Christopher Cox
Christopher Cox; drawing by David Levine

In the early 1950s, at the height of the McCarthy era, “allegations arose” (as the Cox report so vaguely and aptly puts it) that Qian Xuesen, a Chinese-born American rocket scientist, was a spy for the People’s Republic of China. Qian had fled the Japanese invasion of China in 1935, emigrated to the United States, and earned a master’s degree from the Massachusetts Institute of Technology and a doctorate from the California Institute of Technology. He was recognized as one of the world’s foremost experts on jet propulsion, commissioned as a colonel in the US Air Force, and honored for the pioneering work he had done for his adopted country, including development of the Titan intercontinental ballistic missile.

Yet, when the “allegations arose,” Qian was stripped of his security clearances and removed from missile work. In his disgrace, China invited him to return home, and in 1955 the United States let him go. He took four fellow Chinese scientists with him and, together, they created the Chinese ballistic missile system, including the CSS-4 missile, which is currently targeted against the United States. China’s nuclear missiles, which carry about two dozen warheads compared to the eight thousand or so possessed by the United States, are universally believed to be intended as a retaliatory force, threatening the US with, say, the loss of Los Angeles if it ever went to war against China. One credible missile is all China needs to deter the kind of US military operation that was waged against Serbia. This is what Qian gave China.

Qian’s history, recounted in the House investigating committee’s report on Chinese espionage, offers the best explanation for the bizarre sequence of events that has led America once again into a witch hunt for Communist spies burrowed within our most secret defense establishments. Once again, as in the 1950s, the most reckless charges have been made against apparently innocent people, an opponent’s military capabilities have been grossly exaggerated, and unscrupulous politicians have blamed this overdrawn national security threat on the complaisance and possible corruption of their political foes, in this case the Clinton administration.

On the basis of scanty evidence, Republicans and even a few Democrats have demanded the resignations of Attorney General Janet Reno and National Security Adviser Sandy Berger. A Chinese-American scientist at Los Alamos National Laboratory, Wen Ho Lee, has been identified by unnamed government officials—and convicted in parts of the press—as a spy for China on the basis of evidence that is laughably thin. And, most irresponsibly, the Cox report suggests that every Chinese visitor to this country, every Chinese scholar, every Chinese student, every Chinese permanent resident, and even every Chinese-American citizen is a spy, potential spy, or “sleeper agent,” merely waiting for the signal to rise up and perform some unimaginable act of treachery.

Congressional Democrats and the Clinton administration, the intended main target of this Republican-led investigation, refuse to fight back, apparently unwilling to be seen as minimizing the dangers of Chinese espionage. All four Democrats on the Cox committee signed the report, even though one of them, Representative John Spratt of South Carolina, forcefully and eloquently rebutted its leading conclusion that as a result of its technology thefts, China now has nuclear weapons design information “on a par” with the United States.

The leading Democrat on the committee, Representative Norman Dicks of Washington, apparently thought that, if he went along with the report’s broad conclusions, he would both wake up the Energy Department from its past lethargy about admittedly lax security at the nuclear weapons labs and also defuse any Republican attempt to turn the report into a partisan political weapon. A White House official explained the Clinton administration’s supine response this way: “They thought this thing would fall apart of its own weight as soon as it saw the light of day.”

It is true that many nuclear weapons experts dismissed the Cox report’s key findings. “This is completely overblown,” Spurgeon Keeney, president of the privately funded Arms Control Association, told me. “There is nothing to back up the assertion that this information puts China on a technical par with the United States. A lot of the stuff they are making a big deal about you will find in the Nuclear Weapons Databook that was put out [in the 1980s] by the Natural Resources Defense Council, including pictures.” Harold Agnew, a former director of Los Alamos, wrote to The Wall Street Journal that stolen warhead information would have limited value for the Chinese and suggested that any purloined computer codes may have a hidden virus that would thwart attempts to replicate weapons. John Foster, former director of the Lawrence Livermore National Laboratory, told The Washington Post that basic design data—size, weight, shape, and yield—even though highly classified, would not enable the Chinese to build a nuclear weapon.


Charles Ferguson, a nuclear physicist and former Navy officer now at the Federation of American Scientists, described the main Cox charges to me as “scaremongering,” probably part of the ongoing conservative campaign to justify the creation of a costly and controversial national ballistic missile defense system. Similarly, Stephen Schwartz, publisher of the Bulletin of the Atomic Scientists, wrote in an editorial entitled “A Very Convenient Scandal”* that “the story broke at a fortuitous time for advocates of legislation promoting the swift deployment of ballistic missile defenses.” Schwartz also noted that the first public allegations that China had stolen US nuclear weapons technology had been published in 1990, when George Bush was president, and resulted in no similar outcry—and no improvement in security at the labs. And the CIA released a damage assessment that concluded that while China had obtained basic design information on US nuclear warheads, (1) it was not clear how comprehensive its information was, (2) it was impossible to tell how much had been obtained through espionage rather than dogged research, and, in any event, (3) China had not modernized its nuclear weapons arsenal as a result. On June 14, the President’s Foreign Intelligence Advisory Board, headed by former Republican Senator Warren Rudman of New Hampshire, scorned the Cox report in these words: “Possible damage has been minted as probable disaster; workaday delay and bureaucratic confusion have been cast as diabolical conspiracies. Enough is enough.”

But this is a political uproar, not a scientific or military one. Instead of collapsing from its own overreaching, the Cox report has become a feature of talk radio and a staple of the Republican presidential campaign for 2000, which is already underway. Republican politicians and their supporters in the press and television have tried to make a scandal about when, precisely, President Clinton was first briefed on the suspicions of espionage, whether Berger dragged his feet, and whether Reno’s Justice Department improperly thwarted an FBI request for wiretaps on a suspected spy. It does appear that Clinton and Berger were slow to respond, but their defense is that the information they received both was vague and referred to losses that had occurred a decade earlier. Reno herself told a news conference on May 27 that the FBI had never produced evidence that Wen Ho Lee had knowingly committed a crime, which is the prerequisite for ordering a tap on a UScitizen. In the end, Clinton ordered a tightening of security at the nuclear laboratories.

From the information in the Cox report—as opposed to the unsourced and far more dire leaks that preceded it—it is clear that suspicions about espionage were sketchy, and it is not certain to this day whether any espionage, in the traditional cloak-and-dagger sense, occurred. So far, Wen Ho Lee has not been charged with any crime.

The press, which is supposed to function as a check on the excesses of government, has generally suspended its skepticism about the thinly grounded charges—no newspaper has been more credulous than The New York Times—and occasionally exaggerated them in defiance of common sense. In a front-page article on May 14, for example, the Times quoted its sources as asserting that China was installing a small nuclear warhead on a new mobile missile, the DF-31, even though it was clear from reading the story that China has not yet flight-tested the DF-31. According to Howard Diamond, a senior specialist on China at the Arms Control Association, it would be most unlikely that the characteristically prudent Chinese would place nuclear explosives on an untested missile that may or may not go where it is aimed. A US government official says more simply, “The Times just got it wrong.” There has been greater skepticism—and far more balance—in the reporting of The Washington Post (especially that by Walter Pincus and Vernon Loeb), The Wall Street Journal, and the Los Angeles Times. But The New York Times, now a national paper, has single-handedly set the tone for the national debate.

The facts are these: in the late 1970s, China apparently stole information relating to the US neutron bomb, an enhanced-radiation weapon that kills people but is relatively harmless to property. The theft was disclosed in October 1990 by George Carver, former deputy director of the CIA, who told employees of the Lawrence Livermore National Laboratory that Chinese agents had taken the information from their lab. The information was apparently not all that complete; it took the Chinese ten years to test a neutron weapon, and to this day they have not deployed one. Then, between 1992 and 1996, China conducted a series of underground nuclear tests whose seismic patterns and other characteristics led some US weapons experts to believe it had managed to make a modern, small nuclear explosive like the US W-88, an aspherical weapon deployed as one of the multiple, independently targeted warheads on the D-5 missile aboard Trident submarines. In 1995, a Chinese intelligence agent, who claimed to be betraying secrets obtained by China, approached the CIA and handed over a document containing highly classified design information on the W-88 and technical information on other US nuclear weapons.


Much of this information had been published during the 1980s by the Natural Resources Defense Council, an environmental and antinuclear activist group; even though it has been made public, some of it is still classified. But two sensitive pieces of information on the W-88 were not available anywhere, a US government official says. The design and technical information was apparently fragmentary; there is no evidence that actual blueprints or complete designs have been lost to China. In addition—and perhaps the most intriguing aspect of this strange episode—the “walk-in” agent was soon determined to have been a plant, a controlled agent sent by the Chinese Ministry of State Security.

Despite this ambiguity—a double agent providing information that was largely in the public domain—the agent’s document set off a hunt for a spy in US weapons laboratories. Investigators, chief among them the Energy Department’s Notra Trulock, homed in on Wen Ho Lee, who knew how the W-88 was designed and who had visited China on an approved exchange mission in 1988. In addition, “allegations arose” that Wen Ho Lee was once seen hugging a person believed to be a Chinese intelligence agent; and he had once made a telephone call to another suspected spy that sounded as if they were in sympathy. Most damaging, Wen Ho Lee was later found to have downloaded so-called “legacy codes,” a computer database that contains the history of US nuclear weapons development, into an unclassified computer.

There was—and is—no evidence that Lee conveyed these codes to an unauthorized source. He claims, through his lawyer, to have downloaded only unclassified files for his own use, as he was permitted to do by lab policy. Even if he had passed the codes on, however, their usefulness is debatable. These are older, two-dimensional codes that would give the equivalent of a cross section of a nuclear weapon, rather than the three-dimensional codes the US has used since the late 1970s for design and simulated testing. “The difference between two-dimensional codes and three-dimensional codes is the difference between the special effects in a 1950s movie like Destination Moon—a cardboard rocket on a thread—and what you see now in Star Wars,” said John A. Pike, an intelligence and weapons analyst at the Federation of American Scientists. “In computer jargon, the term ‘legacy’ means ‘obsolete.”‘

Nevertheless, the codes were a valuable part of history. “They’re like a B-52 that has been updated and improved over the years,” an Energy Department scientist told me. “There is no justification for downloading them onto an unclassified computer,” as Wen Ho Lee did. He was duly fired from his job at Los Alamos for security infractions. In a comprehensive six-page rebuttal to the leaked accusations against him, his lawyer, Mark Holscher, pointed out that Lee’s trips to China were approved in advance by Los Alamos and the Department of Energy. With Lee’s approval, his wife became an FBI informant and let the FBI monitor her conversations with Chinese scientists. (In addition, the FBI ran a sting operation against Lee, inviting him to spy for China, and he refused.) As for the computer files, Holscher wrote, “In fact, Dr. Lee took substantial steps to protect his computer files and we are confident that federal investigators will also conclude that no third party could have or did access his protected computer files.” On June 15, the Times’s David Johnston reported that the evidence against Lee is so thin that it is unlikely that he will ever face criminal charges for espionage—and may not even be accused of any wrongdoing.

If Lee turns out to be innocent of a crime, as he appears to be, we are left with a spy scandal—the worst case since the Rosenbergs, Republicans have called it—without a spy. It is undoubtedly true that China scours public sources, grills US scientists at seminars, and seeks to “reverse-engineer,” or re-create, whatever US technology it can obtain. In the 1990s, the Department of Energy declassified many nuclear secrets, which then became easily available to Chinese scientists. In an article published on May 30, The New York Times asserted that these declassifications were made by the Clinton administration in hopes that somehow such openness would deter the spread of nuclear weapons. “That was a very bizarre piece,” says Robert S. Norris of the Natural Resources Defense Council, which sent two senior staff members to the Department of Energy to help carry out the department’s “Openness Initiative” of 1993. “It is just a wacky idea that [the then Energy Secretary Hazel] O’Leary was gambling that releasing the information would stop weapons proliferation. This stuff was decades old.” In fact, the declassified “secrets” cited by the Times—information on the construction of hydrogen bombs, minimum amounts of fuel, and how to boost a bomb’s power—can be found in, or inferred from, the NRDC’s Nuclear Weapons Databook. In any case, information declassified in 1993 or later would have been released long after China had supposedly gotten the W-88 secrets and would have come too late to assist in the nuclear tests of the early 1990s.

In addition to the Energy Department’s declassifications, US suppliers have been relaxed about selling technology that might conceivably have military application, like the inertial navigation systems now used on jetliners. But it defies logic to accuse China of espionage for “stealing” what, in an open, mercantile society, is lying around in libraries or databases or is up for sale. Even in the case of nuclear weapons, theoretical information may be of marginal use. “Everyone knows the concepts anyway,” a government official told me. “It’s the engineering that’s tricky.”

The Cox committee was created by former House Speaker Newt Gingrich, a Georgia Republican, in the apparent hope that it would find evidence linking illegal Chinese campaign contributions to the Clinton administration’s approval of or complicity in illegal US aid to the Chinese military. This suspected connection is the subject matter of Year of the Rat, a work that ranges from the dishonest to the just plain goofy. The two authors are a current and a former staff member of the Republican-controlled House of Representatives and they marshal their evidence in the most damaging way possible for the Clinton administration, closing with a demand for his impeachment. Yet, while they recount the unattractive greed of the Clinton fund-raisers and the eagerness of Chinese hucksters to buy influence in the United States, they are never able to make the connection promised in their title, that Clinton compromised US national security for cash. And the very existence of this book begs the question they raise: If their speculations are valid, why did the House of Representatives for which they work—a House whose Republican leaders were eager to charge Clinton with truly impeachable offenses as opposed to the murky intrigues of the Lewinsky affair—fail to impeach Clinton for treason?

The answer is that Timperlake and Triplett have no case. Their notion of what constitutes evidence is truly bizarre: the New York Times columnist William Safire made a vague charge about suspicions of Chinese influence, and Safire’s accusation has never been denied, so therefore for Timperlake and Triplett it must be true. According to the Washington Times, they write, it was “common knowledge” at the Commerce Department that Hillary Rodham Clinton had engineered the hiring of fund-raiser John Huang, so therefore she did. CIA witnesses refuse to answer questions, so therefore the answers must be damaging. The book is filled with sentences like this one: “We can only imagine what would have been shown [about John Huang’s supposedly sinister activities] if the telephone records from all of 1995 had been available….” They work themselves into a dither over the refusal of former Navy Secretary John Dalton, a Clinton political appointee, to agree that the Chinese merchant fleet poses a tactical or strategic threat to the US Navy. Dalton’s position, they reason, in a leap worthy of the Flying Wallendas, means that the United States cannot even plan to target Chinese shipping in the event of war. This makes no sense.

Here are two more examples of grossly misleading claims. The authors charge, as another example of Clinton administration perfidy, that the Chinese use some US-sold satellites for military communication. Yet anyone who knows as much classified information as they claim to possess would also know that when China puts its military signals into the air via a US-built satellite, it exposes its military communications to US interception. That is a strategic gain for the United States, not a betrayal of our national security. The authors don’t mention this; true, government officials are not supposed to discuss such sensitive matters as electronic intelligence-gathering, but they do not give the reader a fair deal if they make their accusation in the knowledge that there is a secret exculpatory explanation.

Second, they persistently warn that China sells weapons to Arab foes of Israel and thereby endangers Israel’s security. But if they know this much about Chinese arms trading, they must also know that Israel is a major military supplier of China, offering, according to the Cox report, aircraft and missile technology, including help for China’s F-10 fighter and airborne early-warning systems. This fact conflicts with their argument, and so goes unmentioned. Timperlake and Triplett boast that they are sources for The New York Times, which may account for the tenor of some of the Times’s coverage of the Clinton administration’s relationship with China.

In setting up the Cox Committee Gingrich—like Timperlake and Triplett—seems to have been fascinated by the fact that Bernard Schwartz, head of the Loral Corporation, is the Democratic Party’s largest single donor, while Loral helped China diagnose the failure of one of its civilian satellite-launching rockets. Similar help was given to China by the Hughes Corporation, whose chairman, C. Michael Armstrong, made trade-promotion trips with former Commerce Secretary Ron Brown. In the event, no connection between campaign money and defense information to China was found and the report makes only passing, seemingly gratuitous, references to campaign contributions.

The entire second volume of the Cox report is taken up with the Hughes and Loral cases, but both can be summed up briefly. After two Chinese rockets carrying Hughes satellites exploded on launch, Hughes determined that the fault was not in their satellites but in the “fairing”—nose cone—of the Chinese rockets. It informed the Chinese of this fact, and Defense Department officials thereupon charged that by helping China improve its civilian rocket, Hughes had given China information that could theoretically be used on military rockets, even though China has no military rockets that use fairings.

The Loral case is almost identical: Loral helped China diagnose the cause—a loose solder connection in a guidance system—after another civilian rocket exploded during a satellite launch. The guidance system is unlikely to be used for military rockets, the Cox report acknowledges. In both the Hughes and Loral cases, the real damage—if it can be called damage—is that China learned how to conduct a diagnostic investigation of a rocket failure. The Cox committee charges that this knowledge allows China to build more reliable military rockets; but it also reports that there is no evidence to support the original allegation—leaked to the press last year—that Hughes and Loral had helped China to make its nuclear missiles more accurate, more robust, and more deadly.

In the press conference on the release of his report, Christopher Cox, a telegenic California Republican, insisted that his committee was releasing only facts, not opinions. His assertion is contradicted, however, on virtually every page of this three-volume report. Here is one sample of its speculative language:

The Select Committee judges that, if the PRC were successful in stealing nuclear test codes, computer models, and data from the United States, it could further accelerate its nuclear development.

And another:

The Select Committee judges that the PRC will in fact use a small nuclear warhead on its new generation ICBMs [intercontinental ballistic missiles].

This judgment is not a matter of “fact”; it is a fairly safe prediction. But what the report fudges is that this new small warhead does not appear, according to US intelligence, to be based on US designs. In any event, a smaller warhead is no more threatening to the United States than the large city-busters China has already. China, with a defense budget one eighth the size of America’s, does not seem likely to develop an extensive, Soviet-style missile force with multiple warheads aimed at US nuclear missile silos.

On a central point, the Cox report is artfully worded to create the most dramatic possible effect:

The stolen information includes classified information on seven US thermonuclear warheads, including every currently deployed thermonuclear warhead in the US ballistic missile arsenal.

“That is the most misleading statement in the report,” a government official told me. “It makes it sound as if there has been total penetration of all our secrets.” But note that the Cox report does not say how detailed this stolen information is. Since virtually everything about nuclear warheads is still classified—even after it has been made public by activists like the Natural Resources Defense Council—the report’s assertion would be technically true even if China had obtained minimal, publicly available information on most of the warheads. On the key question of how much detailed information China has acquired, the report merely speculates, “The PRC may have acquired detailed documents and blueprints from the US national weapons laboratories.”

The Cox report is most reckless in its sweeping assertions against the Chinese community in America. Every one of the 80,000 Chinese who visit the US each year is a suspect, it claims, because “almost every PRC citizen allowed to go to the United States…likely receives some type of collection requirement, according to official sources.” The report “judges” that China has three thousand front businesses in the US, collecting technology and information. Even your local restaurant owner could be a spy. In 1993, the Cox report says, Yen Men Kao, an owner of a Chinese restaurant in North Carolina, was charged with seeking to steal information on US Navy torpedoes, fighter jet engines, and a radar control system. Even Chinese dissidents who fled oppression in their homeland could be “sleeper agents, who can be used at any time but may not be tasked for a decade or more.” Every Chinese scholar, every scientist is subject to appeals to his or her ethnic pride in hopes of recruiting a spy for China.

The Select Committee judges that the PRC is increasingly looking to PRC scholars who remain in the United States as assets who have developed a network of personal contacts that can be helpful to the PRC’s search for science and technology information.

One assertion is flagrantly disingenuous: the committee breathlessly reports that Chinese military intelligence officers “operate directly in the United States, posing as military attachés at the PRC Embassy in Washington, D.C., and at the United Nations in New York.” Imagine that—military intelligence officers posing as military attachés! One may as well charge that the Soviet Union sent spies to the United States posing as KGB agents. Military attachés are by definition intelligence officers, a fact that can be confirmed by noting in the Pentagon’s telephone directory that the US military attaché service is part of the Defense Intelligence Agency.

Somewhat more serious is the report’s concern that high-performance US computers could help China’s defense programs through simulation of nuclear tests. The problem in controlling computer technology, however, is that it improves so rapidly that what was state-of-the-art yesterday will undoubtedly be on tomorrow’s desktops and available from a wide array of suppliers. In addition, in an age when the US Navy launches cruise missiles at Serbia with the click of a computer mouse, almost any computer technology can have a military application. The Cox report warns us, for example, that China might use advanced US computers to make military weather forecasts. Yet the US computer and high-tech industries have pressed all US administrations to relax barriers on exports of sophisticated technology. Most members of Congress—even Republicans like House Majority Whip Tom DeLay, a Texan who is now screaming loudest about China’s military might—have supported the corporations.

Perhaps the most important basic fact in this brouhaha is that in 1996 China accepted the Comprehensive Test Ban Treaty and has not conducted a nuclear test since. Spurgeon Keeney of the Arms Control Association argues that it is inconceivable that China would trust its national security to a new generation of nuclear weapons, based on stolen US codes, without conducting actual tests. Charles Ferguson of the Federation of American Scientists says that the best hope for US security is if this nation ratifies the Test Ban Treaty, which we signed in 1996, as an incentive for China’s continued adherence. But at worst, Ferguson said in an interview, “China does not present any kind of nuclear warfighting threat to us and will not for many years to come.” It is anticipated that China will deploy the DF-31 mobile missile in the next decade, government sources say, but its warhead will be a traditionally shaped spherical one, rather than a copy of the W-88.

To the Energy Department, the greatest danger is that in overreaction to the Cox report’s thirty-eight recommendations for tighter security—many of them valid—Congress will demand Draconian security measures at the labs. Over the years the General Accounting Office has issued repeated warnings that the national laboratories have inadequate security protections, including an untrained guard force, lack of background checks on foreign visitors, and classified computer systems that were not physically isolated from unclassified systems. In addition, scientists who travel abroad, even those who go to China, have been free to share unclassified but potentially harmful information with foreign specialists.

Los Alamos, in the 1940s, was surrounded by double-barbed wire fences and military police. Scientists within were not allowed to tell their wives what they were doing and were not allowed to have telephones, except for the project director, J. Robert Oppenheimer—and his was tapped. Representative Spratt, the South Carolina Democrat on the Cox committee, warns, “You quickly diminish the pool of first-rate nuclear physicists willing to work for the government if you start strapping them to polygraphs.” John Pike of the Federation of American Scientists agrees: “First-rate physicists don’t want to work in prisons.” Howard Diamond, a China analyst at the Arms Control Association, says that exchanges with Chinese nuclear physicists should be maintained because exchanges give the United States knowledge about China’s closed society that could not be obtained in any other way.

If, as the scientific experts assure us, Cox has exaggerated the Chinese threat, why are we going through this self-torture? Politics aside, return for a moment to that mysterious Chinese-controlled “walk-in” agent who offered the CIA information on the W-88. He is the central mystery in this case. Why would China risk exposing its most sensitive sources within the US military establishment by telling us that it has stolen information on the W-88 warhead?

Houston Hawkins, a former Defense Intelligence Agency nuclear weapons expert who is now a top intelligence official at Los Alamos, told The Washington Post he suspected an exquisite Chinese strategy: deliberately trigger a spy hunt, raise suspicions about Chinese scientists in this country, make them feel unwelcome, and thus stop the loss of top Chinese physicists who remain in the US after graduating from US universities. “Are the Chinese sophisticated enough to do this?” Hawkins rhetorically asked the Post. “They’ve been practicing espionage for 2,000 years.”

Even though Hawkins supports his suspicion with an apt citation of Chinese strategist Sun Tzu, his theory sounds too clever by half—until you remember the case of Qian Xuesen, the Chinese-American rocket scientist with whom we began this adventure. As a result of that McCarthyite spy hunt in the 1950s, China got him back—and with him the knowledge to build its most formidable current ballistic missile weapons system. Qian is today presumed to have been a spy but no charges against him were ever proven. He was fired when “allegations arose” and, in disgust, he went back to China. One government official, whose instincts tell him the Hawkins theory is overly convoluted, nevertheless told me, “We may be creating a whole new generation of Qian Xuesens.” If Hawkins is right, the Chinese, desperate to stop a strategic brain drain, have given us exactly the kind of bait that tempted us into hysteria forty-five years ago, and we have swallowed it.

June 15, 1999

This Issue

July 15, 1999