Document created: 18 June 04
Air University Review, January-February 1970

Launch on Warning

A Counter to the Arms Race

Robert I. Widder

 The United States has for some time followed a second-strike assured-destruction strategy in planning and developing its strategic military forces vis-à-vis the Soviet Union. This strategy provides the U.S. with a deterrent capable of surviving a Soviet pre-emptive counterforce attack on the U.S. strategic offensive force (SOF) in sufficient strength to be able to respond with a countervalue attack on Soviet targets that would result in unacceptable damage to the Soviet Union. The SOF structure was designed to provide such a large number (1000) of hardened Minuteman silos that it became virtually impossible for a Soviet pre-emptive attack on them to so degrade the U.S. response as to make the resulting damage acceptable to the U.S.S. R.

The introduction of multiple independent re-entry vehicle (MIRV) payloads on Soviet missiles may present problems to the survivability of the U.S. SOF. It may become possible that a sufficiently massive pre-emptive attack could be mounted against the Minuteman force to damage it too severely to enable the force to retaliate effectively. Whether or not such a situation would ever exist, it certainly is a possible threat which the United States must consider in its force planning. A U.S. force equipped with MIRV would require fewer surviving missiles to inflict the necessary retaliatory damage. Still, a multiple-payload missile attack could be so heavy as to reduce even such a MIRV force below what is considered a necessary survival level.

Several responses are available to the U.S. should it be felt that a pre-emptive counterforce attack would destroy too great a portion of the SOF, including:

1. Proliferation of the SOF in order to have a greater number of them survive, taking the form of additional fixed or mobile land-based missiles, or additional sea-based missiles, carried in either submarines or surface ships; or an increase in the payload of individual missiles.

2. Further hardening of missile silos.

3. Terminal antiballistic missile (ABM) defenses for the land-based missiles.

4. Combinations of the above procedures.

These possible U.S. responses could have serious undesirable effects. They are likely to trigger increases in the level of Soviet forces, thus initiating another round in the arms race. Furthermore, they will not remove the temptation for a pre-emptive first strike; they will merely increase the destructiveness needed for such a strike to succeed. Although economic constraints will eventually put a ceiling on the nuclear armaments each side will deploy, it would nevertheless be worthwhile to examine the alternative approaches for maintaining a credible United States deterrent through an assured-destruction capability.

Clearly, it becomes desirable to maintain the U.S. assured-destruction capability by other means, means which do not proliferate U.S. forces and thus do not generate Soviet responses that would further proliferate their forces but which would reduce the attractiveness of a pre-emptive counterforce first strike on the U.S. One method to achieve these objectives is to adopt publicly a policy of launch on warning1 as a U.S. response option.

A policy of launch on warning can be fraught with danger unless it is circumscribed by a set of rules and safeguards which clearly indicate that it is not simply a pseudonym for pre-emption by the U.S. and thus a destabilizing policy. Within this broad framework, let us examine the rationale for such a policy.

launch on warning

Simply stated, a policy of launch on warning means that when there is incontrovertible evidence of an attack under way which is sufficiently massive to jeopardize the U.S. second-strike assured-destruction capability, the SOF will be launched prior to the arrival of this attack. While such a policy requires knowledge of the characteristics of an attack, as well as the perfection of command and control procedures, these requirements are believed to be achievable. By eliminating the attractiveness of pre-emptive attacks, the policy should tend to stabilize the strategic balance, especially since its bilateral adoption would serve as an effective deterrent. Under such conditions, the impetus for proliferating offensive forces would be removed, and the mutual reduction in offensive forces might become more acceptable.

Objections to a policy of launch on warning (even when all the required safeguards are attached to it) may be made on various grounds, including moral, legal, political, strategic, and technical, with considerable overlap among them. It is believed that most of these can be overcome and that the remainder can, on balance, be tolerated.

The launching of missiles against another country is certainly a hostile act, in particular when the number of missiles launched is sufficiently massive to seriously threaten the attacked country’s force structure. In more specific terms, in case of a pre-emptive counterforce attack on the U.S. SOF the number of threatening objects required to jeopardize the U.S. second-strike assured-destruction capability is clearly going to be large—at least in the hundreds, more likely in the thousands. The collateral civilian casualties resulting from such an attack have been estimated in terms of tens of millions in the absence of fallout shelters. It seems specious to argue that the U.S. should absorb such an attack before retaliating. There are legal precedents for retaliation after the initiation of an attack, without waiting for the effect to be absorbed.

The collateral civilian damage that would accompany a massive Soviet pre-emptive counterforce attack on the U.S. would be sufficiently high to set in motion countervalue retaliation, as has been stated on numerous occasions by responsible U.S. officials. The whole concept of deterrence rests on this foundation. It would be absurd in such a case to await the outcome of the first strike on the U.S. before counterattacking.2 What is important is to maintain deterrence by adopting policies which will remove all temptation for pre-emptive counterforce attack on the U.S. Launch on warning would emphasize the futility of such an attack, regardless of its size.

The political difficulties of adopting a policy of launch on warning seem to center on the potential hostile interpretations that could be placed on it. Propaganda attacks on this policy are to be expected, once it is publicly adopted. These attacks would likely say that the policy is a cover for the U.S. advocating something akin to a pre-emptive policy of its own or that it means that the U.S. is adopting a spasmodic response policy. The first charge can be readily refuted for two reasons: it is quite likely that the U.S.S.R. would respond in kind to a U.S. pre-emptive counterforce attack;3 and should the U.S. seriously contemplate pre-emption as a viable policy, adoption of launch on warning would have no effect on it.

The question of spasmodic response is more difficult to explain, since the whole concept of an assured-destruction second strike is basically of this kind. The whole problem becomes one of detecting the impending attack, identifying the attacker, and assessing the magnitude of the attack, to determine whether it threatens the assured-destruction capability, before ordering a launch. Thus, the solution of the problem is a matter of technology, which must also have provided the necessary command and control system for timely implementation of the decision.

Obviously, extreme reliability in all these procedures is required. It is believed that the presently available detection and identification means are sufficient to deal with a potential Soviet threat. As yet, there are no meaningful threats to the survival of U.S. second-strike capabilities from other countries. It is reasonable to expect that, by the time such threats could develop, improvements in technology will enable the positive identification of any attacker. Thus the possibility of a misdirected attack on warning would be no greater than if the counterattack were launched after arrival of the pre-emptive attack on the U.S. It is obvious, however, that as the number of countries capable of meaningful attack on the U.S. SOF increases, the problems of identifying the attacker(s) and the means of responding to attacks will be vastly increased. This holds true whether or not a launch on warning policy is adopted, and it simply emphasizes once again the political-strategic problems associated with the horizontal proliferation of strategic nuclear missiles.

There is some definite threshold above which a pre-emptive attack would threaten the SOF’s capability in a second-strike assured-destruction role. This threshold value can be influenced by the characteristics of the SOF, of the attack on it, and of the Soviet target complex and its defenses. Because of uncertainties in many of these characteristics, it is not wise to attempt to select a specific threshold value above which a launch signal would be given. It would be sufficient simply to articulate the policy that any attack which would threaten the U.S. second-strike assured-destruction capability by reducing it below an unspecified level would trigger a U.S. launch before the SOF is actually damaged. A policy of launch on warning should in no way change the U.S. response to what is clearly an isolated accidental or unauthorized launch of one or a small number of missiles. This is the type of attack which a light ABM system is meant to cope with.

Several steps are available which would further lessen the likelihood of a spasmodic response prior to the arrival of the pre-emptive counterforce attack on the U.S. Launching on warning could be limited to such a number of U.S. missiles as to guarantee that these missiles, together with the surviving withheld missiles, would provide the assured-destruction capability. (In calculating the required number of surviving missiles, the change in effectiveness of an enemy’s ABM system against attacks smaller than full-scale must be considered.) The launched missiles could be equipped with means of self-destruction in flight, should the expected attack not materialize. Obviously, these would be expensive steps; they would also reduce the size of the SOF until the launchers could be reloaded. However, it is believed that the SOF would still remain at a sufficiently high level to provide the assured-destruction capability considered essential, as can be illustrated by an example:

Let us assume that the U.S. considers it essential that 300 Minuteman missiles be available for launch as a second strike. Let us also assume that a Soviet pre-emptive counterforce attack would be 85 percent effective, leaving only 150 potential survivors from the original 1000. If the U.S. were to hedge and launch 200 missiles upon identifying the attack as one so large as to reduce its second-strike force below the acceptable level, then the withheld force could still be expected to have 120 survivors. Should the attack not materialize, the 200 missiles could be destroyed in flight before they pass beyond the North American continent. The remaining 800 Minuteman missiles would still provide a deterrent; in case of a second attack, they would have to be launched when the detected second attack is large enough to threaten the survival of 300 Minuteman missiles.

Another means of reducing the attractiveness of pre-emptive counterforce attack on the U.S. is to replace fixed land-based missiles with mobile sea-based missiles. The replacement could be accomplished without increasing the total U.S. payload, and from this standpoint it could be considered not to be destabilizing. Bilateral adoption of such basing has been advocated as maintaining mutual assured-destruction capabilities without providing incentives for either side preempting. Arguments against conversion of the U.S. SOF to an all sea-based force would probably center on the possibility of hidden vulnerabilities of a force composed of only one type of weapon system deployed in a relatively small number of ships, and on the added expense over land-based missiles.

The risks, advantages, and technological problems requiring solution for adopting a launch on warning policy are listed for ready reference.

risks

Among the technical, strategic, and moral risks are the following;

False alarms. This risk is probably regarded as the major one. It is certainly not possible to guarantee that false alarms will not occur. It does appear possible, on the other hand, to reduce the likelihood of their occurrence to a very low level, by the use of multiple redundant sensing techniques. In the event that a false alarm would result in the launch of U. S. missiles, it would still be possible to announce that the missiles were launched on warning and that, if the warning was false, they will be destroyed in flight before they pass beyond a specified line (e.g., over North America).

Condemnation of U.S. as callous and playing with fire. In response to this charge, it can be pointed out that adopting a launch on warning policy was publicly announced to emphasize the futility of pre-emptive counter-force strikes on the U.S. SOF. To the extent that the policy strengthens the U.S. deterrent by introducing additional risks for a potential attacker, it is no more callous than threatening a would-be attacker’s viability as a twentieth century nation.

Loss of flexibility. Launch on warning would dictate that the U.S. response be a countervalue one, both in order to deter and because it is unlikely that the U.S. missiles could be retargeted in time to take into account the attacking missiles. The U.S. policy of flexible response, with which launch on warning would have been incompatible, has apparently been replaced with one of assured destruction,4 with which launch on warning appears to be compatible. The loss of operational and targeting flexibility which could result appears to be a price worth paying for the additional deterrence provided.

Vulnerability of the launched missiles.

U.S. missiles in their silos are less vulnerable than during their boost phase; and although it can be argued that, once launched, they may be attacked in flight by some of the arriving missiles, it is unlikely that the attack could readily shift from silos to missiles in flight. Moreover, the attacker will not necessarily know which portion of the U.S. missiles will be launched on warning, so the defense has some latitude in avoiding the attack.

Expense. If, as is suggested here, additional expenditures are incurred for the multiple redundant sensors to reduce the likelihood of false alarms, plus additional expenditures to insure more reliable command and control systems, it may be argued that these sums take money away from the development and procurement of weapons. In other words, they do not increase the U.S. strategic capabilities. At the same time, the incorporation of self-destruct mechanisms into missiles is expensive in terms of cost, payload reduction, and reliability. Once again, it seems that this is a price worth paying, since it would lessen the need for additional rounds of weapon acquisitions.

advantages

A launch on warning policy must offer sufficient advantages to justify the risks and added cost. Among these are:

1. Reduced likelihood of a pre-emptive counterforce missile attack on the U.S. Launch on warning would negate the success of such attacks regardless of their size, thus reducing the temptation for pre-emptive tactics.

2. Increased risks to an attacker, by increased uncertainty about the level of attack which will result in a U.S. launch on warning.

3. Reduced temptation to pre-empt and consequent reduction of incentives for an offensive arms race.

4. Inducement of symmetrical, or perhaps even asymmetrical, arms reductions.

5. Tendency of adversaries to insure tighter command and control procedures in view of the increased risks of retaliatory response.

Technological problems

In order to apply a launch on warning policy successfully, several matters related to technology need improvement, and changes to certain capabilities and systems must be introduced.

1. An extremely high-confidence threat detection, identification, and assessment system to reduce to a minimum the possibility of false alarms. In order to achieve the high confidence levels needed, it is likely that multiple, redundant sensing systems will be needed. These are likely to include over-the-horizon radars, early-warning radars, and optical systems. The systems are likely to be ground-based, satellite-borne, and possibly shipborne and airborne.

2. The threat identification and assessment system must be coupled with a command and control system capable of making rapid decisions and implementing these in extremely short time. Should the threat be sufficiently massive that a portion of the U.S. missile force is launched on warning, it may be necessary to communicate this fact to the U.S.S.R., together with assurance that these missiles will be destroyed in flight, should the alarm turn out to be false.

3. A missile destruct mechanism must be installed to be used in case of launch in response to a false alarm. This mechanism must be controlled from U.S. control centers and must be both highly reliable and tamperproof.

It is believed that publicly adopting a policy of launching U.S. missiles on warning of an attack sufficiently large to jeopardize the ability of the surviving strategic offensive force to inflict unacceptable damage on the attacker will be a useful strategy for the following reasons:

1. A launch on warning policy makes a preemptive counterforce first strike upon the U.S. unattractive, since the attacker cannot be sure of thwarting an assured-destruction second-strike response. Thus the nation’s second-strike capability need not be increased in order to provide a sufficiently strong surviving force.

2. The characteristics of the SOF are such that a launch on warning can be implemented. This option is already available to the U.S., but it is believed that adopting such a policy would increase the credibility of the U.S. deterrent.

3. The launch on warning policy could, in addition to lessening the likelihood of an arms race, provide a means for reducing the size of the U.S. SOF and, reciprocally, the size of the Soviet strategic offense capability.

4. Improvements in the ability to detect and identify attacks and in the command and control systems for the SOF would aid in carrying out the policy. Such improvements could be financed from funds that would otherwise be devoted to the development and deployment of new systems.

5. The potential for stability in U.S.-Soviet strategic offense structures, brought about by removing the temptation for a pre-emptive counterforce first strike, outweighs the potentially adverse criticisms of a launch on warning policy.

6. The uncertainty of the threshold at which a pre-emptive attack would trigger a responsive launch adds to the deterrent provided by the launch on warning policy.

It should be noted that an announced policy of launching U.S. missiles prior to the impact of attacking missiles will discourage a Soviet pre-emptive counterforce ICBM strike. There are other types of attack that could endanger our assured-destruction second-strike capability. Examples include attacks by submarine-launched ballistic missiles, a greatly improved antisubmarine warfare capability that could destroy a high proportion of U.S. SSBN’s (nuclear-powered launch vehicles for Polaris missiles) with high confidence, and a large-scale ground assault on the Minuteman force, which could achieve similar results. Any U.S. policy that would discourage pre-emptive ICBM attacks could be said to encourage such actions. Providing increased protection to the U.S. SOF against non-ICBM threats would appear to be feasible at lower levels of expenditure, and certainly without intensifying the arms race.

Washington, D.C.

 Notes

1. The policy of launch on warning considered in this discussion must be differentiated from that which governs the launch of a missile force deployed in soft launch sites. One of the serious drawbacks of the first-generation U.S. intercontinental ballistic missiles was their vulnerability on the ground, coupled with the rudimentary warning system available and the long time required to launch the force. Under such conditions it became necessary to initiate the launch sequence upon receipt of the slightest warning. Today we have a much more numerous force, deployed in hardened and dispersed silos, a highly sophisticated detection system, and missiles capable of being launched in a very short time. Perhaps a better term for the policy discussed here would be “modified launch on warning.”

2. The purpose of a launch on warning policy is to insure that the U.S. assured-destruction potential is preserved; so it is important only to indicate the intent to launch the missiles targeted for countervalue attack before the U.S. missiles are destroyed in their silos. Should such a situation ever arise, deterrence will have already failed, and simultaneously attacking empty Soviet silos would not make any significant difference.

3. Because of technical considerations, it is believed that the U.S. SOF, consisting primarily of short-reaction-time solid-propellant missiles, is in a better position to launch an attack of its own after identifying the launch of a massive Soviet attack. The launch on warning option has been available to the U.S. since the deployment of Minuteman missiles; however, such a policy has not been publicly stated.

4. The U.S. assured-destruction policy threatens a countervalue retaliation in the event of a massive attack on the U.S. SOF.  


Contributor

Robert I. Widder (B. M. E., Cornell University) is Senior Research Engineer, Battelle Memorial Institute, Washington, D.C. After serving with the Army during the World War II, he held various engineering positions in the aerospace industry until assuming the present position in 1964. At Battelle his research has centered on the technical, economic, and strategic aspects of arms control, advanced weapon systems, and air-cushion vehicle transportation systems. Mr. Widder’s writings have appeared in Systems Designer’s Handbook and U.S. Naval Institute Proceedings.  

Disclaimer

The conclusions and opinions expressed in this document are those of the author cultivated in the freedom of expression, academic environment of Air University. They do not reflect the official position of the U.S. Government, Department of Defense, the United States Air Force or the Air University.


Home Page | Feedback? Email the Editor