ichneumon
4th April 2006, 14:03
!nfect!ousness of terror!st !deology
Darw!n!an National Security: The !nfectiousness of Terrorist !deology
(changed so as not to appear on google)
“The challenge of terrorism is akin to fighting a virus in that we can accomplish a great deal but not eradicate the problem. We can take steps to prevent it, protect ourselves from it, and when an attack occurs, quarantine it, minimize the damage it inflicts, and attack it with all our power.” - Richard N. Haass, former Director of Policy Planning, U.S. State Department
Introduction
Terrorism in the 21st century is unconventional, unpredictable, and potentially unavoidable. In part, this is because contemporary terrorists are increasingly transnational, industrious, unorthodox in their methods, and decentralized (e.g., Ariza 2005, Ehrlich and Levin 2005, Atran in press). In mounting an effective counterterrorism campaign against such cryptic enemies, experts have begun to consider equally unconventional defenses, many of which are discussed throughout this book. Some view terrorism through the lens of epidemiology, where terrorist ideology is analogous to an infectious agent of threat to global public health (in particular, Stares and Yacoubian 2005). This is not the first time that scientists have looked to epidemiology to enhance their knowledge of how memes and ideologies spread, but it may be the most paramount. Similarities between terrorism and pathogen outbreaks have resulted in parallel, but independent, tracking, prevention, and control efforts. The National Strategy for Homeland Security, for instance, “… recognizes that the capabilities and laws we rely upon to defend America against terrorism are closely linked to those which we rely upon to deal with non-terrorist phenomena such as disease...” (Office of Homeland Security 2002: 4).
1
Infectiousness of terrorist ideology
If sufficient parallels exist between infectious agents and terrorist ideology, counterterrorist efforts may be able to draw on the already substantial body of theory developed for public health (e.g., Box 1). In this chapter, we disentangle these parallels in hopes of laying the foundation for future research on the infectiousness of terrorism. We specifically explore the utility of epidemiological theory in increasing our understanding of the dynamics and spread of terrorist ideology. We use the terms infectious agent, parasite, and pathogen interchangeably throughout this chapter. We use our findings to identify key unanswered questions and avenues for future research. In considering this analogy, we do not promote any particular counterterrorist strategy or recommend policy. Terrorist ideology is a type of meme or suite of memes. Memetics, the contagiousness of thought, and the dynamics that govern the spread of ideas have captivated a broad audience for many years (e.g., Dawkins 1976, Lynch 1999, Gladwell 2000, Distin 2005). Dawkins (1976) defines memes as any cultural entity (e.g., a fashion fad, song, idea, religion, language) that is replicated through exposure to humans and which has evolved as an efficient (though not necessarily perfect) copier of information and behavior. Much of the inquiry into memetics has centered on why some memes spread in an “epidemic” fashion. Like genes, memes spread if they provide a clear benefit to an individual, for example, through group identity, enhanced sexual attractiveness, or increased resource acquisition (Sober and Wilson 1998). However, memes that provide less tangible benefits, such as rewards after death, can also spread. Moreover, it has been proposed that memes with no apparent cost or benefit may arise and become ‘fixed’ in a population simply because they possess characteristics that render them more likely to be adopted than another meme (e.g., incorrectly transmitted verbal phrases that contain more audible syllables than their grammatically correct counterparts; Dawkins 1989). Even memes with a seemingly negative impact may spread through their provision of a net indirect benefit. For example, individuals with a “handicap” may be attractive to others due to the perception that the handicapped individual tolerates adverse conditions (Zahavi 1975). One thing that sets terrorist ideology apart from other memes (such as linguistic or religious) and aligns it with infectious agents is an explicit desire by society to control its spread. The infectiousness of terrorism: constructing the analogy Infectious agents depend on one or multiple hosts to support and complete their lifecycle (development, maturation, and reproduction). Host-pathogen dynamics can be viewed at the level of a host individual, host population, or community of species. Analogous to this, a given ideology can be “hosted” by a terrorist, terrorist cell, or terrorist organization. It is within these individuals and groups that ideology is conceived, formed, developed, and honed. Infectious agents need hosts just as ideology cannot exist without the minds that harbor it (though pathogens and ideologies can persist in alternate forms outside of the mind; many infectious agents have free-living resting stages and ideologies can be preserved outside the mind by various media; see below). Neither an infectious agent nor its host can exist in a system where environmental conditions are unsuitable. Indeed, abiotic (i.e., climatic) and biotic (i.e., competition, predation) factors help to shape the boundaries of an infectious organism’s geographic range. An analog to this is the political, social, and economic environment that shapes the region where terrorist ideology thrives. Changes to the environment might foster the evolution of ideology away from terrorism. For instance, the transition of a terrorist group to a political power may lead to ideological or methodological changes better suited for formal governing (Prusher 2006). Finally, it is the transmission of infectious agents and terrorist ideology between hosts/terrorists that maintains the entity’s existence within the system. Infectious agents can be transmitted between susceptible hosts directly, as in the case of sexually transmitted pathogens which require person-to-person contact. Transmission my also occur indirectly, by way of a vector that harbors the infectious agent (e.g., mosquitoes carrying malaria or ticks carrying Lyme disease) or through a contaminated vehicle such as food (e.g., salmonella), or liquid (e.g., giardiasis). The spread of terrorist ideology among individuals occurs via similar routes. Ideology may be transmitted directly, through the oral exchange of ideas from a terrorist to an individual who does not yet harbor the ideology. It may also spread indirectly, via a vehicle/vector such as broadcast media, print media, or the Internet. Ecological and epidemiological weaponry in the war on terror Terrorism as a plague teaches us that plagues are rarely ever wiped out. If we are lucky al-Qaida will be our smallpox, eradicated after a sustained global campaign. More likely, however, it is malaria, a deadly disease which keeps on killing in part because it keeps adapting and changing. So will other diseases and so will other terrorists. -Dr. Erik Steele, Vice President for Patient Care Services, Eastern Maine Medical Center Ecology and epidemiology have a highly developed set of theoretical tools and mathematical modeling approaches for understanding the basic properties of infectious agents. In many cases, such techniques have greatly informed control practices (Smith et al. 2005). Here we compile a theoretical framework from ecology and epidemiology that should be of use in the war on terror.
4
Infectiousness of terrorist ideology
Assembling the theoretical arsenal Basic models of pathogen spread divide individuals into population categories that reflect infection status (e.g., susceptible, infected, recovered, or SIR at the most basic level). SIR models apply specifically to a group of infectious agents called microparasites, which include many viruses, bacteria and protozoa. The rate of change in the abundance of individuals in each category is described by a differential equation constructed from variables such as contact rate, susceptibility, birth rate, mortality, cure, immunization, and the abundance of individuals in the various population categories. The frequency of a particular category can increase, decrease, or stabilize over time. A key to such models is that the differential equations are coupled. For simple formulations, it is possible to solve these equations analytically and gain general insight. In most cases, however, finding analytical solutions for coupled differential equations with more than two categories is difficult. In these cases, epidemiologists have turned to a very effective short-cut called the basic reproductive ratio, or R0. R0 is the estimated number of secondary cases directly arising from one primary case (Anderson and May 1979). In deterministic models, if R0 is < 1, the epidemic will fizzle, while for R0 > 1, an epidemic will occur. Once an epidemic begins, it grows at a rate known as the effective reproductive ratio, R, which tends to decline from R0. R can stabilize around 1 (become endemic in the population) or it can tend toward 0, at which point the pathogen is extirpated from the population (cycling, chaos and other complex outcomes are also possible). It might appear that R0 would not be very informative once an epidemic is already underway because, during an epidemic, we would be most interested in reducing R. However, because R changes with the
5
Infectiousness of terrorist ideology
state of the epidemic, it is more difficult to calculate than R0, which is a set value and applies to a limited set of conditions. Determining the structure of the R0 equation helps identify the variables (and their interactions) that are most likely to determine if infection will spread through an uninfected population. Although an SIR model tracks three classes of individuals over time, only the rate of change of the infected individuals is needed to calculate R0. A simple component of an SIR model which tracks the per-capita rate of change of infected individuals (I) in an SIR population can be written as: dI/Idt = Sβ – (μ + α + r) where S is the number of susceptible individuals, β is the transmission coefficient (how contagious the infectious agent is to susceptibles), μ is the background mortality rate, α is the additional mortality rate suffered by infected individuals, and r is the recovery rate. This model assumes that contacts occur at random and are density dependent (e.g., an individual living in a large city will come in contact with more people per day than an individual living in a rural community). R0 is simply a ratio comprised of the expected rate of transmission through contact with susceptible individuals divided by the sum of the death and recovery rates. The basic reproductive ratio is, therefore: R0 = Sβ/(μ + α + r) 6
Infectiousness of terrorist ideology
By analogy, individuals that are exposed to and “infected” with terrorist ideology become terrorists. In addition, there are non-terrorists in the population who are either susceptible or resistant to the ideology. The number of resistant individuals is not explicitly a component of R0, except that a greater proportion of resistants implies a lower number of susceptibles. Resistant individuals may be susceptibles that become resistant before contact or they may be terrorists that have disavowed their ideology. Understanding the components of R0 for the spread of terrorist ideology may provide greater insight into the types of counterterror strategies that could be used to reduce R0 below 1, and thus diminish the likelihood that terrorism will spread. Efforts to reduce R0 below 1 may concentrate on increasing the death and/or recovery rate of infecteds and/or decreasing the abundance (or susceptibility) of susceptibles, or their contact rate with infecteds. Reducing R0, however, is not the only means of controlling the spread of an infectious agent or terrorist ideology. Outbreaks of infectious agents are oftentimes apparent in the early stages, when very few individuals are infected. This is a direct result of the incentive that an infected individual receives for reporting their condition and thus obtaining treatment. This increases the likelihood that public health officials will learn of enough independent cases to implement control initiatives before the rate of spread increases. Not surprisingly, this scenario is in stark contrast to the spread of terrorist ideology because new converts cannot risk revealing their ideology to authorities. Relative to pathogen outbreaks, such behavior makes it exceedingly difficult to forecast and control emerging terrorist ideological ‘outbreaks,’ though this is not to suggest a complete lack of indicators. Symptoms of such an emerging ideological outbreak may include an increase in the frequency of attacks, the death or capture of militants, or the dissemination of
7
Infectiousness of terrorist ideology
media (Stares and Yacoubian 2005). It is important to recognize, however, that the inception of infectiousness is not necessarily correlated with the appearance of symptoms. This is a critical notion as the relationship between the timing of infectiousness and the onset of symptoms is among the key general properties of emerging infectious agents that public health officials look to when planning control initiatives. A similar snapshot of emerging terrorist ideologies should prove valuable to counterterrorism . Although vaccination is available for many of the pathogens that plague mankind, the evolutionary novelty and unpredictability of emerging infectious agents and outbreaks dampens its viability as a broad scale control strategy. In the absence of vaccines and treatment for pathogens on the verge of an outbreak, public health officials generally favor two basic control practices: 1) isolation of symptomatic individuals and 2) tracing and quarantining their contacts (Fraser et al. 2004). Both the implementation and success of these measures rely on a number of factors, ranging from the epidemiological characteristics of the infectious agent to the communication infrastructure of the public health agencies charged with control. Mathematical modeling of contemporary outbreaks (in particular SARS, HIV, smallpox and influenza) suggests that the success of these control measures is equally dependent on the proportion of transmission that occurs prior to the onset of clinical symptoms (θ) and the inherent transmissibility of the infectious agent (R0, as described in the previous section of this chapter) (see Fraser et al. 2004 for model details). Model simulations suggest that isolation, contact tracing and quarantine are indeed sufficient to control outbreaks of infections when the values of θ and R0 fall below a critical threshold: Control through isolation alone is possible when θ < 1/ R0, but when θ > 1/ R0 contact tracing is also required. When θ is very high, neither control
8
Infectiousness of terrorist ideology
measure (alone or in combination) is able to prevent the outbreak from progressing. The simplicity and efficacy of these control measures and the modeling techniques used to describe them should be attractive features from the standpoint of counterterrorism initiatives. Is intensity important?Adopting an SIR approach to terrorism assumes that all terrorists have the same intensity and capacity for spreading the ideology. However, if ideology is a continuous variable, then some individuals within a population may possess none of the ideology, some may posses a little, and a few may subscribe with high intensity. Individuals that subscribe most to the ideology may be more likely to spread the ideology than those that are merely sympathetic or supportive of the ideology. Is variation in the intensity of terrorist ideology among individuals sufficiently important to justify greater model complexity? A possible alternative is an SIR model where contact rate varies among individuals, allowing some individuals to be more active than others in spreading ideology. This has been of particular interest in understanding the spread of HIV because the number of sexual partners varies greatly among individuals, with the majority of transmission stemming from a small number of highly sexually active people (Johnson et al. 1989). The derivation of models applied to this scenario is beyond the scope of the paper, but the important result is that variation in contact rate among individuals can greatly increase R0 (May and Anderson 1988). This suggests that targeting individuals who are disproportionately active will reduce the spread of terrorist ideology, a conclusion that has intuitive appeal. However, these models describe differences in contact rate
9
Infectiousness of terrorist ideology
that are distinctive to an individual whether or not they are infected (i.e., the link between intensity and the propensity to spread an infection is indirect). Models developed for parasitic worms (macroparasites) might better mimic the spread of terrorist ideology than SIR models. Hosts exposed to many worms typically suffer higher pathology than hosts exposed to few worms. Hosts with many worms also contribute more to the growth of the total worm population. In addition, few infected hosts have the average level of infection: most have light, non-pathogenic infections and the majority of the worms are in a few highly infected individuals (such distribution of parasites in a host population is referred to as aggregation). For this reason, the SIR categorization of hosts as simply infected and uninfected fails to capture important aspects of the population dynamics of parasitic worms. Macroparasite models were developed to account for this level of biological detail (Crofton 1971, Anderson and May 1979, May and Anderson 1979). These models have provided considerable insight into how to target control efforts toward the few, most heavily infected individuals. In evaluating the utility of macroparasite models for terrorism, it is important to describe how macroparasite models only indirectly account for the intensity of infection. Such models do not distinguish infected from uninfected hosts. Instead of tracking infected hosts, they track the total worm population. The differential equation representing the growth of the worm population has a number of new terms. In a very simple formulation, the per-capita rate of change of the worm population is: dW/Wdt = ISβ – (μ + α + d + W(k+1)/Sk) 10
Infectiousness of terrorist ideology
where the variables are the same as above except that W is the total number of worms in the worm population, I is the total number of infectious stages, β represents contact between infective stages and hosts, d is the death rate of worms within the host and k is an inverse measure of the degree of aggregation of worms in the host population. Solving for the basic reproductive ratio of a macroparasite is much more difficult than for a microparasite, and alternative formulations for R0 in macroparasites can be considerably different, depending on the biological details of the model (Roberts et al. 2002). Because macroparasite models are affected by the variation in intensity among infected individuals, the currency in such terrorism models would be the total amount of terrorist ideology in the population. Possible metrics for intensity include attack rates and internet chatter. Attack rate is particularly tempting to consider as a currency because reducing attack rate is an ultimate goal for counterterrorist activities. While variation in ideological intensity suggests that counterterrorism should target terrorist leaders, the network structure of many terrorist organizations contrasts traditional leadership hierarchies and is increasingly “leaderless” (Sageman 2004; Atran in press), thus traditional strategies aimed at organizational “decapitation” are less likely to be effective than predicted by simple mixing models. Extending key epidemiological variables to the spread of terrorist ideology It is not necessarily clear which modeling approach provides a better analogy for the spread of terrorist ideology. However, we can gain further insight by deconstructing, and examining in
11
Infectiousness of terrorist ideology
detail, the major contributing variables in epidemiological models: susceptibility, contact rate, recovery, and mortality. Susceptibility Without susceptibles, an infectious agent cannot spread. Although some infectious agents are able to evolve adaptations to get past host defenses, they are less likely to adapt to two host species that differ greatly in physiology and/or evolutionary history (Combes 2001). As a result, not all host species are susceptible to all pathogens, and host specificity, or the restriction of the number of host species that a pathogen can infect, is widespread. Ideological specificity might arise for the same reasons as host specificity. The specific cultural identity of an individual (religious, national, historical) may predispose it to some, but not other ideologies. If an ideology is recognized as clearly foreign, an individual may more likely reject it as “non self.” Xenophobia is a common aspect of human cultures that probably acts to maintain cultural identity and prevent the spread of foreign ideologies (Boyd and Richerson 2005). For this reason, an ideology that is successful in spreading through a particular population may do so because it contains attributes compatible with that populations’ culture. In contrast, a population may be less susceptible to a terrorist ideology if that ideology differs greatly from the culture’s modal ideology. This suggests that the cross-cultural spread of terrorist ideology may be unusual, and it is probably not reasonable to assume that all cultures are equally susceptible to a particular ideology - just as all species are not susceptible to a particular infectious agent. 12
Infectiousness of terrorist ideology
A second pattern in susceptibility to infectious agents is that, for a single species (or closely related species), populations that have had an evolutionary history often evolve adaptations against an infectious agent while naïve populations can be more susceptible and less tolerant. The evolutionary pressures that result in the evolution of resistance derive from the negative fitness consequences of an infectious agent on its host. This suggests that nations with a history of battling terrorism might be more likely to resist the future spread of terrorist ideology within their population. A tool of modern public health is to reduce the pool of susceptibles through vaccination. Vaccination works by exposing uninfected individuals to dead or attenuated infectious agents, after which the immune system goes through the process of building specific antibodies that then lead to immunity. Even vaccinating only a fraction of susceptibles can reduce R0 below one (herd immunity). Vaccination is most viable as a broad-scale control strategy when used against well-known pathogens. An analogy to vaccination for terrorist ideology is exposing a population to a discredited version of the ideology to make it less likely that newly exposed individuals will find it attractive (Stares and Yacoubian 2005). An in-depth knowledge of a particular terrorist ideology would seem necessary for this approach to succeed. The environment can also alter a host’s susceptibility to an infectious agent. Thermal stress, for example, appears to increase the susceptibility of some marine organisms to different infectious agents (Lafferty et al. 2004). Other physiological stressors, such as lack of food, may force a host to redistribute resources away from defense against infectious agents (Rigby and Moret 2000). In some cases, public health officials can alter the environment, such as increasing nutrition, to
13
Infectiousness of terrorist ideology
reduce the susceptibility of individuals (Solomons and Scott 1994). While stress at the individual level should tend to increase susceptibility to infectious agents, it may, unexpectedly, reduce the spread of a pathogen through a population. This occurs for the following reasons: stress may reduce host abundance (and, therefore, contact rates), and stress can increase pathogen mortality rates either through killing pathogens directly or by leading to differentially high mortality of infected hosts (Lafferty and Holt 2003). The environment could similarly affect the number of individuals susceptible to terrorist ideology. In environments that are unfavorable (poor in justice, resources or freedom), new terrorist ideologies might hold appeal, particularly if they offer the promise of change and current or alternative ideologies do not appear effective. Therefore, efforts that reduce the desire for change or provide alternative pathways for change might decrease susceptibility. Still, environmental conditions that decrease the susceptibility of individuals to a particular ideology will not necessarily prevent the spread of that ideology. For example, improved economic conditions might decrease desire for change in individuals, but also increase opportunities for communication or implementation of a terrorist ideology, making it difficult to predict the net effect. Strategic counterterrorism seeks to reduce the political and operational space where terrorism can develop, spread and sustain (Gunaratna 2005). The targets of strategic counterterrorism include the foundations on which a culture’s education, religion, media, legislation and ideology are set. The mission is to counter terrorism through these institutions by promoting an ethic against terrorist ideology that extends to the greater community. Because strategic counterterrorism is in
14
Infectiousness of terrorist ideology
many ways preemptive, ‘susceptible’ individuals are likely to become infected with an anti-terrorism agenda before they are ever exposed to a terrorist ideology. Like vaccination campaigns, successful strategic counterterrorism would require early implementation, substantial funding and coordination. Contact rate Pathogens and ideologies spread primarily through contact. SIR models assume that in dense populations, individuals are more frequently in physical contact with one another. Another aspect of contact is the extent of movement among populations. This is difficult to capture in simple epidemiological models, but the logic is straightforward: contact between individuals of different populations is likely to lead to a wider spatial spread of an infectious agent. This is why increased contact through diffuse networks related to increases in transportation and modern trade has led to increasing concern for the spread of pandemics (Hufnagel et al. 2004). Current concern about avian influenza is one such example, and rightly so as the 1918 avian flu pandemic may have been associated with large-scale troop movements during World War I (Barry 2004). Efforts to reduce contact can be effective at slowing the spread of infectious agents. Two general approaches are used. The first is to reduce contact rates between individuals irrespective of their infection status. Safe-sex and personal hygiene campaigns are probably the best examples in modern society. For instance, children are taught to wash their hands, use toilets, and avoid spitting in public. The second approach is to specifically reduce contact between infected and uninfected individuals. In public schools, children with head lice are typically sent home to
15
Infectiousness of terrorist ideology
reduce the spread of lice to their classmates. For more serious infectious diseases, public health officials may quarantine infected individuals (and sometimes those that the infected individuals have contacted) in an effort to stop spread (Fraser et al. 2004). Such an approach was used to reduce the spread of SARS. Limits on the movement of infected individuals can be difficult because detection is challenging. In the SARS epidemic, body temperature scans in airports (e.g., Hong Kong) were used to identify individuals with fevers, and these individuals were then subjected to medical examination. Recent epidemiological models are making progress in predicting the spread of pathogens like SARS with evaluation of different control strategies (Hufnagel et al. 2004). How similar is the spread of an ideology to the spread of an infectious agent? Whereas sick people do not try to contact and infect other individuals, it is human nature to share ideas and convince others to agree with our opinions. This is not to suggest that pathogen transmission is passive. Infectious agents that are most likely to persist are those with traits enabling them to spread from one individual to another. Symptoms such as coughing, sneezing, and diarrhea are examples of behaviors induced by infectious agents to facilitate spread (Ewald 1993). It is also likely that ideologies are under analogous forms of selection for characteristics that favor spread (Hochberg, this volume). Successful religions often have doctrines favoring spread, such as active conversion of others, increased reproduction, early indoctrination, and retention. We may expect successful terrorist ideologies to possess adaptive traits for spread or to be aligned with existing religious ideologies. For example, terrorist ideologues have been successful at recruiting members through religious schools (Gunaratna 2005). 16
Infectiousness of terrorist ideology
There are different opportunities for the spread of an ideology than for the spread of an infectious disease. .Electronic communication greatly facilitates the spread of information at speeds and scales far exceeding physical contact. Information contact is increasing due to fewer language barriers, landline and cellular phones, television, radio and the Internet. This decouples the spread of an ideology from local population density (Ariza 2005) while simultaneously favoring decentralization and spatial spread (Atran in press). Cells, in particular, tend to communicate and exchange ideology and information primarily through the Internet (Atran in press). This is not to downplay the importance of direct communication. A study of enlistees into a terror network found that communication of ideologies occurs horizontally - through immediate and secondary friends (~80%). The remaining twenty percent do so vertically - through kinship ties (Atran in press, Sageman 2004). Following enlistment, individuals often self-organize into isolated cells, the preferred size of which is eight members (Atran in press). As mentioned earlier, controlling the spread of an infectious disease can occur through reducing contacts in general or isolating infected individuals. While it may also be possible to reduce the spread of terrorist ideology by generally limiting communication within a population, this could also limit the spread of counterterrorist ideology and have large implications for expected freedoms of expression. It seems more practical to focus on identifying and isolating individuals “infected” with terrorist ideology. While covertness greatly impairs the ability to forecast and control emerging terrorist ideologies, detectable symptoms of an emerging terrorist ideology may include an increase in attacks, the death or capture of militants, or the dissemination of media (Stares and Yacoubian 2005). Active individuals might be identified via their own efforts to communicate their ideology or through identification of existing networks. Strategies presently
17
Infectiousness of terrorist ideology
used to reduce the ability of ideologues to contact others include imprisonment, military isolation, disruption of communication, etc. Unlike strategic counterterrorism, which aims to reduce susceptibility, operational counterterrorism focuses on reducing contact. This approach monitors suspected terrorists, collaborators, supporters, and sympathizers and uses the acquired intelligence to facilitate timely arrests (Gunaratna 2005). By targeting terrorists in the early stages of planning, operational counterterrorism is thought to have reduced what may have resulted in longer-term terrorist activity (Gunaratna 2005). Detecting the spread of terrorist ideologies before the onset of the obvious indicators (e.g., attacks, media dissemination) would better facilitate counterterrorism . For some infectious agents (e.g., Ebola, West-Nile, Rabies), contact with a reservoir host species drives transmission dynamics. Controlling such pathogens in peripheral hosts, such as humans, is greatly hindered because the pathogen is primarily supported in non-human hosts which may be outside the reach of control efforts. Moreover, when the reservoir host suffers little pathology, there is no selection on the infectious agent to evolve reduced virulence. This is the case for many of the zoonotic infectious agents that spill-over from wildlife reservoirs and cause high rates of morbidity and mortality in susceptible human populations. An analog exists in the spread of terrorist ideology. Individuals that abstain from the practice of terrorism may still sustain the founding ideology which they actively fund, harbor, and spread to recruit active terrorists. In doing so, they become reservoirs for the ideology. Like peripheral hosts, the individuals they recruit are more likely to exhibit the symptoms of the ideology
18
Infectiousness of terrorist ideology
(terrorist acts) than they are to spread the ideology or participate in recruitment activities. Decoupling terrorist recruitment from terrorist action makes it less likely that operational counterterrorism specifically directed against terrorists and their activities will reduce the spread of terrorism. In this scenario, the reservoir individuals harboring the ideology would be under-targeted. Terrorist networks with an important reservoir might be better modeled as an infectious pathogen with a reservoir host, with counterterrorism actions targeted specifically against the reservoir. Recovery Increasing recovery is the traditional goal of patient-based medicine, and incidentally results in a reduction in R0. The analogy for terrorism would be to reverse terrorist ideology at the individual/cell level, thus halting its spread and reducing terrorist acts. Anyone old enough to reflect on their childhood appreciates that ideologies change over time, and it seems safe to assume that “terrorist” is not necessarily a permanent ideological state. In borrowing from the SIR model framework, we might divide individuals into distinct ideological classes of terrorist (infected) and non-terrorist (susceptible or recovered/removed). While conversion from non-terrorist to terrorist seems to be analogous to an infectious process, recovery from terrorist to non-terrorist is not analogous to an immune defense. In addition, while an individual’s ideological intensity may wane or increase over time, a person does not necessarily adopt an ideology then lose it over time to a non-ideological state. Ideologies can persist over time, and may coexist with or interact with other ideologies. In other words, there may be a potential series of “infected” states. This is particularly important for the concept of recovery because a change in ideological states would depend on the extent that individuals can be exposed to alternative
19
Infectiousness of terrorist ideology
ideologies. In a sense, this is more like pitting two infectious agents against one another than the effect of an immune system. Although replacement is not the rule for infectious agents, we can turn to flukes (parasitic trematode worms) for insight. Trematodes have complex life cycles that typically involve a snail as a first-intermediate host. Once infected, snails rarely revert to an uninfected state. In many cases, several species of trematode infect the same species of snail, but there is generally only room for one trematode species at a time within an individual snail (few species pairs can coexist). For this reason, trematodes have developed special adaptations for battling with other trematodes inside the snail and, in many cases, one can predict which trematode species will win the internal battle for host occupation (Kuris and Lafferty 1994). If ideologies, like trematodes, have dominance hierarchies, facilitation of dominant ideologies could lead to “recovery.” Strategic counterterrorism would play an important role in developing and implementing such alternative ideologies. Mortality All else being equal, highly pathogenic infectious agents have a smaller R0 because infected individuals with short life spans have fewer opportunities to transmit an infection. A terrorist’s actions may create an analogous situation. Even in the absence of counterterrorist efforts, terrorist mortality rate should increase if militant terrorist activities are inherently dangerous (e.g., weapons training, handling explosives, suicide missions, primitive living). These dangers may therefore make it more difficult for terrorist ideology to spread. However, risky behavior and high mortality rates may also increase ideological spread. This would be particularly true if 20
Infectiousness of terrorist ideology
suicide attacks lead to martyrdom, benefits for the attacker’s family, a desirable afterlife, or if mortality achieves a terrorist goal and inspires others to join. Ecological and epidemiological models also suggest that removing infected individuals from the population can reduce the impact of the infectious agent. Although society does not permit the culling of infected patients, culling infected animals has been shown to be effective in preventing the spread of veterinary and zoonotic pathogens (recent examples include foot and mouth, avian influenza, and Nipah virus) (Barlow 1996). Analogously, counterterrorism efforts often attempt to cull terrorists. This may be motivated by a sense of justice, to directly reduce threat, and/or to help reduce the spread of terrorist ideology. However, as implied previously, culling might ironically enhance the spread of ideology through martyrdom and other factors. In this case, the killing of a terrorist might inspire susceptibles to adopt the very ideology counterterrorist operations aim to defuse. An SIR equation that links death and transmission through martyrdom is: dI/Idt = Sβ(1+αm) – (μ + α + r) and R0 = Sβ(1+αm)/(μ + α + r) where the variables are as before, and m (martyrdom) is the extent that the death of a terrorist (resulting from terrorist activity) increases transmission to susceptibles. 21
Infectiousness of terrorist ideology
We might then wish to explore how culling terrorists affects the spread of terrorism. The partial derivative of R0, with respect to α, indicates the slope of the relationship between culling and spread. This will be positive for m > 1/(r + μ) This means that where m is near zero, increasing terrorist death rate (such as through military intervention or suicide attacks) will always slow the spread of terrorist ideology. Nevertheless, with increasing m, r, or μ, an increase in terrorist death rate can increase the spread of terrorism. This makes obvious sense for m. Less intuitive is how a high background mortality or recovery rate increases the value of martyrdom. This occurs because when the lifespan of a terrorist (physical or ideological) is sufficiently short, their premature loss from the terrorist population has little cost to the spread of the ideology. Therefore, understanding the strength of martyrdom relative to recovery rates and background survivorship might allow better evaluation of counterterrorist strategies. Discussion So in the end, terrorism is just another deadly disease. It must be fought as a disease, as a slow, deadly, tragic, expensive, expansive, personal and national grind over the very long haul. It will be much harder to fight than just another war. -Dr. Erik Steele, Vice President for Patient Care Services, Eastern Maine Medical Center. With this chapter, we have presented a framework and set of models as alternative means of conceptualizing the nature of the terrorist threat. The epidemiological approach and tools
22
Infectiousness of terrorist ideology
presented here form the basis for understanding, controlling and predicting infectious pathogens in human and wildlife populations. By drawing parallels between the spread of infectious agents and the spread of terrorist ideology, we have sought to highlight both the similarities and contrasts between these two threats. In so doing, we suggest how the vast body of knowledge in ecology and epidemiology may be modified and potentially applied to understanding and containing the spread and/or effects of terrorism. Importantly, we propose that consideration of terrorism in this light may stimulate security strategists, policymakers, and others to develop creative solutions to counterterrorism . Nonetheless, while the terrorist ideology-infectious agent analogy has obvious utility, we recognize that is it is both young and imperfect. That ecology and epidemiology have led to advances in the control of infectious agents suggests that a similar approach could be used for counterterrorism . One particularly valuable aspect of such an approach is the discovery of non-intuitive outcomes, for example, when the removal of terrorists facilitates the spread of terrorist ideology through martyrdom via a successful recovery program. In some cases, basic strategies for controlling infectious agents may translate directly to controlling terrorism. In particular, ideological analogs to R0 could be used with coupled differential equations to predict the spread of terrorist ideology. As we have shown, the success of this approach is strongly dependent upon the variables used to construct the equations. Existing generic epidemiological models are not adequately suited for modeling terrorist ideology, just as they are not often adequately suited for modeling specific pathogens. Indeed, a good deal of critical thought will be required to construct adequate definitions for individuals that are ‘susceptible’ to, ‘infected’ with and ‘recovered’ from terrorist ideology, and for the model parameters that link these stages (e.g., transmission rate, death rate and aggregation of the
23
Infectiousness of terrorist ideology
ideology, as well as contact rate between individuals). These variable definitions and associated assumptions would ideally be sufficiently analogous to those in the epidemiological models on which they are based without losing the social dimensions that make them inherently unique. Simple epidemiological models could be customized for terrorist ideology by incorporating a sufficient amount of detail. Such model adaptation might consider the following factors. Infected individuals (terrorists) might be broken down into additional categories lacking analogues in epidemiological models (e.g., militants, educators, leaders). One might consider the possibility of vertical (parent-offspring) transmission of terrorist ideology. Unlike infectious agents that attack a host and, in turn, are attacked by the host’s immune system, infectious ideologies may invade a host that already harbors an ideology. An existing ideology might resist invasion or it may be replaced or altered. In addition, recovery, particularly as attempted through strategic counterterrorism, could also be an infectious process. It seems that models of terrorist ideology would benefit from consideration of heterogeneities in human behavior relating to spread. It would be worth considering how to incorporate terrorist intensity but, like macroparasite models, identify a currency (e.g., attack rate) useful for counterterrorist goals. Moreover, there is much to be gained from the study of novel pathogen outbreaks and the models used to predict their timing, size, and geography. As research into emerging diseases develops, it may prove exceptionally useful to counterterrorism campaigns that aim to use preventative strategies to thwart the spread of terrorist ideologies in the earliest stages of development. Collaboration between ecologists, epidemiologists, and social scientists has proven beneficial to the study of various human pathogens. Keeping with the analogy, such forms of cross-disciplinary collaboration should also benefit the study of terrorist ideology.
24
Infectiousness of terrorist ideology
Summary Box Future Research ● Develop working definitions and assumptions for the variables and parameters that characterize SIR models of terrorist ideology. ● Obtain quantitative data to explore the power of microparasite vs. macroparasite models in describing the spread of terrorist ideology. ● Determine how behavioral, sociological and cultural parameters would be accounted for in these models. ● Consider how the strength of these models vary when temporal (e.g., political cycles) and spatial (e.g., global terrorist networks vs. local terrorist cells) scales are accounted for. Open Questions ● How will ecological and epidemiological models of infectious agent spread be modified to accommodate the diversity of terrorist ideologies? ● What are the ethical considerations of comparing terrorist ideology to an infectious agent and, further, using this analogy to construct counterterror campaigns? ● Are the strengths and benefits of the infectious agent-terrorist ideology analogy sufficient to warrant continued pursuit of its utility? ● How can we improve operational and strategic counterterrorism using insights from the infectious agent-terrorist ideology analogy? 25
Infectiousness of terrorist ideology
Literature Cited Anderson, R. M., and R. M. May. 1979. Population biology of infectious diseases. Part 1. Nature 280:361-367. Ariza, L. M. 2005. Virtual Jihad: The Internet as the ideal terrorism recruiting tool. Scientific American 294(1):18-21. Atran, S. In press. The moral logic and growth of suicide terrorism. The Washington Quarterly 29:127-147. Barlow, N. D. 1996. The ecology of wildlife disease control: simple models revisited. Journal of Applied Ecology 33:303-314. Barry, J. M. 2004. The Great Influenza. Penguin, New York. Boyd, R., and P. J. Richerson. 2005. The Origin and Evolution of Cultures. Oxford University Press, New York. Combes, C. 2001. Parasitism: the ecology and evolution of intimate interactions. University of Chicago Press, Chicago. Crofton, H. D. 1971. A model of host-parasite relationships. Parasitology 63:343-364. Dawkins, R. 1976. The Selfish Gene. Oxford University Press, Oxford. Dawkins, R. 1989. The Selfish Gene. Oxford University Press, Oxford. Distin, K. 2005. The Selfish Meme: A Critical Reassessment. Cambridge University Press, Cambridge. Ehrlich, P.R. and S.A. Levin. 2005. The evolution of norms. PLoS Biology 3:943-948. Ewald, P. W. 1993. Evolution of Infectious Disease. Oxford, Oxford. Gladwell, M. 2000. The tipping point: how little things can make a big difference. Little, Brown and Company, Boston.
26
Infectiousness of terrorist ideology
Gunaratna, R. 2005. Ideology in Terrorism and Counter Terrorism: Lessons from combating Al Qaeda and Al Jemaah Al Islamiyah in Southeast Asia. Conflict Studies Research Center. Fraser, C., S. Riley, R.M. Anderson, and N.M. Ferguson 2004. Factors that make an infectious disease outbreak controllable. Proceedings of the National Academy of Science 101:6146-6151. Hochberg, M.E. 200X. Insights from parasite control models and applications. Pages XXX-XXX in R. Sagarin and T. Taylor, editors. Natural Defense: Darwinian Perspectives on Security in an Uncertain World. University of California Press, Berkeley. Hufnagel, L., D. Brockmann, and T. Geisel. 2004. Forecast and control of epidemics in a globalized world. Proceedings of The National Academy of Sciences of The United States of America 101:15124-15129. Johnson, A. M., J. Wadsworth, P. Elliot, L. Prior, P. Wallace, S. Blower, N. L. Webb, G. I. Heald, D. L. Miller, M. W. Adler, and R. M. Anderson. 1989. A pilot study of sexual lifestyle in a random sample of the population of Great Britain. AIDS 3:34-142. Kuris, A. M. and K. D. Lafferty. 1994. Community structure: larval trematodes in snail hosts. Annual Review of Ecology and Systematics 25:189-217. Lafferty, K. D. and R. D. Holt. 2003. How should environmental stress affect the population dynamics of disease? Ecology Letters 6:654–664. Lafferty, K. D., J. Porter, and S. E. Ford. 2004. Are diseases increasing in the ocean? Annual Review of Ecology, Evolution and Systematics 35:31-54. Lynch, A. 1999. Thought Contagion: How Belief Spreads Through Society. Basic Books, New York.
27
Infectiousness of terrorist ideology
May, R. M. and R. M. Anderson. 1979. Population biology of infectious diseases. Part II. Nature 280:455-461. May, R. M. and R. M. Anderson. 1988. The transmission dynamics of human immunodeficiency virus (HIV). Philosophical Transactions of the Royal Society of London B - Biological Sciences 321:565-607. Office of Homeland Security. 2002. National Strategy for Homeland Security. Office of the President of the United States of America, Washington, D.C. Prusher, I.R. 2006. Will Hamas change course? The Christian Science Monitor, 2/1/2006. Rigby, M. C. and Y. Moret. 2000. Life-history trade-offs and immune defenses. Pages 129-142 in R. Poulin, S. Morand, and A. Skorping, editors. Evolutionary Biology of Host-Parasite Relationships: Theory Meets Reality. Elsevier, Amsterdam. Roberts, M. G., A. P. Dobson, P. Arneberg, G. A. de Leo, R. C. Krecek, M. T. Manfredi, P. Lanfranchi, and E. Zaffaroni. 2002. Parasite community ecology and biodiversity. in P. J. Hudson, A. Rizzoli, B. T. Grenfell, H. Heesterbeek, and A. P. Dobson, editors. The Ecology of Wildlife Diseases. Oxford, Oxford. Sageman, M. 2004. Understanding Terror Networks. University of Pennsylvania Press, Philadelphia. Sober, E. and D. S. Wilson. 1998. Unto Others. Harvard University Press, Cambridge. Solomons, N. W. and M. E. Scott. 1994. Nutritional status of host populations influences parasitic infections. in M. Scott and G. Smith, editors. Parasitic and Infectious Diseases: Epidemiology and Ecology. Academic Press, New York.
28
Infectiousness of terrorist ideology
Smith, K.F., Dobson, A.P., McKenzie, F.E., Real, L.A., Smith, D.L., and M.L. Wilson. 2005. Ecological theory to enhance infectious disease control and public health policy. Frontiers in Ecology and the Environment 3:29-37. Stares, P. and M. Yacoubian. 2005. Terrorism as virus. The Washington Post, 10/23/2005. Zahavi A. 1975. Mate selection-a selection for a handicap. Journal of Theoretical Biology 53(1):205-14. 29
Infectiousness of terrorist ideology
Box 1. Making the metaphor practical In their informative article, Unconventional Approaches to an Unconventional Threat: A Counter-Epidemic Strategy, Stares and Yacoubian (2005) point out what they perceive to be the three most practical applications of epidemiology and public health to combating terrorism (they specify Islamist militancy, while we consider terrorism generically). 1. Epidemiologists observe rigorous standards of inquiry and analysis to understand the derivation, dynamics, and propagation of infectious agents. They seek clarity on the origins, geographical and social contours of an outbreak: where is the pathogen concentrated, how it is transmitted, who is most susceptible to infection, and why are some immune. Applying the same methodological approach to mapping and understanding terrorism can yield immediately useful guidance on where and how to counter it. 2. Epidemiologists recognize that infectious agents emerge and evolve as a result of complex interactive processes between hosts, pathogens and the environment in which they live. To make sense of this complexity, epidemiologists deconstruct the key constituent elements of an infectious agent. This model helps to understand the phenomenon in its entirety and anticipate how it might evolve in the future. The same systemic conception of infectious agents can be adapted to understand the constituent elements of terrorism and their evolution (Figure 1). 3. Epidemiologists view infectious agents as complex, multi-faceted phenomena. Public health officials have thus recognized that success in controlling and rolling back an epidemic requires a carefully orchestrated, systematic, prioritized, multi-pronged effort to address each of its constituent elements. However, it is also recognized that significant progress or major advances can sometimes be precipitated by relatively minor interventions. Again, there are lessons and insights to be learned here for orchestrating a global counterterrorism campaign.
30
Infectiousness of terrorist ideology
Figure 1. The analogy between the spread and dynamics of infectious agents and terrorist ideology is best depicted through a conceptual model of the systems where each entity exists. Adapted from Stares and Yacoubian (2005). Indirect/Direct TransmissionInfectious AgentHost Individual/Population/SpeciesAbiotic/Biotic Environment 31
Darw!n!an National Security: The !nfectiousness of Terrorist !deology
(changed so as not to appear on google)
“The challenge of terrorism is akin to fighting a virus in that we can accomplish a great deal but not eradicate the problem. We can take steps to prevent it, protect ourselves from it, and when an attack occurs, quarantine it, minimize the damage it inflicts, and attack it with all our power.” - Richard N. Haass, former Director of Policy Planning, U.S. State Department
Introduction
Terrorism in the 21st century is unconventional, unpredictable, and potentially unavoidable. In part, this is because contemporary terrorists are increasingly transnational, industrious, unorthodox in their methods, and decentralized (e.g., Ariza 2005, Ehrlich and Levin 2005, Atran in press). In mounting an effective counterterrorism campaign against such cryptic enemies, experts have begun to consider equally unconventional defenses, many of which are discussed throughout this book. Some view terrorism through the lens of epidemiology, where terrorist ideology is analogous to an infectious agent of threat to global public health (in particular, Stares and Yacoubian 2005). This is not the first time that scientists have looked to epidemiology to enhance their knowledge of how memes and ideologies spread, but it may be the most paramount. Similarities between terrorism and pathogen outbreaks have resulted in parallel, but independent, tracking, prevention, and control efforts. The National Strategy for Homeland Security, for instance, “… recognizes that the capabilities and laws we rely upon to defend America against terrorism are closely linked to those which we rely upon to deal with non-terrorist phenomena such as disease...” (Office of Homeland Security 2002: 4).
1
Infectiousness of terrorist ideology
If sufficient parallels exist between infectious agents and terrorist ideology, counterterrorist efforts may be able to draw on the already substantial body of theory developed for public health (e.g., Box 1). In this chapter, we disentangle these parallels in hopes of laying the foundation for future research on the infectiousness of terrorism. We specifically explore the utility of epidemiological theory in increasing our understanding of the dynamics and spread of terrorist ideology. We use the terms infectious agent, parasite, and pathogen interchangeably throughout this chapter. We use our findings to identify key unanswered questions and avenues for future research. In considering this analogy, we do not promote any particular counterterrorist strategy or recommend policy. Terrorist ideology is a type of meme or suite of memes. Memetics, the contagiousness of thought, and the dynamics that govern the spread of ideas have captivated a broad audience for many years (e.g., Dawkins 1976, Lynch 1999, Gladwell 2000, Distin 2005). Dawkins (1976) defines memes as any cultural entity (e.g., a fashion fad, song, idea, religion, language) that is replicated through exposure to humans and which has evolved as an efficient (though not necessarily perfect) copier of information and behavior. Much of the inquiry into memetics has centered on why some memes spread in an “epidemic” fashion. Like genes, memes spread if they provide a clear benefit to an individual, for example, through group identity, enhanced sexual attractiveness, or increased resource acquisition (Sober and Wilson 1998). However, memes that provide less tangible benefits, such as rewards after death, can also spread. Moreover, it has been proposed that memes with no apparent cost or benefit may arise and become ‘fixed’ in a population simply because they possess characteristics that render them more likely to be adopted than another meme (e.g., incorrectly transmitted verbal phrases that contain more audible syllables than their grammatically correct counterparts; Dawkins 1989). Even memes with a seemingly negative impact may spread through their provision of a net indirect benefit. For example, individuals with a “handicap” may be attractive to others due to the perception that the handicapped individual tolerates adverse conditions (Zahavi 1975). One thing that sets terrorist ideology apart from other memes (such as linguistic or religious) and aligns it with infectious agents is an explicit desire by society to control its spread. The infectiousness of terrorism: constructing the analogy Infectious agents depend on one or multiple hosts to support and complete their lifecycle (development, maturation, and reproduction). Host-pathogen dynamics can be viewed at the level of a host individual, host population, or community of species. Analogous to this, a given ideology can be “hosted” by a terrorist, terrorist cell, or terrorist organization. It is within these individuals and groups that ideology is conceived, formed, developed, and honed. Infectious agents need hosts just as ideology cannot exist without the minds that harbor it (though pathogens and ideologies can persist in alternate forms outside of the mind; many infectious agents have free-living resting stages and ideologies can be preserved outside the mind by various media; see below). Neither an infectious agent nor its host can exist in a system where environmental conditions are unsuitable. Indeed, abiotic (i.e., climatic) and biotic (i.e., competition, predation) factors help to shape the boundaries of an infectious organism’s geographic range. An analog to this is the political, social, and economic environment that shapes the region where terrorist ideology thrives. Changes to the environment might foster the evolution of ideology away from terrorism. For instance, the transition of a terrorist group to a political power may lead to ideological or methodological changes better suited for formal governing (Prusher 2006). Finally, it is the transmission of infectious agents and terrorist ideology between hosts/terrorists that maintains the entity’s existence within the system. Infectious agents can be transmitted between susceptible hosts directly, as in the case of sexually transmitted pathogens which require person-to-person contact. Transmission my also occur indirectly, by way of a vector that harbors the infectious agent (e.g., mosquitoes carrying malaria or ticks carrying Lyme disease) or through a contaminated vehicle such as food (e.g., salmonella), or liquid (e.g., giardiasis). The spread of terrorist ideology among individuals occurs via similar routes. Ideology may be transmitted directly, through the oral exchange of ideas from a terrorist to an individual who does not yet harbor the ideology. It may also spread indirectly, via a vehicle/vector such as broadcast media, print media, or the Internet. Ecological and epidemiological weaponry in the war on terror Terrorism as a plague teaches us that plagues are rarely ever wiped out. If we are lucky al-Qaida will be our smallpox, eradicated after a sustained global campaign. More likely, however, it is malaria, a deadly disease which keeps on killing in part because it keeps adapting and changing. So will other diseases and so will other terrorists. -Dr. Erik Steele, Vice President for Patient Care Services, Eastern Maine Medical Center Ecology and epidemiology have a highly developed set of theoretical tools and mathematical modeling approaches for understanding the basic properties of infectious agents. In many cases, such techniques have greatly informed control practices (Smith et al. 2005). Here we compile a theoretical framework from ecology and epidemiology that should be of use in the war on terror.
4
Infectiousness of terrorist ideology
Assembling the theoretical arsenal Basic models of pathogen spread divide individuals into population categories that reflect infection status (e.g., susceptible, infected, recovered, or SIR at the most basic level). SIR models apply specifically to a group of infectious agents called microparasites, which include many viruses, bacteria and protozoa. The rate of change in the abundance of individuals in each category is described by a differential equation constructed from variables such as contact rate, susceptibility, birth rate, mortality, cure, immunization, and the abundance of individuals in the various population categories. The frequency of a particular category can increase, decrease, or stabilize over time. A key to such models is that the differential equations are coupled. For simple formulations, it is possible to solve these equations analytically and gain general insight. In most cases, however, finding analytical solutions for coupled differential equations with more than two categories is difficult. In these cases, epidemiologists have turned to a very effective short-cut called the basic reproductive ratio, or R0. R0 is the estimated number of secondary cases directly arising from one primary case (Anderson and May 1979). In deterministic models, if R0 is < 1, the epidemic will fizzle, while for R0 > 1, an epidemic will occur. Once an epidemic begins, it grows at a rate known as the effective reproductive ratio, R, which tends to decline from R0. R can stabilize around 1 (become endemic in the population) or it can tend toward 0, at which point the pathogen is extirpated from the population (cycling, chaos and other complex outcomes are also possible). It might appear that R0 would not be very informative once an epidemic is already underway because, during an epidemic, we would be most interested in reducing R. However, because R changes with the
5
Infectiousness of terrorist ideology
state of the epidemic, it is more difficult to calculate than R0, which is a set value and applies to a limited set of conditions. Determining the structure of the R0 equation helps identify the variables (and their interactions) that are most likely to determine if infection will spread through an uninfected population. Although an SIR model tracks three classes of individuals over time, only the rate of change of the infected individuals is needed to calculate R0. A simple component of an SIR model which tracks the per-capita rate of change of infected individuals (I) in an SIR population can be written as: dI/Idt = Sβ – (μ + α + r) where S is the number of susceptible individuals, β is the transmission coefficient (how contagious the infectious agent is to susceptibles), μ is the background mortality rate, α is the additional mortality rate suffered by infected individuals, and r is the recovery rate. This model assumes that contacts occur at random and are density dependent (e.g., an individual living in a large city will come in contact with more people per day than an individual living in a rural community). R0 is simply a ratio comprised of the expected rate of transmission through contact with susceptible individuals divided by the sum of the death and recovery rates. The basic reproductive ratio is, therefore: R0 = Sβ/(μ + α + r) 6
Infectiousness of terrorist ideology
By analogy, individuals that are exposed to and “infected” with terrorist ideology become terrorists. In addition, there are non-terrorists in the population who are either susceptible or resistant to the ideology. The number of resistant individuals is not explicitly a component of R0, except that a greater proportion of resistants implies a lower number of susceptibles. Resistant individuals may be susceptibles that become resistant before contact or they may be terrorists that have disavowed their ideology. Understanding the components of R0 for the spread of terrorist ideology may provide greater insight into the types of counterterror strategies that could be used to reduce R0 below 1, and thus diminish the likelihood that terrorism will spread. Efforts to reduce R0 below 1 may concentrate on increasing the death and/or recovery rate of infecteds and/or decreasing the abundance (or susceptibility) of susceptibles, or their contact rate with infecteds. Reducing R0, however, is not the only means of controlling the spread of an infectious agent or terrorist ideology. Outbreaks of infectious agents are oftentimes apparent in the early stages, when very few individuals are infected. This is a direct result of the incentive that an infected individual receives for reporting their condition and thus obtaining treatment. This increases the likelihood that public health officials will learn of enough independent cases to implement control initiatives before the rate of spread increases. Not surprisingly, this scenario is in stark contrast to the spread of terrorist ideology because new converts cannot risk revealing their ideology to authorities. Relative to pathogen outbreaks, such behavior makes it exceedingly difficult to forecast and control emerging terrorist ideological ‘outbreaks,’ though this is not to suggest a complete lack of indicators. Symptoms of such an emerging ideological outbreak may include an increase in the frequency of attacks, the death or capture of militants, or the dissemination of
7
Infectiousness of terrorist ideology
media (Stares and Yacoubian 2005). It is important to recognize, however, that the inception of infectiousness is not necessarily correlated with the appearance of symptoms. This is a critical notion as the relationship between the timing of infectiousness and the onset of symptoms is among the key general properties of emerging infectious agents that public health officials look to when planning control initiatives. A similar snapshot of emerging terrorist ideologies should prove valuable to counterterrorism . Although vaccination is available for many of the pathogens that plague mankind, the evolutionary novelty and unpredictability of emerging infectious agents and outbreaks dampens its viability as a broad scale control strategy. In the absence of vaccines and treatment for pathogens on the verge of an outbreak, public health officials generally favor two basic control practices: 1) isolation of symptomatic individuals and 2) tracing and quarantining their contacts (Fraser et al. 2004). Both the implementation and success of these measures rely on a number of factors, ranging from the epidemiological characteristics of the infectious agent to the communication infrastructure of the public health agencies charged with control. Mathematical modeling of contemporary outbreaks (in particular SARS, HIV, smallpox and influenza) suggests that the success of these control measures is equally dependent on the proportion of transmission that occurs prior to the onset of clinical symptoms (θ) and the inherent transmissibility of the infectious agent (R0, as described in the previous section of this chapter) (see Fraser et al. 2004 for model details). Model simulations suggest that isolation, contact tracing and quarantine are indeed sufficient to control outbreaks of infections when the values of θ and R0 fall below a critical threshold: Control through isolation alone is possible when θ < 1/ R0, but when θ > 1/ R0 contact tracing is also required. When θ is very high, neither control
8
Infectiousness of terrorist ideology
measure (alone or in combination) is able to prevent the outbreak from progressing. The simplicity and efficacy of these control measures and the modeling techniques used to describe them should be attractive features from the standpoint of counterterrorism initiatives. Is intensity important?Adopting an SIR approach to terrorism assumes that all terrorists have the same intensity and capacity for spreading the ideology. However, if ideology is a continuous variable, then some individuals within a population may possess none of the ideology, some may posses a little, and a few may subscribe with high intensity. Individuals that subscribe most to the ideology may be more likely to spread the ideology than those that are merely sympathetic or supportive of the ideology. Is variation in the intensity of terrorist ideology among individuals sufficiently important to justify greater model complexity? A possible alternative is an SIR model where contact rate varies among individuals, allowing some individuals to be more active than others in spreading ideology. This has been of particular interest in understanding the spread of HIV because the number of sexual partners varies greatly among individuals, with the majority of transmission stemming from a small number of highly sexually active people (Johnson et al. 1989). The derivation of models applied to this scenario is beyond the scope of the paper, but the important result is that variation in contact rate among individuals can greatly increase R0 (May and Anderson 1988). This suggests that targeting individuals who are disproportionately active will reduce the spread of terrorist ideology, a conclusion that has intuitive appeal. However, these models describe differences in contact rate
9
Infectiousness of terrorist ideology
that are distinctive to an individual whether or not they are infected (i.e., the link between intensity and the propensity to spread an infection is indirect). Models developed for parasitic worms (macroparasites) might better mimic the spread of terrorist ideology than SIR models. Hosts exposed to many worms typically suffer higher pathology than hosts exposed to few worms. Hosts with many worms also contribute more to the growth of the total worm population. In addition, few infected hosts have the average level of infection: most have light, non-pathogenic infections and the majority of the worms are in a few highly infected individuals (such distribution of parasites in a host population is referred to as aggregation). For this reason, the SIR categorization of hosts as simply infected and uninfected fails to capture important aspects of the population dynamics of parasitic worms. Macroparasite models were developed to account for this level of biological detail (Crofton 1971, Anderson and May 1979, May and Anderson 1979). These models have provided considerable insight into how to target control efforts toward the few, most heavily infected individuals. In evaluating the utility of macroparasite models for terrorism, it is important to describe how macroparasite models only indirectly account for the intensity of infection. Such models do not distinguish infected from uninfected hosts. Instead of tracking infected hosts, they track the total worm population. The differential equation representing the growth of the worm population has a number of new terms. In a very simple formulation, the per-capita rate of change of the worm population is: dW/Wdt = ISβ – (μ + α + d + W(k+1)/Sk) 10
Infectiousness of terrorist ideology
where the variables are the same as above except that W is the total number of worms in the worm population, I is the total number of infectious stages, β represents contact between infective stages and hosts, d is the death rate of worms within the host and k is an inverse measure of the degree of aggregation of worms in the host population. Solving for the basic reproductive ratio of a macroparasite is much more difficult than for a microparasite, and alternative formulations for R0 in macroparasites can be considerably different, depending on the biological details of the model (Roberts et al. 2002). Because macroparasite models are affected by the variation in intensity among infected individuals, the currency in such terrorism models would be the total amount of terrorist ideology in the population. Possible metrics for intensity include attack rates and internet chatter. Attack rate is particularly tempting to consider as a currency because reducing attack rate is an ultimate goal for counterterrorist activities. While variation in ideological intensity suggests that counterterrorism should target terrorist leaders, the network structure of many terrorist organizations contrasts traditional leadership hierarchies and is increasingly “leaderless” (Sageman 2004; Atran in press), thus traditional strategies aimed at organizational “decapitation” are less likely to be effective than predicted by simple mixing models. Extending key epidemiological variables to the spread of terrorist ideology It is not necessarily clear which modeling approach provides a better analogy for the spread of terrorist ideology. However, we can gain further insight by deconstructing, and examining in
11
Infectiousness of terrorist ideology
detail, the major contributing variables in epidemiological models: susceptibility, contact rate, recovery, and mortality. Susceptibility Without susceptibles, an infectious agent cannot spread. Although some infectious agents are able to evolve adaptations to get past host defenses, they are less likely to adapt to two host species that differ greatly in physiology and/or evolutionary history (Combes 2001). As a result, not all host species are susceptible to all pathogens, and host specificity, or the restriction of the number of host species that a pathogen can infect, is widespread. Ideological specificity might arise for the same reasons as host specificity. The specific cultural identity of an individual (religious, national, historical) may predispose it to some, but not other ideologies. If an ideology is recognized as clearly foreign, an individual may more likely reject it as “non self.” Xenophobia is a common aspect of human cultures that probably acts to maintain cultural identity and prevent the spread of foreign ideologies (Boyd and Richerson 2005). For this reason, an ideology that is successful in spreading through a particular population may do so because it contains attributes compatible with that populations’ culture. In contrast, a population may be less susceptible to a terrorist ideology if that ideology differs greatly from the culture’s modal ideology. This suggests that the cross-cultural spread of terrorist ideology may be unusual, and it is probably not reasonable to assume that all cultures are equally susceptible to a particular ideology - just as all species are not susceptible to a particular infectious agent. 12
Infectiousness of terrorist ideology
A second pattern in susceptibility to infectious agents is that, for a single species (or closely related species), populations that have had an evolutionary history often evolve adaptations against an infectious agent while naïve populations can be more susceptible and less tolerant. The evolutionary pressures that result in the evolution of resistance derive from the negative fitness consequences of an infectious agent on its host. This suggests that nations with a history of battling terrorism might be more likely to resist the future spread of terrorist ideology within their population. A tool of modern public health is to reduce the pool of susceptibles through vaccination. Vaccination works by exposing uninfected individuals to dead or attenuated infectious agents, after which the immune system goes through the process of building specific antibodies that then lead to immunity. Even vaccinating only a fraction of susceptibles can reduce R0 below one (herd immunity). Vaccination is most viable as a broad-scale control strategy when used against well-known pathogens. An analogy to vaccination for terrorist ideology is exposing a population to a discredited version of the ideology to make it less likely that newly exposed individuals will find it attractive (Stares and Yacoubian 2005). An in-depth knowledge of a particular terrorist ideology would seem necessary for this approach to succeed. The environment can also alter a host’s susceptibility to an infectious agent. Thermal stress, for example, appears to increase the susceptibility of some marine organisms to different infectious agents (Lafferty et al. 2004). Other physiological stressors, such as lack of food, may force a host to redistribute resources away from defense against infectious agents (Rigby and Moret 2000). In some cases, public health officials can alter the environment, such as increasing nutrition, to
13
Infectiousness of terrorist ideology
reduce the susceptibility of individuals (Solomons and Scott 1994). While stress at the individual level should tend to increase susceptibility to infectious agents, it may, unexpectedly, reduce the spread of a pathogen through a population. This occurs for the following reasons: stress may reduce host abundance (and, therefore, contact rates), and stress can increase pathogen mortality rates either through killing pathogens directly or by leading to differentially high mortality of infected hosts (Lafferty and Holt 2003). The environment could similarly affect the number of individuals susceptible to terrorist ideology. In environments that are unfavorable (poor in justice, resources or freedom), new terrorist ideologies might hold appeal, particularly if they offer the promise of change and current or alternative ideologies do not appear effective. Therefore, efforts that reduce the desire for change or provide alternative pathways for change might decrease susceptibility. Still, environmental conditions that decrease the susceptibility of individuals to a particular ideology will not necessarily prevent the spread of that ideology. For example, improved economic conditions might decrease desire for change in individuals, but also increase opportunities for communication or implementation of a terrorist ideology, making it difficult to predict the net effect. Strategic counterterrorism seeks to reduce the political and operational space where terrorism can develop, spread and sustain (Gunaratna 2005). The targets of strategic counterterrorism include the foundations on which a culture’s education, religion, media, legislation and ideology are set. The mission is to counter terrorism through these institutions by promoting an ethic against terrorist ideology that extends to the greater community. Because strategic counterterrorism is in
14
Infectiousness of terrorist ideology
many ways preemptive, ‘susceptible’ individuals are likely to become infected with an anti-terrorism agenda before they are ever exposed to a terrorist ideology. Like vaccination campaigns, successful strategic counterterrorism would require early implementation, substantial funding and coordination. Contact rate Pathogens and ideologies spread primarily through contact. SIR models assume that in dense populations, individuals are more frequently in physical contact with one another. Another aspect of contact is the extent of movement among populations. This is difficult to capture in simple epidemiological models, but the logic is straightforward: contact between individuals of different populations is likely to lead to a wider spatial spread of an infectious agent. This is why increased contact through diffuse networks related to increases in transportation and modern trade has led to increasing concern for the spread of pandemics (Hufnagel et al. 2004). Current concern about avian influenza is one such example, and rightly so as the 1918 avian flu pandemic may have been associated with large-scale troop movements during World War I (Barry 2004). Efforts to reduce contact can be effective at slowing the spread of infectious agents. Two general approaches are used. The first is to reduce contact rates between individuals irrespective of their infection status. Safe-sex and personal hygiene campaigns are probably the best examples in modern society. For instance, children are taught to wash their hands, use toilets, and avoid spitting in public. The second approach is to specifically reduce contact between infected and uninfected individuals. In public schools, children with head lice are typically sent home to
15
Infectiousness of terrorist ideology
reduce the spread of lice to their classmates. For more serious infectious diseases, public health officials may quarantine infected individuals (and sometimes those that the infected individuals have contacted) in an effort to stop spread (Fraser et al. 2004). Such an approach was used to reduce the spread of SARS. Limits on the movement of infected individuals can be difficult because detection is challenging. In the SARS epidemic, body temperature scans in airports (e.g., Hong Kong) were used to identify individuals with fevers, and these individuals were then subjected to medical examination. Recent epidemiological models are making progress in predicting the spread of pathogens like SARS with evaluation of different control strategies (Hufnagel et al. 2004). How similar is the spread of an ideology to the spread of an infectious agent? Whereas sick people do not try to contact and infect other individuals, it is human nature to share ideas and convince others to agree with our opinions. This is not to suggest that pathogen transmission is passive. Infectious agents that are most likely to persist are those with traits enabling them to spread from one individual to another. Symptoms such as coughing, sneezing, and diarrhea are examples of behaviors induced by infectious agents to facilitate spread (Ewald 1993). It is also likely that ideologies are under analogous forms of selection for characteristics that favor spread (Hochberg, this volume). Successful religions often have doctrines favoring spread, such as active conversion of others, increased reproduction, early indoctrination, and retention. We may expect successful terrorist ideologies to possess adaptive traits for spread or to be aligned with existing religious ideologies. For example, terrorist ideologues have been successful at recruiting members through religious schools (Gunaratna 2005). 16
Infectiousness of terrorist ideology
There are different opportunities for the spread of an ideology than for the spread of an infectious disease. .Electronic communication greatly facilitates the spread of information at speeds and scales far exceeding physical contact. Information contact is increasing due to fewer language barriers, landline and cellular phones, television, radio and the Internet. This decouples the spread of an ideology from local population density (Ariza 2005) while simultaneously favoring decentralization and spatial spread (Atran in press). Cells, in particular, tend to communicate and exchange ideology and information primarily through the Internet (Atran in press). This is not to downplay the importance of direct communication. A study of enlistees into a terror network found that communication of ideologies occurs horizontally - through immediate and secondary friends (~80%). The remaining twenty percent do so vertically - through kinship ties (Atran in press, Sageman 2004). Following enlistment, individuals often self-organize into isolated cells, the preferred size of which is eight members (Atran in press). As mentioned earlier, controlling the spread of an infectious disease can occur through reducing contacts in general or isolating infected individuals. While it may also be possible to reduce the spread of terrorist ideology by generally limiting communication within a population, this could also limit the spread of counterterrorist ideology and have large implications for expected freedoms of expression. It seems more practical to focus on identifying and isolating individuals “infected” with terrorist ideology. While covertness greatly impairs the ability to forecast and control emerging terrorist ideologies, detectable symptoms of an emerging terrorist ideology may include an increase in attacks, the death or capture of militants, or the dissemination of media (Stares and Yacoubian 2005). Active individuals might be identified via their own efforts to communicate their ideology or through identification of existing networks. Strategies presently
17
Infectiousness of terrorist ideology
used to reduce the ability of ideologues to contact others include imprisonment, military isolation, disruption of communication, etc. Unlike strategic counterterrorism, which aims to reduce susceptibility, operational counterterrorism focuses on reducing contact. This approach monitors suspected terrorists, collaborators, supporters, and sympathizers and uses the acquired intelligence to facilitate timely arrests (Gunaratna 2005). By targeting terrorists in the early stages of planning, operational counterterrorism is thought to have reduced what may have resulted in longer-term terrorist activity (Gunaratna 2005). Detecting the spread of terrorist ideologies before the onset of the obvious indicators (e.g., attacks, media dissemination) would better facilitate counterterrorism . For some infectious agents (e.g., Ebola, West-Nile, Rabies), contact with a reservoir host species drives transmission dynamics. Controlling such pathogens in peripheral hosts, such as humans, is greatly hindered because the pathogen is primarily supported in non-human hosts which may be outside the reach of control efforts. Moreover, when the reservoir host suffers little pathology, there is no selection on the infectious agent to evolve reduced virulence. This is the case for many of the zoonotic infectious agents that spill-over from wildlife reservoirs and cause high rates of morbidity and mortality in susceptible human populations. An analog exists in the spread of terrorist ideology. Individuals that abstain from the practice of terrorism may still sustain the founding ideology which they actively fund, harbor, and spread to recruit active terrorists. In doing so, they become reservoirs for the ideology. Like peripheral hosts, the individuals they recruit are more likely to exhibit the symptoms of the ideology
18
Infectiousness of terrorist ideology
(terrorist acts) than they are to spread the ideology or participate in recruitment activities. Decoupling terrorist recruitment from terrorist action makes it less likely that operational counterterrorism specifically directed against terrorists and their activities will reduce the spread of terrorism. In this scenario, the reservoir individuals harboring the ideology would be under-targeted. Terrorist networks with an important reservoir might be better modeled as an infectious pathogen with a reservoir host, with counterterrorism actions targeted specifically against the reservoir. Recovery Increasing recovery is the traditional goal of patient-based medicine, and incidentally results in a reduction in R0. The analogy for terrorism would be to reverse terrorist ideology at the individual/cell level, thus halting its spread and reducing terrorist acts. Anyone old enough to reflect on their childhood appreciates that ideologies change over time, and it seems safe to assume that “terrorist” is not necessarily a permanent ideological state. In borrowing from the SIR model framework, we might divide individuals into distinct ideological classes of terrorist (infected) and non-terrorist (susceptible or recovered/removed). While conversion from non-terrorist to terrorist seems to be analogous to an infectious process, recovery from terrorist to non-terrorist is not analogous to an immune defense. In addition, while an individual’s ideological intensity may wane or increase over time, a person does not necessarily adopt an ideology then lose it over time to a non-ideological state. Ideologies can persist over time, and may coexist with or interact with other ideologies. In other words, there may be a potential series of “infected” states. This is particularly important for the concept of recovery because a change in ideological states would depend on the extent that individuals can be exposed to alternative
19
Infectiousness of terrorist ideology
ideologies. In a sense, this is more like pitting two infectious agents against one another than the effect of an immune system. Although replacement is not the rule for infectious agents, we can turn to flukes (parasitic trematode worms) for insight. Trematodes have complex life cycles that typically involve a snail as a first-intermediate host. Once infected, snails rarely revert to an uninfected state. In many cases, several species of trematode infect the same species of snail, but there is generally only room for one trematode species at a time within an individual snail (few species pairs can coexist). For this reason, trematodes have developed special adaptations for battling with other trematodes inside the snail and, in many cases, one can predict which trematode species will win the internal battle for host occupation (Kuris and Lafferty 1994). If ideologies, like trematodes, have dominance hierarchies, facilitation of dominant ideologies could lead to “recovery.” Strategic counterterrorism would play an important role in developing and implementing such alternative ideologies. Mortality All else being equal, highly pathogenic infectious agents have a smaller R0 because infected individuals with short life spans have fewer opportunities to transmit an infection. A terrorist’s actions may create an analogous situation. Even in the absence of counterterrorist efforts, terrorist mortality rate should increase if militant terrorist activities are inherently dangerous (e.g., weapons training, handling explosives, suicide missions, primitive living). These dangers may therefore make it more difficult for terrorist ideology to spread. However, risky behavior and high mortality rates may also increase ideological spread. This would be particularly true if 20
Infectiousness of terrorist ideology
suicide attacks lead to martyrdom, benefits for the attacker’s family, a desirable afterlife, or if mortality achieves a terrorist goal and inspires others to join. Ecological and epidemiological models also suggest that removing infected individuals from the population can reduce the impact of the infectious agent. Although society does not permit the culling of infected patients, culling infected animals has been shown to be effective in preventing the spread of veterinary and zoonotic pathogens (recent examples include foot and mouth, avian influenza, and Nipah virus) (Barlow 1996). Analogously, counterterrorism efforts often attempt to cull terrorists. This may be motivated by a sense of justice, to directly reduce threat, and/or to help reduce the spread of terrorist ideology. However, as implied previously, culling might ironically enhance the spread of ideology through martyrdom and other factors. In this case, the killing of a terrorist might inspire susceptibles to adopt the very ideology counterterrorist operations aim to defuse. An SIR equation that links death and transmission through martyrdom is: dI/Idt = Sβ(1+αm) – (μ + α + r) and R0 = Sβ(1+αm)/(μ + α + r) where the variables are as before, and m (martyrdom) is the extent that the death of a terrorist (resulting from terrorist activity) increases transmission to susceptibles. 21
Infectiousness of terrorist ideology
We might then wish to explore how culling terrorists affects the spread of terrorism. The partial derivative of R0, with respect to α, indicates the slope of the relationship between culling and spread. This will be positive for m > 1/(r + μ) This means that where m is near zero, increasing terrorist death rate (such as through military intervention or suicide attacks) will always slow the spread of terrorist ideology. Nevertheless, with increasing m, r, or μ, an increase in terrorist death rate can increase the spread of terrorism. This makes obvious sense for m. Less intuitive is how a high background mortality or recovery rate increases the value of martyrdom. This occurs because when the lifespan of a terrorist (physical or ideological) is sufficiently short, their premature loss from the terrorist population has little cost to the spread of the ideology. Therefore, understanding the strength of martyrdom relative to recovery rates and background survivorship might allow better evaluation of counterterrorist strategies. Discussion So in the end, terrorism is just another deadly disease. It must be fought as a disease, as a slow, deadly, tragic, expensive, expansive, personal and national grind over the very long haul. It will be much harder to fight than just another war. -Dr. Erik Steele, Vice President for Patient Care Services, Eastern Maine Medical Center. With this chapter, we have presented a framework and set of models as alternative means of conceptualizing the nature of the terrorist threat. The epidemiological approach and tools
22
Infectiousness of terrorist ideology
presented here form the basis for understanding, controlling and predicting infectious pathogens in human and wildlife populations. By drawing parallels between the spread of infectious agents and the spread of terrorist ideology, we have sought to highlight both the similarities and contrasts between these two threats. In so doing, we suggest how the vast body of knowledge in ecology and epidemiology may be modified and potentially applied to understanding and containing the spread and/or effects of terrorism. Importantly, we propose that consideration of terrorism in this light may stimulate security strategists, policymakers, and others to develop creative solutions to counterterrorism . Nonetheless, while the terrorist ideology-infectious agent analogy has obvious utility, we recognize that is it is both young and imperfect. That ecology and epidemiology have led to advances in the control of infectious agents suggests that a similar approach could be used for counterterrorism . One particularly valuable aspect of such an approach is the discovery of non-intuitive outcomes, for example, when the removal of terrorists facilitates the spread of terrorist ideology through martyrdom via a successful recovery program. In some cases, basic strategies for controlling infectious agents may translate directly to controlling terrorism. In particular, ideological analogs to R0 could be used with coupled differential equations to predict the spread of terrorist ideology. As we have shown, the success of this approach is strongly dependent upon the variables used to construct the equations. Existing generic epidemiological models are not adequately suited for modeling terrorist ideology, just as they are not often adequately suited for modeling specific pathogens. Indeed, a good deal of critical thought will be required to construct adequate definitions for individuals that are ‘susceptible’ to, ‘infected’ with and ‘recovered’ from terrorist ideology, and for the model parameters that link these stages (e.g., transmission rate, death rate and aggregation of the
23
Infectiousness of terrorist ideology
ideology, as well as contact rate between individuals). These variable definitions and associated assumptions would ideally be sufficiently analogous to those in the epidemiological models on which they are based without losing the social dimensions that make them inherently unique. Simple epidemiological models could be customized for terrorist ideology by incorporating a sufficient amount of detail. Such model adaptation might consider the following factors. Infected individuals (terrorists) might be broken down into additional categories lacking analogues in epidemiological models (e.g., militants, educators, leaders). One might consider the possibility of vertical (parent-offspring) transmission of terrorist ideology. Unlike infectious agents that attack a host and, in turn, are attacked by the host’s immune system, infectious ideologies may invade a host that already harbors an ideology. An existing ideology might resist invasion or it may be replaced or altered. In addition, recovery, particularly as attempted through strategic counterterrorism, could also be an infectious process. It seems that models of terrorist ideology would benefit from consideration of heterogeneities in human behavior relating to spread. It would be worth considering how to incorporate terrorist intensity but, like macroparasite models, identify a currency (e.g., attack rate) useful for counterterrorist goals. Moreover, there is much to be gained from the study of novel pathogen outbreaks and the models used to predict their timing, size, and geography. As research into emerging diseases develops, it may prove exceptionally useful to counterterrorism campaigns that aim to use preventative strategies to thwart the spread of terrorist ideologies in the earliest stages of development. Collaboration between ecologists, epidemiologists, and social scientists has proven beneficial to the study of various human pathogens. Keeping with the analogy, such forms of cross-disciplinary collaboration should also benefit the study of terrorist ideology.
24
Infectiousness of terrorist ideology
Summary Box Future Research ● Develop working definitions and assumptions for the variables and parameters that characterize SIR models of terrorist ideology. ● Obtain quantitative data to explore the power of microparasite vs. macroparasite models in describing the spread of terrorist ideology. ● Determine how behavioral, sociological and cultural parameters would be accounted for in these models. ● Consider how the strength of these models vary when temporal (e.g., political cycles) and spatial (e.g., global terrorist networks vs. local terrorist cells) scales are accounted for. Open Questions ● How will ecological and epidemiological models of infectious agent spread be modified to accommodate the diversity of terrorist ideologies? ● What are the ethical considerations of comparing terrorist ideology to an infectious agent and, further, using this analogy to construct counterterror campaigns? ● Are the strengths and benefits of the infectious agent-terrorist ideology analogy sufficient to warrant continued pursuit of its utility? ● How can we improve operational and strategic counterterrorism using insights from the infectious agent-terrorist ideology analogy? 25
Infectiousness of terrorist ideology
Literature Cited Anderson, R. M., and R. M. May. 1979. Population biology of infectious diseases. Part 1. Nature 280:361-367. Ariza, L. M. 2005. Virtual Jihad: The Internet as the ideal terrorism recruiting tool. Scientific American 294(1):18-21. Atran, S. In press. The moral logic and growth of suicide terrorism. The Washington Quarterly 29:127-147. Barlow, N. D. 1996. The ecology of wildlife disease control: simple models revisited. Journal of Applied Ecology 33:303-314. Barry, J. M. 2004. The Great Influenza. Penguin, New York. Boyd, R., and P. J. Richerson. 2005. The Origin and Evolution of Cultures. Oxford University Press, New York. Combes, C. 2001. Parasitism: the ecology and evolution of intimate interactions. University of Chicago Press, Chicago. Crofton, H. D. 1971. A model of host-parasite relationships. Parasitology 63:343-364. Dawkins, R. 1976. The Selfish Gene. Oxford University Press, Oxford. Dawkins, R. 1989. The Selfish Gene. Oxford University Press, Oxford. Distin, K. 2005. The Selfish Meme: A Critical Reassessment. Cambridge University Press, Cambridge. Ehrlich, P.R. and S.A. Levin. 2005. The evolution of norms. PLoS Biology 3:943-948. Ewald, P. W. 1993. Evolution of Infectious Disease. Oxford, Oxford. Gladwell, M. 2000. The tipping point: how little things can make a big difference. Little, Brown and Company, Boston.
26
Infectiousness of terrorist ideology
Gunaratna, R. 2005. Ideology in Terrorism and Counter Terrorism: Lessons from combating Al Qaeda and Al Jemaah Al Islamiyah in Southeast Asia. Conflict Studies Research Center. Fraser, C., S. Riley, R.M. Anderson, and N.M. Ferguson 2004. Factors that make an infectious disease outbreak controllable. Proceedings of the National Academy of Science 101:6146-6151. Hochberg, M.E. 200X. Insights from parasite control models and applications. Pages XXX-XXX in R. Sagarin and T. Taylor, editors. Natural Defense: Darwinian Perspectives on Security in an Uncertain World. University of California Press, Berkeley. Hufnagel, L., D. Brockmann, and T. Geisel. 2004. Forecast and control of epidemics in a globalized world. Proceedings of The National Academy of Sciences of The United States of America 101:15124-15129. Johnson, A. M., J. Wadsworth, P. Elliot, L. Prior, P. Wallace, S. Blower, N. L. Webb, G. I. Heald, D. L. Miller, M. W. Adler, and R. M. Anderson. 1989. A pilot study of sexual lifestyle in a random sample of the population of Great Britain. AIDS 3:34-142. Kuris, A. M. and K. D. Lafferty. 1994. Community structure: larval trematodes in snail hosts. Annual Review of Ecology and Systematics 25:189-217. Lafferty, K. D. and R. D. Holt. 2003. How should environmental stress affect the population dynamics of disease? Ecology Letters 6:654–664. Lafferty, K. D., J. Porter, and S. E. Ford. 2004. Are diseases increasing in the ocean? Annual Review of Ecology, Evolution and Systematics 35:31-54. Lynch, A. 1999. Thought Contagion: How Belief Spreads Through Society. Basic Books, New York.
27
Infectiousness of terrorist ideology
May, R. M. and R. M. Anderson. 1979. Population biology of infectious diseases. Part II. Nature 280:455-461. May, R. M. and R. M. Anderson. 1988. The transmission dynamics of human immunodeficiency virus (HIV). Philosophical Transactions of the Royal Society of London B - Biological Sciences 321:565-607. Office of Homeland Security. 2002. National Strategy for Homeland Security. Office of the President of the United States of America, Washington, D.C. Prusher, I.R. 2006. Will Hamas change course? The Christian Science Monitor, 2/1/2006. Rigby, M. C. and Y. Moret. 2000. Life-history trade-offs and immune defenses. Pages 129-142 in R. Poulin, S. Morand, and A. Skorping, editors. Evolutionary Biology of Host-Parasite Relationships: Theory Meets Reality. Elsevier, Amsterdam. Roberts, M. G., A. P. Dobson, P. Arneberg, G. A. de Leo, R. C. Krecek, M. T. Manfredi, P. Lanfranchi, and E. Zaffaroni. 2002. Parasite community ecology and biodiversity. in P. J. Hudson, A. Rizzoli, B. T. Grenfell, H. Heesterbeek, and A. P. Dobson, editors. The Ecology of Wildlife Diseases. Oxford, Oxford. Sageman, M. 2004. Understanding Terror Networks. University of Pennsylvania Press, Philadelphia. Sober, E. and D. S. Wilson. 1998. Unto Others. Harvard University Press, Cambridge. Solomons, N. W. and M. E. Scott. 1994. Nutritional status of host populations influences parasitic infections. in M. Scott and G. Smith, editors. Parasitic and Infectious Diseases: Epidemiology and Ecology. Academic Press, New York.
28
Infectiousness of terrorist ideology
Smith, K.F., Dobson, A.P., McKenzie, F.E., Real, L.A., Smith, D.L., and M.L. Wilson. 2005. Ecological theory to enhance infectious disease control and public health policy. Frontiers in Ecology and the Environment 3:29-37. Stares, P. and M. Yacoubian. 2005. Terrorism as virus. The Washington Post, 10/23/2005. Zahavi A. 1975. Mate selection-a selection for a handicap. Journal of Theoretical Biology 53(1):205-14. 29
Infectiousness of terrorist ideology
Box 1. Making the metaphor practical In their informative article, Unconventional Approaches to an Unconventional Threat: A Counter-Epidemic Strategy, Stares and Yacoubian (2005) point out what they perceive to be the three most practical applications of epidemiology and public health to combating terrorism (they specify Islamist militancy, while we consider terrorism generically). 1. Epidemiologists observe rigorous standards of inquiry and analysis to understand the derivation, dynamics, and propagation of infectious agents. They seek clarity on the origins, geographical and social contours of an outbreak: where is the pathogen concentrated, how it is transmitted, who is most susceptible to infection, and why are some immune. Applying the same methodological approach to mapping and understanding terrorism can yield immediately useful guidance on where and how to counter it. 2. Epidemiologists recognize that infectious agents emerge and evolve as a result of complex interactive processes between hosts, pathogens and the environment in which they live. To make sense of this complexity, epidemiologists deconstruct the key constituent elements of an infectious agent. This model helps to understand the phenomenon in its entirety and anticipate how it might evolve in the future. The same systemic conception of infectious agents can be adapted to understand the constituent elements of terrorism and their evolution (Figure 1). 3. Epidemiologists view infectious agents as complex, multi-faceted phenomena. Public health officials have thus recognized that success in controlling and rolling back an epidemic requires a carefully orchestrated, systematic, prioritized, multi-pronged effort to address each of its constituent elements. However, it is also recognized that significant progress or major advances can sometimes be precipitated by relatively minor interventions. Again, there are lessons and insights to be learned here for orchestrating a global counterterrorism campaign.
30
Infectiousness of terrorist ideology
Figure 1. The analogy between the spread and dynamics of infectious agents and terrorist ideology is best depicted through a conceptual model of the systems where each entity exists. Adapted from Stares and Yacoubian (2005). Indirect/Direct TransmissionInfectious AgentHost Individual/Population/SpeciesAbiotic/Biotic Environment 31