Transforming U.S. Intelligence

Jennifer E. Sims, Burton Gerber, Jennifer E. Sims, Burton Gerber

• Editorial: Georgetown University Press
• Año: 2005
• ISBN:
  9781589014770
• Categoría: Politica e relazioni internazionali
• Categoría: Intelligence e spionaggio

The intelligence failures exposed by the events of 9/11 and the missing weapons of mass destruction in Iraq have made one thing perfectly clear: change is needed in how the U.S. intelligence community operates. Transforming U.S. Intelligence argues that transforming intelligence requires as much a look to the future as to the past and a focus more on the art and practice of intelligence rather than on its bureaucratic arrangements. In fact, while the recent restructuring, including the creation of the Department of Homeland Security, may solve some problems, it has also created new ones. The authors of this volume agree that transforming policies and practices will be the most effective way to tackle future challenges facing the nation's security.

This volume's contributors, who have served in intelligence agencies, the Departments of State or Defense, and the staffs of congressional oversight committees, bring their experience as insiders to bear in thoughtful and thought-provoking essays that address what such an overhaul of the system will require. In the first section, contributors discuss twenty-first-century security challenges and how the intelligence community can successfully defend U.S. national interests. The second section focuses on new technologies and modified policies that can increase the effectiveness of intelligence gathering and analysis. Finally, contributors consider management procedures that ensure the implementation of enhanced capabilities in practice.

Transforming U.S. Intelligence supports the mandate of the new director of national intelligence by offering both careful analysis of existing strengths and weaknesses in U.S. intelligence and specific recommendations on how to fix its problems without harming its strengths. These recommendations, based on intimate knowledge of the way U.S. intelligence actually works, include suggestions for the creative mixing of technologies with new missions to bring about the transformation of U.S. intelligence without incurring unnecessary harm or expense. The goal is the creation of an intelligence community that can rapidly respond to developments in international politics, such as the emergence of nimble terrorist networks while reconciling national security requirements with the rights and liberties of American citizens.

PART ONE
New Requirements

ONE
The Twenty-first Century Challenge for U.S. Intelligence

ERNEST R. MAY

FROM THE late 1940s to the late 1980s, the chief test of the quality of U.S. intelligence agencies was their capacity to gauge the capabilities and intentions of the Soviet Union and the People’s Republic of China. Retrospective grading of the agencies’ performance ranged from their own “OK” to Senator Daniel Patrick Moynihan’s flat “F,” based largely on their failure to forecast the collapse of the Soviet Union.

Moynihan’s grade was too harsh. His criterion was too narrow. Year after year, the U.S. intelligence community had been right on the mark in counting, locating, and describing Soviet military forces capable of a nuclear strike against the U.S. homeland. Though they had not predicted that the Soviet empire would splinter at the end of the 1980s or that the Soviet Union itself would dissolve soon thereafter, they had done better than most nongovernmental Sovietologists—and better than Mikhail Gorbachev and his colleagues on the Soviet Politburo—in identifying weaknesses in the Soviet system and pointing to the possibility of eventual implosion.¹

“OK” was, however, too generous if applied to overall performance. U.S. intelligence agencies had not done very well at detecting or appraising Soviet military forces other than those positioned for a nuclear strike against the United States. They had failed time and time again to forecast accurately what the Soviet government intended to do. Sometimes this was because a minority in the Politburo overruled the better judgment of its majority, as was the case with the decision to invade Afghanistan in 1979. At other times it was because the agencies’ leaders failed to notice or trust information already in their hands, as, for example, in 1980–81, when they persisted in predicting that the Soviets themselves would crack down on Poland instead of leaving it to the Poles to deal with Solidarity. In the case of China, the community had patted itself on the back for predicting accurately that country’s first test of a nuclear device, only to learn from a postmortem that it had been wrong about every aspect of the test but its timing.² The net performance of the U.S. intelligence community in dealing with major cold war threats probably merits a grade somewhere between B and C.

The Post–Cold War Policy Challenge

Between the fall of the Berlin Wall in 1989 and the terrorist attacks of September 11, 2001, U.S. intelligence agencies faced a very different set of tests. Not only were they challenged to pass judgment on subjects other than the strategic nuclear threat from a rival superpower; they found themselves having to answer to a larger and more varied group of consumers within the “policy community” and to do so during a period of dizzying technological change.

In 1990–91 the chief test for the community had to do with the crisis in the Persian Gulf occasioned by Iraq’s invasion of Kuwait. In the past, the community would have been expected to speak mostly about the intentions of Moscow. No longer. Now collectors and analysts had to size up Saddam Hussein as someone acting outside the earlier bipolar framework.

While questions about China continued to be very much those that the community had prepared itself to answer during the cold war, other questions about Asia were new. In particular, intelligence officers were asked about North Korea, not as a pawn between Moscow and Beijing but as an independent power potentially armed with nuclear weapons. They also had to report on and analyze internal conflicts in Pakistan and India and possibilities for a clash between these nations—a clash that after 1998 would be one between two nations that had both tested nuclear weapons.

Similarly, U.S. intelligence agencies found themselves presented with questions about Africa that no longer involved Soviet or Communist maneuvers among African factions but concerned conditions in Africa itself and the dispositions and intentions of indigenous African leaders. In 1993 they were challenged to help U.S. Special Forces trying to get aid to suffering Somalis despite opposition from the country’s tribal chieftains. Soon afterward, they had to assess for U.S. policymakers massacres approaching genocide in Rwanda and baffling back-and-forth competition among would-be successors to Mobuto Sese Seko in Zaire.

Even regarding Europe, the questions coming to U.S. intelligence agencies were different from those of the past. From the early 1990s until the end of the decade, these agencies had to focus on following and interpreting events within a disintegrating Yugoslavia. First with regard to Bosnia, then to Kosovo, they had not only to assist U.S. negotiators but also to help U.S. and North Atlantic Treaty Organization (NATO) commanders choose the right targets for air operations and the right mixture of forces for peacekeeping. Meanwhile, they had to field questions about European countries formerly dealt with primarily as Soviet satellites and about former Soviet republics, which they had previously all but ignored.

Western Europe itself became a different target for collection and analysis, for policymakers’ questions no longer concerned cooperation against the Soviet Union. Instead these questions had to do with prospects for the European Union’s becoming an economic and political competitor and with rumors of European industrial espionage.

On top of many changes in the character of U.S. regional concerns was the emergence of the drastically altered problem that would later become the dominant one—terrorism. The community had concerned itself marginally with terrorism during the 1970s and 1980s. A good deal of both collection and analysis had, however, focused on the Soviet role. Libya’s role in the Berlin disco bombing of 1986 came to light largely because Libya had worked hand in hand with the German Democratic Republic, and the United States was collecting signals intelligence in East Germany as part of the coverage of the Soviet target.

The first World Trade Center bombing, in 1993, signaled the rise of a new kind of terrorism, not sponsored by any state, not directed toward any particular political end, but intended just to kill as many Americans as possible. By the late 1990s the community had not only recognized this new problem but had also identified Osama bin Laden as the financier and organizer of an international Islamist terrorist network. After the bombing of U.S. embassies in Kenya and Tanzania in August 1998, Director of Central Intelligence (DCI) George Tenet proclaimed this terrorism the highest priority for the community. The events of September 11, 2001, made it the highest priority issue for the nation.

The decade and a half following the end of the cold war thus put before the U.S. intelligence community a whole range of issues different from those of the cold war era. They called for collection on and analysis of governments, most of which had not been important cold war targets. They called also for collection on and analysis of nongovernmental organizations, particularly in the Muslim world. The community had to reorient some of its collection platforms. It had to develop others anew. Most of all, it had to take a workforce skilled at studying the Soviet Union and strategic weaponry and, by a combination of retraining, new recruitment, and managerial stimulation, give it the skills needed for dealing with altogether different sets of problems.

New Customers—and an Emphasis on Military Operations

Difficult in the best of circumstances, this massive reorientation was made particularly hard by a concurrent change in the population of intelligence consumers. During most of the cold war era, the assiduous readers of intelligence within the U.S. government had been mostly diplomatic negotiators or, in the Pentagon, civilians and military officers preoccupied either with budget planning or with high-level war planning.

This changed with the Gulf War of 1991. From the beginning of time, operational military commanders have wanted to know what was “on the other side of the hill.” Improvements in capacity to collect imagery from satellites, combined with improvements in communication between collectors of imagery and collectors of signals intelligence, made it technically possible by 1991 for the U.S. intelligence community to put together actual pictures of the enemy facing U.S. forces opposite Kuwait. The problem was that they could assemble these pictures in Washington but not get them quickly to U.S. commanders in the field.

General Norman Schwarzkopf, in overall charge of operations for the Gulf War, was not known for his even temper. “Stormin’ Norman” unleashed on one and all his fury that these pictures were not arriving on the battlefield in real time. Improvisation improved matters somewhat. After the Gulf War the management of the military establishment insisted that managers in the intelligence community give very high priority to technical and organizational improvements, providing future battlefield commanders with the real-time tactical intelligence that Schwarzkopf had demanded. On Capitol Hill, the armed services committees supported this demand even though, at the time, they were clipping the overall defense and intelligence budgets in order to provide a promised post–cold war “peace dividend.”

Initially the intelligence community leadership put up resistance. Robert M. Gates, who became DCI under the first President Bush, had been a career analyst in the Central Intelligence Agency (CIA). As deputy national security advisor earlier in the Bush administration, he was intensely aware of how hard it had been for the community to answer new questions such as those about the former Soviet republics, India and Pakistan, and Iraq. He was intent on building new analytic capabilities, both substantial and methodological. The leaders of the intelligence oversight committees in the House and Senate had some sympathy for Gates’s objective. But the DCI and the intelligence oversight committees are no match for the military establishment and the armed services committees.

The military establishment itself was in the midst of major internal change. The Goldwater-Nichols Act of 1986 had reduced the autonomy of the separate services, created strong incentives for interservice cooperation, and, among other things, greatly strengthened the chairman of the joint chiefs and the major field commanders in comparison with service chiefs and defense agency chiefs in Washington.³ The combatant commanders, facing actual or potential operations in Africa, the Balkans, and the Korean peninsula as well as in the Persian Gulf region, put intense pressure on the chairman and the secretary of defense to skew spending for intelligence toward meeting their tactical needs.

After the 1992 presidential election, Gates was replaced by R. James Woolsey, then by John Deutch. Woolsey, who had chaired a study that recommended a new generation of reconnaissance satellites, made action on this recommendation his first priority. This irked senior managers at the CIA who thought personnel should take precedence. That Woolsey pursued his goal by allying himself with the combatant commanders and the armed services committees infuriated his congressional overseers. And his relations with President Clinton were so distant that when a protester landed a little plane on the south lawn of the White House, the joke in Washington was that it was Woolsey trying to get into the Oval Office.⁴ Deutch, who had been deputy secretary of defense, made it no secret that his own primary interest remained the military establishment.

The next DCI, George Tenet, who had been Deutch’s deputy, had seen the travails of the community first as a staffer, then as staff director for the Senate oversight committee, and, from 1993 to 1995, as senior director for intelligence programs on the National Security Council staff. He became DCI at just the time when the threat from Osama bin Laden and al-Qaeda began to be recognized as one that might match in urgency the threats faced by combatant commanders in the Balkans and elsewhere. Tenet gave priority to rebuilding the CIA’s clandestine service.

Having only marginal influence over Defense Department spending for intelligence, Tenet could divert no more than a trickle of funds to serve his own priority. In 2004, long after the terrorist attacks of 2001 had released a flood of new resources for intelligence, Tenet said publicly that the CIA was still five years away from having adequate capability for collecting human intelligence even on its number-one target.⁵

The Technological Challenge

The combination of a rash of new issues and pressure to emphasize tactical military intelligence could alone have been a crippling challenge for the intelligence community. But even if these two challenges had not existed—if the end of the cold war had been postponed for a decade or more and the community had continued to concentrate on the strategic threat from the Soviet Union—the U.S. intelligence community would have found itself struggling with potentially overwhelming problems resulting from technological and technology-driven changes.

Because “information revolution” and “digital revolution” have become such trite phrases, we easily forget how real their results have been. At the beginning of the 1980s, high-powered computation still occurred mostly on mainframes owned and operated by large organizations, with the Defense Department’s National Security Agency (NSA) far ahead of all others not only in computational capacity but in research and development. The NSA attracted many of the best mathematicians and programmers in the country. By the turn of the millennium, college students were using personal computers that stored as much data as the NSA had possessed two decades earlier. The Internet had become a medium for global transfer of information on a scale not previously imagined. And right behind the Internet came the cellular telephone, soon so popular that shopping malls began to seem like huge telephone booths.

The implications for intelligence collectors and analysts were obvious and terrifying. The NSA, swamped with data even in the 1980s, had to try to track communications expanding annually by orders of magnitude, much of it traveling through media previously unfamiliar to the agency’s technicians. The agency was hard put to keep up not only with the flow of signals but with techniques of encryption, given that private concerns also spent money to protect their transmissions. General Michael Hayden, the director of the NSA, complained to Congress in 2002 that the terrorist networks the agency was trying to penetrate were the beneficiaries of the $3 trillion global communications industry.⁶

The problem for agencies acquiring imagery was not quite as great, but neither was it trivial. Private concerns and foreign governments put up satellites, acquired imagery, and sold this imagery on the open market. Even if the imagery acquired by the combined resources of the National Reconnaissance Office (NRO) and the National Geospatial-Intelligence Agency (NGA) was of higher resolution, and even if the paths of U.S. reconnaissance satellites remained secret (which was not always the case), the publicly available imagery gave other governments and terrorist organizations useful clues as to how to hide their activities or achieve misdirection. Both India and Pakistan succeeded in concealing from the United States their preparations for the nuclear tests conducted in May 1998.

And it should be said that the changes also affected collectors of human intelligence in the CIA and the Defense Department, for imagery and especially signals intelligence often provide critical leads for officers in the field. This is so much the case that the NSA’s General Hayden has hinted that human and signals intelligence ought perhaps to be combined within a single organization.⁷

The Challenge of Open Sources

The revolutionary character of technological changes obviously affected the ability of intelligence collectors to serve their various customers within the U.S. government. The volume of “take” by the NSA and the NRO became less and less manageable. The meaning of signals became less and less intelligible as information arrived in languages that few analysts commanded.

Worst of all, perhaps, the galloping information revolution strained the capacity of the intelligence community to perform its traditional function of providing policymakers with information and analysis not obtainable from open sources. Intelligence officers had long traded on their ability to give policymakers useful items of gossip not to be found in the morning newspapers. Elliott Richardson, undersecretary of state in the Nixon administration, once described to me the morning briefings he received from the department’s Bureau of Intelligence and Research. He said that an analyst would usually put before him some juicy item of information, “like a cat depositing a mouse.”

But these items tended to open doors for intelligence analysts, gaining them entrée to present their reflections on issues of the moment.

Satellite news collection, the proliferation of newscasts, the ferocity of competition among providers of spot news, and the extent to which post–cold war policy issues called for expertise not reserved to intelligence agencies—all these factors made it difficu...

Índice

• Cover Page
• Title Page
• Copyright Page
• Contents
• Preface
• Introduction
• Part One: New Requirements
• Part Two: New Capabilities
• Part Three: Management Challenges
• Contributors
• Index
• Footnote