|
|
|
CIAO DATE: 7/5/2006
Terrorism Statistics Flawed
Francis Rheinheimer
April 2006
It has become a truism that any attempt to define or quantify terrorism is informed by political trends, and thus subject to fluctuations based not on hard facts but on political fashion. Yet the State Department’s now defunct annual publication, Patterns of Global Terrorism, was the closest approximation of any government effort to provide information in an objective and consistent manner. As a successor to Patterns, the report produced by the U.S. National Counterterrorism Center (NCTC) -- called A Chronology of Significant International Terrorism for 2004 -- effectively ends over 20 years of analytical consistency in the U.S. government’s terrorism accounting practices.
In 1973, Congress passed a law (Title 22, §2656f) requiring the government to produce a report every year detailing the previous year’s trends in international terrorism. The report was to include a general overview, an analysis of the threat by country and region, and a chronology of events the government deemed “significant.” For the first 10 years of the law, the Central Intelligence Agency (CIA) produced an annual report, “International Terrorism,” that explained major trends in terrorism for the government and Congress. Responsibility for the report was transferred to the State Department in 1981. Its report – Patterns of Global Terrorism – has long been a staple for legislators, experts and citizens interested in understanding the causes and effects of terrorism, as well as ways to prevent it. It adhered to a consistent methodology and its findings were comprehensive. It contained regional and country overviews, a report on state-sponsored terrorism and appendices that included a chronology of significant incidents, a list of terrorist groups, a statistical review and charts on total anti-U.S. attacks. Even though the law did not require the State Department to release its report to the public, it became standard practice to publish a glossy booklet accessible to anyone interested.
The NCTC was created in 2004 “to serve as the primary organization in the United States Government for integrating and analyzing all intelligence pertaining to terrorism and counterterrorism.” In April 2005, the NCTC issued its report, Chronology, before the State Department issued its version of Patterns, which was subsequently cancelled as a result of differences with the NCTC over what actually counts as terrorism. In its place, the State Department issued Country Reports on Terrorism, which omits all useful statistical data on international terrorism. Instead, Country Reports recounts the previous year’s trends country by country, both in terms of terrorist activity and state measures to counter the threat. All responsibility for the numbers – domestic and international incidents, injuries and fatalities, and “significant” events – was handed over to the new NCTC, an office of the Director of National Intelligence (DNI).
As a government agency, the NCTC employs the standard U.S. definition of terrorism: “premeditated, politically motivated violence perpetrated against noncombatant targets by sub-national groups or clandestine agents, usually intended to influence an audience.” In order for an attack to be international, it must involve the citizens or territory of more than one country. So when a group of Russian nationals bombed a plane carrying 46 other Russian travelers in August 2004, it was considered purely domestic. A near simultaneous attack on another plane, also perpetrated by Russians, happened to carry one Israeli citizen and was called international. And an incident is judged to be significant if it caused death or serious injury to noncombatants or amounted to more than $10,000 in property damage. The NCTC vetting committee consists of officials from the departments of State, Defense and Homeland Security, as well as the Federal Bureau of Investigation (FBI), CIA and the Defense Intelligence Agency.
Unlike Patterns, the NCTC report does not distinguish between significant and non-significant international terrorism, preferring instead to examine only incidents it considers significant. An examination of the statistics put out by the report points either to questionable methodology or inadequate oversight. The NCTC claimed there were 651 “significant international” terrorist incidents in 2004, up from 175 in 2003 (according to Patterns), and by far the highest level since Patterns was first published in 1983 (see charts). Of that number, 201 were linked to the insurgency in Iraq, and fully 295 to militant action in Kashmir – leaving just 155 incidents outside of those two conflict zones. The NCTC counts 1,907 fatalities, compared with 625 for 2003 and 725 for 2002. And it counts 6,704 wounded – the highest number of any year on the government record.
To add some perspective, the data released by Rand Corporation – which does not differentiate between significant and non-significant international terrorism – shows Kashmir to have endured not a single incident in 2004 (though it has been plagued by domestic terrorism and insurgency for decades). The State Department’s figure for all of India in 2003 was just 53. Although Rand’s figure for Iraq is slightly higher than that of the NCTC report (247 to State’s 201), its total for the year (again, not distinguishing between significant and non-significant) is 396.
Bearing in mind that Rand has historically applied a narrower definition of terrorism than the State Department and as a result shows fewer incidents and deaths, NCTC’s numbers in their present form are still problematic. Its inclusion of Kashmir’s 295 incidents likely reflects the view that the conflict – which involves territorial claims by two states – is inherently international. Thus all armed attacks that were not against active military targets were included in the list. For example, the NCTC report counts three incidents that took place in Kashmir on Aug. 23, 2004, whereas Rand does not. All three incidents involved “armed militants” attacking non-combatant targets (which includes policemen but not soldiers), but none of the brief incident descriptions indicate the participation of international players. The State Department differed with the tendency to include all armed attacks in Kashmir, which is part of the reason why Patterns was cancelled in the first place.
The authors of Patterns applied the criteria for an international attack more selectively and as a result showed just 49 incidents in all of India in 2003, 67 in 2002, and 38 in 2001. Similarly, Rand considered Kashmir a domestic conflict characterized by rampant criminality and violence rather than a region plagued by international terrorism. Certainly, Kashmir is a place of often indiscriminate violence that takes root in certain political realities, but the choice between labeling violence as “terrorism,” as opposed to “insurgency” or “civil war,” is essentially political. And the choice to call it “international” – implying it profoundly affects U.S. security – is just as subject to political judgments.
The NCTC’s decision to include Kashmir’s 295 incidents does not bear significance on whether international efforts to eradicate the real threat of terrorism should continue (they must), but it does raise important questions about trends in and appropriate responses to international terrorism. The NCTC’s data for 2005 are only compiled up to June, but preliminary records show their incident statistics likely to double – from 3,000 in 2004 to 5,200 for just the first six months in 2005. This development points to the NCTC’s evolution from being the heir of Patterns to being a simple database for all incidents involving the threat or use of violence for political ends. It is useless for any analytical undertaking to examine the motives behind three high-school students vandalizing cars with the letters “ELF” (Earth Liberation Front), just as it is useless to count every act of politically motivated violence as terrorism. The list may be more comprehensive, but it is less meaningful. Moreover, how the government counts international terrorism incidents profoundly affects the credibility of the Bush administration’s claim that the United States is engaged in a “Long War” against international terrorism. NCTC’s accounting methods, which show that international terrorism is rapidly getting worse, motivate government officials eager to promulgate their theatrical vision of the conflict in which the United States is now mired. Omitting the NCTC’s more questionable incidents – those in the conflict zones of Iraq and Kashmir – shows terrorism reached its zenith in the mid-1980s, and has been declining since.
Indeed the context within which violence occurs is precisely what makes the effort to quantify terrorism important and doable with a mark of consistency. From a certain vantage point, the conflict in Kashmir – where savage attacks against noncombatants are routine – has international dimensions. It has certainly helped to calcify the India-Pakistan rivalry. Yet Rand excludes most incidents in Kashmir from their database because Kashmir has been in perpetual conflict since the British army left the subcontinent in the late 1940s, and because there are few “foreigners” brave enough to venture into the war zone. Violent acts in Kashmir are frequent, and many amount to cold-blooded terrorism, but calling them international simply because they affect a broader political dynamic is questionable.
It is so questionable that the State Department omitted most of the incidents in Kashmir from Patterns, whose authors were told to stop including the chronology – and to change the report’s name – because the NCTC was granted sole responsibility for the numbers. For the public, this means the loss of a critical component in any effort to understand the threat of international terrorism: the world superpower’s official record of the number of “significant” incidents, and the deaths they caused. Rand applies a different methodology in its database, so the meaning one takes away from its figures must be slightly different than that of Patterns, but its habit of amending its data retroactively is a necessary recognition that terrorism statistics are never completely accurate, nor free from political calculations. The cancellation of Patterns, which arose out of the brash political style of Bush appointees, was like throwing the baby out with the bathwater.
The State Department is expected to release the second issue of Country Reports on Terrorism on April 21, and it may (or it may not) contain statistical information on international terrorism it considers significant. But the government is unlikely to allow two contradicting reports, even if the one ultimately chosen to represent the government’s account of international terrorism is inaccurate, or just useless. There is nothing inherently wrong with the NCTC’s database; it is probably the single best and most accessible resource for finding a specific terrorist incident, even if it is not actually a terrorist incident. But it does not provide the kind of consistent scrutiny that Patterns once provided. Playing politics with the numbers is perhaps to be expected from this administration, but it undermines any effort at honest and reliable analysis.
# # #
CDI Research Assistant Chris Weatherly contributed to this report.
