Canadian Security Intelligence Service logo Main Menu

Franšais{short description of image}


Table of Contents

Biological Terrorism


Perhaps the principal characteristic of biological agents that could make their use attractive to terrorists is their extreme toxicity, even compared to other weapons of mass destruction. This factor has been expressed in a number of different ways:

Berkowitz et al. describe the toxicity of biological agents as follows:

The potency of the pathogens on a weight basis exceeds that of the most toxic chemicals; between a few and a few thousand viable organisms is all that is required to produce infection in many cases. Since pathogens can be prepared in concentrations of the order of 1010 microorganisms per gram, infectious doses range downward from 0.1 microgram per targetindividual. The search capability of the aerosol cloud and the fact that infectious doses are independent of victim bodyweight (because the pathogen reproduces in the host), make thequantity of BW material needed for mass attack quite small indeed. (1972: VIII-54)

A number of studies have compared the amounts of biological agent needed in a particular hypothetical attack to that of various chemical agents. For example:

As the wide range (and sometimes inconsistency) of the above estimates suggests, much remains unknown or uncertain about the precise effects of biological agents. It is clearly highly misleading to extrapolate directly from individual lethal doses of a substance to estimating casualties from mass attacks, given the need for effective delivery of the agent (about which more will be said later). Nevertheless, it cannot be denied that, in terms of sheer lethality, biological agents—in theory—appear to offer a "bigger bang for a buck." In particular, most authors rate them as far more effective than chemical weapons in this respect; some would even extend the comparison to nuclear weapons. In the words of Kupperman and Woolsey: "The terrorist armed with chemical or radiological agents can kill hundreds, possibly thousands of people. By contrast, terrorists armed with biological weaponry can, in principle, kill tens to hundreds of thousands" (Kupperman and Woolsey 1988:5). Elsewhere, Kupperman (former chief scientist at the US Arms Control and Disarmament Agency) has gone even further, stating in 1977 that "biological agents—both toxins and living organisms—can rival thermonuclear weapons, providing the possibility of producing hundreds of thousands to several millions of casualties in a single incident" (cited in Kupperman and Trent 1979: 633); and in 1989 that "the mortality levels from a biological attack could possibly exceed that of a large nuclear explosion" (Kupperman and Kamen 1989: 103)4. Dr. Graham Pearson, Head of Britain's Chemical and Biological Defence Establishment, has been quoted recently as saying that "Anthrax, sprayed from the back of an aircraft on a cool, calm night, could take out all of Washington DC. This could cause up to three million fatalities compared to two million from a hydrogen bomb" (Majendie 1994).

Other Putative Advantages of Biological Weapons

A number of other factors are said to favour the acquisition of biological weapons by terrorists, particularly by comparison with other weapons of mass destruction. Some of these are related to the toxicity issue discussed above. For example, the smaller quantities of agent needed on account of their lethality help reduce the costs and complexity of their production or other acquisition, in turn eliminating the necessity for a large infrastructure of personnel and facilities, which in turn eases the problem of security and avoidance of detection.5 The relative ease and cheapness of their manufacture or acquisition, especially compared to nuclear weapons, is dealt with at greater length later in this paper. Other advantages include:

A sophisticated program designed to produce a fissionable device would probably cost hundreds of millions of dollars, whereas type A botulinus toxin, which is more deadly than nerve gas, could be produced for about $400 per kilogram. A group of C/B experts, appearing before a UN panel in 1969, estimated 'for a large-scale operation against a civilian population, casualties might cost about $2,000 per square kilometre with conventional weapons, $800 with nuclear weapons, $600 with nerve-gas weapons, and $1 with biological weapons.' (1987: 16)

According to Berkowitz et al.: "The cost of equipment and facilities [for BW] is somewhat greater than that described for synthesizing the chemical poisons but appreciably smaller than that involved in INW [illicit nuclear weapon] development, and the problems associated with obtaining a seed culture are trivial compared with those to be surmounted in acquiring a supply of SNM [special nuclear material]" (1972: VIII-66).

Requisite Capabilities

As noted above, specialists appear unanimous in the view that it would be much easier for terrorists to produce or otherwise acquire biological agents than nuclear weapons. However, there remains considerable disagreement over precisely how easy this would be, in particular the level of technical expertise required. At one extreme are authors such as Mullins who assert that:

Unlike with nuclear devices, technical knowledge necessary to use bacteriological agents is practically nonexistent. Almost anyone could use biological agents. There are no special knowledges required, no technical experts required, no hi-technological laboratories required, and cost would be minimal. A large team would not be necessary to manufacture the agents... (1992: 101)

Similarly, Simon maintains that "Several biological agents can be produced either at home or in a small laboratory, without sophisticated scientific knowledge" (1989: 13)10. According to Douglass and Livingstone: "To a knowledgeable person the procedures required to obtain strains or cultures of very dangerous toxins and diseases—and to produce them in sufficient quantities—are about as complicated as manufacturing beer and less dangerous than refining heroin" (1987: 23). Baum adds: "In fact, everything one needs to know to build a biological weapon can be found in a public library" (1993: 17). Ponte maintains that "Terrorists with the technical sophistication of college sophomore biology majors could steal (or even purchase from research supply houses) lethal germs and from them breed batches of disease able to kill millions of people" (1980: 52). Watkins also emphasizes the ease of production:

Simple brewery equipment is all that is needed to mass-produce a weapon from naturally occurring agents of such diseases as anthrax, cryptococcosis, cholera, brucellosis, plague and typhoid. If the organization is sophisticated enough to have tissue-culture capability, the viruses causing smallpox, typhus and yellow fever can be mass-produced using methods similar to those for making vaccines. Generic directions are freely accessible in open literature....Thus, with minimal financing and expertise, a group can equip itself with a devastating weapon of mass destruction. (1987: 195)

Later in the same article, Watkins reiterates that "even the smallest terrorist groups probably have the organization and resources needed to build and deploy such weapons" (1987: 197).

Other authors are more conservative about terrorists' capabilities to produce and employ biological agents. Mengel, for example, emphasizes that the resources required would be "somewhat greater than those for chemical agents," as would be "the extent of the facility and the accompanying cost" (1976: 455-6)11. In his view: "The type of knowledge needed probably is beyond a biologist, necessitating the employment of both a microbiologist and a pathologist....overcoming the problem of the deterioration of the biological agent once it has been released requires extensive skills, even beyond those available to microbiologists and pathologists" (1976: 455-6)12. Mengel believes that the task would likely be beyond the capabilities of any single individual:

It would take a highly trained individual with experience in microbiology, pathology, aerosol physics, aerobiology, and even meteorology to make a reasonable attempt at manufacture and employment of a biological agent. Thus, although it is possible for one individual to undertake this type of technology, it is highly unlikely that this will occur. A larger group of at least three to five members with a full range of capabilities, including training, tactics, technical knowledge, resources, and operational experience, is considered the optimal number required to perpetrate an act using biological technologies. (1976: 456)13

Mengel's views are apparently shared by Root-Bernstein, who writes: "It takes unusual learning to employ bioterrorism. Time, place, and opportunity must coincide. So far, terrorist organizations have apparently lacked the sophisticated knowledge and training to plan and carry out biological terrorism. People with advanced degrees in microbiology, medicine, pharmacology, and agricultural science seem to be rare if not nonexistent among the membership of identified terrorist groups" (1991: 50)14.

Most authors, however, come closer to the opinions expressed by the first group. They emphasize the ready availability of open literature providing details on virtually every step required. In Mullen's words: "Procedures for sampling, screening, identifying, isolating, and culturing almost any biological organism of public health concern are published widely in microbiological texts and manuals, the sampling, care, and feeding of B. anthracis included" (1978: 76)15. As for effective delivery of the agent, according to Kupperman and Smith:

Aerosol dispersal technology is easy to obtain from open literature and commercial sources, and equipment to aerosolize biological agents is available as virtually off-the-shelf systems produced for legitimate industrial, medical, and agricultural applications. With access to a standard machine shop, it would not be difficult to fabricate aerosol generators and integrate components to produce reliable systems for dispersing microorganisms or toxins. (1993: 41)

Most authors also appear to assume that a single individual with a modicum of technical training could acquire the necessary expertise. Kupperman and Smith, for example, note that "biotechnology equipment and expertise have been widely disseminated throughout industrialized and many third world nations....U.S., European, and Asian universities graduate literally thousands of scientists and engineers each year with the technical acumen needed to produce and effectively use biological weapons" (1993: 37).16 Dr. Iris Shannon, President of the American Public Health Association, in 1989 cited "several experts" to the effect that "it would be very easy for an individual to develop his or her own biological weapons and that many new organisms could simply be built in a kitchen and produced in great quantities in a brewery" (SCJ 1990: 114).

Regarding botulinum toxin, Mullen maintains that "with modest facilities, an individual could produce in a relatively short period several hundred thousand human LD50 doses" (1978: 74).17 Similarly, in his view, obtaining a seed culture of anthrax "should present only moderate difficulty to virtually anyone with a background in microbiology or a related discipline" (1978: 75).18 As for ricin, according to Kupperman and Smith: "All that is needed is the castor bean and an adventuresome terrorist willing to extract the toxin from it. The solvent extraction of the protein albuminoid toxin is a well-documented, trivial, two-step procedure" (1993: 40).

According to the US Congressional Office of Technology Assessment (OTA), "The technical requirements for culturing microorganisms or producing toxins for use in bioweapons are not particularly high. Most estimates are that second-year or third-year medical or microbiology students would have enough laboratory experience to prepare an agent with minimal danger to themselves" (1992: 37). Similarly, Jenkins and Rubin estimate that "The technical knowledge required would be that of an experimental microbiologist, probably at the level of at least a master's degree. Some knowledge of aerosol systems would also be required, although for a crude dispersal mechanism, this knowledge need not be advanced..." (1978: 226). Several authors believe that, if the terrorists themselves lacked the necessary technical expertise, it would be relatively easy for them to recruit those who did have it (Simon 1989: 13)19. The OTA study points out that "some states that are suspected or known to have bioweapons programs also are known to have sponsored terrorist groups" (1992: 37).

Likely Types of Agents

Biological warfare agents include both living microorganisms (bacteria, protozoa, rickettsia, viruses, and fungi), and toxins (chemicals) produced by microorganisms, plants, or animals. (Some authors classify toxins as chemical rather than biological agents, but most do not, and they were included within the 1972 Biological Weapons Convention—as reflected in its formal title, the Convention on the Prohibition of the Development, Production and Stockpiling of Bacteriological (Biological) and Toxin Weapons and on Their Destruction). Writers on the subject have produced a long list of BW agents that terrorists could potentially use20. Among those mentioned have been: anthrax, cryptococcosis, escherichia coli, haemophilus influenzae, brucellosis (undulant fever), coccidioidomycosis (San Joaquin Valley or desert fever), psittacosis (parrot fever), yersina pestis (the Black Death of the 14th Century), tularemia (rabbit fever), malaria, cholera, typhoid, bubonic plague, cobra venom, shellfish toxin, botulinal toxin, saxitoxin, ricin, smallpox, shigella flexneri, s. dysenteriae (Shiga bacillus), salmonella, staphylococcus enterotoxin B, hemorrhagic fever, Venezuelan equine encephalitis, histoplasma capsulatum, pneumonic plague, Rocky Mountain spotted fever, dengue fever, Rift Valley fever, diptheria, melioidosis, glanders, tuberculosis, infectious hepatitus, encephalitides, blastomycosis, nocardiosis, yellow fever, typhus, tricothecene mycotoxin, aflatoxin, and Q fever. Some of these agents are highly lethal; others would serve mainly in an incapacitating role21. Some authors have also speculated about the possible terrorist use of new, genetically-engineered agents designed to defeat conventional methods of treatment or to attack specific ethnic groups, for example.

Douglass and Livingstone state that terrorists would be most likely to choose a bacteriological rather than a viral or rickettsial agent, since rickettsial infections can be readily treated with antibiotics and viruses are more difficult than bacteria to cultivate and often do not live long outside a host (1987: 13)22. Toxins, they go on, are more stable and some have the dual advantages of being relatively simple to manufacture and extremely toxic. The main criteria used in selection of an agent by terrorists would presumably include its toxicity; ease of manufacture or other acquisition, cultivation and dissemination; hardiness23; immunity to detection and/or countermeasures; rapidity of effect (this may be desirable in some instances and not in others); and contagiousness.

Despite the great number and diversity of potential agents, most authors focus on a relatively small number of likely candidates. Berkowitz et al., for example, list just eight types: anthrax, brucellosis, coccidioidomycosis, cryptococcosis, pneumonic plague, psittacosis, Rocky Mountain spotted fever, and tularemia (1978: 225). They explain that "Such potential BW diseases as glanders, melioidosis, bacillary dysentery, Q-fever, the various kinds of encephalitis and encephalomyelitis do not appear [in their list] for combinations of reasons related to the terrorist context: low availability, difficulty of cultivation, low resistance to dissemination stresses, problems of self-protection, redundancy of effects with diseases listed, etc.." They do concede, however, that such diseases "might be chosen for an illicit BW attack" (1972: VIII-57). The eight diseases that they did choose to focus on represent "three different patterns of importance," based on the twin criteria of "casualty effectiveness and epidemicity": "plague and psittacosis are potential epidemic agents; anthrax, plague, and Rocky Mountain spotted fever are highly lethal to infected victims; and the remaining diseases are, ideally, non-epidemic incapacitators" (1972: VIII-67).

In terms of their "practicability" from the terrorist point of view, Berkowitz et al. describe their chosen candidates as follows. Of the highly lethal agents:

The anthrax bacillus is not uncommon in the medical environment; can be reliably obtained from its natural reservoirs; is easily grown in quantity; and, by virtue of spore formation, leads to a stable agent for dissemination which offsets its somewhat lower infectivity. Plague bacteria are less common in the medical environment, but perhaps more available in their natural reservoirs than anthrax bacteria; they are also more difficult to cultivate and more subject to aerosolization stresses, but highly infective by both primary and secondary dissemination. The causative agent of cryptococcosis is a yeast isolable from the localized lesions or the cerebrospinal fluid of its victims..., but both its clinical and natural occurrence is limited which may produce an acquisition problem for the terrorist. It grows easily but slowly, and may be difficult to disseminate effectively. (1972: VIII-73)

Of the "incapacitating or low lethality" diseases:

...the organisms of brucellosis (undulant fever) are difficult to grow and to immunize against; but are highly infective, readily available from both clinical and natural sources, and relatively stable to aerosolization....The tularemia organism is extremely infectious and readily available; it is also difficult to grow and is quite delicate when disseminated. The spores of Coccidioides immitis, the desert fever agent, are easily obtained and grown, very stable, and very effective as long as the attack is not made in endemic areas where natural immunity runs as high as 80 percent in some locations....The rickettsia responsible for Rocky Mountain spotted fever is easily obtained and can be protected against, but specialized virological technique is necessary for its production and it is difficult to disseminate effectively while maintaining virulence. Concentrated preparations of psittacosis virus are extremely dangerous and no effective vaccine is available. While the virus could be obtained from an infected turkey farm, the search might be a long one; additionally, quantity production of the virus would be a demanding job. While the mortality rate of treated cases of psittacosis is quite low, untreated cases are frequently fatal....Since the disease can be spread by secondary infection, it is difficult to predict whether widespread primary dissemination would oversaturate available medical resources and result in a large number of fatalities. (1972: VIII-74)

Mullins' list of "agents most likely to be used" is somewhat longer than that of Berkowitz et al., dropping coccidioidomycosis, pneumonic plague, and Rocky Mountain Fever, but adding escherichia coli, haemophilus influenzae, yersina pestis, malaria, cholera, typhoid, bubonic plague, cobra venom, and shellfish toxin (1992: 102). The 1992 OTA study specifically mentions eight likely agents: anthrax, tularemia, yersina pestis, shigella flexneri ("to contaminate water or food supplies of civilian populations"), s. dysenteriae (shiga bacillus), salmonella species such as salmonella typhi (again, "to contaminate food, water and other beverages"), botulinum toxin, and staphylococcus enterotoxin B (1992: 37-8). Kupperman and Smith offer the shortest list of all, restricted to anthrax, botulinal toxin, and ricin (1993: 38-9).

As can be seen, anthrax makes everyone's short list. The reasons for this are not hard to discern; its extreme toxicity and relative ease of production and dissemination have been discussed already. In addition, as Kupperman and Trent explain:

Nearly all microorganisms die quickly when exposed to sunlight. They are adversely affected by high temperatures and succumb easily to desiccation. Simply put, they are fragile. This is not the case for anthrax. It is a hardy organism. In spore form it can live for decades, withstanding wide variations in environment. (1979: 68)

Jenkins and Rubin concur, noting that "Untreated, [anthrax] is invariably fatal, and its lethality is high even with treatment. Anthrax spores can survive in boiling water and have been known to survive in soil for decades after the original contamination" (1978: 226)24.

Among other agents considered possible for terrorist use are the following:

At least two sources consider but dismiss the likelihood of the use of the smallpox virus by terrorists, on the grounds that it is allowed to be stored at only two sites in the world (the Center for Disease Control in Atlanta and Research Institute for Viral Preparations in Moscow), and that these are both secure facilities (Wiener 1991: 130; OTA 1992: 39). However, it is worth noting that a recent newspaper article described extremely poor security conditions at the Moscow site. In the words of Howard Witt of the Chicago Tribune: "Two pleasant pensioners working as doormen, a few strands of rusting barbed wire atop a crooked fence and an old alarm system linked to a distant police station are all that stand between a laboratory housing the deadliest virus in human history and any potential terrorist who might be determined to steal it." The laboratory's deputy director, Vitaly Zverev, is quoted as admitting that "If some terrorist really wants to do something with this smallpox virus, there is nothing that can stop him" (Witt 1994: A10)25.

Finally, as noted above, there has been some speculation about the use of genetically-engineered organisms for terrorist purposes. According to Douglass and Livingstone, "If [terrorists] choose, they can easily enter the uncharted waters of genetic tinkering, and manufacture chemical mutagens that interfere with genetic codes, or diseases that are resistant to any existing antibiotics and for which the body has no known defenses" (1987: 14)26. However, the 1992 OTA study concludes that genetically-engineered agents would have to be "supplied by a state with an advanced offensive biowarfare program" and "even if feasible, would require years of careful work with state-of-the-art technology" (1992: 38-9). Wiener appears to agree with this judgment, noting that "if [terrorists] are using sophisticated weapons made by genetic engineering, they would bring direct attention to state-sponsored terrorism. Use of such weapons would identify specific countries supplying the terrorists [which]....might risk annihilation by a policy involving these weapons" (1991b: 71).

For his part, Watkins simply states that the use of genetically-engineered organisms "is completely unnecessary,...since effective defences are difficult even for naturally occurring diseases" (1987: 195). This also appears to be the view of various "experts" at the US National Institutes of Health and National Science Foundation (NSF) who were interviewed in 1981 by The Futurist magazine. While insisting that "Recombinant-DNA techniques...are simple enough to be mastered by a broad range of people willing to study" and "could be carried out in relatively small laboratories throughout the world," the scientists, according to the article, "do not fear such use of genetic engineering because biological warfare is already so advanced and deadly that gene splicing would simply not be worth the trouble." It went on to quote Herman Lewis, head of recombinant-DNA activities for the NSF: "Any terrorists or people conducting biological warfare who knew what they were doing would not use genetic engineering. There are much deadlier things they can use" (Futurist 1981: 18).

Means of Acquisition

There are a number of ways in which terrorist groups could acquire a "seed culture" or even operationally useful quantities of a biological agent: stealing it from a legitimate existing facility; buying it either on the black market or from legitimate sources through the mail; receiving it from a "friendly" foreign government; or extracting it from the natural environment. In regard to the last of these methods, Kupperman and Smith assert that "Even today, the most effective and easy-to-use agents occur naturally in the environment and are not man-made" (1993: 38). Anthrax, they note, is "endemic to large areas of the world" (1993: 39).

Douglass and Livingstone note that the agents producing anthrax, the plague, brucellosis, tularemia, and smallpox can all be isolated from natural sources; tricothecene mycotoxins derived from corn; aflatoxin from peanuts; and ricin from castor beans (1987: 23). They even provide a kind of recipe for one agent: obtain a potent mixture of tricothecene mycotoxins, the terrorist need only make a slurry of corn meal, add contaminated corn or a strain of the appropriate fungus. The mixture would be put together with nutrients and an antibiotic such as streptomycin to increase the growth rate and yield. Then, the terrorist just sits back and lets the toxin-producing mold grow. At the appropriate time, the mixture is dried, ground, and the tricothecenes extracted with alcohol. The alcohol is then permitted to evaporate. The residue is a potent mixture of a wide variety of tricothecene mycotoxins, more potent than any one individual strain by itself. It is nearly as potent as a nerve agent if aflatoxin (made from peanuts) is added....And the only protections required in the manufacturing process are surgical gloves and a mask. (1987: 23-4)

Mullen suggests that, for a terrorist, acquiring a seed culture from the natural environment may be the preferred method on the grounds of maximum security (1978: 76).(27) Mengel agrees that this is the "most secure source," but adds that it "requires the terrorist to sample, isolate, and identify the organism. Obviously, this...method cannot be accomplished by the layman" (1976: 456). Nevertheless, Berkowitz et al. go into considerable detail in describing how this might be done in the case of BTX:

In general, the steps to be accomplished in order to acquire an adequate supply of the toxin are: obtain, culture, and isolate samples of C. botulinum; characterize the cultures as to type, seeking a type A culture showing high toxicity on animal tests; grow this strain in quantity; and finally, concentrate and purify the toxin produced to minimize the total amount of material needed for the attack....

...The organism or its spores are widely distributed throughout the world, being found in soil from sea-bottom to layers of glacial origin at altitudes of 11,000 feet. It has been isolated from river and lake sediments and from both virgin and cultivated terrestrial soil....Because the procedures for isolating, culturing and characterizing the Clostridia are so important in both the processed food and medical areas (other Clostridia being the causative agents of tetanus and gas gangrene), they have become more or less standardized and are widely documented.... (1972: VIII-46)

The obvious short-cut is to steal or buy the agent ready-made. Potential sources here include biological warfare research facilities, university and public health research laboratories, pharmaceutical research laboratories, and mail-order companies. Many authors express concern about the lack of adequate security surrounding such facilities. Mullins, for example, warns that "biological warfare research...facilities are not as well-guarded as nuclear facilities. Terrorists could penetrate one of these facilities, and unlike in a nuclear facility, quickly steal a biological agent or just release an agent into the atmosphere and then leave" (1992: 103)28. Civilian research facilities presumably are even less secure. Douglass and Livingstone suggest that:

A terrorist posing as a graduate student or medical researcher involved in the field of epidemiology would likely have access to numerous pathogenic cultures that could be stolen with little effort or fear of exposure....Medical and research personnel could also be blackmailed into turning pathogens over to terrorist cells. Most laboratories engaged in biological or related medical research have inadequate physical security, and little screening of personnel is routinely conducted. (1987: 24)

Berkowitz et al. agree with this assessment:

Assuming the terrorist is not a 'qualified professional investigator,' a clandestine visit to a research laboratory in which the desired organism is being used could provide sufficient inoculum for a seed culture; so little is needed that the theft, if carefully carried out, would likely never be noticed. This method of acquisition has the advantage that pure strains of known virulence can be obtained. (1972: VIII-69)

Of course, representatives of the biotechnology industry may dispute this charge. One, testifying before the US Senate Committee on the Judiciary in 1989, maintained that "it would be quite, quite difficult to get into a laboratory and steal a strain of dangerous material....they are aware of the need to supply security...both for public safety reasons and for protecting their commercial property for proprietary reasons" (SCJ 1990: 88).

Another possible route is through mail-order companies supplying organisms for legitimate medical and research purposes. Karisch notes that "Until recently, all desired strains of viruses or bacteria could be purchased without any problem from the American Type Culture Collection (ATCC)" in Rockville, Maryland (1991). According to Jenkins and Rubin, writing in 1978: "Some organisms—including most of those suitable for biological warfare—are restricted, but this means only that the person ordering must be confirmed as a qualified investigator. The controls are not tight: the signature of a laboratory or department head is sufficient proof. It would be about as difficult as forging a prescription for an unqualified person to obtain these cultures" (1978: 226).

Douglass and Livingstone, writing a decade later, describe similar conditions. Declaring in their typical fashion that "marijuana is more closely regulated in the United States than access to and distribution of most deadly biological cultures" (1987:24), they go on to note that professional trade journals routinely carry advertisements for such cultures; "The only stated requirement is that the company providing the cultures must have reason to believe the recipient is appropriately trained and has the proper laboratory facilities to safely handle the pathogens. An acceptable letterhead, and a description of the type of work to be performed and the equipment available, should suffice" (1987: 25). Further, according to Douglass and Livingstone, "the cost of most of the specimen cultures is less than the cost of a handgun. Bacillus anthracis specimens cost about $35. One supply house, promoting the 'sale' of five toxins for the price of four, offered five sample toxins, including T2 toxin (the so-called Yellow Rain in Afghanistan), for $100. Some years ago, a Japanese company sold tetrodotoxin in powdered form for $5.97 per gram" (1987: 25). In 1976, according to another source, a British military lab, the Microbiological Research Establishment, was similarly promoting the sale by mail of infectious bacteriological organisms (Clark 1980: 108).

Douglass and Livingstone point to another danger inherent in civilian research facilities, "far less public and hence preferable to terrorists":

As researchers in the field know, there are no controls over the use of bugs by the recipient researcher or institution, except those that are self-imposed. The only federal controls are those associated with animal or human testing. There are some containment requirements for high-risk experiments in the area of genetic engineering, but in general no one really knows what anyone else is doing, or is particularly concerned....Culture swapping is a common practice, and largely uncontrolled.

...Because of the suspicion of law-enforcement authorities characteristic of many academics, it is questionable whether thefts or misuse of pathogens would be reported. According to the BDM study, Dr. C. Don Cox, chairman of the public affairs committee of the American Society of Microbiology, expressed his belief that university personnel would probably not report improperly documented requests for deadly cultures to law-enforcement officials. (1987: 26)

Finally, a number of authors have discussed the likelihood of terrorists obtaining biological agents from "friendly" governments, especially those with established biological warfare programs of their own (Mullins 1992: 103; Simon 1989: 7, 12, and 22; McGeorge 1986: 59-60). As Simon puts it: "Terrorists contemplating a biological attack would find it easy to obtain virtually any type of biological agent, as well as instructions in how to use such agents, once they acquired government support" (1989: 7). Similarly, McGeorge asserts that "Several potential patron states could easily supply suitable cultures from their own biological warfare stocks, disease control labs or universities" (1986: 59-60). Simon even speculates that "Foreign governments could...give false assurances to terrorists about the safety of using biological agents to get them to carry out an unknowingly suicidal attack that would thus eliminate any connection to the state sponsor" (1989: 12). However, at the same time, he cautions that a state would want at all costs to avoid being linked (and thus held responsible) for an act "with the potential backlash" of biological terrorism, and suggests that another inhibition on state sponsorship would be the fear that "a group that is supplied with biological weapons could some day use those weapons against the supplying state itself or embark upon unauthorized operations" (1989: 7, fn.3).

Means of Delivery

A great number of conceivable means of delivery of a BW agent exist, of course, depending on the target chosen and the scale of the attack. Possible methods include:

Nevertheless, although many fictional scenarios might suggest otherwise, most specialists agree that (at least for mass destruction purposes) the effective delivery of a BW agent is more problematic than its production by a terrorist group (Douglass and Livingstone 1987: 14; Kupperman and Trent 1979: 57-8 and 65; Jenkins and Rubin 1978: 226)29. One popular scenario, for example, would have terrorists poisoning the water supply of a population centre by dumping a biological agent into its reservoir. According to Kupperman and Trent, however: "it is a myth that one can accomplish [mass destruction] by tossing a small quantity of a 'supertoxin' into the water would be virtually impossible to poison a large water supply: hydrolysis, chlorination, and the required quantity of the toxin are the inhibiting factors" (1979: 58 and 65). Mengel agrees: "Contrary to popular belief, the water supply is not a highly vulnerable target....Any of the highly lethal biological agents are not effectively transmitted by water and would be further debilitated by the purification system" (1976: 455). According to Roberts: "...contamination of a municipal water supply would require compensation for a significant dilution factor and hence quantities of biological agent beyond what terrorists might find it easy to acquire or transport (in any case such supplies are already carefully screened for contamination)" (1993: 77-8)30. With regard to botulinum toxin, Jenkins and Rubin write:

An ounce or two in a reservoir of 10 million gallons would theory, kill anyone who drank a half-pint of water. Fortunately, there is a difference between theoretical and practical toxicity. It would be extremely difficult to disseminate the toxin evenly throughout the water supply; its presence would probably be detected, and boiling the water would suffice to make it harmless. Even at lower levels of heat, botulinum toxin rapidly loses its effect. (1978: 224)

On the other hand, there appears to be some disagreement among the "experts" on this subject; one James Reynolds of the London Pharmaceutical Society has been cited to the effect that "a terrorist with no particular expertise could feed the bacteria [referring to escherichia coli] into the water system or other public facilities....while one kilogram 'might not' harm a city the size of London, several kilograms would be enough to defeat the antibacterial agents in the water system" (Clark 1980: 109). Similarly, Lowell Ponte suggests that measures such as chlorination "would not necessarily kill such germs as the hardened anthrax created during the military's CBW program" (1980: 52)31. For his part, Donald Louria (Chairman of the Department of Preventive Medicine and Community Health of the New Jersey Medical School) describes as a "realistic scenario for 1990" the dumping of a vial containing billions of genetically-engineered bacteria into the water supply of a medium-sized city, resulting ultimately (through contagion) in the deaths of "millions." Calling such a weapon "a terrorist's dream," he adds: "The development of antidotes, vaccines, or new antibiotics might take years. Even if a vaccine were developed, the terrorists could introduce another organism with new toxins" (1981: 21).

Several authors posit the contamination of foodstuffs as a likely avenue for terrorist use of BW agents. Kupperman and Kamen, for example, suggest that "they could be inserted into production lines at factories turning out packaged prepared foods, the same foods that come in 'tamper-proof' containers" (1989: 107). Similarly, Mengel notes that "Attacks through bulk foodstuffs or beverages (via dairies, meat processing plants, canning companies, bakeries, and soda and beer bottlers) offer the terrorist a means of attacking either particular groups or a broad cross section of society, depending on the specific facility attacked" (1976: 455). Griffith asserts that "food is especially suitable to chemical or biological contamination for attacking large segments of the population," quoting "national Civil Defense authorities" to the effect that "Because of the characteristics of food manufacturing processes, and the nature of certain foods and their ingredients, many segments of the food industry are extremely vulnerable to introduction of biological and chemical agents" (1975).32 In regard to botulinum toxin, however, Kupperman and Trent, after noting that "As should be expected, the food processing industry is intensely concerned about botulinal contamination," go on to state that: "Although technically feasible, and very frightening, a terrorist attempt to contaminate canned products would be of limited effect" (1979: 65).

Most authors suggest that some kind of aerosolization process would be most likely to be used by terrorists employing biological weapons (Kupperman and Smith 1993: 40; Mullin 1978: 77; Berkowitz et al. 1972: VIII-51 and 78ff).33 Mengel, for example, posits an attacker using anthrax or cryptococcosis "simply driv[ing] through a medium-sized city using a truck-mounted dispenser....Anyone exposed for two minutes would probably inhale enough to be infected. Not all the victims would receive lethal doses, but the medical care problems associated with tens of thousands of cases of anthrax infection in themselves would be catastrophic for a community" (1976: 447).34 Douglass and Livingstone appear to agree, noting, in reference to Q fever, that "a truck or car equipped with a cheap commercial aerosolizer, such as used to spray trees and plants, would pose a major threat to a large city or seat of government" (1987: 17)35.

Douglass and Livingstone also describe a scenario in which a tanker ship with powerful pressurized aerosol generators and external booms circumnavigates Manhattan Island, allowing the terrorists to escape before the authorities could pinpoint the source of the attack (1987: 37-8). Berkowitz et al. depict an almost identical scenario, using anthrax:

Under good meteorological conditions and with a light wind (12 km/h) from the southeast, a small boat could make the 32 km run from Battery Park (the southern tip of Manhattan) to City Island (the entrance to Long Island Sound) in about 3 hours at 6 knots. With a culture containing 109 spores/ml and producing aerosol at the rate of about 500 ml/min, a total of 90 liters (24 gallons) of agent are required....If only half the target personnel are actually exposed; if only half of those develop pulmonary anthrax; if only half the cases result in mortalities (all conservative assumptions), more than 600,000 deaths will ensue. (1972: IX-7-8)36

Livingstone cites an estimate by the Stockholm International Peace Research Institute that, operating in a crop-duster fashion, "one aircraft could infect an area of almost 1,500 square acres with yellow-fever virus in a single outing" (1982: 115).

Still, according to Mullin, there are problems with aerosolization:

A significant problem in the aerosol dissemination of almost any biological agent, is the survival of the agent long enough to infect the intended target. The mechanical stresses in the aerosolization process may kill a significant proportion of the pathogenic agent. Moisture in the air, sunlight, smog, radical temperature changes, and other factors may contribute to reducing, through death of significant numbers of organisms in the agent, the virulence of that agent.37 Thus, as with chemical agents, it is misleading to equate the number of LD50 doses an adversary may possess with the number of LD50 doses delivered, irrespective of a host of other problems associated with aerosol delivery.

Nevertheless, noting the hardiness of anthrax in particular, he continues:

...if an adversary possessed some basic understanding of meteorology, the biological characteristics of the agent he chose to employ, the requirements for and effects of aerosolization, was careful in the selection of the target population, and was aware of the various temporal and spatial conditions which would affect the aerosol dispersal of a particular organism, a significant threat could arise. (1978: 78)

Berkowitz et al. echo these points:

Once the aerosol cloud is released, the viability of a particular organism depends on relative humidity, atmospheric composition, temperature, and radiation. At this point, a little meteorological expertise would contribute to the effectiveness of a terrorist attack. From what has been said, it is evident that evening release of the aerosol on cool, humid days would prove most effective. A cloud released to drift over crowded, large cities during the wintertime, when resistance to respiratory infection is generally low, would probably be most effective. A moderate wind, some turbulence (to prevent the aerosol from settling), and an inversion layer (to confine the cloud to lower altitudes where the target population is found) are other desirable meteorological conditions. (1972: VIII-84)

They also discuss the precise type of disseminating device most suitable, noting that research conducted at Fort Detrick and reported in the open literature (Rosebury 1947)

...describes a variety of atomizer designs, the criteria by which preferred designs were selected, and the procedures for fabricating them. Paint spray devices are rejected because of their large particle size, but artist's air brushes are found to be possibly applicable where small output is sufficient. For limited terrorist or sabotage use, they may be perfectly adequate.

For somewhat larger but still quite limited output the common, self-contained, aerosol dispenser is available. This would be appropriate for attacks in virtually any small to medium size enclosed space such as a command center, submarine, aircraft, legislative chamber, banquet hall, etc. (1972: VIII-81-82)

Douglass and Livingstone point out the difficulties of attempting to selectively use biological agents in an open area, by reference to Israeli settlements in occupied Arab territory: "Because of their great lethality and dependence on winds and weather, the use of biological agents on the crowded West Bank by Palestinian terrorists...would be an act of madness, and one as threatening to the indigenous Arab population and the surrounding Arab states as to the Jewish population" (1987: 14).

Virtually all authors agree that a smaller-scale attack confined to an enclosed area (but still capable of causing massive casualties) would be the most feasible option for terrorists. In the words of Jenkins and Rubin: "The smaller the target, the more likely terrorists are to succeed, especially if the target population resides or works within a sealed building with central air conditioning" (1978: 226-7). Similarly, Griffith maintains that "BW within a city will most likely be directed against specific buildings or specific limited parts of the population. Dangerous bacteriological organisms can be easily introduced into a building a number of ways, the best being the water system (in this case limited in volume and with more direct access) or the ventilation system" (1975). Domed sports stadiums have been described as "ideal" targets for such an attack38; according to Mengel: "Using approximately one fluid ounce of either anthrax or cryptococcosis in aerosol form would result in the inhalation of an infective dose within an hour" (1976: 447).39

Another possibility that has been mentioned in the literature is dispersal through a city's subway system. According to one source, this would require only one to two liters of anthrax spores and "could infect millions of people" (Karisch 1991). One imaginative method of delivery in such a scenario is described by Livingstone:

Live bacteria can be carried safely and conveniently by terrorists in light bulbs. When the release of the bacteria is desired, all the terrorist has to do is leave the light bulb on a subway track or roadway where it will be broken within a short period of time. The delay, however, permits the terrorist to remove himself safely before that happens. (Livingstone 1982: 115)

In fact, the US Army in the 1950s demonstrated US vulnerability to BW attack (and evidently the effectiveness of certain means of delivery) by releasing the harmless bacterium Bacillus subtilis into various New York City subways and Washington, DC, subterranean passageways (Kupperman and Smith 1993: 39; Root-Bernstein 1991: 48; Simon 1989: 3, fn.5; Watkins 1987: 196; Cole 1988). In similar tests during the same period, US Navy minesweepers "attacked" San Francisco with sprays contaminated with Bacillus globigii and Serratia marcescens, reportedly infecting virtually all the residents of the city (Douglass and Livingstone 1987: 38) (117 square miles, according to another report—Watkins 1987: 196). According to Lowell Ponte, who provides the most detailed list of cases,

Between 1949 and 1969, U.S. Army CBW teams carried out 239 open-air tests, simulating germ warfare attacks on the populace, 79 of which infected large numbers of unknowing citizens with disease-causing germs. Among the targets: National Airport and the Greyhound bus depot in Washington, D.C., in 1965; two tunnels on the Pennsylvania Turnpike and a stretch of one Pennsylvania state highway, in 1955; several beaches in Hawaii, Virginia, Florida, and California, including a 1968 test near San Clemente, California; and the New York City subway system, in June 1968. (1980: 52)

Watkins adds that "In 1952, the Army Chemical Corps Special Operations Branch at Fort Detrick contaminated the Pentagon air-conditioning system with a pint and a half of bacteria in a mock attack that demonstrated the vulnerability of large office buildings" (1987: 196). Referring to the 1950 case of Seriatta marcescens in San Francisco, Ponte reports: "Military monitors found that any attacker could thus readily infect millions of people more than 20 miles away, but to this day the Army denies responsibility for the subsequent outbreak of Seriatta marcescens-linked pneumonia that caused the death of one hospital patient" (1980: 52)40. Douglass and Livingstone conclude that "Although an attack on a single building or facility, using the ventilation system to transmit the agent, remains the most likely scenario, nevertheless, a major attack on a U.S. city cannot be ruled out" (1987: 37).

Incidents of Past Use or Threat

Over two dozen specific instances of terrorist use or threat of use of BW agents are cited in the open literature, ranging from apparently empty threats to use such agents without any evidence of their actual acquisition, to serious attempts at acquiring them, the actual discovery of quantities of BW agents in terrorist or potential terrorist hands, and a very few reported cases of the actual use of such agents. Various authors disagree over the specific facts of such cases, including their precise dates, and even apparently over the definition of what constitutes "terrorism" as opposed to, for example, criminal extortion. There have been many reported cases of threats (as well as actual incidents) of contamination of foodstuffs, for example, but most of these have been motivated by strictly financial rather than political gain, and hence are not normally considered as "terrorist"41. The actual use of the toxin ricin in several assassinations or attempted assassinations also appears to be questionable as examples of BW terrorism in that the attacks have been attributed to the intelligence services of particular countries.

Judgments vary as to whether there have in fact been any actual past cases of "bio-terrorism," depending on the particular author's own definition of the term, although all agree that the potential is present and go on to cite reported threats, if not actual use, of BW agents. According to Root-Bernstein, for example, "no overt acts of bioterrorism are yet on record" (1991: 48). Roberts agrees that "there are no recorded instances of successful bw attack by terrorists" (1993: 77)42. The OTA is more equivocal in its judgment, stating simply that "There has been, as yet, no major case of a terrorist attack with biological weapons" (emphasis added)43 and that "There have been many more threats to use these agents than known preparation for use or actual use" (1992: 37 and 40). Douglass and Livingstone note that "There are several suspected cases of biological terrorism, but proof is hard to come by" (1987: 32). They go on to caution that "many incidents are not recognized for what they are and therefore go unreported" (1987: 187).

The extent to which terrorists have even evinced interest in the use of biological agents is also subject to some controversy. At one extreme, Simon argues that "there is no substantive track record of biological-weapons attacks by terrorists—beyond a number of threats and a few 'low-level' incidents....Moreover, neither interviews with terrorists nor terrorists' own writings have made specific reference to the use of biological weapons" (1989: 2; emphasis added). Similarly, ter Haar asserts that "The possibility that biological weapons might be used by terrorists is often mentioned, but there are very few reports that any terrorist group has ever tried or threatened to use them" (1991: 57). On the other hand, according to Jackson, "throughout the 1970s and 1980s the Red Army Faction, Red Brigades, and more extreme Palestinian elements, put considerable effort into attempting to recruit microbiologists, purchasing bacteriological experimentation equipment, and dabbling in sending toxins including anthrax through the post to potential victims" (1992: 520)44. In any case, the record so far appears to support the view, as expressed by Alexander, that at least some terrorists "seriously have considered resorting to biological terrorism" (1983: 230)45.

For purposes of this paper, past publicly-reported incidents will be classed into various categories, according to their degree of "seriousness," as follows: (1) threats to use BW, without any evidence of actual capabilities; (2) unsuccessful attempts to acquire BW; (3) actual possession of BW agents; (4) attempted, unsuccessful use of such agents; and (5) their actual, "successful" use.46 In the first category, then, are the following:

Examples from the second category, attempted acquisition, include:

The third category of the actual, successful acquisition of BW agents is, of course, even more troubling. Reported instances include:

Prosecutors did not specifically identify anyone as a target for the ricin, and charges have not yet been filed against two unidentified men suspected of setting in motion the plan to cultivate castor beans. The two unidentified males, along with Wheeler, belonged to a tax protest group called The Patriots Council. The three had discussed blowing up a federal building, obtaining assault weapons, and killing Internal Revenue Service agents, deputy US marshals, and a sheriff's deputy. In addition, Patriots Council members discussed poisoning US agents by placing ricin on doorknobs. (CSIS 1995); and

The fourth category constitutes attempted, unsuccessful use of BW agents. Only three specific examples can be found in the open literature examined for this report:

Finally, the last category deals with cases of actual, successful use. Although some may be considered questionable as cases of politically-motivated terrorism, as discussed above, the following are cited in the literature:

...established a small biological agent production facility at his northwest Washington, DC home. Using cultures of Bacillus Anthracis (Anthrax) and Pseudomonas Mallei (Glanders) supplied by the Imperial German government, Dilger produced an estimated liter or more of liquid agent. Reportedly, the agent and a simple inoculation device were given to a group of dock workers in Baltimore who used them to infect a reported 3000 head of horses, mules and cattle destined for the Allied forces in Europe. Allegedly, several hundred military personnel were also affected. (1994: 12);

Reasons for Non-Use

It can be seen from the above summary that, although terrorists have at times certainly evinced a serious interest in acquiring and using BW agents, and there have been a few reported cases of their use or attempted use, the record is actually quite sparse. In particular, with the possible exception of the Rajneesh poisoning episode, there appear to have been no cases of the type of mass-destruction attack that has been the subject of so much speculation, as was discussed earlier. A number of authors have queried why this should be the case, why terrorists have not made greater use of BW agents given their assumed technical capacities to do so and the existence of a substantial literature warning of the possible threat.

Many of the possible reasons for non-use cited in the literature have to do with the unpredictability of biological agents, given their capacity to reproduce and to be affected by a wide variety of environmental conditions. For example, Wiener writes that "the ability to contain and control and limit the effects of such weapons is in question. This is a very important thing. These things can go wild, especially if it is an organism that is capable of free living, and not a toxin" (1991b: 70). Similarly, Baum notes that "Such factors as wind patterns and temperature dramatically affect a biological attack" (1993: 17). And Simon: "Unlike conventional or even chemical attacks, which usually can be limited to the intended targets, biological weapons represent virtually unexplored terrain for terrorists. They cannot estimate the consequences of dispersing microorganisms into the atmosphere" (1989: 12). Or Mullins: "As with any weapon, one concern is the ability to control the weapon to the extent that only the target audience is affected. With biological agents, the difficulty in controlling the agents is the major drawback in their usage" (1992: 103, 107).

As Mengel puts it: "...biological technologies are principally characterized inherent variability of effectiveness that makes their application unpredictable....The range of potential lethality within the spectrum of biological agents is indicative, in part, of the difficulty in preparation, delivery and dissemination problems, and resilience under differing environmental and meteorological conditions" (1976: 446). In other words, the problem faced by the terrorist in contemplating the use of a biological agent is uncertainty about whether it will work at all, or whether its effects will be magnified all out of proportion to the original intention. In the words of Jenkins and Rubin: "...biological weapons tend to be highly unpredictable. They can produce negligible results, or a worldwide epidemic that makes no distinction between friend and foe" (1978: 225).51

The factor of unpredictability should not be overly exaggerated, of course. To some extent, it is shared by other types of weapons. Moreover, as Mullins cautions: "The degree of difficulty of control...would be determined in part by how the target audience was defined. If the target audience were the entire population of a major city or the entire population of the United States, then there would be no drawback to the use of biological agents" (1992: 116). Indeed, some forms of terrorism depend for their effect precisely on their level of indiscriminateness. As Watkins puts it:

The uncontrolled tactical effect of biological agents, the high risk they pose to an attacker and the assurance of retaliation have led most national governments to ban their first use. None of these characteristics are likely to dissuade terrorists, and in fact, many characteristics of these agents are highly desirable features only to such terrorist groups. (1987: 191)52

Related to the unpredictability of the weapon is the oft-cited fear of terrorists for their own personal safety. To some extent, this fear may be well-placed; Wiener points out that "There were a lot of casualties at Fort Detrick in the early days, when they were weaponizing systems, even when they were immunized against things like anthrax" (1991b: 70).53 And Mullins warns:

Certain methods of dispersal would be just as risky for the terrorists as for the victims. For example, it would be extremely difficult to control the biological agent using an open area dispersal method. The terrorists would be just as much at risk as the target audience. It would only be a little less risky to release infected animals in a target area. (1992: 103, 107)

Jenkins and Rubin believe that "Fear of catching a dangerous disease probably dissuades many potential users from employing biological weapons." However, they add: "A trained person...would be aware of the necessary precautions" (1978: 226). Similarly, the OTA study concludes: "Such weapons may pose a risk to their users, but this can be overcome, at least to a degree, by the use of protective clothing and masks, or, in some cases, by vaccines" (1992: 37).54 Finally, Baum points out that "there will always be terrorists willing to die for their cause who would not be deterred by the uncontrollable nature of biological agents" (1993: 17).

Another related disincentive to the terrorist use of BW is the danger of collateral damage to non-targets, again depending on the scope of the attack and/or terrorists' compunctions in this regard. For example, Kupperman and Kamen cite as one factor militating against the development or use of biological agents "the likelihood that they might backfire—infecting friends as well as enemies" (1989: 105). In arguing that chemical weapons are likely to be preferred over biologicals, Douglass and Livingstone assert:

The fact that they are not infectious is important to terrorists chiefly from the standpoint of safety. This is important both with respect to fellow members of their own community who may live in proximity to the intended victims of the attack, and to the fact that safe houses where the dangerous agents are grown and processed are likely to be located in areas or neighborhoods populated by support groups, coreligionists, or members of the same ethnic group or race. (1987: 13)

Other authors appear to assume that the terrorists in question will have moral qualms about injuring innocent persons or causing widespread havoc, even affecting future generations. For example, Jenkins and Rubin note that "those most likely to be severely affected by disease are the elderly, the very young, and the infirm, those who by no means could be called combatants" (1978: 225)55. In a similar vein, Douglass and Livingstone write: "What makes plague such a frightening weapon is that once a human chain of victims is started, the disease might continue to spread, unchecked, in secondary and tertiary outbreaks, for years to come, especially if an antibiotic-resistant strain were employed" (1987: 38-9). (Of course, whether this is considered a drawback or a boon by terrorists depends on the latter's motives and ambitions in launching the attack in the first place.) The OTA study appears to be thinking of radical "eco-terrorists" in noting that "Infection of other species (i.e., cattle, rodents, domestic animals) or spread to other neutral countries might be a major problem" that "would have to be taken into account by any state or sub-national group considering use of biological weapons" (1992: 39).

Whether or not terrorists were themselves deterred by moral qualms from using biological agents, they would, presumably, have to take into account the reaction both of the targeted group or government and of the body of potential sympathizers to their cause. Several authors, for example, believe that terrorists might be dissuaded by the anticipated severity of the governmental response to their action, which could lead to their own annihilation (Wiener 1991a: 130; Simon 1989: 11). In Wiener's words: "...terrorists may wish to avoid an extremely severe response against people everywhere....if they are told that terrorists did something like this, they are going to help find those terrorists because they would consider the terrorists to be out of control" (Wiener 1989: 59). Elsewhere, he observes: "...successful use may produce so much civilian terror and loathing that the terrorists' cause is damaged. The terrorists want to get sympathy and publicity for their cause....If there is indiscriminate, widespread killing, they could bring the roof down on themselves" (Wiener 1991b: 70). Similarly, Roberts writes that "Terrorists apparently prefer bloody weapons that make good public theater and therefore draw attention to their claims through the media rather than weapons generally deemed abhorrent that might delegitimize their cause" (1993: 78). The OTA, on the other hand, warns that "terrorists have not balked at mass killing" (1992: 37), while Baum notes that concern about anticipated international revulsion "may not apply to some terrorist organizations which might employ biologics but deny responsibility" (1993: 17).

Leaving aside the possible technical obstacles discussed earlier in the section on "Requisite Capabilities," several other factors, some related to the immediately preceding points, have been cited as inhibiting terrorist use of biological agents:

...unpredictability and the difficulty in ascertaining whether an outbreak of disease is a natural occurrence or is the result of a terrorist act, in the view of some authorities, weigh against such weapons being chosen by terrorist groups. In many cases terrorist activities occur for the purpose of making a political statement, and if the source of a disease cannot be identified and claimed by the terrorist group, the effectiveness of the political statement is lost. (1993: 26);

Current Trends/Likelihood of Future Use

Despite the relatively low incidence of the use or threat of use of bioterrorism in the past, a number of authors speculate, based on various trends, that its likelihood may increase in future (Kupperman and Smith 1993: 45; Roberts 1993: 78; Simon 1989: 22; Mullins 1992: 103; Griffith 1975; Watkins 1987). Only Mullins states outright that "The likelihood of terrorist organizations using biological agents is high" (1992: 103); others are more circumspect. For example, the OTA study declares simply that "future use of these agents cannot be excluded since they already have been used or proposed for use in the past" (1992: 40). Ambassador H. Allen Holmes, the US Assistant Secretary of State for Politico-Military Affairs, testified in a similar vein in 1989: "To date, we have no evidence that any known terrorist organization has the capability to employ such weapons, nor that states supporting terrorism have supplied such weapons. However, we cannot dismiss the possibility that terrorists could acquire BW" (SCJ 1990: 29). Later on in the proceedings, he declared: "...given the stealth with which this material can be developed and hidden and used, we do consider it a serious potential threat" (SCJ 1990: 58). Similarly, US Deputy Assistant Attorney General Ronald K. Noble judged that "the potential for such use is clearly present and poses a danger that should be promptly addressed" (SCJ 1990: 47). US Under-Secretary of State Reginald Bartholomew, in testifying before another Senate committee in June 1992, employed language identical to that used by Ambassador Holmes three years previously, but framed in terms suggesting greater urgency:

We are especially concerned about the spread of biological and toxin weapons falling into the hands of terrorists, or into the arsenals of those states which actively support terrorist organizations....If the proliferation of biological weapons continues, it may be only a matter of time before terrorists do acquire and use these weapons. (US House Committee on Armed Services 1993: 25)

Simon, although also appearing equivocal on the question of whether bioterrorism has in fact occurred in the past, endorses the view that it will likely occur in future: "...there are several reasons to believe that biological weapons will eventually be seriously considered—and probably used in some manner—by terrorists" (1989: 14); and further:

The possibility that terrorists may one day use biological agents cannot be ignored. Although the risk today is quite low, the technological, logistical, and financial barriers to using such weapons are not insurmountable....While it is not necessarily true, as some observers have claimed, that 'time is running out,' we cannot afford to be complacent about the threat of terrorists using biological weapons. (1989: 22)

For his part, given the suitability of biological weapons to terrorist (especially as compared to military) use, and apparently disregarding the option of state sponsorship, Watkins concludes that "the principal biological threat against which nations should try to defend is posed by terrorists rather than by foreign national powers" (1987: 197). Specifically, he believes that the "use of biological agents by terrorist groups poses a significant threat to the domestic security of the United States" (1987: 191).

In judging that the likelihood of use is increasing, Simon and others point to a number of current trends that may be deemed to fall into two broad categories: (1) changes in the nature of terrorism itself; and (2) broader global trends. Included in the first category are the following:

In recent years, terrorists have found it necessary to launch more dramatic and violent attacks to attain the same degree of publicity and government responses that smaller incidents previously generated. With terrorist attacks occurring on an almost daily basis, the public and the media have become somewhat desensitized. And with a multitude of terrorist groups 'competing' for the international spotlight, more dramatic incidents are likely in the future. It would be difficult to conceive of a terrorist incident more dramatic—or guaranteed to gain more attention—than one involving biological agents (Simon 1989: 12)57;

The second category, looking at broader international trends, includes the following:

Candidate Groups

Two of the studies consulted for this report go on to describe the profiles of terrorist groups most likely to employ BW. Jeffrey Simon devotes the greatest amount of attention to this subject, suggesting that candidate groups would "probably exhibit the following characteristics":

(1) "a general, undefined constituency whose possible reaction to a biological-weapons attack does not concern the terrorist group" (1989: 17). On these grounds, he eliminates "nationalistic" groups59 such as the Irish Republican Army (IRA) and ETA that he assumes would be inhibited by public reaction to "the much greater violence of a biological attack and the moral implications of using biological weapons" (1989: 16). On the other hand, he proposes as likely candidates groups such as the Japanese Red Army (JRA), "whose goals and objectives include vague notions about world revolution" (1989: 17); European left-wing groups such as the Red Army Faction (RAF), who "also have undefined constituencies and vague objectives" (1989: 18); and neo-Nazi groups in the US and Europe who "would also be unlikely to feel any restraints because of public opinion" (1989: 18);

(2) "a previous pattern of large-scale, high-casualty-inflicting incidents" (1989: 17). Here, Simon rejects groups such as the Italian Red Brigades, Belgian Communist Combattant Cells, and West German Revolutionary Cells, while acknowledging that they "could be potential perpetrators of 'exotic' types of assassinations (e.g., shooting poison pellets into victims)" (1989: 16). As for "large-scale" biological attacks, however, based on the level of violence of past activity, he proposes groups such as the JRA, Sikh extremists in India, pro-Iranian Shiite fundamentalist groups such as Hizbollah, and Palestinian extremists such as the Abu Nidal organization, adding that "since there are many factions within both the Islamic and Palestinian movements, internal divisions could lead certain factions to decide that a higher level of killing is necessary to preserve their own goals" (1989: 19);

(3) "demonstration of a certain degree of sophistication in weaponry or tactics" (1989: 17). Here Simon cites the Popular Front for the Liberation of Palestine—General Command (PFLP-GC) as "technologically very sophisticated" (1989: 17); and

(4) "a willingness to take risks" (1989: 17).

Nevertheless, Simon cautions that "The nature of terrorism precludes predictions of the exact target, tactic, or weapon that a terrorist group may use, and situations can change or opportunities arise that may make even the most unlikely terrorist group consider using biological agents" (1989: 20).

Like Simon, the US Congressional Office of Technology Assessment (OTA) identifies terrorist groups most likely to use BW as those having "one or more" of four different characteristics. The first three suggested by the OTA are virtually identical to those posited by Simon: (1) "a large base of popular support that they are not concerned about alienating" (this is somewhat different from Simon's first characteristic in presuming "a large base of popular support," as opposed to Simon's fixation on goals, but similar in supposing a disregard—on whatever grounds—for public opinion); (2) "a history of large-scale violence with high numbers of casualties per attack"; and (3) "prior use of sophisticated weapons" (1992: 40). However, the OTA study replaces Simon's fourth characteristic with "state sponsorship." Interestingly, many of the same specific groups suggested by Simon as likely candidates for bioterrorism make the OTA's list: Japanese Red Army, Red Army Faction, US white supremacist groups such as the Aryan Nations, Hizbollah, and the Abu Nidal Organization (1992: 40).

Finally, Simon goes one step further and lists as "indicators that could point to a terrorist group planning a biological-agent attack" the following: "recruitment of new members who have scientific backgrounds; contacts with scientific laboratories or attempts to purchase or steal biological agents; and suspicious inquiries by individuals concerning infectious diseases" (1989: 20-21).

Defence Against Biological Terrorism

Discussions of possible defence against bioterrorism are marked by a curious dualism. On the one hand, there appears to be widespread agreement on the difficulty of "early-warning," i.e., detecting the presence of biological agents in a particular situation in time to take protective action. For example, writing as recently as 1989, Wiener stated outright that:

Early warning systems do not exist. Experimental detectors have been developed [but]....they may sound an alarm often when there is no attack....There are still some attempts at the development of better detectors, but only at a research level. Attempts to detect infectious agents prior to inhalation have not been successful. (1989: 17)

Again: "There is no reliable method of prior detection....the first knowledge of an attack will be a large number of casualties" (Wiener 1991b: 65)60. Similarly, the 1992 OTA study states:

An aerosol attack and food/beverage/medication contamination are not normally detectable by the human senses (the agents are invisible, silent, odorless, and tasteless).

No reliable, sensitive, and specific system, whether based on mechanical, laser, electrical, or chemical detectors, is yet available to detect an aerosol attack in time to allow the target population to put on protective masks and clothing, and thus avoid inhalation and infection. This deficiency means that there is risk even from those agents that produce illnesses that can be successfully treated.

Similarly, there is no testing system in place to ensure against food/beverage/medication contamination. In some cases, attacks may be detected by finding delivery vehicles (bomblets, rockets, or bombs containing remnants of agent) or by intercepting aircraft with spray tanks, but such attacks could be planned for miles upwind of the target and go undetected. (1992: 36)

An earlier OTA study had also pointed out that "Detection of biological agents and subsequent (or, frequently, concurrent) diagnosis of the agent causing the symptoms is relatively undeveloped," adding that "in 1976, it took the full resources of the United States Government seven months to isolate the Legionnaires' disease Legionella pneumophila bacterium when it was discovered" (1991: 52). Phillip Karber of Georgetown University's Center for Strategic and International Studies (CSIS) has been quoted as stating bluntly that the US government lacks "the means of identifying or combating a biological attack on the United States" (Ponte 1980: 53).

Despite these severe difficulties of detection and identification, various authors have provided long lists of defensive measures that could be employed to prevent, or at least mitigate the effects of, a BW terrorist attack. Such measures can be grouped into a number of categories: (1) intelligence collection prior to an attack; (2) measures to prevent the acquisition of biological agents by terrorists; (3) "passive" protective measures prior to an attack; (4) "active" measures to counter an attack in progress; and (5) post-attack mitigative measures. Each of these will now be considered briefly in turn.

(1) intelligence gathering. Among the measures proposed under this heading have been:

(2) counter-acquisition strategies. These could include:

Although some agents can be isolated from the environment or from natural infections, this takes time and expertise. A small number of research and medical laboratories have a legitimate need for pathogenic strains, but this access should be restricted using positive audit control procedures in the same way that access to pharmaceutical drugs is controlled. (1987: 198)61;

(3) passive protection. This includes such measures as:

Watkins also calls for changes in immunization policy, noting that the percentage of Americans susceptible to smallpox (those born since 1980, when the US halted its vaccination of civilians) will grow to about 50 per cent by the year 2000 "unless immunization of the public is again undertaken" (1987: 197). Similarly, Wiener argues that "those at risk can be immunized with vaccines to render them less susceptible to the known threat agents....Massive immunization of civilian populations against an abbreviated list of threat agents is feasible if biological warfare becomes a threat to those populations in the future" (1991a: 131-2)62. However, this approach appears questionable in the context of bioterrorism where so much uncertainty remains regarding the precise nature of the threat and its target;

(4) active defences. Kupperman and Smith are the only authors among those consulted who deal with this area. They suggest enlisting the Los Alamos National Laboratory and its "sister" laboratories in the effort to develop "countermeasure technologies": "...the laboratories can construct simulations that predict cloud movement, horizontal and vertical dilution, and residual virulence. Using meteorological data, active defenses, such as high-power ultraviolet (UV) lasers and 'countercloud missiles,' can be deployed" (1993: 45). Elsewhere, they speak of "bleach-saturated 'counterclouds'" and "counter-clouds of disinfectants" (1993: 44-5).


(5) post-attack mitigative measures. These include:

...the United States might establish an emergency preparedness program similar to that for a nuclear accident or incident (such as the Three Mile Island disaster) or that more recently established for response to chemical accidents (such as the Bophal incident in India). Such an activity by the Federal Emergency Management Agency, in conjunction with the Department of Defense, that would examine command and control, technical, behavioral, law enforcement, search, disarmament and decontamination, and treatment needs, would lay the ground work for response to either a terrorist-initiated or natural biological disaster. (1993: 27)

Root-Bernstein may be the most optimistic of all observers in suggesting that, although "there is very little one can do to deter a biological attack directly," measures of the type outlined above "can lower the potential effectiveness of biological weapons to the point where terrorists would probably not consider their use" (1991: 50).


Table of Contents

[Main Menu] [Other Documents and Releases]

Disclaimer: The Canadian Security Intelligence Service assumes no responsibility for the use of the information at this World Wide Web (WWW) site.

© CSIS/SCRS 1999
Canada wordmark