Much of the federal literature on the risks posed by terrorism —even when it involves the analysis of possible CBRN and WMD attacks—also understates the problem of uncertainty. At least in the open literature, this seems to have contributed to five problems in planning and analysis:

1. a lack of sophisticated pattern analysis and threat characterization;
2. a failure to look beyond past patterns of attack and examine the full range of possible futures;
3. a lack of explicit near- and mid-term net assessments of how the balance of means of attack can evolve relative to defensive and response options;
4. a failure to analyze the nature and impact of the many uncertainties in CBRN lethality and effects data; and
5. a reluctance to explicitly consider the full implications of large-scale and complex attack options for response and defense.

There is nothing new about potential threats to the United States from terrorism or even terrorism using WMD and there were many precedents before September 2001. There were at least fifty-two incidents of terrorist threats to use WMD between 1968–1994. However, there is no agreement within the federal government as to how to count and categorize the past pattern of threats to the U.S. homeland. Furthermore, some departments and agencies count attempts in ways clearly designed to suit their programs without defining the attempts in terms of seriousness and capability. Some threat counts seem to define every person in the United States who writes or speaks the word “anthrax” as a terrorist threat.