Our culture of (over) secrecy

The more we tinker with the way we protect our national secrets, the worse the problem gets. Some congressional committees, and parts of the national security establishment, spent a good part of 2000 overtinkering. With a slightly different cast of characters, they will return to the task in 2001.

Their motivation is partly well-intended to prevent foreign espionage. It partly stems from a visceral dislike and distrust of China. It arose, specifically, from the case of Dr. Wen Ho Lee, the physicist at the Los Alamos Nuclear Laboratory who suffered such shabby treatment by the FBI that the judge presiding over his case publicly apologized to him.

There is abundant evidence that the security system is broken, but efforts to fix it break it more. If we impose stricter standards for handling classified materials, we make access to them so difficult that it becomes almost impossible to work with them. So corners are cut. Busy people put classified data into laptop computers on airplanes. This is what got Martin S. Indyk, the US ambassador to Israel, in so much trouble that his security clearance was suspended. Many officials have done the same thing without notable loss of secrets.

The offices and agencies where these procedures are enforced develop an insular culture. An outside study group concluded that this was the biggest problem with the National Security Agency, which intercepts communications.

Data concerning nuclear energy has been a particular concern ever since the Manhattan Project in World War II.

At that time, there was good reason for tight security. We were racing to build a weapon that would end the war. But the secrets of technology and science are there for anybody to discover, and we can be sure that in time, somebody will. What we were so zealous in protecting in the 1940s was freely available by the 1950s. An Iraqi defector recently told of finding volumes of Manhattan Project data on government shelves in Baghdad, a gift of the US Atomic Energy Commission.

The scientific temperament is incompatible with the constraints of security. The more security restrictions are put on nuclear laboratories, the more difficult it becomes to recruit and retain the scientists who hold our nuclear future in their heads. Earlier this year, Energy Secretary Bill Richardson ordered lie-detector tests for 800 people working on nuclear weapons. Then Congress passed a law upping the number to 15,000.

This will be counterproductive. One thinks admiringly of Secretary of State George Schultz, who announced he would refuse to take a lie-detector test and thereby killed a harebrained scheme of President Reagan to polygraph the whole Department of State looking for the source of a leak.

One of the ways scientific knowledge advances is through meetings of scientists. These ought not to be curtailed in the name of security. It is particularly absurd to say, as a congressional report did last year, that "essentially all Chinese visitors to the United States are potential spies." The Chinese are unlikely to learn as much from their spies as from sending their bright graduate students to MIT or Caltech.

Excessive security requirements blur the difference between what is really secret and what should not have been classified in the first place.

The people who want more restrictions misperceive the threat. Very few, if any, national secrets have ever been lost through lax physical security. Foreign agents are unlikely to break into a government office and ransack files.

Most secrets are lost not through physical break-ins but because of the treachery and greed of disloyal employees. Yet, as estimated by the Information Security Oversight Office, we spend $3.8 billion a year to keep secrets locked up under guard.

Leaks to the media are another engine driving ill-considered ideas about tightening security. Last month, President Clinton vetoed a bill that would have made it a crime to disclose classified material, regardless of its effect, to anyone not authorized to receive it. This would include, strange to say, members of Congress having a quiet talk with an official of the State Department.

This is not protecting national security; it is reinforcing executive control of the flow of information to Congress. The president's veto ended that effort for the time being, but it will be resurrected.

The real problem of handling national-security information is to protect better what needs protecting and to loosen or remove restrictions on the rest.

Pat M. Holt is co-author of 'National Insecurity: U.S. Intelligence After the Cold War' (Temple University Press).

(c) Copyright 2000. The Christian Science Publishing Society