Text Only | Disarmament Diplomacy | Disarmament Documentation | ACRONYM Reports

Disarmament Diplomacy

Issue No. 62, January - February 2002

Opinion & Analysis

Alternative Approaches to Arms Control in a Changing World

By Kerry M. Kartchner and George R. Pitman

I. Introduction: The Appeal of Non-Binding Initiatives

There is growing interest in using approaches other than formal treaties for achieving some US arms control objectives. The lack of progress in negotiating new formal arms control agreements, and in bringing already signed agreements into force, has contributed to a lack of confidence in using formal treaties to achieve key policy goals relating to non-proliferation and strategic arms reductions. In particular: the difficulties, protracted delays, and the current stalemate encountered in trying to bring the START II Treaty into force; the Senate's rejection of the CTBT; the resistance to passing enabling legislation for the CWC; the diverging US and Russian interests in preserving the ABM Treaty; and a desire for greater flexibility in adapting the US strategic forces to the changing international environment have all contributed to this new interest in approaches to arms control that forgo extended negotiations and complicated treaties.

Some believe that formal arms control negotiations or agreements "are simply not needed to get where we need to go."¹ This view is reinforced by the widely held conviction that Russia's nuclear arsenal will continue to decline regardless of whether new formal arms control agreements are reached or not. Some observers hold that some existing treaties are obsolete at best, or, at worst, prohibit developing and fielding those weapon systems most needed to address new and emerging threats. Given the reality of rapid advances in modern weapons technology; the evolving new threats posed by weapons of mass destruction; and the potential for dramatic and increasingly unpredictable international security environment, there is an understandable desire to preserve maximum flexibility in responding to these circumstances. Thus, some believe that it would be unwise for the US to undertake any new formal treaty obligations that would permanently limit or restrict this flexibility.² There are also those who claim that formal arms control agreements have themselves become an obstacle to achieving progress towards the ultimate goal of abolishing nuclear weapons, since they require that states retain otherwise unneeded weapons as bargaining leverage, or for purposes of asserting their rights under a given treaty.³

Finally, the success, albeit qualified, of a previous unilateral initiative - the 1991 US and Russian Presidential Nuclear Initiatives (PNI) under which tactical nuclear weapons were withdrawn from Europe - enhances the attractiveness of such approaches to arms control. All these reasons have been cited for renewed interest in either forgoing formal set-piece negotiations and lengthy complicated treaties in favor of more informal accords, or simply undertaking unilateral steps to reduce force levels.

Yet in spite of these difficulties and setbacks, active bilateral and multilateral cooperation across a wide range of arms control and non-proliferation issues remains vital, and US engagement with other nations on a broad arms control agenda remains a critical component of US global leadership and credibility. Russia retains a large inventory of nuclear weapons and fissile material, and sustaining the process of Russian demilitarisation and defence conversion continues to be an important US priority. In addition to these concerns, the US must also deal with the proliferation of WMD to so-called rogue states who are engaged in developing and deploying capabilities to threaten and deter the US, its allies and friends, and its forces deployed abroad, from actively intervening in regional crises on behalf of US interests.

There are two alternatives to the negotiation of formal, legally binding treaties for achieving US arms control objectives: informal agreements that are politically but not legally binding on their parties; and unilateral initiatives that may or may not be coordinated with other parties but are expected to be reciprocated. When these approaches are considered in the context of the full range of US arms control priorities, it becomes clear that no one approach is necessarily best suited for all arms control objectives. Each entails potential risks and trade-offs that will have to be carefully examined and weighed against the prospective benefits of each approach before the US can match the most effective means with the appropriate arms control ends. This paper defines, discusses, and compares formal and informal approaches to arms control and considers different ways to achieve informal (or legally non-binding) arms control agreements. It also presents some options for limiting or reducing the traditional inflexibility often associated with formal arms control treaties, and discusses their advantages and disadvantages.

II. Defining Formal versus Informal Approaches to Arms Control

The basic distinction between informal and formal methods of agreement is not whether such agreements are written or unwritten, but whether or not the agreement is legally binding on the parties.

Both formal and informal approaches to arms control may involve decisions related to the size, composition, posture, or operational activities of each side's respective military forces. Both involve correlating to one degree or another certain understandings regarding the intentions and actions of the other parties. The most important distinction, however, is that formal arms control agreements are considered under international law to be legally-binding on their parties, while informal accords are considered to be politically binding on their parties. Formal approaches to arms control involve the negotiation and implementation of legally binding treaties between parties. As such, they represent a solemn commitment by each party to comply with the terms of the agreement. In the event of non-compliance by one or more parties to a legally binding agreement, the other parties may have access to various remedies under international law. Such agreements are frequently the result of an intense and protracted negotiation and generally involve participation of the party's legislature in their ratification. The provisions of such agreements often are detailed and explicit, and may contain specific and detailed provisions for verification and monitoring compliance with their substantive provisions and obligations. They also spell out such matters as the specific weapons subject to limitation or elimination, rules of accountability, duration of the agreement, procedures for amending or revising the agreement, and conditions and procedures for withdrawing from the agreement. They may also provide for an implementing body to meet regularly to oversee and discuss the status of the parties' compliance with the terms of the treaty.

It should be noted, however, that there are relatively few provisions under international law for enforcing international agreements and very few treaties contain specific remedies for violations. Nations are expected to abide by the commitments and obligations that they undertake. Remedies to treaty violations may include withdrawal from the treaty by the other parties to it; imposition of sanctions on the offending party; reconstitution of previously reduced military capabilities; military actions; and even war.

Legally Binding Agreements

The US distinguishes between two types of legally binding agreements: treaties and executive agreements. Under international law, the term treaty refers to any international agreement that is legally binding on its parties whether it is called a treaty, an executive agreement, a convention, accord, protocol, pact, or covenant, etc. Additionally, for the US, the term treaty refers to any international agreement that is ratified by the President with the advice and consent of the Senate. Such treaties, together with executive agreements (as discussed below) are legally binding on the US. Treaties are governed by customary international law⁴ as embodied in the Vienna Convention on the Law of Treaties and elsewhere, and by the domestic laws of the state parties to the agreement. Unless the agreement explicitly states otherwise, the parties to it are bound under international law to carry out their obligations as delineated in the agreement.

There are two types of executive agreements: those that the President can ratify on his own accord and those that require approval by a majority of both houses of the Congress. Executive agreements that are pursuant to prior Congressional authorisation or to an existing treaty need not be submitted for Congressional approval. In addition, in determining whether an agreement can be an executive agreement, the Executive Branch, on the recommendation of the State Department, will consider the extent to which the agreement involves the nation as a whole, whether the agreement will affect existing domestic laws, the relevant past practices concerning similar agreements, the preferences of Congress, the degree of formality of the agreement, the agreement's expected duration, and the general international practice with regard to the type of agreement in question.⁵ Under US law and practice, executive agreements are regarded as treaties and are binding on a President's successors. Executive agreements that are ratified by the President without Congressional approval may not prevail against conflicting prior federal statute, but would supersede inconsistent state laws.

The 1972 five-year Interim Agreement on the Limitation of Strategic Offensive Arms is an example of an executive agreement made with the approval of the Congress. When the SALT I negotiations between the US and the Soviet Union commenced in November 1969, the expectation was to reach an agreement on both offensive strategic arms and anti-ballistic missile systems. However, when the negotiations for the ABM Treaty were concluded in May 1972, the US and the Soviet Union had not reached agreement limiting offensive arms and they agreed to freeze the number of strategic offensive missile launchers to those that were already deployed or under construction for a period of five years while the negotiations to limit offensive arms continued. The parties adhered to the agreement until it was superseded in 1991 by the START Treaty. Since the Interim Agreement was not pursuant to any existing legislation or an existing treaty, President Nixon submitted it to both houses of the Congress for approval. Other examples of executive agreements include the 1987 Nuclear Risk Reduction Centers Agreement, the 1988 Ballistic Missile Launch Notification Agreement, and the 1992 Cooperative Threat Reduction (CTR) Arrangements under which the US provided assistance to Russia for the elimination of its nuclear weapons and for the accounting and disposition of the fissile material obtained from those weapons. The latter were not submitted to Congress for their approval since they were pursuant to the Nunn-Lugar legislation to assist Russia in its defence conversion and the elimination of its nuclear weapons and fissile material.

Non-Legally Binding Agreements

Nations can enter into consensual arrangements on whatever terms and with whatever understandings they wish, and the parties' intentions are controlling. Thus, international agreements are not legally binding upon their parties unless they intend them to be.

There are basically two alternatives to the traditional approach to achieving arms control agreements: either to adopt unilateral measures and hope that such steps are subsequently reciprocated; or to seek explicit informal accords that spell out mutual commitments, but which do not impose legally binding obligations on the parties. Unilateral decisions are, of course, easily and quickly enacted, need no prior consultative process, incur no international legal obligations, and can be reversed or modified as circumstances dictate.

On the other hand, there is no guarantee that unilateral initiatives would be reciprocated or accepted by other states. They could engender a perception that the US was acting capriciously, or without regard to the security needs and calculations of its allies. Such initiatives may be seen as deliberate attempts to circumvent legislative advice and consent prerogatives. Some might accuse the US of squandering assets needed for bargaining leverage by giving them away unilaterally. Most importantly, arms reductions that were undertaken unilaterally, outside the context of agreed time schedules and monitored parameters, with no assurances that such initiatives might not be reversed at some point in the future, and without prior coordination or consultation among allies and other affected parties, could paradoxically contribute to increasing unpredictability and instability in the international strategic environment. Some of these risks and disadvantages associated with unilateral initiatives might be avoided or mitigated through informal agreements that incorporate an on-going consultation process, some means of promoting transparency, or which result in coordinated declarations of some type, even if such declarations did not involve legally binding commitments.

Non-binding political agreements have been defined as accords that do not engage the legal responsibilities of the parties and for which non-compliance by a party would not be grounds for legal remedies. When states enter into a non-binding rather than a binding agreement, they may regard their mutual undertaking as only a tentative arrangement and they accept the possibility that each may decide not to comply, without incurring legal obligations. But this does not mean that such an agreement cannot be effective in achieving cooperation between states. Even though the agreement is non-binding, it still enables the parties to spell out what each is expected to do and provides a normative underpinning to support those expectations. It can also be authoritative and controlling on the parties.⁶

Such "informal" approaches to arms control may not involve any explicit agreements among the parties. They may be unilateral actions or may revolve around some form of agreement or coordinated action with another party that relate to each party's respective national security posture. They may be tacit or express, oral or written, and they may be based on declaratory or official policy. They are usually bilateral in nature, but they may also be multilateral, depending upon the subject matter. Such arrangements are generally not considered to be legally binding on the parties and involve few or no provisions for their enforcement.

Informal or non-legally binding arms control arrangements are generally not subject to legislative approval to bring them into force. However, the Congress has severely restricted the use of non-legally binding arrangements to reduce US military forces. In 1993, the Congress amended the Arms Control and Non-Proliferation Act to prevent the President from reducing or limiting "the Armed Forces or armaments of the US in a militarily significant manner, except pursuant to the treaty-making power of the President..., or unless authorised by the enactment of further affirmative legislation".⁷ Furthermore, the Defense Authorization Act of 1993 prohibits the President from reducing US strategic forces below their authorised START levels except if the START II Treaty enters into force. Thus, the President's ability to reduce US forces or armaments by use of informal agreements without Congressional approval is severely limited.

Countries enter into agreements with other countries or with international organisations with the expectation of obtaining a benefit from the agreement. However, like all human cooperative efforts, such arrangements also carry risks, and one of the negotiator's tasks is to find a proper balance between these benefits and risks. For example, it may be possible to quickly arrive at an informal, non-binding agreement for the US and Russia to reduce their strategic nuclear forces without agreeing on provisions to verify the agreement. Such an agreement could be risky because of its lack of verification provisions. On the other hand, negotiating an agreement with extensive verification provisions could involve long, drawn out negotiations and, because of economic pressures, Russia may be forced to reduce its forces below this level before the negotiations are completed, in which case, both parties may lose access to monitoring data they otherwise would have received.

From an international law perspective, informal agreements can be used as a means of limiting or managing the inherent risks of formal agreements in cases where: (a) one or more of the parties desires to maintain a measure of flexibility with respect to its compliance; (b) there are domestic or international obstacles to more formal agreements; or (c) it is anticipated that circumstances might change within the duration of the agreement, as would be the case with a dynamic and unpredictable threat environment.

III. Benefits and Risks of Formal or Legally Binding Agreements

As discussed above, formal, or legally binding, agreements impose legal obligations on the parties to them, and they are to some degree enforceable under international law.⁸ Thus, they provide benefits to their parties that non-legally binding arrangements may not. Some observers believe that Russia still has a propensity towards denial and deception, especially with regards to its most sensitive military and nuclear assets, making rigorous procedures mandatory if the US is to have adequate confidence in its ability to monitor Russian compliance.

There are circumstances in which the US may want to pursue the formal approach to arms control so as to legally bind its negotiating partner to its obligations, e.g., in an agreement with North Korea under which it would give up its ballistic missile programmes and stop exporting missiles and missile technology to other countries in exchange for political recognition and economic aid. In this case, the US commitments may be easily reversible, whereas the threat of legal recourse in event of non-compliance would be a deterrent to non-compliance by North Korea. Agreements that impose asymmetrical obligations on their parties, such as non-proliferation agreements, generally take the form of treaties, in which the rights and obligations of the parties are explicitly spelled out and are legally binding on the parties.

Formal arms control agreements can also provide a degree of reassurance that is important to US friends and allies because they enhance political stability, especially during transitional periods, or periods of political upheaval and realignment. Because of the protracted nature of the NATO policy making process, planning predictability is an important consideration. The NATO allies have repeatedly expressed the importance of preserving the ABM Treaty, or of replacing it with an equally substantive treaty. Because less formal approaches to arms control may be interpreted as evidence of a "less serious" US attitude toward its global security obligations, substituting informal approaches for formal approaches could exacerbate divisions within NATO, roil our friends and allies in Asia, and damage US leadership in other international organisations.

Furthermore, prospective parties to an agreement may require a legally binding format as the price of their participation or cooperation. For example, especially with respect to strategic nuclear weapons, Russia may prefer formal approaches to arms control, and may be less willing to entertain agreements based on informal initiatives. Russia's continuing reliance on nuclear weapons to compensate for its conventional weakness makes it less likely that its political leaders will accede to informal or tacit arms control measures that involve trade-offs in return for self-restraint or unilateral cuts in its nuclear arsenal. Moreover, the declining coverage and reliability of Russia's National Technical Means of verification likely means that Russia is becoming increasingly reliant on the formal on-site verification and information exchange provisions of existing arms control agreements. Russia may be reluctant to sacrifice these benefits for the sake of politically dramatic informal arms control initiatives that do not entail firm commitments of reciprocity in matters of verification and transparency, and which may jeopardize access to such information. Russia may be unlikely to accept adopting new strategic concepts of stability and deterrence without formal, written assurances of reciprocal compliance, and without codified compensation for difficult and potentially risky changes that will be required to adapt its forces to these new constructs. Moreover, Russia may need formal arms control agreements as a symbol of its great power status, and of political parity with the US.

However, if formal agreements could be reached in a timely manner that set the US and Russia on a path toward a new relationship "premised on openness, mutual confidence, and real opportunities for cooperation," as President Bush has called for, and that provide the necessary flexibility to both sides to respond appropriately to changes in the threat environment and to take advantage of new defensive technologies, such agreements would reinforce predictability, provide assurance to other allied nations that their interests were being protected, and serve to enhance the leadership role of the US.

IV. Benefits and Risks of Informal, or Non-Binding, Agreements

The principle advantages that have been alleged for informal, non-binding approaches are that they can be achieved relatively quickly and may be perceived as being relatively simple to implement; they may avoid prolonged and difficult negotiations especially when relations between the negotiating parties are poor; and, since no legally binding obligations are involved, detailed definitions and complicated verification provisions could be replaced with looser transparency measures.

Informal approaches may be desired when there are urgent matters that must be addressed by prompt action that cannot wait for a formal process of negotiation. Such situations may include: the growing weakness of the Russian command and control system; Russia's weak control over its non-deployed nuclear weapons and fissile materials, and the increasing prospects of an accidental or unauthorised launch of Russian nuclear weapons.

Informal approaches may also more easily cope with asymmetrical force structures. According to this argument, formal arms control agreements depend, to some extent, on parity and equality of rights and obligations for their domestic legitimacy, and thus are limited to imposing constraints on comparable or similar systems. Trying to forge agreements encompassing dissimilar or asymmetrical force structures has posed daunting challenges in constructing provisions and limits that are accepted as equal and fair, and often requires difficult trade-offs among alternative negotiating assets. Asymmetries may also involve doctrinal disparities, and, according to this perspective, since both US and Russian nuclear doctrines are currently in flux, it will be much more difficult and time-consuming to fashion and promote formal arms control arrangements that will not limit doctrinal flexibility.

The perceived success of the 1991 Presidential Nuclear Initiatives, whereby Presidents George Bush and Mikhail Gorbachev simultaneously announced that the US and the Soviet Union would both withdraw tactical nuclear weapons from Europe,⁹ has reinforced the belief that informal arms control agreements have great potential to quickly address today's urgent arms control priorities without the need for protracted formal negotiations.

However, such approaches carry their own risks. If the domestic politics of a nation change, it is easier to abrogate an agreement that is not legally binding than one that is legally binding, and violations of a political commitment do not carry the same consequences as violations of legally binding agreements. If the terms of the agreement are vague, each party may interpret its terms differently, and there may be no established channels for resolving such disputes. The legislators of one or more of the parties may object and claim that the use of such measures is an effort to circumvent the formal process of approving agreements.

Informal or non-legally binding arms control arrangements are generally not subject to legislative approval to bring them into force. Because of this the Congress has severely restricted the use of non-legally binding arrangements to reduce US military forces. As was discussed above, the Congress has passed legislation that restricts the President from unilaterally reducing the US strategic forces unless he is authorised by Congress to do so. However, this legislation does not necessarily preclude the US from using informal approaches for achieving its arms control goals. It only means that the President must obtain Congressional approval either prior to or after obtaining an agreement.

V. Managing the Risks of International Agreements, Whether Formal or Informal

As noted above, there is a certain amount of risk associated with any international agreement, whether formal or informal. The primary risk is that circumstances could change over time, leaving a country locked into commitments that are either no longer relevant or that block it from taking steps to respond to the new circumstances. There are several ways of managing the risks of any international agreement, including the following:

Attach reservations to the agreement;
Seek to amend or revise the agreement;
Terminate, withdraw from, or abrogate participation in the agreement;
Restrict the agreement to a political agreement;
Accept only an agreement in principle;
Use parallel policy statements or tacit agreements.

Reservations to International Agreements

If not specifically prohibited by an agreement, states may attach reservations to a treaty that are changes or amendments to the treaty as a condition of its ratification provided that they are not incompatible with the object and purpose of the treaty. A reservation may state the conditions under which a state party may not be bound by the treaty as a whole or by certain of its provisions. Reservations may weaken a treaty, but they may be necessary to obtain a state's ratification. Such reservations must be explicitly or implicitly accepted by the other parties if the treaty is to come into or continue in effect. A party to a treaty may also attach understandings or interpretive reservations to its instrument of ratification stating its own interpretation of certain provisions. During the US Senate's consideration of a treaty, a senator may introduce a reservation, which he knows will not be acceptable to the other parties to the treaty in an attempt to kill the treaty. Such reservations are sometimes called "killer amendments."

By attaching reservations to a treaty, a legislature may inadvertently prevent it from entering into force. When the Russian Duma ratified the START II Treaty in April 2000, it attached a reservation to its resolution of ratification stating the Treaty cannot enter into force until the US ratifies the September 1997 START II Protocol that extends the period for reducing US and Russian nuclear forces under the Treaty until December 31, 2007, and the September 1997 ABM Agreements which would multilateralise the ABM Treaty and establish the parameters for demarcating between ABM interceptors and theater missile defence interceptors. Ratification of these measures requires the advice and consent of the US Senate, but the administration has not yet submitted them to the Senate and is unlikely to do so in the near future.

Amendment and Revisions to Treaties

International agreements generally contain provisions whereby their parties can propose amendments or revisions to the agreement. Usually such revisions must be agreed to by all of the parties and ratified in the same manner as the agreement itself. Thus, for the US, amendments to treaties would generally be ratified by the President with the advice and consent of two-thirds of the Senate. In many cases, the administration has been reluctant to consider amendments because of the requirement for Senate ratification.

However, the parties to an agreement may agree that their executives can agree upon additional measures that would improve the viability and effectiveness of the agreement. For example, the START Treaty establishes the Joint Compliance and Inspection Commission (JCIC) which is authorised to agree upon measures that would improve the viability and effectiveness of the Treaty as it deems desirable without the advice and consent of the Senate. Under the START Treaty, the JCIC is authorised to make changes to the associated documents¹⁰ to the Treaty without the advice and consent of the Senate, but not changes to the main provisions of the Treaty.

The ABM Treaty has been amended once, in 1974, when the US and the Soviet Union agreed to limit each side to one ABM deployment area containing 100 ABM launchers instead of two deployment areas containing 200 launchers as specified in the Treaty. In addition, in 1997 the US and Russia amended the Treaty to include the agreed characteristics of ABM interceptors to intercept long-range strategic ballistic missiles and theater missile defence interceptors.

Termination, Withdrawal, or Abrogation of Treaties

Under extreme circumstances, a nation may find that the benefits it receives from implementing or observing a formal agreement outweigh the costs or risks of complying with it, for example, in the event of a material breach or pattern of non-compliance by another party. In such cases, it may become necessary to terminate, abrogate or (if such an option is provided for in the terms of the agreement) withdraw from such agreements.¹¹ Treaties can be terminated or abrogated in one of several ways.¹² Many treaties contain specific provisions for their termination after a specific period of time or when a specific event occurs. However, some treaties have no provisions for their automatic termination and thus run to perpetuity, unless terminated or abrogated by one or more of the parties to the agreement. Many treaties, e.g. the ABM Treaty, contain provisions for a party to withdraw from it, and may spell out the conditions for executing this option. For example, such provisions may allow a party to withdraw "if extraordinary events related to the subject matter of the treaty have threatened its supreme interests."¹³ Such clauses are called supreme national interest clauses. Generally, a party intending to exercise this option is required to provide the other parties a notification of intent to withdraw a specified time in advance of its actual withdrawal. The ABM Treaty requires a six-month advance notification.

The parties to an agreement may terminate it by either explicit or tacit agreement. Such termination can take the form of a written declaration, whereby the parties agree that the treaty is no longer in effect, or the parties can sign a new agreement that supersedes a given agreement without necessarily mentioning the old agreement in the text of the new. On occasion, treaties have been terminated by tacit agreement among the parties by allowing the treaty to lapse by not continuing to observe its terms. In this case, each of the parties essentially agrees not to object to failure to observe the terms of the agreement on the part of the other parties.

If one party violates a provision of a treaty, another party may be able to declare that the treaty has been abrogated and that the treaty is no longer in force. However, the treaty is not ipso facto considered to be abrogated and is considered valid until the other parties declare that they have abrogated the agreement.

A party may declare a treaty invalid on the grounds that the fundamental conditions on which it is based have changed. This is called the doctrine of rebus sic stantibus, and it has been widely recognised by legal scholars and by the Secretary-General of the United Nations. However, scholars and statesmen cannot agree on when the doctrine can be justifiably invoked. One scholar has observed: "There seems to be no recorded case in which its application has been admitted to by both parties to a controversy."¹⁴

Customary international law recognises that a treaty becomes void if the parties cannot carry out the obligations called for in it. For example, if a country signs mutual defence treaties with two other countries that subsequently go to war with each other, whereupon it becomes impossible for the first country to fulfil its obligations to both of its allies, the treaty becomes void.

If a party to an agreement ceases to exist, and does not merge into a new or an existing state, then there is no doubt that the agreement would be voided. However, if a party to an agreement merges with other states to form a new state, then the new state may be obligated to be bound by the agreements of each of the merged states. Similarly, if a state dissolves into several new states, then each of the new states may be bound by the agreements of the original state. After the Soviet Union broke up into its successor states in 1991, the US and the four successor states which held systems or facilities that were covered by the Treaty, signed agreements which multilateralised the START Treaty, and these states - Russia, the Ukraine, Belarus and Kazakhstan - are now parties to the Treaty. Additionally, an agreement between two states may cease to exist if a state of war between them erupts or is declared.

Whether or not the President alone can abrogate a treaty is an open question. The issue first arose in 1798 during the Napoleonic wars. During its Revolutionary War, the US entered into treaties with France that it found to be entangling and President John Adams asked the Congress to "declare the treaties no longer obligatory on the United States." Congress obliged with a resolution. In 1846, when the US was struggling to define its northern border with Canada, President James Polk asked Congress for authority to withdraw from the Oregon Territory Treaty with Great Britain, and the Congress again obliged with a joint resolution. But, in 1979, President Jimmy Carter abrogated the Defence Treaty between the US and Taiwan as part of the process of normalising relations between the US and the Peoples Republic of China without submitting his action to the Senate. Senator Barry Goldwater brought suit against the President in the Supreme Court alleging that any treaty ratified by the Senate could not be terminated unilaterally by the President. The Court declined to make a decision on the merits of the case, claiming that the issue was a political question inappropriate for judicial resolution.¹⁵

Political Agreements

For an agreement not to be legally binding, it must be made and expressed in a form other than that usually used and recognised in international law as establishing a legally binding contract: by clearly indicating in the agreement itself, or in the negotiating record, that the agreement is to be regarded as non-binding. For example, paragraph 101 of the 1986 document of the Stockholm Conference on Confidence and Security Building Measures and Disarmament in Europe states: "The measures adopted in this document are politically binding" (emphasis added). Such arrangements are sometimes called political agreements and are analogous to a "gentleman's" or "handshake" agreement.

This approach may be used when the parties involved want something in writing, but wish to avoid a legal obligation. This may be out of the desire to retain flexibility, out of a lack of trust, or as a means of circumventing a prospective unfriendly or protracted legislative or international approval process. Examples of such an approach include declarations, resolutions, and joint statements. Such documents, however, may establish strong precedents and may promote the expectation of more formal commitment at a later date. Because of their written form, they impose a higher degree of political and moral obligation on the parties than do tacit agreements.

Another example of a political or legally non-binding agreement is the 1975 Helsinki Final Act, which was intended to establish a broad framework for East-West cooperation. One paragraph states that the parties do not regard the document as a treaty or international agreement eligible for registration under Article 102 of the United Nations Charter. Moreover, during the negotiations, the US and other western nations clearly expressed their understanding that the Final Act was not intended to involve a legal commitment or to be binding upon the signatory powers. Nevertheless, nations have treated the Final Act as embodying important political and moral commitments, and western nations have frequently accused the Soviet Union of violating the Act's human rights provisions. Louis Henkin notes that although the Final Act is non-binding and without legal remedies, it is an international undertaking and violations of it are a "proper basis for international (non-legal) recourse and remedy by the other participants."¹⁶

If a legally non-binding arrangement is broadly accepted and referred to by nations as if it were normative, it may gradually become as obligatory and legally binding as customary international law. For example, while the Universal Declaration of Human Rights, adopted as a resolution by the United Nations in 1948, was clearly intended to be non-binding, over time it has taken on a normative character. In recent years, nations have been accused of violating the standards established by the Declaration. Some scholars contend that at least some of its provisions have become, through wide recognition and acceptance, legally binding as customary international law.¹⁷

In other cases even an implicit agreement can become legally binding as in the case of the right of passage over Indian territory between Portugal and India. In this case, Portugal had established a longstanding practice of India's non-interference with the passage of private persons, officials and goods over Indian territory between India's coasts and Portuguese enclaves within Indian territory. In its 1960 decision, the International Court of Justice (ICJ) ruled that the longstanding practice gave rise to a legal obligation on the part of India not to interfere with such passage.¹⁸

Agreements in Principle

A nation can avoid incurring the immediate risk of a proposed cooperative arrangement by entering into a preliminary type of general understanding which, while it may be legally binding, is devoid of any specific substantive commitments. Agreements to agree are of this character. Nations frequently undertake commitments to negotiate towards solving a common problem, without committing themselves to specific terms, or even to ultimately reach a final agreement. The only obligation that they assume is to negotiate in good faith. Although the agreement does not contain a substantive obligation, an "agreement to agree" can serve a useful purpose. In situations where countries are not yet prepared to commit themselves to an agreement, an "agreement to agree" can at least commit them to the principle of cooperation in the area in question.¹⁹ An informal agreement can be a prelude to a formal agreement, serving as a trial period prior to accepting legally binding commitments. It may serve as an interim accord pending the resolution of detailed procedures to be used in a more formal contest. An example is the 1997 Helsinki Joint Statement by Presidents Clinton and Yeltsin on Parameters on Future Reductions in Nuclear Forces. Under this agreement the two Presidents committed their countries to commence negotiations on a START III agreement immediately after START II enters into force.²⁰

Parallel Policy Statements or Tacit Agreements

One way to avoid becoming prematurely bound by an explicit agreement is for two states to make simultaneous parallel statements of policy. Under this approach, the parties engage in reciprocal, but not necessarily coordinated, behaviour. There is no formal agreement between the parties, and the only written records are the unilateral but parallel statements of the parties. Such approaches may be useful when internal or external constraints make express agreements impossible or impractical. Since they are not legally binding, it is understood that, should circumstances change, the parties are free to alter their respective policies concerning the activities that were subject to the tacit agreement without incurring legal repercussions. (Of course, in practice, other parties are always free to unilaterally retaliate for what may be perceived as reneging on a political commitment.) Among the drawbacks of this approach are that it is uncertain and unreliable. It can apply effectively only to relatively simple types of behaviour or weapon systems and cannot be expected to deal effectively with problems of a complex or enduring nature.

An example of this type of agreement is the 1991 Presidential Nuclear Initiatives under which Presidents George Bush and Mikhail Gorbachev simultaneously announced that the US and the Soviet Union would both withdraw tactical nuclear weapons from Europe, including Germany and Poland. There was no explicit agreement to remove these weapons, and there is no documentation except for the public statements of the two Presidents. The US and Russian actions differed: the US retained some aircraft delivered nuclear bombs in Europe, but withdrew all of its nuclear weapons from South Korea, and removed its nuclear weapons from all US surface naval vessels. Russia withdrew all of its tactical nuclear weapons from Central Europe, but not from its own European territory.

But even policy statements and tacit agreements can become legally binding. In 1974, the ICJ ruled that the unilateral statements by the French Government that France intended to cease conducting atmospheric nuclear tests in the South Pacific could have a legally binding effect. One authority on international law has commented: "The Nuclear Tests cases may have provided a basis for thinking that there are many forms of international commitments that are not treaties in the technical sense of the Vienna Convention, and it may very well be that a customary law of treaties, embracing a much wider range of international agreements than under the Convention, continues to exist and to give binding force to international undertakings of one sort or another."²¹ Furthermore, in 1976 the Legal Advisor of the US State Department took the position that parallel unilateral undertakings by two or more states may constitute as international agreement.

The principle advantage of such statements of policy is that they can be easily made and quickly implemented. Another example was when the US and the Soviet Union both announced that they would continue to abide by the SALT I Interim Agreement after it expired in 1977, while, nevertheless, indicating that no formal agreement would be in effect.

VI. Building Flexibility into Formal Agreements

One method of limiting the risks associated with arms control agreements deserves separate treatment. Formal arms control treaties have been criticised for their lack of ability to respond to changing circumstances. However, nearly all treaties contain provisions for their amendment and some provide their implementing bodies with authority to agree upon measures or revision to improve their viability and effectiveness. Nevertheless, both the amendment and revision process suffer from the fact that the parties must reach consensus on such changes, and they usually require legislative approval.

There are circumstances in which the US may find that formally recording legally binding arms control solutions best serves the national interest. In the case of strategic offensive arms reductions, US domestic legislation prevents the use of non-binding arrangements for militarily significant reductions in US military forces unless authorised by the Congress. Where it becomes necessary to accommodate such requirements, yet retain some protection against the risks by preserving the freedom to respond to sudden changes in the threat environment, or to exploit new more cost-effective technologies, there are several ways to build flexibility into formal agreements. These include the following:

Incorporate flexible terms into the agreement that provide for a range of weapon limits, rather than a single, inflexible ceiling, such as provided for in the START II Treaty for the US and Russia to reduce their respective strategic forces to between 2,000 and 2,500 accountable weapons.
Limit the duration of the agreement. Provide for certain conditions to be met in order for the terms of an agreement to be extended beyond a fixed date. For example, the START Treaty will expire fifteen years after its entry into force unless the parties agree to extend it. The agreement to extend the treaty could include revisions to the treaty, such as reductions in its various ceilings, and other changes that respond to conditions that exist at that time.
Confer greater authority to alter, revise, or suspend provisions of an agreement to the implementing body. Two methods for doing this were incorporated into the original START Treaty: One was to qualify certain provisions of the treaty, including both specific limits and other operational obligations, with the clause "unless otherwise agreed." This formulation is often used in cases where the future needs of the parties could not be confidently predicted by the negotiators, as in the case of limits on space launch facilities. The other means was to note in the text of the treaty that, with respect to certain procedures, limits, or other obligations, "the parties shall agree" on specific terms at a later date.²² This deferred the issue, but was necessary in cases where agreement on complex procedures could not be completed by the time the treaty was signed.
Expand or elaborate on the reasons or causes that would justify termination, suspension, or withdrawal.
If there is no implementation body, mandate frequent reviews, such as the five-year review requirements built into the ABM Treaty, or that were formerly required to extend the Non-Proliferation Treaty.

Other means exist to expedite negotiations while still achieving formal outcomes, such as executive agreements, or attaching reservations or interpretations to agreements. It may also be useful to develop ways to more effectively employ established rights to amend or revise existing treaties.

VII. Conclusion

In the foregoing we have explored the advantages and disadvantages of several approaches to arms control agreements, including formal treaties that are legally binding on their state parties, non-binding political agreements, and parallel or reciprocated initiatives. Each of these approaches has its advantages and has been used by the US to achieve its arms control objectives. Formal treaties are likely to be used in situations where the parties desire to codify long-term obligations, undertake significant limitations of their military forces and their operations, establish rigorous verification regimes, and are confident that the prevailing international environment will not radically change during the course of the agreement. They may also be used in situations in which the parties assume significant asymmetrical obligations, e.g., non-proliferation agreements.

Political agreements that are not legally binding are useful in situations in which the parties desire only a tentative arrangement and are willing to accept the possibility of non-compliance by the other party, or in situations in which the parties wish to provide themselves with flexibility to respond to changing conditions. Such agreements can still be effective in achieving certain goals. The parties can use such accords to enumerate what each is expected to do, and in some cases they can be authoritative and controlling on the parties. Such agreements can be useful between trusting parties and are analogous in status to a gentleman's agreement.

Parallel or reciprocated actions provide maximum flexibility and can be useful in promoting or encouraging reciprocal actions. In some cases unilateral actions may be coordinated and in other cases there may be only an expectation of reciprocity.

We have also explored several ways to build flexibility into formal agreements. It is possible that the US and its prospective arms control partners in the future will find that hybrid approaches combining some elements of formal agreement with some informal arrangements will best meet their respective national security objectives. To retain its effectiveness and legitimacy, contemporary arms control diplomacy must consider the fullest possible range of alternative approaches.

Notes and References

Stansfield Turner, "Timeout on Arms Treaties," Christian Science Monitor, March 6, 2001.
This theme is thoroughly addressed in Keith B. Payne, Rationale and Requirements for US Nuclear Forces and Arms Control, Vol. I: Executive Report, (Fairfax, VA: National Institute for Public Policy, January 2001), pp. 10-15.
See Jonathan Schell "The Folly of Arms Control," Foreign Affairs, vol. 79, no. 9 (September/October 2000), pp. 22-46.
Customary international law refers to legal customs that have developed in international relations. The failure to observe such customs may entail punishment, sanction or retaliation. See von Glahn, Law Among Nations: An Introduction to Public International Law, Macmillan Co. New York, 1965, pp. 15-19, note 8.
State Department Circular No. 175.
Richard B. Bilder, Managing the Risks of International Agreement, Univ. of Wisconsin Press, Madison, 1981, p. 24-26.
"Arms Control and Nonproliferation Act of 1993," 103rd Congress, S 1182.
Much of the material in this section is drawn from the following sources: Charles S. Rhyne, International Law: The Substance, Process and Procedures and Institutions For World Peace with Justice, CLB Publishers, Inc., Washington, 1971; Gerhard von Glahn, Law Among Nations: An Introduction to Public International Law, Macmillan Co. New York, 1965; Thomas Buergenthal and Harold G. Maier, Public International Law in a Nutshell (2nd ed.) West Publishing Co, St. Paul (1990).
The US announced that it would withdraw its nuclear weapons from South Korea and remove them from its naval surface vessels, but would retain 400 air-delivered nuclear bombs in Europe.
The associated documents to an Agreement are those documents - annexes, protocol, memoranda, agreed statements, etc. - that are considered to be integral parts of the agreement.
The American Heritage Dictionary defines "abrogate" as "to abolish, do away with, or annul, especially by authority." Abrogation of a treaty should not be confused with withdrawal from a treaty. Withdrawal from a treaty involves exercising a right pursuant to, and in accordance with provisions written into a treaty, whereas abrogation refers simply to ceasing to observe a treaty's obligations and need not involve a specified notification period.
For a thorough legal and historical discussion, see Arie E. David, The Strategy of Treaty Termination: Lawful Breaches and Retaliations, Yale University Press, New Haven. 1975.
START Treaty, Article XVII, paragraph 3.
Quoted in von Glahn, loc. cit., note 8, p. 443, Footnote no. 5.
"Treaties Don't Belong to Presidents Alone," New York Times, Aug. 29, 2001.
Quoted in Bilder, loc. cit., note 4, p.33.
Ibid., p. 31f.
Ibid., p. 27.
Ibid., p. 34.
"Joint Statement on Parameters on Future Reductions in Nuclear Forces," The White House, Washington, March 21, 1997.
Bilder, loc. cit. note 4, p.29.
The original START Treaty contained seven "shall agree" clauses, mainly related to procedures for using satellite system receivers to determine fixed locations, and procedures for employing radiation detection equipment. All seven clauses have since been fulfilled, or superseded, through agreements negotiated in the Joint Compliance and Inspection Commission, the START Treaty's implementing body.

Kerry M. Kartchner is Senior Representative of the US Department of State to the Standing Consultative Commission (SCC) of the Anti-Ballistic Missile (ABM) Treaty. George R. Pitman is retired from the US Arms Control and Disarmament Agency (ACDA) and currently consults for the US Department of State. The views and analysis expressed in this paper are those of the authors alone and should not be construed to represent those of the US Department of State or any other agency of the US government.