APPENDIX Testimony of President Chemical and Biological Arms Control Institute Subcommittee on International Security, 7 November 2001 Mr. Chairman, Members of the Committee, I appreciate the opportunity to appear before the Subcommittee as it addresses "Current and Future Weapons of Mass Destruction (WMD) Proliferation Threats" and considers the effectiveness of export controls in meeting the threat. My remarks are drawn from work conducted over the last six years by the Chemical and Biological Arms Control Institute (CBACI) on issues related to chemical and biological weapons and CBRN terrorism. My remarks today will focus on chemical and, especially, biological weapons threats. I would like to address three inter-related issues: the need for better threat assessments; the linkage between state and non-state threats; and the need for a strategic response in which export controls continue to play an important role. My starting point is the recommendation of the Gilmore Commission (The Advisory Panel to Assess Domestic Response Capabilities for Terrorism Involving Weapons of Mass Destruction) that we must improve our threat assessments. This is true not only with respect to the threat of terrorism but also for the challenge of proliferation at the state level. Traditionally, threat assessments have been overly simplistic. They have tended to focus on only a single factor such as the agent that might be used or the motivations of the state or terrorist who might use them. In addition, threat assessments have emphasized vulnerabilities rather than risks, which are a combination of vulnerability and likelihood. The emphasis on vulnerabilities that derives from a focus only on a single factor such as the agent has several drawbacks: First, vulnerability assessments, especially those focused on BW, portray dangers that are virtually infinite. As a result they provide no criteria or metric against which to plan. The result is either policy paralysis in the face of an overwhelming challenge or pressure to commit enormous funds that will never be enough. • Second, they foster worst case thinking that skews resources toward highconsequence, low probability contingencies. Third, vulnerability assessments transform “what ifs” into tangible contingencies. They provide no sense of whether what is theoretically possible in fact matches the reality of what is likely to happen. An example of this kind of vulnerability assessment is one that often focuses on the agent smallpox. A scenario is posited that begins, "Assume a terrorist has smallpox...," and it proceeds to describe events that lead to a global smallpox pandemic. While it is possible that use of smallpox could have such consequences, the assessment itself says nothing about the likelihood of such an occurrence. And yet, appreciating the likelihood of an event is critical to effective policy planning. Where, for example, would the terrorist get the smallpox initially? Unlike anthrax, smallpox is not present in nature since it was eradicated as an infectious disease by the World Health Organization. There arc potential sources of smallpox, but the scenario of the kind posited above does not address the issue of acquistion. Would not the issue of availability have some bearing on the likelihood of that particular scenario and, hence, be of interest and concern to a decision maker? Conducting more complex threat assessments is not easy. It demands good intelligence and creative analysis. But a better threat assessment will do three things. First, it describes a "threat envelope” that identifies the most plausible contingencies. Such contingencies may be far-reaching. We have tended to focus on smallpox and anthrax, for example, to the detriment of looking in detail at the implications of use of many other potential agents. These could include such traditional BW agents as plague or hemorrhagic fevers, simple agents such as salmonella, e-coli, or industrial chemicals, or more exotic possibilities that lie at the edge of advancing science and technology. Second, it provides a means to identify those contingencies that require hedging, in that, due to the severity of their consequences, some preparation for them should be undertaken, even if they are relatively unlikely. The combination of the threat envelope and the hedging contingencies should give policy makers some measure for making decisions regarding policy priorities and resource allocations. Third, a good threat assessment will highlight the fact that the threat is not unidimensional; rather, it is composed of several elements, including Who: the actor his motivations, intentions regarding casualties, and capabilities How: issues regarding the mode of attack, such as the dissemination mechanism, and other operational considerations. Each of these elements, in turn, entails a significant array of possibilities. The key to successful threat assessment is disaggregating the threat into these component elements and assessing the possibilities that various combinations of them produce. Some combinations of factors will yield significant consequences; others will produce no consequences at all. Historical examples illustrate how the various elements that make up the threat interact to produce varying results. The Rajneeshis in Oregon in the mid-1980s, for example, combined the goal of incapacitating but not killing a significant number of people with a relatively common agent (salmonella) and simple delivery system (pouring the agent on salad bars) to produce a reasonably effective outcome (from their perspective). In contrast, the Aum Shinrikyo was motivated to take mass casualties, selected an appropriate target, and committed both considerable money and scientific effort to the enterprise. It only had access, however, to an attenuated strain of anthrax and its attempts to use biological weapons were totally unsuccessful. This approach to threat assessment leads to important findings that should inform policy decisions. • First, a key relationship exists between the degree of risk and the level of casualties desired in an attack. This relationship, however, is not the straightforward one that higher risk is associated with catastrophic casualty scenarios. Indeed, the degree of risk declines as the level of desired casualties increases, insofar as it becomes less likely. • Second, despite the low probability of catastrophic attacks in the United States, there is still ample cause for concern because we do not know how "massive" a mass attack has to be. Worst-case scenarios need not happen to stress the response system to the point of collapse. It is unlikely that any regional or local response system, and perhaps even a national one, will be capable of dealing with an attack that produces catastrophic levels of casualties. But it is critical to raise the systems' "breakpoints" by expanding capacity on a realistic basis to deal with low-to-middle size CBW incidents. Moreover, the danger and harm inherent in the use of chemical and, especially, Third, the connections between states with CBW programs and non-state actors |