambers also underlined that those persons whose relatives were killed ‘at security centres suffered long-term emotional suffering, loneliness and material harm’.Footnote ¹⁴⁶ Moreover, they openly acknowledged that:Footnote ¹⁴⁷

These excerpts demonstrate that in the Chambers’ reasoning, the concept of indirect victimhood, despite its broader definitional scope, has been focused predominantly on family relations.Footnote ¹⁴⁸

5.2. Family Relations, Indirect Victimhood and Punitive Justice before the ECCC

As with the other tribunals examined in this article, negative consequences of international victimisation sustained by indirect victims have been viewed by the ECCC as relevant for determining the gravity of crimes. For instance, the SC/ECCC ruled that it is necessary to take note of ‘the impact of the crimes upon them [direct victims] and their relatives’ as well as the ‘outrage these crimes caused in victims, their families and relatives’.Footnote ¹⁴⁹ Further, it added that although the victimising events took place 30 years before the judgment, this fact alone ‘does not weaken the necessity of a high punishment’ as ‘the sufferings of victims and their families and relatives are not in the past, but are continuing and will continue throughout their lives’.Footnote ¹⁵⁰ In another case, the ECCC contended that ‘[m]any of those who survived suffered ongoing physical trauma, as well as mental and psychological disorders. The grave impact of these crimes on the victims and their relatives is both devastating and enduring’.Footnote ¹⁵¹

6.1. Explicit and Implicit Forms of Indirect Victimhood in International Criminal Law

As with direct victimhood, indirect victimhood was not designed as a uniform concept in international criminal law (Sections 3, 4 and 5). These divergences are discernible both in relation to its scope as well as its applicability before IHCTs.Footnote ¹⁵⁶ For, apart from featuring various definitions of victimhood in their documents, some tribunals – such as the ICC and the ECCC – have also invoked indirect victimhood in their procedural decisions (on victim participation) as well as substantive decisions (on sentencing and reparation). Other tribunals (ICTY, ICTR), by contrast, have remained more constrained in this respect. However, even ad hoc tribunals which could not recognise victims as participants in criminal proceedings and award reparations for the harm suffered have made frequent references to the idea of indirect victimhood in their sentencing decisions. Finally, all IHCTs at least once have referred to indirect victims in their reasoning without naming them as such. The scope of indirect victimhood in international criminal law is therefore based not just on the lexical use of the term in question, but on the broader idea that underlies it. Consequently, indirect victimhood that can be derived from judicial practices of IHCTs is internally varied: in some respects it is defined (explicit), and in other respects it remains undefined (implicit). De lege lata, however, its normative relevance is indisputable.

6.2. Avoiding Arbitrariness: Indirect Victimhood in Reparation and Sentencing Decisions

As alluded to above, there is a notable difference in the way in which the concept of indirect victimhood has been recognised in sentencing and reparations decisions of those tribunals which can decide on both aspects of punitive and reparative justice (such as the ICC). The distinction stems from the fact that, in contrast to orders for reparations, sentencing decisions are usually devoid of the substantive reasoning that explains why the harm of those whom we are ready to call indirect victims should provide reasons for imposing a harsher sentence on an offender. IHCTs have followed an idea that the effects of victimisation cannot be confined only to those sustained by direct victims. While this might indeed be correct, on many occasions – regrettably – these findings have been entered by IHCTs with little or no additional explanation. Among others, this is clearly visualised by their continued references to an unspecified category of ‘direct victims and their relatives’. Indirect victimhood in international criminal law may thus first be criticised on account of its arbitrary character, given an insufficiently explained legal basis for its application in specific cases. This viewpoint is premised on the claim that decisions of IHCTs – especially those concerning the determination of punishment – provide insufficient information about the way in which the said concept becomes connected with the crime and its perpetrator.

This should not suggest, however, that indirect victimhood does not raise controversies in the field of reparative justice, especially as regards the conditions of causation. One may wish to argue in this respect that harm suffered by indirect victims is sometimes too remote from the victimising events in question to be recognised as foreseeable to an offender in accordance with binding standards, and in such a case that it should not be deemed reparable by a convicted perpetrator.Footnote ¹⁵⁷ If that objection is indeed legitimate, then the whole concept of indirect victimhood – its components and current scope – could be effectively challenged.

Nevertheless, in their continued recognition of harm sustained by indirect victims, IHCTs have clearly espoused the famous phrase that ‘no man is an island’.Footnote ¹⁵⁸ As virtually every international crime committed against another person's protected interests is going to exert some negative consequences on those who were not directly targeted by it, it may be presumed – as IHCTs have repeatedly done – that every act of direct victimisation is also an act of indirect victimisation, and that behind every direct victim, there is at least one – but usually more – indirect victims.

All this may appear uncontroversial but, pursuant to this assumption, the resulting scope of reparative liability in the context of mass victimisation is inevitably going to be vast, if not totally limitless. Considering the personal scope of indirect victimhood, which no longer is limited only to family members (Section 6.3), it is therefore pertinent that before accepting claims for reparations, IHCTs verify (i) whether an offender could have foreseen that an act of direct victimisation may eventuate in the suffering of indirect victims, and (ii) whether family members and other potential indirect victims have indeed sustained some harm as a result of direct victimisation.

The case law analysis, however, reveals that according to IHCTs an attribution of negative consequences to perpetrators of international crimes will generally meet the condition of foreseeability (the proximate cause standard).Footnote ¹⁵⁹ The ICC, for instance, has followed an approach whereby all harm sustained by indirect victims is potentially reparable under Article 75 of the ICC Statute. It remains debatable, however, whether all harm suffered by indirect victims can be seen as proximate consequences of direct victimisation in every case. To date, the burden of limiting the scope of recognised (indirect) victimhood has thus been put not so much on legal requirements (the but-for test and the proximate cause standard), but on evidentiary standards. While the latter condition has turned out to be effective in those cases in which the Court did not accept reparation claims submitted by indirect victims,Footnote ¹⁶⁰ it would still be desirable for the Court to provide more detailed and explicit ways of reasoning not just in relation to evidentiary requirements as such, but also with regard to binding legal conditions of causation (the legal Dogmatik Footnote ¹⁶¹).

As far as the sentencing regime is concerned, there are two claims – radical and constrained – that may be pressed as critical responses to the function of indirect victimhood in the punitive sphere of international criminal law.

The first claim provides that IHCTs should totally abandon the practice of affording any weight to the suffering of alleged indirect victims in their sentencing decisions. This idea had already been expressed in the case of Krnojelac, referred to above, with the Tribunal making it clear that the negative effects of direct victims’ victimisation sustained by indirect victims ‘are irrelevant to the culpability of the offender, and that it would be unfair to consider such effects in determining a sentence’.Footnote ¹⁶² Accordingly, instead of accepting every act of victimisation that results in a complex status of victimhood covering not just direct victims but also numerous indirect victims, communities and even the international community as a whole, an offender would be held responsible only for such harm that was directly caused by their acts.

All the same, this argument may justifiably be viewed as too limiting, and therefore unjust and unacceptable. This leads us directly to the second – more restrained – viewpoint, which stipulates that indirect victimhood should be considered not simply with regard to reparative justice but also in the domain of punitive justice. To counter the criticism elaborated above, however, IHCTs should explain in more detail why they are willing to recognise the suffering of alleged indirect victims as normatively relevant for these purposes. Further, one could also argue that they should conduct a more detailed examination of both substantive and evidentiary conditions providing for the nexus between offenders, their crimes and the suffering of indirect victims. I therefore claim that so long as sentencing decisions do not feature these additional components, references to indirect victimhood in sentencing practices of IHCTs will remain arbitrary and, in effect, not entirely convincing from a legal point of view.Footnote ¹⁶³

6.3. Whose and Which Types of Harm Should Matter? Family Relations and Substantive Boundaries of Indirect Victimhood

The previous section advocated a more detailed exposition of the function of indirect victimhood against the backdrop of judicial reasoning, as well as factual and legal aspects of specific cases before IHCTs. This section, in turn, focuses on the substantive scope of indirect victimhood; for it is one thing to accept the concept of indirect victimhood in sentencing and reparative regimes (despite criticising its inadequate explication in practice), and another thing to identify and properly delineate the personal scope as well as legally relevant effects of indirect victimisation. As the latter context provides additional reasons for the recognition of the former, in what follows I reflect on the boundaries of indirect victimhood by pursuing two prima facie contradictory arguments (too stringent and too broad).

The concept of indirect victimhood in international criminal law is not constrained by the family criterion. Accordingly, both relatives and non-relatives in some circumstances may be recognised as indirect victims of international crimes in the sphere of punitive and reparative justice. This relaxed interpretation of the contours of indirect victimhood, in my view, merits a positive assessment. Indeed, it would seem rather difficult to defend the position that only family members may sustain such indirect injuries or losses.

In their sentencing decisions IHCTs have frequently made reference either to family members of direct victims (relatives) or some undefined collective victims (inhabitants, communities).Footnote ¹⁶⁴ These decisions, however, do not mention other potential indirect victims, such as those who tried to prevent the victimisation of direct victims, even though they have been accepted by IHCTs (for instance, by the ICC) as a subcategory of indirect victimhood in reparation orders (Section 4.1). Consequently, it looks as if the concept of indirect victimhood has had a different scope and character depending on the context (punitive – narrower; reparative – broader) in which it has been invoked by IHCTs. It seems unfortunate that IHCTs have not explained this discrepancy, especially as it does not contribute to the value of conceptual consistency in their case law. Moreover, the phrasing espoused by IHCTs in their subsequent sentencing decisions (‘victims and their relatives’) is not convincing either. It arguably results in a too stringent understanding of indirect victimhood in the sphere of punitive justice in international criminal law.

A related objection, as far as the scope of indirect victimhood is concerned, may be pressed against reparation decisions as well. Its source lies in the fact that the family criterion, as well as harm sustained by such applicants, have been presumed by IHCTs (for instance, by the ECCC and the ICCFootnote ¹⁶⁵) for the purposes of establishing who may be considered an indirect victim in specific cases,Footnote ¹⁶⁶ thereby creating a duality of victimhood – with family members viewed as typical indirect victims, and other applicants faced with a more demanding evidentiary burden before their status of indirect victims is recognised. This approach may be criticised as too stringent as well, although in a different – evidentiary – sense in that it pursues a very traditional and formalistic understanding of the sources of one's indirect victimisation, with the family depicted as the paradigmatic indirect victim. That said, it is even more disputable whether any presumption of victimisation (family criterion, harm) is defensible in relation to substantive decisions from the perspective of the rights of the accused at all.Footnote ¹⁶⁷ If the latter objection is justified, one may argue that an approach adopted by IHCTs has, in fact, been too broad because they should not have introduced such family- and harm-oriented presumptions in the first place.

Apart from probing into the personal scope of indirect victimhood, it is also necessary to inquire about an operative term of indirect victimisation, namely the concept of harm, and to ask which types of harm of alleged indirect victims should be recognised as legally relevant by IHCTs in their respective sentencing and reparation decisions. Unfortunately, in contrast to reparation orders, which cast light on specific types of injury, these conceptual issues have not been addressed by IHCTs in their sentencing decisions to date. What has been taken for granted in the case law of IHCTs instead is that most – proximate and remote – effects of international crimes should be weighed and assessed in the context of both punitive and reparative justice.

Sections 3, 4 and 5 have demonstrated that a great number of types of harm suffered even by remote subjects, including the international community as a whole, have been taken into account by IHCTs in their decision-making processes. I argue that this is what makes an approach adopted by IHCTs conceptually too broad. To substantiate this position, it is first useful to look at the substantive relationship between harm and victimhood. If one examines the nature of criminal responsibility and reparative liability more closely, nine options seem logically conceivable as far as the recognition of harm and victimhood is concerned:

1. (1) all harm of direct victims and all harm of indirect victims;

2. (2) all harm of direct victims and some harm of indirect victims;

3. (3) all harm of direct victims and no harm of indirect victims;

4. (4) some harm of direct victims and all harm of indirect victims;

5. (5) some harm of direct victims and some harm of indirect victims;

6. (6) some harm of direct victims and no harm of indirect victims;

7. (7) no harm of direct victims and all harm of indirect victims;

8. (8) no harm of direct victims and some harm of indirect victims;

9. (9) no harm of direct victims and no harm of indirect victims.

In as much as these options exclude each other, only one may be applicable to each model of victimisation. In this respect it is worth mentioning that the currently employed unrestricted harm-based model of victimhood is premised on the first formula (victim – Rule 85(a) RPE ICC, civil party – the ECCC), whereas the wrong-based model corresponds with the sixth model (victim – Rule 85(b) RPE ICC, victim – ad hoc tribunals, victim – the ECCC).Footnote ¹⁶⁸ From this it does not follow, however, that other correlations could not be used in practice as well, depending on the elements of the adopted definition of victimhood (for instance, through an exclusion of certain types of harm).

To further visualise this issue in relation to indirect victimhood, suppose that XYZ becomes a victim of rape (war crime). XYZ will most likely suffer from physical and psychological harm as these types of harm are closely related to the nature of rape (consequences – direct victimhood). Further, psychological harm is most likely to be sustained by their family members and other people who remain connected with a direct victim in one way or another (consequences – indirect victimhood). Nevertheless, apart from such proximate consequences of this crime, there may be other repercussions of their victimisation, such as material losses (such as, loss of income). What remains contentious as far as the substantive dimension of indirect victimhood is concerned is whether all such – proximate and remote – negative effects sustained by (direct and) indirect victims should be recognised as sufficient bases for granting reparations and imposing harsher sentences on guilty offenders. The answer to this query inevitably lies in the specific policy of victimhood espoused in international criminal law,Footnote ¹⁶⁹ and only secondarily in the legal requirements of causation and attribution applied by IHCTs.

Against this backdrop, the normative position pursued in this article provides that some but not all types of harm of both direct and indirect victims should be treated as reparable and relevant for sentencing purposes in international criminal law. As such, this proposition is premised on the fifth configuration mentioned above, according to which ‘some harm of direct victims and some harm of indirect victims’ should be given weight in the field of international criminal law. This approach is based on two separate – ex ante and ex post – arguments.

The former ex ante perspective revolves around policy reasons.Footnote ¹⁷⁰ Accordingly, I argue that because of their specific character, some prima facie types of harm (such as transgenerational harm), which underpin the concept of indirect victimhood should be considered ab initio as legally irrelevant in punitive and reparative regimes of international criminal law.Footnote ¹⁷¹ This argument directly limits the scope of indirect victimhood in that it precludes recognition of indirect victims who claim to have sustained, for instance, the aforementioned transgenerational harm. Such an undertaking would need to be completed at the legislative stage through limiting the scope of recognised types of harm in the adopted definition of victimhood (as in Rule 85(b) RPE ICC: material types of harm).

The latter ex post perspective refers, in turn, to causal considerations as well as the character of specific international crimes. As such, it does not narrow the conceptual scope of direct and indirect victimhood ex ante. Instead, it suggests that some types of harm (for instance, monetary losses) could be treated as (i) reparable or non-reparable (reparation orders), or as (ii) relevant or irrelevant (sentencing decisions), and that the answer to this query lies in the circumstances of a particular case as well as adopted normative criteria of responsibility (causation). For instance, although monetary harm may be recognised as a legally material ground of victimhood in a general sense (rejection of an ex ante limitation), especially in relation to offences such as pillaging, when the crime in question is directed at interests unrelated to material goods (for instance, an act of rape), it may be argued that such losses should be excluded because of their remoteness from the offender and their crime (material losses – rape). If so, they should also not be considered in sentencing and reparation decisions. In this case, therefore, the limitation on the scope of indirect victimhood comes ex post, following an examination of the correlation between the specific type of crime and the harm allegedly stemming from it.Footnote ¹⁷²