INTELLIGENCE WARS

American Secret History

From Hitler to Al-Qaeda

By Thomas Powers

New York Review. 450 pp. $27.95

FIXING INTELLIGENCE

For a More Secure America

By William E. Odom

Yale Univ. 230 pp. $24.95

The Sept. 11 terrorist attacks were the intelligence community's gravest failure since Pearl Harbor. Reform, however, has yet to begin. A joint congressional committee inquiry made 19 carefully couched recommendations last December; an independent commission headed by former New Jersey governor Thomas Kean won't deliver its findings until 2004. War, first in Afghanistan, now in Iraq, has delayed diagnosis, pushing critical surgery further into the future.

Over the past year and a half, some of the symptoms of an intelligence community grown sclerotic have come to light. Khalid Sheik Mohammed, the alleged mastermind of the Sept. 11 attacks who was recently arrested in Pakistan, was known to the U.S. intelligence community as early as 1996 because of his part in a plot to blow up American airliners over the Pacific. But tracking him was given a low priority. In 1999, the National Security Agency obtained communications involving individuals who were identified, after Sept. 11, as participants in the attacks that occurred on that day. But that information apparently was not shared or acted on. The CIA tracked two of the hijackers to a meeting of known terrorists in Malaysia 18 months before Sept. 11. But the hijackers were not put on a watch list, and the CIA "was slow to pass on what it knew to the FBI," Thomas Powers reports in Intelligence Wars. Thus the two hijackers easily entered the United States in 2000.

In June 2001, the intelligence community got reports that Mohammed was sending terrorists into the United States, but there was no follow-up. An agent in the FBI's Phoenix office told headquarters in July 2001 that he believed Osama bin Laden was sending students to U.S. civil aviation schools; he proposed an investigation, but FBI headquarters had no interest in it. In August 2001, FBI headquarters refused Minnesota field agents' request to search the belongings of Zacarias Moussaoui. Between Sept. 8 and Sept. 10, 2001, the NSA intercepted communications that might have indicated an impending terror strike; that information was not translated or disseminated.

It is clear, after reading Powers's Intelligence Wars and William E. Odom's Fixing Intelligence -- two markedly different yet equally indispensable books -- that the problems are staggering but their solutions are far from mysterious. Odom, a former director of the NSA, has assembled a bracing set of proposals that would "eliminate entirely the FBI's responsibility for counterintelligence" and the CIA's oversight of the intelligence community and create "a national counterintelligence service." Powers's book helps us understand why Odom's recommendations make sense.

A Pulitzer Prize-winning journalist, Powers has written a highly regarded biography of former CIA director Richard Helms and an ambitious, controversial history of the failed Nazi atom-bomb program. Intelligence Wars collects 24 reviews (most of which appeared in the New York Review of Books) of more than 60 intelligence books. These discerning essays span 25 years and provide a revealing history of the victories, defeats and ambiguities of Cold War and post-Cold War intelligence gathering.

Powers portrays in vivid human terms repeated FBI failures in counterintelligence, the intelligence agencies' inability to infiltrate terrorist groups, chronic reluctance to share information and a management structure that leaves no one in charge of and accountable for the entire effort. "Paralysis when confronted by internal problems is no minor matter," he writes, "because the ready excuse of secrecy makes it difficult even for authorized outsiders -- congressional committees, oversight boards of distinguished public citizens, presidential advisers -- to address any problem which an intelligence organization chooses to conceal. It is as if a bank had lost millions to an embezzler year after year, but continued to block the auditor at the door."

As Powers's book makes clear, a bewildering array of agencies and missions makes up the intelligence community. In Fixing Intelligence, William Odom dissects this bureaucratic jellyfish with remarkable precision. He keeps the historical context pithy. Starting with the American Revolution, Odom writes, U.S. clandestine intelligence operations were generally a wartime matter, discontinued in peacetime. That changed with World War II and the establishment of the Office of Strategic Services, the CIA's precursor. After the war, William J. "Wild Bill" Donovan, the OSS's flamboyant leader, was unable to persuade President Truman to make the OSS permanent. The FBI took over domestic counterintelligence, and the Army took over foreign counterintelligence. But only two years later, in 1947, Congress created what Donovan had wanted -- the Central Intelligence Agency -- to "prevent 'another Pearl Harbor' " and to shadow the Soviets.

In 1952, the Army, Navy and Air Force intelligence capabilities combined to form the National Security Agency. In 1960, the National Reconnaissance Office was created. In 1961, the Defense Department created the Defense Intelligence Agency. (DIA provides threat assessment in order to formulate requirements for U.S. weapons developers and to improve training and organization; it also provides intelligence support to military operations.) "Since that time, the Intelligence Community has remained essentially unchanged in its general outlines," Odom writes. Revelations of CIA excesses led to inquiries from a U.S. Senate select committee, convened in 1973 and headed by Idaho Sen. Frank Church, and to limitations on surveillance at home, assassination abroad and recruiting criminals anywhere.

Odom and Powers point out that these reforms were not structural; on top of that, the structure that remained intact was configured to track Soviet military developments and penetrate a government run by well-educated, diplomatically sophisticated people. Now the intelligence community faces nomadic operatives supported by improvisational, transnational opposition groups from North Africa to the Philippines.

Evaluated one way, Powers writes, the CIA's Cold War record can be seen as a string of embarrassments. "The CIA's history . . . is rich with failures to predict major events, among them the first Soviet atomic bomb, the North Korean and Chinese invasions in Korea, the Hungarian revolt, Fidel Castro's victory and Khrushchev's subsequent placement of missiles in Cuba, the invasion of Czechoslovakia, and the invasion of Afghanistan. Above all, the CIA failed to predict -- failed even to imagine -- the collapse of Soviet communism and the end of the cold war."

But Powers concludes that "the one thing the CIA over the years consistently got right" was "the steadiness and pervasiveness of a kind of close contact which will unmistakably reveal an opponent's real capacities and approximate intentions. Intelligence services touch, watch, and listen to each other at a thousand points. The intimate knowledge revealed by the wrestler's embrace freed both sides from ignorance, rumor, and outbreaks of panicky fear that spark big wars no one wants."

How can we achieve a "wrestler's embrace" with terrorists? Reading Odom and Powers, it seems clear that it can be done. Odom proposes that signals intelligence -- "intelligence derived from intercepted electronic communications" -- should be centralized in the NSA. The National Reconnaissance Office, for example, should no longer exercise budgetary authority over the NSA. That way, the NSA could focus on technological research and development. The Sept. 11 intelligence failure demonstrated that the NSA lacks sophisticated technology to create a refined state-of-the-art database that would give other parts of the intelligence community access to intercepted information.

Odom proposes a similar consolidation of all the intelligence functions -- imagery intelligence, human intelligence and counterintelligence -- in which each would have a national manager who would report to a national director of the entire intelligence effort. This would address the current diffusion of functions among different agencies. It would also halt an outdated practice of having the CIA director also be the national director of central intelligence (DCI), which gives the CIA control over the entire community. (The congressional joint inquiry also recommends doing away with this configuration.)

In practical terms, this new structure would take the FBI out of counterterrorism and counterintelligence, a step Odom sees as critical. "The FBI predominately uses three methods against criminals: telephone taps, informers in criminal circles, and heavy-handed interrogations. Although these tactics often work against criminals, they do not work against spies and terrorist organizations." A new national counterintelligence service -- an NCIS -- should be created that would do away with the fragmenting of counterintelligence among agencies that rarely share information with one another.

The new NCIS service should not have the authority to make arrests. Odom concludes that such a service, free from law enforcement responsibility, "would have had much better prospects for putting together the pieces of information that were accumulating from many sources in the United States and Europe" prior to Sept. 11.

Powers notes that "bringing legal cases against terrorists takes time and dries up intelligence sources." The existing congressional inquiry makes the same point, criticizing the 1980s strategy that had "law enforcement often emerging as a leading tool" with which to fight terrorism.

Powers also brings to the fore important shortcomings that are not structural. For instance, he reports that "a dissident claims" that the NSA "has only one Pashto speaker" (Pashto being "the language of the principal ethnic group in Afghanistan, including most of the leadership of the Taliban"). Reliance on the intelligence services of Egypt, Jordan, Saudi Arabia and others to overcome such deficiencies has not worked. (The Congressional inquiry also criticized "excessive reliance on foreign liaison services.")

There has been a tidal wave of books attempting to understand the Sept. 11 intelligence failure, many of them written from various self-interested positions, with a limited feel for the intricacies of the intelligence world they criticize. Powers brilliantly conveys the ethos and culture of intelligence agencies -- a complexity he has been studying and writing about for almost 30 years. Odom produces a detailed reform proposal that is also born out of decades of puzzling out intelligence conundrums. Without self-congratulation or animosity, both men have made a formidable contribution to the difficult work ahead in re-aligning the intelligence agencies' Cold War-vintage structure. *

Lorraine Adams reviews frequently for Book World.

An RAF officer examines photographs of a Soviet spy trawler (1968).