START GTD GlobalTerrorismOverview2019 July2020
START GTD GlobalTerrorismOverview2019 July2020
OVERVIEW
In 2019, there were nearly 8,500 terrorist attacks around the world, which killed more than 20,300 people, including 5,460
perpetrators and 14,840 victims. The patterns described here provide a general overview. We encourage readers to explore the
Global Terrorism DatabaseTM and consider contextual information for a comprehensive assessment.1
▪ 2019 was the fifth consecutive year of declining global terrorism since terrorist violence peaked in 2014 at nearly
17,000 attacks and more than 44,000 total deaths. The total number of terrorist attacks worldwide decreased 50%
between 2014 and 2019, and the total number of deaths decreased 54%. Regional trends varied substantially.
▪ While the Taliban engaged in peace talks with the United States, Afghanistan experienced 21% of all terrorist attacks
worldwide in 2019, and 41% of all people killed in terrorist attacks (including assailants) in 2019 were killed in
Afghanistan.2
- The number of terrorist attacks in Afghanistan increased 2% between 2018 and 2019.
- The total number of deaths due to terrorism in Afghanistan decreased 16%; however, this change was largely a
result of the 32% decline in the number of assailants killed during attacks, which is typically especially high in
Afghanistan (38% of all deaths in 2019). The number of victims killed in terrorist attacks in Afghanistan
declined 2% and the number of victims killed in attacks carried out by the Taliban increased 9% between 2018
and 2019, essentially sustaining the escalation that has taken place in Afghanistan in recent years.
1 Please see the Methodological Note at the end of this report for important information about the potential impact of changes
in access to translated source materials.
2 Maizland, L. (2020, March 2). U.S.-Taliban Peace Deal: What to Know. Council on Foreign Relations. Retrieved from
https://www.cfr.org/backgrounder/us-taliban-peace-deal-agreement-afghanistan-war
START Background Report © University of Maryland, July 2020 1
TERRORIST ATTACKS AND TOTAL DEATHS WORLDWIDE, BY MONTH, 2012 – 2019
7000
6000
5000
4000
3000
2000
1000
3 Consistent with START’s practice of including in the GTD only those attacks that have been reported by at least one high-
validity source, these statistics represent those incidents that were reported by independent news outlets. These data provide
conservative estimates of terrorism in locations where unbiased media coverage is limited, such as Syria and Yemen.
4 Note that two of the nine mass casualty attacks in Western Europe were marked as “doubt terrorism proper” in the GTD,
meaning there is conflicting information about whether the definitional criteria were satisfied.
START Background Report © University of Maryland, July 2020 2
explosives, one involved incendiary weapons, and one was an unarmed assault in which four people were
injured.
▪ In 2019, there were 64 terrorist attacks, killing 51 people in the United States. Terrorism in the United States continued
to be characterized by diverse, sometimes complex, and often ambiguous ideological influences, typically without clear
ties to formal, named organizations.
- Perpetrators of the 10 lethal terrorist attacks in the United States in 2019 included white
supremacists/nationalists, anti-Semitic extremists linked to the Black Hebrew Israelite movement, an al-Qaida
in the Arabian Peninsula operative, and a conspiracy theory extremist.5
- However, the vast majority of terrorist attacks in the United States in 2019 were non-lethal (84%, excluding
perpetrator deaths), and these attacks were also motivated by diverse ideological influences, including anti-
fascist, anti-government, anti-LGBT, anti-Muslim, anti-Semitic, anti-white, left-wing, pro-choice, and white
supremacist/nationalist extremism.
▪ Between 2015 and 2019, 286 people were killed in terrorist attacks in the United States (excluding assailants). Nearly
all of the victims (95%) were killed in attacks involving firearms. Firearms were used in 27% of terrorist attacks in the
United States during this time period.
▪ Exceptionally deadly attacks targeting Hispanic Americans in the United States and Muslims in New Zealand in 2019
marked a sharp increase in the lethality of “racially and ethnically motivated terrorist attacks” (REMT), many of which
were motivated by white supremacy, xenophobia, and anti-immigrant beliefs. At least 86 people were killed in such
attacks in Australasia, North America, and Western Europe in 2019, compared to 52 in 2018.
▪ The deadliest terrorist attacks in 2019 took place on Easter Sunday in Sri Lanka. More than 250 people were killed and
at least 500 others were injured when eight assailants carried out suicide bombings at seven different crowded
locations, including hotels and churches. A pipe bomb was defused at an eighth location. Sri Lankan authorities
indicated that National Thowheeth Jama'ath and Jammiyathul Millathu Ibrahim operatives were responsible for the
attacks. However, Islamic State reportedly posted a video purportedly showing the assailants pledge allegiance to the
group, and reports indicate that one or more of the assailants trained with Islamic State.6
Terrorist violence remained heavily concentrated in certain locations and coincided with other types of political violence. More
than half of all attacks took place in five countries: Afghanistan (21%), Yemen (9%), Iraq (8%), India (7%), and Nigeria (6%). As in
2018, half of all deaths due to terrorist attacks in 2019 took place in two countries: Afghanistan (41%), and Nigeria (8%).
TERRORIST ATTACKS AND TOTAL DEATHS, COUNTRIES WITH MORE THAN 150 ATTACKS, 2019
Total % of % Change Total % of % Change
Country Attacks Total from 2018 Killed* Total from 2018
Afghanistan 1804 21% 2% 8249 41% -16%
Yemen 771 9% 55% 1219 6% 8%
Iraq 642 8% -53% 798 4% -44%
India 623 7% -30% 311 2% -25%
Nigeria 507 6% -22% 1718 8% -33%
Philippines 464 5% -23% 396 2% -10%
Pakistan 362 4% -25% 416 2% -40%
Somalia 360 4% -32% 860 4% -25%
Syria 292 3% 25% 1102 5% -29%
Democratic Republic of the Congo 250 3% 53% 642 3% -35%
Colombia 228 3% 11% 134 1% 2%
Nepal 200 2% 102% 7 0% --
Burkina Faso 159 2% 127% 797 4% 587%
Worldwide Total 8473 100% -14% 20309 100% -13%
*Includes perpetrator deaths
Source: Global Terrorism Database
5 Note that three of the ten lethal attacks in the United States were marked as “doubt terrorism proper” in the GTD, meaning
there is conflicting information about whether the definitional criteria were satisfied.
6 Mandhana, N., Taylor, R., and Shah, S. (2019, April 29). Sri Lanka Bomber Trained in Syria with Islamic State. Wall Street
PERPETRATORS
Information on the perpetrator of the attack was reported for 66% of all attacks worldwide in 2019. In 59 attacks, the individual
perpetrator or perpetrators were identified, but they were not known to be affiliated with a particular group or organization. This
represents a decline in the number of attacks carried out by unaffiliated individuals, which peaked at 102 in 2017. The lethality
of attacks carried out by unaffiliated individuals also decreased, from more than 200 victims killed in 2016 to 98 in 2019.
PERPETRATOR GROUPS RESPONSIBLE FOR MORE THAN 100 TERRORIST ATTACKS, 2019
Total Change Total Change
Perpetrator Group Attacks from 2018 Killed* from 2018
Taliban 1375 9% 7531 -11%
Houthi extremists (Ansar Allah) 579 55% 983 11%
Islamic State of Iraq and the Levant (ISIL) 461 -37% 1252 -44%
Boko Haram 348 43% 1954 47%
Al-Shabaab 330 -33% 864 -25%
Maoists/Communist Party of India - Maoist (CPI-Maoist) 238 -11% 146 -23%
New People's Army (NPA) 192 -32% 155 -18%
Communist Party of Nepal - Maoist (CPN-Maoist-Chand) 134 148% 7 --
Fulani extremists 118 -61% 426 -64%
*Includes perpetrator deaths
Source: Global Terrorism Database
7Inside Cameroon’s 100-year old Anglophone conflict. (2019, October 1). Agence France Presse. Retrieved from
https://www.africanews.com/2019/10/01/inside-cameroon-s-100-year-old-anglophone-conflict/
START Background Report © University of Maryland, July 2020 4
▪ In 2019, 265 groups and organizations carried out terrorist attacks worldwide. This number has been steadily declining in
recent years since it peaked at 377 groups and organizations identified as perpetrators of terrorist attacks in 2016. The
perpetrator groups responsible for the most attacks in 2019 are shown in the table above.
▪ The Taliban in Afghanistan was responsible for more terrorist attacks in 2019 than any other group by a wide margin, and
those attacks resulted in more deaths than the next ten deadliest perpetrator groups combined. Between 2018 and 2019,
the number of terrorist attacks carried out by the Taliban increased 9% and the total number of deaths decreased 11%.
However, it is important to note that attacks carried out by the Taliban often involve many perpetrator deaths. In 2019 the
number of perpetrator deaths in Taliban attacks decreased 31% and the number of victim deaths increased 9%.
▪ Aside from the Taliban, several of the perpetrator groups that increased terrorist violence in 2019 included Houthi
extremists in Yemen (55% increase in attacks; 11% increase in total deaths; 54% increase in victim deaths), Boko Haram
(43% increase in attacks; 47% increase in total deaths; 37% increase in victim deaths); and the Communist Party of Nepal
– Maoist – Chand (148% increase in total attacks; seven people were killed in 2019 including one perpetrator, up from zero
in 2018).
▪ Islamic State and certain affiliated groups were among those whose terrorist violence continued to decrease in 2019.
These include Islamic State of Iraq and the Levant (attacks decreased 37% and deaths decreased 44%) and Islamic State-
Khorasan Province (attacks decreased 52% and deaths decreased 68%). However, this network of organizations remained
very deadly and their attacks resulted in thousands of casualties in locations around the world. Attacks by Islamic State
operatives in Iraq and Syria killed more than 1,000 victims in 2019. The group also claimed responsibility for deadly attacks
in Lebanon, the Philippines, Sri Lanka, Tajikistan, and Tunisia.
▪ Boko Haram increased terrorist violence in 2019, not only in Nigeria, but also in neighboring countries Cameroon, Chad,
and Niger. Specifically, attacks attributed to Boko Haram resulted in 241 total deaths in Cameroon (up from 76 in 2018),
189 in Chad (up from 106 in 2018), and 315 in Niger (up from 50 in 2018). In addition, Boko Haram (Barnawi faction,
Islamic State West Africa Province) claimed responsibility for an attack in Burkina Faso in 2019, in which assailants killed
24 soldiers.
▪ In 2019 there was a sharp increase in the lethality of terrorist attacks that were racially or ethnically motivated (REMT). This
increase was largely a result of exceptionally deadly attacks targeting Hispanic Americans in the United States and Muslims
in New Zealand. The table below illustrates recent patterns in REMT attacks, many of which were motivated by white
supremacy, xenophobia, and anti-immigrant beliefs, and which took place in 23 different countries between 2015 and
2019.
TRENDS OVER TIME IN RACIALLY AND ETHNICALLY MOTIVATED ATTACKS, 2015 -- 2019
Total Total Perpetrators Total Perpetrators
Year Attacks Killed Killed Injured Injured
2015 34 26 2 22 0
2016 42 18 2 64 1
2017 84 16 1 51 1
2018 33 54 2 79 4
2019 54 89 3 146 2
Total 247 203 10 362 8
Source: Global Terrorism Database
▪ In recent years, perpetrators of terrorism in the United States have been very loosely organized, rarely affiliated with a
formal organization, and motivated by a diverse array of ideologies. These ideologies often overlap, to the point that it can
be difficult to classify the identity of the perpetrator or perpetrators in a straightforward way. The table below lists the
identity of perpetrator “groups” that were active in the United States between 2015 and 2019. Note that many of these
attributions are generic identifiers, not meant to exhaustively capture all ideological characteristics of the assailant(s), but
typically based on the ideological influences most relevant to the specific attack in question.
Perpetrators
Perpetrator Group Name Total Attacks Total Killed Killed
Al-Qaida in the Arabian Peninsula (AQAP) 1 4 1
Anarchists 1 0 0
Animal Liberation Front (ALF) 1 0 0
Anti-Abortion extremists 7 3 0
Anti-Arab extremists 1 0 0
Anti-Fascist extremists 1 1 1
Anti-Government extremists 7 64 3
Anti-Immigrant extremists 1 0 0
Anti-LGBT extremists 6 0 0
Anti-Muslim extremists 32 5 0
Anti-Police extremists 7 11 3
Anti-Republican extremists 2 1 1
Anti-Semitic extremists 16 17 2
Anti-Sikh extremists 1 0 0
Anti-Trump extremists 4 0 0
Anti-White extremists 12 10 1
Black Hebrew Israelites 3 0 0
Citizens for Constitutional Freedom 1 0 0
Conspiracy theory extremists 3 1 0
Court Reform extremists 1 0 0
Environmentalists 3 0 0
Incel extremists 2 13 2
Jihadi-inspired extremists 23 86 8
Ku Klux Klan 3 0 0
Left-wing extremists 1 0 0
Male supremacists 3 0 0
Muslim extremists 4 7 2
Neo-Nazi extremists 4 12 1
Pro-choice extremists 2 0 0
Pro-LGBT Rights extremists 1 0 0
Pro-Trump extremists 16 0 0
Right-wing extremists 2 0 0
Shield Wall Network (SWN) 1 0 0
Sovereign Citizen 1 0 0
United Constitutional Patriots 1 0 0
Unknown 99 17 3
White Rabbit Three Percent Illinois Patriot Freedom
2 0 0
Fighters Militia
White supremacists/nationalists 34 64 2
Total 310 316 30
Source: Global Terrorism Database
▪ When considering perpetrator group patterns, it is important to note how groups and organizations evolve over time, often
breaking into factions and splinters, sharing members, changing names and aliases, locations and personnel, and forming
mergers, alliances, and “franchises.” As a result, groups are a problematic unit of analysis that can lead to shortsighted
inferences. Combining perpetrator organizations, groups, and individuals with shared goals into “movements” helps capture
40
30
20
10
60
40
30
20
10
8Miller, E. (2016, August). Patterns of Islamic State-Related Terrorism. START Background Report. Retrieved from
https://www.start.umd.edu/pubs/START_IslamicStateTerrorismPatterns_BackgroundReport_Aug2016.pdf
START Background Report © University of Maryland, July 2020 7
ADDENDUM: METHODOLOGICAL NOTE REGARDING NON-ENGLISH SOURCES
Understanding the importance of including multi-lingual sources in the data collection process, the GTD team typically
supplements the English-language content in the LexisNexis Metabase news feed with content from Open Source Enterprise
(OSE). Although the GTD research team does include analysts with non-English language skills, at this time the team does not
have resources to translate content at scale or conduct artificial intelligence processing on non-English language source
documents. OSE—a U.S. government media aggregation and translation service originally established in 1941—was formerly
known as Open Source Center, World News Connection, and the Foreign Broadcast Information Service. This product has been
used in the GTD data collection effort for decades and, although it is impossible to know what percentage of GTD records would
not exist without OSE sources, approximately 15-25% of all records in the GTD leverage information from OSE or its
predecessors in some way. Despite the fact that the translated content on the OSE platform was “open source” content curated
from publicly available news feeds around the world, access to OSE was not available to the general public. The GTD team
accessed the OSE platform at no cost by virtue of sponsored research agreements with U.S. government agencies.
On June 3, 2019 the GTD team learned that the OSE platform would be decommissioned on June 28. The GTD team later
learned that the open source media translation effort and the content itself were apparently not decommissioned, but that the
platform (now known as Open Source Data Layer and Services (OSDLS)) would no longer be available to those who are not U.S.
government employees with Joint Worldwide Intelligence Communications System (JWICS) credentials. According to an OSE
customer service representative, the reason for this shift, which was clearly detrimental to open source research efforts, was to
improve the security of access to the platform as it was moved to cloud servers.
GTD staff worked with U.S. government and private sector partners to attempt to understand the implications of this decision for
the research team’s access to the critical translated content needed to promote comprehensiveness and consistency in the
GTD collection process. Ultimately, efforts to make decisionmakers aware of the impact of this change and/or identify a
pathway to accessing the content on OSDLS were not successful in reversing or mitigating the policy change.
Concurrently, the team began exploring contingency plans, the most promising of which by far was purchasing subscription
access to BBC Monitoring. BBC Monitoring is a media aggregation and translation platform with which the GTD team has long
been familiar. In fact, for many decades, BBC Monitoring worked in close partnership with OSE, and the GTD research team has
had access to the BBC Monitoring content that was included in the OSE platform and LexisNexis Academic. Although BBC
Monitoring is the only resource remotely comparable to OSE, little information is available about the scope of unique or
overlapping translated content across the two platforms. In December 2019 the GTD team gained subscription access to the
BBC Monitoring platform and began to incorporate this content into the data collection workflow.
The team has since completed the backlog of triaging for the 2019 BBC Monitoring source documents and reviewed the data
with the aim of providing an assessment of the likely impact that the loss of OSE and replacement with BBC Monitoring had on
the resulting data. The initial data on the total count of source documents provided by BBC Monitoring by month in 2019,
compared to the same month in 2018, are instructive.
The table above indicates that the total amount of content accessed through the BBC Monitoring subscription in 2019 was
much smaller than the total amount of content accessed through the OSE platform in 2018. This difference is evident for the
source articles that match the GTD’s initial keyword filters, and for the source articles that are classified as potentially relevant
by the machine learning model (candidates for triaging). Although the decrease in supplemental content is stark, this
information is not sufficient to illustrate the degree and nature of its impact. Which source documents were ultimately used to
support GTD records? Which source documents from OSE were still accessible in either BBC Monitoring or LexisNexis
Metabase? How do these deficits differ by language or location? How frequently might particular events still be recorded in the
database even if certain sources or source articles were lost? Ultimately, although BBC Monitoring presents the best opportunity
to supplement source documents following the loss of OSE content, concerns about the consistency and comprehensiveness of
data without OSE content remain.
Likewise, the table below shows the number of cases either added to the GTD or updated with new information based on the
triaging of the BBC Monitoring content. Although OSE was decommissioned in early July, the GTD research team triaged BBC
We have completed an initial review of the 2019 data with the intention of identifying any specific locations that may be
particularly impacted by this change in access to translations of non-English source documents. Although there do not appear to
be any countries where data trends clearly coincide with the shift from OSE content to BBCM content, that does not rule out the
possibility that this change suppressed otherwise increasing trends or exacerbated already decreasing trends in certain
locations. In general, we urge analysts to interpret trends over time with caution due to ongoing fluctuations in the availability of
source documents. Based on our familiarity with the source documents formerly provided by OSE, the research team urges
analysts to pay particular attention to the potential impacts on trends in Somalia and Yemen. Additional analysis of regional
trends, trends in types of attacks, or trends in supporting sources may provide greater insight.
Readers familiar with the U.S. State Department’s Country Reports on Terrorism may recall that START provided the “Statistical
Annex” for that report each year from 2012 through 2017. For consistency with the definition of terrorism established in the
U.S. Code, we used a restricted application of the GTD’s inclusion criteria for the analysis in the Statistical Annex.
Despite a productive partnership with our colleagues at the Bureau of Counterterrorism and Countering Violent Extremism, in
2018 the State Department did not award START a contract to continue providing data and analysis for the Statistical Annex.
We recognize that any data or analysis the Statistical Annex includes for 2018 and 2019 is not consistent or comparable with
previous years’ data and analysis produced by START. To support analytical continuity, we have re-produced several key tables
using the restricted Statistical Annex version of the GTD. Specifically, the following tables exclude any attacks that do not meet
all three GTD inclusion criteria, and any attacks that were classified as “doubt terrorism proper” by GTD analysts.
Table 1: Terrorist attacks and casualties worldwide by month, 2019
Kidnapped/
Month Total Attacks* Total Deaths* Total Injured* Hostages
January 540 1284 1388 486
February 568 1108 917 700
March 636 1123 1269 503
April 582 1268 1625 489
May 740 1366 1238 358
June 599 1386 1047 387
July 625 1444 1938 225
August 497 1020 1181 281
September 542 1104 1266 207
October 509 983 912 458
November 393 716 814 164
December 491 1020 947 406
Total 6722 13822 14542 4664
*Includes perpetrators
Source: Global Terrorism Database (Statistical Annex Dataset)
Houthi extremists (Ansar Allah) 446 277 436 258 609 260 394 410
Islamic State of Iraq and the Levant (ISIL)** 343 560 954 1601 1319 1423 41 970
Boko Haram 245 171 1069 732 378 589 304 376
Maoists/Communist Party of India - Maoist
226 247 136 175 87 154 94 135
(CPI-Maoist)
* Includes perpetrators
** Excludes attacks attributed to branches of ISIS or ISIS-inspired individuals
Source: Global Terrorism Database (Statistical Annex Dataset)
Erin Miller is the primary author of this report. Please direct questions to
[email protected].
The data presented here are drawn from START’s Global Terrorism Database (GTD) and reports from news media. The GTD
contains information on more than 190,000 terrorist attacks that occurred around the world since 1970. For more
information about the GTD, visit www.start.umd.edu/gtd.
The National Consortium for the Study of Terrorism and Responses to Terrorism (START) is a university-based research,
education and training center comprised of an international network of scholars committed to the scientific study of terrorism,
responses to terrorism and related phenomena. Led by the University of Maryland, START is a Department of Homeland
Security Emeritus Center of Excellence that is supported by multiple federal agencies and departments. START uses state-of-
the-art theories, methods and data from the social and behavioral sciences to improve understanding of the origins, dynamics
and effects of terrorism; the effectiveness and impacts of counterterrorism and CVE; and other matters of global and national
security. For more information, visit www.start.umd.edu or contact START at [email protected].