Conference Report:
Identification, Characterization, and Attribution of Biological Weapons Use
London , UK July 12-13, 2006
Conference organized by the Center for Contemporary Conflict, U.S. Naval Postgraduate School and King’s College London for the Advanced Systems and Concepts Office of the Defense Threat Reduction Agency
Prepared by Dr. Anne L. Clunan, Dr. Peter R. Lavoy, and Ms. Elizabeth L. Stone Bahr
For a printable version of this report, please click here.
Introduction
The Center for Contemporary Conflict (CCC) conference on Identification, Characterization, and Attribution of Biological Weapons Use brought together an international group of scholars and practitioners to discuss the various issues associated with attributing the perpetrator(s) of attacks using of a biological weapons (BW) agent with the aim of developing recommendations for improving the accuracy, timeliness, and ease of the attribution process.
The conference was held in London, United Kingdom, in collaboration with King’s College London, and with support from the Advanced Systems and Concepts Office of Defense Threat Reduction Agency, U.S. Department of Defense. The participants were a mix of academics, policymakers, intelligence analysts, and practitioners from the United States and the United Kingdom. The conference was held under the explicit understanding that all proceedings were conducted on a non-attribution basis; therefore, this report characterizes the discussion without identifying particular speakers—except for those that made presented papers and made formal remarks.
The conference was convened as a first step in helping U.S. and UK policymakers to better understand the requisite elements composing the process of attributing BW use and in creating attribution policies and organizations. All participants concurred on the pressing need for governments to establish such policies and personnel. Over the course of the two days, the participants discussed the lessons that could be learned from historical cases of confirmed, suspected, and alleged BW use, and naturally occurring outbreaks involving BW agents. The participants focused on the tripartite challenge that makes up the process of attribution:
- Identifying whether and which BW agent has caused an event,
- Characterizing the event as intentional or unintentional (natural or accidental), and
- Attributing the attack to the perpetrator(s).
The conference participants came away from the event with a much deeper awareness of the challenges practitioners and policymakers face in developing and conducting attribution policy, as well as a set of practical ideas to help them do so.
Conference participants Theo Farrell and Peter Lavoy.
Importance of Biological Weapons Attribution
In his introductory remarks, CCC director Peter Lavoy emphasized that the use of biological warfare (BW) agents by states or terrorists is one of the world’s most frightening security threats. One of the many attractions of BW agents is the wide variety of effects they can produce on humans (from incapacitating a population to causing mass deaths), animals, and agriculture without the likelihood of detection. Unlike nuclear or possibly chemical attacks, which could be traced back to a particular source, the initiators of a BW release have reason to believe they would not be attributed as perpetrators of the attack due to lack of identifying “signatures.” Moreover, because the effects of biological attacks often take days to manifest, and are potentially similar to the effects of naturally occurring diseases, there may be uncertainty if a deliberate BW agent release has occurred. The history of biological warfare is thus understandably full of uncertainty, controversy, and innuendo. Bio-defense planners, therefore, face significant challenges in identifying whether a deliberate BW attack has occurred, characterizing the agents involved, and attributing the perpetrators.
Reflecting the importance of the topic, U.S. Assistant Secretary of State for Verification, Compliance, and Implementation Paula DeSutter opened the conference with a keynote speech (available at http://www.state.gov/t/vci/rls/rm/69313.htm ). She emphasized the seriousness with which the United States views the threat of biological weapons use, and the critical need to develop means to attribute their use to the states or non-state actors employing them. Assistant Secretary DeSutter closed her remarks by asking the participants to identify tools that could assist in identifying better baseline data, developing reliable global databases, so that scientists and policymakers can more quickly and accurately identify what strains of disease were responsible for particular outbreaks, and precisely where those strains originated. With that tasking, the conference participants turned to methodologies and historical case studies of BW attribution in order to assess what lessons can be drawn for contemporary policy.
Assistant Secretary of State Paula DeSutter.
The Problem of BW Attribution
Conference participants first discussed the methodological problems in analyzing cases involving BW use. Conference authors observed that given the great ambiguity of the historical record, determining how many cases of actual, alleged or suspected BW use have occurred over time is very difficult, complicating assessment of the real scope of the problem of BW use. As a result, they remarked that that qualitative analysis of the limited number of cases presented at the conference represented the best methodology for assessing failures and successes in BW attribution. The use of counterfactual analysis—the effort to run “experiments” on history through the limited use of changing initial conditions—was recommended as an important tool for elucidating lessons for current BW attribution policy because effective quantitative measures are too unreliable.
A recurrent theme throughout the conference was the lack of any common definition of the “attribution problem.” At the end of the two days, the participants agreed that the conference had made progress in outlining practical definitions of the problem and how it should be addressed. Based on the conference results, follow-on research would produce refined policy recommendations on the standards to be used and mechanisms for improving the process attributing the use of BW.
Elizabeth Stone and Leonard Cole.
Lessons Learned from Historical Analysis
Several important themes recurred throughout the two days of discussion that provide the basis for interesting policy lessons. At the heart of the attribution problem lie the issues of credibility and legitimacy attached to allegations of BW use. The historical cases amplified the extent to which allegation and counter allegation, rumors, political agendas, socio-economic factors, and human ecology shape and complicate the attribution process. Reliability of information in the face of tremendous uncertainty is at the root of all these lessons. The first lesson derived from the historical cases concerns the role of politics in allegations and investigations of BW use. The second concerns the critical importance of standardized evidence collection and independent analysis for successful attribution—both in domestic and international settings. The third concerns the significance of accurate and comprehensive baseline information for accurately and rapidly assessing a potential BW event. The fourth and final lesson concerns managing scarce resources in the face of uncertainty and risk.
UK conference participants.
The Role of Politics
The importance of the availability and standards of evidence were reflected in the significance participants attached to the political context in which BW attribution occurs. Jeanne Guillemin’s presentation on the activities of the Japanese Imperial Army’s Unit 731 in China during the 1930s and 1940s demonstrated the need for measures to preserve trial-worthy evidence under battlefield conditions and how political and strategic considerations interfered with the collection, preservation, and public airing of such data. Unit 731 developed and deployed BW agents on Chinese civilians and Soviet prisoners of war, and conducted experiments on live human patients. Despite widely held knowledge of the program, Japanese officials were not tried during the Tokyo war crimes trials in exchange for turning information on the program and its experiments over the United States. This case represents a missed opportunity for developing and implementing a methodology for identifying, characterizing, and attributing BW use that would meet legal standards.
Jeanne Guilleman.
Similarly, the political context of the controversial U.S. allegations against the Soviet Union regarding the use of Yellow Rain (trichothecene mycotoxins) was highlighted in the presentations by Elisa Harris, Rebecca Katz, and by Matthew Meselson and Julian Perry Robinson. The Yellow Rain case was pushed forward in the early 1980s at a time of heightened Cold War tension. Conference participants, particularly those who had been involved in sample analysis at the time, challenged the original U.S. allegation and Rebecca Katz’s conclusion that tricothecene mycotoxins or other toxins could have been used in Southeast Asia. Meselson and Robinson, seconded by UK conference participants, argued that the evidence collected did not support U.S. allegations.
These cases suggest that decisions to collect and analyze data are often instigated or halted for political reasons and that this can severely compromise the legitimacy of the attribution process. In her presentation on the history of U.S. policy in this realm, Elisa Harris noted that despite the high levels of political attention that were paid to the Yellow Rain incident, the investigation and attribution of the issue was never a serious policy priority for the U.S. government. Harris concluded that the failure to make the Yellow Rain investigation a policy priority impacted every aspect of the U.S. effort to characterize and attribute the perpetrator of the incident. In Washington, only the intelligence community was involved, and as result, the investigation lacked any credible technical or scientific evidence.
To circumvent politicized allegations, Harris suggested that the development of impartial international investigatory teams were essential for attribution to succeed in cases involving international conflict zones and other non-permissive environments. She emphasized that governmental officials should not go public with any accusations or allegations without the possession of a strong, technically sound case, as well as the existence of corroborating evidence that can hold up to intense international scrutiny.
Several conference participants observed that BW attribution has almost always been a much lower priority for the governments in question than other political or military issues, and consequently the process of attribution was likely to be dependent on other overarching strategic and political objectives. Such objectives usually impede collection and analysis of evidence that can meet scientific or legal standards.
Participants emphasized that independent scientific peer examination of all evidence collected was essential for credible attribution and provided a remedy for potential politicization. Participants agreed that international and independent analysis was essential for allegations of BW use to be considered legitimate. Most agreed that multilateral mechanisms for investigations and independent scientific peer review were essential. The attribution process, in the view of most participants, cannot be carried out solely within the intelligence and policy communities.
Julian Perry Robinson, Matthew Meselson, Rebecca Katz, and Graham Pearson.
Standards and Availability of Evidence regarding BW Use
Conference participants emphasized that the credibility and legitimacy of claims of BW use rely on the availability of and standards of evidence. Scientists, law enforcement officials, intelligence analysts, and policymakers may believe in different standards of evidence for attribution. These issues arose repeatedly throughout the cases examined at the conference, with most participants, both practitioners and scholars, insisting that the only standard worth using was a legal standard that would hold up in a court of law. One U.S. participant pressed for a lower standard, “a reasonable man” standard,” in order to push for further international investigations to confirm or deny allegations. The historical cases highlighted the wide divergence of evidence available and standards used to assess alleged use of BW and the factors that have impeded the establishment of common evidentiary standards and prevented the collection of evidence.
Victor Asal, Gary Ackerman, and Anne Clunan.
Standardized Data Collection and Analysis
Participants agreed that data collection, preservation, and analysis were vital to credibly attribute the perpetrator(s) of a BW attack. In the case of Yellow Rain, despite their disagreement on the veracity of U.S. allegations, the Katz and Meselson and Robinson presentations emphasized the deleterious effect that poorly conducted investigations and laboratory analysis can have on the ability to present credible evidence. Both presentations stressed that identification, characterization, and attribution of BW use requires prompt access to alleged outbreak sites in often hostile or non-permissive conditions. The non-permissive collection environment meant that inadequate sample and control data were collected and that collected samples were often highly problematic. Such problems were compounded by biased or non-standardized interview techniques of alleged victims and biased sampling of those interviewed. Reliance on untrained or otherwise questionable collectors of data (in this case, U.S.-supported Hmong insurgents, often motivated by hopes of political asylum) resulted in samples that were of unknown origin, subject to tampering and improper handling and storage. Subsequent to data collection, laboratory analysis of the data was poorly executed.
The participants highlighted the importance of preserving the chain of custody of evidence in order to dismiss charges of tampering. If the problems that plagued the Yellow Rain case are not seriously examined and corrected, the United States would likely face almost all of the same hurdles and problems in a contemporary international BW investigation of the same magnitude. In her presentation, Harris echoed these calls for the development of high evidentiary standards to present a scientifically sound case.
Even in permissive environments in advanced industrial countries, the lack of standards can substantially impede the attribution process. In his discussion of the anthrax attacks in the United States during September-October 2001, Leonard Cole highlighted that a significant cause of the delay in characterizing the appearance of anthrax as a deliberate attack was the inability of the Center for Disease Control to come up with a definition of what an “anthrax case” actually was. There were many discrepancies as to who was actually “infected with anthrax” to a level that would register on the CDC’s outbreak radar. This inability to have a working definition of anthrax infection during the outbreak, and the failure to effectively disseminate information on anthrax infection prevented multiple physicians from immediately recognizing symptoms caused by anthrax.
Kerry Kartchner, Susan Martin, and Peter Zimmerman.
Availability of Evidence
The problems of availability and standards of evidence were also emphasized in the presentations of Milton Leitenberg, Jonathan Tucker, and Ray Zilinskas. Leitenberg examined the Soviet and Chinese false allegations of U.S. employment of biological warfare agents during the Korean War. The United States, the United Nations, and the International Committee of the Red Cross were unable to gain access to the sites of the alleged attacks in order to investigate these claims due to Soviet opposition. This case study suggests how difficult it is for countries to prove the negative—that an alleged use of BW did not occur—when they are unable to collect evidence.
Jonathan Tucker noted that several multilateral mechanisms currently exist for investigating allegations of BW and toxin use, but most of them are severely underutilized or atrophied. Such mechanisms include Articles V and VI of the Biological Weapons Convention (BWC), Article XI of the Chemical Weapons Convention (CWC), and use of the UN Secretary General to investigate alleged violations of the 1925 Geneva Protocol banning the use of chemical and biological weapons. He emphasized that the BWC consultative mechanisms lack means for impartial scientific investigation, the CWC challenge mechanism is potentially effective but has never been used, and the UNSG mechanism has atrophied from lack of use and needs to be updated and strengthened. Harris emphasized that the UN Secretary General mechanism in particular offered good prospects for BW investigations.
Conference participants Milton Leitenberg and Ray Zilinskas
Ray Zilinskas discussed Cuba’s allegations that the U.S. aircraft had used thrips palmi, or dengue fever, and other biological agents against Cuban territory from 1961-1996 and its use of Article V of the BWC to gain an international hearing. Since the BWC does not define standards of evidence that must be presented for making such allegations, Cuba presented no evidence at the meeting. Zilinskas critiqued the U.S. government for not requesting an international investigation to collect data and for presenting insufficient evidence of its own at the meeting. As a result, the meeting resulted in the Cuban allegations not being dismissed, due to lack of evidence. Zilinskas warned policymakers that this case suggests that the BWC must develop standards of evidence and standard operating procedures for official investigations to prevent opportunistic false allegations. Had the United States gone to greater lengths to provide a compelling case against the Cuban allegations, this would have made the BWC consultative meeting mechanism a more robust venue for credibly and legitimately handling alleged BW use.
Jonathan Tucker, Anne Clunan, and Kay Mersish.
The Centrality of Information and Separating Signals from Noise
Several conference presenters highlighted the crucial role that background knowledge and the availability of baseline data are for accurate and timely attribution of BW use. In her presentation, Anne Clunan laid out the central problem for attribution: determining whether the appearance of an identified biological agent was the result of deliberate action by a state, non-state organization or individual, an accidental release or a naturally occurring outbreak. She remarked that the problem of identification of an agent or outbreak is a medical and/or scientific problem, in which there generally will be a signal-to-noise problem. It is essential for scientists and the medical community to be able to possess background data, to help decipher this signal-to-noise problem. She stressed that at the heart of this problem is collecting and identifying genuine indicators of deliberate biological agent use from the overwhelming background data and multiplicity of possible causes of a biological outbreak. Clunan noted that international organizations such as the WHO are critical in this role, for they are the type of NGO that will already possess a global outbreak alert system, and assist scientists and medical professionals in determining whether on outbreak is legitimately natural or unnatural. The signal-to-noise problem is particularly acute in unannounced use of BW. For example, in the U.S. anthrax case, identification of the agent as anthrax and characterization of the attack as deliberate were relatively easy tasks, given that the envelopes containing the anthrax announced its presence. Yet, as Leonard Cole noted in his presentation, the number of anthrax hoaxes contributed to the month-long delay in public acknowledgement that anthrax had appeared not just in the mail rooms of the intended targets, but in the U.S. postal system as well.
Clunan emphasized that efforts to focus primarily on technological tools such as air sniffers and syndromic surveillance were insufficient responses. Education of first responders, particularly health care workers familiar with the health profiles of their patients and the region, is essential for picking up indicators of BW use, a point underscored in Leonard Cole’s presentation on the 2001 U.S. anthrax attacks, as well as Katz’s presentation regarding humanitarian medical workers operating in refugee camps. Barrett and Clunan highlighted that a developed public health system is the best provider of baseline data on the affected population, as well as the best measure to prevent and curb a biological outbreak.
Ronald Barrett.
Ronald Barrett’s presentation on false allegations that plague had been deliberately introduced in Surat, India in 1994 emphasized the importance of a broad spectrum of baseline knowledge for accurate and rapid attribution to the accidental importation of pneumonic plague from a neighboring region. Significant background noise, including a dramatic rise in rat infestations, appearances of a related but different disease (bubonic plague), ethnic tension and previous ethnic violence, widespread rumors of poisoned wells, and the lack of public trust in government and health officials all contributed to an extremely difficult environment for characterizing and attributing the outbreak of plague.
Barrett demonstrated that a thorough understanding of the social demography and economic conditions of the outbreak site and its surroundings are necessary to correctly identify the causative biological agent, characterize the outbreak as unintended, and attribute the outbreak to the migratory patterns of rural workers seeking urban employment. Zilinskas’ presentation also highlighted the need for baseline regional knowledge on the endemic and new appearance of particular biological agents and the consideration of alternative hypotheses to assess the causes of alleged BW use.
Lawrence Freedman.
Inaccurate Attributions, Credibility, and Scarce Resources
An important lesson derived from the conference was the negative consequences that politically motivated or flawed attributions have on a country’s ability to manage the threat of biological weapons use. Graham Pearson, former head of the UK chemical and biological defense institution, Porton Down, made this the focus of his presentation on lessons learned from UK attributions.
All of the conference papers will be edited for publication as a book in 2007. As information on that publication becomes available, it will be posted on the Center for Contemporary Conflict website.
Graham Pearson and Ray Zilinskas.