Abstract
Objective: In this study, we investigate extremists’ appraisals of and sensitivity to perceived sanction risk during the commission of arson. We pay specific attention to the decision-making processes of extremists leading up to and during the offending opportunity. Methods: We examined data collected from self-reported communiqués (n = 275) describing acts of arson committed by radical environmental extremists. Results: We found that extremists, like other criminals, are sensitive to situational factors that affect the certainty of apprehension. Additionally, extremists work to reduce the risk of detection by engaging in crime-specific risk management techniques prior to and during the offending opportunity. Conclusions: Analysis of the communiqués is consistent with recent works on extremism, situational crime prevention, and restrictive deterrence. We discuss our findings in the context of rational choice and situational crime prevention theory and the advancement of preventative policies aimed at ideological and political crime.
Although it is not difficult to articulate the ideological motivations behind a particular act of extremism, it remains unclear how to best reduce these ideologically motivated “threatened or actual use[s] of illegal force” (LaFree and Dugan 2007:184). This issue has motivated a recent growth in criminological research on violent extremism, which often takes one of two forms (see Freilich, Gruenewald, and Mandala 2019). One line of research examines similarities in psychological and social risk factors between extremists and nonideological offenders, including personality, childhood adversity, involvement with deviant peers, lack of stable employment, and a history of mental illness (e.g., LaFree et al. 2018; Simi et al. 2016). This research can be summarized as a “root-cause” approach, which attempts to identify criminogenic risk factors that predict the onset, continuation, and desistance from extremism and ideological criminal involvement.
Although it is important to understand what propels individuals into (and out of) extremism, others have pointed to the intractability of individual dispositions in efforts to predict and ultimately prevent specific acts of extremism (Clarke and Newman 2006; Freilich and Newman 2009). Consider low self-control and related psychosocial characteristics, which are consistently associated with criminal involvement generally. Whether these concepts are related to ideological crime is less clear. Whereas many common crimes are “acts of force and fraud in pursuit of self-interest”, extremism is a behavior that usually “reflect[s] commitment to a political cause” (Hirschi and Gottfredson 2001:94; see also LaFree and Dugan 2007). And unlike many other crimes, violent extremism is often done without immediate gratification (or at least not primarily for it) (Dugan, LaFree, and Piquero 2005; Victoroff 2005). Thus, it seems the unique motivational properties of ideological crime can confound the influence of traditional criminogenic dispositions on violent extremism, potentially limiting the efficacy of individual-level interventions (Silke 2008).
An alternative approach is to move beyond dispositions and focus on the context in which specific acts of extremism occur. That is, the immediate environmental and situational features of a specific criminal opportunity offer additional sources for manipulable policy intervention (Nagin, Solow, and Lum 2015). This approach draws on environmental perspectives of crime (e.g., crime pattern theory, routine activity theory) to elucidate temporal and spatial elements that shape offender decision making and patterns of victimization, necessarily inculcating rational choice theory (RCT) and situational crime prevention (SCP) (Clarke and Cornish 1985). Here, crime reduction is achieved through immediate, environmental manipulations that increase the risks and efforts needed to successfully commit crime, thereby reducing potential rewards of crime.
Research on terrorism has provided promising evidence for the efficacy of RCT and SCP-inspired interventions; however, the effects of these interventions vary considerably across data sources, intervention strategies, and movements (e.g., Carson et al. 2020; Dugan and Chenoweth 2012; Fisher and Becker 2021; Perry et al. 2017). Moreover, much of this research has been evaluative in nature—how much, if at all, do situational interventions affect rates of ideological crime? The mechanisms (e.g., perceived sanction threats) by which any effect is achieved is often left in the background (see, e.g., Nagin et al. 2015).
The aim of the current study is to compliment macro- and meso-level research on extremism by examining actual accounts of ideological crime, described by the offenders themselves, to gauge situational decision-making processes and behavioral patterns before and during a carried-out attack. We contextualize our theoretical discussion by examining politically motivated arson perpetrated by one of the most active movements globally and in the timeline of U.S. domestic extremism: radical environmental extremists (Ackerman 2003; Carson et al. 2012; Carson, Dugan, and Yang 2020; Leader & Probst, 2003). 1
A primary goal of this movement is to “protect” the environment through the property destruction of perceived environmental threats (e.g., oil companies, animal farms). Such targets, however, are typically accompanied by high levels of risk for detection and arrest. This underscores the importance of careful planning, skilled execution, and quick situational decision-making to ensure the success of an attack and, in turn, advance the intended political message (Robinson, Marchment, and Gill 2019). The accounts offered by the extremists in our sample provide support for RCT and SCP processes highlighted in prior works and contribute to evolving conversations about the complexities of displacement and adaptation (e.g., Freilich et al. 2019), while underscoring the importance of sanction risk perceptions in extremists’ decision-making and strategic behaviors (Gibbs 1975; Jacobs 2010).
Rational Choice Theory, Situational Crime Prevention, and Extremism
An important element of ideological crime is that attacks often involve the exploitation of opportunities through determination, careful planning, and deliberate decision-making processes (e.g., Clarke and Newman 2006; Gill, Horgan, and Deckert 2014; Smith and Damphousse 2009). Consequently, some have suggested rational choice and deterrence models are important for understanding extremism (see LaFree, Dugan, and Korte 2009). Here researchers are interested in the extent to which governmental policies make these crimes riskier and less rewarding.
For instance, Dugan, LaFree, and Piquero (2005) examined trends in ariel hijackings using data on global terrorist events. The authors found that certainty-based strategies aimed at reducing the perceived likelihood of success through increased presence of metal detectors and law enforcement at passenger checkpoints significantly reduced the number of new hijacking attempts albeit with an important caveat. These interventions had a stronger effect on nonterrorist hijackings compared to those attempted by terrorists. The authors concluded that greater attention to the rewards of terrorism is necessary to fully capture processes of ideological crime. Towards this end, Dugan and Chenoweth (2012) examined the joint and differential effects of repressive actions related to increased punishment (e.g., retaliation) and conciliatory actions that increase the utility of abstaining (e.g., allowing greater freedoms) on subsequent Palestinian terror attacks on the Israeli state. While repressive action was generally unrelated to subsequent attacks, conciliary action by way of increasing the benefits of nonaction produced significant decreases in terror attacks.
In a related study on terror attacks against Israel, Bejan and Parkin (2015) did not find support for conciliatory action on subsequent terror, but instead demonstrated that increases in repressive actions lead to a significant reduction in terrorist attacks. Moreover, Carson et al. (2020) examined the effect of both the costs and benefits on the radical eco-movement and found that improvements to the environment (i.e., conciliatory action) did not suppress the radical eco-movement, but federal actions that put the environment at risk do fuel the movement. Interestingly, however, illegal behavior of this movement did decrease as the number of arrests for these crimes increased (see also Yang and Jen 2018).
Although these studies suggest governmental action towards extremism matters, the way they matter appears to vary considerably by intervention type and target, challenging the efficacy of large-scale or “universal” deterrence-based counterterrorism efforts (Fisher and Becker 2021; Perry et al. 2017). One factor leading to inconsistent findings is that the “consequences” of interventions that rely on punishment by the state may be perceived as too distal, especially compared to other more immediate considerations around the time of an offense (see Cusson, 1993; Freilich and Chermak, 2009). Further complicating this issue are qualitative differences across extremist groups and movements. For example, the primary goal of the radical environmentalist movement (REM) is to protect the environment and animals. While REM extremists support the use of illegal actions to promote their cause, the use of interpersonal violence is rare (Carson et al. 2012). In general, specific extremist movements often rely on specific crimes because they are selective about their targets due to varying resources and skills (Gill et al. 2017; Tench et al. 2016). Thus, global government interventions are unlikely to produce uniform deterrent effects across groups.
Instead, proponents of SCP shift empirical focus from broader geographical and organizational contexts to the context of specific criminogenic events—situations conducive to criminal victimization. Rooted in opportunity and decision-making theories, SCP seeks to understand how actors take advantage of situated opportunities to commit crime based on situational cues about the risks, costs, and benefits of a particular crime (Clarke 1983; Clarke and Cornish 1985). Theoretically, SCP interventions attempt to make targets less vulnerable and less attractive by increasing the risks of offending, as well as the effort required to successfully carry out the offense. To illustrate, consider one of the most important aspects in the decision to engage in extremism—target selection. There are a broad range of targets for groups to choose from, yet not all offer the same opportunity for an attack. The attractiveness of a target varies based on operational goals of the group, situational contingencies, and attributes of the target (Gruenewald et al. 2015). Therefore, under the SCP framework, reduction in vulnerability can be achieved through defensive measures such as target hardening where the safeguarding of public spaces is achieved through efforts like positioning fences and installing screening devices (Hsu and McDowall 2017; Newman and Clarke 2017).
Research on extremism generally shows a negative association between ideological crime and SCP interventions like target hardening (Lasley and Guffey 2017; Mandala and Freilich 2018; Romyn and Kebbell 2018). Nevertheless, an important issue remains—the assumed mechanisms (e.g., perceived sanction certainty) by which reduction is achieved are often left in the background. That is, research has tended to focus on if interventions reduce ideological crime rather than how and why. This has led some to challenge the efficacy of SCP efforts suggesting they may not reduce crime but instead displace it in terms of geography, time, target selection, and/or crime types (see Freilich et al. 2019; see also Guerette and Bowers, 2009).
To be sure, the displacement of ideological crime has been investigated, and researchers generally show displacement is not a guarantee—in many cases, environmental intervention even creates a diffusion of benefits (e.g., Hsu and Apel 2015; Hsu, Vásquez and McDowall 2018; Yang and Jen 2018; see also Clarke and Weisburd 1994). Still, it is difficult to define the geographical and temporal boundaries of a “situation” in which an act of extremism may occur, further complicating issues around displacement. This is in part because “there is currently no first-hand understanding within the terrorism literature on offender decision-making in the context of risk” (Gill et al. 2020:146). Thus, a primary goal of the current study is to assess how extremists themselves define and evaluate criminal opportunities in hopes of complimenting important insights from macro- and meso-level RCT and SCP research on ideological crime.
Situational Decision-Making and Extremism
The central premise of environmental approaches to crime is that situational characteristics of the physical and social environment affect the vulnerability of potential criminal targets, whether human or physical, to victimization. Importantly, emerging insights from perceptual deterrence research show that generally, would-be offenders are sensitive to environmental cues, which inform key decision-making perceptions (Pickett, 2018; Pogarsky et al. 2017; see Apel 2021 for review). Equally important, as information changes and evolves, people adjust their evaluations of the risks inherent to the opportunity and alter behavior accordingly (Barnum et al. 2021; Thomas et al. 2018).
Whether these situational processes apply to extremists, however, is less clear. There is evidence demonstrating extremists do not operate entirely in line with their nonideological criminal counterparts when assessing the risks and rewards for a crime opportunity. For instance, Robinson, Marchment, and Gill (2019) recently compared property crime trends between extremists and nonideological criminals and found that extremists were “less concerned” about the presence of environmental factors that signal increased levels of risk compared to their nonideological counterparts. The authors further identified many cases where extremists victimized targets with a greater number of risk factors present than did their nonideological counterparts who tended to select more innocuous targets.
One interpretation of this finding is that the presence of security measures signal to extremist movement members that the intended target is high profile and therefore likely to send a strong political message. Ostensibly, the focus is on anticipated benefits where considerations about potential risks are ancillary (see Caplan 2006). Notwithstanding these motivations, successful attacks are most likely to send the desired political message. As such, it would seem, the risks associated with an attack are indeed pertinent to the decision-making process. Jacobs and Cherbonneau (2017) articulated this phenomenon by arguing that because “committed offenders do not respond to the socio-moral forces that inhibit most everyone else from crime, sanction threats may be the only force that influences them” (p. 192, original emphasis). However, unlike more traditional offenders, highly motivated offenders such as extremists are less likely to be absolutely deterred by such threats, a process Gibbs (1975) and subsequently Jacobs (1996, 2010) defined as restrictive deterrence—where perpetrators engage in crime in the face of sanction threats by engaging in adaptive behaviors aimed at mitigating risk.
Restrictive deterrence is defined by four behavioral responses to perceived sanction threats: (1) the offender reduces the number of crimes she commits over a period of time; (2) the offender commits less serious crimes than originally planned; (3) the offender engages in situational measures to decrease the probability that the committed crime will result in detection; and (4) the offender commits the same crime in a different place or time (also referred to as displacement). The current study is concerned with the latter two dimensions in which the core assumption is that would-be offenders must first recognize and be sensitive to the inherent risks (e.g., the presence of security interventions) of a particular criminal opportunity. In turn, extremists are more likely to engage in precautionary measures leading up to and during an attack (e.g., discretionary target and/or weapon selection, attempting to “blend in”, changing or altering tactics, and committing the act in a different space or time) (see, e.g., Cherbonneau and Jacobs 2019; Jacobs and Cherbonneau 2014; Topalli, Jacques, and Wright 2015).
Scholars of extremism have highlighted the use of risk management strategies, including choosing inconspicuous members to conduct certain operations, wearing disguises, selecting appropriate weapons, and, as operations become increasingly dangerous or draw unwanted attention, attack less frequently (Clarke and Newman 2006; Freilich et al. 2019; Gill et al. 2020). In what follows, we draw on the accounts of successful attacks carried out by various environmental extremist group members to assess the extent to which situational decision-making processes guide the strategies and tactics used to evade capture during an attack. We discuss our findings in the context of RCT and SCP, and the efficacy of the policy interventions motivated by these perspectives.
Methods
Data for this study come from communiqués written by REM extremists claiming responsibility for arson against individuals, corporations, or institutions perceived as contributing to the suffering of animals or the destruction of the environment. 2 More specifically, communiqués refer to texts written by anonymous individuals employing illegal action for ideological purposes (Liddick 2013; Loadenthal 2017a; Logan and Hall 2019). Communiqués serve as a means of “credit-claiming” and provide detailed information about an attack, including motivational factors, details about the operation, and a description of damage caused. It is vital for radical environmentalists to share this information via communiqués, or the incident may go undetected thereby undermining the intended goals of the movement (Brown 2020). Loadenthal (2017a) best summarizes the significance of communiqués by arguing that “a strike without explanation is akin to telling a joke without a punch line” (p. 260). Since destruction perpetrated by REM extremists is performative, in the sense that it serves as a symbolic message to their target audience, communiqués ensure the audience understands the event through its intended frame.
Communiqués were collected on April 17, 2017, from the “Diary of Actions” page in Bite Back Magazine (directaction.info), an online publication that serves as an outlet for the radical elements of REM (Liddick 2013). Bite Back is an online repository of REM activity and publishes communiqués, general news (e.g., prisoner releases), and media reports to unclaimed attacks. Although other websites exist, we selected Bite Back since it contains the largest collection of self-reported REM crimes (Loadenthal 2017b). Documents submitted to the website, including communiqués, are originally submitted to a website administrator for publication approval; however, the inclusion criteria are unclear. Despite this, several communiqués were validated by independent press accounts and included supplemental evidence (e.g., photos, videos) suggesting that the communiqués are authentic (Liddick 2013; Robinson et al. 2019). Furthermore, there is little reason to doubt the validity of communiqués since Bite Back is sustained by the most radical elements of the REM and generally unknown to outsiders.
Data were collected using an automated webpage collection program to harvest webpages linked to the seed URL (see Derrick et al. 2016). The automated program collected each documented linked to the Diary of Actions page on direct action.info and stored it in an Excel file. In total, 2,876 relevant documents, including communiqués, media reports, and general REM news posted on the website were collected and stored with each entry containing a unique identifier, the title (as seen on the website), the date posted, and the country of origin. Next, a team of three graduate student coders were instructed to read through each document and extricate any communiqués referencing arson. This included both attempted and successful acts of fire setting. A total of 275 communiqués described arson and served as the final sample for this study.
The reliability of the data was ensured via the use of coder training, consistency checks, and tests of inter-rater reliability. 3 For example, each rater underwent 40 hours of training on political violence and open-source data collection prior to coding. Additionally, when asked to code a subsample of 20 communiqués for a wider range of 36 different crime types, including arson, a Cohen's Kappa of .67 illustrates that reliability issues were minimal, and the raters were adequately consistent in their assessment of REM crime types (Logan and Hall 2019). 4 Finally, we should note that most arson communiqués were already classified as an arson incident by the website administrator of Bite Back, which we then used to safeguard the consistency of our coding.
Data Analysis and Coding Procedure
The current study relies on a modified version of grounded theory to identify patterns, concepts, and theoretical explanations regarding extremist decision-making, target selection, risk perceptions, and sanction avoidance strategies. While modified grounded theory is an inductive approach where theory emerges from the raw data, it also allows researchers to rely on existing literatures and frameworks to guide the research and help interpret the findings (Charmaz 2014; Glaser and Strauss 1967; Miles and Huberman 1994). Utilizing a modified grounded theory approach, the communiqué data were coded in three stages by the authors of this study (see Simi et al. 2017). 5
In the first stage, the data were analyzed line-by-line and each communiqué received substantive or initial codes (Charmaz 2014; Glaser and Strauss 1967; Miles and Huberman 1994). Inductive codes emerged from the line-by-line analysis and deductive codes were extracted based on the REM, criminal decision-making, and related literatures (Berg 2006). Consider the following example: “On wednesday, 06/03/13, we paid a visit to yet another highway expansion site. We looked for a decent target and, upon finding one, put 6 litres of gasoline wired to a timer inside driver's cockpit.” The segments of text where extremists discuss the type of weapon employed (i.e., gasoline wired to a timer) received an inductive code termed, weapons selection. Inductive codes related to target type, the presence of formal guardians, the timing of the offense, number of offenders, and target familiarity also emerged during the first stage of coding. The goal of the initial stage of coding was to “cast a wide net” while paying attention to similarities within and across communiqués.
In the second stage, coding was more focused, selective, and conceptual (Windisch et al., 2020). Specifically, our analysis emphasized substantive codes that were identified from the line-by-line coding to understand their linkage to the overall data. By substantive codes, we refer to codes that emerged across communiqués that relate to existing research on REM and criminal decision-making. At the same time, substantive codes were also labeled and organized into categories. A category is a higher order abstraction that links related codes in the text (Vaismoradi et al. 2016). For example: “On the night of December 25 an incendiary device was placed in a meat market belonging to the murderous Friosa company in Santiago. The device consisted of a can with gasoline, attached to sponges, activated by a simple chemical timer.” This example highlights a segment of text that initially received a weapons selection code and was later categorized under the category termed, improvised incendiary devices (IIDs).
In the final stage, we arranged these categories into themes. A theme is more implicit and abstract relative to a category and often requires interpretation based on the underlying meanings discovered at the interpretative level and elements of the subjective understanding of the data (Ryan and Bernard 2003). In developing each theme, we compared categories to one another as well as to existing literatures and frameworks to help interpret the findings. For example, sanction avoidance strategies, emerged as a theme in this study after comparing segments of data where REM extremists discussed managing the risk of detection via planning and the use of IIDs.
Sample Characteristics
The final sample of arson communiqués included an average of 177 words (Md = 128 words; SD = 165 words). The shortest communiqué included eight words and the longest had 964 words. As shown on Table A1 of the Online Appendix, the arson communiqués were written by self-identified REM extremists across 26 different countries. Most arson communiqués were written by extremists located in Mexico (22 percent), Italy, (11 percent), Russia (10 percent), Sweden (10 percent), the United Kingdom (8 percent), the United States (7 percent), France (6 percent), and Germany (5 percent). The geographical diversity is a unique strength of these data and departs from the overwhelming focus on criminal decision-making among offenders in the United States.
The arson communiqués were written between 2002 and 2016–although most communiqués (66 percent) were penned between 2008 and 2012. The frequency in which the communiqués in the current study were published by year is consistent with other similar analyses of REM arson incidents (e.g., Chermak et al. 2013). On average, there was around a 13-day lag (SD = 53) between the date of the arson incident and when the communiqué was published online. 6 While some communiqués were published the same day, the longest lag time was 719 days. Finally, most arson communiqués were written by self-identified members of prominent REM-affiliated groups such as the Animal Liberation Front and Earth Liberation Front. 7 Slightly over half of the arson communiqués (55 percent) were written by self-identified members of the Animal Liberation Front, while members of the Earth Liberation Front accounted for 19 percent (see Table A2 of the Online Appendix for full list group names identified in the communiqué data).
Results
Findings from this study highlight two broad conceptual themes related to the situational or momentary decision processes occurring before and during an act of extremism. First, we examine situational risk appraisal of extremists including sensitivity to objective environmental factors that define the level of sanction certainty. Second, we observe several sanction avoidance strategies highlighting long-term, adaptive processes leading up to attacks and momentary, situational adjustments during the commission of arson; both of which are consistent with themes from SCP and restrictive deterrence literatures. It is important to note that the excerpts from the communiqués are not meant to provide a formal test but rather to illustrate conceptual categories that emerged from our analysis (Simi et al. 2016). The prevalence for each theme is presented in Table A3 of the Online Appendix.
Situational Risk Appraisal
The first step of the analysis was to assess the efficacy of situational decision-making theory for understanding REM attacks. Overall, the recognition of and sensitivity to legal and extralegal risk factors was a major theme in the data. Consider the following passages. “It was 3 oclock in the night, but the workers were awake. They [the workers] were going back-and-forth around the encampment, one quarry excavator was raping nature while guards were watching TVs on their posts drinking beer. In our experience the camp was pretty large: 6–7 living sheds, several sheds for tools, 2 passenger cars, 3 excavators and 1 grader. Having sneaked between sheds and vehicles, we selected two excavators that looked as the most recent addition to local park (i.e., their loss will be most expensive). After that we followed routine procedure: put some rags around engine parts and oil pumps, soaked them with gasoline, etc. After we left the area, we tarried for some time to enjoy the night view. Both excavators were trailing huge columns of smoke into the air.” (Earth Liberation Front/Informal Anarchist Federation/Informal Revolutionary Front, 04/09/2014).
“The night of Tuesday 14th to Wednesday 15th of January of this new year (2003) at 3:00, our ALF-ELF cell attacked with incendiary devices a mink farm in Lubia, a small town of soria (Spain). The farm was composed by two different areas separated with a distance of about 250meters. Each part was provided with his own security mesasures [sic]. Both of them had a wire fence of about 2 meters height with a barb wire on the top, and a wall of concrete bricks of about 1,5 m height in the interior part, separated with the fence with a distance of another 1,5m aprox. The farm house was in front of one of the two areas, but between the area and the house was a narrow way, and it was built in the middle of a garden with two mastines del pirineo (big dogs)…We decided to go to the area which didn’t had dogs because we didnt wanted to drug them, even though the risk was bigger because the house was closer. We parked the vehicle in a copse [sic] two kilometers away the farm because we didn’t wanted to be detected by a Guardia Civil patrol (spanish rural police)…” (Animal Liberation Front/Earth Liberation Front, 03/28/2003).
These examples highlight the sheer number of situational factors that inform extremists about the suitability of targets, the attendant sanction threats, and, importantly, the ensuing strategies to mitigate these risks (Apel 2013; Nagin et al. 2015). In this way, labeling extremists’ perceptions as “situationally dependent” undersells their situational awareness as they simultaneously evaluate a myriad of micro-situational and contextual cues within a single opportunity—e.g., the passages allude to key factors such as time of day, level of guardianship, structural layout, and proximity of potential targets.
Moreover, the extremists in our sample demonstrated the ability to rank order these environmental cues by potential risk consistent with a model of target selection put forth by Nagin et al. (2015; see also Barnum et al. 2021). That is, extremists identified specific “high-risk” areas to avoid while in proximity of the target they intended to arson. The following two passages show how the actors identify multiple targets and discuss them in terms of risk (e.g., the ease of destroying a truck versus breaking in the slaughterhouse in the first passage). “On the night of 11/24/2012 we cut fences at the slaughterhouse in San Giovanni Valdarno (Florence). After causing more damage we burned a refrigerated truck used to transport meat, since it was impossible to get inside the slaughterhouse because of security cameras and alarms.” (Animal Liberation Front, 11/28/2012)
“we arrived at the property in the late hours of the night. With some bolt cutters we cut the metal fencing that had protected the farm with much care so that they would not hear us and so that the dogs who were guarding the place would not chase us. We opened the fencing from top to bottom and entered. Several attempts were made to open the barred window in the shed where chinchillas were, but these were in vain. The shed was medium-sized and was very close to the house belonging to those guarding this business of torture and domestication, so we decided to focus on the empty shed… When we were far away from the target, the fire could be clearly seen as it burned” (Animal Liberation Front, 07/14/2010)
Assessments about by-stander confrontation and other elements of the social environment were also paramount concerns informing the perceived distribution of risk within an offending opportunity. For example, “On July 9th the ALF of Mexico carried out an attack on an old enemy who we had visited before, a KFC in Mexico State… In the late hours of the night after watching the manager in one of the windows, we went to the target, spraying gasoline inside a small warehouse where the multinational stored various logistical equipment and we set it on fire. Because of the gas released from the gasoline inside the warehouse it made a noise like an explosion, and immediately we left the area. Police and firefighters did not delay and they began the search for those responsible, but had no success.” (Animal Liberation Front, 07/12/2009)
These examples highlight the complex decision-making extremists engage in to assess the attendant levels of risk when selecting suitable targets at the slaughterhouses, farms, or construction sites they intend to victimize. Interestingly, REM extremists in our sample emphasized a temporal ordering of assessing and attending to risk before “reaping” the benefits of their actions, often demonstrating self-control and patience by waiting to “enjoy the night view” of their fires from a safe distance. This is somewhat inconsistent with research showing the tendency of nonideological offenders to overweight potential immediate benefits of their actions while discounting consequences (e.g., Nagin and Pogarsky 2004), suggesting possible differences in intertemporal decision-making between extremists and other offenders.
It is important to note that these examples also allude to criminal expertise. The REM arsonists relied heavily on experience to identify attractive targets. Their ability to identify relevant situational characteristics that may (otherwise) interfere with the execution of an offense appears to be key mechanisms for successful attacks. We expand on this below.
Sanction Avoidance Strategies
The findings below reveal two distinct but interrelated decision-making processes consistent with notions of displacement and restrictive deterrence used by REM extremists to evade detection, which are informed by the processes highlighted in the aforenoted section. These include long-term adaptation and more proximate, situational processes in the form of momentary adjustments.
Adaptation: Target and Weapons Selection
Extremists in our sample engaged in at least two long-term, preparatory decision-making processes to manage the risk of detection (e.g., Cothren et al., 2008; Smith et al., 2006). First, REM extremists described the process of target selection through intensive surveillance. This process highlights the acquisition of detailed information over time that is used to select a target and determine its vulnerability and inherent level of legal and extralegal risk and is consistent with extant extremism research on geographical and temporal patterns of preparatory conduct (see also Gruenewald et al., 2019; Smith et al., 2006). “On Friday 23rd September 2005, a group calling themselves the Oxford Arson Squad, acting on information, located an unoccupied house owned by Christchurch College, broke in and planted a number of large incendiary devices. The message should be very clear to Oxford University: 1. We have spent 15 months researching you, and we know every weakness you have. 2. You have people within your ranks acting against you. 3. You cannot build the South Parks Lab without incurring massive losses. 4. We are stronger than you, we have more resolve than you and we never give up!” (Animal Liberation Front, 09/25/2005).
“It was a silent night in the early hours of September 20th, 2011, under a fine rain at the Santa Catarina Federal University (UFSC), Florianopolis, Brazil. The ALF broke into the campus after cutting open the fence that separates it from the avenue….
These examples highlight the lengthy surveillance process, which often involved multiple types of information gathering efforts. For instance, the extremists of the Animal Liberation Front in the second passage expressed a deep understanding of the surroundings because the attack was not their “first visit as this action has been planned during several months” (emphasis added). Moreover, the extremists in the first passage engaged in 15 months of physical reconnaissance of their target. Unsurprisingly, target surveillance increased in intensity and duration for high value targets. Reconnaissance becomes an essential preparatory step when the destruction of property has potential to cause substantial financial loss for the businesses involved. In such instances, the ability to adapt plans according to environmental interventions is crucial as high value targets undoubtedly have an increased presence of enhanced security features (e.g., Ekblom and Tilley 2000; Gruenewald et al. 2015).
Although physical reconnaissance was a common form of target surveillance identified in our sample, certain locations, such as fur farms and animal processing facilities, are regularly located in geographically remote areas not accessible to the public. Here, REM extremists are forced to adapt alternative surveillance tactics to supplement physical reconnaissance. For example, “Late at night on July 5 of this year we attacked a chinchilla farm… For this action a lot of time and effort was dedicated in surveying the area; pictures and videos were taken, information and maps were compiled from the internet, so that we'd have more information about the area and the possible escape routes. With this preparation completed we decided to attack and make our hardest strike for animal liberation… … Two days after this action, after the workers and the owners saw our graffiti and threats, after they had seen the charred empty shed and that the fencing was cut, we sent a threat by e-mail to the owners of the farm. They must stop operating their business of animal exploitation.” (Animal Liberation Front, 07/14/2010)
This shows how the use of the Internet can be used as a complementary source to compile “information and maps.” The Internet provides vital information about targets when it was difficult to collect this information serendipitously or during routine activities. As illustrated in the previous example, this form of “online learning” was especially important for identifying escape and get-away routes prior to the attack to increases the chances of evading detection.
Beyond direct information gathering strategies, target surveillance also included clandestine efforts referred to as “normalcy illusions” where offenders attempted to act normal and blend in with the environment (e.g., wearing disguises) to avoid detection and gain information regarding the target (Cherbonneau and Copes 2006; Jacobs and Cherbonneau 2014). Consider the following passage, “After more than one year of study and planning and a few weeks surviving in the harsh reality of a slaughterhouse that kills 500,000 birds a day, we reached our goal… You can invent a cause and argue that the fire was not an act of sabotage, but what matters is the result and nothing more…You can increase your security, but you will never be sure that you're safe. We have made sure that the doors will always be open for those who have a good reason to enter…You can not hide and you can not stop us… During the weeks we had the privilege of being inside this company (you can look, you will not find us), we had the opportunity to document some facts with photos, videos and photocopies, which will be released to selected institutions at the appropriate time.” (Animal Liberation Front, 04/01/2009)
The use of normalcy illusions highlights self-control and prudence on behalf of extremists, in this example, committing more than a year to observing a potential target within the “harsh reality of a slaughterhouse.” Importantly, prior research has focused on the use of normalcy illusion tactics during or just after the commission of a crime (e.g., Cherbonneau and Copes 2006; Jacobs and Cherbonneau 2014). Our findings add that normalcy illusion tactics are useful during the preparatory stage as well (see also Gill et al. 2020).
The second adaptive theme identified in the communiqués, which directly builds on target surveillance, involved weapons selection (Clarke and Newman 2006). In the context of arson, a would-be offender needs at minimum some type of ignition source and flammable or combustible substance to successfully take advantage of a criminal opportunity (Ekblom and Tilley 2000). Such a rudimentary approach, however, increases the offender's risk of detection by bringing unwanted attention due to the ignition process and structural fire. Alternatively, many extremists in our sample used IIDs, which are fire-starting weapons that typically consist of some sort of timing mechanism, ignition source, and flammable or combustible fuel in a concealed container (Raynis 2006). The use of IIDs is exemplar of adaptive processes used by extremists to ensure their success of victimization as the development and implementation of time-delayed IIDs require procedural expertise and technical skills acquired through practice, testing, and experience over time.
Importantly, extremists in our sample emphasized the utility of IIDs to distance themselves from the target before a fire ignites—providing a protective veil without lessening the intended force behind their political message. For example, consider the following passages: “Late yesterday, December 8, we armed our incendiary devices; we were conspiring to attack the machines that devastate ecosystems and create civilization. The target: a group of planet-destroying machines…a large floodlight illuminated the machines so that no one could get near them. But in spite of it we did. At night we jumped the fence that divided the federal highway from the pedestrian walkway. We slipped underneath the machines to avoid being seen and
“On the night of Saturday Oct 28, a new cell of volunteers visited Deans Farm near Witney where tens of thousands of hens are kept in battery cages. For the first time in UK
These examples highlight how IIDs allow attackers to enter the grounds of a target and swiftly exit undetected, effectively bypassing the active interventions (e.g., floodlights) that otherwise deter cumbersome attacks (e.g., manually setting a fire). These examples also offer a nuanced distinction between characteristics of extremism and many nonideological crimes. Whereas many instrumental crimes require “co-presence” with victims and targets to enact (e.g., auto theft), the arsonists in our sample prefer to remove themselves from the situation before the act. Adapting the use of IIDs appears to be an important tool for achieving this goal. Consider further, “So that night, camouflaged by the darkness and with only the moon lighting our insurgent steps, we arrived at the property in the late hours of the night. With some bolt cutters we cut the metal fencing that had protected the farm with much care so that they would not hear us and so that the dogs who were guarding the place would not chase us…we placed an incendiary device made of about four liters of gasoline at the side of the empty shed. …We set the timer on the device and left through the torn wire mesh. When we were far away from the target, the fire could be clearly seen as it burned. The abolitionist's flame was present once again on a tranquil night shaken by the Frente de Liberacin Animal in Mexico.” (Animal Liberation Front, 07/14/2010)
Not only did the use of IIDs provide attackers in our sample with a sense of security to carry out attacks from a safe distance using a timer. Importantly, the use of timed IIDs appeared increase the chances the intended message about their cause was delivered to the “enemy.” For example, “During the evening of Saturday, October 31, we, from the cell against animal domination (F.L.A-C.C.D.A), armed with an incendiary device and a fake bomb, went to a rodeo which was set to happen in the contreras borough…Upon reaching the location, we noticed that this action would be relatively easy since there was no light or security, so we moved into action right in front of the noses of the damned anthropocentric people who were there. An incendiary device was placed in a tire of a truck belonging to civil protection and a message was left in it warning that more devices had been left elsewhere on the property” (Animal Liberation Front, 11/03/2009)
The adaptive strategies highlighted in this section underscore the long-term decision-making processes REM extremists rely on in the days, months, even years leading up to an attack. Next, we outline complementary situational processes in which attackers adjust to (unexpected) situational cues on the fly during the commission of their crime.
Situational Adjustment Due to Unexpected Contingencies
The adaptive processes highlighted the preparatory processes before an attack is carried out. However, extremists often face surprises while carrying out an attack forcing them to adjust in the moment. The REM extremists in our sample consistently identified the unexpected presence of human guardianship as key factor upending their plans. For example, “In the night of June 8, we, the Frente de Liberacin de la Tierra, along with some anarcho-Insurrectional individuals who are committed to constant conflict with the state and its institutions, decided to carry out an action together in the city of Ecatepec in Mexico State….When we arrived there,
In the early morning hours of May 7, a cell from the Frente de Liberacin Animal was planning to hit two targets in the commercial area of the City of Coacalco in Mexico State, but
These examples highlight an asymmetrical sensitivity to unexpected human guardianship compared to other inanimate threats such as security systems. That is, the extremists reassess risk and adjust accordingly when faced with the unexpected human guardianship whereas they less influenced by other interventions.
Interestingly, and in line with arguments by Nagin et al. (2015), attackers integrated relevant information acquired through adaptive preparatory procedures to help inform situational, “on the fly” adjustments. For example, “During the night of October 10, we left an incendiary device on the roof of a small poultry farm…
Here, knowing where to hide “to avoid being seen by the people who were passing by and the shitty police” when faced with unexpected contingencies enabled extremists to carry out their attack when the “coast is clear.” This process combines the latter two elements of restrictive deterrence: situational adjustments and temporal displacement. In general, the ability to utilize surrounding environmental features of a target proved to be an important adaptive element when the extremists in our sample faced “surprises,” a finding that overlaps with the concept of “tactical displacement” with an emphasis on perceived sanction threats as a potential mechanism. For instance, the following example illustrates how self-identified members of the Animal Liberation Front took advantage of their immediate environment to avoid apprehension: We went to an ostrich farm in Zumpango (Mexico State), where the animal's skin is cut off to be sold. But we did not arrive with empty hands; we brought a large incendiary device that we planned to place in one of the trucks.
And, when faced with unexpected human agents during the commission of their attack, the ability to adjust and carry out an attack quicker than intended proved essential when victimizing high-profile targets. Last Thursday in the early morning hours we attacked a McDonald's located at the intersection of Pajaritos and Las Torres Avenues, in the Maip area, with Molotov cocktails. We threw about 10 bottles filled with sulfuric acid and gasoline, along with a self-igniting phosphorus mixture.
Another consistent theme in the communiqués related to situational adjustment pertains to group offending. Specifically, extremists often applied a division of labor by assigning group members specific roles focusing on a specific element of an attack to quickly transmit key information about the evolving circumstances of the offending opportunity. For example, “One night,
Whereas “actors” contribute directly to setting the fire, “lookouts” stand guard and signal information to the actors (Uhnoo 2016). The lookouts are essential for the adjustment processes required to avoid unexpected risk during an offense because, compared to solo offending, the information distributed by lookouts and similar roles (e.g., scout) during an offending window allow the “fire setters” to focus on their task and make quick adjustments when necessary. “In the midnight of Jan 27th, a group of ELF activists set out to the construction yard, situated in the forest in the southern suburbs of Moscow. Despite the site being brightly lit by numerous light masts, we managed to crawl right up to the vehicles (Caterpillar bulldozer and another unidentified one) with 10liters of gasoline and some rags. Guards were either drunk or asleep in their special sentry car, so up in flames the dozers went.
The ability to momentarily evaluate risk and adjust to unexpected contingencies appears to play a key role in whether an attack is successfully carried out. This remains a consistent theme in active offender research (e.g., Jacobs and Cherbonneau 2014), in which highly motivated offenders attempt to manipulate, mitigate, and even outright reject sanction threats and offend despite the inherent risks associated with their actions.
Discussion and Conclusion
Research on violent extremism has provided promising evidence for the efficacy of RCT and SCP-inspired interventions. To date, however, much of this research has been evaluative in nature meaning the theoretical mechanism underlying these processes have largely been left in the background. The goal of the current study was to explore potential mechanisms leading extremists to either abstain or carry out an attack in high-risk situations. Drawing on recent developments in situational decisions making, including restrictive deterrence, our analysis of arson attacks carried out by REM extremists provides a potential backdrop from which to interpret some recent works on SCP and extremism (Freilich et al. 2019). That is, we add to the growing evidence that extremists are sensitive to sanction threats, and from the point-of-view of extremists themselves demonstrate that rather than produce an absolute deterrent effect, these perceived threats appear to inform how they define and ultimately act within and around a criminal opportunity.
Theoretical and Policy Implications
A key finding in the current study, which we believe compliments recent insights from RCT and SCP scholarship on extremism and terrorism, is that perceptions of sanction threats—both in terms of recognition and sensitivity—appear to be foundational to situational decision-making and the strategic behaviors of politically motivated arsonists. That is, our findings not only provide support for prior tests of SCP but shed further light on at least one key mechanism underlying the use of strategic behaviors used to mitigate perceived risk. Specifically, the arsonists in our sample showcased the ability to simultaneously recognize and assess several features of a criminal opportunity that define the attendant level of risk. In turn, they displayed an essential willingness to momentarily adjust their assessments as the situation evolved and sanction threats changed during an attack—a process foundational to recent models of situational decision-making in criminology (e.g., Barnum et al. 2021; Pogarsky et al. 2017; Nagin et al. 2015; see also Paternoster 1989). In this way, the ability to recognize and be sensitive to sanction threats served as a catalyst for ensuing strategic behaviors employed by extremists to navigate SCP interventions and transgress in the face of (sometimes) high levels of risk (Gibbs 1975; Jacobs 2010).
Building on themes related to situational risk appraisal, findings highlighted in the communiqués revealed two forms of strategic adaptation and adjustment for mitigating risk. A common theme among our sample of extremists was the use of advanced target selection strategies, including physical reconnaissance, online learning, and various clandestine efforts such as illusions of normalcy. These strategies were essential for selecting vulnerable yet high-profile targets, identifying environmental interventions designed to reduce the risk of victimization, choosing effective escape routes, and informing key contingency plans to increase the probability of a successful attack (see also Clarke and Newman 2006:120 for a discussion on “tools” of terrorism).
Importantly, the current findings, coupled with the findings from prior works in this area, provide a theoretical link between ideological offenders and other highly motivated and experienced offenders. Studies examining active offenders involved in various instrumental crimes such as drug dealing, auto theft, and burglary consistently highlight the use of preparatory behaviors to evaluate “target-imposed costs” and the risk of detection during the target selection process (Deitzer et al., 2021; Jacobs and Cherbonneau 2014, 2016; Logie, Wright, and Decker 1992; Nee and Meenaghan 2006; Nee and Taylor 2000; Tedeschi and Felson 1994; Topalli 2005). Indeed, prior research on pre-incident preparatory behaviors of extremists and terrorists mirrors these processes. For example, Smith et al. (2006) highlight the temporal and geographical patterns of preparatory conduct leading up to an attack. These start with recruitment into extremist groups, continue through preliminary organization and planning, and shift closer in time and geography to the attack through the direct preparatory behaviors highlighted above (see also Cothren et al., 2008). Given the consistency of this body of research with the current findings, preparatory stages may serve as a key point of intervention for reducing extremist attacks. 8 That is, a deeper understanding of the preoperational activities extremists engage in may help law enforcement agencies recognize suspicious behaviors prior to attacks (see, e.g., Clemmow et al., 2020; Gruenewald et al., 2019; Mazerolle et al., 2020).
The current findings on weapon selection, an additional long-term adaptive strategy, also have strong ties to the “pillars of terrorism” central to Clarke and Newman's (2006) extension of SCP and extremism (i.e., their “MURDEROUS” acronym summarization; see also Freilich et al. 2019). Specifically, our findings map onto several dimensions, including extremists’ tendency to rely on weapons that can be employed in a variety of situations, are undetectable, easily handled, destructive, enjoyable, reliable, and safe (i.e., removes the actor from harm's way). In particular, the extremists in our sample highlighted the reliability and safety of IIDs as their primary weapon. The current findings add additional nuance to Clarke and Newman's theoretical extension. Specifically, and in line with the situational appraisal findings, the importance of timing mechanisms for IIDs, including the expertise and training that goes into implementation, ensure safe distancing from the target before detonation so that they can “enjoy” the benefits of their attack only after they are at a safe distance.
Interestingly, this finding reveals unique intertemporal decision-making processes in which extremists place extra weight on the immediate risks associated with arson therein tactfully discounting future benefits. Theoretical and practical evaluations of extremist decision-making, including the efficacy of SCP interventions for reducing attacks would greatly benefit from a closer examination of the intertemporal tradeoff between perceived risks and rewards among extremists. This may be especially important as many SCP policies were designed to reduce broader traditional crime types, which are often characterized by myopic and impulsive actors who discount future consequences in favor of immediate rewards (Gottfredson and Hirschi, 1990; Nagin and Pogarsky, 2001, 2004).
In addition to adaptive preparatory decision-making prior to an attack, our findings highlight extremists’ tendency to (at least) momentarily adjust to situational sanction threats and carry out the planned attack in the wake of new, evolving, or unexpected risk factors. The extremists in our sample demonstrated the ability to adjust to unexpected characteristics of a target, namely the presence of human guardianship, through attentive situational evaluation and key strategies like a division of labor (see Shover 1973). Importantly, these situational adjustments, which have recently been explicated in crime decision-making studies using various experimental methods (e.g., Barnum et al. 2021; van Gelder et al., 2019; van Sintemaartensdijk et al., 2021), appear to be informed by the preparatory strategies and provide support for restrictive deterrence as an additional lens to understand extremist behavior.
From a theoretical standpoint, the situational adjustments highlighted in the current study, whether informed by preparatory strategies prior to an attack or in-the-moment information transmission strategies such as using a lookout and dividing labor, expand on the discussion around displacement in SCP research. On the one hand, there is substantial overlap between the situational adjustment identified here and the different versions of displacement (see Johnson et al. 2014 for review). The tendency to make behavioral alterations during an attack by quickly changing targets, altering the timing of an attack, or even changing the crime entirely have long been documented by SCP scholars. The current accounts by extremists add to the discussion by emphasizing the role of sanction risk perceptions as a key mechanism, which appears to motivate these behavioral modifications, a finding consistent with Jacobs’s (2010) extension of restrictive deterrence. This finding dovetails with the concept of tactical displacement.
To SCP scholars, tactical displacement occurs when extremists use methods or tools not previously planned or employed—e.g., entering the back of the property instead of the front due to an unexpected presence of a bodyguard. Our findings suggest that situational adjustments, while certainty including this definition, may have an additionally refined level as many of these “adjustments” were informed by preparatory conduct prior to the attack, or through momentary reconnaissance via updated information from co-offenders. In this way, extremists constantly reevaluate and update their perceptions of the inherent sanction threats leading up to and during an attack and then, if necessary, make tactical adjustments. So, while we concede there is overlap between the concept of situational adjustment and tactical displacement, our emphasis here is on the cognitive process of recognizing and adjusting to new and evolving threats such that some version of the original plan is enacted, and the attack is carried out. Perhaps a line of research that could yield novel insights for decision-making theory and further inform SCP policy would be to compare processes where attacks carried out by extremists include detailed contingency plans with those where attacks relied more on in-the-moment (even impulsive) alterations to the plan.
From a policy standpoint, the acute awareness of risk and the reliance on avoidance strategies by extremists necessitate the use of more clever approaches to dissuade terrorism and ensure displacement does not occur (Gill et al. 2020). Combining insights from our study with findings from prior RCT and SCP studies, prevention strategies might be most successful when they are tailored towards specific offending opportunities as opposed to broader “blanket” level policy. Indeed, REM arsonists consistently pointed to preparatory strategies and their experience and expertise as risk-mitigating factors—i.e., less brazen, and less experienced actors tend to perceive greater risk for offending (see, e.g., Loughran et al. 2009; Nee et al. 2019). Consequently, traditional security measures are less likely to sufficiently increase eco-terrorists’ perceptions of sanction risk to a level that prevents target victimization. In line with Clarke and Newman's (2006) assertion that we must “outsmart” extremists, more research is needed to unpack extremist decision-making processes to inform newer and more clever interventions that leverage sanction risk perceptions as a deterrent rather than an input for sanction avoidance strategy.
One potential intervention could be to leverage insights on ambiguity aversion and legal compliance (Nagin 1998; Sherman 1990). Here strategies that randomize security patrols and other surveillance tactics may disrupt extremist planning tactics and, importantly, introduce a level of ambiguity around their chances of success. This ambiguity can serve as a deterrent net of sanction risk as people tend to be ambiguity adverse (Pickett and Bushway 2015). Further, and central to the processes outlined here, ambiguity around probabilities on the lower end of the probability spectrum can increase the deterrent effect of ambiguity (Barnum and Nagin, 2021; Loughran et al. 2011; Midgette et al. 2021). Such approaches may make situations that are otherwise perceived to be defined as low risk appear riskier, ultimately reducing the vulnerability of a previously desirable target.
Limitations and Future Directions
Although the results of this study are promising, there are several areas for future works to expand on. Most notably concerns the validity and generalizability of the communiqué data. For example, it is difficult to ensure the author of each post is the same person or group who committed the crime. This is in large part because our analysis was limited to the information provided in each post without the ability to probe further with follow-up questions. 9 At the same time, is it possible that some communiqués do not include important details regarding the incident for fear of future apprehension since the website is almost certainly known to law enforcement. It is also possible that the language and behaviors outlined in the communiqués may be performative or emphasize specific aspects of arson incidents. However, this issue reflects a broader concern with studying active offenders. For instance, Jacobs and Wright (2008) note that “the validity of interviews that seek information about illegal activities will always be open to question. It is possible that some interviewees lied or embellished their accounts to protect themselves or to impress us” (p. 515). Despite not being able to 100 percent guarantee the authenticity of each communiqué, several were at least partially validated by independent press accounts and supplemental evidence (Liddick 2013; Robinson et al. 2019) and, importantly, the rich diversity in terms of number REM groups identified and countries where attacks took place across the communiqués affords additional confidence in the current interpretations. Still, it is vital future works replicate our findings across alternative sources of data, and with the use of more direct methodologies to ensure the theoretical and policy implications.
It is also important to note the current sample is necessarily dependent on the self-proclaimed attackers’ decision to post about an attack in the first place. This is especially relevant for the study of extremism, a rare but serious behavior, because the accounts in the communiqués presented here primarily report successful attacks. In other words, findings from this study reflect the decision-making processes involved in successful arson incidents. Although we believe this “self-selection bias” does not necessarily take away from the primary goal of complimenting macro-level research with situational insights on decision-making, there are undoubtedly invaluable insights to be gained by comparing successful and unsuccessful attacks by ideological offenders, especially regarding the adaptive and adjustment tactics employed (or not) prior to and during an attack.
Finally, we would be remis to ignore the limited generalizability of our findings since the parameters of the total population are unknown (Jacobs, 1996; Wright and Decker 1994, 1997). At the same time, the communiqués examined in this study came from a singular source—Bite Back Magazine’s online repository of REM communiqués. While Bite Back Magazine contains one of the largest collections of self-reported REM crimes, it is not the only source of REM communiqués—ultimately limiting the scope of our findings. Nevertheless, future works could build on our findings, and the findings of other qualitative studies to inform broader data collection efforts that directly query the decision-making tendencies of extremist group members to further inform ideological crime prevention.
It is also important to consider the uniqueness of radical environmental extremism related to other types of extremist movements. For example, radical environmentalists deliberately avoid interpersonal violence in favor of property destruction, which is a stark contrast from other extremist types (Carson et al. 2012). Although similar risk management strategies have been found among extremists engaged in interpersonal violence (see Gill et al. 2020; Windisch et al., 2018), the decision-making processes found in this study reflect strategies used by extremist engaged in property destruction. Future research should continue to investigate the similarities and differences in how extremist evaluate and manage risk when engaged in interpersonal violence and property destruction.
REM members also tend to be different from other extremist group members in important ways, which may influence the extent current findings can be generalized. Individuals involved with the REM tend to be older, more highly educated, and involved in committed relationships relative to other extremist types (Chermak and Gruenewald, 2015; Logan and Ligon 2021). In turn, REM extremists represent a “uniquely rational and deterrable demographic” and therefore may be particularly sensitive to the consequences of punitive actions, further motivating situational analyses across a range of extremist group member (Carson et al. 2020:705; also see, Liddick 2013; Logan, Windisch, and Simi 2022).
Nevertheless, we remain confident that our analyses of communiqués by REM extremists, and the use of communiques more generally, offer unique opportunities for the study these often inaccessible offenders. In contrasting the limitations of communiqués with their unique advantages, we draw on Nagin et al.’s (2015) call for criminologists to carefully examine the mechanisms underlying environmental approaches to crime control (see also Barnum et al. 2021; Chalfin et al., 2022). Since Clarke and Newman's (2006) influential book, Outsmarting the Terrorists, researchers have taken an environmental approach to the study of terrorism and extremism, examining the extent to which various interventions deter (or displace) ideological crime. While this work has demonstrated the “success” of many interventions, the reasons for this success have been left in the background. Communiqués offer a novel source of information akin to ethnographic observations and interviews with active offenders (e.g., Jacobs 2010; Topalli, 2005; Wright and Decker 1997). Researchers can directly observe the factors that shape offender's decisions at the time of the offense and in turn shape their behaviors. So, while communiqués may be limited in generalizability, they provide complimentary insights to macro-level environmental studies on active political offenders by shedding light on key mechanisms. We believe communiqués and related methodologies can help bridge the gap between research on terrorism and extremism with traditional criminological research by offering new inputs on the decision processes and behaviors of this hard-to-reach population. Not only does this allow for nuanced theoretical comparisons between extremists and their nonideological criminal counterparts, but the insights gained from these data sources can help refine environmental approaches to crime and the subsequent interventions employed to reduce target vulnerability.
Conclusion
The current study sought to complement existing RCT and SCP research on extremism by more closely examining the situational decision-making processes of REM arsonists. Our findings add to the growing support of these perspectives for understanding extremism and terrorism and highlight the role of perceived sanction threats as an important mechanism. Given the consistency of our results with prior works, we challenge criminologists and policymakers alike to continue unpacking the situational decision-making processes of extremists to further advance theoretical explanations of and societal responses to these acts of political destruction and violence.
Supplemental Material
sj-docx-1-jrc-10.1177_00224278231152439 - Supplemental material for Radicalized Environmental Extremism and Situational Decision Making
Supplemental material, sj-docx-1-jrc-10.1177_00224278231152439 for Radicalized Environmental Extremism and Situational Decision Making by Timothy C. Barnum and Michael K. Logan in Journal of Research in Crime and Delinquency
Footnotes
Acknowledgments
The authors would like to thank Starr Solomon, Steven Windisch, and Greg Pogarsky for invaluable feedback on the manuscript. We would also like to thank Gina Ligon and Doug Derrick for their support during the data collection process. Finally, we appreciate the comments and feedback from all three anonymous reviewers.
Disclosure Statement
The authors declare that there is no conflict of interest.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Supplemental Material
Supplemental material for this article is available online.
Notes
Author Biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
