TERRORISM
The September 2001 attacks led to a greatly increased U.S. emphasis on terrorism, which became the primary national security issue, although not necessarily dominating intelligence activities in the same way as did the Soviet issue during the Cold War.
HISTORICAL CONTEXT. The intelligence community’s interest in terrorism pre-dates the 2001 attacks. First, there had been a series of earlier attacks by al Qaeda on U.S. interests, beginning with the first attack on the World Trade Center in New York in February 1993. Second, it is also important to remember that terrorism is a recurring phenomenon in international politics. In the late 1890s there was a series of anarchist assassinations, killing President Sadi Carnot of France (1894), Empress Elisabeth of Austria-Hungary (1898), and President William McKinley of the United States (1901). In the United States from 1917-1920 there was the Red Scare, largely a series of bombings by anarchists, labor radicals, and pro-Soviet individuals. In the 1970s and 1980s there were several strands of terrorism: European and Japanese radicals (West Germany’s Baader Meinhoff Gang or Red Army Faction; Italy’s Red Brigades; Japan’s Japanese Red Army); various Middle East terrorist groups (Black September, the Abu Nidal Organization, and others); and state-based terrorism (including Libya, Iran, North Korea). These strands sometimes came together in cooperative terrorist attacks. Thus, one can argue that the U.S. intelligence community has had more than thirty years of experience with terrorism.
However, unlike the consistency of the Soviet target, terrorism has been a shifting target as groups rise and fall or are defeated, and as the locus of terrorism changes. Therefore, it may be fair to say that there is more generic experience with terrorism than specific experience. Moreover, the earlier terrorist campaigns all were political in nature. The current terrorist threat has a self-selected religious basis, which makes it much more difficult to discuss as a policy issue because of our concerns about religious freedom and our understandable desire not to blame an entire religion because of the acts of a faction within that religion. The religious aspect of modern terrorism also poses an analytic challenge in that western states (with the exception of Northern Ireland) largely stopped fighting about religion in the seventeenth or early eighteenth centuries. Ironically, the next “-ism” that some assumed would come about to replace communism as a foe—religious fanaticism—may actually be a historical throwback, at least in terms of western experience.
LESSONS FROM THE COLD WAR. To understand the difficulties inherent in tracking and forestalling terrorism, one must recall the intelligence legacies of the cold war. Terrorist groups, unlike the Soviet Union, do not operate from large, easily identifiable infrastructures and do not rely on extensive communications networks. As more becomes known publicly about U.S. intelligence sources and methods, terrorists have made greater efforts to avoid detection. For example, al Qaeda leader Osama bin Laden reportedly gave up the use of cell phones and fax machines to avoid being located by the United States. Also, terrorist groups do not conduct large-scale repetitive exercises, as do organized military forces. Thus, the visible signature of terrorists is much smaller than is that of the Soviet Union or any nation-state. But, the intelligence community still has to some extent a cold war legacy collection system developed to track a large political-military structure. Another major distinction between the Soviet target and the terrorist target has been noted by John McLaughlin, former deputy DCI: in the case of the Soviets we had a good sense of their capabilities but not their intentions. In the case of the terrorists, we know their intentions but not their capabilities.
Analysts sometimes refer to
chatter
when they describe intelligence on terrorism. “Chatter” is a difficult term to define. It refers less to precise intelligence than to patterns of intelligence: communications and movements of known or suspected terrorists. As chatter increases—more messages, even those that may not contain direct references to attacks—or as suspects suddenly drop from sight, an increased urgency is felt about the possibility of an attack. In that sense, chatter is much like indications and warning (1&W)—anything that represents a change in observed patterns is the subject of increased attention. But chatter is also imprecise and, as terrorists learn more about how the United States collects intelligence, chatter can decrease for reasons other than pending operations.
In the aftermath of the September 2001 attacks, familiar claims were made that the United States was overly reliant on technical intelligence (TECHINT) and needed more human intelligence (HUMINT). Although HUMINT can, theoretically, collect terrorist-related intelligence that TECHINT cannot, the realities of terrorism must again be examined. Terrorist groups, and certainly their leadership cells, tend to be small and well known to one another. They have tended to operate in parts of the world where the United States does not have ready access. Even if trained agents were available who knew the required language and could be provided with a plausible cover story for their presence in one of these areas, penetrating the terrorist organization would remain problematic at best. One does not simply show up in Kabul, ask for the local al Qaeda recruiting office, and then request to see the person in charge. (The press made much in this regard of the activities of the American John Walker Lindh in Afghanistan. Lindh was captured fighting for the Taliban, not for al Qaeda. Recruitment into the Taliban was fairly simple. One had to be a self-professed Muslim willing to carry a gun—a far easier task than joining al Qaeda.) Finally, if HUMINT penetration were to be achieved, the new recruit would likely be asked to take part in some operation to prove his or her commitment to the cause. This raises important moral and ethical issues for intelligence. How far would the United States be willing to go to sustain a HUMINT penetration—putting an agent’s life at risk by taking part in a terrorist operation?
Some advocacy for more HUMINT was odd in that it seemed to treat HUMINT as a numbers issue: that is, if enough agents were sent, penetrating the target would prove inevitable. Such a scenario shows a fundamental misunderstanding of how HUMINT operates and the nature of the terrorist target. HUMINT is not an en masse activity. It relies on precision.
Finally, much of U.S. HUMINT against the Soviet Union was carried out in foreign diplomatic posts outside of the Soviet Union, where Soviet officials were present and more accessible. Terrorists do not have this same overt overseas presence and thus present a smaller accessible target.
This does not mean that human penetrations of terrorist cells are impossible. As with the Soviet Union, a walk-in may occur. This apparently was the case with Ilich Ramirez Sanchez, a Venezuelan-born terrorist better known as Carlos the Jackal. He apparently was betrayed either by someone in his organization or by his Sudanese hosts. Walk-ins remain fortuitous, although, as discussed later, they can come as a result of ongoing successes against terrorists.
Beyond HUMINT, the terrorist target puts a premium on several types of intelligence:
• Signals intelligence (SIGINT): Very broadly defined, to include a wide variety of communications, including a presumed extensive presence on the Worldwide Web
• Open-source intelligence (OSINT): To collect and dissect the many public statements made by terrorist leaders and factions
• Measurement and signatures intelligence (MASINT): To collect against acquisition of various types of WMD
Finally, terrorism is an intelligence issue in which foreign liaison is very important, as is true for all transnational issues.
STATE SPONSORSHIP OF TERRORISM. State sponsorship of, or at least acquiescence to, terrorists makes the intelligence issue more complicated. The intelligence community must collect not only against the terrorists but also against other governments and their intelligence services. At one level this is easier than is the terrorist collection itself, as it falls within more common intelligence practice. However, it also puts an additional strain on intelligence resources. Liaison relationships may be questionable in such cases. For example, the government of Pakistan has been supportive of U.S. operations in Afghanistan up to a point, but the Pakistani intelligence service had been a longtime sponsor of the Taliban. State sponsorship also raises the issue of the failed states. Here it is useful to know if the terrorists are actually being hosted by the government, as was the case in Sudan and Afghanistan for bin Ladin, or whether the lack of internal order simply provides an atmosphere where terrorists can work relatively freely, perhaps without official sanction.
Closely related to state sponsorship is the even murkier question of relations between and among terrorist groups. For example, Libya had contact with factions of the Irish Republican Army. The Japanese Red Army worked with the Popular Front for the Liberation of Palestine (PFLP). Members of an Irish Republican Army faction were arrested after spending time with the revolutionary armed forces of Colombia (Fuerzas Armadas de Colombia, FARC). Such ties are both important and difficult to track or disrupt. The issue of ties among terrorist groups is important in the current campaign against terrorists both as a means of assessing threat and of assessing success. For example, most of the al Qaeda members who planned the September 11 attack are either captured or dead; al Qaeda’s safe haven in Afghanistan has been overrun. One of the concerns raised by these successes has been the effect on al Qaeda’s command and control. Is it still a unitary group, planning and ordering attacks from wherever its leaders are, or has it, in effect, become a franchised activity, with like-minded cells inspiring one another and occasionally working together but not necessarily in direct command and control? Similarly, once an attack has occurred it is important to know if it has been planned or ordered by an external group or has been carried out by indigenous individuals. The September 11 attack clearly was carried out by terrorists who entered the United States. However, the attacks in Madrid (2004) appear to have been directed from terrorists in Morocco, for whom a direct connection to al Qaeda has not been proven. The 2005 attack in London also appears not to be connected directly to al Qaeda, although some of the bombers had been in Madrassas in Pakistan. A conclusion of this sort may be more troubling because it indicates an indigenous problem that will be much more difficult to identify: radicalized, home-grown terrorists. This question of connections among various terrorist groups also indicates why so much emphasis is put on
link analysis,
that is, establishing connections between various people to get a sense of their broader social networks. This is also one of the major types of information gleaned by phone surveillance, connections between people, in addition to the actual content of their conversations.
WAR ON TERRORISM. The intelligence services have two roles in the campaign against terrorism: defense and offense. Defense consists of preventing future attacks by disrupting them or deterring them. This means, in turn, trying to obtain both detailed intelligence about any attacks that are being planned as well as ongoing intelligence about terrorist organizations. One of the most difficult aspects of defense is learning to think like a terrorist. This means not only being able to conceive of attacks that many analysts would consider too horrific to contemplate for long but also to appreciate the importance of randomness, which is a key ingredient of terror. It has been suggested that terrorist analysts focus too much on specific dates and events (holiday travel periods, major sporting events, national holidays). Although these dates have symbolic value or indicate periods when large numbers of people are either traveling or gathering in one place, they also may be easier to defend against. Of the major terrorist attacks that have occurred to date, only one—the failed Millennium attack at the beginning of 2000—was tied to an iconic date. In other words, this may be another case of mirror imaging. How successful are analysts at thinking like terrorists versus thinking like Westerners thinking like terrorists?
Offense consists of identifying, locating, and then attacking terrorists. These activities are important not only for eliminating terrorists but for introducing uncertainty into their activities and making it more difficult for the terrorists to organize, plan, and train. Offensive activities go from analysis into operations and raise questions about assassinations, renditions, and detentions. The war on terrorism adds another intelligence burden: support to military operations. This requirement encompasses both the usual military-related support and new activities. For example, the press has reported that the Central Intelligence Agency (CIA) has a Special Activities Division in the National Clandestine Service (formerly the Directorate of Operations, DO) that was engaged in operations against the Taliban and al Qaeda. Although little is known publicly about the division, it would appear to occupy a niche between Special Forces and the National Clandestine Service’s paramilitary activities in support of indigenous groups, such as the contras or the Mujaheddin. Also, some important developments have been made in geospatial intelligence with the use of unmanned aerial vehicles and commercial imagery.
One of the most difficult aspects of the campaign against terrorism is trying to gauge the relative degree of success. Unlike conventional wars, there are no battle fronts moving one way or another. Nor is it clear that the absence of another attack entirely means success. Again, it is possible that the nature of the terrorist organization has changed under the pressure of the U.S. response since 2001, going from a more centrally controlled structure to a looser one in which there may be many small centers of activity rather than a central one. If this is so, then the intelligence agencies face uncertainty about what this means for the future of terrorist attacks and for the best way to counter terrorists, both defensively and offensively. It is known that al Qaeda has fairly long planning cycles. Therefore, a quiescent period may simply be somewhere in this cycle. Also, it matters how one thinks about the terrorist issue. Although the United States has not been attacked since 2001, the other attacks—Bali (2002 and 2005), Madrid (2004), London (2005 and attempted 2007), Algeria (2007), and several others all suggest that it is better to look at the terror issue on a global basis. In July 2007, the director of national intelligence (DNI) released unclassified Key Judgments (KJs) of a national intelligence estimate (NIE),
The Terrorist Threat to the US Homeland,
which assessed that al Qaeda’s ability to attack the United States had been constrained but that the group remained a threat and “that the United States currently is in a heightened threat environment.” The NIE also noted the problem of alliances among various terrorist groups and the problem raised by al Qaeda’s apparent “safe haven” in Pakistan’s far western region.