Revamping the CIA
The terrorist attacks have once again exposed wide-ranging flaws in the agency’s operations.
One week after the terrorist attacks on the Pentagon and the World Trade Center, national security adviser Condoleeza Rice told the press: “This isn’t Pearl Harbor.” No, it’s worse. Sixty years ago, the United States did not have a director of central intelligence and 13 intelligence agencies with a combined budget of more than $30 billion to produce an early warning against our enemies.
There is another significant and telling difference between Pearl Harbor and the September 11, 2001, attacks: Less than two weeks after Pearl Harbor, President Franklin D. Roosevelt appointed a high-level military and civilian commission to determine the causes of the intelligence failure. After the recent attacks, however, President George W. Bush, Director of Central Intelligence George Tenet, and, surprisingly, the chairmen of the Senate and House intelligence committees adamantly opposed any investigation or post mortem. Sen. Bob Graham (D-Fla.), chair of the Senate Select Committee on Intelligence, said it would not be “appropriate” to conduct an investigation at this time; his predecessor, Sen. Richard Shelby (R-Ala.), agreed that any investigation could wait another year. The President’s Foreign Intelligence Advisory Board normally would request such a study, but the board currently has only one member, because the president has not yet replaced members whose terms have expired. The president’s failure to appoint a statutory inspector general at the Central Intelligence Agency (CIA) deprives the agency of the one individual who could have requested an investigation regardless of the CIA director’s views. Overall, the unwillingness to conduct an inquiry increases the suspicion that there may have been indicators of the attacks that went unheeded.
The failure to anticipate the attacks is merely the latest in a series of CIA failures during the past 10 years. The CIA spent nearly two-thirds of its resources on the Soviet Union but did not foresee the Kremlin’s collapse. Yet there was no investigation or post mortem of what went wrong in the CIA’s directorate of intelligence, nor were there major changes in the CIA’s analytical culture.
There was also the incredible but true saga of Aldrich Ames, the CIA officer who spied for the Soviet Union and the Russian Federation for nearly a decade, flaunting his KGB-supplied wealth and betraying the entire U.S. spy network inside Moscow. The Ames saga did lead to a 1994 study of the CIA’s clandestine culture that concluded, in the words of then-director James Woolsey, “It is a culture where a sense of trust and camaraderie within the fraternity can smack of elitism and arrogance.” A year later, in fact, then-director John Deutch learned that the CIA payroll included a Guatemalan colonel implicated in the murder of a U.S. citizen and, as a result, initiated efforts to reform the directorate of operations and to remove the thugs from the payroll. Predictably, the old boy network rallied in the name of the directorate and tried to stymie Deutch’s efforts.
Demilitarize intelligence gathering
Previous directors, particularly Deutch and Robert Gates, have done great harm to the CIA and the intelligence community by deemphasizing strategic intelligence for use in policymaking and catering instead to the tactical demands of the Pentagon. The CIA began to produce fewer national intelligence estimates and assessments that dealt with strategic matters and placed its emphasis on intelligence support for the war fighter. Gates, moreover, ended CIA analysis of key order-of-battle issues in order to avoid tendentious analytical struggles with the Pentagon; Deutch’s creation of the National Imagery and Mapping Agency (NIMA) at the Department of Defense (DOD) enabled the Pentagon to be the sole interpreter of satellite photography. This is particularly important because the Pentagon uses imagery analysis to justify the defense budget, to gauge the likelihood of military conflict around the world, and to verify arms control agreements. In creating NIMA, Deutch abolished the CIA’s Office of Imagery Analysis and the joint DOD-CIA National Photographic Center, which often challenged the Pentagon’s analytical views.
In its short history, NIMA has been responsible for a series of major intelligence disasters, including the failure to predict Indian nuclear testing in 1998, the bombing of the Chinese embassy in Belgrade in 1999, and more recently the exaggeration of the missile programs in North Korea and Iran. The failure to anticipate and record Indian nuclear testing stemmed from the Pentagon’s downgrading of South Asian intelligence collection and DOD’s low priority for counterproliferation. Open sources did a far better job of predicting the nuclear tests than did the U.S. intelligence community. To make matters worse, CIA Director Tenet told the Senate that the CIA could not monitor and verify the Comprehensive Test Ban Treaty and, for the first time in 80 years, the Senate failed to ratify a major international treaty.
The bombing of the Chinese embassy was attributed to the faulty work of NIMA as well as the inability of the CIA to conduct operational targeting for the Pentagon. Consequently, when the crew of a U.S. B-2 Stealth bomber skimmed over Yugoslavia and dropped three bombs on a building in downtown Belgrade, it actually believed that it had made a direct hit on the country’s arms procurement headquarters. Instead, three people were killed and 20 wounded, creating a diplomatic crisis with Beijing and key members of the NATO coalition. The CIA had never been responsible for operational targeting before, and as a result of the Belgrade disaster, Tenet has made sure that the agency stays out of the targeting business.
Leaving imagery analysis in the Pentagon’s hands allows the military to exaggerate strategic threats to the United States. Throughout the Cold War, military intelligence consistently exaggerated Soviet strategic power, particularly the quantity and quality of Soviet strategic forces and the capabilities of key weapons systems. The Air Force was particularly guilty of exaggerating Soviet missile forces, presumably in order to gain additional resources for U.S. missile deployment. At the same time, the uniformed military was not enamored with the intelligence capabilities of satellite photography and such surveillance aircraft as the U-2, and if it had not been for lobbying by the CIA and civilian scientists, the United States would not have had access to such technology until much later. When the CIA tried to create its own Foreign Missile and Space Analysis Center in 1963 to provide detailed intelligence information on offensive missile systems, senior Air Force generals unsuccessfully tried to stop it.
New intelligence priorities
Although the collapse of the Soviet Union and its Eastern European empire fundamentally altered the strategic environment, there has been no major effort to redefine U.S. national security and intelligence needs. The Soviet collapse created new areas of instability and policy challenges in the Caucasus, central Asia, and southeastern Europe, where the United States and the intelligence community possess few intellectual resources. And nontraditional security problems, which will define U.S. policy choices in the 21st century, have been given short shrift. These problems include water scarcity in the Middle East, social migration caused by coastal flooding in South Asia, infectious diseases in Africa and Russia, and contamination caused by nuclear and chemical weapons stored in the former Soviet Union.
The nontraditional national security problems that confront the United States could give the CIA a competitive advantage because of its data storehouse on oil reserves, demographics, and water supply. The CIA is in a position to provide information on a variety of environmental issues, using baseline data from satellite photography documenting global warming, ozone depletion, and environmental contamination. Spy satellites already provide key environmental data on Earth’s diminishing grasslands, forests, and food resources. Yet the CIA has not been forthcoming with its data, and the only politician who has ever made a serious effort to obtain such data and analysis–former vice president Al Gore–is on the sidelines. To make matters worse, there is a satellite sitting on the ground that is designed to collect such data, but the Bush administration will not pay to launch it.
With the proliferation of international peacekeeping missions, the intelligence community is a natural resource for providing political and military data to peacekeepers in places such as Afghanistan, Bosnia, Cambodia, and Somalia. The CIA should have assisted the United Nations (UN) monitoring programs in Iraq rather than running its own operations against Saddam Hussein. War crimes tribunals also require funds and expertise for collecting data on political and military officials, which would be a less difficult task if the political and biographic assets of the CIA could be used. And it is unlikely that global institutions such as the International Atomic Energy Agency can successfully monitor strategic weapons production in North Korea, Iraq, and Pakistan without support from the CIA.
Unfortunately, the CIA has shown little inclination to take on these tasks. Woolsey was lukewarm at best to the idea of sharing intelligence with international agencies. Deutch was stubbornly opposed to providing information to the UN, even though it would have been helpful in peacekeeping situations. And current director Tenet also does not have much interest in these activities.
Problems with covert action
There is no absolute political and ethical guideline delineating when to engage in covert action. However, Cyrus Vance, secretary of state in the Carter administration, articulated a standard two decades ago when he recommended covert action only when “absolutely essential” to the national security of the United States and when “no other means” would do. The CIA observed this standard in the breach when it placed world-class criminals such as Panama’s General Manuel Noriega, Guatemala’s Colonel Julio Alpirez, Peru’s intelligence chief Vladimiro Montesinos, and Chile’s General Manuel Contreras on its payroll. The CIA’s favorite “freedom fighter” in Afghanistan in the 1980s, Gulbuddin Hekmatyar, was also the country’s chief drug lord.
In addition to playing a role in overthrowing the democratically elected government of Chile in the 1970s, the CIA hired and protected Contreras despite his involvement in assassination plots in South America and the United States, including the car bombing in the nation’s capital of former Chilean Ambassador Orlando Letelier and his U.S. associate, Ronni Karpen Moffitt. Recently released documents demonstrate that the CIA placed Contreras on its payroll despite its acknowledgement that he was the “principal obstacle to a reasonable human rights policy” in Chile.
These unsavory assets had nothing to do with the collection of sensitive intelligence but were important to the CIA for the conduct of covert actions in South America that usually were counterproductive to the interests of the countries involved as well as to the United States. Montesinos, for example, was responsible for two decades of human rights abuses in Peru. Yet the CIA helped him flee the country in September 2000 to avoid standing trial for crimes that included the massacre of innocent civilians in the early 1990s. The CIA station in Amman approved an arms deal between Jordanian officials and Montesinos, although he was involved in a 1998 transfer of arms from Jordan to leftist guerrillas in Colombia, perhaps Washington’s most notorious enemies in Latin America. There is probably no stronger evidence of the ineptitude of the CIA’s directorate of operations.
We learned in 1999 that the United States and the CIA used the cover of the UN and the UN Special Commission (UNSCOM) to conduct a secret operation to spy on Iraqi military communications as part of an effort to topple Saddam Hussein. Neither the UN nor UNSCOM had authorized the U.S. surveillance, which Hussein cited as justification for expelling the UN operation. As a result, the most successful effort to monitor and verify Iraq’s nuclear, chemical, and biological programs was lost, and the credibility of multilateral inspection teams around the world was compromised.
Separating intelligence and operations
Any reform of the role and missions of the CIA must recognize that the agency performs two very different functions. The CIA’s clandestine operations, particularly covert action, are part of the policy process. Yet when paid agency assets are also the sources of intelligence reporting, the finished reports may be seriously flawed. CIA’s covert operations are approved and often designed by the White House and the State Department to support specific policies. The Bay of Pigs in 1961, which the inspector general of the CIA described as the “perfect failure,” and Iran-Contra in the 1980s, which violated U.S. law, demonstrated the ability of the directorate of operations to corrupt the analysis of the directorate of intelligence.
The CIA’s intelligence analysis, including national estimates and current reporting, must provide both an objective exploration of the situation for which policy is required and an impartial assessment of alternative policy options. Intelligence should play a role in setting the context for policy formulation, but it should never become an advocate for a specific policy. CIA Director William Casey and his deputy for intelligence, Robert Gates, slanted intelligence reporting in the 1980s to support operational activity in Central America and southwest Asia. In his memoirs, former Secretary of State George Shultz charged that the CIA’s operational involvement “colored” the agency’s estimates and analysis. The CIA’s distortion of Soviet strategic policy skewed the public debate on the Star Wars program in the 1980s, and similar distortions of the strategic capabilities of so-called rogue states have factored into the debate on national missile defense.
The decline of wizardry
During the worst days of the Cold War, the strategic position of the United States was enhanced by the scientific and technological successes of the CIA, which designed and operated some of our most important spy satellites as well as the U-2 spy plane. The CIA was heavily involved in the collection of signals intelligence and helped pioneer the technical analysis of foreign missile and space programs. Secret CIA installations eavesdropped on Soviet missile tests and gathered intelligence that was crucial to the success of arms control negotiations in the 1970s and 1980s. As a result, the CIA had advance knowledge of every Soviet strategic weapons system and up-to-date intelligence on the capabilities of these systems.
Unfortunately, the technological frontier has moved from Langley, Virginia, to Silicon Valley, and as a result, the CIA has lost much of its technological edge. In 1998, the CIA abolished its Office of Research and Development (ORD), which had been responsible for much of the agency’s success in the fields of technical collection and analytical intelligence. The CIA will no longer be on the cutting edge of advanced technology in the fields of clandestine collection and satellite reconnaissance and will be heavily dependent on the technology of outside contractors. ORD led the way in major breakthroughs in the area of overhead reconnaissance, including optics and imagery interpretation, which presumably are paying dividends in Afghanistan. Previous ORD technology, such as sophisticated facial recognition, will help in the war against terrorism but only if that technology is shared with the Immigration and Naturalization Service (INS), the FBI, and the Drug Enforcement Agency.
In addition to the weakening of the CIA in important areas of science and technology, the National Security Agency (NSA), which is responsible for collecting and interpreting signals and communications intelligence from around the world, has been weakened by a series of management decisions that have created serious problems. The NSA has been caught off guard by a series of new communications technologies that have compromised its intercept capabilities, including fiber optic cables that cannot be tapped, encryption software that cannot be broken, and cell phone traffic that is too voluminous to be processed. There is no question that a managerial revolution needs to take place throughout the intelligence community.
A new intelligence infrastructure
What the CIA and the intelligence community should be, what they should do, and what they should prepare to do are all less clear than at any time since the end of the World War II and the beginning of the Cold War. Throughout the Cold War, the need to count and characterize Soviet weapons systems against which U.S. forces might find themselves engaged, as well as the search for indications of surprise attack, focused the CIA’s efforts. Such clarity disappeared with the fall of the Soviet Union. The following steps are needed in order to design an intelligence infrastructure to deal effectively with the new and emerging national security problems.
Demilitarize the intelligence community. The mismatch between the tools of the past and the missions of the future has given rise to an increased militarization of the various intelligence agencies and an excessive reliance on CIA support for the war fighter. It is essential that the major intelligence collection agencies–NIMA, NSA, and the National Reconnaissance Office (which designs spy satellites), with their collective budget of at least $10 billion–be taken from DOD and transferred to a new office that reports to the director of central intelligence. This move would allow more leeway for spending the intelligence budget on analysis and sharing of information gathered by satellites, rather than the current emphasis on building satellites and other data collectors. According to press reports, retired general Brent Scowcroft, who is conducting a comprehensive review of the intelligence community for President Bush, favors such a transfer of authority, but Secretary of Defense Donald Rumsfeld and high-ranking members of the Senate Armed Forces Committee oppose it.
Revive oversight. The decline of the CIA during the past decade coincides with reduced oversight of the intelligence community by the Senate and House intelligence committees. Beginning with the chairmanship of Senator Shelby in 1994, the Senate committee has become less effective in providing oversight and in advancing much-needed reform. It is unusual to have more than two or three senators present at any given time, even at important hearings, and Senate committee members are limited to an eight-year term. (The House has a six-year term limit.) The number of open intelligence oversight hearings has dropped significantly, as has the number of nongovernmental witnesses invited to testify. Because the authorization bill for the intelligence community is imbedded in the defense budget, the Senate Armed Forces Committee is able to significantly modify the authorizations of the intelligence committee. The system worked when former Senators Sam Nunn and David Boren, who were close colleagues, chaired the armed services and intelligence committees, respectively, in the 1980s, but the system has broken down in the 1990s. The House intelligence committee chair, Rep. Porter Goss (R-Fla.), is a former CIA case officer who has acted as an advocate for the intelligence community and not a reformer.
There has also been an astonishing exchange of personnel between intelligence committee staffs and the agencies they oversee. Tenet and his chief of staff formerly served as the majority and minority staff chiefs, respectively, of the Senate intelligence committee. Other staff members went on to serve in a variety of other CIA posts: inspector general, chief of the legislative counsel’s office, chief of the Foreign Broadcast Information Service, deputy director of the Counter-Proliferation Center, and director of resource management for the directorate of operations. The current head of the NRO and the NRO’s inspector general both came from the Senate intelligence committee, as did the deputy director of intelligence programs at the National Security Council. It is unprecedented for one congressional committee to supply staff to so many senior positions at a major executive agency, which raises a disturbing question: Who will oversee the overseers?
Reduce covert action. Covert action could be radically reduced without compromising national security. CIA propaganda has had little effect on foreign audiences and should end immediately. The CIA should never be allowed to interfere in foreign elections.
Many problems that have been considered candidates for covert action were ultimately addressed openly by unilateral means or cooperatively through international measures, both of which are preferable to clandestine operations. Nuclear proliferation problems created by missile programs in Iraq and North Korea in the 1990s led to congressional calls for covert actions, but in both cases overt multilateral activity with the United States in a pivotal role contributed to denuclearization. The U.S. military was successfully involved in secret denuclearization of the former Soviet Union, clandestinely removing strategic weapons and nuclear materials from Georgia, Kazakhstan, and Moldova in the 1990s.
Separate operations and analysis. It is time to debate whether it is preferable to separate the CIA’s operational activity from its analytical work or continue running the risk of tainted intelligence. The issue is one of advocacy, ensuring that the provider of intelligence is not in a position to advance its own point of view in the policy process. The CIA’s heavy policy involvement in the war on terrorism will certainly call into question the worst-case views of the directorate of intelligence on terrorist threats at home and abroad.
Because there are few institutional safeguards for impartial and objective analysis, the intelligence community ultimately depends on professional personnel of the highest intellectual and moral caliber. Yet Walter Lippmann reminded us more than 70 years ago that is essential to “separate as absolutely as it is possible to do so the staff which executes from the staff which investigates.” If Washington is serious about “reinventing government,” Lippmann’s admonition is a good place to start for the intelligence community.
The intelligence directorate has become far too large and unwieldy and, because of its failures during the past decade, has become permeated with the fear of being wrong or second-guessed. Hiring smarter, more informed people would help. In recent years, the CIA’s rigorous security standards have often filtered out analysts who have traveled and lived abroad and have collegial relations with their foreign counterparts. Not surprisingly, the intelligence directorate thus lacks people with the language skills and the regional expertise needed for dealing with today’s intelligence challenges.
The operations directorate also needs to be revamped. Its modus operandi is based on placing relatively junior people abroad, working out of U.S. embassies with State Department cover. Yet the directorate will not be able to substantially increase the amount of crucial information it collects unless it is willing to take greater risks by assigning experienced people abroad without diplomatic cover. Only then would intelligence personnel have the wherewithal to encounter the unsavory people who threaten our interests. In addition, the operations directorate must rely more heavily on foreign liaison services that have access to sensitive intelligence on terrorism and criminal activities abroad. Doing so would allow the CIA to concentrate clandestine collection efforts on countries where no access currently exists, such as Somalia, Sudan, and Yemen.
Just as the U.S. military could be used to perform clandestine actions in wartime, State Department foreign service officers could collect intelligence more effectively than their clandestine counterparts. However, recent budget cuts have seriously eroded the department’s capabilities. At the same time, the demands of an unstable and fractious world have created additional demands on the department, which must supply an ambassador and staff to 192 independent countries. Because of budget cutbacks, the department has been forced to close important posts in Zagreb, Medellin, Lahore, Alexandria, and Johannesburg, to name just a few, and has had to post political amateurs with deep pockets to key embassies in Europe and Asia. The staffs of most of these embassies could collect intelligence openly and less expensively than could their CIA counterparts, freeing the agency to concentrate on the collection of intelligence on terrorist networks, technology, and weapons of mass destruction in closed areas. One of the CIA’s first and most prestigious directors, Allen Dulles, emphasized that “the bulk of intelligence can be obtained through overt channels” and that if the agency grew to be a “great big octopus” it would not function well. The CIA has about 16,000 employees–more than four times as many as the State Department.
Increase intelligence sharing. The CIA must strengthen links across the intelligence community in order to share intelligence. Today, information tends to move vertically within each of the 13 intelligence agencies instead of horizontally across them. The CIA’s emphasis on the compartmentalization of intelligence and the need to know also serve as obstacles to intelligence sharing. In addition, the CIA must become more generous in sharing information with organizations that will be on the front lines in the war against terrorism, including the INS, the Federal Aviation Agency, the Border Guards, and the Coast Guard.
The intelligence community, particularly the CIA, faces a situation comparable only to that of 55 years ago, when President Truman created the CIA and the National Security Council. As in 1947 and 1948, the international environment has been fundamentally recast and the threats have been fundamentally altered. The institutions created to fight the Cold War must be redesigned. This is exactly the task that the new FBI director, Robert Mueller, has established for himself and his agency, and a failure to do so at the CIA could mean a repeat of the intelligence failures of September 11, 2001, and an additional erosion of CIA credibility. A reconstituted directorate of operations and directorate of intelligence could be the linchpin of a reform process that will restore a central and valued role to intelligence in the making of national security policy.
Melvin A. Goodman ([email protected]) is professor of international security at the National War College, senior fellow at the Center for International Policy, and author of The Phantom Defense: America’s Pursuit of the Star Wars Illusion (Praeger, 2001) and The Wars of Eduard Shevardnadze (Brassey’s, 2001). From 1966 to 1990, he was senior Soviet analyst at the CIA and the Department of State’s Bureau of Intelligence and Research