Terrorism in North America (Canada, United States & Mexico), 1970-2010: A Research Note

Terrorism in North America (Canada, United States & Mexico), 1970-2010: A Research Note

 by Richard J. Chasdi



The shooting on September 5, 2012 at Parti Quebecois headquarters in Montreal that police now describe as a full blown assassination attempt against this party’s “premier-designate” Pauline Marois, resulted in the death and injury of two persons. Montreal police believe the suspect, Richard Blain, was spurred to action by the underlying prospect of increased French influence in Quebec.[1] That event, reminiscent of the shooting of U.S. Representative Gabrielle Giffords (D-Arizona) and others in Tucson in 2011, compels us to re-examine the issue of terrorism trends in Canada from a comparative analysis perspective that takes into account “long-haul” terrorism trends in Canada, the United States, and Mexico between 1970 and 2010 as well as a select number of more contemporary events.

A rigorous comparative analysis of terrorism trends within and between those countries is clearly beyond the scope of this brief Research Note. Its purpose is to present a basic first pass comparison of trends that might serve to illuminate relationships between variables for future research, and improve data compilation and coding procedures to produce richer and more multifaceted research.[2] The reasons this research direction is important are threefold: first, the emergent reality that terrorist assault “attributes” in each country continue to evolve and may have links to not only to endogenous variables but also to “systems factors” such as 9/11, and the “global war on terrorism” in response. Second, there are newer developments such as the increasingly powerful set of modalities between common criminal activity and terrorism (e.g., so-called “organized crime,” “gang activity”) within certain countries.[3] Third, there is continued interest in the role that “structural factors” such as “political regime-type” (e.g., “western style liberal democracy,” “authoritarian democracy,” “authoritarian”) and non-state actors might play to influence the structural shape of terrorist assault patterns.[4]

At the same time, such comparative analysis would reflect good research design as it makes it possible to isolate and identify “contextual factors” at work in each country as there is control for region (i.e., North America), and political regime type because all three countries under consideration are examples of Western-style liberal democracy. Furthermore, each is an example of a “federalist system” where, as Thomas Patterson tells us, there exists a condition within the political framework of “shared sovereignty” between national government at the federal level and states or provinces at the regional level.[5]

The data used for this presentation of terrorism trends in Canada, the United States, and Mexico are compiled from data produced by the National Consortium for the Study of Terrorism and Responses to Terrorism (START) at the University of Maryland, with supplementary qualitative summaries of counter-terrorist actions and trends from “Country Reports on Terrorism 2011” issued by the U.S. Department of State in 2012.[6] The framework for discussion involves a presentation of basic findings for each country, some impressionistic observations about possible trends and relationships in the data to explore in the future, and comments about coding and data compilation to enhance comparative analysis of terrorism trends across North America.



In the case of Canada, the “Global Terrorism Database (GTD) from START provides a broader perspective of terrorism trends for a forty year period with a breakdown of terrorist “incidents” by year. From the start, it is clear the pattern for terrorist event frequency in Canada is “cyclical” in nature, itself reflective of a condition that Eric Im, Jon Cauley, and Todd Sandler describe as the constant interplay between terrorist tactics and innovation, counter-terror response, and the terrorist tactic innovation that follows.[7] Peak years include 2008 with 6 events, 1995 with 5 events, 1982 with 5 events, and 1972 with 4 events. There were several years where no terrorist events in Canada were chronicled; those years include 1971, 1973, 1975-1979, 1981, 1984, 1987-1991, 2001-2003, 2005, 2007.

One recent summary of Canadian counter-terrorist efforts is supplied by the United States Department, “Country Reports on Terrorism”; it reports that Canada continues to work in effective and sustained ways with the United States in jurisprudential and interdiction/ “disruption” spheres. At a functional level, Canadian “judicial” and “disruption” counter-terrorist activities involved the detention of a Canadian with an Iraqi background, a close relative of an Al-Qaeda senior official, two Liberation Tigers of Tamil Eelam (LTTE) activists, a member of the so-called “Toronto18,” and an Al-Shabab recruit who was detained in Lester Pearson Airport in Toronto.[8]

Figure 1

“GTD Search Results” – “Incidents Over Time” “Country: (Canada)”

The relatively small n-set (n=58) for data on Canada from GTD makes it possible to review incidents for certain “peak” years. For terrorist incidents chronicled in 2008, clusters of similarities in locale were noticeable. Indeed, four of the six terrorist assaults in 2008 happened in Dawson Creek, Canada, a “city” of some 10,944 people found in the province of British Columbia’s north-eastern region, some 760 miles from Victoria and close to the border with Alberta.[9] Those terrorist assaults (which might be better termed ‘acts of sabotage’) were carried out against “utilities” and involved the detonation of devices by “unknown” persons with no deaths or injuries reported. The three other incidents include an Animal Liberation Front (ALF) assault against a “business” (i.e. civilian) target in the “town” of Aldergrove, British Columbia, and two terrorist assaults with small numbers of injuries carried out by “unknown” perpetrators against civilian targets in larger “urban” locales, namely in Calgary and in Edmonton, Alberta.

In comparison, terrorist assaults chronicled for 1995 appear to be more evenly dispersed across “urban-rural” distinctions and geographical region of the country, with two attacks by “unknown” assailants against civilian targets in Toronto, one assault in Ottawa against a “government” target, one anti-abortion related attack with one injury in Ancaster, Ontario carried out by the “Army of God,” and one attack in Beaconsfield, Quebec, against a government target that resulted in one death.[10] In turn, the START data chronicles five terrorist assaults in Canada in 1982 where all recorded acts are attributed to three groups, namely “Action Directe (1 act) “Justice Commandos for the Armenian Genocide” (1act ), the “Armenian Secret Army for the Liberation of Armenia” (3 acts).  Moreover, those terrorist assaults happened in two large “urban” areas: Toronto (3 acts) and Ottawa (2 acts). In 1972, “Black September” conducted two terrorist assaults against “government” targets, one in Montreal and the other in Ottawa, while the “Young Cuba”organization carried out one terrorist assault in Ottawa against a civilian target. In addition, there is one chronicled terrorist assault against a government target in Ottawa by “unknown” assailants in 1972.

Specific techniques to carry out attacks seem to cluster in specific time intervals. For example, the use of detonations was most frequently noted in several time intervals. A preference for attacks against “facility/infrastructure” seems to have peaked in 1998 and 2008.  For example, terrorist assaults that utilized “bombing/explosion” tactics were recorded most frequently in 1972 (4 acts), 1980 (3 acts) and 1982 (3 acts), with two detonation acts 1985.  There were three detonation acts in 2008. In the case of “assassination,” peak” years are 1982 (2 acts), 1995 (2 acts), while, by contrast, 2008 was the “peak” year for acts of “armed assault” (2 acts).  In turn, one “hostage taking barricade” terrorist assault happened in 1985. In comparison, the “peak” year for “hostage taking kidnapping” assaults was 1970 with 2 acts which is consistent with the FLQ campaign (1963-1970) passing into eclipse. Attacks against “facility/infrastructure” peaked in 1998 and 2008 with two incidents in each year, while there was one chronicled “unarmed assault” in 1985. One finding that seems significant is there were no aircraft skyjackings recorded for this forty year interval in Canada.

Figure 2

“GTD Search Results” – “Attack Type” “Line Chart” “Country: (Canada)”

In the case of “Attack Type,” data from the” the Global Terrorism Database” inform us that for the forty year period under consideration, “bombing/explosion” (32 events) were the most predominant technique used to conduct terrorist assaults. While the GTD attack type “facility/infrastructure” is a peculiar misnomer (because “facility/infrastructure” is a target-type, rather than a technique, that “attack-type”), it ranked second with 9 events. “Assassination,” which itself is a rather soft designation for “attack type” as it is not mutually exclusive with regards to other “attack type” categories offered, ranks third with 8 events. At the other end of the spectrum, two (2) “hostage taking kidnappings” are reported, while one (1) “hostage taking barricade” incident and one (1) “unarmed assault” are chronicled.

Figure 3

“GTD Search Results” - “Attack Type” “Bar Chart” “Country: (Canada)”

In the case of a breakdown of target-type for the Canadian terrorism experience, data from the GTD illustrate that terrorist assaults against “government (diplomatic)” targets (13 acts) were the single most predominant choice of target over the past forty years. Assaults against “business” targets ranked second with 10 incidents, while what GTD calls “private citizens,” and attacks against “utilities,” both ranked third with 9 attacks each. Conversely, terrorist assaults against “religious figures/institutions,” (1 act) “journalists, media,” (1 act) and “airports, airlines” ranked lowest in terms of relative frequency of events. [11]The lack of distinction between quintessential terrorist attacks targeting explicitly civilians on the one hand and other forms of political violence (e.g. assassinations where the victims is also the intended target – contrary to terrorism where the de-individuated murder of the victim serves to generate a message for the ultimate target) is one of the weaknesses of this type of data gathering. Nevertheless, it offers us a first take on violent politics in any given country.

Figure 4

“GTD Search Results” – “Target Type” “Bar Chart” Country: (Canada)”


United States of America

In the case of the United States, the terrorist assault patterns found for that “long-haul” forty year interval are different in several ways from those illuminated by the START data analysis for Canada. First, the range of terrorist events is substantial and appears to span from nine events (2009) to nearly 480 events (1970). Even though patterns are noticeably different in many respects, such as much larger relative frequency rates by year for example, the patterns established for that forty year “time line” after 1975 and through 2010 appear to have a similar “cyclical” shape of “peaks” and “troughs” that is also the hallmark of the Canadian case and what Im, Cauley, and Sandler describe. [12]From the start, what is noticeable is a perceptible drop in the rate of terrorist assaults from 1970 through 1973. It would be useful to have data before 1970 to determine the overall structural shape of trends for what appears to be a previous cycle. While an authoritative interpretation of those empirical results is not available, it is probably no exaggeration to say the structural shift away from the political instability and social unrest that marked the late 1960’s are reflected in those results. Terrorist groups that were active in the political fray in that early period chronicled by START include, but are not limited to, the “Weather Underground,/Weathermen,” “Armed Revolutionary Independence Movement (MIRA),” Jewish Defense League (JDL), “Students for a Democratic Society (SDS),” “the Puerto Rican Revolutionary Movement,” the “Black Panthers,” and the “Black Liberation Army.”

The peak year for terrorist assaults carried out in the United States for this forty year time interval is 1970, followed by much smaller “peaks” in 1975, and 1977. Corresponding “trough” years include 1972 and 1973 for that early period, while after 1975 “trough” years include 2006, 2004, 2010, 1998, and 1991.

Figure 5

“GTD Search Results” -“Incidents Over Time” “Country: (United States)”

In the case of terrorist assault “attack type,” the START data inform us that as in the case of Canada, acts with “bombing /explosion” methods were the most common type of U.S. attack chronicled for much of the 1970’s. One underlying research issue is whether or not this might be an example of Benjamin Starr and Harvey Most’s “contagion effect” across borders.[13] It would be useful to have or code disaggregated data for event by distance to the border in an ordinal level scale to delve into that question. What seems significant is that from 1970 through about 1972, terrorist assaults that revolved around detonations were predominant and again between 1974 through 1983, when that trend seems to pass into eclipse in favor of “facility/infrastructure” assaults by around 1989. Nonetheless, in 2002, detonation attacks in the U.S. slightly outpaced “facility/infrastructure” terrorist assaults.

Figure 6

“GTD Search Results” - “Attack Type”  “Line Chart” “Country: (United States)”

In turn, when U.S. terrorist assault data for the 1970-2010 time period are broken down based on relative frequency of “attack type,” the START data results reveal that the overwhelming number of terrorist attacks during this time revolved around “bombing/explosions” with 1,197 acts. In comparison,  “facility/infrastructure” assaults comprised 782 acts or 65% of the total number of “bombing explosions” incidents recorded. Conversely, those data results reveal that the least frequent terrorist “attack type” employed during that period was “hostage taking-barricade” with only 9 events, followed by “hostage taking-kidnappings” with 15 chronicled acts.

Figure 7

“GTD Search Results” – “Attack Type” “Bar Chart” “Country: (United States)”


When a relative frequency of “target-type” is produced using the START data for the United States, it is clear that an overwhleming number of terrorist assaults carried out in the U.S. were directed against “business”, with 657 acts. By contrast, the 295 chronicled incidents carried out against “government (general”) targets placed a distanct second, while attacks against “private citizens, property” ranked third with 283 events. At the other extreme, there was one “maritime” terrorist assault chronciled in the START data followed by only 2 attacks against “food or water supply.” What START calls “abortion related” assaults accounted for a full 244 assaults, while 104 terrorist incidents involved “government/diplomatic” targets.[14] Given the fact that the START database is US-based (expanding on the original private detective firm Pinkerton’s database, except for 1993, the year for which Pinkerton data went missing), recording for the United States is more dense than for its northern and southern neighbour which makes comparisons problematic.

Figure 8


“GTD Search Results” – “Target Type” “Bar Chart” “Country: (United States)”



In the case of terrorist assaults carried out in Mexico for the forty year period between 1970-2010, the START data results illuminate patterns that showcase “peaks” and “troughs”and most notably, a significant spike in terrorist events starting around 1992, a trend that culminates with 95 incidents five years later. The range of terrorist assault incidents by year spans from 1 terrorist event in 1972 to 95 terrorist incidents. In 1997, terrorist assaults were conducted by groups that include, but are not limited to, “the Zapatista National Army,” “the Popular Revolutionary Party,” “Institutional Revolutionary Party,” “Popular Revolutionary Party,” and the “Justice Army for the Defenseless Peoples.”[15] Other “peak” years include 1994 with 42 events, and 1978 with 38 events. By contrast, “trough years” include, but are not limited to, 1972 (1 act), 1970 (2 acts), and 1973 (6 acts). In 2012, the U.S. Department of State was unequiocal in its assertion that, “no known international terrorist organization had an operational presence in Mexico and no terrorist group targeted U.S. citizens in or from Mexican territory.”[16] The terror inflicted by Mexican drug cartels on the public, the government and each other’s members, did not count in the calculation of the State Department that still makes an increasingly unrealistic distinction between criminal and political  violence directed against civilians.

Figure 9

“GTD Search Results” - “Incidents Over Time” “Country: (Mexico)”

When the data for Mexico are broken down in the START analysis to reflect “attack type” over time passage, it appears that “bombing/explosions” incidents were predominant in parts of the 1970’s, 1991, 2001, 2008, and 2010, in contrast to “armed assaults” that seem predominant for certain years in the 1990’s, and 2007. At the same time,  “hostage taking-kidnapping” was commonplace to note for certain years in the 1990’s.  It also appears that “assassination” was also used frequently to carry out terrorist assaults and patterns for “assassination” in Mexico closely parallel the pattern for “hostage taking-kidnapping” in Mexico. By contrast,  as with the cases of the United States (5) and Canada (0), “hijackings” were extremely infrequent events.[17] In Monterey, Mexico, one “hijacking” in 1972 was chronicled  in the START data.[18] What seems significant here is a lack of experience with hijackings for all three countries under consideration in ways that resonate with terrorism and counter-terrorist “substitution” dynamics that Im, Cauley, and Sandler isolate and identify, and those are at least consistent with the 9/11 “attack-type” experience.

Figure 10

“GTD Search Results” – “Attack Type” “Line Chart” “Country: (Mexico)”

When the relative frequencies of “attack type” are analyzed and presented by START, it is found that “armed assaults” were the most common place to note in the Mexican terrorism experience with 142 acts, followed by “bombing/explosion” incidents with 89 acts and “hostage taking-kidnapping” acts with 89 acts. Terrorist incidents that involved “assassination” ranked closely behind with 84 acts over that forty year period. In contrast, there was one (1) hijacking event and three (3) “hostage taking-barricade” situations chronicled.

Figure 11

“GTD Search Results” – “Attack Type” “Bar Chart” “Country: (Mexico)”

In the case of a breakdown of target-type for terrorism in Mexico, the START data reveal that the highest number of Mexican terrorist assaults were directed against “business”, with 120 acts, followed closely behind by terrorist assaults against “ private citizens/property” with 102 acts. Terrorist assaults conducted against “police” ranked third with 62 incidents. At the other extreme, as in the case of the the United States, there was only one (1) “maritime” incident chronicled, and only two terrorist assaults against “tourists” recorded in the START database.[19] To continue, there were only five (5) Mexican  terrorist assaults undertaken against “religious figures/institutions” and nine (9) attacks aimed at “transportation” chronicled for the forty year interval covered.

Figure 12

“GTD Search Results”- “Target Type” “Bar Chart” “Country: (Mexico)”


Nowadays, it seems prudent to develop a framework of analysis of the similarities and differences in terrorism trends across the three nation-states that comprise North America. This rather impressionistic, first pass at a review of terrorist assault trends provides a tantalizing glimmer what more full blown empirical studies might reveal. Terrorism trends may be influenced by certain exogenous varibles that are “systemic” or “structural” in nature such as the “global war on terrorism” and country specific endogenous factors that all work with interactive effects across levels. Having said that, the GTD analysis  essentially relies on the use of aggregated data and that severly limits the utility of the analysis as presented. Aside of improvement in certain coding procedures that all too frequently miss the mark in terms of mutually exclusive categories imperative for meaningful analysis, what is needed is disaggregated data that measures changes in frequency rates and terrorist assault “attributes” by region of the country, proximity to borders, and “urban/rural distinctions.” That is important as authorities such  as Siri Aas Rustad and Helga Malmin Binningsbo essentially describe the link between political instability and social unrest and geographic conditions such as heavily forrested locales as well as urban areas that  drive “sanctuary effects” and help illuminate more effective counterterror measures.[20]

At a theoretical level, analysis based on disaggregated data can illuminate whether or not what Starr and Most call “cross border” and “contagion”effect” exist through analysis of terrorist group comparison and terrorist act similarities near borders. In addition, such empirical work could provide insight into whether or not empirical evidence supports or refutes Im, Cauley, and Sandler’s “substution” thesis that account for underlying change in terrorism-counter-terrorist action-reaction dynamics.  In essence, disaggregation of data compels us “to think smaller” in one sense, but a comparison across three countries,that views North America as an integrated operational environmental , compels us “to think larger” about the sources and origins of terrorist assaults and campaigns.[21] After all, 9/11 did not emanate from Canada or Mexico, but that is not to say that new operations might not make use of the set of political, social, and economic interconnections between these three countries. As such, the finding that all three countries have little experience with “maritime” incidents (other than those linked to drug trafficking) might be important.  Eleven years after 9/11, a more reasoned “multifacted approach” to data on acts of violence is called for. The case for a more holistic monitoring effort is getting stronger but bureaucratic compartmentalisation is often standing in the way. To cross such boundaries, academia should integrate databases combining political and criminal violent actor data as well as those based on monitoring official and inofficial government uses of force and some of the more muscular activities of  private security firms.


About the Author: Richard J. Chasdi is an Adjunct Assistant Professor at the Center for Peace and Conflict Studies at Wayne State University and an Adjunct Associate Professor of Political Science at the University of Windsor. He is also a Member of the Editorial Board of ‘Perspectives on Terrorism’.



[1] Ian Austen. 2012. “2 Shot at Election Party in Quebec: Assassination Try Suspected.” New York Times, 6 September, A-11.

[2] Harvey Starr and Benjamin Most. 1976. “The Substance and Study of Borders in International Relations Research.” International Studies Quarterly 20 (4): 581-620; Harvey Starr and Benjamin Most. 1983. “Contagion and Border Effects on Contemporary African Conflict.”Comparative Political Studies 16(1): 92-117. This conceptualization presupposes and derives from a grant proposal for a comparative study of the U.S. and Canada submitted to the Canadian Studies Grant Programs written by Dr. Frederic S. Pearson and Dr. Richard J. Chasdi in 1996 entitled, “Terrorism in North America: An Empirical Cross Country Comparison of Canadian and American Terrorist Group-Type and Group Behavior.”

[3] Cf. the work of Louise I. Shelley.

[4] Richard J. Chasdi. 2010. Counterterror Offensives for the Ghost War World: The Rudiments of Counterterrorism Policy, Lanham, MD; Lexington Books,

[5] Thomas E. Patterson. 2000. We the People: A Concise Introduction to American Politics. Boston: McGraw Hill, 57-85, A-65.

[6] Those “Global Terrorism Database (GTD)” reports produced by START were sorted out by searches for “country” and subsequently by “attack-type” and “target-type,” for those specific countries. National Consortium for the Study of Terrorism and Responses to Terrorism START. 2011. Global Terrorism Database (http://www.start.umd.edu/gtd); United States Department of State 2012 “Country Reports on Terrorism 2011” (http:www.state.gov/documents/organization/195768.pdf).

[7] Eric Iksoon Im, Jon Cauley, and Todd Sandler. 1987. “Cycles and Substitutions in Terrorist Activities: A Spectral Approach.”  Kyklos 40: 238-255; Walter Enders, Gerald Parise, and Todd Sandler. 1992. “A Time Series Analysis of Transnational Terrorism: Trends and Cycles.” Defense Economics 3: 305-320; Richard J. Chasdi. 2002. Tapestry of Terror: A Portrait of Middle East Terrorism, 1994-1999. Lanham, MD: Lexington Books, 53-54.

[8] U.S. Department of State. 2012. “Country Reports on Terrorism 2011” – Canada, 153-156; Cf. the work of Jeffrey Ian Ross.

[9] START’s database for Canada is N= 58. “Dawson Creek Tourism BC- Official Site” (www.hellobc.com/dawson-creek.aspx); “Dawson Creek, British Columbia” (www.city-data.com/canada/Dawson-Creek.html); GTD Global Terrorism Database “Search Results”- Canada. Indeed, two out of three terrorist assaults chronicled for 2009 happened in Dawson Creek and in nearby Pouce Coupe.

[10] GTD Global Terrorism Database “Search Results” - Canada

[11] Plainly, both of those categories, “private citizens” and “utilities” are subsets of broader “civilian” targets as are other GTD target categories such as “journalists, media,” “educational institution,” and “business.”

[12] Im, Cauley, and Sandler 1987, 238-255; Enders, Parise, and Sandler 1992, 305-320; GTD Database “Search Results” – United States. In the case of START’s United States database, N= 2347.

[13] Starr and Most 1976, 581-620; Starr and Most 1983, 92-117. – For a documentation of  contagion effects from Canada to the United States and beyond in the case of so-called parachute hijackings, following the 12 November 1971 hijacking of an Air Canada DC-9 en route from Calgary to Toronto, see Alex P. Schmid and Janny de Graaf. Violence as Communication. Insurgent Terrorism and the Western News Media. London: Sage, 1982, pp.133-136. Schmid & De Graaf documented 26 copycat parachute hijackings in a six year period -  21 of them in the first year after this new technique of extortion was invented. In each case the hijacker(s), after collecting ransom money, tried or intended to escape by parachuting in mid-flight from the hijacked plane.

[14] Ibid.

[15] GTD Database “Search Results” – Mexico. In the case of START’s Mexico database, N=444.

[16] U.S. Department of State 2012. “Country Reports on Terrorism 2011” – Mexico, 162-163.

[17] Indeed,  “hijackings” were extremely rare terrorist assault occurrences– a hand count of GTD Database “Search Results” - United States included: – (1) a “left-wing militant”event# 19700314004 “Date: 1970-03-14”; (2) a “JDL” event# 197009270001 “Date: 1970-09-27”; (3) a “Black Panthers” event#  19720602002 “Date: 1972-06-02”; (4) a “Blank Panthers” event# 197207310001 “Date: 1972-07-31”; (5) a Croatian Freedom Fighters event# 1976091100005 “Date: 1976-09-10.”

[18] GTD Database “Search Results – Mexico - “Unknown” event# 197211080002 “Date: 1972-11-08.”

[19] In the START data, there are no “maritime” terrorist assaults chronicled for Canada.

[20] Siri Aas Rustad and Helga Malmin Binningsbo. 2012. “A price worth fighting for? Natural resources and conflict recurrence.” Journal of Peace Research, 49 531-546 (http://jpr.sagepub.com/cgi/content/abstract/49/4/531 ).

[21] Etzioni, Amitai. 2004. From Empire to Community: A New Approach to International Relations. New York: Palgrave Macmillan Ltd; Follett, Mary Parker 1918. The New State Group Organization the Solution of Popular Governance. New York: Longmans, Green and Company, 250.

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