Spy watch: Behind closed doors at the national security agency. (Culture & Reviews).

Body of Secrets: Anatomy of the Ultra-Secret National Security Agency From the Cold War Through the Dawn of a New Century, by James Bamford, New York: Doubleday, 721 pages, $29.95

OSAMA BIN LADEN is a dutiful stepson. This mundane bit of information took on particular importance following the September II attacks against the World Trade Center and the Pentagon. As James Bamford reminds us in Body of Secrets, his latest book on the shadowy National Security Agency (NSA), officials at the organization would routinely play intercepted telephone conversations between bin Laden and his stepmother to congressmen in order to acquire more funds for eavesdropping activities.

As Bamford could not have known when he wrote his book, the passage highlights what is both right and wrong in America's reliance on communications intelligence. One such call, taped in early September, allegedly constitutes part of the evidence that bin Laden was involved in the mass homicides in Washington and NewYork. At the same time, the United States, despite its ability to listen in on its arch foe, was unable to prevent the attacks from taking place, underscoring the chasm between the NSA's technological prowess and the intelligence community's capacity to absorb, analyze, and act on information gleaned.

The NSA was formally established on October 24, 1952, to replace the shaky and ineffective Armed Forces Security Agency (AFSA). The United States had begun collecting signals intelligence (or Sigint) before World War II, but had avoided creating a single authority to handle it all. Instead, the armed forces services collected Sigint separately, and even when the AFSA was set up to combine these efforts, the services maintained control over their specific code breaking and intercept activities. This fragmentation proved catastrophic during the Korean War--the North Korean invasion took the U.S. completely by surprise--and led to the NSA'S urgent creation. Though the agency reports to the secretary of defense, it became early on a semi-sovereign entity.

The NSA is the largest of the U.S. intelligence agencies, with a staff of some 38,000 people, an additional 25,000 non-staff personnel in listening posts, and an annual budget estimated at $7.3 billion. It is headquartered in an enormous complex known by some agency employees as Crypto City, located off the Baltimore-Washington Parkway near Annapolis Junction in Maryland. The atmosphere there is reminiscent of the Polish science fiction novelist Stanislaw Lem's outlandish The Bullding in his Memoirs Found in a Bathtub (1971), where spies search for secret meanings in Shakespeare and one character exclaims, "A cracked code remains a code. An expert can peel away layer after layer. It's inexhaustible. One digs ever deeper into more and more inaccessible strata. That journey has no end."

To Bamford's credit, he single-handedly has done a considerable share of excavation into the NSA'S inaccessible strata. He first did so in his much-acclaimed The Puzzle Palace (1982). The numerous government documents he managed to obtain for Body of Secrets confirm the earlier book's underlying premise: The NSA is both a remarkable and disquieting embodiment of the awesome power of the American government. While Bamford never draws explicit political conclusions from this observation, he is acutely sensitive to the illegal behavior an institution like the NSA can help generate. Granted partial access to NSA officials, he is also rarely taken in by his subject or sources, constantly playing off inconsistencies in quotes by some agency members against those from others. This is laudable when one has been provided unique information, as Bamford was, on an ultra-secretive organization--one increasingly conscious, however, of the advantages of partial transparency after decades of stony silence.

One of Bamford's most damning accusations is that the NSA failed to do what it was mainly designed to do: break high-level Soviet ciphers. In the immediate aftermath of World War II, the U.S. Sigint effort had some sterling successes. A team of experts, known collectively by the codename TICOM, was able to get hold of the USSR'S "Fish" cipher machine, one of which had been captured by the Germans, and therefore read Soviet communications. The system worked until 1948 when, overnight, the USSR'S encrypted lines went dead. (An AFSA linguist, William Weisband, was suspected of having warned Moscow, but he was never convicted.)

The array of the NSA'S duties is vast and complex. Though high-level Russian codes remained unbroken, the NSA had greater success penetrating and unscrambling Soviet communications traffic (Comint, in the professional jargon). It also gathered much vital electronic intelligence, or Elint, meaning those signals put out by radar, missiles, and other devices. When the Cold War ended, the NSA shifted its focus away from the former Soviet Union. Though the NSA eavesdropped on most countries from the moment it began operating, the agency's principal mission had changed by the mid-1990s and it spent most of its time listening in on friendly states and allies.

Some allies would prove to be more equal than others. One of the peculiar byproducts of the NSA's activities was the formation of an Anglo-Saxon fraternity of snoops, UKUSA, named for a communications intelligence agreement originally signed between the NSA and its British counterpart. The grouping, which now includes the United States, the United Kingdom, Australia, and New Zealand, is sometimes inaccurately known as Echelon, for the software program integrating the Sigint capabilities of the member states.

As Bamford writes, the idea behind Echelon was that "agencies would be able to submit targets to one another's listening posts and, likewise, everyone would be allowed to share in the take--to dip their electronic ladles into the vast cauldron of intercepts and select what they liked."

Bamford doesn't take kindly to this invasion of the privacy of others, whether the others are foreign states or individuals. He discerns threatening patterns that can, in extreme cases, have a nefarious impact on domestic American life. The NSA is legally barred from spying within the continental United States, or even, in most cases, on American citizens. Nevertheless, it has on numerous occasions engaged in domestic surveillance, leading in one noted case in the late 1990s to the arrest of Nasser Ahmad, an Egyptian immigrant, and his detention in solitary confinement for three years. only when Ahmad was finally allowed to see a portion of the secret evidence against him was he able to gain his release.

Such misuse of power has always lurked in the NSA'S past, even as elected officials have tried to expand its legal range of activities. Richard Nixon, for instance, tried to empower the NSA to spy inside the U.S. (The effort was derailed by, of people, J. Edgar Hoover, who didn't want anyone competing with the FBI.) Yet one of the most infamous examples of political manipulation by a branch of the U.S. government did not directly involve the NSA. Bamford wisely includes a discussion of the benignly named Operation Northwoods. He suggests that the political system that could spawn the NSA was also one that could take the mania with communism to the repulsive extremes revealed by that scheme.

Northwoods was a secret and illegal plan drawn up by the joint Chiefs of Staff, chaired by Gen. Lyman Lemnitzer, during the Kennedy administration. Bamford succeeds in showing that all those '6os and '70s films about generals with a screw loose and a taste for Armageddon weren't entirely fictional. The idea was to provoke violent incidents inside the United States, including murders, bombings, and hijackings, that could then be pinned on Cuba, thus justifying military action to overthrow Fidel Castro. Northwoods was ultimately rejected by Kennedy's defense secretary, Robert McNamara, but the fact that the plan could have reached the upper echelons of the administration reveals that Kennedy's top brass felt the president could stomach considerable misconduct.

Like all bureaucracies, the NSA is in perpetual search of more funding to generate ever-larger amounts of information with less and less practical use. To be fair, it is not the NSA's brief to analyze what it accumulates--that is the role of the CIA and various government departments, and their respective intelligence arms. In the past, notably during the Vietnam War, the agency's fine intelligence was simply ignored by those to whom it was directed, most prominently Gen. William Westmoreland. (The NSA warned, for example, of the 1968 Tet Offensive.) The problem is that the volume of information gathered by the NSA today far outreaches the intelligence community's capability to process it. As former CIA director Robert Gates put it: "Sometimes I think we just collect intelligence for the thrill of collecting it....We have the capacity to collect mountains of data that we can never analyze. We just stack it up."

Despite the NSA's colossal budget and its tendency toward information overkill, the agency's deputy director for services, Terry Thompson, could complain in 1999: "One of the reasons we don't get more support on the Hill for the budget is that we don't have a strong lobby in the defense industry....We spend our money on four hundred or four thousand different contracts and it's hard to get a critical mass of people who want to go down and wave the flag for NSA when budget deliberations are going on." Thompson was speaking in the wake of successive budget cuts at the NSA, so perhaps he had a point. But then one gets a distinct sense that he would consider any amount of money for the NSA to be somehow too little.

Whatever its faults, the NSA has suffered a fair share of casualties in its relatively short history. The most costly episode, which Bamford describes in detail, was the Israeli attack against the NSA spy ship, USS Liberty, off the Sinai Peninsula during the 1967 Arab-Israeli war. The Liberty incident led to the death of 35 crew members, but was long ago swept under the rug of U.S.-Israeli relations. Bamford tears down the official explanation, that it was all an accident. He writes that NSA officials with access to secret intercepts from the episode "were virtually unanimous in their belief that the attack was deliberate." Liberty crew members noted that during the attack, the Israelis first went after the ship's communications apparatus, which required knowledge of its makeup. According to intercepted radio traffic, the Israelis positively identified the Liberty and the markings painted on the ship's side. In contrast, the attacking airplanes were unmarked, undermining the Israeli claim that their pilots co nfused the Liberty with an Egyptian vessel, one that they, incidentally, knew to be much slower than the moving U.S. ship. Bamford's hypothesis is that the Liberty recorded radio communications between Israeli units discussing the extensive execution of Egyptian prisoners-of-war. He believes the Israelis sought to destroy the ship to cover up their war crimes.

Some have questioned Bamford's allegation. For example, New York Times reviewer James Finder disingenuously wrote that the Egyptian POW theory was based on slender evidence, and mentioned a single Israeli journalist as Bamford's source. In fact Bamford cites three sources, all Israelis. One of them was a participant in the attack and another was an eyewitness. The third, an Israeli military historian, concluded (on the basis of interviews with dozens of soldiers who themselves had killed prisoners) that as many as 1,000 Egyptians were shot. Bamford's argument is surely plausible, as anyone who has surveyed a half-century of Israel's wartime behavior will admit. The only part of Bamford's theory that is dubious--and here Finder's protest is in order--is that it would have been foolish for Israel to cover up one massacre by another. Yet where Finder sees this as evidence of Israeli blamelessness, readers will conclude that an explanation for the undeniably deliberate assault must lie elsewhere.

The Last two chapters of Body of Secrets are devoted to detailed descriptions of Crypto City, of life at the NSA, and of past progress in the organization's successive supercomputer programs. Those parts don't make for particularly compelling reading, but they represent a major accomplishment, since Bamford is the first reporter to ferret out such details, which have been secret for decades. Published in a year when the U.S. Congress sought--and then postponed--passage of an official secrets act that would have criminalized the unauthorized disclosure of any type of classified information by federal employees, Body of Secrets is a valuable reminder of the enduring siren song of concealment, the enemy of all true democracies.

Michael Young ([e-mail]) is a writer living in Lebanon.