Document created: 3 September 03
Air University Review, July-August 1975
There are two sets of theories about war and weapons that have taken on the character of myths; that is, they are accepted on faith and almost never really examined, despite their implications for the state of the world. First, there are the theories about arms races, which deal with the dynamics of changes in force size and composition and how those changes are instigated and effected. Second, there is a partially overlapping set of theories that deals with the state of the strategic balance at any one time. These theories concern static measures of the strategic balance: Who is ahead and why? . . . What are the relative static capabilities of each side? . . . What are the relevant measures for establishing such capabilities? . . . And what are the triggers for change in the existing balance?
Many of these theories have become accepted without challenge, primarily by
dint of constant repetition. It is only recently that work has begun on
challenging their premises and conclusions.1
In this article I summarize and comment on a number of the premises and conclusions that appear regularly and widely in the public literature about the strategic balance.2 These premises have generated myths and misunderstandings that are commonly used as bases for major conclusions about the state and dynamics of the strategic balance between the U.S. and the U.S.S.R.
Premise: Numbers do not count;
once you have reached a minimum assured destruction capability, virtually
nothing else matters—all else is overkill.
This argument has been virtually beaten to death, but it still manages to arise with regularity. For example, in the May 1974 Scientific American lead article, Barry Carter says that the “numerical relation” of the two countries does not count beyond the level at which each side could absorb an attack by the other and then retaliate to the “assured destruction” level against population and industry.3 And in Foreign Affairs, April 1974, no less an authority then General Maxwell Taylor argues that:
. . . numbers of missiles are
important only up to a certain point. The tremendous destructiveness of megaton
warheads soon produces a needless “overkill” to which the addition of more
weapons would amount to sheer wantonness.4
The fact that the SALT I agreements codified numerical disparities in the numbers of missiles—and SALT II, while so far appearing to provide for equal numbers of missiles, allows wide disparity in throwweight—suggests an explanation for the increase in frequency of statements of this genre. But repetition of an argument does not increase its validity.
On the contrary, numbers do count—both in the military arena and in the political one. The political value of numbers is obvious from the intensity of the debates about the meanings of “superiority” or “parity” or “essential equivalence. “If there were no importance, why such fervor? If numbers of weapons were not of value, why would the Soviet Union continue to buy more? One would expect that it would be awkward at best to believe simultaneously that we must stop a strategic arms race in order to save the world from oblivion and that the numbers don’t matter. Yet some people seem to be able to accept the paradox.
The military importance may be less clear. The premise that all is overkill beyond a small minimum (York’s 10 bombs on 10 cities example)5 has some validity, but only if the sole purpose for strategic weapons is to threaten to kill women and children. I don’t believe that threatening to kill women and children is a legitimate purpose of the strategic force at all, from either a political or moral point of view. But even if it were, it would not be the sole purpose. This is particularly clear since an antipopulation response in a context of parity is not credible as a support for guarantees to allies. In that respect, killing —or overkilling—people is an irrelevant measure of effectiveness and, more obviously, an irrelevant measure of sufficiency. Thus something other than spasm response against population is necessary. And small less-than-spasm responses against other-than-population targets look more credible for the side with a numerical edge, since this side would not be putting itself at even more of a disadvantage by so using a small portion of its forces. Moreover, there is a large number of nonstrategic military targets that could be attacked with strategic weapons in a discriminating fashion if such weapons were sufficiently accurate. In such a situation the use of strategic weapons could have significant advantages in timeliness and in their control. Again, numbers clearly count.
Premise: Moreover, the number of strategic on both sides does not measure
reality because “it does not include the 7000 U.S. tactical weapons in Europe
and the nuclear weapons aboard aircraft carriers.”
Not only does this premise fail to consider the “nonstrategic” requirements for the weapons in Europe and aboard carriers; it also ignores the presence of substantial numbers of Soviet tactical nuclear weapons in Eastern Europe—about 3500 of them6—and the several types of shipborne nuclear surface-to-surface cruise missiles that the Soviet Union has developed and deployed (and the U.S. has not), not to mention the Soviet intermediate-range (IR) and medium-range ballistic missiles (MRBM’s), which the U.S. tac nukes are considered to balance in part, at least politically.
The common Soviet SALT distinction between “central” and “noncentral” systems, which defines central systems as those that can target the “homeland,” ignores the fact that many Soviet cruise missiles are capable of being targeted against the continental United States. (I believe that Soviet distinction to be a very indistinct one, and I do not agree with it.
Moreover, the Russians have insisted throughout SALT that the nuclear forces of our NATO allies should be counted in the U.S. force totals. If this were to be accepted, then the territory of the NATO allies should be included in the definition of “our homeland.” And then Soviet weapons capable of reaching the territory of the NATO countries—the IR/MRBM’s, the short- and medium-range bomber force, etc.—should be incorporated into Soviet force totals.
Premise: Limiting damage from a nuclear attack is not a matter of policy
choice.
This myth is most clearly stated in the following:
Calculations indicate that an
attack against all of the U.S. Minuteman silos would result in casualties (U.S.
or Canadian) in the multi-million range from fallout alone; from all causes the
actual numbers would be still larger.7
Elsewhere it has been stated that even under favorable conditions an attack against all of the U.S. Minuteman silos would result in fallout casualties in the ten-million range. This is a relatively clear—and thus testable—claim. And we have tested it. What we find is that numbers in the range of ten million can only be derived by assuming a configuration of Soviet forces and their employment in complete disregard for any criteria of avoiding by-product damage.8 Moreover, these numbers of casualties can result only if, in addition, the U.S. completely ignores some simple, sensible protection measures. A “multi-million” figure would have to result from extreme assumptions about protecting people—that is, that people remain exposed in the open after the fallout cloud arrives. These assumptions are hardly most favorable conditions.
It is within the capabilities of the attacker to reduce fatalities. Different force characteristics and targeting doctrine— smaller-yield warheads, detonation in air rather than on the ground, and cleaner weapons—would reduce fatality levels to tens of thousands without significantly reducing military effectiveness. Reasonable assumptions about population protection would reduce fatalities further and would be warranted, since the cities would not be directly under attack and there would clearly be some warning of the coming fallout. One hundred percent of the civilians need not be assumed to be standing out in the open gawking at the fallout cloud night and day.
Thus it is quite plain that there are no fixed results of attacks that are not sensitive to attacker and defender actions. It is clear that the fallout from an attack with large numbers of large-yield nuclear weapons, by either side, could be horrendous, although U.S.S.R. forces tend in general to have larger weapons than U.S. forces. But the key word is “could.” Both the employment of weapons during a strike and the earlier force planning and procurement decisions are matters of policy choice.
Premise: The U.S. qualitative advantage in weaponry is such that it can
offset substantial numerical disadvantages in a number of categories.
First, this premise, when used to justify U.S. acceptance of numerical disadvantages, assumes that all, or nearly all, qualitative advantages rest with the U.S. This simply is not true. There are a number of areas in which the Soviets are more advanced than the U.S. They are considerably ahead in large ICBM technology. And some Russian applications of lasers are far ahead of ours. Electro-optical devices in advance of ours are widely deployed throughout the Red Fleet. Furthermore, the Soviets are definitely ahead in cruise missiles, particularly in antiship cruise missile launchers on both submarines and surface ships. (While generally interpreted as being aimed at ships, some of these are also potentially useful against a large variety of land-based targets in the American homeland.) The Soviets are also considerably ahead in the range they have achieved for sea-launched ballistic missiles (SLBM’s); the SS-N-8, which has a range of at least 4200 nm, will not be matched by the U.S. until about 1980. Additionally, they are ahead in the areas of chemical and biological warfare weapons and defenses and in some types of ocean surveillance systems. While some of these examples are nonstrategic, the restriction of comparison to strategic technologies is arbitrary and ignores overlapping roles and missions as well as transfer of types of technology.
Second, this premise assumes the permanence of U.S. qualitative advantages, even under conditions of imposed quantitative ceilings. The existing SALT I agreements concern only quantities and not qualities.* The qualitative advantages that the U.S. currently enjoys are not conferred by the agreement or recognized by it. The premise assumes the permanence, nonetheless, during a period in which Soviet resources can be focused on quality rather than quantity.
*With the exception of the restraints on ABMs. And the ABM Treaty does not codify any disparity in qualitative characteristics.
Third, quantity and quality are assumed to be completely substitutable in satisfying military requirements. But this is certainly not true. For example, take the matter of covering targets: one of the significant advantages of quantity lies in having enough weapons to use against all desired targets; on the other hand, one of the significant advantages of quality—particularly when it involves improved accuracy—is the increased probability of killing the desired target while avoiding by-product damage. These aspects are not interchangeable.
Premise: The net effect of qualitative improvements beyond where we are
now is to leave us worse off than before.
In an attempt to establish that the current strategic balance is stable, theories often try to establish that all future changes are destabilizing, as compared to past changes that have brought us to this stable situation. The theories often focus on the question of eliminating or restricting future qualitative improvements, since these are often seen as the impetus for changing the existing force balance.
The main argument behind this premise is that qualitative improvement is dangerous in the sense that it increases effectiveness primarily in first-strike rather than second-strike capabilities. However, virtually no important technological changes of the past have fitted into the neat distinction between first-and second-strike capabilities. The development of missiles based in hard silos or in submarines increased second-strike survivability on the one hand; but on the other hand, since the missiles replaced bombers, it reduced potential warning time for the Soviets and thus added to U.S. first-strike capability. In short, there is no evidence to suggest that potential future technological changes are destabilizing or that unilateral choices made by bureaucracies tend to be destabilizing while bilateral choices of technology made by negotiation tend to be stabilizing.
Furthermore, self-denial does not necessarily affect the advance of technology in general. It is frequently argued that “if we develop something, the other side will too” (the H-bomb, for example). But what is implied by such a statement is that if we don’t, then they won’t. There is a tendency in much of this discussion of unilateral restraint to adopt the “leader-follower” model of duopoly theory, with the U.S. as the leader and the Soviet Union as the follower. Unfortunately this appears to be wishful thinking. Unilateral restraint just hasn’t worked; it has remained unilateral. The Soviet Union has clearly gone ahead despite U.S. restraint in numerous instances, e.g., by its development of large H-bombs and in its deployment of large numbers of ICBM’s.
Another argument holds that it is not clear whether, if we said that we had no programs for improvements in accuracy, for example, the Soviets would believe us or, from their point of view, could rely on the belief. If there is any validity to the relationship between accuracy and stability, it is a matter not of real U.S. or Soviet capabilities but of the perceived capabilities.
Moreover, it is not at all clear that qualitative improvements which might appear destabilizing by one definition—the first-strike/second-strike distinction—may not be stabilizing by another definition. For example, in the matter of accuracy: improvements in accuracy, particularly when combined with low-yield nuclear or nonnuclear warheads, may be stabilizing precisely because they allow the achievement of an objective and only that objective to a degree not heretofore achievable. The capability of destroying a particular military target without causing unwanted civilian damage is an important result. If, despite all, something happens by accident, the less civilian collateral damage the greater the possibility of containing the conflict. Furthermore, the capability to limit collateral damage leaves an option to meet nuclear attack without unleashing holocaust but also without surrendering.
Another stabilizing result is the potential substitution of high explosives for nuclear weapons in some situations, which would raise the nuclear threshold.
Premise: The U.S. must not build new weapons because to do so would fuel
the arms race.
In contrast to the focus of the previous question on military stability of arms decisions, this premise is concerned with preparatory and budgetary effects. It asserts that change will result in “arms spirals” or “ever-accelerating arms deployments.” This is the issue often called “long-term or ‘arms’ race stability or instability” in contrast to “crisis stability or instability.”
The only real test of this proposition is to look at history. It turns out that U.S. restraint has not in fact resulted in any observable Soviet restraint. The U.S. strategic budget and U.S. forces have decreased by substantial percentages in the past fifteen years: for example, the strategic offense and defense budget has declined steadily from a peak in 1959 that was more than 2 ½ times the 1974 budget in real terms: the total destructive power of all U.S. strategic offense forces has declined since 1964, when it was 72 percent higher than in 1972, the lowest year since 1956. Since 1964 Soviet strategic budgets and forces have increased: the Soviet budget has increased by roughly 2/3; their ICBM’s alone have increased by a factor of 14, while their heavy bombers have remained roughly constant in number; their SLBM’s have increased by about a factor of 5; and the destructive power of these strategic forces has increased manyfold.
It is hard to envision a situation in which an additional U.S. deployment would have fueled more Soviet deployment!
Of course there is some relationship between their forces and ours. This is only natural, given the basic adversary nature of the relationship. All available evidence, however, indicates that the rate of Soviet development and deployment is determined largely by “independent considerations” or internal factors (for example, the Soviet gross national product is strongly correlated with their defense budget9) rather than factors imposed by the U.S. There is no evidence to suggest that they would respond specifically to marginal changes in U.S. programs. They have clearly not responded, despite the views of many arms controllers, to cuts or halts in U.S. programs. On the contrary, there is much evidence to suggest that Soviet restraint has appeared mostly when the U.S. has demonstrated determination, as in our development of counters to Soviet area antiballistic missiles. And, contrary to common arms control theories, there is evidence to suggest that the Soviet buildup of strategic offense forces may have stemmed in part from an awareness of Secretary of Defense Robert McNamara’s policy of restraint in U.S. ICBM deployments.
Premise: Impetus for destabilizing change comes from the military, which
always overestimates Soviet forces and capabilities to justify increases in
U.S. forces and capabilities.
This is the “annual Pentagon scare story” myth. Albert Wohlstetter, David McGarvey, and I have examined this premise in detail. The Foreign Policy article by Professor Wohlstetter summarizes part of this work.10 The research shows conclusively that the history of predictions by the Secretaries of Defense during the past fourteen years regarding Soviet strategic offense force growth has been not one of overestimation but in fact has been overwhelmingly one of underestimation. While the short “missile gap” period may have formed the basis for this myth, even that was not an overestimate but was a bit of ethnocentrism on our part—U.S. estimates of ICBM’S were off by about as much in one direction as the estimates of IR-MRBM’s were in the other. The evidence on this clearly demonstrates that the “missile gap” was a U.S. misreading of Soviet priorities.
The Pentagon, to be sure, does talk of the Soviet threat it expects to face when it is presenting its own force program. When else would it be more relevant? On the other hand, the evidence clearly establishes the falsity of the notion that the estimates it then presents are invariably high. Out of the 51 estimates of future Soviet capabilities that we examined in the Secretary of Defense Posture Statements since 1962, the mid-range of only two exceeded the actual achieved deployments, and the high end of only nine exceeded the reality.
What is important about all this? The belief in the myths affects decisions in three major classes of issues: assured destruction, accuracy improvement, and arms control. The major conclusions often based on the assertions I have noted are that assured destruction is enough; accuracy is bad; and arms control agreements are so important that major quantitative disadvantages can be accepted to achieve them, because our qualitative advantages can offset quantitative disadvantages and will endure. A finding that these myths are incorrect lends credibility to the opposite conclusions:
· Assured destruction—if it is relevant at all-is insufficient.
· There are many reasons why increasing accuracy may be beneficial.
· Arms control agreements that codify quantitative advantages to the Soviets and are justified by qualitative advantages to us—which are unlikely to last and are not codified in the agreement—could be highly disadvantageous to the U.S.
Arlington, Virginia
Notes
1. See the forthcoming book, Competition or Race? Innovation and the Changing Size of Strategic Forces, by Albert Wohlstetter, Amoretta Hoeber, Fred Hoffman and David McGarvey; also, drawn from this work, Albert Wohlstetter, “Is There a Strategic Arms Race?” Foreign Policy, no. 15, Summer 1974, and “Rivals but No ‘Race,’ “Foreign Policy, no. 16, Fall 1974.
2. This article is an outgrowth of the joint work cited in note 1. It was presented in modified form at the Military Operations Research Society meeting at West Point in June 1974. I am indebted to the critiques of Albert Wohlstetter and David McGarvey and the comments of Donald Brennan.
3. Barry Carter, “Nuclear Strategy and Nuclear Weapons,” Scientific American, vol. 230, no. 5, May 1974.
4. Maxwell D. Taylor, “The Legitimate Claims of National Security,” Foreign Affairs, vol. 52, no. 3, April 1974.
5. Herbert York, “Deterrence by Means of Mass Destruction,” Pacem in Terris III, September 6, 1973.
6.The Military Balance l971-1972, The International Institute for Strategic Studies, London.
7. Wolfgang K. H. Panofsky, ‘‘The Mutual Hostage Relationship between America and Russia,” Foreign Affairs, vol. 52, no. 1, October 1973.
8. John Warner of Science Applications, Inc., has also done some detailed analysis of this question. I draw on his analysis here as well as that of David McGarvey and myself.
9. John Després of the RAND Corporation has done considerable work on correlations of Soviet defense budgets with many factors. This conclusion is drawn from his work.
10. Albert Wohlstetter, “Is There a Strategic Arms Race?” Foreign Policy, no. 15, Summer 1974.
Mrs. Amoretta Hoeber (B.A., Stanford University) is a Senior Analyst, Systems Planning Corporation, Arlington, Virginia. Previously she was Director, Department of Military Policy Analysis, General Research Corporation; consultant for several defense firms, including General Research Corporation, Lockheed, RAND Corporation, Hudson Institute; analyst, Strategic Branch, Analytic Services, Falls Church, Virginia; and Research Assistant, Strategic Studies Center, Stanford Research Institute.
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.
Air & Space Power Home Page | Feedback? Email the Editor