As noted in the previous section, the toxicity of chemical agents generally falls somewhere in-between that of the more deadly biological agents and that of conventional weapons, or at the lower end of the scale for weapons of mass destruction. For example, Kupperman and Trent estimate that, based on "the weight required to produce heavy casualties within a square-mile area under idealized conditions," fuel-air explosives require 320 million grams; fragmentation cluster bombs, 32 million; hydrocyanic acid, 32 million; mustard gas, 3.2 million; GB nerve gas, 800,000; a "crude" nuclear weapon (in terms of fissionable material only), 5,000; Type A botulinal toxin, 80; and anthrax spores, 8 (Kupperman and Trent 1979: 57). Similarly, it has been estimated that it would take 100 grams of the "V" nerve agent, or almost 40 pounds of potassium cyanide, to have an effect on a water supply equivalent to just one gram of typhoid culture (SCJ 1990: 3-4). Put another way, to incapacitate or kill a person drinking less than half a cup of untreated water from a 5 million-liter reservoir would require no less than 10 tons of potassium cyanide, compared to just 1/2 kg of Salmonella typhi (OTA 1991: 52).
As in the case of biological agents, different types of chemical agents vary considerably in their lethality. Of the two principal categories of chemical toxins, fluoroacetates and organophosphorous compounds, the latter are widely considered the more lethal64. At one end of the scale is DFP (diisopropyl fluorophosphate), described as a "relatively mild poison" (Mullen 1978: 69). Another possible chemical agent, the organophosphate TEPP, is the most toxic of the commercially available insecticides (Jenkins and Rubin 1978: 224). The nerve agent sarin, on the other hand, when taken orally, is ten times as toxic as TEPP to humans; according to Berkowitz et al.: "a small quantity of Sarin splashed on the skin is likely to produce a vapor concentration high enough to exceed the inhalation LD50 [mean lethal inhalatory dose] with a single breath" (Berkowitz et al. 1972: VIII-25).65 They go on:
Far more toxic again are the V-agents; "VX, when inhaled, is ten times as toxic as sarin, but dermally it is 300 times as toxic" (Kupperman and Trent 1979: 6566). According to Douglass and Livingstone, "The amount of VX (a nerve agent) that one can place on the head of a pin is sufficient to produce death in a human being" (1987: 17)67. Livingstone reports that "In tests conducted by the army, one drop of VX absorbed through the skin was enough to kill a dog" (1982: 110).
Also as in the case of biological agents, it would be equally misleading to extrapolate directly from individual lethal doses to estimates of casualties from mass attacks, given the need for effective delivery. As Mengel notes:
Similarly, Mullen maintains that "by any measure, it does not seem credible that a chemical threat could be mounted that could result in the magnitude of destruction potentially possible with nuclear or biological weapons....to have some probability of success in causing thousands of casualties in a military operation, even so-called nerve gas gases must be dispersed in quantities of hundreds to thousands of kilograms" (1978: 78, 83).
Chemical weapons such as nerve agents are generally credited with being capable of causing casualties in the range of hundreds to a few thousand (Kupperman and Trent 1979: 63 and 84; Kupperman and Woolsey 1988: 5; Mengel 1976: 446). A few authors put the total much higher, in the same range as for biological or even nuclear weapons. Thus, for example, Douglass and Livingstone write that "Four tons of VX is enough to cause several hundred thousand deaths if released in aerosol form in a crowded urban area" (1987: 17). Clark goes even further, stating that "A canister [of VX] dropped from any tall building or sprayed over a large city from a private plane would kill millions" (1980: 110)68. However, most authors appear to agree with Berkowitz et al. that "even with the best chemical agents available, if the attack effort is kept within the bounds of reason, its impact probably cannot exceed exposure of a few thousand target individuals at one time" (1972: IX-7).69 Berkowitz et al. conclude: "Therefore, this is one of the lesser superviolent threats, but its small resource requirements and the great availability of necessary skills must be kept in mind" (1972: IX-7).
The final characteristic of chemical agents that should be noted here is that, in contrast to biological agents, their effects can be virtually instantaneous. In Mullen's words: "Death from organophosphate poisoning may be so rapid that the afflicted individual may be entirely unaware of what is happening" (1978: 71). According to another source, a one-milligram dose of a nerve agent "can usually kill within 15 minutes" (Joyner 1990: 137).
Despite not being as toxic as the most lethal biological agents, chemical weapons have certain other advantages that may make them more attractive to terrorists. A number of authors maintain that they are cheaper than biological agents (Douglass and Livingstone 1987: 12-13; Alexander 1983: 229; Mullins 1992: 116). For example, Livingstone cites one estimate that "the cost of producing 1,000 kg of GB (nerve agent), based on small laboratory purchases of raw materials, would be in the neighborhood of $200,000" (1986: 143).70 On the other hand, Douglass and Livingstone appear to contradict themselves later in citing a 1969 estimate that, "for a large-scale operation against a civilian population," casualties might cost about $600 per square kilometre with nerve-gas weapons, as compared to just $1 with biological weapons (1987: 16). There can be no doubt, of course, that the manufacture of chemical weapons would be much less expensive than the manufacture of nuclear weapons, for terrorists or for anyone else.
It has also been said that chemical agents are "easier to use" than biological agents (Douglass and Livingstone 1987: 12). This rather vague claim could refer to a number of different aspects. Among those noted by Douglass and Livingstone are their "stability" and the fact that they are "more containable," easily dispersed, and "controllable" ("inasmuch as they are not contagious") (1987: 12-13). Alexander agrees that "their delivery systems are manageable, and their dispersal techniques are efficient" (1983: 229); Mullins that "dispersal is easy and widespread," as well as being "fairly easy to control" (1992: 116). On this latter point Mullins elaborates: "The use of chemical agents could be controlled to a much greater extent than could nuclear or biological agents. The delivery of chemical agents could be accomplished with exact precision, thus insuring that only the target audience was affected" (1992: 111, 116). On the other hand, Mengel argues that, by comparison to biological agents, "chemical technologies...are practically limited by delivery problems" (1976: 446). The issue of deliverability will be dealt with at greater length below.
In contrast to what was said above about the effects of some chemical agents being virtually instantaneous, Mullins maintains that "One major advantage of chemical agents over nuclear devices or biological agents is that by using the right chemicals, any effects could be delayed for a period of time. That is, the agent could be dispersed and it could be days or weeks before any effects appeared" (1992: 111). Why this should be greater in the case of chemical than of biological weapons is unclear. In any case, it is presumably at least partly on this basis that Mullins goes on to cite another putative advantage in using chemical weapons, that "there is minimal risk of detection" (1992: 116)71. Perhaps also related to this factor (or to that of "controllability," in the sense of being non-contagious), Mullins also states that chemical agents "offer low risk for usage" (1992: 116).
In discussing the presumed advantages of chemical weapons, Mullins appears to contradict himself on another point, however. On the one hand, he argues:
However, Mullins goes on to cite as an additional advantage of chemical agents that "most chemical agents rapidly disperse. Thus, the target area would be clear for the terrorists to enter at a later date" (1992: 116). Apart from the apparent internal contradiction here, again it is unclear whether chemical agents would necessarily have an advantage over biological agents in these respects, either of quick dispersal or of relative longevity; presumably it would depend on the particular agent in question. Both chemical and biological agents would certainly appear to have an advantage over nuclear weapons of mass destruction in allowing for comparatively early access to the site of the event, although it is unclear whether this would be more to the benefit of the terrorists or of counter-terrorist forces.73
Douglass and Livingstone have also referred to the relative "ease of manufacture" of chemical, as compared to biological, agents (1987: 13)74. At the same time, however, in another apparent contradiction, they argue that "Whereas chemical weapons require a 'moderately advanced chemical technique,' the raw materials for a biological weapon are readily accessible in most countries and should present little difficulty to terrorists" (1987: 23). Alexander simply describes chemical weapons as "relatively easy to obtain" (1983: 229), thus leaving open the question of manufacture or acquisition by other means. As to the latter point, Ye. Primakov, the head of the Russian Foreign Intelligence Service, has noted that "an additional temptation for the employment of chemical weapons for terrorist purposes is the rather wide employment of toxic substances by the police and special purpose forces of a number of countries" (1993: 5). It is unclear here, however, whether Primakov is referring to the opportunities thereby created for the theft of such material, or rather to the precedent of its being used. Regarding the first of these alternatives, Mullins suggests that, due to the putatively lower level of security surrounding chemical weapons storage sites as compared to nuclear or biological facilities, chemical agents would be the easiest to steal (1992: 109).
In their assessment of the comparative advantages of chemical and biological agents for terrorist use, Kupperman and Trent note that "there is limited commercial availability of deadly pathogens. Moreover, the growth, care, and dispersion of biological agents require more technological sophistication than does the dispensing of chemicals" (1979: 85). Similarly, Mengel refers to chemical agents as "[r]equiring the least amount of resources to manufacture of the technologies examined" (1976: 446). On the other hand, Hurwitz avers, without further elaboration, that "It may be even easier for terrorists to acquire biological weapons than it would be for them to acquire chemical weapons" (1982: 38). Mullins appears to agree with this assessment by locating "chemical terrorism" "on the continuum midway between the technology required to manufacture a nuclear device, and the ease of using biological agents" (1992: 108).
Clearly, there remains disagreement among the authors consulted as to the relative merits for terrorists of chemical and biological agents. Mullins, perhaps the biggest "booster" of chemical agents, declares that "For NBC Terrorism, chemicals are the ideal weapon....Chemical agents...offer the greatest probability of success....It is believed that the most serious threat from NBC terrorism comes from chemical agents" (1992: 116). Similarly, Thornton maintains that "theft or production of a nuclear device is exceptionally difficult and biologicals are inherently unpredictable; therefore, chemical weapons present the terrorist with the best range of possible options" (1987: 6)75. Mullen, by contrast, argues that, compared to biological weapons, "the mounting of a credible clandestine mass destruction threat...would appear more difficult with potential chemical agents" (1978: 78). Similarly, in their detailed study of the subject, with reference to mass destruction, Berkowitz et al. conclude "that chemical poisons represent a relatively ineffectual threat, but that the nuclear weapon and the biological pathogens constitute threats of comparable seriousness with the latter the more practicable of the two" (1972: VIII-89). Later, they explain: "Attack with toxic chemicals offers the terrorist many options with only small resource requirements, but coupled with this is strong dependence on specific target vulnerabilities, severe problems associated with agent dissemination, and a net impact very much less than can be achieved with a nuclear weapon even in the best situations" (1972: IX-5). If the comparative advantages of chemical and biological agents are not always clearcut, however, those between chemical and biological weapons on the one hand, and nuclear weapons on the other, in regard to such aspects as ease of manufacture or other acquisition, as well as selectivity in targeting, appear obvious.
Finally, some advantages of biological agents outlined in the previous section may well apply equally to chemical agents. These may include, for example, indetectability to traditional anti-terrorist sensor systems (Root-Bernstein 1991: 50), whether for interdiction or for early warning of (and hence protection against) an attack (Kupperman and Trent 1979: 89)76; the lack of a "signature," thus allowing for the possibility of anonymous attacks (OTA 1992: 37); confining the damage to human beings or other living things, leaving other material and structures intact (Wiener 1991b: 65; Joyner 1990: 136); and, notwithstanding Mengel's attempt to distinguish between chemical and biological agents in this respect, their adaptability to demonstration attacks on small, isolated targets, while retaining the capacity of a more devastating attack (Mengel 1976: 446).
Many of the observations made above in respect to the capacity of terrorists to produce biological agents apply equally as well to chemical agents. Virtually all authors emphasize how easy it would be to obtain the relevant information from the open literature, acquire the necessary chemicals, and prepare the agent (Barnaby 1992: 85). On the first point, for example, it is often noted that both the US and Britain have declassified (and, according to some accounts, "widely published") the formula for making VX nerve gas (Clark 1980: 110; Thornton 1987: 7). According to Kupperman and Trent, "its method of preparation was first published by the British Patent Office" (1979: 65). Mullen adds that "Enough information has appeared also in the U.S. press to deduce both the formula and the preparatory routes to its manufacture" (1978: 71). Alexander quotes a 1978 report that "terrorists wanting to make deadly nerve gases can still find the formulas at the British Library despite attempts by the Government to remove them from public access" (1981: 346, quoting The Observer (London) of 19 November 1978). Douglass and Livingstone, in their characteristically sweeping manner, declare:
Ponte reports that "England and the United States declassified the formula for making VX nerve gas in 1971, and the United States read it into the widely published record of the Geneva disarmament conference." He adds:
Elsewhere, Ponte recounts an incident where "At a 1969 teach-in in England a professor scrawled the formula for making VX on a blackboard in front of hundreds of radical students, and thereafter it was widely circulated" (1977: 79).
Mullen, in referring to the availability in the open literature of information concerning chemical agents, states that "There are literally tens of thousands of professional papers, monographs, and books in this literature. A trained clandestine adversary has virtually at his fingertips, at almost any university library, all the information he would need to synthesize toxic chemical agents from raw materials or intermediates" (1978: 67).
As for the actual manufacture of agents, Hurwitz explains:
Kupperman and Trent also note that "Relatively small changes in chemical structure can produce an order-of-magnitude change in toxicity" (1979:64). Mullins begins his discussion of the subject on a cautionary note: "The development and use of chemicals would require the terrorists to have some technological sophistication. Very few of the chemical agents terrorists would be likely to use are naturally occurring....Most chemical agents would have to be produced in the laboratory." However, he goes on to disparage the level of technical expertise required: "With a basic working knowledge of chemistry, however, this would not be a difficult task for the terrorist. One of the deadliest chemicals known, VX Nerve Gas, can be produced with books from the local library, and requires no special materials or knowledge. Ball-point pen ink is only one chemical step removed from Sarin" (1992: 108-9)77. The OTA suggests that the relatively low level of expertise required should not be surprising in view of the long history of the subject, noting that "classical chemical munitions and delivery technology were used effectively in World War I, some 75 years ago, and were further developed by several nations by the time of World War II" (1991: 32). Lowell Ponte describes how "Crude World War I poison gases can be made from common commercial ingredients or even from items around the home. Want deadly chlorine gas of exactly the type that killed doughboys in the trenches of France? Just put Drano and Clorox liquid bleach in a bottle together, shake and run for your life" (1977: 79)78.
Despite the relative ease of manufacture of chemical agents, a number of authors warn that it can be a hazardous undertaking. Kupperman and Trent, for example, while noting that "a moderately competent organic chemist, with limited laboratory facilities, can synthesize sarin and VX," caution that "The operation would not be without considerable personal risk" (1979: 65). Douglass and Livingstone elaborate on this point at greatest length:
Of course, certain types of chemical agents would be more difficult to produce and use than others. Douglass and Livingstone, for example, state: "though the more sophisticated nerve agents are difficult and dangerous to manufacture, there are many varieties that are no more difficult to make than insect sprays and, subsequently, relatively easy to weaponize" (1987: 12). Indeed, Mengel suggests that insecticides themselves could be used as terrorist weapons (1976: 455). Of the two principal categories of chemical agents, it is the generally less toxic fluoroacetates that would be easiest to produce, according to Mullen:
For more toxic compounds, according to Mullen,
Berkowitz et al. generally agree with this assessment, while positing a somewhat lengthier production period:
Turning to the organophosphates, Mullen notes that, though "less common or absent from normal commercial channels," they "could be manufactured in a clandestine laboratory":
Unlike the case with biological agents, there appears to be a widespread consensus on the level of skill required for the production of a chemical agent: namely, that of a graduate student in chemistry (Clark 1980: 110; Jenkins and Rubin 1978: 223; Mullen 1978: 72; Hurwitz 1982: 38).79 In this regard, many authors refer to the need for nothing more than a "moderately competent chemist" (Kupperman and Trent 1979: 64; Barnaby 1992: 85-6), or even "any competent scientist" (Clark 1980: 110). There does, however, appear to be some difference of opinion over whether a single individual is likely to be capable of both producing a chemical weapon and employing it effectively in a terrorist attack. According to Mengel,
However, he goes on:
As in the case of biological agents, a large number of chemical substances have been identified as being of potential interest to terrorists. According to Kupperman and Kamen, "There are literally tens of thousands of highly poisonous chemicals" (1989: 101). Mullen cites an estimate of "well over 50,000" for the number of different organophosphate compounds alone (1978: 69). Those agents specifically mentioned in the literature on CB terrorism include: insecticides such as nicotine sulfate, DFP (diisopropylphosphorofluoridate), parathion, and TEPP; herbicides such as 2,4D and 2,4,5T (against plants), TCDD (dioxin), and benzidine (112-14); "blood agents" such as hydrogen cyanide and cyanogen chloride; "choking agents" such as chlorine, phosgene (carbonyl chloride), and chloropicrin; "blistering agents" such as sulfer mustard, nitrogen mustard, and lewisite; and "nerve agents" such as tabun, sarin, VX, and soman. Other chemicals mentioned include: Prussic acid (hydrocyanic acid), lysergic acid diethylamide (LSD), aminazin, pheromones, pure nicotine, phosgene oxime (CX), arsenic, Cobalt-60, compound 1080, arsine, nickel carbonyl, sodium fluoroacetate, and strychnine.
As the above list indicates, some authors have even speculated about the possible terrorist use of "psychochemical" agents or mind-altering drugs. According to Douglass and Livingstone,
Similarly, Joyner points out that "it is not necessary to kill to accomplish the main purpose intended, i.e., to inflict severe psychological distress upon a population. Certain chemical agents are available which only incapacitate victims temporarily, allowing for subsequent full recovery" (1990: 136).
Most authors (including Douglass and Livingstone), however, consider nerve agents to be the likeliest weapon of choice, given their lethality (see the section on "Toxicity" above). In their study of the subject, Berkowitz et al. focus on a "select few of the OPA [organophosphorous anticholinesterases] poisons which might interest a terrorist," namely "TEPP, because it is the most toxic of the commercially available insecticides; Sarin (GB), because its standardization as a US chemical weapon vouches for its effectiveness; and certain organophosphorous choline derivatives, because their published toxicity levels make them the most potent synthetic poisons known" (Berkowitz et al. 1972: VIII-24). Barnaby goes the furthest in narrowing the choice to a single type of agent, insisting that "Of the nerve gases, Tabun is the easiest to make and is, therefore, the most likely candidate for chemical terrorism" (1992: 85). Other factors to consider would be the ready availability of certain agents, such as insecticides sold commercially or chemical weapons stored or transported by the military.
A previous section has described in some detail the general capabilities needed for a terrorist group to be able to manufacture chemical agents on its own. There are other ways by which it might acquire chemical agents, however, including direct use of commercially-available poisons; the theft of chemical munitions held by the military; or the receipt of ready-made chemical weapons from a state sponsor. Regarding the first of these paths, insecticides, rodenticides, or other industrial or pharmaceutical chemicals such as Cobalt-60, compound 1080, TEPP, hydrogen cyanide, cyanogen chloride, carbonyl chloride (phosgene), arsine, nickel carbonyl, and parathion are widely available through commercial channels, and could be bought or stolen (Alexander 1990: 10; OTA 1992: 34; Jenkins and Rubin 1978: 224; Ketcham and McGeorge 1986: 31; Kupperman and Trent 1979: 56 and 63-4; David 1985: 146; Mullins 1992: 109; Bremer 1988: 10; McGeorge 1986: 59; Joyner 1990: 139). Douglass and Livingstone, for example, note that "Terrorist organizations can...bypass the manufacturing problem by simply purchasing suitable toxic chemicals such as parathion or phosgene that are readily available at many agricultural or industrial chemical supply stores" (1987: 12). As for CX or phosgene oxime, one of the original chemical warfare agents and featured prominently in the Soviet chemical weapons arsenal: "it is now more widely known simply as a toxic industrial chemical, and, as such, it is manufactured, stored, shipped, and sold throughout the United States like dozens of other toxic chemicals" (Douglass and Livingstone 1987: 16).
According to Kupperman and Trent: "For small, not widely destructive terrorist acts, household cleaning agents could prove lethal. Certainly, the more toxic insecticides, such as parathion or TEPP, although requiring an exterminator's license, are essentially unregulated items" (1979: 84). The latter two agents, by another account, are "almost as toxic as their military counterparts" (Kupperman and Woolsey 1988: 4). Berkowitz et al. highlight the danger of theft of such materials, noting that "Truckload quantities of Parathion are on the highways daily" and that "a hijacked truckload certainly poses a potential threat" (Berkowitz et al. 1972: VIII-32).80
Some authors note the risk of detection as a possible disincentive for terrorists to rely on commercially-acquired chemical agents. Mullen, for example, writes: "...if it is important that there are no outward indicators of an effort to employ a clandestine chemical weapon until it is time to do so, and if the terrorist wishes to inflict a higher proportion of fatalities per unit of material disseminated than is possible with some commercially available fluoroacetates, then the preparation of a fluoroacetate may be indicated" (1978: 68). Similarly, Barnaby, in noting that "It is not difficult to buy on the open market moderate quantities of the chemicals used in the preparation" of tabun, goes on: "If terrorists were nervous about buying the precursor chemicals, they could make them....These chemicals [used to make the precursors] are easier to get hold of..., and their purchase would give rise to less suspicion" (1992: 85). McGeorge confirms that "Sarin or other agents can be manufactured from relatively innocuous substances such as isopropyl alcohol and phosphorous trichloride, thereby helping to preserve secrecy" (1986: 59).
A number of authors have also expressed concern about the possible theft of chemical weapons from military installations or disposal sites in the US (a threat which is presumably even greater in the states of the former Soviet Union). According to Livingstone, "the U.S. government [has] acknowledged that a small amount of its inventory of VX is presently unaccounted for" (1982: 111). Clark charges that "There have been known instances of its being rather casually offered for sale in New York City," and goes on:
Mullins agrees with this assessment:
Also on this theme, Marshall goes so far as to claim that "Chemical weapons in particular are relatively easy to purchase on the black market, particularly since they were so widely deployed during the Iran-Iraq War in the 1980s" (1990: 372). Joyner also emphasizes the danger that
Finally, the literature is replete with references to the possibility of rogue states supplying chemical weapons to terrorist groups. Those potential culprits mentioned most often are Libya, Iraq, Iran, Russia (or the former Soviet Union), Syria, North Korea, and Cuba (Alexander 1990: 10; OTA 1991: 52 and 1992: 34; Ketcham and McGeorge 1986: 31; Jackson 1992: 520; Kupperman and Kamen 1989: 99-100; Revell 1988: 16; Mullins 1992: 109; APN 1988: 16; Joyner 1990: 138-9). According to Jackson: "There is now considerable evidence of Soviet-derived chemical arms having been deployed in several regional conflicts throughout the Third World, ranging from hybrid chemical/explosive toxic 'firebombs' with a phosphine base to weapons with traces of organic cyanide and strontium" (1992: 520)82. McGeorge reports that "Iraq allegedly turned over control of Soviet supplied agents to known PLO members" (1986: 60), although Douglass and Livingstone maintain that the exchange occurred in the opposite direction, with Moscow making use of the PLO as an intermediary to transfer chemical and biological agents to Iraq (1984: 18).
The US Congressional Office of Technology Assessment notes the ability of Libya, Iraq, and Iran to produce chemical weapons together with the fact that "all of these countries have sponsored active terrorist groups that have attacked civilian populations with the aim of producing many deaths" (1991: 52). Joyner, clearly concerned about the danger of a rogue state providing a terrorist group with chemical weapons, observes that "As yet, no state is known to have done so, though this does not mean such reluctance necessarily will be perpetual" (1990: 138). Kupperman and Kamen express perhaps the strongest view of this potential linkage, declaring that "Chemical attacks by terrorists will almost certainly be driven by the proliferation of chemical arsenals among their state sponsors" (1989: 99).
As in the case of biological agents, most authors consider the effective delivery of chemical agents to their target as being more difficult than their manufacture (Jenkins and Rubin 1978: 226; Kupperman and Trent 1979: 64; Kellett 1988: 56; Loehmer 1993: 62; Mengel 1976: 445-6; Mullen 1978: 76-7; Berkowitz et al. 1972: I-12).83 Mengel explains:
Also as in the case of biological agents, the popular scenario of contamination of a large water supply is unlikely to be a feasible method of terrorist attack with chemical agents. Jenkins and Rubin point out that "Organophosphorous compounds....do hydrolyze in water,...making them unsuitable for most scenarios involving the contamination of water supplies" (1978: 224).84 Even Clark admits that, in connection with a 1972 threat to poison New York City's water supply, an "Army expert" had "advis[ed] that it would take tons of nerve gas to poison the 31-billion-gallon reservoir," although Clark goes on to insist that "Still, there are many chemicals that can, quite easily, make an area's water system lethal" (1980: 113). Mengel, by contrast, calculates that "based on personal consumption as opposed to other uses and a four billion gallon reservoir, if each member of a community of 20,000 were to drink 16 ounces of water, it would require in excess of 14 billion lethal doses to deliver one dose per person. If the best suited chemical, fluoroacetates, were used, it would require 600 metric tons" (1976: 455)85. Similarly, Hurwitz argues that "Introducing an agent into a municipal water supply would not be a credible threat because of the huge volume of water that would need to be contaminated and the numerous steps in the filtration and purification process" (1982: 39). After discussing the dilution problem, Mullen adds:
Berkowitz et al. posit four likely methods of dissemination of chemical agents by terrorists: "(1) covert contamination with bulk agent of foodstuffs or beverages selected to avoid conditions which would destroy the poison; (2) covert generation in enclosed spaces of lethal vapor concentrations from volatile agents; (3) covert dissemination in enclosed spaces of aerosols of non-volatile agents; and (4) overt attack with bursting munitions or thermogenerators" (1972: IX-5). As an example of the first of these, they note that
In a somewhat eerie foreshadowing of the 1995 Tokyo gas attack, Berkowitz et al. go on to note that "For vapor dissemination, of the agents investigated only Sarin is sufficiently volatile," while "the involatility of the V-agents and BTX require that they be disseminated as aerosols." Finally, they point out: "All the agents except BTX could be effectively incorporated into either bursting munitions or thermogenerators. It is doubtful that an unwarned and untrained target group would comprehend the nature of the threat to which it is exposed; its first reaction would likely be to interpret the explosion as a conventional bomb and attempt to render aid to the nearby victims" (1972: IX-6).
Most authors agree that the most feasible "mass" chemical attack would be one limited to the enclosed spaces of a single, discrete facility such as a hotel, office building, or convention center (Jenkins and Rubin 1978: 224)86, with a resulting casualty toll ranging between a few hundred and several thousand. At the lower end of the scale, Mullen argues:
Hurwitz, by contrast, puts the likely number of casualties as the result of a chemical attack on a "large auditorium" at "several thousand" (1982: 36).
Livingstone posits a number of likely scenarios against government facilities. For example: "...a truck loaded with drums or canisters containing a nerve agent like VX or Sarin could be crashed into an embassy and exploded, turning the deadly substance into a fine mist which would envelop the entire facility" (1986: 143). Or, if targeted against a military base: "mortar bombs, if filled with a V-series nerve agent, would force the evacuation of the entire area and probably inflict a large number of casualties. If the target were an airbase, it would, in all likelihood, be shut down for a matter of days" (1986: 144). The vulnerability of even the highest-value, discrete targets has been demonstrated by various US Army "mock attacks" on government buildings in Washington, DC, as recounted by Lowell Ponte:
Elsewhere, Ponte cites an earlier example of the same type:
Other possible means of delivering chemical agents to their targets, though on a smaller scale, would be through the contamination of foodstuffs or by direct contact (as in the case of the ricin-tipped umbrellas discussed in the previous section). Livingstone, for example, suggests that "it would...be possible to inject...a chemical poison into a victim by means of a hypodermic needle concealed in the tip of an umbrella" (1982: 111). Mullins adds that "Chemical agents could be used effectively as contaminants for projectiles such as bullets, flechettes, and shrapnel" (1992: 111).
A considerable number of threats or incidents involving the terrorist use of chemical agents have been reported in the open literature88. As in the previous section concerning biological agents, these may be ranked in terms of seriousness or severity (in ascending order) as follows: (1) threats to use CW, without any evidence of actual capabilities; (2) unsuccessful attempts to acquire CW; (3) actual possession of CW agents; (4) attempted, unsuccessful use of such agents; and (5) their actual, "successful" use. In the first category, the following cases have been reported:
The following reported cases fall into the second category, of unsuccessful attempts by terrorists to acquire chemical agents:
Into the third category, reports of the actual possession of chemical agents by terrorists, fall the following:
A number of cases are reported of apparently unsuccessful attempts by terrorists to actually use chemical agents:
The successful use of chemical agents by terrorists (but without inflicting "mass destruction") has been reported in the following cases:
Alexander adds that, as a result of the incident, "Israel had to cut back its orange exports by 40%" (1990: 10);
There have, of course, been many other reported instances of product contamination, perhaps the most notorious being the 1982 Chicago case of cyanide being placed in capsules of the pain remedy Tylenol, which resulted in seven deaths (Kellett 1988: 57). However, the vast bulk of these acts have apparently been committed with no political motivation in mind, and hence should not be classified as "terrorist" in nature. Furthermore, even where a political motive has been present, as Jenkins reminds us: "In none of these latter cases was it the intent of groups to cause death. Their weapon was the alarm that would be caused and the consequent loss of revenue. This is true in most cases..." (1989: 2).
In sum, there is sufficient evidence in the public domain to indicate that terrorist groups have indeed displayed an interest in acquiring chemical agents; have made threats to use such agents; have in some instances actually succeeded in acquiring such agents; have at times attempted to make use of them; and in some cases actually "succeeded" in such attempts, though without inflicting mass casualties in the process.
Only a few authors speculate about the reasons why terrorists have so far not made greater use of chemical agents in particular. Kellett notes that "The desire for political legitimacy acts as a considerable constraint on" the use of both biological and chemical weapons, but "particularly on the employment of chemical weapons, whose application has been widely condemned by public opinion and proscribed by treaty." Why this factor should carry greater weight in the case of chemical than biological weapons (which are also proscribed by treaty and, if anything, considered even more abhorrent by the general public) is unclear. In any case, Kellett goes on: "Terrorists have retaliated against corporations for chemical spills and industrial accidents, adding to the restraints they may feel on using such weapons" (Kellett 1988: 56). The latter consideration, it may be surmised, would apply at best only to so-called "ecological" terrorists, and even then would probably not affect their propensity to use chemical agents in "low-level" incidents directed at individuals or small numbers of people that would not result in widespread environmental impacts.
A more convincing explanation for the relative non-use of chemical weapons by terrorists is that provided by Jenkins, which applies at least equally well to BW: "With an explosion, you get a bang and some blood and you can calculate it pretty much. In the case of chemical weapons, there's a lot of uncertainties. Terrorists tend to abhor uncertainty" (quoted in Marshall 1990: 372-3). Other reasons offered for the comparative non-use of chemical weapons match those discussed earlier in respect to BW: the lack of any desire to kill large numbers of people; the fear of alienating the general public, or provoking ruthless suppression by governments; the preference for "sharp, dramatic impactsto exploit an event's immediate shock value" (versus the prolonged suffering anticipated as the result of a chemical attack, as well as the lack of a "stark explosion" or "bloody evidence"); and the desire for "premeditated control...over an event," made difficult by the unpredictability of chemical weapons (Joyner 1990: 137). Finally, the US House Armed Services Committee notes that (in contrast to BW) "Chemical agents must usually be employed in relatively large quantities to be effective" (1993: 26).
Those authors who have speculated about the future terrorist use of chemical agents in particular have generally rated its likelihood as quite high. Barnaby, for example, declares that "If terrorists manufacture weapons of mass destruction in the near future, they are likely to opt for chemical rather than biological or nuclear weapons" (1992: 85). Similarly, Jackson refers to the "real and growing threat" that terrorist organizations will employ chemical weapons (1992: 520). And according to Joyner: "chemical terrorism remains just over the horizon as a distinct possibility" (1990: 135). Thornton similarly argues that "The use of chemical agents by terrorists is a definite possibility," going so far as to add that "from the terrorist standpoint, it is a virtual necessity" (1987: 2). It is Mullins, however, who expresses himself most definitively on the issue of future use, declaring simply: "There is a high probability that terrorists will rely on chemical agents in the near future to achieve their goals" (1992: 116).
Bremer contrasts the possible use of chemical weapons with the relatively minimal threat of nuclear terrorism, stating that "Chemical substances on the other hand have been used for malevolent purposes by a variety of groups and individuals and must be considered as presenting a somewhat more likely terrorist choice." While denying that any such, politically-motivated cases have yet occurred in the US, he judges that "the possibility of terrorists ultimately using this tactic [of product contamination] against us or other nations is sobering" (1988:8). Finally, he notes that "The world community has shown little outrage at the recent use of chemical weapons by both Iran and Iraq in their war," and speculates: "Perhaps a psychological barrier has already been broken for terrorists to use them" (1988: 12). Jenkins agrees, suggesting that
Other trends suggesting a greater likelihood of the use of CW by terrorists mirror those discussed in the earlier section on BW. They include: a growing number of less-discriminate, high-casualty attacks beginning in the mid 1980s; the "severe brutalization" of some terrorists in the course of their "long struggle against society or the state"; the growing desensitization of the public to more traditional methods of attack, requiring "raising the level of violence...to regain the public's attention"; greater technical proficiency on the part of terrorists, as demonstrated, for example, in the use of sophisticated timing mechanisms to bring down airliners; and the growing state sponsorship of terrorists (Joyner 1990: 138; Thornton 1987: 8).
Of the authors focusing on chemical terrorism in particular, Joyner devotes the greatest amount of attention to "candidate groups." In his words, those that "stand out for the stark viciousness of their terror-violence...in general would appear the most likely candidates for resorting to chemoterrorism." In this regard, he suggests that "The most dangerous region, and the one offering the greatest opportunity for chemoterrorism, is the Middle East." In particular, he singles out the Abu Nidal Organization ("the most wide-ranging and dangerous among the Palestinian terrorist groups"), noting that "it is not inconceivable that Abu Nidal could acquire chemical weapons from Muammar Qadhafi once Libya attains production capability" (1990: 139). The second Middle Eastern group identified by Joyner as possibly "find[ing] chemoterrorism tempting" is the Popular Front for the Liberation of PalestineGeneral Command (PFLP-GC). Referring to suspicions of its being behind the December 1988 Pan Am 103 bombing, he writes:
According to Joyner, "In Western Europe, no single terrorist group presently appears inclined to move to the chemoterrorism threshold." However, he goes on to identify as groups that "ultimately might resort to using chemical weapons because of their extreme viciousness and radical ideology," both Direct Action in France and the Red Army Faction in Germany. On the latter, he notes:
Finally, Joyner considers that no current terrorist groups in Asia or Latin America "seem viable candidates for chemoterrorism" (1990: 140).
Another author who speculates about which terrorist groups would be most likely to employ chemical weapons is Alexander, who writes: "the Hezbollah (Islamic Jihad or the Party of God), operating with the support of Iran, might employ chemical terrorism against Western interests in the Middle East or against other adversaries such as Iraq or Saudi Arabia" (1990: 10). Similarly, Thornton notes that "Iran has been providing major support to Lebanese Shiite Muslims; chemical weapons are a distinct possibility as a part of its anti-US and anti-West crusade" (1987: 7).
As in the case of biological agents, most authors are quite pessimistic about the feasibility of defences against terrorist use of chemical weapons. In the words of Kupperman and Trent:
Those few authors who do speculate on possible defences against chemical terrorism in particular focus on preventing terrorist access to the most likely chemical agents, and developing better early-warning detection methods. For example, Jenkins and Rubin speculate that "It may be possible...to identify specific chemical compounds that ought to be subject to licensing procedures with penalties for unauthorized possession" (1978: 228). The OTA notes that "In the chemical area, rapid 'early warning' multiagent detectors are being developed" (1992: 5). Even Kupperman and Trent acknowledge that "Generally speaking, chemical detectors could be used to interdict selected chemicals were close-in inspections feasible. Obviously, if chemical detectors were distributed widely, they could give warning during the first minutes of such an attack" (1979: 85). However, they go on:
Without getting into specifics, Thornton emphasizes the need for contingency planning against chemical terrorism, including "studies on damage limitation," arguing that "The heated period following such an attack is neither conducive to diplomatic and military constraint, nor to the sound management of the situation itself" (1987: 9).
Finally, in the realm of political measures to help contain the threat of chemical terrorism, Joyner is a strong proponent of international arms control agreements, in particular the (then-emerging) Chemical Weapons Convention (CWC):