1. Introduction
The number of human rights non-governmental organisations (NGOs) globally has increased from only a few dozen in the 1970s to several thousand today.Footnote 1 Initially, these NGOs focused on monitoring and reporting on state compliance with international human rights norms. For example, Human Rights Watch (HRW), one of the most influential of these groups, was founded ‘to support the citizens groups formed throughout the Soviet bloc to monitor government compliance with the 1975 Helsinki Accords'.Footnote 2 Amnesty International (Amnesty) was founded in 1961 to campaign for prisoners of conscience and to protect the rights of freedom of opinion and religion from state interference.Footnote 3 Amnesty and HRW were selected for this study because they are global in scope and are the largest and most influential NGOs working on issues relating to international humanitarian law (IHL).Footnote 4
The primary form of Amnesty's advocacy was through postcards sent by activists to government officials in the appropriate states.Footnote 5 During the 1980s, however, particularly towards the end of the Cold War, HRW, Amnesty and many other NGOs that were promoting human rights agendas shifted their focus to armed conflict and IHL.Footnote 6 Emphasis on conflict and IHL compliance enabled the organisation to draw attention to the human rights agenda of the United States (US) government and to highlight allegations of complicity in abuses by allied governments during the Cold War.Footnote 7 Amnesty and other human rights NGOs then broadened their mandates, also adopting IHL as a major focus of their work and, like HRW, positioned themselves as military and IHL experts.Footnote 8
Through this process, HRW, Amnesty and other like-minded organisations have become major actors in the world of IHL. For example, NGOs played a significant role in the adoption of the 1997 Ottawa Mine Ban Treaty,Footnote 9 the 1998 Rome Statute for the International Criminal Court,Footnote 10 the 2008 Convention on Cluster MunitionsFootnote 11 and the 2014 Arms Trade Treaty.Footnote 12 These organisations have permanent representatives based at the United Nations (UN), the International Criminal Court (ICC), the European Union (EU), and other international organisations where they lobby and consult on IHL issues with diplomats and UN employees.Footnote 13
Every year, Amnesty, HRW and other human rights NGOs issue hundreds of publications which purport to document violations and to promote IHL enforcement. These publications, accompanied by press statements with quotes from NGO officials, are ubiquitously cited (often in place of law professors or military experts) in the media in articles on armed conflict,Footnote 14 and are used as source material for governmental and UN inquiries, quasi-judicial bodies, the ICC, academic studies and other frameworks.Footnote 15 NGO officials seamlessly move to positions in government and international organisations and back again.Footnote 16 The resources available to many of these organisations often dwarf those of major corporations and government budgets.Footnote 17 Their staffing ranges from hundreds to thousands of employees; they hire top PR and media consultants, as well as the most prestigious law firms to promote their campaigns.Footnote 18
Yet, despite the increase in the number, role and influence of NGOs working on IHL enforcement, it is difficult to correlate NGO advocacy with a measured decrease in armed conflicts or significantly fewer civilian deaths. According to Uppsala University's Department of Peace and Conflict Research (UCDP/PRIO Armed Conflict Dataset) the number of state-based conflicts has risen from under 30 between 1946 and the mid-1970s to between 35 and 55 from 1976 to the present.Footnote 19 Genocide and mass killings have taken place in Rwanda and Bosnia, and armed conflict has killed hundreds of thousands in the Democratic Republic of Congo, Syria, Iraq, Colombia, Sri Lanka, the Balkans, the former Soviet Republics, Algeria, Sudan and elsewhere. In 2015 alone, more than 100,000 civilians were killed in such conflicts. The London-based International Institute for Strategic Studies has similarly concluded that conflicts are becoming more deadly in terms of the number of fatalities, the lethality of weaponry used and the increased displacement of populations.Footnote 20 The vast majority of these conflicts involve parties, including non-state actors, which have engaged in wholesale disregard of IHL.
In other words, there appears to be little direct association between the proliferation of organisations claiming to monitor and document violations during armed conflict with improved IHL compliance, less conflict and reduced civilian casualties.Footnote 21
Among the factors that blunt NGO impact in these areas is their weakness in the realm of fact-finding in armed conflict. As has been noted by many scholars, this type of reporting is not only about collecting facts, but about interpreting, analysing, and putting those facts into what is usually a very complex and contentious context.Footnote 22
In their 2009 publication, the expert group convened by the International Bar Association's Human Rights Institute and the Raoul Wallenberg Institute highlighted the need for clear fact-finding standards for NGOs. The participants found that the ad hoc processes used by NGOs, UN bodies and international agencies were inadequate and counter-productive.Footnote 23 However, this initiative and other similar efforts have not produced noticeable changes in the practice of fact-finding in armed conflicts.Footnote 24 As a result, the need for substantial reform in this area remains pronounced.
This article will begin by examining the essential components of NGO fact-finding in armed conflict, as examined in the relevant academic literature. These include clarity of mandate and methodology, mitigation of selectivity, consistently applied legal standards, requisite military expertise, and sourcing – most notably how witness testimony is gathered and evaluated. Within each category, the article will analyse both objective and subjective elements and how they impact on the credibility of NGO fact-finding inquiries. Specific examples will be explored through the case studies of Amnesty and HRW publications on the conflicts in Yemen, Ukraine and the 2014 Gaza War. The article will conclude with recommendations for NGOs and the actors with which they interact.
2. Fact-Finding Standards and NGO Methodologies
In their seminal 1980 article in the American Journal of International Law, Thomas M Franck and H Scott Fairley wrote that the key components for credible fact-finding are independence, impartiality, thoroughness and promptness. They listed five indicators of impartiality for fact-finding missions: ‘(1) choice of subject, (2) choice of fact finders, (3) terms of reference, (4) procedures for investigation, and (5) utilization of product’.Footnote 25
David Weissbrodt, Professor at the University of Minnesota, and James McCarthy, attorney, in the Virginia Journal of International Law (1981) emphasised the need for standards in NGO fact-finding because these organisations ‘live from press release to hasty drawn report, without time for methodology’.Footnote 26 In an abridged version of the article they noted that by utilising ‘well-accepted procedures’, NGOs could establish ‘general confidence in the fairness, impartiality, and wisdom of the organization’.Footnote 27 The abridged article by Weissbrodt and McCarthy was published as part of a 1982 volume, International Law and Fact-Finding in the Field of Human Rights, edited by Bertrand Ramcharan, a senior UN official and human rights scholar.Footnote 28 The purpose of Ramcharan's work was to establish ‘ground rules’ and strengthen ‘the quality and integrity’ of fact-finding reports.Footnote 29 In his introduction, Ramcharan stresses that the ‘impartial determination of facts’ is an ‘essential requirement’ for effective fact-finding.Footnote 30 In addition to recommendations by Weissbrodt and McCarthy, other requirements highlighted in the volume include procedural due process; evidentiary standards; witness interviews; transparency regarding mandate, choice of subjects and financing; and limiting the process to a statement of facts rather than drawing broader conclusions or legal culpability from those facts.Footnote 31
Building on this foundation and following several years of intensive study, in 2009 the International Bar Association and the Raoul Wallenberg Institute issued ‘Guidelines on International Human Rights Fact-Finding Visits and Reports by Non-Governmental Organisations' (Lund-London Guidelines).Footnote 32 The Guidelines emphasise objectivity, transparency and proper sourcing.Footnote 33 It was hoped that the adoption of international standards would improve the quality of NGO fact-finding and reporting in conflict situations.Footnote 34 It was also hoped that adherence to the Guidelines would provide objective measures by which courts, policy makers and international agencies could assess the credibility of NGO fact-finding publications.Footnote 35
In contrast to these theoretical standards, Professor Robert Charles Blitt noted that, in practice, many NGO missions investigating IHL violations lack terms of reference, clear agendas, selection procedures for team members, detailed descriptions of on-site methodology, the distinction between direct evidence and inferences from indirect evidence, and an explanation of legal standards used.Footnote 36
Blitt also noted the absence of certification or other prerequisites for engaging in human rights monitoring and investigation. As a result, the author cautioned that the absence of standards to distinguish legitimate human rights NGOs from ‘any other third party, “necessarily detracts” from the ability to authoritatively ascertain truth or falsity [in NGO publications]’ and ‘from the industry as a whole’.Footnote 37
Neither HRW nor Amnesty have endorsed the Lund-London Guidelines or other professional fact-finding standards. The HRW website includes a general statement about selecting locations for field research and the interviews it conducts, but it does not refer to or cite professional standards or otherwise indicate that the organisation's methods are based on generally accepted procedures.Footnote 38 Amnesty is even more opaque about its methodologies to the extent that they exist or are articulated, though at times staff members have provided glimpses.Footnote 39 Some of the NGO reporting in the Ukraine, Yemen and Gaza conflicts includes brief mentions of methodology, although these are ad hoc and lack detail. Moreover, there is little evidence that they apply any systematic guidelines or standards.
3. Mandate and Purpose
According to Franck and Fairley ‘a fact-finding mission should not begin its quest without clearly defined terms of reference’.Footnote 40 It is critical that these ‘terms of reference should be neutrally stated in the form of questions of fact’ and that the mission be ‘insulated from socio-political passions and assumptions’.Footnote 41 Terms of reference must be ‘nonconclusory and nonprejudicial to the mission's objectivity’.Footnote 42
In contrast, in their reports dealing with alleged IHL violations in situations of armed conflict, HRW and Amnesty rarely provide the terms of reference, the mandate or objectives in specific fact-finding undertakings. Readers are therefore unable to determine the content and scope of NGO investigations, and whether they are being conducted for advocacy purposes, fundraising, or for some other reason. The broader priorities, strategies, objectives and research agendas of these organisations, along with resource allocations, depart from the objective criteria outlined by Franck and Fairley.
HRW, for instance, on its FAQ page claims that its ‘efforts lay the legal and moral groundwork for deep-rooted changes in policy, law, and public opinion’.Footnote 43 Yet, the organisation does not identify which ‘deep-rooted changes’ it is looking to implement. It is left to the consumer of HRW products to determine how any individual report or campaign fits in with this agenda.
Similarly, Amnesty, in the context of the 2014 Gaza War, conducted an evaluation of its ‘fact finding, documentation and research activities’ to determine whether they were ‘strategic’ for advancing Amnesty's wider campaign of promoting the prosecution of Israeli soldiers in the ICC.Footnote 44 This objective differs significantly from reporting that is designed to establish facts in situations of armed conflict, and to examine the degree to which military actions by the parties provided the best available response, in the context of IHL.
Failure to disclose the purpose of fact-finding exercises raises serious questions regarding the credibility and reliability of NGO reports. Publications issued for advocacy and/or publicity (which are essential for fundraising) are particularly problematic: an NGO's desire to garner public support for its agenda may conflict with accurately conveying the complex situation on the ground. The more dire a situation the NGO can portray, the more likely it will be able to generate media coverage and outrage, prompting a desired action (government pressure, sanctions campaigns) and/or increased donations. There can be a strong incentive for NGOs to exaggerate and overstate allegations, in contrast to conducting careful, thorough and accurate investigations.Footnote 45
A rare acknowledgement of this problem from within the NGO community was provided by William Arkin, a former military analyst for HRW. Analysing claims made during the 2006 Lebanon War, Arkin found that NGOs – including HRW and Amnesty – often exaggerated damage allegedly caused by Israel and ‘consistently described things as having been “destroyed” when they were not destroyed or only peripherally damaged’.Footnote 46
Another example is seen in a 2007 study published by University College London and the Colombian think tank, the Conflict Analysis Resource Center (CERAC). Researchers found that the portrayal by HRW and Amnesty of the armed conflict in Colombia was marred by almost automatic pronouncements that the conflict was intensifying, even when, in fact, the available evidence indicated fluctuation.Footnote 47
HRW has emphasised the issue of cluster munitions in its reporting on conflicts both in Ukraine (20 out of 55 publications) and Yemen (60 out of 99 publications), even though the use of this weaponry is responsible for only a small proportion of the casualties and is not the central source of IHL violations in these conflicts.Footnote 48 Rather than placing cluster munitions within context, the publications appeared to be aimed at bolstering HRW's core campaign of increasing signatories to the Convention on Cluster Munitions,Footnote 49 which in turn helped to enhance the organisation's influence and prestige. Like earlier HRW campaigns on white phosphorus and ‘killer robots’,Footnote 50 cluster munitions are flashy, and generate significant media coverage and PR for the organisation (discussed in the next section). In turn, this attention helps the NGO to position itself as a source of expertise, allowing it to have a more influential role in the policy-making process.
3.1. The Role of Media Coverage
As a number of detailed studies have demonstrated, media attention is a primary driver of the reporting agendas and campaigns of HRW and Amnesty related to armed conflict. According to Aryeh Neier, co-founder and first executive director of HRW, the NGO's efforts to time the release of publications ‘to secure maximum public attention’ is one of the distinguishing characteristics, setting it apart from other human rights organisations.Footnote 51 Amnesty officials have stated that they felt the organisation was often in competition with HRW for media coverage.Footnote 52 In this environment, as noted by James Ron and Howard Ramos, NGOs like HRW and Amnesty tend to follow the media and political agendas: ‘It's easier to sell people what they already want than to try to create new demand’.Footnote 53 Ron and his colleagues analysed ‘data on poverty, repression, and conflict to identify some of the worst places on earth’.Footnote 54 They then correlated this data with NGO reporting and found that ‘few of these countries were covered much by either Amnesty or Human Rights Watch’.Footnote 55
A 2007 study conducted on the priorities of Amnesty and HRW by Ramos and Ron,Footnote 56 along with Oskar Thoms, expands on these findings.Footnote 57 For instance, they note the significant media influence on priority setting for NGOs, because many scholars believe that ‘the most influential activists are those capable of packaging their concerns in ways that appeal to the media’.Footnote 58 In other words, NGOs will not invest significant resources in countries that do not garner media attention, even if the level of abuse is high.Footnote 59
3.2. Selectivity
Highly selective focus on particular issues or conflicts, in contrast to comparative analysis of behaviour based on universal norms, also mars the credibility of NGO fact-finding in armed conflict. NGO networks constitute like-minded communities reflecting narrowly defined frames of reference – their closed institutions are criticised in terms of a democratic deficit, and they are generally not subject to independent accountability measures or checks and balances.Footnote 60 NGOs often set their priorities and resource allocation according to ideological and political agendas, as well as particular individual and institutional interests. They are answerable and primarily accountable to their donors, boards and members, so that, as noted, their priorities may or may not correspond with the gravity of a given conflict or other human rights-based considerations.Footnote 61 In addition, NGOs are able to campaign without having to do the difficult work of balancing the rights and concerns of many constituencies – in contrast to the governments of states.Footnote 62 Thus, NGOs can promote narrow and highly idealistic policies without regard to the wider implications and the demands of realism, which includes exemption from the need to procure the necessary financial resources required to implement them. If their policy prescriptions turn out to be wrong, even disastrously, they are not held accountable.Footnote 63
The distorted agendas of influential NGO actors, which often reinforce the biases of UN frameworks such as the Human Rights CouncilFootnote 64 and other institutions, can have major and detrimental impacts. Groups suffering from discrimination but unable to gain the attention of the more powerful NGOs often go ignored by the international bodies, which may result in additional abuse or facilitate continuing violations.Footnote 65 In some cases NGOs represent groups with financial and political motives that diverge from human rights justifications. They may be in conflict with a rival ethnic group over land and resources and choose to frame the narrative as one of human rights in order to gain public and media attention, or to obtain a political advantage over rivals.Footnote 66
These concerns often manifest themselves in NGO reporting on IHL issues. Franck and Fairley acknowledge that fact-finding rests on a ‘fragile assumption of fairness and credibility’, and is often ‘employed not to discover evidence of real probity, but to amass whatever evidence there may be – even of doubtful probity – to re-enforce predetermined political conclusions’.Footnote 67 Often there is difficulty in distinguishing ‘between objective facts and slanted information provided for partisan purposes’.Footnote 68 In order to prevent NGO fact-finding from being merely a ‘chimera’, therefore, strictly applied standards of due process must apply. These standards are not just ‘desirable but a functional prerequisite’.Footnote 69 Otherwise, NGO fact-finding will ‘solely be used for propaganda purposes and to support generally pre-conceived political views on the situation investigated’.Footnote 70
3.3. Which Conflicts?
The methodological deficiencies initially present themselves in the choice of subject for reporting and, in most cases, are not explained by the NGOs. For example, why are some conflicts the subject of frequent reporting while other seemingly similar (or even worse) situations receive comparatively less attention? As noted above, and in the research by Ramos, Ron and Thoms, NGO agendas often follow the lead of media platforms and political factors, contributing to situations of obsessive attention for certain conflicts on the one hand, and neglect on the other. Political objectives and ideological filters can also influence NGO agendas.
These processes are illustrated in an analysis of the reports published by HRW and Amnesty on the conflicts in Ukraine, Yemen and the 2014 Gaza War between June 2014 and September 2016. Table 1 presents the number of publications posted on the websites of these NGOs. While a review of the number of publications issued is not a perfect metric, it provides a readily observable and consistent measure of attention and resource allocation by the NGO to a given conflict.
Table 1 Number of HRW and Amnesty Publications (June 2014–September 2016)*
Note
* The list of publications is on file with the authors. See n 4 for a description of the reviewed publications.
A review of this data raises several issues, particularly in the comparative realm. The conflict in eastern Ukraine intensified into armed conflict in June 2014, while the conflict in Yemen began at the end of March 2015 (though fighting of varying intensity had been taking place for more than a decade). As of the completion of this research (March 2017), both conflicts were still ongoing. The 2014 Gaza War began in mid-June 2014, intensified in July, and ended with a ceasefire on 26 August 2014. In terms of casualties the UN has reported that in both Ukraine and Yemen approximately 10,000 people were killed, while in the Gaza War it reported approximately 2,200 deaths.Footnote 71 HRW has not provided an explanation as to why – as measured in terms of duration, intensity or casualties – the Ukrainian conflict has received approximately the same attention as Gaza and far less attention than Yemen. Similarly, it is unclear why the volume of Amnesty's reporting for Yemen is the same as that for Gaza and is far less on Ukraine than either conflict.
These questions are even starker when looking at the NGO publication rate from July and August 2014, when both the Gaza and Ukraine conflicts were at the highest level of intensity. Table 2 includes comparative data from other NGOs and armed conflict situations occurring concurrently. North Korea is included for comparative purposes as the most serious human rights situation globally.Footnote 72 Again, there is a significant disparity in the reporting of both HRW and Amnesty on Gaza (and Ukraine in HRW's case) in comparison with more intense conflicts (such as Iraq, Syria and South Sudan), and also regarding Amnesty's reporting of Gaza and Ukraine during that time period.Footnote 73
Table 2 Number of NGO Publications Issued (July–August 2014)
Note
* International Federation for Human Rights.
Similarly, the topics covered by the NGO publications during this time are also telling. HRW, for instance, issued only one short statement three days after the Russian-backed separatist downing of civilian airliner MH17 in Ukraine, killing 298 people, including 80 children. Yet, it issued three condemnations over six days – the first being released less than 24 hours after the event – regarding the deaths of four boys on a beach in Gaza in an Israeli airstrike.Footnote 74 Amnesty also issued only one short statement on the airliner.Footnote 75 In contrast to the dozens of condemnations, lobbying campaigns, embargo calls, demands to end the weapons blockade, and extensive campaigning at the ICC targeting Israel during the Gaza War,Footnote 76 HRW and Amnesty have done little to no follow-up on the MH17 atrocity.Footnote 77
3.4. Which Parties?
The publication records of HRW and Amnesty for the Ukraine, Yemen, and Gaza conflicts also raise the question of how and why the NGOs choose specific parties in a particular conflict to emphasise in their reporting – notably which groups of combatants and which other actors that provide weaponry or other military assistance.
Past studies on NGOs have found a reporting bias in criticising state parties for IHL violations as opposed to non-state actors. The texts of the NGO publications also adopt harsher language, a presumption of violation and allegations of intent to commit civilian harm when discussing state parties. The practice of IHL education and training, as well as the role of IHL advisers throughout the chain of command in many state military forces, is mostly, if not entirely, ignored. State militaries that employ IHL compliance procedures are often equated to terror groups and other non-state actors who have no IHL observance.Footnote 78 This bias is even more pronounced if US support is involved, as discussed below, and specifically in the cases of Gaza and Yemen.
Catherine Fitzpatrick, a former research director at HRW during the 1980s and 1990s, has noted how HRW prioritises its work in the Middle East:Footnote 79
The chief problem of Human Rights Watch in this and other matters related to the Middle East is that it sees itself as the sole honest arbiter of what constitutes compliance with human rights. Yet it does so in a highly politicized manner, not recognizing the essential ‘political’ act of picking and choosing cases and priorities, and engaging with or rejecting this or that regime … human rights groups would do better to ‘go where the violations are’ instead of ‘endlessly balancing the saddle bags’.
Similarly, she offered the following assessment of HRW's coverage of Ukraine after her correspondence with an HRW staff member:Footnote 80
I encountered that well-established biased frame [focusing on state activity] regarding HRW in the coverage of Mariupol … I asked, incredulous, why [the HRW researcher] could not include in her ‘Dispatch’ the fact that armed people in civilian clothing popped up in the crowd of unarmed bystanders and shot directly at the Ukrainian troops – this is clearly documented on several videos … [and] makes it absolutely clear that far from using the ‘excessive force’ that HRW loves to talk about with every state, they used minimal force, returned fire only when fired on themselves, and shot at the ground, not people.
[The HRW researcher's] response … That this couldn't be reported because it would exonerate the Ukrainian army from charges of excessive use of force, and therefore ‘couldn't’ be part of ‘human rights reporting’. Circular reasoning if there ever was one, but I'm glad for that debate because it gave me a great insight into the mind of Human Rights Watch: it cannot accept an exonerating back story, background, mitigating circumstances, factors of armed provocateurs and armed rebels, etc. because to do so would mean they were justifying what might be human rights violations by a state.
The 2007 UCL/CERAC study (referred to in Section 3) also found that HRW and Amnesty placed significantly more emphasis on the actions of the Colombian government and paramilitaries than on left-wing guerrilla groups like the FARC.Footnote 81 In response to this assessment, Amnesty answered that its primary concern was the actions of the government:Footnote 82
[F]or strategic reasons AI's focus has to be on changing government policy. They are the signatories to international human rights treaties and should hold a monopoly of power. As such, they will continue to be the main, but not exclusive, focus of AI's work.
Notwithstanding the applicability of IHL to both governmental and non-state actors, in this, as in other examples discussed below, the sharp contrast between the claims made by NGOs as promoters of universal human rights and humanitarian law, on the one hand, and the attempt to justify a desultory approach in the face of mass violations by non-state actors, on the other, is legally questionable and morally untenable.Footnote 83 Not only does the double standard and inconsistency erode the rule of law by minimising or excluding entire categories of violations and violators from scrutiny and accountability, but such NGO practices also legitimise continuing non-compliance by non-state actors. The one-sided focus may also incentivise compliant states to digress in their IHL observance. Morally, this approach leaves many victims without a remedy and faced with continued exposure to attack.
Beyond generally prioritising criticism of state actors compared with guerrilla or terror groups, HRW and Amnesty are seen as particularly and disproportionately active when those actors are Western governments. Professor Kenneth Anderson (a former consultant for HRW) reviewed reporting by NGOs in the 2006 Lebanon War, and found that human rights organisations focused to ‘near exclusion’ on Western armies in asymmetrical conflicts.Footnote 84 He also criticised NGO reporting for failure to present the ‘full range of factual and legal’ information in such conflicts, instead offering the media and public ‘essentially lawyers’ briefs that shape the facts and law toward conclusions that [they] favor’.Footnote 85
Brookings Institution scholar, Ben Wittes, echoes Anderson's sentiments:Footnote 86
In my view, we are talking about [asymmetric war] today not simply because of the barbarities of any groups but because of the reaction over time to the behavior of those groups by NGOs, international organizations, activist groups, and many members of the legal academy – for whom systematic violations of the law of armed conflict by insurgent groups is just not quite as troubling as is the reaction to those violations, often taken in rigorous compliance with the LOAC [Law of Armed Conflict or IHL] or in good faith attempts to comply, by organized state militaries, particularly those of the United States and Israel.
Robert Bernstein, HRW's co-founder, also remarks on this dynamic:Footnote 87
In my opinion, over the last few years … human rights organizations are trying to act like a referee at a sports event … They come across like a group of litigator lawyers playing a game of ‘Gotcha!’ mostly with the Israeli Defense Forces and occasionally with Hamas, Hezbollah, and from time to time, Iran.
These patterns are present in the reporting of HRW and Amnesty in Gaza and Yemen, in particular: for instance, Israel is the focus of 34 out of 47 (72 per cent) HRW publications on the Gaza War, and 39 out of 47 (83 per cent) in Amnesty's publications. With regard to Yemen, publications directed at the Saudi-led coalition comprised 75 out of 99 (76 per cent) publications for HRW and 31 out of 47 (66 per cent) for Amnesty.
The reviewed publicationsFootnote 88 also reflect greater weight given to the transfer of arms and other assistance from Western third-party states to Israel and to the Saudi-led coalition, compared with scant discussion of military support provided to the other combatants. Seventeen HRW publications and ten Amnesty documents call directly for an end to assistance to the Saudi-led coalition, and to Saudi Arabia in particular, naming the US and the UK. None of these NGO publications on the Yemen conflict specifically discuss Iran or call directly on Iran to end its close involvement and arms transfers to the Houthis. Similarly, regarding Gaza, several of the NGO publications call for an end to third-party support; however, Israel is the only country where specific third-country (the US) assistance is named. Massive lethal Iranian, Turkish and Qatari support for Hamas and other Palestinian combatant groups – including the transfer of weapons, such as rockets used in indiscriminate attacks – goes unmentioned. While this might be explained in terms of the NGO claim to be able to influence perceptions and policies in open, democratic societies, as opposed to the lack of access and influence in Iran and other closed states, the double standards are nevertheless very blatant. Moreover, the one-sided focus and emphasis on advocacy is clearly inconsistent with NGO claims of objective and credible fact-finding.
As noted, the publications of both HRW and Amnesty focus preferentially on state parties to the conflict in Yemen, despite non-state responsibility for hundreds of deaths and contributing to the instability and lack of IHL enforcement. Al Qaeda and ISIS are implicated in more than 200 terror attacks during the Yemen conflict, and thousands of dead and wounded.Footnote 89 On 24 September 2015, for example, 25 were killed at a mosque by an ISIS suicide bombing;Footnote 90 yet neither NGO appears to have reported on this mass-casualty event. By ignoring this major aspect of the conflict, the NGOs are providing a very incomplete and distorted assessment of the conflict, and are erasing critical factors that impact centrally on the legal and factual analysis of the situation and may, in turn, distort policy discussions. While artificially narrow reporting might be beneficial for the NGO's advocacy purposes (such as HRW's focus on cluster munitions), this objective is generally invisible. Furthermore, from a moral perspective, ignoring the role of non-state actors prolongs conflict, deprives large segments of the civilian population of international attention to their plight, and ultimately may obstruct protection efforts and effective remedies.Footnote 91 In other words, the failure to systematically and substantially address IHL violations committed by non-state actors reinforces and contributes to a culture of impunity in those actors.
The selectivity in reporting on state parties by Amnesty and HRW, particularly where Western countries are involved in conflicts, in contrast to reporting on non-state actors, is also reflected in the language used by the NGOs. Often in the publications relating to the latter, NGOs minimise or ignore violations and give the benefit of the doubt to non-state actors, praising minimal reforms and presenting them in positive terms, as discussed below. This ‘soft approach’ appears to be based on the theory that it is supposedly a more effective way to implement change. One reason for this approach is ideological: non-state actors, such as guerrilla movements, are often viewed sympathetically in the post-colonial ideologies often adopted by NGO officials.Footnote 92 Another reason is the desire for these groups to obtain access to closed areas, and the fear that critical reporting might endanger staff.Footnote 93 This would appear to be the case all the more so with areas tightly controlled by terror organisations and militias like ISIS, Hamas, the Donetsk People's Army and the Houthis.
An example of sympathetic language regarding violations by non-state actors is provided in Amnesty's Q&A on the Israel/Gaza conflict in July 2014 regarding the extensive use by Hamas of human shields. Instead of confronting the issue directly, the NGO ascribes humanitarian motives to the forcing by Hamas of civilians to stay in areas of impending Israeli operations:Footnote 94
During the current hostilities, Hamas spokespeople have reportedly urged residents in some areas of the Gaza Strip not to leave their homes after the Israeli military dropped leaflets and made phone calls warning people in the area to evacuate. However, in light of the lack of clarity in many of the Israeli warnings … such statements by Hamas officials could have been motivated by a desire to avoid further panic.
Similarly, Bill van Esveld, an HRW researcher, in comments to a journalist downplayed the extent to which Hamas and Palestinian armed groups embedded weaponry within civilian infrastructure:Footnote 95 ‘I don't think there's any doubt urban areas were used to launch rockets from in the Gaza Strip. What needs to be determined is how close to a populated building or a civilian area were those rocket launches’.Footnote 96 The implication was that while Hamas rocket launches routinely took place within dense residential neighbourhoods and were aimed at Israeli population centres, some or perhaps many of these instances could somehow be described as not in violation of IHL.
4. Applicable Law
Another serious deficiency in NGO fact-finding on armed conflict relates to the law that is cited and applied. Often, the NGO interpretation does not represent generally accepted interpretations, as discussed below with the example of cluster munitions or with ‘effects-based’ analysis military operations.Footnote 97 Non-expert readers are not informed that the organisation is adopting an obscure or even invented position; nor are they provided with alternative viewpoints, as shown in the examples below. In many cases, citations to legal texts or decisions are not provided. Conflicting legal principles are also ignored. Consequently, many legal experts recommend that institutions involved in fact-finding avoid developing legal conclusions and instead leave this dimension to the international institutions and courts.Footnote 98
Inconsistency also plagues the legal analysis provided in NGO fact-finding publications. Definitions for the same legal term differ across conflicts, different reports on the same conflict, and even offer multiple definitions of a legal term within the same document. The UCL/CERAC study on Colombia, for example, found that among its reports Amnesty and HRW had inconsistent definitions for the terms ‘internally displaced persons’, ‘political assassinations’ and ‘massacres’, creating highly skewed data and reporting.Footnote 99 As noted, the NGOs lack systematic sources and metrics for obtaining and evaluating how those variables are measured and sourced.Footnote 100 These problems also characterise the reporting by HRW and Amnesty on Gaza, Ukraine and Yemen, as will be shown below.
4.1. Cluster Munitions
HRW's publications that address cluster munitions provide a stark example of the NGO's inconsistently applied legal standards and confusion between its policy preferences with the existing law. This NGO was a major proponent of the 2008 Convention on Cluster Munitions and frequently lobbies to increase the number of signatories.Footnote 101 Reflecting this agenda, publications involving cluster munitions, as discussed above, comprise a significant portion of the reporting on Yemen and Ukraine (60 per cent and 36 per cent respectively) even though none of the parties involved in these conflicts are signatories.Footnote 102 In addition, these weapons are reportedly responsible for around one per cent or less of casualtiesFootnote 103 and, without additional investigation, it is not known whether they were used in indiscriminate or disproportionate attacks.
Even when discussing the same conflict, HRW provides conflicting claims regarding the legal status of this weapon (demonstrated in Table 3). In some publications, HRW mentions that the parties are not signatories to the Convention, but in others this critical fact is absent. In other cases, HRW presents its policy preferences as the existing law (‘cluster munitions are prohibited by international treaty’).Footnote 104 The result is a highly misleading narrative, which invokes the principles of IHL but without the necessary rigour required for a credible document based on consistent legal standards.
Table 3 HRW Statements on Cluster Munitions*
Note:
* Emphasis added. In contrast to HRW claims, there is no international agreement stating that cluster munitions ‘violate the laws of war’, nor is there any customary law preventing their use. An international treaty was adopted in 2008, whereby many countries voluntarily agreed to stop the use of cluster munitions in order to minimise civilian harm. This treaty was silent, however, on whether cluster munitions violate the laws of war. In fact, as noted by the Prosecutor for the ICC, ‘[t]he Rome Statute contains a list of weapons whose use is prohibited per se (Article 8(2)(b)(xvii)-(xx)). Cluster munitions are not included in the list and therefore their use per se does not constitute a war crime under the Rome Statute’: Luis Moreno-Ocampo, Chief Prosecutor of the ICC, ‘Letter to Senders regarding Iraq’, 9 February 2006, http://www.icc-cpi.int/NR/rdonlyres/04D143C8-19FB-466C-AB77-4CDB2FDEBEF7/143682/OTP_letter_to_senders_re_Iraq_9_February_2006.pdf.
Sources:
(a) HRW, ‘Eastern Ukraine: Questions and Answers about the Laws of War’, 11 September 2014, https://www.hrw.org/news/2014/09/11/eastern-ukraine-questions-and-answers-about-laws-war.
(b) HRW, ‘Q&A on the Conflict in Yemen and International Law’, 6 April 2015, https://www.hrw.org/news/2015/04/06/q-conflict-yemen-and-international-law.
(c) HRW, ‘Ukraine: Rising Civilian Toll in Luhansk’, 1 September 2014, https://www.hrw.org/news/2014/09/01/ukraine-rising-civilian-toll-luhansk.
(d) HRW, ‘Yemen: Cluster Munitions Wounding Civilians’, 14 February 2016, https://www.hrw.org/news/2016/02/14/yemen-cluster-munitions-wounding-civilians.
(e) HRW, ‘Ukraine: More Civilians Killed in Cluster Munition Attacks’, 19 March 2015, https://www.hrw.org/news/2015/03/19/ukraine-more-civilians-killed-cluster-munition-attacks.
(f) HRW, ‘Targeting Saada’, 30 June 2015, https://www.hrw.org/report/2015/06/30/targeting-saada/unlawful-coalition-airstrikes-saada-city-yemen.
(g) HRW, ‘Technical Briefing Note: Cluster Munition Use in Ukraine’, June 2015, https://www.hrw.org/sites/default/files/supporting_resources/ukraine_clusters_briefing_note_final.pdf.
(h) HRW, ‘Q&A on the Conflict in Yemen and International Law’, 6 April 2015, https://www.hrw.org/news/2015/04/06/q-conflict-yemen-and-international-law.
(i) HRW, ‘Dispatches: More Cluster Munition Use in Ukraine’, 4 February 2015, https://www.hrw.org/news/2015/02/04/dispatches-more-cluster-munition-use-ukraine.
(j) HRW, ‘Yemen: Cluster Munition Rockets Kill, Injure Dozens’, 26 August 2015, https://www.hrw.org/news/2015/08/26/yemen-cluster-munition-rockets-kill-injure-dozens.
(k) Mark Hiznay, ‘Dispatches: More Cluster Munition Use in Ukraine’, Human Rights Watch, 4 February 2015, https://www.hrw.org/news/2015/02/04/dispatches-more-cluster-munition-use-ukraine.
(l) HRW, ‘Yemen: US Bombs Used in Deadliest Market Strike’, 7 April 2016, https://www.hrw.org/news/2016/04/07/yemen-us-bombs-used-deadliest-market-strike.
(m) HRW, ‘Ukraine: Widespread Use of Cluster Munitions’, 20 October 2014, https://www.hrw.org/news/2014/10/20/ukraine-widespread-use-cluster-munitions.
As is shown in Table 3, when discussing the use of cluster munitions in Ukraine, HRW sometimes claims that such weapons are prohibited and at other times acknowledges that this is not the case. For instance, a September 2014 document claims that ‘cluster munitions are prohibited by international treaty’ while a March 2015 publication states that cluster munitions are only ‘widely banned’.Footnote 105 Some publications present cluster munitions as a violation of IHL while others present it as a moral issue. In an October 2014 publication HRW says that the use of cluster munitions ‘violates the laws of war’, while a February 2015 report simply states that it is ‘disturbing that Ukraine still uses cluster munitions’ and that it is ‘time’ to ‘foreswear’ their use.Footnote 106
These inconsistencies are also found in HRW publications on Yemen. In contrast to its claims in other publications that say cluster munitions ‘violate the laws of war’ or are ‘prohibited’ (see Table 3), in an April 2015 publication on Yemen, the authors state their belief ‘that cluster munitions should never be used’.Footnote 107 Several HRW documents on Yemen claim that the Saudi coalition is using cluster munitions despite their being ‘banned’ (for example, April 2015, June 2015, February 2016) but others acknowledge that the members of the Saudi coalition are not in fact signatories to the Convention on Cluster Munitions and therefore are not bound by the treaty.Footnote 108 In some cases, both claims appear in the same HRW document (such as August 2015).Footnote 109
4.2. Effects-based Claims
The NGO approach to examining the legality of targeting in asymmetric conflicts reflects another problematic dimension in fact-finding. Under IHL, targets must be confined to military objectives, defined as objects whose ‘nature, location, purpose or use make an effective contribution to military action and whose total or partial destruction, capture or neutralization, in the circumstances ruling at the time, offers a definite military advantage’.Footnote 110 According to the International Committee for the Red Cross (ICRC) commentaries, ‘military objectives’ are ‘all objects directly used by the armed forces: weapons, equipment, transports, fortifications, depots, buildings occupied by armed forces, staff headquarters, communications centres etc.’.Footnote 111 Thus, legitimate military objectives can also include strategic infrastructure sites and buildings that are not located in military facilities. Moreover, as noted by the ICRC, ‘most civilian objects can become useful objects to the armed forces. Thus, for example, a school or a hotel is a civilian object, but if they are used to accommodate troops or headquarters staff, they become military objectives’.Footnote 112
Once a military objective is identified, it must be determined if the attack would be proportional. If the attack ‘may be expected to cause incidental loss of civilian life, injury to civilians, damage to civilian objects, or a combination thereof, which would be excessive in relation to the concrete and direct military advantage anticipated’, an attack would constitute a violation.Footnote 113
No clear standards exist, however, to determine with any consistency what constitutes excessive harm, and the term is used inconsistently, particularly by NGOs.Footnote 114 Whether an attack complies with the principles of distinction and proportionality requires consideration of many factors. For instance, anyone undertaking such an assessment must have knowledge of what was known to military commanders prior to an attack, including enemy locations, presence of military objects, presence of civilians, anticipated harm to civilians, the military advantage expected, and evidence of intent to cause civilian harm. These factors are evaluated prospectively rather than based on the outcomes of a strike.Footnote 115
Few NGOs possess the expertise (see below) or access to information that would allow them to make these assessments. As a result, their reports and campaigns almost invariably claim that strikes were unlawful solely based upon the effects of an attack, such as civilian deaths or damage to objects that the NGOs claim are of a civilian nature.Footnote 116 Legal scholar Geoffrey Corn has remarked that focusing on effects as a ‘touchstone of LOAC [laws of armed conflict] compliance’ is an ‘endemic’ methodological flaw.Footnote 117 Professor Laurie Blank characterises the NGO tendency even more starkly: ‘An effects-based analysis – that is, using the numbers of casualties and extent of destruction to make legal claims – is simply incorrect’.Footnote 118 Similarly, as the former Prosecutor of the ICC, Luis Moreno-Ocampo, explained:Footnote 119
The death of civilians during an armed conflict, no matter how grave and regrettable, does not in itself constitute a war crime. International humanitarian law and the Rome Statute permit belligerents to carry out proportionate attacks against military objectives, even when it is known that some civilian deaths or injuries will occur.
Because of their lack of military expertise and access to military information, and in their near exclusive focus on effects, Amnesty and HRW frequently disregard acceptable justifications for military action, and instead impute malevolent intent, particularly, as noted by Anderson and Wittes, when the attacking side is a Western (or Western-backed) army.Footnote 120
For instance, HRW and Amnesty often declare that areas of intense or widespread damage are the result of indiscriminate or disproportionate targeting, rather than considering other explanations, such as a large concentration of targets in one area or targets located in or near civilian structures. Sometimes damage caused by an attack might be as banal as a technical weapons failure or a simple mistake (tragic, but not illegal). For the NGOs, however, the default conclusion in almost every strike involving reported casualties is consistently attributed to extreme reckless disregard or malicious intent to harm civilians.Footnote 121
HRW and Amnesty publications on the conflicts in Gaza, Ukraine and Yemen generally feature some anecdotes and use those incidents as proof of a widespread pattern of violations. The authors do not disclose why these examples were selected as opposed to others, and do not provide data that could demonstrate the opposite of their contentions (as required under the principle of falsifiability). In the Gaza war, for example, there were thousands of Israeli targeting decisions that did not result in civilian harm.Footnote 122 In contrast, HRW and Amnesty selected a small percentage of cases in order to ‘prove’ a pattern of deliberate intent to attack civilians (a clear example of selection bias).Footnote 123 Cases (and there are many) where the Israel Defense Forces (IDF) chose to cancel an attackFootnote 124 are not included in a comprehensive way, which is also attributable to the substantial gaps in the information on which the NGOs draw.Footnote 125 In some cases, civilians in Israel were reportedly harmed (even killed) because of IDF decisions to limit harm to Palestinian civilians.Footnote 126 Again, these are not mentioned.
Context is central in assessing the legality of any use of force in a conflict situation, but NGOs such as HRW and Amnesty generally decontextualise the fighting by focusing on the alleged violations of one side rather than on the active combat environment. For example, in their publications condemning Israeli actions, HRW and Amnesty do not provide details of Hamas military operations (including actions that were contemporaneous with Israeli operations), the locations of Palestinian combatants, where weaponry was deployed, and many other factors (see examples below). This artificial separation between the actions of the parties in the conflict emphasises a narrative of aggressors and victims, including allegations of collective punishment against civilians.
This practice also has the effect of distorting and erasing the context of combat, which is central in asymmetric or urban warfare. Writing on the 2006 Lebanon War, Arkin noted that NGO reports were written in such a way that ‘the force Israel was battling was nonexistent’. Fitzpatrick has criticised HRW reporting on the Ukraine conflict for the same failures.Footnote 127 Providing an accurate picture of the facts on the ground and the complexities that are involved would complicate the simplistic legal and emotional narrative that NGOs often promote.
Amnesty's 2015 Gaza Platform is an egregious example of this practice of selective reporting.Footnote 128 This interactive map, created in conjunction with the Forensic Architecture group, claims to give access to ‘precious first-hand information’ to ‘help make sense of it’. The map is ostensibly ‘aimed at uncovering the truth about the attacks on Gaza and contributing to accountability efforts for crimes’. The impressive graphics notwithstanding, the information provided is limited to purportedly mapping ‘attacks by Israeli forces on Gaza’ without context regarding Palestinian forces and operations. Attacks by Palestinian combatants on Israel are similarly missing. The Gaza Platform provides no ‘first-hand information’, but rather is based on unverified allegations and hearsay provided by Gaza-based Palestinian NGOs, which in turn appeared to source much of their information from Hamas.Footnote 129
In ignoring the location of Palestinian combatants, weapons infrastructure, military engagements between the IDF and Palestinian armed groups, and rocket attacks targeting Israeli civilians, the so-called Platform makes no contribution to assessing the legal issues. More troubling, the omission of such information could be considered ‘downright fraudulent’, as characterised by Arkin when examining similar reporting problems during the 2006 Lebanon War.Footnote 130 Expanding on why these types of NGO claim were highly tendentious, Arkin found that ‘the prominent international human rights organizations … reported they found little or no evidence of previous Hezbollah presence … suggesting Israeli intent to destroy Lebanon's infrastructure’, and that this ‘dominant and conventional accounting’ was ‘grossly exaggerated, misleading, or patently false’.Footnote 131 Arkin's work on the 2006 Lebanon War is instructive in analysing the 2014 Gaza War (and applicable to aspects of the conflicts in Yemen and Ukraine), reflecting the many parallels in NGO reporting on the two conflicts, as discussed throughout this article.
For the most part, powerful NGOs systematically ignore this criticism and the documentation of their fact-finding failures. Neither HRW nor Amnesty responded to Arkin's analysis, and this pattern has been repeated many times. When Judge Richard Goldstone repudiatedFootnote 132 his own report on the 2008–09 Gaza conflict, citing the false assumptions on which the report was based (largely quoting NGOs), the NGOs fiercely denounced Goldstone, although did not respond to the substance of his criticism.Footnote 133
One important exception is the example of Donatella Rovera, who is Amnesty's head of field investigations. Rovera acknowledges the credibility issues surrounding witness testimony in situations of armed conflict, in which ‘interested parties go to extraordinary lengths to manipulate or manufacture “evidence” for both internal and external consumption’.Footnote 134 She describes how evidence cited in NGO reports is often ‘rapidly removed, destroyed, or contaminated’ and that this ‘bad’ evidence can ‘lead to wrong assumptions or conclusions’.Footnote 135
According to Rovera, these concerns were particularly acute in the publications relating to Gaza, but there is no reason to think that similar dynamics do not operate in other conflict zones:Footnote 136
In Gaza, I received partial or inaccurate information by relatives of civilians accidentally killed in accidental explosions or by rockets launched by Palestinian armed groups towards Israel that had malfunctioned and of civilians killed by Israeli strikes on nearby Palestinian armed groups’ positions. When confronted with other evidence obtained separately, some said they feared reprisals by the armed groups.
These deficiencies are compounded by the absence of references to comparative cases that examine rules of engagement, customary law, casualty data and use of weaponry. The regulations and training which forces receive regarding IHL and the procedures in the chain of command used to ensure compliance are also largely missing from Amnesty and HRW reports.Footnote 137 While it perhaps could be argued that it is unrealistic to expect such information in publications that are intended to give ‘real time’ accounts or to provide ‘early warning’ (though such comparative studies were readily available in the Israeli case at the time of the NGO reporting),Footnote 138 this does not justify unsubstantiated claims, or the absence of the context (such as combatant operations) necessary to make valid factual and legal conclusions. In addition, comparative data is also missing in lengthy ‘post-mortem’ publications issued by Amnesty and HRW months after making their initial claims.Footnote 139 The failure to include such data, even in the absence of time pressure or an active conflict, calls into question the argument that expediency prevents the NGOs from referring to it at the outset of the conflict.
Instead, as described in the examples below, the NGO assessments often read as if they were composed on the basis of a checklist: if, based on their own standards, the NGO fails to find what it considers to be a military objective in a given incident,Footnote 140 the NGO will declare an indiscriminate attack and a deliberate violation of international law. As demonstrated in the examples below, when there is a clear military objective that cannot be explained away, advocacy NGOs often label the attack as disproportionate. If the collateral damage is not particularly high, they will find no military necessity or allege that feasible precautions were inadequate.
The following is a representative example. According to a detailed Israeli report, on 23 August 2014 the IDF targeted a weapons cache in Gaza. Because of an apparent malfunction in the guidance system, the launched bomb struck 100 metres from the intended target and instead hit the home of Hayel Abu Dahrouj.Footnote 141 Abu Dahrouj, a member of Islamic Jihad, was killed along with four other members of the family. Shortly thereafter, a second strike was launched and the cache was destroyed.
Amnesty described this event as one of Israel ‘dropp[ing] aerial bombs on or launched missiles at homes they knew or should have known had civilians inside’.Footnote 142 The NGO further claimed that ‘the apparent target was a member of a military group, targeted at a time when he was at home with his family’; that ‘the mass casualties and extensive destruction of civilian objects that could have been foreseen were in excess of the military advantage anticipated by these attacks’; and that ‘the fighters who were the apparent targets could have been targeted at a different time or in a different manner that was less likely to cause excessive harm to civilians and destruction of civilian objects’.Footnote 143
The publication also stated: ‘Amnesty International has not managed to ascertain the importance of Hayel Abu Dahrouj in the military hierarchy, for instance, if he was a simple member or someone more important’.Footnote 144 These allegations were accompanied by two emotive ‘witness’ accounts which minimise Abu Dahrouj's role as a combatant for Islamic Jihad, one of which discussed how Abu Dahrouj was home because he just ‘missed his kids’.Footnote 145 The NGO appeal to emotion and how it affects factual analysis is discussed in detail below.
Amnesty's claims highlight the fundamental limitations of NGO assessments of military targeting, and demonstrate the dangers of unsupported allegations of war crimes. In the Abu Dahrouj incident, based on the Israeli account, the NGO's assumptions regarding the target were incorrect and, as a result, numerous erroneous conclusions followed: ‘they knew or should have known [the targeted homes] had civilians inside’; ‘the apparent target was a member of a military group’; ‘mass casualties could have been foreseen’; ‘in excess of the military advantage’; ‘targets could have been targeted at a different time or a different manner’.Footnote 146 At no point did Amnesty refer to the possibility of technical failure – far from unique in combat situations.
Another example is found in a second interactive application (also in cooperation with Forensic Architecture) issued by Amnesty, and given the hyperbolic headline ‘Black Friday: Carnage in Rafa’.Footnote 147 In one instance Amnesty claims that Israel struck an ambulance in a drone strike that was ‘carrying two medics, Atef Zamali and Youssef Sheikh al-Eid, and a young volunteer, Youssef Darabih’. Missing from the highly emotive narrative was the information that the two ‘paramedics’ and ‘young volunteer’ were combatants for Islamic Jihad. Instead, Amnesty concealed this information behind a small marker which opens a separate pop-up window.Footnote 148
Rather than simply presenting the terror link in the pop-up, however, Amnesty tries to discount this information, which directly rebuts Amnesty's accusation that Israel committed a war crime by allegedly targeting civilians in an ambulance. First, Amnesty attempts to deny the Islamic Jihad affiliation, despite publicly available photos of the men in uniform, claiming without citing a source that it was ‘not the case’ that they were members.Footnote 149 Second, Amnesty alleges, again without a source, that even if they were combatants, they were not ‘directly participating in hostilities’ but ‘exclusively performing their duties as paramedics’.Footnote 150
In contrast, Israeli NGO B'Tselem, which is not generally supportive of Israeli government policies or the IDF,Footnote 151 noted that Darabih, described by Amnesty as a ‘young volunteer,’ was 27, and that all three of the men ‘participated in hostilities’ and were members of the ‘military wing of the Islamic Jihad’.Footnote 152
In a wider sense, and referring to Anderson's comment that NGO publications are crafted like ‘lawyers briefs’,Footnote 153 it appears that no military responses to rockets and terror attacks would be legal according to HRW or Amnesty.
5. The Question of NGO Expertise
Knowledge of military operations is critical for credible fact-finding and for analysing whether IHL violations have occurred. However, in the years during which NGOs shifted their focus to these issues, as discussed below, they did not make up for their lack of military expertise. This has a direct impact on fact-finding capabilities in the context of armed conflict, significantly reducing the quality and credibility of NGO reporting.
Investigation of even a single airstrike requires extensive knowledge of weapons capabilities and availability, weather and environmental factors, chain of command, targeting and collateral damage assessments, location of forces, capabilities and operations of the opposing force, the context of the strike within the operation as a whole, and many other variables.Footnote 154
NGOs, including HRW and Amnesty, do not possess this knowledge. On occasion, they will hire individuals whom they claim to be military experts, but the qualifications for this title are subjective. For instance, some consultants may have some knowledge of weapons systems in the abstract, but have little experience with military operations and lack the ability to assess how those weapons perform in a dynamic combat environment.Footnote 155
Salil Shetty, Amnesty's Secretary General, acknowledged in an Al Jazeera interview that the NGO does not possess military expertise and should not ‘pontificate on issues we don't really understand’.Footnote 156 Yet, Amnesty publishes hundreds of claims relating to military matters each year, including dozens in its reports on the fighting in Gaza, Ukraine and Yemen.
In its publications ‘Families under the Rubble’ and ‘Nothing is Immune’, regarding the 2014 Gaza War, Amnesty claims that ‘the organization consulted on the interpretation of photos and videos’ purporting to show the impact of IDF airstrikes with ‘military experts’.Footnote 157 These experts are not named, and Amnesty fails to disclose which photos and videos were shared, or their provenance (were they taken by Amnesty field workers? residents? journalists? Hamas members?).Footnote 158 Nor does Amnesty disclose what steps it took to authenticate the materials and whether it provided information from the IDF to its ‘experts’.
To the extent that Amnesty provides the conclusions of these ‘experts’, they invariably relate to guesses as to the weapons used in a particular strike. For example, ‘Amnesty International shared photos of the damage and the weapon fragments at the site with military experts who examined them and described them as remnants of large guided missiles which were likely to have been launched from the air’;Footnote 159 ‘A military adviser who supported the work of Amnesty International confirmed that the damage was consistent with that which would be caused by a large air-dropped bomb’:Footnote 160
A military expert helping Amnesty International's investigations examined photographs of the destruction and the shrapnel found in the rubble. He said that the scale of destruction could only have been caused by an air-dropped munition and fragments suggested a large air-dropped bomb of the Paveway type, guided by either laser or GPS, which, in the case of the Israeli army, is likely to have been equipped with a Joint Direct Attack Munition (JDAM).Footnote 161
Yet, beyond the questions regarding the veracity of the photos, and the highly problematic efforts to identify the weapon used and the party using it, such claims alone provide only partial information as to the context of the incident, and do little to clarify the target, location of combatants, and intelligence known to military commanders regarding the presence of civilians. All of these additional factors are essential for assessing whether a particular military action constitutes a violation of IHL.
Like Amnesty, HRW officials have ‘little expertise about modern asymmetrical war’.Footnote 162 HRW makes extensive claims regarding military operations throughout its reporting, but generally does not disclose the role of military experts, if any. To the extent that HRW names individuals, many appear to have no military experience but rather are hobbyists or ‘citizen journalists’.Footnote 163 There is little evidence that HRW researchers have actual expertise or experience in military tactics, strategy or operations. HRW's ‘senior military analyst’, Marc Garlasco (employed from 2003 to 2009), who frequently issued publications on Israel, had no combat experience (he claims to have done some targeting work for the Pentagon).Footnote 164
During the Gaza War in December 2008 to January 2009, Garlasco made numerous claims regarding the use of white phosphorus by the IDF. These included charges that ‘the IDF had a readily available and non-lethal alternative to white phosphorous … to the same effect and dramatically reduced the harm to civilians … which produce the equivalent visual screening properties without the incendiary and destructive effects’.Footnote 165
Yet, in contrast to HRW's claims, the suggested alternative, the M116A1 shell using hexachloroethane (HC) smoke, is not equivalent to white phosphorus. The United States Army Field Manual explicitly states that ‘[p]hosphorous compounds are considered to be better screening agents than HC’ and approximately 42 per cent more efficient than HC, allowing for better smoke generation using lighter munitions.Footnote 166 US Lt Col Raymond Lane testified before the 2009 UN Goldstone Commission that ‘the quality of smoke produced by white phosphorous is superb. You will never match it’.Footnote 167
Importantly, because white phosphorus smoke is more efficient and longer lasting, fewer rounds need to be fired, thereby reducing the potential for injury from smoke-carrying shells.Footnote 168
Instead of credible evidence and military assessments that reflect knowledge and experience, NGO publications like those of HRW and Amnesty use language and emphasise claims that are inaccurate or irrelevant, but present the facade of expertise. These include references to satellite imaging, precise GPS coordinates and weapons specifications. The use of extreme detail in describing these factors can be used, therefore, as a means to transfer credibility to other claims made that are speculative, not well founded, or even patently false.Footnote 169
In its reporting of the 2014 Gaza War, for example, HRW disclosed in a 2017 publication that it had no experts on the ground, but rather ‘a consultant and a research assistant in Gaza’ who were in contact ‘via telephone and Internet communication’. According to HRW ‘the consultant and research assistant would send photos and sketch the damage from the bombings for the weapons expert [unidentified] to review remotely’.Footnote 170
As discussed, the NGO emphasis is on details of technical weapons and damage assessments derived from brief on-site surveys, interviews and sketches.Footnote 171 Arkin, in contrast, in his investigation of the 2006 Lebanon War, claims to have been ‘mindful that images of bomb damage and enumerations of a relentless effort could also end up conveying exactly the opposite of the actual meaning’. Indeed, he writes, ‘divining Israeli and Hezbollah intent through examining destruction on the ground … can, if one is not careful, convey a much distorted picture’.Footnote 172
6. NGO Reliance on Eyewitness Testimony
Interviews with residents in war zones comprise the bulk of HRW and Amnesty reporting on armed conflict. These few selected interviews are then almost exclusively used to create a factual narrative as well as to underpin military and legal claims. The accounts are highly emotive and frequently gruesome not only to generate sympathy for those who are genuinely suffering, but also to impute credibility to the legal conclusions of the NGOs. Juxtaposing anecdotes of grieving families shields an NGO from scrutiny for the accuracy of its legal claims of war crimes or collective punishment.Footnote 173
The NGOs also use testimonials to prove their claims of indiscriminate or disproportionate attacks, as the witnesses almost always claim that there were no combatants or war objectives anywhere in the vicinity of military strikes and that there was no possible justification for the attacks.
In their publications the NGOs omit many key issues related to witness testimony – issues that are of particular concern in areas tightly controlled by armed groups like ISIS, the Houthis or Hamas. These issues include how witnesses were selected, whether the interviewees were members of or related to members of armed groups, whether members of armed groups were present during the interviews, whether interviewees were contacted by armed groups either before or after meeting with the NGOs, and so on. Former staff members of the Prosecutor's Office at the International Criminal Tribunal have pointed out additional problems often found in NGO interviews, such as lack of accuracy and clarity, and whether statements reflect the ‘actual words of witnesses’ rather than the NGO's interpretation.Footnote 174 They noticed ‘a clear difference between the standards adopted by ‘professional police or criminal prosecutors’ and that of the NGOs’.Footnote 175
Hearsay is also a significant problem in these interviews as it is often not clear if the witnesses actually saw or heard what they were reporting or were simply relaying what others had told them.Footnote 176
As cited above, in a rare moment of candour, Amnesty's Donatella Rovera has acknowledged these problems, and the extraordinary efforts made ‘to manipulate or manufacture “evidence” for both internal and external consumption’.Footnote 177
7. Conclusions and Recommendations
Over the past two decades NGOs have emerged as influential players in reporting on armed conflicts, and on how these conflicts are addressed by the international community. They serve as filters of information, in addition to framing issues and creating narratives, while often advocating political and ideological agendas. NGO narratives are adopted by the media, policy makers and international institutions.
Yet, as demonstrated, in their reports, press releases, videos, social media posts and other activities, NGOs conspicuously lack international fact-finding standards, and in most cases do not possess the knowledge and expertise to make the claims that they do. NGO officials generally do not demonstrate the humility and self-reflection to admit that there is much they do not know or that their conclusions might be wrong. In the numerous instances in which allegations by HRW, Amnesty and many other organisations have been shown to be incorrect or speculative, there are very few examples in which clear corrections have been made. For example, when HRW suspended its ‘senior military analyst’, Marc Garlasco,Footnote 178 the NGO did not remove from its website his numerous reports that alleged Israeli war crimes or add any caveats. While Donatella Rovera, Amnesty's head of field investigations, has acknowledged that in its work witnesses provided the organisation with partial, inaccurate and even manufactured evidence in Gaza, this powerful NGO did not change its procedures and continued to rely primarily on witness claims.
The many problems highlighted in this article regarding the reporting by Amnesty and HRW on Gaza, Ukraine and Yemen are not limited to these NGOs, nor to the armed conflict situations examined. They are endemic and reflective of issues in almost all NGO reporting on armed conflict. These methodological failures are compounded by their incorporation into UN reports and other policy documents that uncritically cite these publications, as demonstrated in the 2009 UN inquiry into the Gaza War (the Goldstone Report), among other examples.Footnote 179
Significant and systematic errors in human rights reporting have important negative consequences in terms of protecting populations from harm in armed conflict, and contribute significantly to what Brookings Institution scholar Ben Wittes has defined as a ‘crisis in the law of armed conflict’.Footnote 180 According to Wittes such faulty reporting contributes to ‘ever increasing demands’ being placed by ‘self-appointed guardians of IHL’ on the ‘most legally scrupulous militaries in the world to achieve something close to perfection in civilian protection’.Footnote 181 At the same time these ‘guardians’ (including influential global NGOs such as Amnesty and HRW) grant near impunity to non-state groups that flout the laws of war as ‘a matter of core strategy’.Footnote 182
To break this negative cycle, Robert Bernstein, founder of HRW, aptly observes:Footnote 183
While there should certainly be oversight over democratic forces in battle, I question whether human rights organizations, unless they change their methodology and in my view, their attitudes, and are more accountable in terms of accuracy, are the right parties to do this. If they wish to continue as judges of democratic armies whose lives are at risk, they must be accountable.
These deficiencies and failures in human rights reporting have been addressed in a number of frameworks designed to provide antidotes, particularly in the Lund-London process led by the International Bar Association and the Raoul Wallenberg Institute; however, this process was largely stillborn and did not produce significant change.Footnote 184 Most NGOs that claim to provide credible fact-finding in reporting on armed conflict, including the leading groups such as HRW and Amnesty, have ignored this effort. NGOs have yet to adopt standards, and other efforts by academics and international bodies to address fact-finding do not appear to have influenced NGO reporting or resulted in visible changes.Footnote 185
Other initiatives directed at the UN or other domestic or international fact-finding commissions have focused on the technical aspects of logistical concerns, confidentiality and translation, but require greater attention to the impact of bias, substantive reporting and sourcing guidelines.Footnote 186 Similarly, these initiatives refer to critical fact-finding principles such as impartiality, but offer little substance as to how to implement them in practice.
The requirements for reliable fact-finding, and expanding on the Lund-London guidelines, include the following:
Separation between political or ideological advocacy and fact-finding
Politics, ideology and advocacy are inseparable from bias, particularly in conflict situations. In pre-judging the ‘innocence’ or ‘guilt’ of the parties, and assigning some the role of victim and others the role of aggressors, NGO fact finders are automatically to be suspected of attempting to validate their assessments. In this context, the Lund-London guidelines call for ‘all members of the delegation’ to be ‘aware that they must, at all times, act in an independent, unbiased, objective, lawful, culturally appropriate and ethical manner’.Footnote 187
One way in which NGOs could potentially mitigate the tension between fact-finding and advocacy could be to establish separate and independent entities to carry out the distinct activities of fact-finding/research and advocacy, utilising different personnel and reporting hierarchies, and subjecting the fact-finding arm to periodic audit.
Transparency
Many of the proposed guidelines for fact-findingFootnote 188 highlight that the mandate and terms of reference for fact-finding missions must be clearly articulated. However, transparency for NGO fact-finding must go beyond these basic elements in order to assess credibility and evaluate potential biases in the reporting. In addition to the mandate and terms of reference, NGOs engaging in fact-finding must be transparent with regard to:
• purposes (fundraising? advocacy? part of an official inquiry or judicial process?);
• selection of subject;
• research methodologies;
• staffing;
• legal and evidentiary standards;
• experts consulted;
• selection of witnesses;
• sources reviewed; and
• financing.
Lack of transparency should be grounds for exclusion in official, judicial and quasi-judicial frameworks.
Regarding the extensive reliance on eyewitness testimony in NGO reports, the use of witness and other anecdotal testimony requires particular scrutiny. The choice of individuals chosen as sources of evidence and the process by which these claims are verified become central issues in evaluating credibility. When relying on witness testimony NGOs should provide detailed information as to how witnesses were selected, what questions were asked, and who was present during the interviews. Bias in witness selection in order to fit a particular agenda, as well as a failure to verify the claims, which includes comparing them with other accounts, would be grounds for excluding the allegations from any official framework.
Accountability
Accountability in NGO fact-finding is also critical. NGOs should commit and adhere to professional fact-finding standards. NGOs should sign up to established frameworks or create their own industry-wide guidelines, and promote such standards within the human rights field.
NGOs should view themselves not only as accountable to their members and donors but also to the public. NGOs must acknowledge mistakes and correct the record in a timely fashion. Inaccurate reporting should be removed from an NGO's website or posted with a visible correction at the beginning of the publication.Footnote 189
In order to enhance their credibility and to serve as a check, NGOs should insist that independent review and compliance mechanisms be included as part of NGO fact-finding standards. This could include the creation of a professional journal or other frameworks to periodically review and offer constructive criticism to published fact-finding reports. The appointment by military and legal experts of an independent committee to serve an audit function would also be useful.
Larger and more influential human rights NGOs, such as HRW and Amnesty, could also create such mechanisms in-house to function much like an ombudsman office, or public editors in some newspapers and media platforms.
Utilisation of independent, professional fact-finding bodies
Independent, internationally established fact-finding bodies such as the International Humanitarian Fact-Finding Commission (IHFFC) – established in 1991 pursuant to Article 90 of Additional Protocol I to the 1949 Geneva Conventions – could also be a credible alternative.Footnote 190 Based in Switzerland, the Commission is a permanent international body designed to be used to investigate alleged violations of IHL and serves as a focused state-based framework without a major advocacy component.Footnote 191 Despite its creation more than 25 years ago, it was only first invoked in May 2017 to investigate an attack on an Organization for Security and Cooperation in Europe (OSCE) vehicle in Ukraine.Footnote 192 Experts and states will closely watch this investigation to see if it can better negotiate the challenges of IHL fact-finding. If this effort proves to be successful, perhaps the IHFFC will start to be utilised more frequently.
Promoting pluralistic discussion
In this context, consideration should be given to initiating a public, open, professional and pluralistic dialogue of the Lund-London Guidelines in the human rights community.Footnote 193 As part of this process NGOs should be encouraged to openly discuss their fact-finding processes and engage with their critics. This discussion should include officials from NGOs and UN institutional frameworks, governments and academics, and would provide an important step towards correction of the many deficiencies analysed in this article. Indeed, the inclusion of this issue as a permanent agenda item in the quarterly meetings of the UN Human Rights Council would provide an important platform for promoting such discussions. This process could be initiated formally by the High Commissioner or a group of UNHRC member states, or through side events sponsored by accredited NGOs and governments, acting separately or in concert. Importantly, however, any discussions or mechanisms would require broad-based and ideologically diverse input from multiple sectors, including critical voices.
In addition, in the large human rights NGOs the establishment of branches assigned to the task of promoting credible fact-finding, and flagging failures in this regard, would mark an important development.
In order to proceed towards such developments, the human rights community, in general, and the NGOs that constitute an important element of this community, in particular, must first acknowledge the questions of credibility in their fact-finding in situations of armed conflict. After this acknowledgement has taken place, it will be possible to implement the various concrete measures proposed in this article and in other frameworks.
