In terms of operational practice as distinct from political rhetoric, institutionalized security policy in the United States is based on two presumptions: that imperial aggression is the principal form of threat and that the countervailing threat of deterrent retaliation is the most decisive method of protection. This formulation was established during the Cold War and has been retained in its aftermath. Because no country can plausibly threaten immediate imperial aggression against U.S. territory, the formulation is now justified as a hedge against the rise of an unnamed peer competitor. For many other countries, however, the United States itself is the most credible potential embodiment of such a threat.
Although most prefer not to dwell on that practical fact, it does make the security assurances first issued by the Carter administration 30 years ago a matter of immediate interest. In the traditional formulation, a reliable deterrent should be based on the capacity to retaliate under all circumstances, not merely on the expressed intention to do so. Capacity displayed in immediate operational capability is considered to be more credible and less subject to deception or rapid change than are expressed intentions. That hard standard is far more readily achieved by the United States than by any other country. Because most countries do not have and cannot reasonably acquire unquestionable nuclear deterrent capability, they are dependent on agreed restraint and forced to rely more heavily on judgments of intention. As a result, the authoritative declarations of intention embodied in security assurances have greater global significance than is usually assumed in U.S. discussion of the topic.
The declarations of greatest significance are those issued in connection with the nuclear Nonproliferation Treaty (NPT). Those countries who agreed not to acquire nuclear weapons under the terms of the treaty had reasonable expectation that they would in return receive protection from the five states allowed to retain nuclear weapons. That basic rule was not included in the treaty text, however, because it was difficult to reconcile with U.S. and Soviet alliance commitments. By the time the NPT was negotiated in the late 1960s, the original principle of the UN Charter that each member state would be protected by all the others had been superseded by Cold War arrangements whereby protection was preferentially provided to formal alliance members. The United States believed at the time that the defense of Western Europe against conventional attack might require initiating the use of nuclear weapons against non-nuclear-weapon state-parties to the treaty that belonged to the opposing Warsaw Pact. U.S. officials considered it imperative to extend the same commitment to Japan and South Korea as well.
Unable to include the expectation of general protection directly in the treaty language, the non-nuclear-weapon states resorted to UN Security Council Resolution 255, passed on June 19, 1968. The resolution “recognizes” that aggression with nuclear weapons against a non-nuclear-weapon state-party to the treaty would create an immediate obligation for action under the UN Charter and “welcomes the intention expressed by certain states” to provide assistance under those circumstances. That language suggests not merely the promise not to attack (negative assurance) but also to provide active defense if someone else did (positive assurance). By virtue of the Security Council vote, the concept was implicitly accepted by all five NPT nuclear-weapon states, although France abstained from the vote, stating that declaratory assurances would be inadequate without corresponding disarmament provisions.
In 1978, eight years after the NPT officially entered into force and 10 years after approval of Resolution 255, the United Nations convened its first special session on disarmament, and the United States used the occasion to issue an authoritative statement of intention. Speaking to the special session on behalf of President Jimmy Carter, Secretary of State Cyrus Vance stated:
The United States will not use nuclear weapons against any non-nuclear- weapon state party to the NPT or any comparable internationally binding commitment not to acquire nuclear explosive devices, except in the case of an attack on the United States, its territories or armed forces, or its allies, by such a state allied to a nuclear weapon state, or associated with a nuclear-weapon state in carrying out or sustaining the attack.
That statement of negative assurance, endorsed by all succeeding administrations and reiterated at the 1995 NPT review conference that indefinitely extended the treaty, has come to be regarded as the standard formulation of U.S. declaratory policy. Its main feature is that the United States reserves the right to initiate the use of nuclear weapons in response to an attack of any sort as long as another nuclear-armed state is somehow implicated in the attack.
That qualified assurance had earlier been given formal legal standing for those states adhering to the nuclear-weapon-free-zone treaties for Latin America, the South Pacific, and Africa. In the Latin American case, the United States signed and subsequently ratified protocols to the treaty of Tlateloco extending negative security assurances to the states-parties to the treaties; but, as part of the ratification process, President Richard Nixon issued a proclamation laying out the qualification that was subsequently issued more generally but less formally in the 1978 statement.
Then, in 1996 with a Soviet invasion of Europe no longer a plausible concern, an aide to President Bill Clinton introduced additional qualifications at a White House press conference announcing that the United States had signed the corresponding protocol to the Treaty of Pelindaba, which sought to establish an African nuclear-weapon-free zone. He stated that attacks by other weapons of mass destruction would be justifying conditions for nuclear retaliation, and that was understood but not directly stated to mean chemical or biological agents. In Senate testimony two weeks earlier, Secretary of Defense William Perry had also implied that the use of nuclear weapons would be considered in responding to a chemical weapons attack. Neither statement addressed the central question as to whether the uniquely destructive power of nuclear weapons could be appropriately applied to substantially lesser forms of “mass destruction.”
In citing the doctrine of belligerent reprisal as justification for the additional qualifications, Clinton administration officials evoked a traditional legal rule holding that the provisions of one treaty would be suspended if those of related one were violated. Hence an attack using chemical or biological agents in violation of the 1925 Geneva protocol by a party to the Pelindaba treaty would invalidate its protection against a nuclear response. They generally did not mention, however, that the doctrine would allow only for a response that could be demonstrated to be necessary and proportionate.
The evident efforts of Nixon, Carter, and Clinton to minimize legal limitations on U.S. deterrent operations all reflect an underlying institutionalized commitment to preserve as much uncertainty as possible in the minds of all conceivable opponents as to how far the deterrent effect of nuclear weapons might be extended. The incremental supplement to deterrence supposedly achieved against a willful aggressor by that deliberate ambiguity outweighed the legitimate interest of the non-nuclear-weapon states in establishing categorical security assurances, an interest on which they have repeatedly said their continued adherence to the treaty ultimately depends.
The U.S. extended deterrence doctrine has been emulated and thus reinforced by France, Russia, and the United Kingdom, all of whom have advanced similar qualifications to the security assurances they have issued. Because the nuclear arsenals of Russia and the United States contain more than 95 percent of the nuclear weapons believed to exist, those two countries together effectively set the global standard. There is, however, a meaningful difference between them. Russia, whose conventional force capabilities are not competitive with those of NATO, has far more reason to rely on the extended deterrent effect of nuclear weapons for its own defense than does the United States. In fact, with the world’s most capable military establishment and least vulnerable national territory, the U.S. doctrine is especially subject to serious moral and practical questions.
Those questions are implicitly posed, moreover, by China’s nuclear weapons policy. China has historically had potentially antagonistic security relationships with Russia and the United States and is entangled in one of the world’s most significant remaining disputes over sovereign jurisdiction: the question of the ultimate status of Taiwan, which could become a trigger for active confrontation. China has nonetheless maintained the most limited deployment of the five official nuclear powers and has proclaimed the most restricted doctrine of use. China has repeatedly issued categorical assurance that it would not to be the first to use or threaten to use nuclear weapons at any time or under any circumstances, and those assurances are reflected in its deployment pattern. The most authoritative information indicates that China has less than 100 operational nuclear delivery systems, less than 25 of which could reach the United States and none of which are ever brought to immediately available alert status. Academic speculation aside, nothing in the historical record suggests that this minimal deployment posture and exclusive doctrine of retaliation has provided inadequate deterrent protection for China, despite the fact that it had the most demanding burden during its formative period and arguably still has.
The extended deterrence doctrine has acquired nearly axiomatic status in U.S. security culture, to an extent that even its most glaring defects are generally ignored. Although few if any individuals seriously believe in the existence of an implacable aggressor held in check only by the threat of massive retaliation and poised to exploit any ambiguity of resolve, the continuation of deterrent operations designed against a hypothetical threat of that sort is considered prudent beyond question. That the resulting operational coupling of large U.S. and Russian alert forces creates the possibility of inadvertent catastrophe is dismissed as a negligible risk despite the directly applicable folk wisdom embodied in Murphy’s law: if something can go wrong, it eventually will.
Even more remarkably, the fact that large dispersed deployments increase the risk that terrorists might seize a weapon or an equivalent amount of explosive material is accepted despite frequently articulated fears of exactly that possibility. Current mainstream discussion appears willing to consider reducing active deployments to the 1,500-warhead level but not below 1,000. That latter figure might begin to impose meaningful constraint on the destructive potential of the U.S. deterrent force, but the former would not. There is no prominently articulated support for emulating China’s categorical assurances even from those who have recently advocated the elimination of nuclear weapons as an ultimate goal.
Despite its impressive hold over the U.S. security establishment, however, the extended deterrence doctrine is not assured indefinite reign. It has been undermined within the security bureaucracy by a radical assault from an intense minority faction promoting decisive military superiority as an ultimate substitute for deterrence. Successfully evoking the authority of President George W. Bush, they have advanced a doctrine of preventive war euphemistically labeled pre-emption. In an address to the West Point graduating class in 2002 and in an ensuing review of nuclear weapons policy, Bush reserved the right to initiate the use of force, including nuclear weapons if necessary, to prevent the acquisition of nuclear weapons by states considered to be inherently hostile to the United States. In his 2002 State of the Union address, Bush named Iran, Iraq, and North Korea as instances; in the perception of the world, he implemented the doctrine in the 2003 invasion of Iraq, unjustifiably as it turned out. Meanwhile, his administration issued a series of military planning documents proclaiming the intention to dominate space for national military advantage and began integrating both long-range precision strike weapons and ballistic missile defense into nuclear weapons operations, all of which projects a serious intention to engage in preventive operations.
In the estimate of attentive military planners in China and Russia, the prevention/pre-emption doctrine backed by a rate of military investment that far exceeds their own is an inherent threat to their own deterrent forces, not merely to the named adversaries. If they come to believe that the doctrine has been established in the United States, they will be forced to counteract it in some fashion. That in turn is a threat to those in the professional core of the U.S. establishment who understand that the doctrine is inherently unrealistic and dangerously provocative. It is the unsustainable political project of an ideological minority whose actions have been discredited in Iraq and whose domination can be expected to diminish within the U.S. political system whatever the outcome of the election.
It is distinctly possible that adverse international reactions to the prevention/pre-emption doctrine will ultimately catalyze an accommodating revision of U.S. security policy that sweeps aside the traditional doctrine of extended deterrence as well. That outcome would be a bitter irony for the advocates of decisive superiority, but the threat of terrorist access to nuclear explosives provides a major incentive for that to occur. Admittedly, any hostile nuclear explosion anywhere in the world might trigger an insensate U.S. political reaction, reinforcing the aspiration for military dominance and giving it longer political life. Such an event would not render the aspiration achievable, however, and belligerent emotion would have to yield eventually to implacable fact.
Fortunately, it is more likely that some breach-of-security incident will command sufficient attention to give standing to the obvious common sense solution, namely, an arrangement whereby all nuclear weapons are removed from active operational deployment to secure storage where they can be continuously monitored and accurately counted. Because they could be reactivated if necessary, that arrangement would preserve all of the deterrent effect plausibly required and would essentially eliminate the inherent danger of inadvertent catastrophe. It would also provide much more reliable protection against terrorist exploitation, especially if all weapons-grade nuclear explosives are subjected to the same monitored storage conditions. Under such an arrangement, conveying reassurance of responsible management would be the principal objective. Preservation of a residual deterrent effect, readily achieved by the existence and operational potential of the stored weapons, would be subordinate to that objective.
The relentless underlying reality is that the process of globalization has rendered massive imperial aggression against any of the major established states virtually infeasible but is subjecting all of them to the disruptive threat of dissident violence emanating from the breakdown of legitimate authority in fragile jurisdictions. Common interest in preventing severe disruption of the global economy gives all major states strong reason to established higher standards of control over nuclear explosives and other menacing technologies. Those common interests are more significant than any residual threat of imperial aggression or coercive projection of influence. The imperatives of the situation require extensive collaboration for mutual protection rather than belligerent competition for national advantage. The ultimate implication is that reassurance replaces deterrence as the central purpose of security policy.
Defenders of the U.S. extended deterrence doctrine will be quick to point out that it plays an important role in conveying reassurance, as indeed it does for formal allies. Under emerging circumstances, however, it will be vitally important to recognize that extended deterrence does not require and cannot tolerate the initial use of nuclear weapons under any circumstance and that its legitimacy depends on global rather than culturally preferential application. To sustain the traditional doctrine in appropriately subordinated form, the United States will not only have to issue categorical negative security assurances, as China has done, but also globally inclusive positive assurances. It will have to indicate it is prepared to protect any country from unprovoked imperial attack and that it will not itself undertake pre-emptive attack without international authorization. That is the only viable way of justifying the degree of military superiority it will retain for quite some time.
The U.S. political system does not concentrate executive authority to the degree that would be required to change operational security doctrine. The president and other authoritative agents can and do make doctrinal pronouncements, but institutionalized implementation depends on consensus created by seminal formative experience and lengthy evolution. The Cold War formulation still largely prevails because the United States has not yet encountered a comparably powerful formative experience. Unfortunately, it is an open question whether institutionalized policy can be appropriately altered in response to fundamental changes of circumstance without the motivating and organizing focus of global war, real in the case of World War II and widely imagined during the ensuing Cold War period.
It is not evident that entanglement in civil conflict and fear of associated terrorism will alone drive a fundamental reformulation. The threat involved is much smaller in scale in any given instance and different in character but not so different as to compel consensual reformulation. Nonetheless, current agonies of policy in Iraq and Afghanistan will clearly be formative experiences of some enduring consequence, and the lessons eventually drawn are likely to feature the importance of global collaboration. It is very unlikely that acceptable stabilization can be achieved in either instance without constructive engagement with Iran and significant assistance from Russia and China in accomplishing the degree of engagement required. That in turn will require fundamental accommodation with all three countries involving categorical negative assurances and substantial positive ones as well. Most immediately, however, that process will probably be burdened by the Russian action in Georgia, which is actually a localized event resulting from the failure of accommodation but is being depicted by defenders of the established doctrine as an ominous indication of broader imperialist impulse.
It is more likely that the consequences of global warming will eventually become imposing enough to force fundamental accommodation and the reformulation of policy necessary to accomplish it. Reliably effective action requires a massive transformation to energy generating technologies that do not emit carbon gases, which is unlikely to be accomplished on sufficient scale without substantial expansion of nuclear power generation. That in turn could not be safely or practically accomplished without new reactor designs, much more secure management of the fuel cycle, and intimate collaboration between the sources of finance and technology—primarily the European Union, Russia, and the United States—and the leading venues of application—China, India, and the developing world generally. Fundamental security accommodation would be an indispensable precondition.
Whatever the degree of accommodation eventually achieved, it is reasonable to expect that negative and positive security assurances will have increasing prominence in the regulation of security relationships. The disparities in military investment and resulting operational capability between the United States and everyone else virtually preclude equitable restriction of military capabilities and therefore make credible restraint on behavior essential. It will be increasingly incumbent on the United States to convey reassurance about the rules and circumstances under which its military establishment operates. It will be incumbent on potential adversaries to accept restrictions on behavior that preclude justification of coercive action. It will be incumbent on all to work out routine documentation of compliance more continuous and more convincing than traditional methods of adversarial verification. That is a predictable trend, however discouraging immediate events and traditionally interpreted history may appear to be.
John Steinbruner is director of the Center for International and Security Studies at Maryland (CISSM) and chairman of the board of directors of the Arms Control Association.
1. “Statement of Secretary of State Vance: U.S. Assurance on Non-Use of Nuclear Weapons, June 12, 1978,” Department of State Bulletin, August 1978, p. 52, ACDA, Documents on Disarmament, v. 1978, p. 384.
2. George Bunn, “The Legal Status of U.S. Negative Security Assurances to Non-Nuclear Weapon States,” Nonproliferation Review, Vol. 4, No. 3 (Spring-Summer 1997), pp. 1-17.
3. A conceptual distinction is often made between extending the deterrent effect from one state to another and extending it to forms of attack other than nuclear weapons. The operational doctrines of France, Russia, the United Kingdom, and the United States all include both forms of extension.
4. Office of the Press Secretary, The White House, “President Bush Delivers Graduation Speech at West Point,” June 1, 2003.
5. Office of the Press Secretary, The White House, “President Delivers State of the Union Address,” January 29, 2002.
6. Nancy Gallagher and John D. Steinbruner, “Reconsidering the Rules for Space Security,” American Academy of Arts and Sciences, 2008, pp. 25-29.