U.S. Intelligence in the Twenty-First Century

The Conflict

On September 11, 2001, the United States suffered a major surprise attack when terrorists hijacked and crashed four passenger airliners into buildings in New York City, Arlington, Virginia (near Washington, DC), and in a field in Pennsylvania. The U.S. intelligence community was criticized for not putting together intelligence on the planning of this most recent and deadly attack before it hit. It is commonly agreed that the U.S. intelligence community, whose legal charter dates from 1947, was created to prevent another strategic surprise like that of Pearl Harbor in 1941, to which the terrorist attack was likened. The 2001 attack raised serious questions about the capabilities of the intelligence community and its future role, giving new emphasis to a debate that began with the end of the Cold War in 1991.

Counterterrorism

• The United States has been actively engaged against international terrorism since the early 1980s, but the September 11, 2001, attacks underscored a new, more immediate threat and a higher sense of focus.
• In this struggle against a shadowy, internationally dispersed opponent, a greater burden falls on the U.S. intelligence community, which is still transforming itself from its 50-year Cold War posture to one that will be able to respond to a more diverse set of twenty-first century threats.

On September 11, 2001, the United States suffered the second major surprise attack in its history. Unlike the first attack, by Japan at Pearl Harbor, Hawaii, in December 1941, the 2001 attack was not perpetrated by a nation. This new attack was the most recent and most deadly in a series of terrorist attacks by the al-Qaeda terrorist organization, headed by Osama bin Laden. Although separated by almost sixty years, the two events are linked for U.S. intelligence. It is commonly agreed that the U.S. intelligence community, whose legal charter dates from 1947, was created to prevent another strategic surprise like Pearl Harbor. The 2001 attack raised serious questions about the capabilities of that intelligence community and its future role, giving new emphasis to a debate that had begun with the end of the Cold War in 1991. The world and the challenges facing the United States had clearly changed. After a decade in which no issue seemed paramount, terrorism became the main focus of all national security policy. How well the intelligence community responds to the post-September 11, 2001, world will have a major effect on the future role intelligence is asked or allowed to play.

Historical Background

The Development of the U.S. Intelligence Community

As noted, the U.S. intelligence community officially dates from 1947, when the National Security Act revamped the U.S. government structure with three important changes. First, the act created a National Security Council (NSC), designed to bring together diplomatic and military policy into an advisory group under the president. The NSC currently consists of the president, vice president, and secretaries of state and defense, with the chairman of the joint chiefs of staff as the military adviser and the director of central intelligence as the intelligence adviser. Second, the War and Navy Departments were merged into a new unified structure, which became the Defense Department and included a separate Air Force as well. Third, the position of Director of Central Intelligence (DCI) was given legal status (it had existed earlier by presidential order) and placed under the NSC. Further, a Central Intelligence Agency (CIA) was created under the DCI.

There had been a handful of U.S. intelligence organizations prior to 1947. Both the navy and army had created intelligence branches in the late nineteenth century, and during and after World War I (1914-18), the United States had a very successful code-breaking operation. During World War II (1939-45), President Franklin Roosevelt (1933-45) had created the Office of Strategic Services (OSS), which served both operational and analytical functions, but it was disbanded in 1945. There had never been a peacetime national intelligence organization. President Harry Truman (1945-53) had agreed to the creation of the CIA in order to coordinate better the disparate intelligence he was receiving from diplomatic and military sources. Very quickly, however, the CIA began to fill vacuums in analysis, collection, and operations as the United States began to respond to Soviet pressure in the early days of the Cold War. By the Korean War (1950-53), the CIA was a full-blown intelligence service, not just a coordinator.

Between 1947 and 1962 the intelligence community continued to evolve and to grow. The National Security Agency (NSA) was created by presidential order in 1952, and given responsibility for collecting information from other nation's signals (that is, communications, test data, etc.) and for protecting the communications of the United States. With the advent of so-called spy satellites in the 1960s, the National Reconnaissance Office (NRO) was created, responsible for designing an array of intelligence satellites, including those that took images or photos and those that enabled the NSA to do its job. In 1962 the Defense Intelligence Agency (DIA) was formed to provide broader defense-related intelligence beyond that provided by the intelligence units of each military service for their more narrow needs. Other members of the intelligence community include four military service intelligence units: the State Department's Bureau of Intelligence and Research (INR), the Federal Bureau of Investigation's (FBI) National Security Division—responsible for counterintelligence— and, most recently, the National Imagery and Mapping Agency (NIMA), created in 1996 to oversee imagery intelligence.

This is a complex and highly specialized structure. Many agencies serve similar roles, with the two dominant roles being collection or analysis. There is witting redundancy in the structure. There are many collection agencies because each one specializes in a different type of highly specialized collection. There is more obvious redundancy in analysis because of two major operational concepts. The first concept is the view that each of the senior policymakers—the president, the secretaries of state and defense—has unique intelligence needs that can best be served by a specific intelligence agency. Thus, the CIA serves the president, DIA serves the secretary of defense and the chairman of the joint chiefs, and State/INR serves the secretary of state. This does not mean that the agencies do not share analysis or do not allow others outside their agency or department to see it. It does mean that each has a "primary" policymaker customer whose needs these agencies are particularly attuned.

The second operating concept is competitive analysis. Competitive analysis is based on the view that by having many analysts with disparate viewpoints come together on a specific issue, the intelligence community is more likely to come to analytical conclusions that are sound because they are broadly based and of the greatest value to those same policymakers. Intelligence, more often than not, is about uncertainty and ambiguity. After all, if something were known to be true, we would not need intelligence to collect information about it or to decide what that information meant. It is the uncertainty that drives intelligence, along with the fact that other nations or international actors seek to deprive us of the intelligence we seek, and we, in turn, seek to keep secret how we try to obtain that same information.

When looking at the intelligence community, an obvious question is: Who is in charge of it all? There are two answers, simultaneously: the DCI, or no one. The DCI is the nation's senior foreign intelligence official and the president's senior intelligence adviser. He/she is responsible for coordinating the activities of the 14 agencies that comprise the intelligence community. However, his ability to do so is also severely limited, as the DCI has line authority and budget responsibility over only two components, the CIA and the National Intelligence Council (NIC). The bulk of the intelligence community—NSA, DIA, NIMA, and the service intelligence units—comes under the purview of the secretary of defense. Some estimate that upwards of 80 percent of the intelligence community is under the secretary of defense rather than the DCI. Thus, there is a widely acknowledged gap between the DCI's responsibilities and his authority.

The Cold War

The other important factor in the growth and development of the intelligence community was the Cold War between the United States and the Soviet Union, which lasted from the end of World War II in 1945 until the end of the Soviet Union itself in 1991. For this entire period, the intelligence community was at the center of a global struggle that was political, military, economic, scientific, and cultural. Even though the intelligence community was not created specifically to fight the Cold War, this long struggle has had lasting effects on how the intelligence community functioned far beyond the end of that struggle itself. Former DCI Robert Gates (1991-93) estimated that half of all intelligence activities during the Cold War were devoted to some aspect of the Soviet issue. First and perhaps foremost among the effects, the intelligence community emphasized a variety of national technical means by which it could remotely collect intelligence. This choice was dictated in large part by the physical and political nature of the Soviet state. The Soviet Union sprawled across the Eurasian landmass, many parts of it remote and inaccessible. It was also a police state in which everyone, foreigner and citizen alike, was under surveillance. But foreigners were also restricted to certain areas and denied access to many others. These two factors, in combination, made the usual means of intelligence collection—espionage—of only limited utility. Thus, the United States used a series of remote means—balloons, aircraft and finally satellites—to overfly Soviet territory in order to collect the needed intelligence. The two major types of technical intelligence were imagery (also called IMINT) and signals (SIGINT). There was also ongoing espionage (HUMINT), a nontechnical form of intelligence collection.

Second, the major area of emphasis was military capabilities. The Soviet Union tested its first atomic weapon in 1949 (greatly helped by Soviet espionage) and achieved the capability to launch an object into space and, by inference, the ability to hurl one across continents, in 1957. Thus, for the first time in well over a century, the United States was vulnerable to a potentially devastating attack. Even without intercontinental nuclear weapons, the vast conventional forces of the Soviet Union and its satellites in eastern and central Europe were of concern. Given the origins of the intelligence community in Pearl Harbor, this military emphasis was understandable. But it also led to an emphasis on studying tangible aspects of power—military forces, economic data (much of it meaningless for the Soviet Union), industrial production, and so on.

The Post-Cold War Transition

The Soviet Union collapsed in 1991, ironically driven over the brink by a failed coup staged by Soviet conservatives opposed to the reforms of the last Soviet leader, Mikhail Gorbachev. Although the United States and its allies had truly won the Cold War, there was something disquieting about the victory for the entire national security network. "The enemy we had come to know and love," as one senior defense intelligence official called the Soviet Union, had ceased to exist. What would be the new area of emphasis?

In an initial attempt to answer this question, then-DCI Robert Gates asked various departments and agencies for their intelligence needs, including many agencies that had not been viewed traditionally as intelligence community customers. It is important to remember that intelligence is a service provided to policymakers, and that they are supposed to establish the intelligence agenda. Gates was sincere in his effort to canvass for requirements and priorities, but critics accused him of "shopping" for new work to replace the lost Soviet target or of pandering to policy customers. Gates' tenure was cut short by the defeat of President George H. Bush (1989-93) by Bill Clinton in the 1992 election. The new Clinton administration (1993-2001), however, faced the same problem—what should the focus of the intelligence community be?

Some facilely answered this question by stating that the mission of the intelligence community had changed. This response only betrayed misunderstanding as to the purpose of the intelligence community. The mission of the intelligence community—supplying intelligence to policymakers so as to reduce their uncertainties as they make decisions— had not changed. What changed was the intelligence target and policymaker requirements. The mission was and is constant.

After several false starts in terms of his overall approach to foreign policy, in 1995 President Clinton signed Presidential Decision Directive 35 (PDD-35), which set out his administration's intelligence priorities. The first was support to military operations (SMO), meaning any and all intelligence support to the military in all of its roles. The second priority was a group of so-called Hard Targets, that is, issues that potentially threatened U.S. interests and were also difficult against which to collect intelligence. The Hard Targets included the so-called rogue states of Cuba, Libya, Iran, Iraq and North Korea; and the transnational issues. The transnational issues—proliferation of weapons of mass destruction (WMD), terrorism, crime, narcotics, and environmental issues—are not necessarily located in one nation or region and may not even involve nation states. All the other issues fell into what was called Global Coverage. It was understood that the Hard Targets would get the primary emphasis and Global Coverage less emphasis. Beyond this taxonomy, there were also tiers that described the relative priority of each issue. Issues could move up and down in the tier structure, depending on their importance. PDD-35 offered a more coherent and comprehensive outline of intelligence requirements and priorities, but the nature of international relations was such that on any given day any of several issues could claim to be the most important. The focus forced upon the intelligence community by the Soviet threat could not be recaptured, even under PDD-35.

The Terrorist Problem

Terrorism has been a recurring problem in the modern world. In the late nineteenth century, anarchists assassinated several world figures, including U.S. president William McKinley (1897-1901) in 1901. In the aftermath of World War I and the Russian Revolution, there was a new outbreak of terrorism, including bombings and assassination attempts in the United States, the so-called Red Scare.

Terrorism had been a U.S. intelligence concern for decades before the 2001 attacks. In the 1970s the focus was on independent radical groups, such as the Baader-Meinhoff gang in Germany, the Japanese Red Army Brigade, or the Italian Red Brigade, who either staged attacks for reasons of their own or allied themselves with other groups, particularly radical Arabs. During the Reagan administration (1981-89), the focus became state-sponsored terrorism. There was both the belief and intelligence to indicate that certain states were harboring terrorists, training them, and providing a range of logistical support. The United States sought to identify these states and to punish some of them for their actions. Several Soviet satellites (communist nations with ties to, and often bordering, the Soviet Union) were implicated, as were Libya and Iran.

In response to Libyan involvement in the bombing of a disco in West Berlin frequented by U.S. servicemen, the United States conducted an air raid in 1986 against the compound of Libyan leader Muammar Qadhaffi. Libya, in turn, planted a bomb on the Pan American flight 103 from London that exploded over Lockerbie, Scotland, in 1988. In the mid-1990s, however, the emphasis of U.S. policy on terrorism shifted to independent terrorist groups. Certain state sponsors were still of concern, but the emphasis had shifted to groups with increasingly potent capabilities and shifting bases of operations. For a variety of reasons, many of these groups had Arab connections, either as a direct result of the Arab-Israeli conflict or, as in the case of al-Qaeda, spawned in the successful U.S.-supported effort to force the Soviets out of Afghanistan.

Terrorism as an Intelligence Target

Several aspects of terrorism make it a difficult intelligence target. Terrorism is covert (meaning planned and carried out in secret), it is highly mobile, and it can operate with a minimum of capabilities that are distinguishable. All of these attributes run counter to the intelligence collection array that has been developed over decades, primarily to collect against the old Soviet target. Signals intelligence is usually a major means of collection against nations. Most modern states have easily identified telecommunications capabilities that provide a variety of useful intelligence. Militaries communicate regularly and extensively. They also tend to hold exercises according to a set annual schedule. Terrorist groups may rely on a variety of modern communications—cellular phones, faxes, the Internet—but these are never the same extensive networks that are found in nations. A further complication is the fact that U.S. intelligence collection capabilities are constantly becoming better known in unclassified sources, particularly the press. These revelations or leaks provide any potential target of U.S. collection with a ready-made guide on how to avoid that collection. For example, terrorists know to avoid using cell phones or faxes as much as possible to evade both yielding information and being located. Cell phones, having become relatively cheap, can be used once and then discarded, which also foils collection.

Imagery, which was so useful against the Soviet Union, also suffers when used on terrorism. The Soviet Union was a large, complex, and highly visible target with a large permanent infrastructure. Military bases are often large and usually are built in familiar and recurring patterns. Military exercises often involve large units, whose deployments are observable. Many terrorist facilities have few distinguishing characteristics to assist in identifying them. Large training camps may be an exception, but these are not overly valuable targets as terrorist groups rarely have significant assets concentrated in any one place. Terrorism itself most often operates with very small groups, not large formations.

Many people familiar with intelligence capabilities argue that espionage is the best way to collect on terrorism. There has been a long, recurring debate in the U.S. intelligence community—even before the 2001 attacks—over the degree to which U.S. intelligence is dependent on technical collection when, for certain types of intelligence, espionage would be better. The basic argument runs along the following lines: technical intelligence is very useful in collecting on capabilities (IMINT and SIGINT) and can offer insights into plans (via SIGINT), but it cannot get close to its target. Moreover, all technical collection depends on there being something to collect—either something to be imaged or communications to be intercepted—both of which can be avoided. They are, in effect, "where and when" intelligence sources. Inserting a spy into an organization offers the possibility of insights into plans and also has more continuous access to the target. This debate resurfaced in the aftermath of the September 2001 attacks.

HUMINT

Generically, all of the above arguments have validity. But it is simplistic to argue that improved espionage capabilities are an easy remedy to collecting on terrorism. First, spies are not an "on the shelf" capability who can be immediately ordered to this or that part of the world to collect intelligence. Successful HUMINT requires time for preparation. The spy needs to have proficiency in the language or dialects of the place in which he or she will be operating. Combating modern terrorism, unfortunately, calls for language skills that tend to be rare in the United States, including Arabic. The spy needs to blend in with his or her surroundings in terms of ethnicity and have a plausible reason for being in the place where the espionage will take place.

Typically, there are two ways to accomplish this. Some spies are posted to an embassy, where they have a "daytime" job that provides them with official cover. In the case of terrorism, however, we are often dealing with places where there is no embassy or consulate out of which to work. The other means is what is called non-official cover (making the spy a "NOC," pronounced "knock"). The NOC usually has a job or other plausible reason for being in this location and has no overt connection to the embassy. NOCs can have a variety of jobs—lawyer, salesman, etc.—that give them a plausible reason for being in that location and has no overt connection to a foreign government. In the case of terrorism, however, this, too, can be a problem. For example, there were few, if any, plausible reasons to be in Kabul, Afghanistan, in the months preceding the terrorist attack that would have provided a NOC with cover. Two final factors also make HUMINT more difficult. Even if one can establish a plausible reason for being in the right location, most terrorist organizations do not run major recruiting campaigns. Nor do they have an overt presence in many other states, the way nations do via their embassies. There is still the question of being able to locate and penetrate the group. Even if one penetrates the group, the inner circles tend to be very small and very well known to one another. Access to the key terrorists may still be difficult if not impossible. Finally, it is highly likely that a new recruit will be asked to prove himself. This means, in short, taking part in some sort of terrorist activity. Here we enter an ethical realm with few sure guideposts. Is there a level of activity or violence beyond which a spy should not be allowed to go? What is it—kidnapping, bombing, a single murder, multiple murders?

One means of extending HUMINT capabilities is to form partnerships with the clandestine services of other nations. This type of relationship is known as foreign liaison. Each nation has distinct HUMINT advantages in terms of regional knowledge, political relationships, history, and so on. Foreign liaison relationships are based on mutual exchanges of information, if not simultaneous, then over time. They also depend on a willingness to trust one another, given the sensitivity of the intelligence being exchanged—both requirements or actual intelligence. As useful as these foreign liaison relationships can be, their utility against terrorism may be more limited, as few intelligence organizations that are friendly to terrorists are likely to be helpful to the United States. The ambiguous stance of Pakistan's military intelligence, Inter-Service Intelligence (ISI), is a case in point. ISI had good relations with the Taliban, which should have been useful to the United States. But Pakistani president Pervez Musharraf clearly had doubts as to ISI's loyalty to the Pakistani leadership, finding it necessary to remove some senior ISI officers after he decided to support U.S. military operations responding to the September 2001 attacks. He has since taken steps to reduce greatly the size and influence of ISI.

There are also important policy issues, which must always be paramount when reviewing intelligence. As was noted above, intelligence exists solely to service policy. Intelligence exists to provide policymakers with as much information (intelligence) as possible so as to reduce the uncertainty inherent in most decisions. Intelligence officers do not, however, make policy. In U.S. practice, at least, there is a strict "line" separating intelligence from policy. The policy function is always the dominant one.

To put it in very stark terms, policymakers can exist and function without an intelligence community, but an intelligence community can not exist and function without policymakers. Intelligence has no independent function. Prior to the September 2001 attacks, the United States appeared to respond to terrorism on a local level, rather than globally. As noted, the United States staged a bombing raid against Libya. There were also several instances in which cruise missiles were used as a response to terrorism, attacking specific facilities. These included the military intelligence headquarters in Iraq in retaliation for the attempted assassination of former President George H. Bush; a suspected chemical weapons factory in Sudan; and al-Qaeda facilities in Afghanistan in retaliation for attacks prior to September 2001. This is not to suggest that the intelligence community was not active. In the mid-1990s, for example, the United States, working with France, captured the terrorist known as Carlos the Jackal. All of these responses, however, were far less than that seen after the 2001 attacks. It is also important to recognize that prior to the 2001 attacks there probably did not exist sufficient political support for a more aggressive policy against terrorism, which was another important limitation.

The political nature of terrorism also imposes limits on intelligence collection and analysis. When dealing with other nation states, there is usually the expectation of some level of give and take, or of fairly rational action/reaction. Even when dealing with states whose avowed aims are to change the international status quo, such as Nazi Germany or the Soviet Union, there was the expectation, up to a certain point, that some basis for dialogue or negotiation could be achieved. The very factor of dealing with a nation state and the possibility of dialogue predicates certain types of intelligence collection and certain lines of analysis. These might include an assumed level of rationality, the likelihood of a certain amount of give and take, an assumed unwillingness to put large portions of one's own population at risk, and so on. Little or none of this holds true when dealing with terrorist groups as opposed to their nation state supporters. Terrorists, by definition, cannot adhere to the status quo. To do so removes their very reason for being. They may issue political demands, but few states are going to be willing to meet these demands because they are either impossible, or because of the fear that this will only validate the terrorism and inspire more of it. Thus, many of the policy assumptions that may guide intelligence when dealing with nation states do not hold true for terrorists.

The September 2001 Attacks

It is difficult to argue that the terrorist attacks in September 2001 did not represent some level of failure, but the nature of, and reasons for, that failure are important and not wholly discernable as yet. More went wrong than simply intelligence. There were flaws in airport security (the responsibility of the Federal Aviation Authority, or FAA); the Immigration and Naturalization Service (INS); and the FBI, responsible for combating terrorism within the United States. This list does not mean that there are not also responsibilities of the intelligence community. But the failure was more systemic than that.

Were the attacks a "new Pearl Harbor," as many have argued? The basis for the comparison is easy to understand, but it also ignores important differences between the two events. In both cases the hostility of the perpetrator towards the United States was well known. U.S.-Japanese relations had been deteriorating since at least 1937, when Japan commenced its overt war in China. Similarly, Osama bin Laden had been overtly hostile to the United States since 1990, when the United States deployed troops into Saudi Arabia in reaction to Iraq's invasion of Kuwait. There had also been previous attacks by bin Laden, including the bombing of two U.S. embassies in Kenya and Tanzania in 1998 and an attack on the USS Cole in Yemen in 2000. In both cases there were expectations of increased hostility. In the case of Japan, U.S. policy-makers widely expected some new aggression, but wrongly assumed it would be directed against vulnerable European colonies in Southeast Asia or the East Indies, not against the United States. In the case of bin Laden, DCI George Tenet issued a series of warnings of future attacks. In both events, despite these expectations, the attacker achieved surprise. At this point, however, the comparisons are no longer apt, for reasons that are also significant. Japan's attack was strategic in nature, aimed at crippling the U.S. Pacific Fleet, which had been deployed to Hawaii to deter future Japanese aggression—a classic case of a failed deterrent that became, instead, the target. Bin Laden's attack in September 2001 was mostly symbolic in nature and did not affect U.S. power. Finally, Japan and its Axis allies had the capability to defeat the United States militarily and destroy its way of life. Neither of these factors is true for al-Qaeda. This may not ameliorate all of the intelligence aspects of the September 2001 attacks, but it does place the attacks in their proper context.

As emotionally devastating as the 2001 attacks were, it is also important to keep in mind the very small number of people who had to be involved in the planning and execution of the attacks. There were 19 hijackers (with discussions as to whether or not there is a "missing" twentieth hijacker), not all of whom may have known the true nature of the operations, perhaps suspecting it was a "traditional" airplane hijacking as opposed to a suicide attack. We can assume that bin Laden and several of his chief lieutenants knew of the plan. We can even posit, solely for the sake of argument, that there may have been others in the United States in a variety of support roles who may or may not have been witting as to the attack's true nature. This still may total fewer than three dozen people, a very small number with extremely low visibility.

The intelligence community has spent decades developing the concept of indications and warning, meaning being able to detect the precursor signs of an activity. Indications and warning, or I&W as it is known to intelligence professionals, is a direct outgrowth of the origins of the intelligence community in Pearl Harbor. I&W is primarily a tool of military analysis, seeking signs that will alert one to an impending attack. During the Cold War, there was a heated debate about the I&W that might be seen before a conventional Soviet attack in central Europe. Some analysts held that there would be tell-tale signs: supplies moving forward, reserves being called up, and so on. Others argued that the Soviets had sufficient forces and supplies in place to attack from what was called a "standing start," without any precursor activity. Neither theory was put to the test, fortunately. But this debate does give one a feel for what is involved in I&W. For terrorism, however, there is little, if any, reliable I&W. The groups involved are small, as are the weapons of choice—previously, bombs mounted in common modes of transportation, and later the actual airplanes themselves. Thus, there is a large gap between knowing that a group is hostile to the United States, or even that it has carried out attacks and is likely to do so in the future, and knowing the location and means of the next attack. With a society as open as is that of the United States, and with as many overseas interests as the United States has (embassies, bases, military units), terrorists will always have some advantage of choosing the time, location, and means of their next attack. Which is why, after the 2001 attacks, the George W. Bush administration (2001-) decided that the best strategy was to take the war directly to the terrorists rather than to try to defeat each attack.

Intelligence and the War on Terrorism

The first two tasks of the intelligence community in the immediate aftermath of the September 2001 attacks were (1) to determine if there were more attacks coming; and (2) to determine who was responsible for the attacks. We do not know from the public record if other attacks were planned or not, or if some were thwarted either by U.S. operations or by the sudden clamp down on security in the United States. Based on the currently available information, including the tapes of bin Laden, it would seem that no immediate follow-on attacks were planned. Although the terrorists apparently did not expect the level of destruction they achieved, neither did they expect the eventual U.S. response. Intelligence linking bin Laden to the attacks apparently was established fairly quickly.

According to press accounts that relied on interviews with senior policymakers, once President Bush had decided upon a strategic—as opposed to a tactical—response, the intelligence community was quickly ready with an operational plan to link up with Afghan opposition forces and attack al-Qaeda's supporters, the Taliban. In other words, rather than launching new, limited retaliatory strikes, as had been used in the past, the Bush administration decided to destroy al-Qaeda and its supporters. This plan, which involved using U.S. paramilitary experts to link up with the Northern Alliance to provide military advice and to serve as targeteers for the bombing campaign, reveals another aspect of intelligence—the operational arm. Intelligence operations can involve a large range of types, from propaganda campaigns to economic disruption to political intervention to coups to paramilitary campaigns.

Any and all of these types of operations (also called covert actions) can be controversial, although the paramilitary operations have tended to be more so than the others. There are several reasons for this. Paramilitary operations involve sizeable armed forces and combat. They tend to take longer than other types of operations and also run a greater risk of being inconclusive. They also raise questions about the obligations of the provider of paramilitary support to the indigenous forces in the field, a key issue, especially when an operation shows no sign of success or is an actual failure. The political costs of failure in paramilitary operations are also very high. The United States has experienced paramilitary operations that failed, such as the attempted invasion of Cuba at the Bay of Pigs (1961); operations that were inconclusive, such as the Contra war in Nicaragua (1980s); and operations that were successful, such as support to Afghanistan to resist the Soviet invasion (1980s). These last two operations are interesting because they were contemporaneous, and yet one, Afghanistan, enjoyed wide political support in Congress, while the other, the Contras, was the subject of often rancorous debate. Also, one of the results of the Afghan operation was the eventual rise to power of the Taliban due to the internal fighting that erupted after the Soviets were defeated.

The CIA's operational plan for Afghanistan after September 2001, which was melded with the use of regular U.S. military Special Forces units, was tremendously successful. The Taliban was defeated with a minimum of U.S. ground force involvement and with the cooperation of indigenous troops, which is important to the political future of Afghanistan.

One interesting aspect of the post-attack debate on intelligence was a revival of the discussion over the propriety of assassination as an operational tool. This debate goes back to the mid-1970s, when a Senate select committee investigating allegations of intelligence community wrongdoing issued a report detailing U.S. government involvement in or witting knowledge of assassination attempts against foreign leaders. The targets included Fidel Castro; Patrice Lumumba, first premier of the Congo; Ngo Dinh Diem, president of South Vietnam; General Rene Schneider of Chile; and Rafael Trujillo, dictator of the Dominican Republic. The report also noted that none of the victims specifically targeted by the United States—primarily Castro—died as a result of U.S. actions. The report did result in strong public reaction and the prohibition of the use of assassination in three successive executive orders, signed by presidents Gerald Ford, Jimmy Carter, and Ronald Reagan.

The executive order signed by President Reagan in 1981 is still the governing order for the intelligence community. There was anecdotal evidence to suggest that public support for the ban had been waning. In the aftermath of the terrorist attacks, many called for a lifting of the ban, with bin Laden and his lieutenants as likely targets. In actuality, the point was moot. The United States declared itself to be at war with terrorism and under those conditions, terrorists or their leaders became legitimate military targets whose deaths would not fall under the ban. Still, the debate was an interesting indicator of one reaction to the attacks and of the greater license that many would subsequently allow intelligence operations.

Intelligence Innovations

Two intelligence innovations in the Afghan phase of the terrorism war are worth noting, both involving imagery. First, the United States continued to make advances in its use of unmanned aerial vehicles (UAVs), or drones. These pilotless aircraft, operated remotely from long distances, have the ability to fly over areas of interest for long periods of time (unlike satellites, which fly in rapid orbits), and without risk to pilots (unlike U-2s or other spy planes). Thus, UAVs can provide more continuous coverage and, when using video cameras, can supply "real-time" coverage. A second UAV innovation has been the mounting of air-to-surface missiles on the UAVs. This allows targets to be attacked as soon as they are identified, rather than waiting for aircraft to be called in and then flown to the target.

The other imagery innovation has been the greater use of commercial imagery by the intelligence community. Imagery satellites launched and flown by private companies have now achieved resolutions (that is, the ability to identify an object of a certain size in a photograph) of 0.8 meters—just over 31 inches—or better. In order to take advantage of this capability, NIMA purchased the exclusive and perpetual use of commercial images of Afghanistan taken by the Ikonos satellite, operated by Space Imaging, a private U.S. company. This not only augmented imagery collection, reserving even more powerful intelligence satellites for tasks that only they can perform, but also denied this imagery either to current or potential foes or to nations eager to assess the progress and capability of U.S. forces. This purchase also denied the use of the images to the press, which might have been eager to use them as part of its own independent analysis of the war.

The rapid success of U.S. forces in Afghanistan has, to some extent, helped to undo some of the damage done to the intelligence community's reputation by the September 2001 attacks. This is not a unique occurrence. In 1961 the CIA was badly damaged by the failure of the abortive Bay of Pigs invasion. In 1962 the intelligence community's performance in the Cuban missile crisis undid much of that damage in terms of the intelligence community's relationship with policymakers by providing sufficient warning of the emplacement of the missiles to allow President Kennedy (1961-63) to act before the missiles were operational. The intelligence community also provided U.S. policymakers with detailed intelligence about Soviet military capabilities, which also gave them greater confidence in their decisions.

Recent History and the Future

The Future of U.S. Intelligence

The terrorist attacks revived the ongoing but sometimes unfocused debate over the future of the intelligence community. This debate is almost as old as the intelligence community itself, and is driven by many factors. Chief among them are the relative novelty of intelligence as a permanent function of government, the levels of expectation about intelligence performance, and the inherent difficulty of managing a broad and diverse community that deals in the often intangible commodities of analysis and interpretation.

The most recent debate over intelligence had come in the mid-1990s, in the aftermath of the Cold War. As discussed above, some felt that the intelligence community's mission had changed and that it had to be re-examined. Others understood that the mission had not changed but were concerned that the major influence of the Cold War on the intelligence community had left forms and practices that might be outmoded. Several major reviews were conducted, the two most prominent being the Commission on the Roles and Capabilities of the United States Intelligence Community, also known as the Aspin-Brown Commission after its two chairmen, former secretaries of defense Les Aspin and Harold Brown, and the House Intelligence Committee's study, IC21: The Intelligence Community in the 21st Century.

The two studies did not use similar approaches and came up with different recommendations—as well as some that agreed—but it is striking that both reports urged that the intelligence community function more like a true community, that is, one in which all the disparate agencies function as part of a more integrated whole. This remains the most central issue in what is called "intelligence reform," the degree of authority exercised by the DCI over the intelligence community. According to press reports, a study ordered by President Bush and headed by former national security adviser Brent Scowcroft recommended that the DCI have greater authority over the "national" agencies—particularly NIMA and NSA. Predictably, the intelligence staff in the Defense Department have been opposed to this concept.

The debate over the future of the intelligence community goes much further, however, than the authority of the DCI or organizational shifts. There are many other significant issues. Some argue that the intelligence community has to break away from its current bureaucratic structure—built around either geographic or topical offices—and adopt a more fluid approach, with groups of specialists coming together and disbanding as the issues require, perhaps working in "virtual" proximity rather than in the same actual building. The intelligence community began moving towards a somewhat more adaptive structure in the 1990s, when DCI Gates created a series of centers responsible for some of the more pressing transnational issues, in which analysts from several agencies could be brought together. The centers are generally seen as being successful, although some question the degree to which they are true "community" centers, given the invariable CIA dominance in them.

Such an approach also addresses the important issue of "surge," or the ability of the intelligence community to respond to sudden crises. The United States has global responsibilities and interests. The intelligence community, in theory, should be able to respond to and support any of those interests. In reality, however, all intelligence resources—like all other government resources—are finite. Therefore, the intelligence community picks and chooses among the issues it can cover in depth, those that can get some coverage and those that are largely left wanting. The choices tend to reflect policymaker requirements. When a crisis breaks, however, it often comes in one of those areas where there has been little or no coverage. The intelligence community must then "surge" resources from current coverage to this new area. During the Cold War, this was less problematic given the predominance of the Soviet issue. In the post-Cold War period, it has become more of a problem, as there has been no predominant issue. One of the other suggestions advocated to address surge is the greater use of outside experts and retirees in what would be an intelligence reserve. External regional experts, for example, could be especially useful when attention shifts to a region whose languages or cultures are not widely studied or known. Authority to use such a reserve exists, but the intelligence community's reaction has been tepid at best.

Some would take these new organizational concepts even further and treat intelligence like a commodity to be purchased by interested policy-makers. Advocates of this approach believe that the introduction of "market forces," even in an economy as small and as specialized as this, would mandate changes, weeding out inefficient processes and forcing the intelligence community to focus only on those issues that really matter to policymakers. A problem with this approach is that it puts almost the entire emphasis of the intelligence community on what is called "current intelligence," that is, the issues that are on the "front burner." This flies in the face of decades of experience, which tell us that at least a few of next year's crises will come in areas that had previously been backwaters. One of the major contributions that intelligence makes at times like those is the ability to call on veteran analysts who have been following these secondary or tertiary issues for years and who can bring everyone up to speed on them. The market force model would seem to have little room for this type of background coverage, which has no immediate utility but may pay off handsomely in the years to come, reflecting the inherent uncertainty of the overall intelligence agenda. For example, as late as September 10, 2001, no U.S. policymaker would likely have been able to envision circumstances that would lead to a major U.S. military engagement in Afghanistan. This, however, is the essence of intelligence.

Intelligence Collection Issues

There are several collection issues in the reform debate. The issue of the proper balance between technical and human collection operations has already been discussed above. Another important reform collection issue is the imbalance between the amount of technical intelligence that is collected and the much smaller amount of images or signals that is processed and exploited into useable intelligence for analysts. Technical collection systems do not take a photograph or record a conversation per se. Rather, these are captured in digital form, which then must be processed into an image or audio tape. The collection systems tend to act like large vacuum cleaners, sweeping up much more than can be used. This is sometimes referred to as the TPEDs problem, which stands for tasking, processing, exploitation, and dissemination, of which the middle two—processing and exploitation—are the most problematic. Congress has been especially vocal on this issue, holding that it is difficult to allocate more funds for ever more powerful collection systems if the same levels of intelligence—or less—are ultimately made available to analysts.

Another collection issue is the internecine competition that goes on among the various collection disciplines. Although each of the INTs, or methods of intelligence collection, has its own strengths and weaknesses, there are bureaucratic and budget imperatives that foster an ultimately wasteful competition among the INTs. No one has the authority to adjudicate among the INTs, or to determine which ones should collect, or not collect, on a given issue. Some, including the House Intelligence Committee's IC21 study, have recommended that a single individual have this authority, at least for the technical collection disciplines.

This is a rich agenda and one that engages the many practitioners, students, and aficionados of intelligence. The venue in which these concepts might be debated with an eye towards reform is unclear. The only scheduled review of intelligence community performance in the events leading up to September 11, 2001, is that planned by the House and Senate Select Committees on Intelligence. The actual agenda of that investigation is not yet decided upon. The breadth or narrowness of that inquiry will determine the range of the eventual recommendations. It is also likely that there will be other investigations as well. (Pearl Harbor was the subject of multiple investigations both during and after World War II.) It will be important, in any recommendations that follow from these investigations, to be able to show how recommended changes address specific problems within the intelligence community. Pet causes or fashionable intellectual hobbyhorses will not survive without this. It is also important to work from a reasonable set of expectations about intelligence performance. Omniscience is beyond the capability of any intelligence organization. What level of success should be expected against a threat as difficult to track as terrorism? Repetitions of the September 2001 attacks are not acceptable, but a "zero tolerance" policy is also not realistic. Finally, it is important to remember that the intelligence community has to address more than terrorism. Any proposed reforms should create a more capable intelligence community across the board, rather than a response to one issue alone. After all, there are still more issues on the agenda than just terrorism.

Conclusion

The future of U.S. intelligence in the twenty-first century will depend on more than any new investigations, although these will be a factor. It will also depend on how well the intelligence community performs in the war against terrorism, while also keeping watch on the other issues on the national security agenda. It is not just a question of increased budgets, or more satellites or more spies. The intelligence community has to manage well within a fixed amount of resources, covering the necessary issues while maintaining a certain degree of flexibility. Ultimately, intelligence is one of the most difficult government functions to assess in terms of success and effectiveness. There is much about intelligence that is inefficient. The intelligence community collects more than it can process, and it tries to cover issues that, to an outsider, may look secondary or worse. Because of the requirements of security, much that goes on in intelligence is seen by very few. Finally, as noted above, it is important to have a reasonable standard of expectations for intelligence. Intelligence can perform well or poorly on the various issues it covers, but perfect knowledge or warning are not options, especially when dealing with a set of national security interests as broad and diverse as those of the United States.

Bibliography

Balz, Dan and Bob Woodward. "Ten Days in September," Washington Post, January 27-February 3, 2002.

Berkowitz, Bruce and Allen Goodman. BEST TRUTH: Intelligence in the Information Age. New Haven, CT: Yale University Press, 2000.

Betts, Richard K. "Fixing Intelligence," Foreign Affairs,January/February 2002, vol. 81, no. 4: 43-59.

Cilluffo, Frank J., Ronald A. Marks, and George C. Salmoiraghi. "The Use and Limits of U.S. Intelligence," The Washington Quarterly, 25, no. 1, (Winter 2002): 61-74.

Commission on the Roles and Capabilities of the United States Intelligence Community. Preparing for the 21st Century: An Appraisal of U.S. Intelligence. Washington, DC: U.S. Government Printing Office, 1996.

Johnson, Loch K. Bombs, Bugs, Drugs, and Thugs: Intelligence and America's Quest for Security. New York: New York University Press, 2000.

Lowenthal, Mark M. Intelligence: From Secrets to Policy.Washington, DC: Congressional Quarterly Press, 2000.

U.S. Congress. House Permanent Select Committee on Intelligence. IC21: The Intelligence Community in the 21st Century. Staff Report. Washington, DC: U.S. Government Printing Office, 1996.

Mark M. Lowenthal

Chronology

1940-45 Office of Strategic Services (OSS) is established as the first U.S. national (rather than military) intelligence service.

December 7, 1941 The Japanese surprise attack on PearlHarbor is seen by many as the main cause for the creation of a post-war intelligence community.

1947 Passage of National Security Act creates the Central Intelligence Agency and gives de jure status to the Director of Central Intelligence.

1952-62 The U.S. intelligence community is developed with the creation of most of its major components, including the National Security Agency, State Department Bureau of Intelligence and Research, and Defense Intelligence Agency.

1947-91 The Cold War ensues between the United States and the Soviet Union, in which the U.S. intelligence community plays a major role and which also has long-lasting effects on the intelligence community.

1980s-90s U.S. counter-terrorism policy emphasizes state-sponsored terrorism.

1990s-present U.S. counter-terrorism policy emphasizes independent terrorist groups.

August 7, 1998 Al-Qaeda terrorists attack U.S. embassies in Kenya and Tanzania with bombs, killing 12 U.S. citizens and 240 Africans.

October 2, 2000 Al-Qaeda terrorists attack the USS Cole in Yemen, killing 17 and wounding 39.

September 11, 2001 Al-Qaeda terrorists engage in a multi-plane attack on the United States, hitting the Pentagon in Arlington, Virginia, near Washington, DC, and destroying the World Trade Center in New York.

October 7, 2001 Operation Enduring Freedom begins against al-Qaeda and the Taliban in Afghanistan.

The U.S. Intelligence Community

The intelligence community is complex and diverse. The following are the major components, listed alphabetically:

• CIA: Central Intelligence Agency, responsible for collection (espionage), analysis, and operations (known as covert actions).
• CMS: Community Management Staff, assists the DCI in carrying out his community-wide responsibilities.
• DIA: Defense Intelligence Agency, primarily responsible for analysis affecting more than one military service, and for collection (espionage) via the Defense HUMINT Service (DHS).
• INR: The State Department's Bureau of Intelligence and Research, responsible for analysis.
• Military Service Intelligence Components: army, air force, navy and marine intelligence, each serving the specific needs of their service.
• NIC: National Intelligence Council, a group of the most senior analysts (called national intelligence officers, or NIOs), who prepare and coordinate national intelligence estimates (NIEs).
• NIMA: National Imagery and Mapping Agency, responsible for imagery intelligence and for defense mapping needs.
• NRO: National Reconnaissance Office, builds and operates intelligence satellites.
• NSA: National Security Agency, responsible for all signals intelligence and for the protection of U.S. communications.

Types of Intelligence Collection

There are a variety of means of collecting intelligence, sometimes referred to as collection disciplines or as INTs.

Imagery (IMINT or PHOTINT) is pictures, although there are a variety of ways of taking them, including electro-optical, which is akin to a standard picture; infrared, which reads the heat coming off objects and surroundings; and radar, which can locate objects through cloud cover.

Signals (SIGINT) actually has several subsets. Communications (COMINT) refers to actual exchanges between people, most often by telephone. It can mean voice, fax, or computer communications. Electronics (ELINT) refers to electronic emissions from weapons or tracking systems. Telemetry (TELINT) are the data released by weapons during tests.

Espionage (HUMINT) is the most familiar collection type, also known as spying.

Measurement and signatures (MASINT) is a somewhat obscure collection type, referring to weapons capabilities and industrial activities obtained from multispectral and hyper-spectral frequencies and some fairly exotic processing.

Open source (OSINT) refers to any information that is neither classified nor proprietary and includes news media, academic papers, government reports, the World Wide Web, and so on.

The U.S. Intelligence Community

Intelligence Issues for Congress

The U.S. Intelligence Community continues to adjust to the post-Cold War environment. Congressional and executive branch initiatives have emphasized enhancing cooperation among the different agencies that comprise the Community by giving greater managerial authority to the Director of Central Intelligence (DCI).

Priority continues to be placed on intelligence support to military operations and on involvement in efforts to combat narcotics trafficking and, especially since September 11, 2001, international terrorism. Growing concerns about transnational threats are leading to increasingly close cooperation between intelligence and law enforcement agencies. This relationship is complicated, however, by differing roles and missions as well as statutory charters. The September 11, 2001 terrorist attacks, for which no specific warning was available, have led to increased emphasis on human intelligence, better cooperation between law enforcement and intelligence agencies, and on consideration of organizational changes to the Intelligence Community.

Intelligence Community leadership and congressional committees have expressed determination to enhance analytical capabilities. A major concern is an imbalance between resources devoted to collection and those allocated to analysis, with collected data much exceeding analytical capabilities.

In several regional crisis areas, the role of the U.S. Intelligence Community is especially important. Provisions for U.S. intelligence to monitor security arrangements between Israelis and Palestinians have been a significant factor in efforts to resolve Middle East tensions. Intelligence efforts have also been important in attempting to enforce U.N. sanctions on Iraq and monitoring peace agreements in Bosnia. Cruise missile and bomb attacks on Afghan targets in the campaign against the Taliban, and on Serbian targets during the Kosovo crisis have been heavily dependent upon precise targeting data provided by intelligence sensors. The mistaken attack on the Chinese Embassy in Belgrade resulted from faulty information provided by the Intelligence Community.

A particular concern for many in Congress has been the Intelligence Community's assessment of the missile attack capabilities of foreign countries, especially North Korea. Some believe that U.S. vulnerability to missile attack may arrive sooner than has been estimated by intelligence agencies.

CRS Issue Brief for Congress, received through the CRS Web. "Intelligence Issues for Congress." Updated January 8, 2002. Richard A. Best, Jr. Foreign Affairs, Defense, and Trade Division. Congressional Research Service. The Library of Congress.