The Shalikashvili Report on the Comprehensive Test Ban Treaty

Report Defense

The Shalikashvili Report on the Comprehensive Test Ban Treaty

May 25, 2001 20 min read
Baker Spring
Baker Spring
Former Kirby Research Fellow in National Security Policy
Baker is a former Kirby Research Fellow in National Security Policy

Arms control advocates are pressing the Senate to reconsider its 1999 decision not to ratify the Comprehensive Test Ban Treaty (CTBT), which President Bill Clinton signed in 1996.1 They point to recommendations in a report released in January by retired U.S. Army General John M. Shalikashvili, the former Chairman of the Joint Chiefs of Staff appointed by President Clinton to examine the concerns over the treaty.2

The Senate had rejected the treaty because of serious concerns that it is unverifiable and unenforceable, and would likely force the United States into unilateral nuclear disarmament. This would jeopardize U.S. security even as many of America's enemies and adversaries are gaining increasingly more sophisticated nuclear capabilities.3 Regrettably, while the Shalikashvili report admits that the treaty presents inherent problems, it mistakenly claims that ratifying the CTBT will not reduce the U.S. nuclear deterrent and asks the Bush Administration to begin implementing its provisions as rapidly as possible.

Nuclear deterrence and nuclear non-proliferation have been essential elements of national security policy since 1945. Nevertheless, in 1999, after six former Secretaries of Defense voiced their strong opposition to the CTBT,4 its supporters in the Senate fell nearly 20 votes short of the 67 votes needed for ratification.5 Now the Shalikashvili report is being used both to convince the Administration that the "risks" inherent in the CTBT can be managed effectively and to persuade the Senate that it should ratify the treaty. The report is misguided. Members of the Senate should continue to oppose ratification and implementation of the CTBT.

The preamble to the CTBT states that the treaty's prohibition on nuclear testing "constitutes an effective measure of nuclear all its aspects."6 The Shalikashvili report paints a different picture. It states that the CTBT "allows the United States to keep a strong nuclear deterrent."7 Seemingly oblivious to this basic contradiction, the report recommends several steps that it claims would preserve a U.S. nuclear deterrent absent testing. Its recommendations could lead to charges from other signatory states that the United States was violating the treaty; and given the clear statement of objective in the treaty's preamble, such charges would be difficult to rebuff.

The CTBT itself assumes that global nuclear disarmament will occur at an unspecified time following treaty implementation. Whether this is true or not, given the array of ongoing nuclear weapons programs around the world, disarmament would be certain for the United States because it would be impossible for the U.S. to maintain its arsenal of highly sophisticated weapons, as well as its appropriately high standards for their safety, reliability, and effectiveness, without testing.8 A country relatively unconcerned about achieving a high degree of safety, reliability, and effectiveness could well field a nuclear arsenal without testing. The assumptions and recommendations of the CTBT are inconsistent with a policy of nuclear deterrence.

Like the CTBT, the assumptions and recommendations of the Shalikashvili report will not deter nuclear proliferation. There are eight key reasons why the Senate should ignore the report and not reconsider its 1999 decision against ratification of the treaty.

Reason #1: The Shalikashvili report fails to acknowledge that testing existing nuclear weapons is necessary to ensure their safety and reliability.

The report mentions that the CTBT restricts testing on new nuclear weapons, such as weapons to destroy hardened or deeply buried targets, and asserts that parties to the treaty would "be free" to keep their nuclear stockpile "safe and reliable through...testing...up to the point where the core nuclear explosive package would go critical."9 In fact, the treaty's supporters claim that testing up to this point will be sufficient.

Maintaining high levels of safety and reliability for weapons in the U.S. nuclear stockpile is necessary to keep them operational. Currently, strategic systems that fail to meet safety and reliability standards are withdrawn from active inventory until the problems are fixed. But discovering the problems frequently requires that weapons be tested. As of 1987, one-third of all nuclear weapons in the U.S. stockpile had undergone post-deployment testing to resolve problems. In 75 percent of these cases, the problem was discovered as a result of explosive testing beyond the point where the weapon would "go critical."

Periodic explosive testing of weapons in the stockpile yields information that scientists, engineers, and technicians at the nation's nuclear laboratories may not have known or reveals problems they otherwise would have no reason to suspect. A permanent ban on testing would deny them complete knowledge and thereby degrade the safety and reliability of weapons in the stockpile. Worse, the extent of degradation would be unknown.

In 1995, to assure that a high level of safety and reliability would be maintained after the CTBT entered into force, the Clinton Administration established the Science-Based Stockpile Stewardship (SBSS) program.10 Today, however, scientists in the national laboratory system have begun to suggest that the goal of the SBSS program is not achievable. In the words of Dr. Merri Wood, a senior weapons designer at the Los Alamos National Laboratory, stewardship of the U.S. nuclear weapons stockpile without testing is "a religious exercise, not science."11

The Shalikashvili report recommends a number of management and budgeting improvements for the SBSS program. For example, it recommends that SBSS program leaders focus on near-term reliability, enhanced weapons surveillance and monitoring activities, multi-year budgeting, and a mechanism for consulting with scientists outside the program. While these recommendations might improve the SBSS program to some degree if implemented, however, they cannot answer the fundamental question of whether it is possible to assure the safety and reliability of nuclear weapons in the stockpile absent explosive testing.

Reason #2: The report includes no specific proposal for designing new weapons to meet future military requirements.

The SBSS program was not designed to maintain militarily effective weapons in the face of changing military requirements. Both proponents and opponents of the CTBT in the United States have acknowledged that a permanent ban on testing would block the entry of new weapons into the U.S stockpile. The best that can be hoped for is that an existing weapon could be repackaged to meet a new requirement.12 Ratifying the CTBT would make it virtually certain that the nuclear arsenal's effectiveness would decline over time and result in a serious military requirement's going unmet in the future.

In his cover letter accompanying the report, General Shalikashvili states that the U.S. nuclear arsenal is "able to meet all stated military requirements." He goes further and asserts that "For as far into the future as we can see, the U.S. nuclear deterrent can remain effective [under the CTBT] assuming prudent stockpile stewardship."13 His assumption of a static military environment, however, is seriously flawed. By this logic, somebody could say that it will not rain tomorrow because it is sunny today: in other words, that the best indication of tomorrow's weather is today's. The repeated application of this assumption would lead to the absurd assertion that it will never rain because each prior day was sunny. This kind of illogic essentially assumes that new military requirements will never arise.

Though the report does state that a circumstance might arise that would lead the United States to withdraw from the CTBT and resume testing, such a statement in effect urges the Senate to ratify a deficient treaty from which, by objective standards, the United States would likely find a reason to withdraw in the future. Such a proposition undermines the international standing and reputation of the United States.

Finally, just the presence of treaty obligations undermines objective military decisions. The CTBT would act as a quiet force driving the military both to avoid finding new requirements for nuclear weapons and to put off recommendations to build new weapons in order to avoid the possibility of treaty withdrawal. Military decisions regarding new weapons requirements and modernization should not be made in a vacuum. Regrettably, under the CTBT, the final decisions would be the result not of an objective process, but of a politically sensitive one.

Reason #3: The Shalikashvili report does not satisfactorily address the problem of aging missile systems.

The bulk of the U.S. land-based missile force consists of 1970s-vintage Minuteman missiles. Their age should be a concern. Aging rockets will become more difficult and expensive to maintain, and this will raise questions about their safety, reliability, and military effectiveness.

The United States in the past would modernize its nuclear missile force by replacing existing missile systems as they aged with new ones. In this process, a warhead would be designed and built specifically for a missile, and the missile specifically for the warhead.

Under the CTBT, however, the process for modernizing missiles at the point that existing missiles must be retired will depend on building a missile to the requirements of existing warheads. This approach will severely constrain or, as a practical matter, bar outright the procurement of more modern missiles and result in an atrophying nuclear deterrent.

The CTBT will make it difficult to replace aging nuclear missile systems. The omission of any discussion of this problem leaves open serious questions about what strategic missile systems the United States would deploy when the currently deployed Minuteman, Peacekeeper, and Trident missiles are retired. Barring the replacement of existing systems as they retire will result in a continuous reduction in the size of the U.S. deterrent force until there are no missiles left. Maintaining a more modern strategic missile force at an established size by building new missiles designed around the warheads in the stockpile would be difficult. Far from ideal, this approach would undermine the overall effectiveness of the U.S. strategic nuclear force.

For the U.S. government, neither alternative is acceptable. The U.S. strategic missile force should be modernized with entirely new missiles and with warheads designed and built for those missiles as necessary, which requires testing.

Reason #4: While the report recognizes problems with the aging warhead stockpile, it proposes no effective remedy.

In the past, old warheads could be replaced with new ones. Curtailing modernization of U.S. nuclear weapons will only exacerbate the problems that accompany an aging stockpile. The fissionable materials at the core of nuclear weapons are highly volatile; their key properties change over time, even to the extent that questions could be raised about whether a warhead will reach critical mass at the time of detonation. Left unanswered, such questions could lead to the withdrawal, or "stand-down," of a particular class of warheads from the active inventory, as has been done in the past. With a smaller strategic arsenal that depends on fewer classes of warheads, a future stand-down could undermine the U.S. deterrent posture.

The problems of an aging weapons stockpile would not be present if the United States continued to conduct nuclear tests. The Shalikashvili report recommends addressing these with the same management reforms it recommends for the SBSS program, such as focusing on near-term problems and adopting multi-year budgeting. The SBSS program faces daunting technical hurdles in detecting and solving its problems, not merely management hurdles, and the proposed management reforms will not solve technical problems with the program. Similarly, the report offers no substantive remedy for the problem of aging warheads.

Reason #5: The report provides only general explanations for how the United States could clear the scientific hurdles and maintain a nuclear deterrent under the CTBT.

The Clinton Administration's program for maintaining the U.S. nuclear deterrent under the CTBT faces serious scientific hurdles. In 1997 testimony before a Senate subcommittee, one nuclear weapons designer portrayed the major goals of the SBSS program as significant scientific achievements, were they to be reached.14 For example, some of the machines being designed to monitor nuclear weapons in the stockpile must create pressure and temperature environments that previously were associated only with nuclear weapons and stellar objects. Failure would not be the result of a lack of effort or management problems; it would result from the fact that science had not advanced far enough to achieve these goals.

The report provides only the most general explanations for how such scientific hurdles could be cleared. General Shalikashvili states in the report that

In my judgment, the challenges facing the [SBSS] Program can be managed, and the safety and reliability of the U.S. nuclear deterrent can be maintained indefinitely, so long as future administrations and congresses provide high standards of accountability and sufficient resources to keep uncertainty at an acceptable level.15

The report had already stated that nothing about the annual SBSS certification process for confirming the safety and reliability of nuclear weapons in the stockpile was "sugar-coated."16 In reality, the report includes neither a detailed list of the scientific hurdles the SBSS program must clear nor any explanation of how those hurdles would be cleared. The report, in other words, sugar-coated the likely success of SBSS by resorting to generalities.

Reason #6: The report offers no detailed assessment of the risks associated with decreasing diversity in the nuclear weapons stockpile.

In 1985, the U.S. nuclear weapons stockpile contained 30 kinds of warheads; today, it contains nine. The discovery of an unexpected problem in one kind of warhead could result in the temporary withdrawal of a large portion of the total warheads from the useable inventory. One way to address that shortcoming would be to develop and manufacture new warheads to diversify the stockpile, but the CTBT permanently bars the development and manufacturing of new warhead designs. The test ban increases the risk of catastrophic failure in the overall nuclear posture by decreasing the stockpile's diversity.

The Shalikashvili report does not address the diversity question. Indeed, it acknowledges that moving forward with the construction of a facility for remanufacturing existing nuclear weapons designs must await the determination of the
composition of the long-term stockpile. The report should have included a specific recommendation for diversifying the stockpile by adding new types of warheads.

Reason #7: The report inadvertently admits that the temptation to "fudge" facts about the state of the nuclear stockpile will be greater under the CTBT.

President Clinton acknowledged in 1998 that his Administration would "fudge" findings of fact relative to the activities of foreign governments in order to avoid the triggering of economic sanctions.17 Senators were concerned in 1999 that the mere presence of the treaty obligation and the Administration's desire to preserve it in the future could result in a similar fudging of facts about the state of the U.S. nuclear stockpile. Problems with weapons in the stockpile could be swept under the rug rather than confronted. Such a policy of ignoring problems would undermine the U.S. nuclear deterrent over time and lead ultimately to functional nuclear disarmament.

The Shalikashvili report inadvertently makes clear that the temptation to fudge facts about the nuclear stockpile will be greater under the CTBT. It does this by arguing that the non-proliferation costs of even ending the current voluntary U.S. moratorium on testing are too high. By recommending that the United States not resume nuclear testing now--something that has been barred by unilateral policy since 1992--it in effect justifies the concern that problems with weapons in the stockpile will be ignored under the CTBT. The reason: The non-proliferation costs of resuming testing following ratification of the CTBT, which would require abrogating the treaty, will be exponentially higher than the costs of ending the current moratorium.

If the Shalikashvili report cannot find justification for resuming testing under current circumstances, the presence of the treaty obligation will all but guarantee that such a justification will never be found following ratification. Problems with nuclear stockpiles will be ignored even more assiduously than the existing problems with the stockpile are ignored in the Shalikashvili report.

Reason #8: The report includes an inadequate proposal for preventing political pressures from undermining the stockpile stewardship program.

Both foreign and domestic governments and policymakers who favor U.S. nuclear disarmament see the CTBT as a vehicle for achieving that outcome. They argue that essential elements of the SBSS program are contrary to the spirit and intent of the CTBT and, in the context of U.S. ratification, violate a treaty obligation.

The report merely revives an earlier Clinton Administration proposal on how to protect the stockpile stewardship program against political pressures. That proposal, presented in 1995, established safeguards that could be used to trigger a decision by the United States to exercise its right to withdraw and resume testing. The Senate in 1999 correctly judged these safeguards to be inadequate to protect the U.S. nuclear deterrent, because the United States has rarely withdrawn from arms control treaties. The Shalikashvili report recommends that the Administration and the Senate agree to an intensive joint review of the treaty's value to national security at 10-year intervals (following its entry into force) as a way to assuage Senate concerns about the effectiveness of safeguards.

The assumption is that the mere presence of this joint review will serve to discipline the SBSS program and the certification process, and allow for a more serious examination of the abrogation option. But it will fail to ease political pressures in favor of nuclear disarmament, particularly if such pressures emanate from inside the Senate and the Administration. Indeed, ratification of the CTBT will increase the legitimacy of the arguments advanced by advocates of U.S. nuclear disarmament and enhance their ability to achieve their goal.

On the one hand, the report states unequivocally that preservation of the U.S. nuclear deterrent is essential.18 On the other hand, its specific proposals either will do nothing effective to preserve the deterrent against the forces that want to undermine it or will actually strengthen those forces. CTBT ratification will decisively change the balance of power in favor of those who want to eliminate the U.S. nuclear deterrent.

The Shalikashvili report compounds its errors by proposing two interim steps, both of which are inimical to U.S. security and legally questionable: (1) continue the U.S. moratorium on testing and (2) take steps to implement provisions of the CTBT prior to ratification.

The Administration should reject these interim steps. Instead, it should:

  • Recognize that the moratorium on testing has lapsed. President George H. W. Bush established the moratorium on nuclear tests in statute in 1992. That law, however, also stated that the moratorium would be lifted if another country conducted a nuclear test. Both India and Pakistan conducted nuclear tests in the spring of 1998. Nevertheless, the Shalikashvili report recommends continuing the moratorium.19 This would turn a policy of mutual restraint into a policy of unilateral restrictions on the United States. The report recommends that the Administration treat the moratorium as if it remained a legal requirement, which it does not.

The Administration should issue a statement that the tests by India and Pakistan negate the intent of the moratorium and that the law has lapsed. It should state that while the United States at present has no need to conduct a test, it retains the right to do so when such a test is necessary to ensure the safety, reliability, and effectiveness of its nuclear stockpile.

  • Refrain from implementing any provisions of the CTBT without Senate consent. The CTBT requires participating states to construct facilities as part of an International Monitoring System (IMS) to detect nuclear explosions anywhere in the world.20 This system is key to the ability to verify treaty compliance on testing. The Shalikashvili report recommends that the U.S. participate in building the IMS, even though the Senate has not ratified the treaty and the United States is not a party to it.21 This recommendation is unacceptable. If the executive branch were allowed to implement the provisions of treaties before their ratification, the Senate's constitutional role in the treaty-making process would be nullified.

The Senate should insist that U.S. activities to construct monitoring facilities similar to those of an IMS be a national effort. The Administration should consider negotiating a separate treaty on international cooperation in constructing an IMS as a confidence-building measure to help stem nuclear proliferation. Such a treaty would be subject to Senate review.

The Comprehensive Test Ban Treaty is incompatible with U.S. security and nuclear non-proliferation goals. This was true in 1999, and it remains the case today because the treaty will result in U.S. nuclear disarmament over time but will not prevent those countries that are willing to accept nuclear weapons designs that are unsafe and not reliable from proliferating.

The Shalikashvili report is misguided because it assumes that the treaty will not result in U.S. disarmament. Senator Jon Kyl (R-AZ), who examined the national security implications of ratification prior to the CTBT's rejection in 1999, has noted that the report "recycles the same flawed arguments that six former Secretaries of Defense and the majority of the Senate rejected [in 1999]."22 The Senate should ignore the report and the calls of arms control advocates to reconsider ratifying the Comprehensive Test Ban Treaty.

Baker Spring is F. M. Kirby Research Fellow in National Security Policy in the Kathryn and Shelby Cullom Davis Institute for International Studies at The Heritage Foundation.


1. See, for example, Coalition to Reduce Nuclear Dangers, "Bush, Senate Should Study Report of General Shalikashvili and Reconsider CTBT, Experts Say," January 5, 2001.

2. General John M. Shalikashvili (USA Ret.), Report on the Findings and Recommendations Concerning the Comprehensive Test Ban Treaty, January 4, 2001, at

3. For a detailed discussion of why the CTBT will lead to U.S. nuclear disarmament, see Baker Spring, "The Comprehensive Test Ban Treaty and U.S. Nuclear Disarmament," Heritage Foundation Backgrounder No. 1330, October 6, 1999.

4. Current Secretary of Defense Donald Rumsfeld, who also served as Secretary of Defense during the Administration of President Gerald R. Ford, was among the six former Defense Secretaries who voiced their opposition to the CTBT in 1999.

5. Assuming that all 100 Senators vote. The Senate voted 51 to 48 to reject the treaty on October 13, 1999.

6. U.S. Arms Control and Disarmament Agency, "The Comprehensive Nuclear Test-Ban Treaty," as posted on ACDA's Web site at on January 19, 1999.

7. Shalikashvili, Report on the Findings and Recommendations, p. 6.

8. In the past, the Department of Energy thought it necessary to achieve reliability goals for nuclear warheads of between 99 percent and 99.5 percent. Neither proponents nor opponents of the CTBT have advocated lowering reliability standards for U.S. nuclear weapons.

9. Shalikashvili, Report on the Findings and Recommendations, p. 6.

10. James Glanz, "Testing the Aging Stockpile in a Test Ban Era," The New York Times, November 28, 2000, p. F1.

11. Ibid.

12. Such re-packaging was undertaken with the B61-7 nuclear bomb, which was redesignated as the B61-Mod 11, to make it capable of meeting a new military requirement for destroying deeply buried bunkers.

13. Shalikashvili, Report on the Findings and Recommendations, p. 6.

14. Robert B. Barker, testimony before the Subcommittee on International Security, Proliferation, and Federal Services, Committee on Governmental Affairs, U.S. Senate, 105th Cong., 1st Sess., October 27, 1997.

15. Shalikashvili, Report on the Findings and Recommendations, p. 16.

16. Ibid., p. 14.

17. Elaine Sciolino, "Clinton Argues for `Flexibility' Over Sanctions," The New York Times, April 28, 1998, p. A1.

18. Shalikashvili, Report on the Findings and Recommendations, p. 3.

19. Ibid., p. 22.

20. "The Comprehensive Nuclear Test Ban Treaty," at, pp. 12-15.

21. Shalikashvili, Report on the Findings and Recommendations, p. 22.

22. Senator Jon Kyl, press release, "Shalikashvili Report Unable to Paper Over CTBT's Deep Flaws," January 5, 2001.


Baker Spring
Baker Spring

Former Kirby Research Fellow in National Security Policy

The Heritage Foundation's President's Club Keynote Address by President Donald Trump

Watch live now