What Do The Terrorism
Numbers Really Mean?
Terrell E. Arnold
After release of the annual report, "Patterns of Global Terrorism," in April 2004, the State Department and the Bush administration took enormous flak for understating attacks and casualties in 2003. The principal spin put by critics on this gaffe is that the April numbers were designed to put the best possible face on results of the War on Terrorism. Supporting that spin were remarks of the State Department coordinator for counterterrorism, J. Cofer Black, that the low reported numbers represented a "remarkable achievement." But political spin from the critics is also obvious, because California Representative Henry Waxman probably launched the complaints mainly because four attacks in Istanbul, Turkey that targeted synagogues had been left out.
State's revised report increases the number of reported attacks from 190 to 208, while more than doubling the number of deaths (from 307 to 625) and the number of wounded (from 1,593 to 3,646). These numbers are totally abstract unless we understand the data series as well as understand where terrorism fits in the scheme of violence against humans. For starters, in 2003 worldwide terrorist attacks killed and wounded fewer Americans than were killed and wounded by lightening strikes in the United States. Except for 9/11, that is pretty much the average case for the past decade.
Secretary of State, Colin Powell, tried to put the data problem in a correct perspective when he said there was no intent to "make our efforts look better or worse;" rather he cited problems in compiling the data. In that vein there are numerous and persistent problems with building this report. As one of the developers of the report in the early 1980s, this writer wrestled with problems of data collection, terrorism definition, biases of reporting organizations and individuals, design limitations on report coverage, pressures from regional and policy peers in various organizations, and sheer lack of data, to name only some of the difficulties. Most of those problems are neither easily solved nor likely to go away.
On its face, data collection sounds easy, but numerous problems make the global process error prone. To start with, open source data collection is a people intensive process. Computers are great compiling tools once they are fed information, but they lack judgment, and terrorism data do not lend themselves to automated collection. Only experience will demonstrate source or data set integrity, and over time the integrity of a reporting organization may change. For example, today would you trust the Bush team or the Pentagon on weapons, terrorist numbers or casualty figures for Iraq? Exactly! Knowing when sources have become unreliable or incomplete requires a human judgment.
In the past 25 years the United States has added diplomatic representations to about 40 new countries that have joined the present 192-member United Nations. That is almost a quarter of the current family of nations. But the number of US diplomats who observe and report in those missions is smaller and more thinly spread than it was 20 years ago. A second problem is that there are no uniform international standards for gathering and reporting such data. That is partly a problem of methodology"ways of selecting and compiling information. Chances are pretty good therefore that data collected from a global hodgepodge of sources will not all mean the same things, and certainly they will not mean the same things to all users.
An equal if not greater problem is definitions. Who is a terrorist for purposes of inclusion in "Patterns of Global Terrorism?" Since 1983 the Department of State has used the following definitions: The term terrorism means premeditated, politically motivated violence perpetrated against noncombatant targets by subnational groups or clandestine agents, usually intended to influence an audience. The term international terrorism means terrorism involving citizens or the territory of more than one country. The term terrorist group means any group practicing, or that has significant subgroups that practice, international terrorism.
"Significant" international terrorist incidents are the meat of this report. As defined in the report, a significant international terrorist incident " is one that results in loss of life or serious injury to persons, major property damage, and/or is an act or attempted act" that could result in those outcomes. Of the 208 incidents reported in 2003, 161, or 77% of all reported incidents"the highest proportion in some time--were classed as significant. Significant incidents occurred in 40 countries, but 30% of them were in India alone, and roughly 60% of them were in five countries: India, Afghanistan, Colombia, Iraq, and Israel, in that order.
Domestic incidents"those that do not involve either foreign terrorists or foreign victims--in each country are discussed, but no data are included, because the report does not require them. However, according to the report "domestic" incidents may be more numerous that international ones. Thus problem one in making the cut is to know for sure which incidents actually meet not only the first requirement, but also the two international terrorism definitions. That is a fact-finding problem, and since the criteria must be applied rigorously in the 75 or so countries where terrorism routinely occurs, there is ample room for error.
The second definition problem centers on the actors who commit international terrorist attacks. They are defined for the report as "sub-national groups or clandestine agents". This by nature excludes acts carried out by the armed forces of any country. It might include acts by the clandestine agents of countries, but that is not the apparent intent of the definition, and thus a good clandestine operation"one in which the foreign hand is invisible--is more likely to be classed as domestic terrorism. Moreover, even when it is clear that, for example, Mossad is conducting operations against Palestinians, killing alleged Hamas members or destroying alleged terrorism-related facilities, those incidents do not get counted as terrorism. Nor do operations of the Israel Defense Force against civilian targets in Palestine or of Coalition forces against civilian targets in Iraq get counted. In short, the presently applied definitions of international terrorism are a lowest common denominator that is delimited by the political imperatives of the defining state.
The incidents annually reported in Patterns of Global Terrorism, therefore, may be accurate for the cases cited, but the picture painted will always be less than the sum of the visible parts. More is going on than is ever counted, and less is always said than needs to be said. The picture that exists for purposes of assessing worldwide politically motivated violence is inevitably incomplete.
Carefully fenced in the reporting practices of every government are the operations of their armed forces. The definition used in Patterns excludes such operations. But how can one legitimately exclude such acts as those of the Khartoum government that are directly and indirectly carried out by agents to exterminate the non-Muslim and black peoples now scrambling to cross into Chad? Just how, or should, one count Russian attacks in Chechen? How can one exclude the targeted assassinations of the Israelis against Palestinians or the willful Israeli destruction of townships such as Jenin or Rafah? The issues become matters of legitimacy and propriety, and the ground beneath a range of official actions becomes very slippery.
This very slipperiness in the eyes of the beholder has caused the Bush and earlier US administrations to seek immunity from prosecution in the world court. Whereas, in earlier years, the US was concerned that charges against US forces could be capricious and/or politically motivated, the risk now is that operations against civilians in Fallujah and elsewhere in Iraq, as well as torture at Abu Ghraib, provide real and convincing evidence of misconduct. A simple truth about terrorism data, therefore, is that they may be clear in the eye of the beholder, but they can be something else indeed in the calculus of the politician. How to keep as far from this problem as possible is a challenge for the compilers of terrorism data.
The comment that used to send this writer up the wall, first as an official and later as a speaker and writer, was: "One man's terrorist is another man, s freedom fighter". Unfortunately, that is a flippant version of a real problem. If the data collector government is sponsoring groups such as the Contras in Nicaragua, or UNITA in Angola, or the precursors of al Qaida in Afghanistan, how do the activities of those groups get handled in the statistics? Tricky, especially when the head of one such group was invited to meetings with White House senior staff. Nor has it been easy at any time in recent history to shut off private US funding sources for the IRA, even though the IRA and its offshoots are treated in Patterns as terrorist groups. It is sufficient for our purpose here to say that the terrorist/freedom fighter perception casts shadows over terrorism data for several countries that make accurate data collection difficult, while making the presentation of such data to say the least politically sensitive.
References to al Qaida complicate the matter even further. The current gut reflex is to say that any act of terrorism anywhere is al Qaida related. Proof neither regularly nor automatically follows this charge. Because the US has designated al Qaida as terrorism enemy number one, maybe the unique enemy, the suggestion that al Qaida members, agents, affiliates, or sympathizers carried out an attack tends to shut off inquiry. Recent attacks in Saudi Arabia have been blamed on and claimed by people who say they are al Qaida, either foreign or domestic, but the situation is not necessarily so clear-cut, because Saudi Arabia has dissidents who have grievances that have nothing to do with al Qaida. First blame on the Bali bombings in Indonesia was laid on al Qaida, but the culprits were indigenous Islamic militants with purely local grievances against the government, maybe with some help/inspiration from al Qaida agents. This is a data management problem, because it matters who did these things, and blaming al Qaida for somebody else's work is little better than refusing to include incidents because they were carried out by governments or were politically sensitive. In such cases, responsible governments are off the hook, but the terrorism goes on.
Major problems with the meaning of the terrorism data concern the magnitudes. The Bush administration declared War on Terrorism based on a single incident, but how does terrorism compare with numerous scourges of civilization? Since the US-led invasion of Iraq, US and Coalition deaths in Iraq have exceeded 1,000, and US wounded have exceeded 5,000. This does not count Iraqi casualties that exceed 14,000. US deaths due to vehicle accidents in 2003 exceeded 42,000. Over 17,000 of those deaths were alcohol related. Suicides numbered upward of 30,000, and there were over 16,000 homicides in the United States in 2003. Deaths and injuries from foul weather in Haiti this year have greatly exceeded the 2003 global terrorism numbers, or even the 9/11 numbers. Far more people were killed and injured in earthquakes in both Iran and Turkey during 2003 that died or were wounded in terrorist incidents.
The terrorism numbers are by no means in the same class as the foregoing causes of death and injury. As noted above, worldwide terrorism casualties during 2003 were 625 dead and 3,645 wounded. American casualties in 2003 totaled 52, including 35 dead and 17 wounded. The numbers were down by 10 from 2002 when deaths numbered 27 and injuries numbered 35. Less than 6% of the worldwide terrorism deaths in 2003 were Americans, and less than .05% of the wounded were Americans. In short, total US deaths and injuries due to terrorism in 2002-2003 were fewer than those caused by lightening strikes in the United States (according to the National Weather Service, an annual average of 93 deaths and 300 injuries).
These numbers pose some rude questions: Just how much of our national attention should be devoted to terrorism? What share of our national resources should be spent on defenses against terrorism? How much of our law enforcement budget should be spent on it? What part of our military budget should be sunk into it? How many of our intelligence resources should we use on it? What part of our diplomatic energy should be devoted to it? The final questions are (a) just what difference will it make in our national safety if we remain preoccupied with this problem, and (b) Just how realistic is the pattern of fear that the Bush team uses to keep us on the side of the War on Terrorism?
The terrorism numbers themselves shed a peculiar light on those questions. The truth is most of the world's terrorism is not directed against the United States. Nor has it ever been. It is also clear that other human behavior patterns are more deadly to life and limb than terrorism. None of the 2003 attacks occurred in the United States, and 9/11 aside it has been the pattern for some time that few international terrorist attacks occur in the United States. To be sure, as demonstrated by 9/11, a determined and successful attacker can wreak great havoc. But even the limited facts that are now available show (1) that the focus and attention of Washington leadership were more at fault than our diplomatic and intelligence networks that compile terrorism data, and (2) that the most powerful military capability on earth did not deter the attackers. Actually our military budget went up sharply in 2003, but the number of significant incidents increased sharply as well.
What do these experiences tell us? The asymmetric nature of the terrorism weapon is such that military power alone will never deal with it. The random pattern of terrorism is such that some of it will inevitably slip through our defenses. Our current approach to terrorism, the War on Terrorism, therefore is impractical, expensive, and unlikely to succeed. Weapons of mass destruction (WMDs) figure in the terrorism statistics only in a backhanded way. It is obvious that possession of WMDs is not a deterrent to terrorist attacks, because 64 incidents, representing 40% of the significant attacks during 2003, occurred in nuclear powers (India, Pakistan, Israel, France, and Britain). India alone experienced 48 attacks. Difficulties with obtaining WMD materials and problems of safe use, as well as the very provocative nature of their use offer significant deterrence to their use, but rumors persist of at least al Qaida efforts to obtain such a weapon. The apprehension raised by the possibility of such an attack is so great that the mere threat may constitute an attack. It certainly raises the prospect that target countries or groups could be intimidated by such a threat, even though the threatening group may have neither intent nor capability to deliver.
What can we do? The numbers tell us that our present posture is at best extreme and at worst irrelevant to the risks that beset us. Available data tell us that we should not fear terrorism more than other, more common causes of death and dismemberment. Rather, we should fit the risks of terrorism more effectively into our preparations for the range of crises generated by acts of man and acts of God. We should work diligently to assure that our society is properly equipped to deal with the whole spectrum of likely emergencies. We should equip our emergency response systems and our emergency responders with the tools and training they need to manage the consequences of those potential emergencies. Most of the tools needed to respond to the effects of a terrorist incident are the same as or similar to those needed in other emergencies.
On the international front, our main energies"political, economic, diplomatic, and intelligence"should be devoted to reducing the causes of terrorism. Terrorism numbers should be primarily a tool for focusing those energies. The numbers should tell us which countries need that energy most. Encouraging and assisting those governments to deal constructively with grievances will reduce terrorism risks. Our military capabilities should be used to maintain global stability conditions, only as a last resort to chase terrorists. We can do the most good over time by assuring that fewer people have reason to resort to terrorism to get our attention, or the attention of their own governments.
The writer is a retired Senior Foreign Service Officer of the US Department of State. He will welcome comment at



This Site Served by TheHostPros