Below follows an excerpt from of “Anything But Simple: Arms Control and Strategic Stability,” Dr. Ford’s chapter in a book published this month by the Strategic Studies Institute at the U.S. Army War College entitled Strategic Stability: Contending Interpretations (Elbridge A. Colby & Michael S. Gerson, eds.) (Carlisle, Pennsylvania: U.S. Army War College Press, 2013). As a “teaser” for the full text of the book – and in consideration of readers’ patience on this website – the selection below consists only of the opening sections of Dr. Ford’s chapter.
The book can be had in PDF form from SSI by clicking here. (Dr. Ford’s contribution appears as Chapter 6 in that volume.) In addition to Mssrs. Colby and Gerson, other contributors to the book are Thomas Schelling, Dale Walton, Colin Gray, James Acton, Ronald Lehman, Jeffrey McCausland, Matthew Rojansky, Laura Saalman, and Austin Long. NPF strongly recommends the book.
Chapter 6: “Anything But Simple: Arms Control and Strategic Stability”
“Strategic stability” does not appear to have any generally-agreed definition. Contributors to this volume, for instance, take a range of positions – from focusing very specifically upon the incentives nuclear-armed powers face to alter their nuclear force posture for fear of pre-emptive strike, to very broad understandings that sweep within their reach almost the entire spectrum of interstate violence. This chapter will outline one particular conception of strategic stability – a definition focusing upon the incentives for general war between great powers – before exploring the relationship between this idea of stability and arms control policy.
I will argue herein that despite the common assumption in the U.S. and global policy communities that arms control is essential to strategic stability, the reality is that the two concepts actually have an ambivalent relationship, and that arms control some- times fosters stability and sometimes undermines it. Moreover, stability, per se, is of indeterminate value. In assessing whether to seek strategic stability and whether to use arms control in its pursuit, one can- not rely upon a priori assumptions but must instead carefully examine the circumstances involved and the interests served by various different policy options – including nontraditional forms of arms control, or perhaps none at all.
I. “Strategic Stability” and its Implications
A. A Working Definition
This chapter conceives strategic stability in the geopolitical arena as being loosely analogous to a military “Nash Equilibrium” between the principal players in the international environment (i.e., the “great powers”) as it pertains to the possibility of their using force against each other. It defines strategic stability as being a situation in which no power has any significant incentive to try to adjust its relative standing vis-à-vis any other power by unilateral means involving the direct application of armed force against it. General war, in other words, is precluded as a means of settling differences or advancing any particular power’s substantive agenda. The environment is thus strategically stable if no player feels itself able to alter its position by the direct use of military force against another player without this resulting in a less optimal outcome than the alternative of a continued military stalemate and the pursuit of national objectives by at least somewhat less aggressive means.
This model, of course, is – like all social science models – only an imperfect description of any situation in the real world, and does not purport to in- corporate every relevant component of, or possibility for, state behavior. It revolves, for instance, around a general assumption of rationality, presuming that decisions on matters of war and peace usually occur as the result of calculations about the costs and benefits of contending courses of action, and not simply randomly, accidentally, or as a matter of emotional reflex (e.g., visceral hatred or exuberance). This Nash-inspired approach does not well accommodate these latter possibilities. Accidental war, for instance, might yet occur between powers in a “stable” relationship – a question that has arisen with particular acuteness in the era of nuclear weaponry.
This model also tends to assume that players are generally at least passably knowledgeable about their adversaries’ capabilities – that is, that they are not radically incorrect in the beliefs they hold and assumptions they make about other players. I do not assume perfect information, of course, and indeed, as we shall see, this model explicitly envisions that confidence-building measures may be able to lessen misperceptions and at least partly attenuate the security dilemma created as uncertainty about one’s opponent drives behaviors that themselves elicit seemingly threatening countermoves by that opponent. Nevertheless, this model has some difficulty accommodating the possibility of dramatic misapprehension, for in extreme cases divergences of perspective may become the functional equivalent of eliminating my assumption of basic rationality, for neither side would really be responding to the actions and position of the other at all.
Despite its flaws, however, I believe this Nash-inspired conception of stability is useful in the way that good models are supposed to be. As a heuristic, it provides a way of describing important aspects of real world behavior, identifying characteristic trends or tendencies, and providing a valuable tool with which policy choices and outcomes can be evaluated. As we shall see, this model offers a valuable prism through which to think both about stability dynamics within the international system, and about the potential benefits and costs of arms control.
It is important, however, to be clear about what the model actually envisions. Its focus upon the preclusion of general war between the great powers, for instance, does not imply that all means of conflict are ruled out. Indeed, strategic stability may create incentives for other types of competition, or for more indirect military clashes, if basic political or systemic rivalries are displaced into other arenas that carefully stop short – or are at least intended to stop short, for statesmen do not always get their calculations right, of course – of direct military conflict. This is what tended to happen during the Cold War, when both the United States and the Soviet Union became in various ways ensnarled in proxy wars, either themselves fighting adversaries supported by the other superpower or becoming involved in sponsoring the opponents of such forces.
Nor does my Nash-inspired concept mean that change in the major powers’ relative positions is ruled out, nor even one or more powers’ encouragement of other (nonmilitary) dynamics calling into question the very existence of another power’s government. If such “existential” challenges arise by means not involving the direct application of another power’s military force, I would still be willing to say that the environment remains strategically stable. This concept of strategic stability does not envision freezing a global status quo in place forever, but merely ruling out certain modes of competition and conflict – specifically, general war. Struggle may and in a sense must continue withal, and great powers may rise or fall by other means and for other reasons.
The persistence of some warring, even on a small scale, clearly makes it impossible, as an analytical matter, to rule out the escalation of minor conflicts into larger ones. The point is not that general war between major states is impossible, however, but that certain configurations seem to make it less likely than others. As demonstrated by the U.S.-Soviet rivalry of the Cold War, it is apparently quite possible for low-level proxy conflicts to occur without such combatants dragging their sponsors into the fray. The Korean War of 1950-53, however, illustrates the potential for problems, having brought Chinese and American forces into direct conflict – albeit one contained to a particular theater which did not escalate into a broader or more “existential” clash between these powers. One may deem a system strategically stable to the degree that relationships between the great powers are merely resistant to such escalatory pressure. Without recourse to a crude determinism, one can do no more than identify tendencies and likelihoods.
Nor, of course, is it inevitable that a strategically stable configuration will always remain so, for it may be that economic or other trends generate instabilities over time, such as by dramatically changing the balance of military power between states and thus making seem feasible direct military actions that might previously have been “unthinkable.” This does not make present-day strategic stability meaningless, however, for what it takes to create such a turnaround will presumably vary, with a more stable status quo ante requiring more to change before it will degenerate into instability than would be necessary to degrade a less stable initial situation. Here again, stable systems will tend to be resistant to change, but this does not mean that none can occur.
It should also be recalled that the definition of strategic stability offered here only focuses upon the principal players in the international system: the states one might call the great powers. Through this lens, small players may perhaps face existential military crises from time to time without the stability of the system as a whole being affected. Their particular trajectories might be unhappy indeed, but it does not necessarily follow that international politics as a whole is thereby strategically unstable. It would surely set the bar too high to define system stability as the complete absence of all violent conflict. A Nash-inspired notion of strategic stability might usefully apply as between smaller powers in their local context, of course, but that is not our task here. For present purposes, we shall be discussing the global strategic aggregate, and confining our analysis to major states because major states are those that can materially affect that aggregate in the most direct and important ways.
As it is used herein, the concept of strategic stability is value-neutral. This is not to suggest that there is necessarily anything inherently “good” about its achievement, though of course this may frequently be the case. Especially where nuclear weapons are widely possessed among the great powers, for instance, the argument seems compelling. In most circumstances, ruling out general war is presumably a very good idea. But I would stop before saying that strategic stability is a per se good.
Indeed, strategic stability might sometimes impose tremendous costs, for it tends to privilege the status quo between the powers in question. How one evaluates the merits of such stability will depend upon who one is in the constellation of players, what status quo that stability enshrines, and what it serves to permit.
- For a power that seeks fundamental change in the strategic environment, strategic stability is probably unwelcome, for it imposes sharp limits on how change may be sought.
- Nevertheless, in some circumstances, strategic stability could serve to protect an aggressive rising power while it prepares itself for a future military challenge to the global order. (War by France and Britain against Adolf Hitler’s Germany over the Austrian Anschluss or the invasion of Czechoslovakia in 1938 would technically have been an affront to strategic stability in Europe, but might have prevented greater stability challenges still to come.)
- Even where one might think strategic stability to be a salutary objective, moreover – as, for instance, in a balance between very powerful states whose clash could be catastrophic – it may have significant justice costs, such as by essentially “immunizing” a tyrannical regime against well-deserved foreign efforts to replace it by direct military means. (Strategic stability between the Axis and Allied powers on the eve of World War II, for instance, would have consigned much of Asia permanently to the Japanese yoke, and much of Europe to the jackboots of the Gestapo and the Nazi death camps.)
- Arrangements to ensure strategic stability might facilitate aggression against smaller powers, as occurred in 1939 when the Molotov-Ribbentrop Pact opened the door for aggression in erasing Poland from the map and dividing it between Nazi Germany and the Soviet Union. Even if not thus pre-arranged, furthermore, the immunity strategic stability tends to offer a power against direct military challenge from other important states could encourage unilateral external aggression against systemic “small fry” – or at least those lacking strong military alliance relationships with other major states, at any rate – by leading an aggressor to believe that the victims of such predation will not be saved or avenged by outsiders. This may to some extent have been the case with the Soviet invasion of Afghanistan in 1979, or with the North Korean invasion of South Korea in 1950 after U.S. officials created the impression – not just in Pyongyang but also in Moscow and Beijing – that the Republic of Korea was outside America’s “defense perimeter.” (As for those smaller states that do have strong alliance relationships with great powers, how- ever, one might argue that strategic stability is a precondition for their security, for it may be that these relationships provide deterrents to aggression only to the extent that they enable a minor player to participate in the stability of a great-power balance. Post-1953 South Korea may be a case in point.)
Depending upon the circumstances, therefore, stability can have decidedly unpleasant results. Though stability is presumably indeed often “good,” it can in other circumstances help empower the perpetrators of both internal and external aggression, coexist with local violence and instability, act as an enabler for aggression, protect the instigators of brutal internal repression, or serve to protect a power during its rise to a position from which it can challenge the existing great-power balance. A policy of seeking strategic stability is not, therefore, necessarily a sign of international benevolence and virtue. Details matter, and the point here is that it is not substantively or morally sustainable to argue that strategic stability is a per se good. It may be good, or it may be, on the whole, harmful. In order to assess its net value, one needs to know a good deal more than simply that things were “stable.”
B. Strategic Stability and Nuclear Weapons
Though the term comes up frequently in discussion of nuclear weapons and arms control policy, moreover, I do not envision strategic stability as being inherently about nuclear weaponry. That said, of course, nuclear weapons are of special salience in this arena, because they may seem to offer some states a real hope of achieving security – that is, of leading other powers to conclude that general war against them is inadvisable – to a great extent independent of the state’s actual ordinary (i.e., conventional) military strength. Nuclear weapons may have an enormous impact upon strategic stability, in other words, but the stability question neither begins nor ends with them. (Indeed, particularly with regard to new possessors among the minor states, nuclear weapons might provide relative security to some individual countries at a cost to strategic stability as we have defined it here, if such proliferation helped increase the risk of conflict between major powers – e.g., through the escalation of regional conflicts made more ugly and/or more likely by a proliferator’s emboldenment, or if major states were forced to undertake policies in response to proliferation that affect their capabilities vis-à-vis other great powers.)
Here lies a broader point. The impact of nuclear weapons is probably especially great in geopolitical terms precisely because they aren’t useful only to deter other such weapons – though many in the disarmament community would have it otherwise. They are important because they also deter conventional weapons, and nuclear weapons’ possessors often hope to use them as a sort of fast-track road to security without the expense and inconvenience of having to defend themselves by other means. The United States and its North Atlantic Treaty Organization (NATO) allies relied upon nuclear deterrence to make up for a perceived disadvantage vis-à-vis Warsaw Pact conventional forces in Central Europe during the Cold War, for instance, and nuclear weapons seem today to be prized – or sought – by planners in Moscow, Beijing, Pyongyang, and Tehran alike for their presumed ability to counterbalance others’ advantages in sophisticated conventional arms. Nor should one forget that nuclear weapons were first used not against a nuclear power but in order to help win a bitter conventional war.
Accordingly, one would argue the need to decouple the concept of specifically nuclear stability from strategic stability more generally. They are to some extent analytically distinct concepts, and conflating them would tend to obscure important points – such as the reasons why many states have pursued nuclear weapons in the past, why some seek them today, and an important reason that a country might use “the Bomb” (i.e., to win or to stave off defeat in an otherwise conventional conflict). Theoretically, moreover, a nuclear balance characterized by “complete” stability in nuclear terms – that is, a case in which more than one power possessed nuclear weapons but no circumstances existed in which these devices would be considered “usable” – might well be unstable under the definition of strategic stability used here: if asymmetries of conventional force or other circumstances made war attractive, nuclear weapons in this case might not deter it. This is why hopes for strategic stability in a nuclear-armed world presuppose that participants’ nuclear arsenals are not entirely “self-canceling.” In a multi-nuclear world, to deter general war with nuclear weapons requires some real possibility of weapons use – which is another way of saying that the success of nuclear deterrence requires that it be, to some degree, imperfect. Strategic stability and the specifically nuclear aspects of power-balancing are clearly related, but should not be confused.
In any event, on the assumption that this Nash-inspired concept of strategic stability is both coherent and useful, the following discussion will offer some thoughts on its relationship to arms control ....
The remainder of the chapter (which runs for a further 58 pages) goes on to discuss the relationship between arms control and strategic stability. First, it breaks down arms control into three basic analytical categories: (a) bilateral and multilateral agreements and arrangements related to limiting, reducing, proscribing, and/or dismantling some sort of weaponry or other military-related technology (i.e., capability-regulatory measures); (b) efforts to develop and promote “best practices” or codes of conduct pertaining to the use of certain types of technology or capability (i.e., behaviorally-regulatory measures); and (c) steps related to transparency and confidence-building (i.e., information-concessive measures). Rejecting the a priori assumptions so commonly made in the arms control community that strategic stability is per se good, and that arms control is also both per se good and inherently strategically stabilizing, the chapter discusses how and when arms control can contribute to stability – as well as when it can produce instability.
Among the potential pitfalls of arms control, the chapter identifies “capability ‘lock-in’” as a possible incubator of strategic instability – such as when an agreement binds parties to a force posture that no longer fits the needs of deterrence in a changing world, prevents adaptation to new or evolving threats, or otherwise impairs adaptation to circumstances. Another pitfall lies in “displacement effects,” where an agreement displaces competitive behavior into new areas, which may potentially create stability problems that turn out to be worse than those “solved” by the agreement in the first place. Also discussed is the possibility of “strategic manipulation,” in which one party is able to craft the terms of an agreement in such a way as to restrain its opponent in an area of that power’s would-be competitive advantage while leaving itself relative freedom of action in an area of its own.
In terms of the benefits of arms control from the perspective of strategic stability, the chapter discusses how “lock-in” effects can sometimes help create stability by: (i) constraining one side’s potentially-destabilizing advantages (or by imposing “cross-domain restraints upon mutually-asymmetric comparative advantage”); (ii) retarding or preventing the adoption of potentially-destabilizing technologies or postures by existing (or new) players; and (iii) reducing distrust and false perceptions, or recourse to provocative behaviors, that might otherwise spur the parties to adopt policies or acquire capabilities that could destabilize the balance between them.
The chapter concludes:
... In designing arms control regimes, many traps await the unwary or the credulous. Subtle shadings of circumstance can turn a well-designed and stability-promoting arrangement into a destabilizing geopolitical canker, and indeed one’s negotiating partner may be working very hard to skew stability dynamics in his favor. Capability-regulatory arms control can impose destabilizing rigidities as easily as it can restrain dangerous competitive dynamics, and the balance between such effects may also shift over time. At the same time, capability-focused regimes are certainly capable of providing real value, as can behavioral and information-centered approaches in their own distinctive ways.
For the policy community, then, the key lesson may simply be to avoid ideological complacency, remembering that arms control is neither inherently bad nor inherently good. It is simply a tool, and if one wishes to promote strategic stability – and to avoid engendering instability – there are many variables to take into consideration, and many dynamics of which one must be aware. Arms control theory needs to be de-theologized if arms control is to be practiced well, and the endeavor needs to be approached with an intellectual humility rooted in awareness that the strategic environment is difficult to shape, that effects are hard to predict, and that the world has a stubborn habit of changing over time in ways that sometimes make yesterday’s certainties implausible or even counterproductive. One could do worse than to approach the task of arms control planning with a wary eye.
-- Christopher Ford