|
|
|
|
|
BERNARD BRODIE in 1946Thus far the chief purpose of our military establishment has been to win wars. From now on its chief purpose must be to avert them. It can have almost no other useful purpose.
The destruction of Hiroshima and Nagasaki in 1945 demonstrated to the world the unprecedented scale of devastation which could now be achieved from a single attack. The death and disruption which had needed vast numbers of aircraft, over a number of hours, using conventional high explosive bombs, could now be obtained with a single weapon in an instant. Even this power was soon to be dwarfed by the theoretically limitless energy of the thermonuclear bomb. The American strategist, Bernard Brodie, appreciated the - significance of this development from the earliest days of the atomic age. He could visualise no effective defence against such weapons and so nuclear deterrence could be the only practical future strategy.
The decision to go to war is an act of choice by a state. The choice may be made for offensive purposes, in order to gain some advantage; or it may be for defensive reasons, in order to protect some vital interest from an aggressor. In either event, the decision to go to war has been based on an assessment, which is not necessarily correct, that there is more to be gained, or less to be lost, by going to war rather than refraining from war. The essence of deterrence is the ability to convince a state, which has it in mind to go to war with you, that no such advantageous profit and loss assessment is possible. To achieve this happy position, a number of conditions must be met:
- Capability. The deterring state must have the capability to ensure that the potential enemy will profit less by going to war than by refraining from war.
- Will. The deterring state must have the will to use its capability if necessary.
- Credibility. The potential enemy must know of, and believe in, both (i) and (ii).
- iv Rationality. Each side must be sufficiently rational to base its conduct on an understanding of (i) to (iii).
By examining each of these four conditions in detail, the various threads and paradoxes of deterrent theory emerge.
What capability is necessary to ensure that no potential enemy can make the wrong deduction about his inevitable losses, should he go to war? Many had expected that the conventional 1 strategic bombing, which was to come in World War II, would cause unacceptable devastation. In the event, the damage was not a decisive factor and the capability was not a deterrent. Brodie maintained that nuclear weapons are different because; of the lack of warning time, the instantaneous destruction, the uselessness of shelters and the additional casualties from the delayed effects.2 The United States Senate Committee on Foreign Relations commissioned a report, which was published in 1980, into the effects of a nuclear war. The report examines, as objectively as possible, the various outcomes of a range of conflicts in which nuclear weapons are used. The stark summary of the effects on either of the superpowers of a large scale attack shows the quite different deterrent capability of nuclear weapons from high explosive:
Summary of Effects Case 4: Attack on range of military and economic targets using large fraction of existing arsenal. Main causes of civilian damage: Blast and fallout; subsequent economic disruption; possible lack of resources to support surviving population or economic recovery. Possible breakdown of social order. Possible incapacitating psychological trauma. Immediate Deaths: 20,000,000 -160,000,000 Middle-term effects: Enormous economic destruction and disruption. If immediate deaths are in the low range, more tens of millions may die subsequently because the economy is unable to support them. Major question whether economic viability can be restored - key variables may be those of political and economic organisation. Unpredictable psychological effects. Long-term effects: Cancer deaths and genetic damage in the millions; relatively insignificant in the attacked areas, but quite significant~cant elsewhere in the world. Possibility of ecological damage.(3)
This summary describes what amounts to the destruction of the state. Indeed, elsewhere it explores the economic problems following an attack and points to the possibility that 'the future of civilisation in the nations attacked would be in doubt'.(4) Other research has suggested that the use of between 500 and 2000 nuclear warheads could trigger a climatic catastrophe, which would carry with it the risk of the extinction of the human race.(5) In the early days of deterrence theory, attempts were made to quantify levels of assured destruction necessary to deter; modern nuclear arsenals leave little room for doubt that these levels have been achieved.
Nuclear weapons can therefore provide a level of destruction which, by any criteria, will leave a state worse off if it suffers a nuclear attack than if it takes a course of action which will not result in such an attack. One such course is to be deterred, the other is to prevent the nuclear attack from happening by destroying the opponent's capability. Thus to preserve a deterrent capability, a state must be able to deliver its weapons under all possible circumstances. Much concern has been expressed by the theorists over the possibility of preventative or preemptive attacks in which the deterring state is deprived of its retaliatory capability. A threatened state could reduce its risk of unacceptable damage by making the first strike against the other state's nuclear forces: a counter-force strike. As the French strategist, General Beaufre, said:
The conclusion was therefore that capacity for riposte was the key to nuclear deterrence, whereas capability to reduce the riposte was the key to nuclear initiative.(6)
What is necessary is the ability to absorb a first strike, and yet still retain the capability to return a sufficiently devastating attack against all that the enemy values: a counter-value strike. There is general agreement that the invulnerable second strike capability against the counter-value targets (which in practice would be cities) underpins any deterrent posture.
If the political will to use the nuclear weapon capability does not exist, then deterrence can only be achieved through bluff. As Herman Kahn explained:
If we are only pretending that we would do it, the credibility and therefore the deterrent value of our force is i almost certain to be lessened by the automatic and : inevitable leaks.(7)
The question of credibility will be discussed later, but the factors which affect the real determination to use nuclear weapons in extremis are now considered.
Deterrence of a non-nuclear state by a nuclear weapon state ought to be straightforward. There should be no doubt that the nuclear state can devastate its non-nuclear opponent, while remaining unscathed itself. It is by no means clear, however, that any nuclear state, with the possible exception of Israel, has the will to use its nuclear weapons against a non-nuclear enemy. Leaving aside the examples of Hiroshima and Nagasaki, each time that the opportunity to use nuclear weapons has arisen, the nuclear state has limited itself to the use of conventional weapons only; even at the price of accepting limited defeat. It appears that the political will is lacking for practical, political and ethical reasons.
Brodie suggested a number of different reasons why nuclear weapons were not used in the Korean War.(8) The limited American stockpile was kept in reserve for Europe; the weapons were considered inappropriate for tactical targets; there was political pressure from the United Nations allies; the Soviets might use their bomb; there were distasteful racial implications in using atomic weapons first against the Japanese, and then against the Koreans and Chinese. Thomas Schelling, in 1960, highlighted the importance of a tradition of the non-use of nuclear weapons.(9) By making a 'tacit bargain' through the continued non-use, then the expectation grows that all crises will be solved without recourse to nuclear weapons, and stability is enhanced. Whether this restraint has been the conscious - establishing of a tradition, moral repugnance or fear of other super- power involvement, the taboo has grown. It appears unlikely, unless the survival of a country was at stake - as could be the case in Israel - that nuclear weapons will be used against an enemy armed only with conventional weapons.
The political will to use nuclear weapons against another nuclear-armed state raises many of the fundamental questions of nuclear deterrence. Once each of two opposing states has invulnerable retaliatory forces, deterrence is mutual. The question- then is whether, given the retaliatory power of the opponent- , a state would ever have the will to use its weapons. Herman Kahn explored the consequences of a nuclear exchange in the 1960s, and considered that measures could be taken to limit the damage expected sufficiently to reinforce the will to use nuclear weapons. His view that civil defence measures, air defence improvements and targeting the enemy's nuclear weapons on the ground could so limit damage that the 'survivors will not envy the dead (10) has lost credibility in the light of the growth of nuclear capabilities. Henry Kissinger, writing three years earlier, could not conceive of a meaningful victory in a total war, but he was concerned with the effect that this nuclear stalemate would have on his nation's strategy and diplomacy:
The costs of an all-out war are too fearful for it to be our only response to a challenge. Even if it could be ~won, we should seek to achieve our objectives at smaller sacrifice. Strategy can assist policy only by developing a maximum number of stages between total peace (which may mean surrender) and total war. It can increase the a willingness of policy-makers to run risks only if it can demonstrate other means of preventing amputations than the threat of suicide. (11)
There is a widespread assumption that a government would have the will to use its nuclear capability in response to an all-out attack on the homeland. The rationality of this action will be considered later. This spasm response is the basis of mutual assured destruction deterrence. The worry that Kissinger was expressing was the need for appropriate responses to threats at levels beneath that of the destruction of the homeland. The extending of deterrence to cover a range of conflicts, including the defence of allies, is a recurring problem in the writings of strategists. If deterrence works, then a country will be self-deterred if it has no alternative but all-out war in any conflict. What is required is a range of options which will give the politician the chance to combat aggression with the appropriate level of defence. A conventional attack is met with a conventional defence; tactical nuclear weapons meet a nuclear response; limited strategic options of, for example, the destruction of one city, are returned in kind. The political will to use a capability to reply in kind is more likely to exist than the political will to cause a total war for limited objectives.
The range of options can be described as a 'flexible response' doctrine or a 'nuclear war-fighting' capability, depending on the view of the writer as to the motives of that state. In either case, it has a number of attractions to the political, military, scientific and strategic establishment. It gives the politician a wide range of options; it allows the military to equip and plan for a wide range of contingencies; it gives the scientist scope to develop new weapons; and the strategist can devise scenarios to justify the range of options. However, if every level of engagement is to be catered for, the potential cost can become very great. In admitting that military resources must be finite, the politician must admit also that the appropriate response available for any given level of aggression may be inadequate. The option then is to concede or to escalate. A conventional attack would be met with a conventional response, to the limits of the defending forces. If the defences do not hold, then battlefield nuclear weapons would be used to prevent defeat. This type of escalation scenario envisages a graduated increase in the level and extent of the battle until one side stops, having realised that it cannot win. In a symmetrical conflict between nuclear powers, it is difficult to see how the escalation of violence by one side can do anything except raise the level of the response. With the exception of the important tradition of the distinction between conventional and nuclear weapons, there is no natural stopping place on the slope to all-out war. Even if such fire breaks could be tacitly agreed, one study by Desmond Ball (12) gives little prospect for the practicalities of control of a war, once the nuclear threshold has been crossed.
If the practicalities of a particular flexible strategy are such that there is a strong possibility of any response to aggression leading to a strategic nuclear exchange, then the problem of the political will has not been solved. This difficulty is particularly acute in NATO strategy. Any use of nuclear weapons in Europe, at whatever level, is likely to result in massive devastation to the country which is the scene of the conflict. The political will might well exist in the United States to use battlefield nuclear weapons to stem a possible conventional defeat on European soil, if there were a strong belief that further escalation would not occur. In Europe, the attractions of limiting the destruction to the continent itself might be less strong. Indeed, there would be a desire to shelter under the greater deterrent threat of the American strategic guarantee. This would increase the pressure for rapid escalation if the nuclear threshold were crossed. The problems of the deterrent posture for Western Europe have given rise to two schools of thought. There are those who follow General Beaufre (13) Henry Kissinger (14) and Glenn Snyder (15) and maintain that the tactical nuclear weapons provide the essential graduated link between the strategic nuclear forces and the conventional forces. The expectation that the political will exists to meet aggression at the conventional level, but is likely to escalate to the strategic level, couples the strategic deterrent to all the other possible response levels. Those of the other school - Richard Garwin,(16) Lord Carver (17) and Fred Ikle,(18) for example - argue that the strategic deterrent is effective. A strong conventional force is usable, whereas the prospect of a nuclear battlefield in Europe would sap the political will of West European leaders.
While there is some force to both proposals, it is unlikely that either would make a great deal of difference in practice to the kind of decision which politicians would have to make in a crisis. While two of the European NATO powers - Britain and France - have their own deterrent forces, it would seem likely that at least one nuclear power would be prepared to honour its guarantees, and this aspect is discussed further when credibility is considered. The argument that conventional forces need to be increased to raise the nuclear threshold, and thus enhance deterrence, is valid only if it dissuades the opposition from adventurism. Conventional forces as a deterrent have had little success in the past, and if an increase in them signifies a reluctance to use nuclear weapons, then this may reduce the overall deterrent effect.
If the stark alternatives of suicide or surrender sap the political will, and the graduated devastation of the countries where the battle is fought is also politically controversial, the third strategy to consider is that of limited strategic options. In this case, the response to an act of aggression would be to carry out, either with or without warning, limited strategic strikes. The strikes could be on selected military, industrial or urban targets. If the action, plus the threat of further strikes, is sufficient to cause the adversary to withdraw, then this is a powerful form of deterrence. Escalation is, of course, still possible and indeed could be more likely following even a single strike on an enemy city, but it can be argued that the war would be more easily controlled than the fast moving events of the nuclear battlefield. One investigation in 1962 into the political will to carry through such a policy of limited strategic war concluded that it did not seem to offer a viable alternative to the normal counter-force strategy which existed.(9) However, since then, the technical capability to carry out accurate strategic attacks has improved considerably with the advent of better navigation systems, rapid retargeting facilities and a wide range of warhead yields. It is also entirely likely that the leaders of any nuclear countries would wish in a crisis to keep close personal control over the use of nuclear weapons, at least in the early stages. This would mean that the effective strategy might well become one of threats for very limited strikes, with deadlines set, followed by careful execution of the threat. Limited strategic options appear to be the more likely way nuclear weapons would be used, whatever the declared strategy or the force structure. In this case, there is some merit in questio°ning the usefulness of a tactical nuclear capability if it has an adverse effect on non-nuclear capability.
In summary, the political will to use the nuclear capability remains an uncertain area. The political will may be there for a spasm retaliation to an attack on the homeland. At lesser levels, the political will is questionable, whatever the force structure, given the expectation of escalation. In the event of a crisis and the failure of deterrence, the capability to threaten limited strategic strikes may serve to offer sufficient prospect of control to reinforce the political will.
The credibility of a particular deterrent posture cannot be assigned some absolute value. It is a measure of the probability, as perceived by the potential enemy, that the deterring state has both the will and the capability to carry out an action which will make the planned aggression not worth doing. It is interesting to note that there is an interaction between the perception of capability and of will. For example, it is entirely credible that a conventionally armed invasion force would, normally be met with a conventional defence: the political will is credible. However, the deterrent effect of this conventional defence posture is reduced if its capability to stop the invasion is perceived by the invader as low: the capability is not credible. At the other extreme, to threaten massive nuclear retaliation in response to a minor incursion may lack credibility, not through any doubt of the capability to make such an incursion less than worthwhile but because of a disbelief in the political will to carry through such a policy. Where two states have an assured retaliation capability, it will seem especially unlikely that the defender will risk retribution by taking such heavy handed action against a small aggression. In general, the harsher the nuclear response, the more improbable that it will be employed, for fear that it will draw similar or worse retaliation.
In an attempt to rationalise the mutual perceptions of probabilities of action, some strategists have reduced the relationships between states to mathematical models. While no precision is claimed for the values attributed to specific probabilities, and to the profit and loss values of different outcomes, their authors claim that 'to express it in mathematical terms can provide a useful check on intuitive judgment' (20) The methodology of game theory could be of value, when examining past crises, as a way of pointing to factors which merit detailed analysis. It is of limited application when dealing with the complexities of the decision-making processes, to produce answers to future problems which are as yet ill-defined. The Cuban missile crisis was a rare occasion on which there was some evidence of the perceived probabilities involved, and where the deductions from game theory give rather unconvincing results:
President Kennedy said that he thought the chances of war during the Cuban missile crisis were at least one in three. If he believed this, he either placed an incredibly high value on prevailing, or else he did not understand probability-utility calculus. It is hard to accept the notion that Kennedy thought the costs of war combined with twice the gains that would accrue if he won (the payoff for 3 standing J~firm when the odds were two to one that the Russians would back down) were anything like equal to the value of losing combined with twice the value of a compromise (the payoff for backing down with the odds as stated) (21)
As both ignorance and the conscious ignoring of probability- utility calculus are likely to be continuing features of crisis management, game theory offers little help in formulating deterrent strategy for the wide range of possible future~ scenarios. We shall not consider it further, but return to the subjective approach.
To make a deterrent threat credible, the deterrent state must signal its potential enemy that there is a significant probability that the appropriate retribution will be forthcoming. As we have seen, a convincing readiness to fight a nuclear war at any level can improve credibility. Measures such as civil defence, air defence and the complete integration of nuclear arms into battle plans make this posture more credible. As Kahn (22) pointed out, the deterrent posture would become completely credible if it could be made to be a totally automatic form of retaliation. Moves towards this can be made by adopting 'launch on warning' techniques and delegating nuclear release authority to the lowest levels. While these measures may improve credibility, they have two major drawbacks: they make accidental war more likely; and they give worse outcomes should deterrence fail. Responsible governments are, rightly, likely to depend on declared policy, rehearsed in exercises, with authority retained at the highest level, to give as convincing a posture as possible.
The credibility of a nuclear response is also increased as the number of independent nuclear weapon states increases. If more than one such state is threatened, then a lack of political will in one place can be compensated for by firmness in another. Multiple centres of independent decision-making increase the credibility of an alliance deterrent. The converse situation arises in an alliance when more than one state must agree before nuclear weapons can be used. In this case, a lack of will on the part of any of the chain of decision-making states would affect the response of them all; and thus the credibility of the deterrent posture of such an arrangement is reduced. 'Dual-key' nuclear systems may offer reassurance to the public, but they are not as effective as deterrent forces.
In discussing capability necessary to deter, the necessity for invulnerable retaliatory forces was highlighted earlier in this chapter. Not only must these forces be invulnerable but - of greater importance - the potential enemy must believe that they are. This can be done by making the problem of their simultaneous preemptive destruction as difficult as possible. Sheer numbers of systems compound the problem of their destruction. Different basing systems - submarine, air-launched- , ground-based, ship-carried, low altitude transit and ballistic flight - all serve to complicate the attacker's disabling strike plan, and hence increase the credibility of the deterrent posture. The multiplication of weapon systems also acts as an insurance against future technological developments, which might render a particular basing system too vulnerable to retain credibility.
The final aspect of credibility is concerned with the potential aggressor's perception of the defender's areas of vital interest. A deterrent posture will fail if the potential enemy does not believe that it applies in a particular crisis. By failing to make an unambiguous security commitment to South Korea when their troops withdrew in 1949, the Americans could not expect their deterrent posture to be as effective as if they had given a guarantee. One study by George and Smoke (23) concludes that deterrence failed because the Communists did not believe Korea was an area of vital interest to the United States. The problem does not end with the defining and signalling of areas of vital interest. Future credibility depends on whether these areas are defended when necessary. A failure to meet an alliance or treaty obligation, or the appearance of a lack of resolve in one crisis, can reduce the overall credibility of a deterrent posture for the future.
In summary, credibility deals with mutual perceptions and cannot be an absolute science. A massive risk needs only to be slightly credible to deter; a less intimidating threat may by its nature be more credible, yet not deter as much.
For deterrence to operate, a certain level of rationality must be presumed in the decision-making process of the state which is to be deterred. At minimum, this needs to be an assumption that the enemy will refrain from war if he believes that he will lose more, by going to war, than by refraining from war. It can be argued that the penalties of nuclear retaliation are so great that a form of rationality will be necessarily imposed on what may otherwise be a less than rational government. Professor Waltz has explored the question of whether, had nuclear weapons been available, Hitler's behaviour could have been moderated. He concludes that Germany, either through Hitler or his generals, would have been deterred.24 Even if this speculation were true, it is equally possible to postulate irrational enemies, nuclear terrorists for example, who would not be deterred. For those who lack the rationality to understand the rules of the game, deterrence theory has no answers.
When mutual deterrence is operating, similar assumptions must be made about the rationality of both sides. If these assumptions of rationality are followed through, an interesting paradox emerges. Kahn made the point, since made by many others:
In most deterrence situations, once deterrence has failed, it is irrational to carry out the previously made warnings or threats of retaliation since that action will produce an absolute or net loss to the retaliator.(25)
If the carrying out of the threat is irrational, and there is an assumption of rationality, then the threat will not deter, and deterrence fails. Fortunately, the knot tied by these assumptions of consistent rationality is not difficult to cut through. Patrick Morgan summed up the problem as:
The ultimate problem in devising a theory of deterrence is to simultaneously postulate that officials and governments are sometimes in some ways conscious and rational, yet take into account that certain forces work to make their behaviour unconscious and irrational. (26)
Governments cannot be aware of all the possible outcomes of any decision: they have less than perfect information, intellect and control; they are subject to conflicting pressures; and their alternatives may be constrained in different ways at different times. If deterrence were to fail, there is little prospect that an attacked nation would act in a purely rational way. Deterrence can operate because of the entirely rational fear that the °opposition will act less than rationally, once it has suffered loss. It is interesting to see how Bernard Brodie modified his views as a result of the success of nuclear deterrence. In 1959, he wrote:
Over the long term a policy of deterrence threatens to founder on the fact that too few people are sufficiently rational or sufficiently wise, with respect to either diplomacy or strategy to make it work.(27)
Fourteen years later he said:
It is a curious paradox of our time that one of the foremost factors making deterrence really work, and work well, is the lurking fear that in some massive confrontation crisis it might fail. Under these circumstances one does not tempt fate.(28)
The rationality paradox is solved by the understanding that it is entirely normal for human beings to act irrationally, especially when under pressure.
Deterrence theory has emerged through the work of social scientists, historians, scientists, military thinkers, politicians and the followers of many disciplines. It is by no means a cohesive internally consistent theory which can predict best strategies for every scenario. Indeed, some deductions are mutually exclusive. Nevertheless, it can show which factors may be of importance when attempting to make a deterrent posture more effective.
The examination of the requirements necessary to meet the four conditions for deterrence showed an edifice built on mutual perceptions and assessments of risk in an unpredictably rational world. While theory could point to ways to improve deterrence posture, the results often made the outcome, if deterrence failed, less controllable. The realities of the world can make the structure and strategy of the deterrent force quite different from the ideal. In Part II, this practical application of nuclear weapons to strategy is examined.
|
|
|
|
|
