Keep the Facts of the Cox Report in Perspective

Representative John M. Spratt, Jr.

Back to Cox Report Main Page

The Select Committee on U.S. National Security and Military/Commercial Concerns with the People's Republic of China (PRC) was chartered in the wake of allegations about U.S. satellites launched on PRC rockets. Three rocket failures over 38 months were followed by accident investigations that were not licensed by the State Department and resulted in technology transfers to the PRC. It was initially alleged that U.S. encryption technology had been compromised when the Loral-built Intelsat 708 satellite crashed shortly after launch. This rumor turned out to be unfounded. It was also alleged that Motorola had helped the PRC design a platform for off-loading Iridium satellites that was a precursor to a post-boost vehicle for off-loading multiple independently targetable reentry vehicles (MIRVs). This too was unfounded. Finally, it was alleged that Hughes and Loral had helped the PRC improve the accuracy, range and payload capacity of its rockets and missiles. The committee investigated the launch failure investigations in detail, and concluded that China's rockets and missiles may have gained reliability as a result of transfers, but nothing to increase or improve range, payload capacity or accuracy.

If the committee had ended its work here, the report would have raised concerns about exports controls and technology transfer, but revealed nothing earthshaking. In mid-October, however, committee staff learned that design information about the W-88 warhead had been lost to Chinese espionage. The committee heard testimony about these losses at hearings on November 12 and December 16. Due to the shortness of time, the record on this subject is much thinner than the extensive record compiled on Loral and Hughes.

The committee reports that in 1995, the CIA learned from a PRC "walk-in" of nuclear secrets lost to Chinese espionage. The national security advisor, Samuel R. Berger, was informed of espionage at Los Alamos National Laboratory in April 1996, but the president was not briefed until early 1998, and Presidential Decision Directive 61 was not issued until February 1998. The administration is criticized for not acting quickly or vigorously. But in July 1996 the House and Senate Intelligence committees received a similar briefing, instigated by the national security advisor, and neither sounded an alarm nor took action. The lag seems to be due in part to fragmentary, uncertain facts and an inconclusive investigation. The select committee required three staff briefings, one from the FBI and two from the Department of Energy (DOE), two hearings, and multiple telephone calls to get the facts for this report, which was prepared two years after the national security advisor was first briefed.

One is still baffled by how this investigation has been handled, but security lapses and miscues between the FBI and the DOE date back to the early 1980s when information about the W-70 was lost from Livermore and red flags were raised but not heeded. The W-70 is an enhanced radiation warhead, sometimes called the "neutron bomb." The first clear call for beefed-up security and counter-intelligence against China came in the early 1980s when loss of W-70 design data was strongly suspected. When China tested a neutron bomb in 1988, the need for stronger security should have been unmistakable, yet nothing was done.

Among those beating the drum for tighter security were the General Accounting Office (GAO) and the House Energy and Commerce Committee. The Carter administration set up an interagency group reporting directly to DOE's inspector general on security at DOE facilities. This group engaged experts in anti-terrorism and protection of classified material. In 1981, the assistant secretary for defense programs was replaced by the Reagan administration with an appointee who disbanded the interagency group. The oversight subcommittee of Energy and Commerce, under Chairman John Dingell, kept up the drumbeat for an independent Office of Safeguards Evaluation, and was answered by Secretary of Energy John S. Herrington with assurances of "significant progress." In 1991 and 1992, the oversight Subcommittee received six GAO reports critical of DOE's safeguards and security. These reports cited lax internal controls and weak accountability for classified documents. It was during this period, between 1984 and 1992, that design information about the W-88 warhead was stolen by Chinese agents.

All of these facts raise serious concerns, but they need to be kept in perspective. There was too little time to conduct an independent investigation of this subject before the committee's charter expired. The committee relied heavily on a few witnesses, and did not substantiate their testimony with weapons experts at the national labs or interagency review. I do not fault the witnesses we heard from; they performed a valuable service by alerting us to serious failures in security. But they did not have the technical background to fully assess the nature or value of the information lost. The committee did not have time to call the senior statesmen of the nuclear labs, like Harold Agnew and Johnny Foster, for their perspective. Partly because of haste, there are statements in the report that will not stand scrutiny. I objected to a number of these in our mark-up of the "Overview," but not all were deleted or revised, and some of the revisions are still inadequate, in my opinion.

The first page of the "Overview," for example, makes one statement that is simply incorrect. It says that the United States has not deployed an enhanced radiation warhead or a neutron bomb. In fact, we have deployed three: the W-70 on the Lance tactical missile; the W-66 on the Sprint interceptor; and the W-79 on an 8-inch artillery round. This is not a serious mistake, but a report of this importance should not contain such mistakes. Two other sentences are much more serious. The opening sentence states that "The PRC has stolen design information on the United States' most advanced thermonuclear weapons." That is a sweeping charge. It is followed by the statement that "The stolen U.S. nuclear secrets give the PRC design information on thermonuclear weapons on a par with our own." That is even more alarming, but is it accurate?

Here is the intelligence community's best estimate of what the PRC has obtained:

• Late 1970s: Design information on the W-70 enhanced radiation warhead.

• 1984-1988: Design information on the W-88 warhead and its reentry vehicle.

• 1984-1988: Classified (but not design) information on reentry vehicles and weight-to-yield ratios of the W-62 (Minuteman II), W-76 (Trident C-4), W-78 (Minuteman III), and W-87 (Peacekeeper).

The only design data we know for sure to be lost is data on the W-70 and W-88, and it is hardly accurate to call the W-70 one of our "most advanced thermonuclear weapons." The W-88 is unquestionably one of our most advanced warheads, but a reader can come away from the report with the impression that China obtained design information on all of our most advanced nuclear weapons. The select committee did not find evidence of that.

The intelligence community simply does not know whether blueprints or design documents have been stolen. The W-88 has several thousand parts and a highly sophisticated arming and firing system. A full set of blueprints would be voluminous, and most of those who know find it hard to believe that China obtained all of this data. Still, the loss of any restricted data on nuclear weapons is serious and should not happen.

As for the loss of neutron or enhanced radiation technology in the mid-1990s, the evidence is limited. It is based on a single piece of intelligence, whose reliability has not been determined. There is no settled conclusion within the intelligence community that information was lost; but the information, if stolen, is esoteric physics that may have little military application.

The "legacy codes" downloaded from Los Alamos' classified computer, if they have been stolen, are an alarming loss. Just their transfer to an insecure computer raises serious questions about computer security in the national labs. A lot of what scientists know about nuclear materials is empirically based rather than scientifically understood. The legacy codes are mathematical equations that model physical phenomena observed in the explosion of a thermonuclear weapon. They record, for example, how neutrons and protons move through matter, how shock waves go through materials and how heat is transferred. This is a mother-lode of empirical data, and if the PRC has obtained these codes, their ability to model nuclear explosions will be enhanced.

But these legacy codes are not blueprints or three-dimensional models of bombs or warheads, and even if these codes have been lost to the Chinese, it is a reach to say that "stolen U.S. nuclear secrets give the PRC design information on thermonuclear weapons on a par with our own." The United States has conducted 1,045 nuclear tests and built more than 30,000 nuclear weapons. China has conducted fewer than 50 nuclear tests, and built a fraction of the weapons we have built. The PRC has tested a neutron bomb and a W-88 derivative, but the PRC has not replicated the W-70 or the W-88. In addition to a complex physics package, the W-88 has sophisticated electronics and thousands of parts, which the PRC would find hard to replicate. In view of the enormous differences in experience and capability, it is simply not accurate to say that China's design is on a "par" with our own. I disagreed when the witness said this before our committee; I disagreed when our committee marked up the report; and I disagree now.

Harold Agnew was director of Los Alamos Scientific Laboratory when the W-88 was developed. He offered his assessment of the potential loss in a letter to The Wall Street Journal on May 17, 1999:

The design of the actually quite old. The basic test was done by Los Alamos Scientific Laboratory when I was director, and I retired 20 years ago. It is a 'delicate' and neat package. Having the computer printouts, as I remember them, gives a general idea, but actually being able to manufacture the total system from a computer code is a different matter. No nation would ever stockpile any device based on another nation's computer codes.

The committee surmises that China wanted W-88 design data to develop a small warhead with sizable yield for its a road-mobile missile which has been in development for years and is expected to be their next deployed ICBM. In addition to seeking our designs, the Chinese were apparently listening to our debate about the MX, and concluded that their large, liquid-fueled CSS-4s were vulnerable to a pre-emptive first strike. A road-mobile missile is a survivable alternative, and with a single warhead, it is essentially a second-strike retaliatory system.

Had the committee made time to hear from witnesses like Dr. Agnew, some of the statements cited above probably would not have survived the cut. There are, unfortunately, places where the report reaches to make a point and exaggerates. The report states, for example, that "the PRC has stolen a specific U.S. guidance technology used on current and past generations of U.S. weapons systems." The guidance component in question is employed on aircraft like the 747 and has been commercially available for years, though it does require an export license. I moved in mark-up to strike this reference, but to no avail.

The report also states that the PRC has stolen data about an electromagnetic gun, and that "such technology, once developed, can be used for space-based weapons to attack satellites and missiles." This program was started in the late 1970s by Defense Advanced Research Projects Agency, and funded for a while by Strategic Defense Initiative but dropped from its budget. It is in basic research today as an Army program, and is so experimental that the Army does not expect to have a weapon system using this technology before the second decade of the next century.

These exceptions do not alter the report's basic finding. The Chinese are seeking advanced technologies. They are doing it aggressively and by means that include espionage because they are not on par with us. But they are capable. The Chinese first developed and tested nuclear weapons in the 1960s and built missiles with megaton warheads that have been able to reach the U.S. since the early 1980s. And of 28 U.S. satellites launched on Chinese rockets, 25 have been placed in orbit.

Congressman Norm Dicks, the committee's ranking Democrat, calls the report "a worst case scenario." It assumes the worst to make its message clear: Our security and counter-intelligence systems have fallen short of the challenge and need to be strengthened, restructured from the bottom up. All members of the committee, five Republicans and four Democrats, reached that conclusion. Not only our national labs but satellite exporters and others have allowed security to lapse and sensitive technologies to be lost. The losses do not significantly shift the balance of power between the U.S. and China. We still have an overwhelming advantage in military technology in both nuclear and non-nuclear systems. But the transfers should not have occurred, and most probably would not have occurred if a stronger security and counter-intelligence system had been in place.

In the end, no security measures, counter-intelligence efforts, or export controls are likely to be totally foolproof or completely adequate. Their chief purpose is to limit and impede the pace at which China develops things like new warheads with higher yield-to-weight ratios. One means to this same end is the Comprehensive Test Ban Treaty (CTBT). I wanted to add to the report a few sentences observing that the CTBT would impose major impediments on China's development of new nuclear weapons, without seriously diminishing the reliability or safety of our nuclear weapons. The committee did not adopt that statement, but even without it, the report makes a case for the CTBT.


Representative John M. Spratt, Jr. (D-SC) was a member of the Cox Committee.