If arms control agreements are to contribute to the security of the United States and to international stability, they must be adequately verifiable. Verification-the attempt to determine whether the other parties to an arms control agreement are complying with its obligations-is a critical element of the arms control policy of the United States.

The verification of arms control agreements has several purposes. First, verification serves to detect violations of an agreement, or to provide evidence that violations may be occurring. Second, by increasing the risk of detection, it helps to deter violations. Finally, by providing evidence that an agreement is in fact being observed, it enhances domestic and international confidence in that agreement, contributes to mutual trust among the parties, and creates an atmosphere conducive to further progress in arms control.

Verification depends to a considerable extent on sophisticated techniques of intelligence collection, but there are important differences in the objectives of verification and intelligence. While arms-related intelligence seeks to determine the numbers, characteristics and activities of an opponent's military forces, verification attempts to prove a negative that certain force levels are not being exceeded, that certain activities are not taking place. For the purposes of verification, then, it is necessary to pay attention not only to military deployment and testing areas normally used by the other parties, but also to areas which might be so used. For verification must begin with the possibility that violations of an agreement may occur-and assume that a concerted attempt will be made to conceal them.

If standards for verification are set too low, violations may be invited, and the constraints set by an agreement may become a liability for the side that observes them. If verification standards are too high, however, agreement may be impossible, and the benefits of a treaty lost. Methods that are capable of monitoring treaty compliance under most imaginable circumstances might be unable to detect in time an elaborate, costly, and ingenious program of concealed violations. Therefore almost every arms control agreement involves some risk and uncertainty. But if violations can be made sufficiently difficult, an

agreement may still be judged adequately verifiable, and hence compatible with national security.

It is important to stress that verifiability is not a matter of black and white. Almost no agreement that proposes to limit modern weapons can be verified with total certainty through present or currently conceivable techniques of verification. Verifiability is a matter of judgment, and requires a balancing of considerations of different kinds.

To estimate the verifiability of a given agreement, it is necessary to assess both the capacity of U.S. detection systems and the ability of the other party to evade detection. An agreement might be difficult to monitor with the techniques at our disposal; but if evasion would involve high risks or heavy expenditures with little prospect of a commensurate advantage, such an agreement could well be considered adequately verifiable. Conversely, an agreement limiting certain kinds of arms might be relatively easy to monitor given the practices and operating procedures currently used by the other party; yet if those procedures could be easily altered (or substitutes devised) in ways that could greatly facilitate covert violations, the agreement might have to be considered inadequately verifiable.

Assessments of verifiability must take into account, in the first place, the character of the restrictions to be imposed. Bans, for example, are in general more easily verified than numerical limits, and bans on testing or deployment are easier to verify than bans on research. Of equal importance are the characteristics of the weapons or forces to be constrained. Objects that are large and stationary (for example, missile silos) are easier to count and keep track of than objects that are small and mobile (for example, soldiers); limitations on discrete or countable objects are, as a rule, easier to verify than limitations on qualitative changes in technology. Thus the SALT I Interim Agreement, which imposed numerical limits on objects relatively easy to count, poses fewer difficulties for verification than an agreement on reduction of forces in Europe, or an agreement involving qualitative limits on nuclear warheads.

It is one thing to determine to what extent an arms control agreement is verifiable; it is quite another to decide whether its verifiability is adequate to safe

guard our security interests. The latter requires a political rather than a technical judgment. A critical aspect of the verification process is insuring that the technical facts of verifiability are properly conveyed to those who must assess them in the light of the political questions which any arms control agreement must raise. Much will depend on an assessment of the past record of the other parties and of the current state of our relations with them. Much will depend on the risk posed by possible violations and on our ability effectively to counter them. Much will depend as well on our own foreign policy choices. Some lessened degree of verifiability may be acceptable if the political benefits of a treaty are judged to be sufficiently important.

Verification by "national technical means"-by the employment of modern techniques of intelligence gathering which do not require agreed access to the territory of the parties being monitored-is the main approach currently used. The SALT I agreements were the first arms control agreements to make explicit provision for verification by national technical means, and to forbid interference with such verification methods as well as the use of deliberate concealment measures designed to impede their operation. Although technical verification is subject to definite limits (the type of warhead contained in a missile, for example, cannot be distinquished except through observation of test firings), a continuing effort to improve such methods may permit us to expand the scope of arms control agreements in the future.

It is, of course, equally important that we maintain our present capabilities. Many of the technical systems that contribute importantly to the monitoring of current agreements (and that could prove valuable in verifying future agreements in the areas of strategic arms and nuclear testing) are located in other countries; it is extremely important to insure their continued availability. We must also take the necessary steps to protect the secrecy of our intelligence techniques and procedures. If the detailed characteristics of our verification capabilities remain uncertain, violations are more likely to be detected; if they become known to our adversary, he is provided with a blueprint for violating an agreement with little risk. of detection.

Arms control requires effective intelligence, not only for verifying agreements in force, but also for assessing the intentions and capabilities of nations which are parties to agreements. The ability to make such assessments in an accurate and timely manner is essential if arms control policies are to harmo

nize with changing security requirements. Intelligence on intentions and trends is particularly critical as it bears on developments affecting the prospects for nuclear proliferation.

Provisions for access to the territory of, or to facilities controlled by, the other parties to an agreement also play a role in verification. Onsite inspection is a feature of the 1959 treaty banning military activity in Antarctica, and inspections of stations operated on that continent by the Soviet Union and other signatory nations are carried out on a regular basis by U.S. observer teams. The most recent inspection, in January 1975, involved aerial as well as onsite observation of Soviet, British, Chilean, and Argentinean facilities and revealed no evidence of violations of the treaty.

Future progress in some areas of arms control may well depend on a greater readiness on the part of the Soviet Union and its allies to consider arrangements of this kind. In evaluating the role of onsite inspection and related measures in future agreements, however, it is important to distinguish between the symbolic or political value of such measures and their actual worth for verification. In most cases they will act primarily as a supplement to national technical means.

A third category of verification methods-one that is neither wholly unilateral nor wholly cooperative involves the exchange and analysis of data concerning an agreement. A protocol to the Threshold Test Ban Treaty of 1974 provides for the exchange of geological and other data to assist verification by national technical means. A similar verification method currently being explored is the analysis of economic data (whether derived from open sources or directly supplied under a future agreement) that bear on military budgeting practices. Such methods are an important step toward active international cooperation in the verification of arms control agreements. Though not wholly reliable taken by themselves, these methods could act as a useful check on information obtained through other verification channels.

Certain kinds of arms control agreements are best verified by an international monitoring system. The most important agreement of this kind concluded to date is the Nonproliferation Treaty. The fundamental interest of all nations in the peaceful use of nuclear energy and in preventing further proliferation of nuclear weapons points to the particular importance of an effective system of verification for this treatyas well as one that can deal on an impartial basis with the sensitive issues it involves. The Interna

tional Atomic Energy Agency, the organization responsible for administering the safeguards required under the treaty, may serve as a model for analogous verification arrangements in multilateral agreements of the future.

Deciding whether a violation of an agreement has actually occurred is perhaps the most critical step in the verification process. It can also be a difficult one. Ambiquities may arise in connection with the interpretation of the evidence available or the meaning of the agreement itself-or both. This is a particular hazard in the case of agreements dealing with matters as complex as strategic arms. It is important that both sides be aware of the importance of avoiding actions which can be reasonably interpreted as noncompliance-actions which have the appearance of crowding or exceeding the limits of an agreement or which strain its interpretation. Such actions do not contribute to building the confidence essential for meaningful progress in arms control in the future.

Verification is a central feature of our arms control policy today; it will become increasingly important in the future as the United States and other nations seek to regulate larger numbers and greater varieties of weapons. As we begin to look beyond the immediate task of limiting arms competition and contemplate an eventual reduction of military forces, we must remember that the risk that treaty requirements will be evaded may increase as the advantages of evasion become greater. At the same time, the verification of agreements is likely to become more difficult. The rapid development of new technologies is shifting the focus of arms control from numbers to quality, and much will depend on our ability to monitor controls on technological improvements in weaponry. Marked improvement-perhaps bold innovation-in our verification methods may be needed if we are to achieve further progress. And yet we should not allow uncertainties or possible risks to deter us from pursuing agreements which serve the national interest. Every assessment of

CONFERENCE OF THE COMMITTEE ON DISARMAMENT (CCD)

The Geneva-based Conference of the Committee on Disarmament held two sessions during 1975 that proved to be among the most active in the 13-year history of this principal multilateral forum for arms control negotiations. Its 31 member-nations represent a geographic and political cross section of the world. The representatives of the United States and the Soviet Union serve as Co-Chairmen. Ambassador Joseph Martin, Jr. led the U.S. Delegation.

The Committee's agenda included several new issues that had been referred to it by the 1974 U.N. General Assembly-the question of nuclear weapon-free zones, arms control implications of nuclear explosions for peaceful purposes, and environmental warfare-in addition to the topics of chemical weapons restraints and a comprehensive nuclear test ban that have been on the agenda for several years.

Environmental Warfare

The highlight of the sessions was the August 21 tabling by the U.S. and Soviet representatives of identical texts of a draft "Convention on the Prohibition of Military or any other Hostile Use of Environmental Modification Techniques."

In 1974 the U.S. Government undertook an indepth study of the military aspects of weather and other environmental modification techniques. An interagency review of this study led to the Administration's decision to seek agreement with the Soviet Union to explore the possibilities of an international convention—a decision consistent with recommendations made by the Congress.

During the Summit meeting in Moscow in 1974, the heads of the two nations formally agreed to hold bilateral talks to seek the most effective means of overcoming the dangers of the use of environmental techniques for military purposes. Three sessions were held in 1974 and 1975, resulting in agreement on the language of the draft tabled in the CCD.

Basically, the convention would commit its parties not to engage in military or any other hostile use of environmental modification techniques having wide

spread, long-lasting, or severe effects as a means of destruction, damage, or injury to any other party to the convention, or to assist any other nation to do so. At present, use of environmental modification techniques for hostile purposes does not play an important role in military planning. (Dispersal of fog at airfields under one's own control is the only technique which appears practical at present, and this is not considered to be in the category of environmental warfare.) However, in the future environmental modification techniques might become a new arena of military competition. Changes in weather or climate patterns, ocean currents, the state of the ozone layer or ionosphere, or an upset in the ecological balance of a region are some of the effects which might result from the use of such techniques.

The draft convention does not restrict environmental modification techniques for peaceful purposes, such as altering weather to reduce the destructive effects of hurricanes, floods, or droughts. The U.S. research effort in this area received total Federal funding in fiscal year 1976 of just over $18 million.

The draft convention does not prohibit research on or development of environmental modification techniques, because of the difficulty of determining whether these activities have peaceful or military purposes.

Nuclear Explosions for
Peaceful Purposes (PNEs)

Acting on a request made by the General Assembly during the 1974 session of the United Nations, the CCD convened informal meetings of experts in July to study the arms control implications of PNES. The practice of holding such informal meetings of experts has proved successful over the past several years in aiding the CCD in its consideration of technical matters. An ACDA officer was a member of the U.S. group.

The experts focused on two basic questions. The first was whether it is possible for a state to develop a nuclear explosive capability that could not have military application. There was general agreement that it would not be possible to acquire the capability