Eco-Terrorism? Countering Dominant Narratives of Securitisation: a Critical, Quantitative History of the Earth Liberation Front (1996-2009)


Countering Dominant Narratives of Securitisation: a Critical, Quantitative History of the Earth Liberation Front (1996-2009)

by Michael Loadenthal


The Earth Liberation Front (ELF) has carried out acts of political violence and ‘economic sabotage’ characterized by a pattern of behaviour reflecting tactical and targeting selections, communications strategies, and geographic location. The movement’s attacks typically focus on the destruction of property located in ‘soft targets’ associated with commercial and residential construction, the automotive industry, and a variety of local, national and multinational business interests. These sites are routinely targeted through a variety of means ranging from graffiti to sabotage to arson. Geographically, the movement has focused its attacks in the United States and Mexico, and, to a limited extent, countries on the European, South American and Australian continents. The findings presented in this article were developed through a statistical analysis of the movement’s attack history as presented through its above-ground support network. This is discussed in critical contrast to assertions about the movement’s alleged terrorist behaviour found in most academic and government literature. This study seeks to present an incident-based historical analysis of the ELF that is not situated within a logic of securitisation. In doing so, it challenges traditional scholarship based on statistical findings.

Keywords: Earth Liberation Front, eco-terrorism, single-issue terrorism, quantitative analysis


In the wake of the 9/11 attacks, the often-linked fields of Terrorism Studies and Security Studies have witnessed a boom, accompanied by the more general rise in university studies directed at Islam, political Islam, terrorism and Middle Eastern politics[1]. Subsequently, new approaches have been developed within a host of “critical” fields, including Critical Terrorism Studies[2] and Critical Security Studies[3]. These attempt to problematise and clarify a methodology for those seeking to investigate political violence and responses to it through a non-orthodox, non-realist lens in an attempt to move beyond a purely securitisation focus. These approaches, while offering a host of new points of concern and criticism, often continue to base their study on raw data produced by state institutions. Thus while such studies may critically examine findings, new scholarship is needed that draws its conclusions from the wealth of data offered by the social movements themselves.

Within this post-9/11 era of terrorism scholarship, a new class of “terrorism experts”[4] emerged, poised to corner the academic market, often at the service of law enforcement, state-centric think tanks and a wider statecraft of securitisation. The present study is not meant to serve as yet another quantitative tool for criminal and behavioural profiling. Instead it is meant to act as an example of a methodological break, wherein one surveys the difficult data offered by the practitioners of political violence or their supporters themselves. The analysis contained herein is not meant to be a tool for law enforcement but rather to serve as a counter balance to the statist narrative concerning tactical trends and their relation to the criminalisation of dissent. Overblown, inaccurate, and fear mongering depictions of bomb-throwing masked vigilantes occupies much of the discussion of “eco-terrorism”. In response, scholars have been careful to begin developing counter-narratives to discuss these movements within a more accurately nuanced language. Activist-aligned journalists[5] and academics[6] have also begun to offer critiques of the terrorist framing of these movements in an attempt to offer an alternative explanation to state rhetoric. Within state rhetoric, we see the ELF (and its sister entity the Animal Liberation Front) termed “eco-terrorists” since around 2002, when Federal Bureau of Investigations (FBI) Domestic Terrorism Chief James Jarboe invoked the label twelve times in a speech entitled “The Threat of Eco-Terrorism.” The same year, Dale Watson from the FBI’s counterterrorism and counterintelligence division, reported to the United States Senate that the ALF/ELF represented a serious terrorist threat, characterising them as the most active extremist elements in the US.

The ELF has been active in the US since 1996, and through the use of decentralised, self-contained, underground cells of activists, has managed to not only carryout scores of attacks on property, but remain relatively immune from arrest. Activists inspired and motivated by the politics of the ELF are free to carry out acts of property destruction and claim them via the ELF moniker, provided they meet the movement’s guidelines. According to widely circulated guidelines, ELF actions must economically harm the adversary, aim to educate the public, and avoid harming both human and (non-human) animal life. Therefore if an individual, or a small grouping of activists agree with these three simple points, they are encouraged to act independently and to claim their attack through the ELF moniker.  This has most often been done through written communiqués bearing the ELF name. In those rare instances when a communiqué is not issued, the letters E.L.F. have appeared in paint at the site of the incident. The various actors and cells that constitute the ELF should therefore we understood as an ideologically-aligned network of autonomous, decentralised nodes, who share a strategic and tactical vision. They are not a unified movement in the traditional sense, nor are they a membership-based organisation. They are a tactical, strategic and praxis-informed tendency supported by a similarly decentralised, ad hoc network. As a result, scholarship that insists on understanding such groupings as nothing more than radical splitters of traditional social movements (e.g. Earth First!) will continue to be inherently flawed.

In exploring these networks of clandestine eco-saboteurs and arsonists, one is often tempted to construct a definition of terrorism, and following that, present one’s case comparatively to that set of parameters. The goal then becomes to decide if the evidence presented qualifies the object for inclusion within the terrorist designation. Conversely, this study seeks to present evidence which can then be held up against a variety of definitions of terrorism that have in common a focus on deliberate attacks against unarmed human beings in order to intimidate, coerce or otherwise influence a larger audience. Therefore the ‘dominant narratives’ this study seeks to challenge are those that present the ELF not as a strategic social movement utilising targeted property destruction, but as a violent terroristic threat to the nation state. It is not the main intention of this study to refute the FBI’s classification of the ELF as domestic “eco-terrorists”, but rather to discuss how data is represented through a divergent lens, and subsequently used to embolden such claims for the purposes of securitisation. The intention of this study is to provide quantitative evidence to public, above-ground activists and scholars who seek to offer support in the creation of counter narratives; explanations of an emergent social movement not based in state-centric terrorism rhetoric.

The following study seeks to determine the tactical, targeting, messaging and associated behavioural characteristics of the Earth Liberation Front (ELF) through a data set drawn from the movement’s self-constructed mouthpiece. This analysis will draw from the movement’s 13 (1996-2009) year history of global attacks in order to answer the question: What does a typical ELF attack look like, and secondly, how often are atypical incidents claimed under the ELF moniker? In order to develop such a behavioural profile, a series of statistical findings will be reviewed. These findings are drawn from an analysis of the movement’s attack history as presented via their above-ground support structure, the North American Earth Liberation Front Press Office (NAELFPO)[7]. All data analysed was gleaned from pubic (i.e. non-classified) sources and as such, provides little to no utility for law enforcement as such entities compile their own incident databases from a host of clandestine (e.g. Law Enforcement Sensitive, Classified, etc.) sources. This attack chronology, documented by the NAELFPO, was used to develop a database of 707 events, each coded for 11 variables. Each attack was coded through a standardised decision tree based on the description provided by the NAELPO, as well as communications issued by the ELF cell directly. The data was then split into six datasets and analysed. These six distinct datasets were developed to account for the presence or absence of repeating events (e.g. one cell breaks the windows of four banks, claimed in one communiqué) and commonly occurring, distinct national locations. Throughout the discussion contained herein, the findings have been compared to studies presented in academic journals, as well as government reports, in an attempt to evaluate the ELF’s attack history in the light of assertions made about the movement’s behaviours.[8]

Methodology: Process & Limitations

The database utilised throughout this analysis was created from the “diary of actions” hosted on the NAELFPO’s website. According to the NAELFPO, “The actions contained on the pages below comprise a complete history of ELF actions in North America and globally.”[9] From the “diary of actions,” a 707-entry database was created, each entry representing one ELF-linked attack. These 707 entries were comprised of 211 distinct events, and 496 repeating attacks (e.g. a single cell vandalising multiple, distinct targets in one outing claimed through a single communiqué) occurring between 14 October 1996, and 23 November 2009. The attacks were carried out across 14 countries, including 28 US states while there were as well four attacks without a clearly discernable location. The data was coded manually, and used to create a database via the “SPSS Statistics 17.0” software suite. Each event was assigned distinct values within 11 variable fields.[10] Many of the attack descriptions and cell communiqués are exceedingly descriptive regarding the tactics utilised and target selection, though in some cases this descriptive richness was lacking. Occasionally, attacks were recorded in the NAELFPO diary with only a single descriptive sentence, making the process of coding for 11 variables difficult. The variable categories were developed with such limitations in mind, and thus, “the goal in developing the [coding] taxonomy was to build a set of classes broad enough to capture the range of terrorist behaviour, but still simple enough to use, given the limitations in the descriptive data available on each individual terrorist incident”[11].

In cases where elements of the description necessary for coding were absent, attempts were made to estimate a reasonable scenario, and to describe it through the most accurate terms available.[12] For example, if a description stated that a laboratory was “attacked,” “trashed,” or “monkey wrenched,” the attack was recorded as an act of “sabotage/vandalism/graffiti,” as the broad nature of this tactic category was developed to allow for the coding of such events; events where the exact nature of the damage and tactics are unclear. If the description stated that the target was “covered in slogans,” “paint bombed,” “tagged” or used similar language, the tactic was recorded as “graffiti,” despite the fact that such a term was not included in the communiqué text. Throughout the coding process, attention was paid to the stated motivation for attacking a target. For example, when a Wal-Mart or Nike shop was attacked and criticised for its global policies, it was recorded as an attack targeting a “multinational corporation,” whereas the office of a regional energy company was recorded as a “business property.” Similarly, two attacks, both targeting automobiles could be coded differently depending on the owner’s position vis-à-vis the larger ELF policy. The vandalism of a sports utility vehicle (SUV) in a dealership was recorded as targeting a “SUV/automobile,” whereas similar vandalism of a specific individual’s (e.g. CEO of targeted company, researcher engaged in controversial experimentation) car, targeted because it belonged to that individual, was recorded as the targeting of “personal property.” In coding the data, the aim was to use as little interpretation as possible, and to transparently decipher the language of the description and/or communiqué into the coding values through a standardised decision tree.

In order to accurately represent the scale of some attacks, some single events are recorded as multiple entries. For example, if four SUVs are firebombed, the events were recorded as four acts of arson because four targets were attacked[13] Conversely, the breaking of four windows of one office/SUV/home/etc., was counted as a single attack. However, if one window was broken on each of four separate offices, this was recorded as four attacks since multiple targets serve as the determining factor. Occasionally, the exact number of targets attacked was unclear. If the description stated that “numerous vehicles” or “a row of homes” was attacked, that incident was recorded as two entries despite the possibility that many more targets were attacked. Lastly, if an attack utilised two distinctly different tactics, the event was recorded as two incidents[14]. This was done when both tactics fell outside of the “sabotage/vandalism/graffiti” category, such as in the case of an “animal liberation” that also involved the arson of the building. In this example, the event would be recorded as one “distinct incident” and one “multiple entry”[15]. Because of the tendency for such a coding procedure to artificially inflate the appearance of some attacks, calculations were conducted separately within multiple data sets, one wherein multiple entries are included, and another wherein only distinct (non-recurring) attacks are included.  In this second, distinct incident data set, multiple entries were condensed to single attacks. For example, if saboteurs were to slash the tires of four vehicles and claim it in a single communiqué, this would be recorded as one tire-slashing incident in the distinct incident data set, and recorded as four tire-slashing incidents in the multiple entry data set.  For the purposes of analysis, the sample was split into three location-based categories. Each of the three datasets were then split into subsets (multiple entries and distinct incidents). For the purposes of reporting and discussion, these six datasets will be referred to throughout, abbreviated as DB1, DB2, DB3, DB4, DB5 and DB6 and are detailed in the endnotes.[16] All numerical findings were rounded to the nearest whole number when presented in the in-text data tables, and in doing so, some total to more than one hundred percent.

In rare cases, the NAELFPO’s data included attacks that were carried out by a known group that was not affiliated with the ELF. For example, between 10 October 2008 and 31 October 2008, four attacks were carried out in Canada targeting the EnCana Corporation[17]. These attacks were not claimed by an ELF cell despite the presence of a communiqué hosted by NAELFPO, and thus, these attacks were excluded from the sample. Occasionally a cell adopted the ELF name after the initiation of an attack campaign. For example, starting in June 2009, attacks in Mexico ceased to be claimed by the ELF, and were instead claimed by "Eco-arsonists for the liberation of the earth" (EpLT).[18] Thus attacks claimed by the EpLT were excluded from the study sample, and only those prior attacks signed with the ELF name were included. Around the same time, additional attacks in Mexico were being carried out by “Luddites Against the Domestication of Wild Nature,” (LADWN); these were similarly excluded as LADWN represented a distinctly new, non-ELF moniker.[19] However, on 20 July 2009, LADWN announced in a communiqué that it now "form[ed] part of a cell of the Frente de Liberación de la Tierra,”[20] thus from that date onwards, the group’s actions were recorded as attacks of the ELF[21]. Finally, when a group used the ELF name but specified a distinct unit within the movement, this attack was simply recorded as being carried out by the ELF. For example, on 5 March 2001, a graffiti attack was claimed by the “ELF Night Action Kids,”[22] and was recorded in the database as being carried out by the ELF. These methodological decisions were made to allow for focus on the deployment of the ELF moniker - not the wider constituency makeups. For example, despite the ideologically-shared proclivities of the ELF and the EpLT, since the latter chooses an explicitly non-ELF moniker to claim its actions, it is excluded despite tactical, strategic and ideological similarities.

In coding for the “communiqué” variable, the presence of a communiqué linked from the NAELFPO website was recorded as such, just as the lack of a communiqué present on the website was recorded as “no communiqué issued,” despite the possibility that a communiqué was available in another source.[23] In order for an attack to be marked as ELF-linked without the presence of a communiqué, it was required that the letters “E.L.F.” were left at the scene of the attack either through graffiti, a banner, note, or similar visual/written communication. For example, on 11 October 2004, a “package with the letters ‘ELF,’”[24] was left on a road in Philadelphia and treated as a possible improvised explosive device, thought the box turned out to be harmless. Thus in the database, the attack was credited to the ELF as a “bomb threat.”

Limitations exist in the data acquisition and categorisation methodology employed. Of particular importance are concerns regarding the validity of the NAELFPO’s “diary of actions” as the data was provided by an organisation with vested interests in promoting the best image of the ELF. Despite other databases available, such as those created by the Foundation for Biomedical Research[25][26] the North American Animal Liberation Front Press Office reports[27], and numerous scholarly articles[28], this study sought to utilise a single data source, thus eliminating the need to synthesise conflicting information. In an attempt to remove the judgment of the researcher from the acquisition of a data sample, only the NAELFPO “diary of actions” was used despite the understanding that such a source may contain inherent bias. As previously discussed, the lack of descriptive detail present in accounts of some attacks led to the development of coding categories that were more broadly defined then would have been necessary if complete incident descriptions were present for all attacks. To correct for this tendency, the variable coding categories were defined broadly enough to be inclusive of the uncertainty present in the data, while attempting to maintain distinctions. The categories were designed so that a single event could only be classified within one category. Despite these limitations, the NAELFPO dataset provides a singular and complete source for analysis while avoiding the need for the researcher to decide which sources are legitimate and which are to be excluded. At its best, the “diary of actions” represents an accurate, well-researched history of the ELF and affiliated movements. At its most limited, this study analyses the manner in which the ELF’s press office presents the movement to a wider audience; how the press office frames the cells’ actions via their intended messaging.

Findings & Discussion: Targeting

This first section will analyse the targeting pattern present in the ELF data. Target data was coded within 24 targeting types ranging from the common (e.g. 208 attacks on automobiles) to the obscure (e.g. one attack on an advertisement). The results from the data analysis concerning targeting vary dependent on the portion of the sample utilised. When the complete 707-entry[29] dataset (DB1) is analysed, the following findings emerge as the eight most commonly attacked[30] target types:

  • SUV/automobile: 29%
  • Phone booths: 17%
  • Homes under construction/model homes: 12%
  • Company vehicles: 11%
  • Construction/industrial equipment: 10%
  • Business property: 5%
  • Farm/ranch/breeders: 2%
  • McDonalds restaurants: 2%

The other 16 target types each account for less than 2% of the total attacks, and collectively comprise only 12% of the total targets.[31]

When the global dataset excludes the multiple entries (DB2), the predominance of attacks on automobiles and phone booths is reduced, as these targets are typically attacked in groups. In the 211 entry DB2 dataset, the 12 most commonly attacked targets are:

  • Construction/industrial equipment: 14%
  • Home under construction/model homes: 13%
  • Business property: 12%
  • SUV/automobile: 10%
  • Phone booth: 8%
  • Business vehicle: 8%
  • McDonalds: 5%
  • Farm/ranch/breeder: 5%
  • GMO experiment/research: 4%
  • Government property: 3%
  • Trees: 3%
  • Government vehicle: 2%

The other 11 target types each account for less than 2% of the total attacks, and collectively comprise only 14% of the total targets.[32]

When these findings are compared to that of prior scholarship, points of congruency and disagreement can be seen. Though no alternative study could be located using the exact sample data source, in an article based on a dataset consisting of “database of [109] ELF attacks,” occurring between 1996-2001, Leader and Probst [33] report that the most commonly attacked ELF targets are:

  • “corporations”: 33 (36%)
  • “Urban sprawl/development”: 30 (33%)
  • “logging & related”: 18 (20%)
  • “genetic engineering/biotech research facilities”: 14
  • “facilities that threaten animals”: 6 (7%)
  • “government facilities”: 5 (5%)
  • “symbols of global economy”: 3 (3%)

As one can see, a close comparisons becomes quite difficult between the two studies as they adopt different frameworks for categorisation; one based in the nature and identity of the targeted object, and the other in a broader business type category.

Similarly, a US Department of Homeland Security (DHS)-commissioned study, carried out by Helios Global Inc., reported comparable, though more generic results. The Helios study focuses on a conflated history of the ELF and the Animal Liberation Front (ALF) over a longer timeline, from 1981-2005.[34] Because their sample incorporates the ELF as well as the ALF, and their timeframe predates the ELF’s founding by 15 years, an exact comparison is not possible. Regardless of such limitations, according to the Helios study[35], the most frequently attacked “primary targets” of the ALF/ELF are:

  • “Commercial enterprises and/or individuals engaged in housing and urban development
  • Commercial enterprises and/or individuals involved in the logging industry
  • Sports utility vehicle (SUV) dealerships
  • Commercial enterprises and/or individuals involved in the production, sale, and distribution of animal products (leather and fur producers, sellers, and distributors; restaurants; and meat, poultry, and fish producers
  • Animal research facilities and personnel
  • Commercial enterprises and universities involved in genetic engineering.…”

From the Helios study, though different terminology is employed, the targeting findings are quite similar to those contained in the present study. Target category 1, overlaps this study’s category termed, “house under construction/mobile home,” whereas the remaining five categories similarly overlap this study’s use of categorical terms such as: “SUV/automobile,” “business property,” “farm/ranch/breeder,” “laboratory” and “GMO crops or research” respectively.

Returning to the data presented herein, in this ELF-specific study, one can now examine the remainder of the datasets concerning targeting typologies. When the dataset is further reduced to only attacks carried out in the United States, including multiple entries (DB3), the results are largely the same, with the same target types occupying the higher echelons. Notable changes include the exclusion of attacks on “phone booths,” as all such attacks occurred in Mexico, as well as the rising presence of the targeting of “trees” (via tree spiking) as the eighth most common target type, comprising 2% of the total attack pool. In the 462-entry DB3 dataset, the most commonly attacked targets are “SUV/automobile”, “House under construction/model home”, “Business property”, “Construction/industrial equipment”, “Business vehicle”, “McDonalds”, “Farm/ranch/breeder” and “Trees: 2%”.  The other 13 target types each account for less than 2% of the total attacks, and collectively comprise only 12% of the total targets.[36] When this dataset excludes multiple entries (DB4), the results are largely the same. With the exclusion of multiple entries, the targeting types are more evenly distributed, with 15/20 target types accounting for 2% or more of the total pool. Comparison between these two datasets is displayed below:

The other 11 target types each account for less than 5% of the total attacks, and collectively comprise 21% of the total targets.[37]

When targeting type is further reduced to the pool of attacks carried out only in Mexico, including the incorporation of multiple attacks (DB5), the exceedingly high proportion of attacks on “phone booths” is visible. In this sample, 78% of all ELF attacks in Mexico targeted a Telmex phone booth. This is by far the most singularly-focused targeting seen in any of the datasets. In the DB5 dataset, the second most commonly attacked target type is “business property,” comprising only 7%.[38] When multiple attacks are excluded from the dataset (DB6), attacks on phone booths still remain the most common target type, but only account for 39% of the total attacks. Attacks on business property similarly remain the second most commonly attacked target type, now accounting for 24%.[39]

Throughout the findings in regards to targeting, there is a pattern of ELF cells attacking unguarded, ‘soft target’ sites such as vehicles, phone booths and construction sites. In general, such properties would be located in public areas with little or no security. In contrast, target types such as laboratories, ski resorts, banks and government property consistently occupy the lower levels of target selection, possibly because such areas would more commonly employ electronic surveillance systems or human guards. Regardless of the dataset examined, and independent of the inclusion or exclusion of multiple attacks, business properties and construction sites are routinely targeted. In the US datasets, there is a dominant pattern of targeting homes under construction and model homes, though this pattern is not seen in other national settings. This is likely a reflection of the growing “sprawl” critique as seen in anti-globalisation, anti-gentrification, anti-capitalist and anarcho-primitivist movements in the US, a site where many ELF activists find their ideological groundings[40]. In other nations, such as Mexico, there is no record of the targeting of “sprawl” sites, as Mexican cells have focused their attention on attacking phone booths as part of a larger campaign against Telmex, a company described in communiqués as “earth destroying,”[41] and guilty of “biocide”[42].

Throughout the data collection and coding process, attention was paid to determining if the ELF attack carried out was part of a larger stated campaign, thus leading to the specific location being targeted. In the US, two campaigns were identified. The most prominent was the anti-sprawl campaign in Long Island, N.Y, comprised of eight distinct attacks (42 multiple entry attacks), occurring from September 2000 to January 2001. Such attacks accounted for approximately 8% of the total attacks carried out in the US. Craig Rosebraugh, former ELF spokesman, notes the prominence of the Long Island anti-sprawl campaign, writing that it “constituted the most focused and intensified campaign the ELF had ever undertaken,”[43] consisting of 11 “major” attacks including five arsons of homes and condominiums under construction [44]. The second largest campaign targeted affiliates of animal research supplier Huntingdon Life Sciences (HLS).[45] Six distinct anti-HLS attacks were carried out in the US, comprising less than 2% of the overall attacks. Such campaigns do not appear to be prominent in the ELF’s targeting system as approximately 91% of all US attacks were not part of a stated campaign.[46] In Mexico, this trend dramatically changes, as over 81% of all multiple entry attacks, and over 51% of all distinct entry attacks are part of the campaign targeting Telmex. Mexico also was the venue for one anti-HLS attack.[47]

In examining atypical events, some of the data that comprises the rare incidents, the outliers, though not statistically significant, are deserving of brief discussion. In only one distinct attack a ski resort was targeted, yet despite its rarity, the Vail, Colorado arson is often the most commonly heard of ELF attack, possibly because it caused approximately $26 million in damages [48]. At the time the datasets were being constructed, a photograph of the Vail fire was featured on the ELF’s main page, cataloging the movement’s “diary of actions”[49]. A second atypical targeting discovery focuses on cells’ decisions to target human life and not solely property. Throughout the movement’s 13-year history, only one attack directly targeted a human being. On 3 June 2009, an Australian cell of the ELF “hand delivered” a written threat to the home of a Hazelwood Power Station CEO, located in Melbourne[50]. The note threatened the individual’s property not his person, but because the threat was addressed towards a specific person, the incident was recorded as an attack targeting an individual not their property. Finally, in the targeting of fast food establishments, 15 attacks were directed at McDonald’s restaurants, while in only one attack, a Burger King was targeted.  This particular action was taken in 2002 by an ELF cell in the US city of Richmond, Virginia [51].

Findings & Discussion: Tactics

The findings related to the tactics employed by ELF cells show little variation when the different datasets are examined comparatively. In every dataset, regardless of national location or the inclusion of “multiple entries,” the top three tactical choices are: “sabotage/vandalism/graffiti,” “arson” and “graffiti.” When viewed across datasets, the proportionality of these tactics changes, as does their usages vis-à-vis one another (but regardless of these variations), these three tactical choices consistently occupy the top three positions in the tactical tool belt. This cross-dataset trend can be viewed in the comparison chart below (Table #1) wherein the frequency of the three most common tactical typologies are compared across DB1-DB6:

In five out of six datasets, “sabotage/vandalism/graffiti” is the most commonly employed tactics, followed by “arson,” and finally, “graffiti.” In only one instance does “arson” calculate as the most commonly employed tactic.[52] After the first three most commonly employed tactics, the breakdown across the various datasets begins to show greater diversity. Below is a comparison (Table #3) of the fourth, fifth and sixth most commonly utilized tactics within the six datasets:

When examined together, these tactical trends show some similarity, with the breaking of windows, tree spiking [TS], attempted arsons [abbreviated  as “At ar”] and animal liberations proving common in the first four datasets, while the Mexican datasets (DB5-6) show identical results.[53]

The findings concerning tactics of this study can be compared to similar attempts in the scholarly literature. The Helios study summarises the tactical choices of “ecoterrorists,” a broad category including but not limited to the ELF, and represents the totality of such attacks within five tactical typologies. According to the Helios study[54], the tactical breakdown of “ecomilitant tactics carried out between 1981 and 2005,” can be summarised as:

  • “Vandalism”: 45%
  • “Theft”: 23%
  • “Harassment”: 15%
  • “Arson”: 10%
  • “Bombing”: 7%

The Helios results share some findings with this study, as both agree that “vandalism” (termed “sabotage/vandalism/graffiti” in this study) as broadly defined, is the most commonly employed tactic, and “bombing[s]” (termed “IED” in this study), is the least commonly used tactic. Both studies also agree that within these extremes, “ecomilitants” use other tactics including arson and theft. Though not detailed in the study, one can assume that “Theft” for Helio encompasses the removal (or release) of live animals from slaughterhouses, breeding facilities, laboratories, etc. The Leader and Probst (2003) article reports similar findings but utilises smaller categorical groupings. Based on 92 attacks, according to the article[55], the three most common tactic types are:

  • “Vandalism”: 36 (33%) [56]
  • “Arson”: 32 (29%) [57]
  • “Sabotage”: 19 (17%) [58]

Once again, such findings support those of this study, in that both report the most commonly utilizsed tactics combine sabotage, vandalism, graffiti, arson and attempted arson.

In attempting to identify inaccuracies within the literature—such as those positioned to embolden security debates—the tactical descriptions of the ELF are likely the most important areas to examine. The Leader and Probst article asserts “ELF’s prime weapon is arson”,[59] though this claim is not supported by their own data nor the research presented herein, as more general sabotage and vandalism tactics generally show a higher predominance, as they do in datasets DB1-DB5, excluding DB6, where arson does surface as the most commonly utilised tactics in “distinct incident” attacks carried out in Mexico. In Leader and Probst’s own findings, vandalism occurs slightly more commonly than arson, and thus the statement that the movement’s “prime weapon is arson” appears hyperbolic for the sake of rhetoric. Linked to the tactics chosen for attack, are issues of lethality and threat to life. Casualty data was collected for every incident in the datasets. Throughout the 707, multinational, all incident dataset (DB1), and thus all secondary datasets, no ELF attack is reported to have caused any injuries or fatalities to human beings. This finding is supported by the scholarly literature[60] in every example surveyed and places a big question mark as to the rationale for the categorisation of ELF activities under ‘terrorism’.

Findings & Discussion: Claims of Responsibility

The presence of a communiqué documenting an attack is common throughout the different attacking cells. In cases where a formal communiqué is not issued, attackers sometimes leave ELF ‘calling cards’ such as the group’s name scrawled in graffiti, notes, or banners. In the following chart (Table #4), the comparison of communiqués and ‘calling cards’ is shown across the six datasets:

When compared across the six datasets, the trend is relatively uniform. In all cases, communiqués are more commonly issued then not, though their existence has varying degrees of regularity. In the first four datasets (DB1-4), communiqués are issued for more than half of all attacks (56% on average), and when a communiqué is not issued an ELF ‘calling card’ is present in approximately 19% of all cases.[61] In the Mexican datasets, communiqués are issued 100% of the time.[62]

Although this study has focused on attacks carried out by cells self-identifying as members of the ELF, occasionally, attacks are claimed in the name of a cooperative endeavor by the ELF and ALF through either a communiqué or through the NAELFPO’s description of the attack. The following chart (Table #5) shows the proportion of attacks claimed solely by the ELF, as compared to those claimed mutually by the ELF/ALF:

This cross-dataset comparison shows that the claiming of ELF/ALF joint actions is marginally more common in the non-US, non-Mexican arena. In the Mexican-only datasets, such cooperative claims of responsibility are exceedingly rare, accounting for only one attack in the entire country’s history.[63]

Findings & Discussion: Location

According to the data collected, the ELF is active in fourteen countries across the continents of North America, South America, Europe and Australia. The highest concentration exists in the English-speaking, ‘Western’ world of North America and Western Europe, though the presence of active cells in Mexico appears to be increasing [64]. In DB1, the countries with the highest rates of attacks are the United States, Mexico, United Kingdom and Canada. The other ten national locations account for less than 2% of the total attacks each, and collectively comprise only 7% of the total events.[65] Making up this collective 7%, are attacks carried out in Spain, Italy, the Netherlands, Sweden, Iceland, Russia, Australia, New Zealand, Chile and Colombia. When the dataset is reduced, examining only “distinct incidents,” the results are the same, both in the nations identified, as well as their general ranking vis-à-vis one another. The most common countries remain the US, Mexico, Canada and UK and the other ten nations account for less than 1% of the total attacks each, and together comprise less than 3% of the total events.[66] A comparison of these two similar findings is displayed below:

These findings are similar to those reported in the Helios study wherein the authors state, “despite their global presence…acts of terrorism appear to be most prevalent in North America and Western Europe”[67]. Helios' exclusion of Mexico as a target of ELF attacks is expected, as the study was published in 2008, the year when Mexico began to experience activity from ELF cells.

The United States is overwhelmingly the focus of the ELF’s international campaign, despite the fact that the movement as it exists today emerged in England[68]. The availability of information relating to the location of ELF attacks occurring in the US lends itself well to analysis, as there is no need for the researcher to equitable develop categories or to extrapolate variable labels from attack narratives. Data on location as it pertains to state was available for all but one “distinct incidents,” and regional divisions were determined based on mapping provided by the US census report.[69] In the chart below (Table #7), attacks within the US (DB3- DB4) will be compared in regards to region and regional division:

From this data, one can see that the Western region (specifically the five state Pacific division), has been a particular center of activity, concentrated in the states of California, Oregon and Washington. Similarly, the Middle Atlantic three state division within the Northeast region has been particularly active, as numerous attacks have been carried out in the states of New York and Pennsylvania. The least active region appears to be the Midwest despite its large geographic area. The least active regional division is the West South Central, four state grouping of the southern region.[71]

In surveying the literature concerning the ELF, only the Helios study includes a detailed discussion of location as it pertains to regions throughout the US. Although the Helios sample is different from the one employed herein, the findings are similar. The Helios study concludes “ecoterrorists” are “particularly active in the Western and West Coast states. In particular…Oregon, California and Washington…the Midwest and East Coast have a smaller percentage of eco-terrorist incidents”[72]. Certainly, the data contained herein supports the Helios claim that such movements are especially active in the “Western and West Coast states,” and state-specific data supports the claim that high levels of activity is seen in Oregon, California and Washington.[73] The claim, that the Pacific Northwestern region “is the most prominent environmental hot spot in the nation” is also offered, by former ELF spokesmen Craig Rosebraugh[74]. Also, both the Helios study and this study agree that the Northeast (called East Coast in Helios study) and Midwest occupy the lower regions for ELF activity.


This article draws on the case study of the ELF to demonstrate the analytical potential of conducting a quantitative tactical analysis of activism of social movement groups in order to debunk hyperbolic tropes of "terrorism." For example, methodological decisions related to categorisation, coding and data sourcing can be used to skew data towards hyperbole, fear-mongering and securitisation or they can be used to approach greater accuracy, nuance and balance. The preceding dataset challenges the framing of radical environmental groups as terrorist threats to the nation-state. This rhetorical framing—especially that dealing with tactics and targeting—supports the increased government repression of leftist movements through targeted legislation such as the Animal Enterprise Terrorism Act of 2006 and the larger atmosphere of the “Green Scare”[75]. Future research can challenge limits to dissent through quantifying movement group actions, and calling into question government tropes about "radical" movements. With such political and methodological concerns in mind, the data presented can allow scholars to develop an incident-driven history of the ELF beyond broad state framings as bomb-throwing ‘terrorists’ and ‘arsonists.’

The ELF, a transnational movement of direct action eco-saboteurs, follows a definitive targeting and tactical pattern, focusing its attacks on unguarded properties associated largely with commercial and residential construction, automobiles (especially SUVs), and various regional, national and multinational business interests. ELF cells target such entities clandestinely, and with low-tech tactics, often striking multiple sites within one target type in rapid succession. For example, it is common for one cell to vandalise dozens of SUVs in one outing. The targeting patterns follow regional indicators concerning campaigns developed through attack histories in that locale. In the US, such attacks have focused on targets associated with “sprawl” and residential development, SUV sales and ownership, and construction sites. In Mexico, attacks have focused on a campaign targeting Telmex phone booths and other affiliated properties. The majority of ELF attacks are not part of larger attack campaigns, though in about 5% of US attacks, and 34% of Mexican attacks, the communiqué stated that the target was chosen as part of a long-term campaign; focusing strikes on a specific set of entities, linked thematically.

Tactically, ELF cells tend to rely on varied combinations of vandalism (including graffiti), sabotage and arson. Throughout all of the data, a combining of vandalism and sabotage has dominated the tactical history, with arson occurring as the second most commonly employed tactic. In extremely rare instances (six attacks out of 707 equaling 0.85%), cells have used tactics that direct violence against humans, or present the threat of violence against humans through the use of improvised explosive devices, bomb hoaxes, and written threats.

Through a combined analysis examining targeting as it compares to tactics, one witnesses the defining modus operandi of ELF cells.[76] When distinct attacks are examined globally, the arson of residential units, as well as the sabotage and vandalism of construction vehicles and other business properties emerges as the most dominant attack patterns.[77] When the US is examined separately, one sees the same pattern of homes being targeted through arson, business properties targeted through sabotage and vandalism, and SUVs targeted with graffiti.[78] In the purely Mexican context, targeting and tactics collide at the vandalism, sabotage and arson of Telmex phone booths, and to a far lesser extent, the arson of construction and industrial equipment.[79]

Concerning cells’ claims of responsibility for attacks, in the global context, attacks are claimed via a formal communiqué sent to either an aboveground press office or other media outlet more than 70%, of the time.[80] In other instances, the ELF name is left at the scene of the crime to indicate the movement’s claim of responsibility. In only approximately 17% of ELF attacks is the incident linked to the movement, but not formerly claimed via a communiqué or other form of communication.[81] The ELF movement rarely reports that its cells have jointly carried out attacks cooperatively with cells affiliated with the Animal Liberation Front. The ELF/ALF cooperative moniker is seen globally in approximately 6% of cases, with a greater frequency seen in attacks occurring outside of North America.[83]

Trends in attack location indicate that the ELF, while focused primarily in the United States, is having an expanded sphere of activity in Mexico. Other sustained areas of activity include Canada—centered in Ontario province—and the United Kingdom—especially within England. Within the US, attacks have focused on the eastern and western coastal areas, centered in the states of California, Oregon, Washington, New York and Pennsylvania. Outside of North America and the UK, sparse attacks have been documented in the continents of Europe (especially Western Europe and Scandinavia), South America and Australia. At the present time, there are no reports of ELF attacks within Antarctica or the African or Asian continents.

Finally, the preceding analysis has attempted to diagram the attack history of the ELF through a transparent methodology. In doing so, one is able to comparatively evaluate its findings alongside that of other, more opaquely-authored studies. While it is true that the preceding findings were constructed around an a-priori agenda—namely providing defensible data for furthering nuanced and well-informed debates regarding emergent social movements—this is no different than scholarship that came before or will likely come after. If the critical analysis of state and academic scholarship is seen as ‘having an agenda’ could one not say the same thing of well-circulated papers built around an a-priori agenda of securitisation? For example, the 2013 DHS-funded report authored by the National Consortium for the Study of Terrorism and Responses to Terrorism (START), while academically rigorous, cannot be described as politically impartial.  The report explicitly describes its mission in the opening pages stating:

This report is part of a series in support of the Prevent/Deter program. The goal of this program is to sponsor research that will aid the intelligence and law enforcement communities in assessing potential terrorist threats and support policymakers in developing prevention efforts.[83]

Studies such as those conducted by START or Helios are funded by, and produced explicitly for, the policing of dissent, as they are identified as projects of the DHS. The collection and publication of such data fits into a larger American, post-9/11 shift in domestic policy -  a shift from policing to national security. Scholarship of this nature can be as rigorous (or manipulative) as any academic pursuit, but to contend that it does not possess a pre-existent ideological framework and political agenda is to misunderstand statecraft as a neutral endeavour.

The data suggests that the label “eco-terrorism” has been misapplied to a form of political militancy that falls short of what can reasonably be called “terrorism” since there have been practically no deliberate deadly attacks on civilians that would warrant the use of such a loaded term. While the actions of the perpetrators are often unlawful since these tend to involve acts of vandalism, arson or sabotage, and while these acts are meant to convey a message to a wider audience, that is still a far cry from the bloody terrorism of, for example, salafist jihadists. The terrorism label loses its potency if it is stretched beyond credibility. It should be used sparingly, rather than loosely and be limited to certain categories of gross violations of human rights and international humanitarian law – roughly the peacetime equivalent of war crimes.[84]

Acknowledgement: The author would like to thank Editor-in-Chief of Perspectives on Terrorism for his kind encouragement and guidance throughout this process, as well as the anonymous reviewers, especially the individual who challenged the article’s “a-priori agenda.” Lastly, the author would like to thank the past, present and future anonymous and known contributors to the North American Earth Liberation Front Press Office, the North American Animal Liberation Press Office, Bite Back Magazine, the Talon Conspiracy and others.  Without their dedicated archival efforts, the histories of these movements would have been far more difficult to reconstruct.

About the Author:  Michael Loadenthal is a Doctoral Fellow at the School for Conflict Analysis and Resolution, George Mason University (Arlington, VA) and teaches ‘Terrorism and Political Violence’ at the Program on Justice & Peace at Georgetown University (Washington, DC).  He can be reached at [email protected] and regularly posts papers, presentations and other work at Michael has been surrounded by ‘radical’ social movements as both a student and teacher, participant and observer, for more than fifteen years. He can usually be found reading to his daughters, writing about political violence, or conspiring to build a better world.

The appendices for this article can be found in full in the PDF version.


[1] Richard Jackson. Constructing Enemies: ‘Islamic Terrorism’ in Political and Academic Discourse, Government and Opposition, 42, 3, 2007, 394–426.; C Kurzman and C.W. Ernst. Islamic Studies in U.S. Universities (presented at the Social Sciences Research Council workshop on “The Production of Knowledge on World Regions: The Middle East”). Chapel Hill: University of North Carolina, 2009, 1-34.; David Miller and Tom Mills. The Terror Experts and the Mainstream Media: The Expert Nexus and Its Dominance in the News Media, Critical Studies on Terrorism, 2, 3, 2009, 414–437, Magnus Ranstorp. "Introduction: Mapping Terrorism Research," in Mapping Terrorism Research: State of the Art, Gaps, and Future Directions. London: Routledge, 2007, 1–28; Jessica Shepherd. The Rise and Rise of Terrorism Studies, The Guardian, 3 July 2007,; Andrew Silke. “Contemporary Terrorism Studies Issues in Research,” in Critical Terrorism Studies: A New Research Agenda. New York: Routledge, 2009, 34–48.; Yasir Suleiman and Ayman Shihadeh. Islam on Campus: Teaching Islamic Studies at Higher Education Institutions in the UK, Journal of Beliefs & Values, 28, 3, 2007, 309–329.

[2] Bob Brecher, Mark Devenney, and Aaron Winter (Eds.) Discourses and Practices of Terrorism: Interrogating Terror. New York: Routledge, 2010; Richard Jackson, Eamon Murphy, and Scott Poynting (Eds.) Contemporary State Terrorism: Theory and Practice. New York: Routledge, 2010; Scott Poynting and David Whyte (Eds.) Counter-Terrorism and State Political Violence: The “War on Terror” as Terror. New York: Routledge, 2012; Jacob L. Stump and Priya Dixit, Critical Terrorism Studies: An Introduction to Research Methods. New York, Routledge, 2013; Mikkel Thorup, An Intellectual History of Terror: War, Violence and the State. New York: Routledge, 2012.

[3] Rita Floyd. When Foucault Met Security Studies: A Critique of the ‘Paris School’ of Security Studies (presented at the 2006 British International Studies Association Annual Conference, University of Cork, Ireland), 2006; Rita Floyd. Towards a Consequentialist Evaluation of Security: Bringing Together the Copenhagen and the Welsh Schools of Security Studies, Review of International Studies, 33, 2, 2007, 327–350., Mark Salter and Can Mutlu (Eds.) Research Methods in Critical Security Studies: An Introduction. New York: Routledge, 2012.; Paul Williams and Alex Bellamy (Eds.) Critical Security Studies, in International Society and Its Critics, 2004.; Nick Vaughan-Williams and Columba Peoples. Critical Security Studies: An Introduction.  New York,: Taylor & Francis, 2010.; Laura J. Shepherd (Ed.) Critical Approaches to Security: An Introduction to Theories and Methods. New York: Routledge, 2013.

[4] Glenn Greenwald. The Sham ‘terrorism Expert’ Industry, 15 August 2012,

[5] Will Potter. Green Is the New Red: An Insider’s Account of a Social Movement Under Siege. San Francisco: City Lights Publishers, 2011.

[6] Dara Lovitz. Muzzling a Movement: The Effects of Anti-Terrorism Law, Money, and Politics on Animal Activism. Herndon, Lantern Books, 2010; Colin Salter. Activism as Terrorism: The Green Scare, Radical Environmentalism and Govermentality, Anarchist Developments in Cultural Studies, 2, 1, 2011, 211–238.

[7] Since completing the data collection for this study, the website of the North American Earth Liberation Front Press Office has been taken offline. While much of the data is available at other sources such as the North American Animal Liberation Press Office ( and Bite Back Magazine ( I also have access to an offline archive of the NAELFPO website as it appeared on 15 January 2010.

[8] In August 2013, the National Consortium for the Study of Terrorism and Responses to Terrorism (START), affiliated with the Department of Homeland Security’s Science and Technology Center of Excellence at the University of Maryland issued a comprehensive study of ELF and ALF bombings and arson. While a complete comparison between these findings and the findings presented herein is not included, in brief, START’s findings are surprisingly similar when a similar data parameters are added to my data set. Though this study examined only the Earth Liberation Front, START also reviews Animal Liberation Front activity as well. A comparable dataset was produced from my own research fitting START’s requirements: 1.) a bombing or arson 2.) occurring between 1995-2000, 3.) in the US, 4.) and fitting a radical environmental or animal liberationist politic.  Including these parameters, I located 285 incidents. The START study includes 239. The similar number of incidents points to similar methodologies for locating resources as described herein as well as in START’s report. A complete comparison of these findings will be discussed in a subsequent paper.

[9] North American Earth Liberation Front Press Office. Earth Liberation Front Diary of Actions. Earth Liberation Front Press Office, 2009,

[10] The variables recorded for each attack are: 1.) date, 2.) group responsible, 3.) tactic type, 4.) target type, 5.) casualties, 6.) national location, 7.) state or province for attacks occurring in the US and Canada, 8.) damage estimate in USD, 9.) multiple or distinct incident, 10.) presence of communiqué or other claim of group responsibility, and lastly, 11.) was the attack part of a larger ELF campaign? Distinct values were assigned in all cases except for #8, damage estimate in USD, as this data was often not provided and could not be estimated.

NB: The dataset analysed in this study is a subset of a much larger project to be analyzed and published at a later date. The larger dataset is comprised of approximately 30,000 global “eco-terrorist” incidents, occurring from 1972-present, and coded for twenty-two variables.

[11] Brian Jackson and David Frelinger. Rifling Through the Terrorists’ Arsenal: Exploring Groups’ Weapon Choices and Technology Strategies, Studies in Conflict & Terrorism, 31, 7, 2008, 587.

[12] Vague dates were estimated in a standardized manner, thus when only a month and year were given in the description, the 15th was chosen as it represents the middle of the month. For example, a date reported as “06/04-07/04” was recorded in the database as occurring on 15 June 2004.

[13] In the case of multiple attacks, when damage estimates were available, the dollar value for the damage incurred was divided evenly between the events, thus an arson spread across four targets, causing $10,000 in damage, was recorded as four arsons of $2,500 each.

[14] Brian Jackson and David Frelinger. Rifling Through the Terrorists’ Arsenal: Exploring Groups’ Weapon Choices and Technology Strategies, Studies in Conflict & Terrorism, 31, 7, 2008, 602.

[15] Ibid.,  587.

[16] The DB1-DB6 abbreviations used throughout constitute:

  • DB1: All countries, multiple entry inclusive dataset, (707 incidents)
  • DB2: All countries, “distinct incident” dataset, (211 incidents)
  • DB3: US only, multiple entry inclusive dataset, (462 incidents)
  • DB4: US only, “distinct incident” dataset, (134 incidents)
  • DB5: Mexico only, multiple entry inclusive dataset, (150 incidents)
  • DB6: Mexico only, “distinct incident” dataset, (41 incidents)

[17] North American Earth Liberation Front Press Office. Earth Liberation Front Diary of Actions- 2008. Earth Liberation Front Press Office, 2009,

[18] North American Earth Liberation Front Press Office. Earth Liberation Front Diary of Actions- 2009. Earth Liberation Front Press Office, 2009,

[19] The exact name of the group, in Spanish is “Ludditas Contra La Domesticacion de la Naturaleza Salvaje.”

[20] “Frente de Liberación de la Tierra,” is the Spanish term for the ELF, literally translated it is ‘Front of Liberation of the Land’ (FLT)

[21] La célula del ELF-LCDNS México. Earth Liberation Front Communiqué 07.20.09: ELF Targets Telmex Branch Office in Mexico. Earth Liberation Front Press Office, 2009,

[22] North American Earth Liberation Front Press Office. Earth Liberation Front Diary of Actions- 2001. Earth Liberation Front Press Office, 2009,

[23] No additional data sources were searched (outside of the NAELFPO “diary of actions”) to determine if a communiqué was issued.

[24] North American Earth Liberation Front Press Office. Earth Liberation Front Diary of Actions- 2004. Earth Liberation Front Press Office, 2009,

[25] Foundation for Biomedical Research. Illegal Incidents (1997 - Present). Foundation for Biomedical Research, 2009,

[26] In addition to the “Illegal Incidents Map,” the Foundation for Biomedical Research was contacted via email and provided over 70 pages of data it had collected documenting attacks carried out by the ELF and the Animal Liberation Front targeting the ‘medical industry.’ This data was excluded from the sample.

[27] North American Animal Liberation Front Press Office. 2008 Year- End Report on Animal Liberation Activities in North America. North American Animal Liberation Front Press Office, 2009,; North American Animal Liberation Front Press Office. 2001 Year-End Direct Action Report. North American Animal Liberation Front Press Office, 2002),

[28] Stefan H. Leader and Peter Probst. The Earth Liberation Front And Environmental Terrorism, Terrorism and Political Violence, 15, 4, 2003, 37–58.; Helios Global. Ecoterrorism: Environmental and Animal-Rights Militants in the United States. Washington DC: Helios Global Inc./U.S. Department of Homeland Security, 2008.

[29] In addition to the “Illegal Incidents Map,” the Foundation for Biomedical Research was contacted via email and provided over 70 pages of data it had collected documenting attacks carried out by the ELF and the Animal Liberation Front targeting the ‘medical industry.’ This data was excluded from the sample.

[30] For the purpose of discussion, percentages have been rounded to the nearest whole number. The exact percentages as well as the actual number of attacks they represent are included in the corresponding data tables provided in the appendix.

[31] For the complete targeting DB1 data table, see appendix chart 1.

[32] For the complete targeting DB2 data table, see appendix chart 2.

[33] Stefan H. Leader and Peter Probst. The Earth Liberation Front and Environmental Terrorism, Terrorism and Political Violence, 15, 4, 2003, 43.

[34] The dataset used in the Helios study, is likely most similar to the DB2 dataset, as it represents a global movement and makes no mention of a multiple entry feature factoring into their calculations.

[35] Helios Global. Ecoterrorism: Environmental and Animal-Rights Militants in the United States. Washington DC, Helios Global Inc./U.S. Department of Homeland Security, 2008, 7.

[36] For the complete targeting DB3 data table, see appendix chart 3.

[37]For the complete targeting DB4 data table, see appendix chart 4.

[38] For the complete targeting DB5 data table, see appendix chart 5.

[39] For the complete targeting DB6 data table, see appendix chart 6.

[40] Craig Rosebraugh. Burning Rage of a Dying Planet: Speaking for the Earth Liberation Front. Herndon: Lantern Books, 2004, 121–126.; Sally and Peter. “ELF Claims Vandalism Against New Housing Developments in Philadelphia Suburb,” in: Igniting a Revolution: Voices in Defense of the Earth. Oakland: AK Press, 2006, 415–416; Karen Ziner. Burning Down the Houses - Earth Liberation Front Tries to Stop Urban Growth in Rural Long Island, E: The Environmental Magazine, 2001,

[41] Frente de Liberación de la Tierra. Earth Liberation Front Communiqué: 10.04.09 Earth Liberation Front Torches Telmex Cell Tower in Mexico State. North American Earth Liberation Front Press Office, 2009,

[42] FLT/ELF- Mexico. Earth Liberation Front Communiqué, 01.07.09: Mexico: ELF Burns 5 Telmex Phone Booths. North American Earth Liberation Front Press Office, 2009,

[43] Craig Rosebraugh. Burning Rage of a Dying Planet: Speaking for the Earth Liberation Front. Herndon: Lantern Books, 2004, 157.

[44] Craig Rosebraugh. Burning Rage of a Dying Planet: Speaking for the Earth Liberation Front. Herndon: Lantern Books, 2004, 161.

[45] The campaign against Huntingdon Life Sciences has been led aboveground by Stop Huntingdon Animal Cruelty (SHAC), and most anti-HLS attacks have been claimed by the Animal Liberation Front. Generally, within this campaign SHAC helps to identify and publicize HLS-related primary and secondary targets, and attacks are carried out against these entities by underground, clandestine cells with attacks often being claimed by the Animal Liberation Front. An example of a SHAC produced targeting list, visit:

[46] For the complete campaign DB3-DB4 data tables, see appendix, chart 7-8.

[47] For the complete campaign DB3-DB4 data tables, see appendix, chart 9-10.

[48] Earth Liberation Front. Communiqué 10/19/98. North American Earth Liberation Front Press Office, 1998,

[49] North American Earth Liberation Front Press Office. Earth Liberation Front Diary of Actions. Earth Liberation Front Press Office, 2009,

[50] Earth Liberation Front. Earth Liberation Front Communiqué: 06.03.09 Earth Liberation Front Leaves Message for Hazelwood CEO Graeme York in Australia. North American Earth Liberation Front Press Office, 2009,

[51] North American Earth Liberation Front Press Office. Earth Liberation Front Diary of Actions - 2002. Earth Liberation Front Press Office, 2009,

[52] These entries have been marked by *s in the data table #1.

[53] For the complete tactical data tables DB1-DB6, see appendix charts 11-16.

[54] Helios Global. Ecoterrorism: Environmental and Animal-Rights Militants in the United States. Washington DC: Helios Global Inc./U.S. Department of Homeland Security, 2008, 14.

[55] Stefan H. Leader and Peter Probst. The Earth Liberation Front And Environmental Terrorism, Terrorism and Political Violence, 15, 4, 2003, 41.

[56] Defined as “low level damage such as breaking windows, spray painting of slogans, etc.” (Leader and Probst, 41)

[57] Defined as “Deliberately starting a fire or attempting to start a fire to cause property damage.” (Leader and Probst, 41)

[58] Defined as “deliberate damage to equipment, vehicles, crops, buildings etc.” (Leader and Probst, 41)

[59] Stefan H. Leader and Peter Probst. The Earth Liberation Front and Environmental Terrorism, Terrorism and Political Violence, 15, 4, 2003, 41.

[60] Gary A. Ackerman. Beyond Arson? a Threat Assessment of the Earth Liberation Front, Terrorism and Political Violence, 15, 4, 2003, 162.; Randy Borum and Chuck Tilby. Anarchist Direct Actions: A Challenge for Law Enforcement, Studies in Conflict & Terrorism, 28, 3, 2005, 212.; Stefan H. Leader and Peter Probst. The Earth Liberation Front And Environmental Terrorism, Terrorism and Political Violence, 15, 4, 2003, 44;  Bron Taylor. Religion, Violence and Radical Environmentalism: From Earth First! to the Unabomber to the Earth Liberation Front, Terrorism and Political Violence, 10, 4, 1998, 3,8.

[61] For the complete communiqué data tables DB1-DB4, see appendix charts 17-20.

[62] For the complete communiqué data tables DB5-DB6, see appendix charts 21-22.

[63] For the complete group claim data tables DB1-DB6, see appendix charts 23-28.

[64] John Ross. Our Fires Illuminate the Night: Wave of Anarchist Bombings Strikes Mexico, Counterpunch, 6 October 2009,; John Ross. The Mexican Revolution at 100: Mexico Welcomes 2010 With Bombs and Riots, Counterpunch, 11 January 2010,

[65] For the complete national location data table DB1, see appendix charts 29.

[66] For the complete national location data table DB2, see appendix charts 30.

[67] Helios Global. Ecoterrorism: Environmental and Animal-Rights Militants in the United States. Washington DC, Helios Global Inc./U.S. Department of Homeland Security, 2008, 11.

[68] Noel Molland. “A Spark That Ignited a Flame: The Evolution of the Earth Liberation Front,” in Igniting a Revolution: Voices in Defense of the Earth. Oakland,:AK Press, 2006, 49.

[69] Descriptions of the US regions and divisions are taken from the US “Census Regions and Divisions” as of 14 June 2000. A graphical version of this breakdown, including the regional and divisional splits is available at:

[70] The appendix’s “Definition of Abbreviations and Terms, Region and Division” contains all necessary formation regarding the territorial breakdown of the US census regions and regional division terminology.

[71] For the complete US state regional and divisional data tables DB3-DB4, see appendix charts 31-32

[72] Helios Global. Ecoterrorism: Environmental and Animal-Rights Militants in the United States. Washington DC, Helios Global Inc./U.S. Department of Homeland Security, 2008, 11.

[73] For the complete, US state individual data tables DB3-DB4, see appendix charts 33-34.

 [74] Craig Rosebraugh. Burning Rage of a Dying Planet: Speaking for the Earth Liberation Front. Herndon: Lantern Books, 2004, 76.

[75] Will Potter. Green Is the New Red: An Insider’s Account of a Social Movement Under Siege. San Francisco: City Lights Publishers, 2011.

 [76] For a complete series of tables relating TACTIC to TARGET, for DB1-DB6, see appendix charts 35-40.

[77] These calculations are based on the DB2 dataset.

[78] These calculations are based on the DB3-4 datasets.

[79] These calculations are based on the DB5-6 datasets.

[80] These calculations are based on an average of all (DB1-DB6) datasets.

[81] These calculations are based on an average of all (DB1-DB6) datasets.

[82] These calculations are based on an average of all (DB1-DB6) datasets.

[83] National Consortium for the Study of Terrorism and Responses to Terrorism. Overview of Bombing and Arson Attacks by Environmental and Animal Rights Extremists in the United States, 1995-2010. College Park, MD: U.S. Department of Homeland Security Science and Technology Center of Excellence, 2013, 1.

[84] Cf. Alex P. Schmid.’ Frameworks for Conceptualising Terorrism’. Terrorism and Political Violence, 16 (2), 2004, 197-221.

Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.


Perspectives on Terrorism is  a journal of the Terrorism Research Initiative and the Center for Terrorism and Security Studies

ISSN  2334-3745 (Online)

Disclaimer, Terms and Conditions