Section 2656f(b) of Title 22 of the U.S. Code requires the State Department to include in its annual report on terrorism “to the extent practicable, complete statistical information on the number of individuals, including United States citizens and dual nationals, killed, injured, or kidnapped by each terrorist group during the preceding calendar year.” While NCTC keeps statistics on the annual number of incidents of “terrorism,” its ability to track the specific groups responsible for each incident involving killings, kidnappings, and injuries is significantly limited by the availability of reliable open source information, particularly for events involving small numbers of casualties. Moreover, specific details about victims, damage, perpetrators, and other incident elements are frequently not fully reported in open source information.

• The statistical material in this report, therefore, is drawn from the incidents of “terrorism” that occurred in 2007 as reported in open sources information, which is the most comprehensive body of information available to NCTC for compiling data that it can provide to satisfy the above-referenced statistical requirements.

In deriving its figures for incidents of terrorism, NCTC in 2005 adopted the definition of “terrorism” that appears in the 22 U.S.C. § 2656f(d)(2), i.e., “premeditated, politically motivated violence perpetrated against noncombatant targets by subnational groups or clandestine agents.”

• Through 2004 NCTC compiled statistical data on the basis of a more limited methodology tied to the definition of “international terrorism,” which is also contained in 22 U.S.C. § 2656f.
• Because of the change in methodology during 2004, the NCTC data is only comparable starting with the 2005 calendar year data.
• Subject to changes in reporting statutes, NCTC anticipates that future statistics provided by it will continue to be tied to the broader definition of “terrorism.”

To record and update attack records, NCTC has continued to post information in the repository for the U.S. Government's database on terror attacks, the Worldwide Incidents Tracking System (WITS) that was unveiled in 2005. A data management system with a more comprehensive dataset than those used in years prior to 2004, WITS is accessible on the NCTC website at www.nctc.gov for the public to have an open and transparent view of the NCTC data. NCTC will insure that the data posted to the website is updated as often as necessary by regularly posting information about new or prior attacks.

Considerations for Interpreting the Data

NCTC cautions against placing too much emphasis on any single set of attack data to gauge success or failure against the forces of terrorism. Furthermore, NCTC does not believe that a simple comparison of the total number of attacks from year to year provides a meaningful measure.

• Tallying attack data necessarily involves relying exclusively on frequently incomplete and ambiguous information—information for these statistics is not derived from federal government collection programs created or operated specifically to obtain attack data. The quality, accuracy, and volume of open source reporting can vary greatly from country to country. As a result, determining whether an attack meets the NCTC criteria for a terror attack is often difficult and highly subjective. This is particularly true if the attack does not involve mass casualties because little information is typically available on these incidents that usually receive less media coverage. Furthermore, in parts of the world where there is little press coverage and little non-governmental organization presence, terror attacks often go unreported.
• Attack tallies exclusively do not provide a complete picture of the magnitude or seriousness of the terrorism challenge confronting a country or region. Moreover, different factors weigh more heavily than others in assessing the dangers posed by terrorism. For example, an attack that kills 100 civilians is likely to be considered more alarming than an attack that damages a pipeline but harms no one; however, each attack is simply tallied as one incident. We also note that 50 percent of the attacks in the NCTC database involve no loss of life.
• Counting protocols matter and inevitably require judgment calls that may have an impact on results. For example, NCTC protocols dictate that events identified as simultaneous and coordinated would be recorded as one attack, as would be secondary attacks that targeted first-responders. For instance, on the morning of August 17, 2005, there were approximately 450 small bomb attacks in Bangladesh, and because they were coordinated according to a central plan, NCTC counted them as a single event. Other valid counting protocols would register these attacks as 450 separate attacks.
• Analyzing attack data from year-to-year to identify patterns and notable deviations or trends in the data is problematic, and may not be meaningful in some cases. The availability, quality, and depth of open source reporting vary making it hard to isolate whether a rise or fall of a particular data element from one year to the next is due to an increase or decrease of this open source reporting or whether actual events are behind the change in the data.

Despite these limitations, tracking and analyzing incidents can help us understand some important characteristics about terrorism, including the geographic distribution of attacks and information about the perpetrators, their victims, and other details. Year-to-year changes in the gross number of attacks across the globe, however, may tell us little about the international communities’ effectiveness either for preventing these incidents, or for reducing the capacity of terrorists to advance their agenda through violence against the innocent.

Methodology Utilized to Compile NCTC’s Database of Attacks

Starting with the 2005 results, NCTC, working with a panel of terrorism experts, adopted a revised methodology for counting terrorist incidents, basing it on the broader statutory definition of “terrorism” rather than that of “international terrorism,” on which the NCTC based its incident counting in years prior to 2004. For 2007, we continued using this broader definition of “terrorism” and overall this broader definition and improvements in cataloging have resulted in a larger, more comprehensive set of attack data, all of which can now be found on NCTC's website, www.nctc.gov.

The data provided on the website is based on the statutory definition set forth in the Developing Statistical Information section to this Annex. Accordingly, the attacks NCTC has catalogued in the database are those that, based on available open source information, meet the criteria for “premeditated, politically motivated violence perpetrated against noncombatant targets by subnational groups or clandestine agents.” Determination of what constitutes a terror attack, however, is sometimes based on incomplete information and may be open to interpretation. The perpetrator's specific motivation, whether political or otherwise, is not always clear, nor is the perpetrator’s identity always evident. Moreover, additional information may become available over time affecting the accuracy of initial judgments about attacks. Users of this database should therefore recognize that expert opinions may differ on whether a particular attack constitutes terrorism or some other form of political violence.

NCTC has made every effort to limit the degree of subjectivity involved in the judgments. In the interests of transparency NCTC has adopted counting rules that require that terrorists must have initiated and executed the attack for it to be included in the database; foiled attacks, as well as hoaxes, are not included in the database. Spontaneous (i.e., non-premeditated) hate crimes without intent to cause mass casualties are excluded to the greatest extent practicable.

What is a “noncombatant”?

Under the statutory definition of terrorism that NCTC uses to compile its database, the victim must be a “noncombatant.” However, that term is left open to interpretation by the statute. For the purposes of the WITS database, the term “combatant” was interpreted to mean military, paramilitary, militia, and police under military command and control, in specific areas or regions where war zones or war-like settings exist. Further distinctions were drawn depending on the particular country involved and the role played by the military and police, e.g., where national security forces are indistinguishable from police and/or military forces. Noncombatants therefore included civilians and civilian police and military assets outside of war zones and war-like settings. Diplomatic assets, including personnel, embassies, consulates, and other facilities, were also considered noncombatant targets.

Although only acts of violence against noncombatant targets were counted as terror attacks for purposes of the WITS database, if those incidents also resulted in the death of combatant victims, all victims (combatant and noncombatant) were tallied. In an attack where combatants were the target of the event, non-combatants who were incidentally harmed were designated “collateral” and the incident excluded from the posted data set. For example, if terrorists attacked a military base in Iraq and wounded one civilian bystander, that victim would be deemed collateral, and the attack would not be counted.

In the cases of Iraq and Afghanistan, it is particularly difficult to gather comprehensive information about all incidents and to distinguish terrorism from the numerous other forms of violence, including crime and sectarian violence, in light of imperfect information. The distinction between terrorism and insurgency in Iraq is especially challenging as terrorist groups target combatants and non-combatants and also engage in tribal and sectarian violence. Therefore, some combatants may be included as victims in some attacks when their presence was incidental to an attack intended for noncombatants. We note, however, that because of the difficulty in gathering data on Iraq and Afghanistan, the dataset does not provide a comprehensive account of all attacks of terrorism in these two countries.

What is “politically motivated violence”?

The statutory definition also requires the attack to be “politically motivated.” NCTC has adopted a series of counting rules to assist in the data compilation. Any life threatening attack or kidnapping by any “Foreign Terrorist Organization” or any group designated by other authorities is deemed politically motivated. Similarly, any serious attack by any organization or individual against a Government/Diplomatic official or a Government/Diplomatic building is deemed politically motivated and is therefore considered terrorism. On the other hand, any attack that is primarily criminal or economic in nature or is an instance of mob violence is considered not to be “politically motivated.” Similarly, any terrorist organization actions that are primarily intended to enable future terrorist attacks (robbing a bank or selling narcotics for the purpose of raising money, for example) are not considered terrorism.

In between these relatively clear-cut cases, there is a degree of subjectivity. In general, NCTC counting rules consider that attacks by unknown perpetrators against either unknown victims or infrastructure are not demonstrably political and therefore are not terrorism. However, there are exceptions to this general rule: if such an attack occurs in areas in which there is significant insurgency, unrest, or political instability, the attack may be considered terrorism; or if the attack occurs in a region free of such political violence, but involves something more than a shooting (for instance, improvised explosive device, beheading, etc.), the attack may, depending on the circumstances, be considered terrorism. Finally, if low level attacks against noncombatant targets begin to suggest the existence of a chronic problem, the attacks may be considered terrorism.

Incidents of Terrorism Worldwide*

	2005	2006	2007
Terror attacks worldwide	11,156	14,570	14,499
Attacks resulting in at least one death, injury, or kidnapping	8,032	11,322	11,125
Attacks resulting in at least one death	5,137	7,434	7,258
Attacks resulting in the death of zero people	6,019	7,136	7,241
Attacks resulting in the death of only one person	2,882	4,142	3,993
Attacks resulting in the death of at least 10 people	227	295	355
Attacks resulting in the injury of at least one person	3,839	5,798	6,259
Attacks resulting in the kidnapping of at least one person	1,154	1,349	1,158
People killed, injured or kidnapped as a result of terror attacks	74,309	75,211	72,066
People worldwide killed as a result of terror attacks	14,616	20,872	22,685
People worldwide injured as a result of terror attacks	24,853	38,455	44,310
People worldwide kidnapped as a result of terror attacks	34,840	15,884	5,071

Incidents of Terrorism in Iraq and Afghanistan*

	2005	2006	2007
Terror attacks in Iraq	3,469	6,628	6,212
Attacks resulting in at least one death, injury, or kidnapping	2,837	6,026	5,574
People killed, injured, or kidnapped as a result of terrorism	20,731	38,863	44,008

	2005	2006	2007
Terror attacks in Afghanistan	491	969	1,127
Attacks resulting in at least one death, injury, or kidnapping	366	695	892
People killed, injured, or kidnapped as a result of terrorism	1,540	3,557	4,673

* In all cases limited to attacks targeting noncombatants. 2005 & 2006 numbers were updated since last year’s publication on the Worldwide Incidents Tracking System at www.nctc.gov.

NCTC Observations Related to Terrorist Incidents Statistical Material

Approximately 14,000 terrorist attacks occurred in various countries during 2007, resulting in over 22,000 deaths. Compared to 2006, attacks remained approximately the same in 2007 while deaths rose by 1,800, a 9 percent increase from last year’s number. As was the case in the previous two years, the largest number of reported attacks and deaths occurred in Near East and South Asia. These two regions accounted for about 87 percent of the 355 casualty attacks that killed 10 or more people—only 45 casualty attacks occurred in Africa, East Asia & Pacific, Europe & Eurasia, and Western Hemisphere.

• Of the 14,499 reported attacks, almost 43 percent—about 6,200—occurred in Iraq where approximately 13,600 fatalities—60 percent of the worldwide total—were reported for 2007.
• Violence against non-combatants in Africa, particularly related to attacks associated with turmoil in or near Somalia, Kenya and Niger, rose 96 percent in 2007, totaling 835 attacks in comparison to approximately 425 attacks reported for 2006.
• Fighting in Afghanistan intensified during the past year, resulting in 1,127 attacks and a 16 percent increase over the approximately 970 attacks reported for 2006.
• The number of reported attacks in 2007 fell in the Western Hemisphere by 42 percent, in Europe and Eurasia by 8 percent, and in South Asia by almost 7 percent.

The number injured during terror attacks rose in 2007, as compared with 2006, by 15 percent; this is largely attributed to a more than doubling of the reported injuries in Africa from 2006. Kidnappings in 2007 fell two-thirds, with the largest decline in Nepal where peace negotiations during the year apparently curtailed hostage taking by 95 percent.

Attackers

The perpetrators of over 9,200 terrorist attacks (64 percent of total attacks) in 2007 could not be determined from open source information. Of the remaining attacks, as many as 130 various subnational groups—many of them well-known foreign terrorist organizations—or clandestine agents were connected to an attack in various ways, including as a claimant, as the accused, or as the confirmed perpetrator. In most instances, open source reporting contains little confirmed or corroborating information that identifies the organizations or individuals responsible for a terrorist attack. In many reports, attackers are alleged to be tied to local or well-known terrorist groups but there is little subsequent reporting that verifies these connections. Pinpointing attackers becomes even more difficult as extremist groups splinter or merge with others, make false claims, or deny allegations.

• According to open source reports, Islamic State of Iraq (aka al-Qa’ida in Iraq), more than any other subnational group, claimed they conducted attacks with the highest casualty totals.
• By contrast in 2007, the Taliban claimed the most attacks, but had a lower casualty total. The Taliban also took hostages more often than any other group.

No terrorist attack occurred last year that approached the sophistication of planning and preparations that were characteristic of the 9/11 attacks. However, open source reporting alleges that Islamic extremists played an important role in a 2007 UK bombing plot that was foiled when vehicle bombs were discovered outside several night clubs, as well as a disrupted German bombing plot that targeted American interests. Reporting points to a steadfast al-Qa'ida that is planning attacks in northwest Pakistan and was able to expand its propaganda campaign in 2007 to invigorate supporters, win converts, and gain recruits while its al-Qa’ida linked groups carried out several successful attacks.

• Al-Qa’ida in the Islamic Maghreb (AQIM) attacked a United Nations facility in Algeria killing over 40 people and wounding over 150.
• According to open source reporting, al-Qa’ida may have been behind the Benazir Bhutto assassination that killed the former prime minister along with 153 others and wounded approximately 250.
• AQIM launched a coordinated vehicle bomb (VBIED) attack in Algiers and Bab Ezzouar, Algeria targeting a government palace and a police station killing over 30 people and wounding over 200.

Types of Attacks

As was the case in 2006, most 2007 attacks were perpetrated by terrorists applying conventional fighting methods such as bombs and weapons including small arms. However, technology continues to empower terrorists and effective methods of attack are offsetting countermeasures against terrorism. Terrorists continued their practice of coordinated attacks including secondary attacks on first responders at attack sites with uniquely configured weapons and other materials to create improvised explosive devices (IED), including the introduction of chemical IEDs in 2007.

• Bombing incidents increased approximately 4 percent from those in 2006, while the death and injury tolls in these incidents rose by about 30 percent and 20 percent, respectively. Overall, suicide bombing attacks rose by about 50 percent and suicide car bombings about 40 percent. Suicide bombers operating outside of vehicles increased by about 90 percent, and the ability of these attackers to penetrate large concentrations of people and then detonate their explosives may account for the increase in lethality of bombings in 2007.
• A new terrorist attack method in Iraq emerged in late 2006. According to an Iraqi explosive expert, a 2006 attack included chemical weapons in a Sadr City car bombing, signaling a dangerous shift in tactics for Islamic radicals that carried over into the first half of 2007.

Victims and Targets of Attacks

As was the case in 2006, substantial numbers of victims of terrorist attacks in 2007 were Muslim.

• Approximately 67,000 individuals worldwide were either killed or injured by terrorist attacks in 2007. Based upon reporting and demographic analysis of the countries involved, well over 50 percent of the victims were Muslims and most were victims of attacks in Iraq.

Open source reporting identified approximately 70 percent of the approximately 67,000 killed or injured victims as “civilians,” and therefore actual tallies of significant types of victims cannot be specifically determined. However, the reporting does yield some insights about the demographics of these victims.

• Reporting identified a 13 percent increase in police killings and injuries, totaling over 9,400 killed or injured in 2007 in comparison to the 8,350 killed or injured in 2006.
• Killings of educators dropped slightly in 2007; 145 deaths were reported in 2007 as compared to 151 last year. Reporting cited almost 800 student victims killed or injured in attacks, an increase of over 80 percent.
• Over 2,400 children were reported as either killed or injured in terrorist attacks, an increase of more than 25 percent from 2006.
• The percentages of attacks involving journalists increased 22 percent in 2007 and deaths and injuries resulting from those incidents increased by 12 percent from 2006. Hostage situations involving journalists increased from 47 in 2006 to 79 in 2007.

In addition to the human toll, over 19,000 facilities were struck or were targets of terrorist attacks last year. Since the data’s baseline in 2005, the most common types of properties damaged or destroyed during an attack were vehicles and residences, but in 2007 communities were frequently attacked, an approximate increase of 45 percent, from just over 930 attacks in 2006 to well over 1,300 attacks in 2007. The percentage of attacks for other types of property damage or destruction, such as those associated with energy, transportation, education, government, and other enterprises, remain at single digit levels with a few notable exceptions.

• Almost 100 mosques were targeted during an attack in 2007, in most cases by Islamic extremists, representing a 55 percent decline. This is indicative of a return to levels in 2005, before the al-Qa’ida bombing of the Samarrah (Golden Dome) mosque that sparked a wave of sectarian violence in Iraq.
• Attacks on schools rose by 22 percent, but of those attacks reported, there was a 67 percent increase in the number of school facilities hit in those attacks.
• Attacks against political parties rose 64 percent with a 91 percent increase in party facilities being hit in those attacks.
• Attacks on communities increased 45 percent from 2006, and 158 percent from 2005. The ratio of the number of communities hit to the number of attacks was about 1:2, 2:3, and 1:3 for 2005, 2006, and 2007, respectively. This drop in 2007 may indicate that communities have taken security measures to protect themselves.
• Attacks on trains and subways rose 28 percent from 2006, and 83 percent from 2005, indicating terrorists are still intent on targeting them. However, the number of trains and subways actually hit in these attacks dropped 24 percent from 2006, indicating that security measures may be effective.

An Academic’s Perspective of Statistical Data

“Congress mandated a review and annual reporting on international terrorism starting in 1987, and an opportunity was missed to create a national accounting of terrorism with the same standing as the accounting for inflation and unemployment. In 2004, the government acknowledged mistakes in the 2003 report and made efforts to improve future reporting. Out of the improvements emerged the Worldwide Incidents Tracking System (WITS) and the enumeration of data has markedly improved in the past several years. Pursuing the data is fascinating, and some assumptions made about terrorism must be reconsidered. Gaps in our knowledge of terrorism remain, and however professional the WITS team has become, the data lacks the standing that national statistics merit. Two areas of improvement are of vital concern: time series data and reducing missing values, both crucial for statistical inference. Time series data allows us to rigorously investigate what actions work best to reduce terrorism. Making the public and academics guess about crucial relationships in data due to missing values, reduces rather than enhances the usefulness of the data. Classified information could be used to inform missing values without exposing confidential sources. National statistics are a great public service, and there is no reason that our national security policy should not be conditioned by trends and patterns with the same accuracy used to track our economy.

Building up a small but dedicated staff of enumerators and subjecting their procedures to the annual critiques of a Brain Trust (on which I serve), the WITS product has markedly improved in the past several years. It has produced a web-based data service that can help public officials, journalists and citizens answer basic questions about the scope and source of terrorism in today’s world.”

David Laitin
Stanford University
April 30, 2008

The full letter of Dr. Laitin is available in the 2007 NCTC Report on Terrorism, available via the Internet at www.nctc.gov.

Terrorism Deaths, Injuries, Kidnappings of Private U.S. Citizens, 2007
Provided by the Bureau of Consular Affairs, U.S. Department of State

The term "Private U.S. Citizen" refers to any U.S. citizen not acting in an official capacity on behalf of the U.S. Government; therefore these figures do not include, for example, U.S. military personnel killed or injured in a terrorism-related incident while on active duty or employees of the Department of State and other federal agencies. Members of U.S. Government employees’ households are considered private U.S. citizens.

Although every effort was made to include all terrorism-related deaths and injuries involving private U.S. citizens, the figures below reflect only those cases reported to, or known by, the U.S. Department of State, and may not reflect actual numbers of injured, which may not always be reported depending on their severity. As NCTC also notes, in the cases of Iraq and Afghanistan, it is particularly difficult to gather comprehensive information about all incidents and to distinguish terrorism from the numerous other forms of violence.

U.S. citizens worldwide killed as a result of incidents of terrorism: 19
U.S. citizens worldwide injured as a result of incidents of terrorism: 0
U.S. citizens worldwide kidnapped as a result of incidents of terrorism: 17

• In all cases limited to incidents targeting noncombatants.

TERRORISM DEATHS OF PRIVATE U.S. CITIZENS IN 2007 (BY COUNTRY)

Country	Date of Incident	Number	Location
Afghanistan	February 27	1	Bagram
	June 28	1	Pol-i-Charki
Iraq	January 17	1	Baghdad
	January 18	1	Baghdad
	January 23	5	Baghdad
	March 27	1	Baghdad
	March 31	1	Diwaniyah
	June 12	2	Baquba, near Tikrit
	July 19	1	Baghdad
	September 18	1	Baghdad
	November 26	1	Baghdad
	December 9	3	Al Kut

TERRORISM KIDNAPPINGS OF PRIVATE U.S. CITIZENS IN 2007 (BY COUNTRY)

Country	Date of Incident	Number	Location
Afghanistan	September 22	1	Kabul
	Before November 7	1	Kabul
Chad	October 1	1	Bardai
Iraq	January 5	1	Al-Haritha
	January 27	2	Sulaymaniyah
	February 1	1	Baghdad
	March 3	1	Baghdad
	April 25 (approx)	1	Kuwait-Iraq border
	May 25	1	Kuwait-Iraq border
	August 17	1	Al Amarah
Nigeria	January 7	1	Okan Oil Field, Delta State
	May 1	1	Okan Oil Field, Delta State
	May 9	4	Okan Oil Field, Delta State