Role and Impact of NGO Cooperation in Human Rights Reporting
Timely human rights reporting proved to be a decisive condition precedent for international intervention in Kosovo. In the slow build-up to international intervention, many human rights organizations publicized the occurrence of violations, including numerous reports of large-scale human rights abuses. However, the international community responded cautiously.
Arguably, the single most significant event in moving the international debate in Kosova/Kosovo toward intervention was the January 1999 massacre at Recak/Racak. It was significant not because it was an atrocity previously unheard of in the conflict, but because of the speed and accuracy with which an unbiased third party—OSCE-Kosovo Verification Mission (KVM)—was able to catalogue and report the events. When the Yugoslav government tried to explain the killings of civilians as collateral damage incurred in a legitimate police action, the international community rejected this claim because the OSCE reports clearly contradicted it. International resolve to act hardened immediately thereafter. This event serves as an example of the power of human rights reporting when it possesses the respect and attention of the international community.
Given the debate surrounding the question of intervention generally, it is likely that, when future human rights crises erupt, the international community will continue to insist on increasingly sophisticated documentation for human rights abuses on which to base their decisions. The NGOs in the human rights community will be the key players called upon to address this challenge. One way in which NGOs can increase the impact of their efforts is to quantify their work in a way that permits biases to be checked and adjusted for in statistical analyses. As discussed in Section II, the comparison of datasets and pooling of information is an essential component of this process. While the human rights reporting of many NGOs is not biased in the sense of being politicized, the various NGOs nevertheless address certain issues and collect data within conditions that limit researchers to a smaller number of data collection methods or to subsets of the victim population. Often these methods limit the extent to which the resulting data may be quanitified and generalized to the entire population of interest. With cooperation, NGOs are able to expand the range of information they can report on and the projections they may make.
Statistical Necessity for Broader NGO Cooperation
From a data analysis perspective, there are a number of benefits that can result from cooperation among NGOs collecting human rights data. One such benefit is the ability to overcome real or perceived bias in the data. When a critic charges that a human rights study is biased, s/he can often mean that the study is too intently focused on violations committed by one perpetrating group. This criticism is taken to imply that the analysis has ignored or undercounted violations committed by some other perpetrating group.32 For example, a critic might charge that a particular project documented only the violations committed by an insurgency while ignoring violations committed by state forces. With multiple, independent organizations surveying the same human rights situation such as that studied here, investigators may be less subject to criticisms that their work is politically biased. In addition, drawing from diverse, multiple sources enables researchers to test for and reduce statistical bias in the data. (See “Sources of Bias in Estimates” in Section II.)
Information from multiple sources also allows statisticians to make sounder estimates of the overall amount of violations committed, as well as totals for each perpetrator, period, or region despite different levels of reporting intensity or foci. The method employed to estimate the total numbers of victims in these analyses relies on information about how often witnesses report the same incidents or victims to different projects. In places (or periods, or among types of perpetrators) where the same victims are reported in many projects, one can estimate that there were relatively few cases that were entirely undocumented. In places where there were relatively more cases that were documented by only one project, one can estimate that there were more unreported cases. Adding information from projects that document relatively few additional cases does not substantially affect the estimates. Adding data from projects that report thousands of cases improves the estimates by bringing the number of documented cases closer to the actual total. That is, adding large projects tends to increase the overall coverage of all incidents and hence reduce remaining uncertainties.
The estimates are also improved by adding data from projects that collected their information in different ways. Human rights information can be collected by surveys and other interviews, exhumations, medical records, and migration records, among other sources. The ideal estimate would combine data collected by very different methods, such as a survey with an exhumation. The victims more likely to be identified by these two methods are not necessarily related; the more unrelated the two methods, the greater the likelihood that any biases in the data produced in one study (e.g., based on exhumation) would be independent of the biases in data produced by a second study (e.g., based on a survey).
The data used for the statistical analysis in this study could be combined with other datasets, thereby reducing both the estimated error and the amount of any statistical bias that still remain. Additional data would particularly improve the estimates of numbers of people killed in each municipality, over time, and by each perpetrator category.
Finally, there is a clear need for new types of data and analyses to aid in the process of holding perpetrators responsible for their actions. Statistical analyses, such as those described above, can be of particular use in this endeavor as they can often identify patterns and trends in abuses that may not necessarily be gleaned from anecdotal information alone. Knowledge of these patterns can help to place responsibility on people in authority positions by helping to identify abuses that result from official policies. For example, the AAAS study, Policy or Panic,33 provides evidence for the argument that Yugoslav authorities executed a coherent program of ethnic cleansing, in contradiction to their repeated official statements. The ABA/CEELI-AAAS study provides further evidence in support of this assertion. Without cooperation among the contributing NGOs, the corroboration provided by this report would not have been possible. Human rights organizations would benefit from conducting more joint analyses of this type. While the challenges may be substantial, fuller cooperation among NGOs is clearly an essential first step necessary to achieve these results.
Prospects for Enhanced NGO Cooperation
Human rights organizations generally gather information in formats tailored to the particular needs of the organization. These organizations operate according to differing mandates and collect information toward differing, yet often complementary, ends. Despite this diversity, efforts at coordination among these groups can be beneficial for all involved. In the Kosova/Kosovo conflict, there was a notable effort to harmonize data collection with ABA/CEELI, The Center for Peace Through Justice, International Crisis Group, and OSCE-KVM, agreeing to follow a standardized format that had the imprimatur of the ICTY. This information, while preserved in a harmonized format and thus more useful to the ICTY, was not adequate for further statistical analysis.
In this regard, it is important to note that most human rights organizations that gathered data were not focusing on statistics as a goal. Rather, they were interested in two main objectives: 1) portraying the nature and variety of human rights abuses to galvanize public opinion; and 2) supporting the investigative efforts of the ICTY. In addition, these organizations were operating under varying timeframes. Some had the goal of addressing the ongoing conflict, while others were focused on contributing to an accurate historic record. To accomplish these goals, most organizations chose a narrative approach that focused on personal details, putting a human face on the tragedy as it unfolded. The narrative approach performs an essential function, and ABA/CEELI and AAAS do not intend to understate its value.
What differentiates this report is that it seeks to expand the range of available information and analyses outside of the narrative context. Access to a wider variety of information is one benefit of enhanced human rights data collection methods and statistical analysis. As the international community deliberates its response to a crisis and its aftermath, it is important to have access to the broadest range of data.
To ensure that such information is available, it is important for the international community to support the efforts of NGOs to collect rigorous, quantifiable data that can be represented in computer databases. Currently, ABA/CEELI, in collaboration with the Chicago-Kent College of Law and the AAAS, is developing a violations documentation database. The design of this database is directly informed from the experience of these organizations in Kosovo, as well as previous AAAS work in Guatemala, South Africa, Haiti, and elsewhere. The goal of this initiative is to provide free database software that will enable interested NGOs to process their data in formats suitable for subsequent statistical analysis. Furthermore, the format and structure will allow independent groups to merge some or all their data, at any point, in order to conduct larger analyses with greater accuracy, controlling for biases. These analyses could serve a variety of purposes, such as providing a reliable picture of the events for the international community or aiding in the prosecution of perpetrators, and they would therefore be beneficial for all organizations involved.
In this report, ABA/CEELI and the AAAS have relied on the cooperation and support of a a group of like-minded human rights NGOs, including The Center for Peace Through Justice, Physicians for Human Rights, and Human Rights Watch. Inter-organizational technical cooperation is a promising development. It follows similar coordination that has in the past been convened under the auspices of truth commissions. The results detailed herein demonstrate the utility of pooling information for its own sake. ABA/CEELI and AAAS hope that this study spurs additional scientific investigation of civilian suffering during the conflict between Yugoslavia and NATO in 1999. Ideally, in future conflicts, NGOs will begin their work equipped with a knowledge of the basic rules needed to collect good data, adequate software, and an understanding of the power of collaboration. ABA/CEELI and AAAS invite like-minded NGOs to join them in this ongoing initiative.
32 This section follows Patrick Ball, Herbert Spirer, and Louise Spirer, Making the Case (2000), ch. 1. A related form of this bias results when a critic challenges the objectivity of an organization’s work, arguing that “violations were committed on both sides” when in truth nearly all violations were committed by one side. Such claims are based on the attribution of moral equivalence, and are often made by diplomats, the press, commissions of inquiry, and other quasi-official processes professing objectivity.
 |
|
|
|
 |
 |
 |
|