The Science of Secrets

or most government executives, it would be a nightmare. For Energy Secretary Bill Richardson, it's daily life. In the wake of revelations about Chinese espionage at nuclear weapons laboratories, the Energy Department has come under unrelenting attack-not only in legislation sponsored by Republicans led by Sen. Jon Kyl, R-Ariz., but also from President Clinton's own Foreign Intelligence Advisory Board, chaired by former Sen. Warren Rudman. Even one of Richardson's own subordinates, Victor H. Reis, the highly respected assistant secretary of Energy for defense programs, turned against his boss.

The Kyl and Rudman plans-which Reis quietly endorsed-would strip the Secretary's office of day-to-day control over national security functions. Richardson found this idea so threatening that he forced Reis to resign and promised to recommend a Presidential veto. But within two weeks of Reis' ouster, Richardson agreed, with reservations, to Kyl's and Rudman's idea of a semi-autonomous Agency for Nuclear Stewardship within Energy.

Yet while Richardson and Energy's Washington headquarters were fighting for their institutional lives, their subordinates at the laboratories were gaining ground. The Energy Department is bitterly divided. The deepest of its many fault lines runs between federal employees, who work at Washington headquarters and a nationwide network of field offices, and contractor employees, who do the actual science at the labs. The labs resent federal control; the feds resent the labs' independence.

Both sides share blame for the security failures, which experts believe let China steal key design data for nuclear weapons. The labs had security lapses, while headquarters ignored more than a decade of dire warnings about those lapses. In Rudman's report, the federal employees fared far worse than the contractors. And both Rudman's recommendations and Kyl's legislation call for reforms that grant the labs greater independence.

"You must get the weapons laboratories out from under the general bureaucracy . . . and the only way you can do that is to set up a semi-autonomous [Agency for Nuclear Stewardship]," says Rudman.

Both sides agree the department has been mismanaged for years, but they offered opposing visions of how to fix it. Rudman and Kyl both proposed consolidating the department's scattered national security functions, including its weapons labs, in a new agency. Its director would have substantial autonomy; independent security, counterintelligence and oversight ability; and, on most issues, a blanket exemption from headquarters control. By contrast, Richardson's reforms relied on a series of new headquarters offices-a counterintelligence chief, a "security czar" and so on-with ever-stricter centralized control not only over national security functions, but also over all of Energy's disparate activities. As of mid-July, Richardson's acceptance of a semi-autonomous agency was still conditional on those offices retaining oversight authority (which Rudman says was his intention all along). At stake are the power of the Secretary, the careers of senior officials and the entire structure of the department charged with securing America's nuclear arsenal.

Counterintelligence

At the core of the Clinton administration's vision for a reformed Energy Department is a new Office of Counterintelligence created in 1998 to radically upgrade national security. Its chief, veteran FBI spy-catcher Edward J. Curran, not only reports directly to Richardson, but also has direct access to FBI Director Louis J. Freeh and Central Intelligence Director George J. Tenet. "I was sent here because the intelligence community was very concerned," Curran says. A former intelligence agency official puts it more bluntly: "Security has always been atrocious. . . . The labs had a terrible reputation."

In February 1998, before Richardson's arrival, President Clinton issued Presidential Decision Directive 61, which ordered Energy to address a 12-year pileup of reports warning of security lapses at the labs. Since 1988, the General Accounting Office had repeatedly criticized the department's foreign-visitors program.The program allowed foreign scientists to visit the labs-ostensibly to engage in strictly peacetime research carefully quarantined from weapons work. A 1997 GAO report warned that two of the three weapons labs, Los Alamos and Sandia, both in New Mexico, had obtained waivers exempting their foreign visitors from even simple cross-references with intelligence agency databases for suspicious information. Only 5 percent of visitors from "sensitive" countries were checked at all. As a result, says Curran, in many cases "the lab director didn't have a clue who was at his site." The waivers recently were revoked.

In addition, lab scientists were often poorly briefed on the potential for security breaches when they went abroad-as when suspected spy Wen Ho Lee went to China in the 1980s, for example. "We would send scientists from the labs over to have private meetings with the Chinese nuclear weapons scientists," says the former intelligence official, [but] there wasn't any significant intelligence briefing that went on that said, 'look out.' "

A former weapons scientist recalled hearing colleagues laugh out loud at a required pre-trip training film, which featured a ludicrously sultry Soviet seductress-and was still being shown as recently as 1995. Nor was travel follow-up any better: Despite a requirement to list potentially compromising contacts, says the scientist, "it would be amazing if 10 percent of the people who did foreign travel did their trip reports." Curran adds, "We had people who'd been over to Moscow many, many times [who told me], 'I've never been debriefed by anybody.' "

Headquarters officials responded to years of warnings about the labs with promises, but little action. In 1997, Energy's then-deputy secretary, Charles Curtis, made a serious attempt to tighten up security. But Curtis' six-point directive was never fully implemented, and it apparently was discarded when he departed later that year. Elizabeth Moler, acting Energy Secretary in 1998, has testified under oath that she was never briefed on or even told of Curtis' plan.

Even G-men could drown in Energy's organizational mire. "I'm the fourth FBI agent sent over here," says Curran. "But the structure in which they sent them over . . . was null and void. They were not able to do their job."

Before Clinton's 1998 directive, the Energy Department's counterintelligence unit was just part of its overall Intelligence Office, which in turn belonged to the Office of Nonproliferation and National Security - putting counterintelligence three levels down from the Secretary.

Likewise, at most labs, counterintelligence staffs reported to the site security office, which in turn reported to the facilities manager-putting counterintelligence three levels down from the lab director and in the same category as night watchmen and janitors.

"The [counterintelligence] program was buried in the security structure," says Curran. "Security people look at many issues as 'guards, gates, and guns.' Counterintelligence is not that at all. . . . They wanted to do a good job, but they didn't have the training, the background."

What Curran wants now is professional investigators reporting directly to the highest levels of authority. To that end, and to the discomfort of some scientists, he has hired retired FBI agents to take over new counterintelligence offices at five of the most sensitive labs.

"I don't want to demean the staff that was here," says Ken Schiffer, the former FBI agent now heading counterintelligence at Los Alamos, Wen Ho Lee's old lab. "I certainly inherited a good professional staff. [But] what I am trying to do is attract individuals like myself, who have . . . a little more experience in interviews, a little more exposure to cases."

Schiffer, like his fellow lab counterintelligence chiefs, now reports to the lab director-with a second line straight to Curran. This arrangement was jokingly likened by one Energy source to the Soviet Union's system of commissars. "No," Curran laughs. "[It's simply that] I have to be informed and the lab director has to be informed." But certainly the effect of such a structure is to diminish the labs' cherished independence.

Centralizing Control

The burial of counterintelligence beneath layers of bureaucracy was just one aspect of Energy's disorganization, and Curran's commissars are but one aspect of the effort by the Secretary's office to centralize control. Historically, many sections of the department did not report to the bureaus that disbursed their funding; the lines of accountability were blurred or nonexistent.

Consider the plight of Reis. As chief of Defense Programs (DP), he paid the contractors at the three weapons labs and the field office employees who oversee them, but he was clearly frustrated by how little real control he had. "The labs didn't report to DP. The field didn't report to DP. Nothing reported to DP, except the immediate staff . . . even though I was the assistant secretary and I was responsible," Reis recalled in an interview before his ouster.

On April 21st, that changed-though apparently not enough for Reis. Richardson announced that three lead bureaus at Washington headquarters-Defense Programs, Environmental Management and Science-would control the activities they funded. This simple decision reversed years of confusion in Energy's organizational chart.

Tangled Web

That confusion grew from the hybrid and evolutionary nature of the Energy Department. Starting with the creation of the first federal lab, Los Alamos, in 1943, the government deliberately decided not to get into the science business directly, but rather to hire private contractors with extensive experience in administering research programs. The government gave contractors one clear goal-build better bombs-then gave them lots of money and got out of their way. The results, which included the H-bomb, were impressive, but the foundation for accountability and oversight was weak.

Over the years, the labs reported to a series of progressively larger and bureaucratically more distant bodies:

• First, to the tightly focused Atomic Energy Commission;
• Beginning in 1974, to the more diffuse Energy Research and Development Admini- stration;
• And finally, in 1977, in response to the oil crisis, to the sprawling Energy Department, charged with everything from regulating nuclear power to researching electric cars and better thermal insulation.

Lab scientists, technicians, security guards and support personnel all worked for private entities-the University of California for the Los Alamos and Lawrence Livermore labs; AT&T and later Lockeed Martin for Sandia. The contractors, in turn, reported to the nationwide network of Energy field offices, staffed by federal employees, who in turn reported to an Office of Field Management in Washington.

But the contractors got their budgets not from Field Management, but from other offices in Washington-and because most labs ran several different types of programs, they got money from several different sources.

What's more, another set of headquarters offices issued policy directives to the field offices and labs without regard to Field Management orders or budgets, and without even consulting their headquarters colleagues.

Now, says Energy Undersecretary Ernest J. Moniz, "we have reconfigured to get responsibility and authority and accountability lined up." The Office of Field Management has been abolished. Each field office now reports directly to one, and only one, of the three headquarters offices-Defense Programs, Environmental Management and Science-which have always funded most lab activity. Policy directives from all other headquarters offices must be cleared by the new Field Management Council, led by the three main offices and top Energy officials.

A nice start, but "you still have these overlaying bureau- cracies," argues Rep. Mac Thornberry, R-Texas, who introduced a companion bill to Kyl's in the House. While Rich-ardson reformed the field offices, "he does not put all of the weapons labs . . . directly under the [assistant] secretary for defense programs," Thornberry says, whereas the Agency for Nuclear Stewardship plan would give Defense Programs full authority over the laboratories.

The Rudman report recommends eliminating the field offices altogether, calling them a useless layer of bureaucracy. "Frankly they're not part of the solution, they're part of the problem," Rudman says. "It's going be interesting to see what [Congress does] with the field offices," which represent thousands of home-state jobs. "Any time you have jobs involved it's a tough one," Rudman says.

'Cynicism and Disrespect'

No matter how Energy's formal structure ends up, all agree the more profound problem is the intangible one. "Never have the members of the Special Investigative Panel witnessed a bureaucratic culture so thoroughly saturated with cynicism and disrespect for authority," the Rudman report said. Cobbled together from disparate agencies and assaulted by abolitionists since President Reagan's election in 1980, the department long ago lost control of its component subcultures and became increasingly unable to stop them from appealing directly to allies on Capitol Hill.

While the labs are infamous for such end-runs, one high-level Energy official says this syndrome was most pronounced in "a culture of dissent and maverickness . . . among the security people." Federal officials who leaked reports of failures to secure nuclear materials or guard the gates looked like heroes to the press and Congress. But the Secretary's office believed they were undermining the department in order to win bigger budgets for their own independent and uncoordinated operations.

"What you had," says Richardson, "was a security structure with no accountability [that was] constantly carping at the programs and at the political appointees. This is what I'm going to change."

On June 17, Richardson appointed a new "security czar"-retired Air Force general Eugene E. Habiger, who once controlled all the military's nuclear weapons as chief of the U.S. Strategic Command. While Habiger is an adviser to Richardson without operational activities under his direct control, he is supposed to shape the security budget for every entity within the department and then closely monitor implementation-much as Curran does for counterintelligence.

Habiger's appointment, like Curran's, fits into Richardson's pattern of asserting central control over the disparate elements of the department, especially the labs. "Clearly we have to make the labs more accountable, and that's happening," said Richardson in May. "The fight is over, and I have won. They work for me."

The Labs Win

Richardson was not reckoning with Rudman or Kyl, however. Their reform proposals bear the stamp of lab allies. Kyl's co-author, powerful Republican Sen. Pete Domenici, R-N.M., has long protected the Los Alamos and Sandia labs, which are in his home state. Rudman relied on what he called the "invaluable contributions" of Sidney Drell, a career lab scientist who served on the Intelligence Board's investigative panel.

One Rudman recommendation is particularly telling. Not only does it call for freeing the labs from the much-resented field offices, but it also would slash Energy's new Office of Intelligence. The report recommends leaving only a few liaison officers who would still report to the Secretary, which Richardson quickly said he would accept.

Richardson fought longer against the heart of Rudman's and Kyl's plans, however: creation of a new agency to consolidate all Energy's national security functions in an organization that technically reports to the Secretary but possesses substantial autonomy and is largely outside the day-to-day control of Energy headquarters.

The weapons labs have long had mixed feelings about such plans. The upside: They would no longer be lost in a vast, diversified department, but once again would be the centerpiece of a focused agency, much as in the old Atomic Energy Commission days. The downside: Such a focused agency might not support the labs' extensive work outside the nuclear weapons field. Skeptics consider such "multifunction" research a distraction, while the labs insist it is a vital bulwark of their core mission.

Both Rudman and Kyl give the labs the best of both worlds: a special agency built around their weapons mission, but with strict instructions to collaborate with Energy's 24 civilian labs and with outside agencies on non-weapons work as well. It was to preserve "this exceedingly productive relation," scientist Drell says, that "I strongly favored [creating] a semi-autonomous agency . . . freed of a lot of the complications and baggage [but] still within DOE." In other words, the labs would still get department support where they want it, but not department control where they don't.

Though both the Kyl legislation and the Rudman report do have the newly created nuclear weapons agency reporting directly to the Energy Secretary. But as detailed by Rudman, the director would also have his or her own budget, a general counsel and a guaranteed tenure of five years-five more than any Energy Secretary can count on. Richardson initially insisted such a scheme would fatally compromise his efforts to impose discipline upon the department.

But is such discipline really possible? "I don't think there's any chance that headquarters can control the labs," scoffs a former Energy official. "And do we really want to? . . .You cannot govern DOE through some kind of central, top-down control structure."

Scientific Independence

This is the Energy Department's central dilemma: While such activities as spy-catching and security are amenable to centralized control, cutting-edge science is not.

"I don't want to hurt the labs-their productivity, their interactions with other nations, with other scientists," says Richardson. "We can't over- react." Undersecretary Moniz adds, "I don't believe it's true that there is a balance between security and science in the sense of a seesaw-one goes up, one goes down."

Yet an overreaction-increasing security at the expense of science-is precisely what many current and former lab scientists fear in Richardson's actions. "It's a tragedy that they haven't found the balance," says the former weapons scientist.

Exhibit No. 1 is the computer stand-down that occurred in April, when the classified computer systems at the three weapons labs were shut down for 12 days for security upgrades and training. "In terms of man-days, this was probably the largest" stoppage of scientific work in the department's history, acknowledges Reis. "[But] we found things that really required some fixing." And in response to the Rudman report, Richardson hastily ordered another stand-down for two days in late June.

Scientists also are skeptical of the new permanent security measures, such as increased lie-detector testing and Curran's proliferating FBI agents. "A lot of the lab people were resisting the polygraph and additional intelligence folks at their labs," Richardson acknowledges, "But I said we've got to do this. We're doing it."

Just how disciplined lab scientists are, or can be, with sensitive information is a bitterly contested question. "The folks that I had a chance to work with . . . were obviously very bright and capable people in their particular fields," says Patrick Eddington, a former CIA analyst, "[but to a scientist], the concept of counterintelligence is foreign." The former weapons scientist was harsher, calling colleagues "naive."

"Although we found arrogance and naivete in some cases, these are academics whose basic culture is to share and to be open and to collaborate," says Rudman. "Even though they may have some attitudes about security and counterintelligence that are really quite antithetical to good management, nonetheless, that's really not their primary responsibility."

But Schiffer, the ex-FBI agent now at Los Alamos, says, "I was pleasantly surprised at the degree of security awareness of the Los Alamos scientists and employees. . . . They were more security-conscious than I was."

There's more to the security issue than whether or not scientists can keep a secret. The classified core of nuclear weapons science is supported by a broad base of unclassified information. "The laboratories aren't the font of all computer science and all chemistry and all physics," notes Reis. Especially now that nuclear tests are banned, the labs must supplement their own research by keeping careful watch on their academic colleagues, who may quite unwittingly discover something that can be turned to a military use. While the most important pieces of the puzzle are stamped "secret," others are scattered throughout the open scientific literature.

So how can the security leaks be stopped? Just "put a tourniquet on it," says Richard C. Shelby, R-Ala., chairman of the Senate Select Intelligence Committee: If the free flow of information is so uncontrollable, then just cut it off.

But lab scientists and Energy bureaucrats unite when threatened with such Draconian proposals. "You really have to interface with the outside," Reis says. Science requires the exchange of ideas, he says-"you can't do it in a monastery." Kyl and Rudman appear to agree, so in the battle to reform the Energy Department, the winners are unlikely to destroy nuclear weapons science in an effort to protect it.

Sydney J. Freedberg Jr. is a National Journal reporter.