Manipulation of Risk: The Strategy of Brinkmanship
What are nuclear weapons worth when they can't be used without inviting mutual destruction?
The fourth and final part of this series on nuclear strategies addresses the strategy of Manipulation of Risk, or brinkmanship. Based in game theory, this strategy is both the most complex and the one with the greatest potential for actual use in that it involves nuclear weapons, but not necessarily nuclear war. Prominent American economist Thomas Schelling provided the theoretical basis for this strategy in his seminal work Arms and Influence. In this, he argues that nuclear weapons are useful not in their actual employment, but in the danger of annihilation that they represent. States can increase the chances that a nuclear exchange will occur to deter or compel its opponent. It is this use of heightened chance of unintentional escalation that constitutes the strategy of Manipulation of Risk.
Schelling begins by drawing the distinction between “coercion” and “brute force.” Coercion he defines as the use of suffering and destruction to compel an enemy to concede an object. In coercion, force is most useful when it is latent, not yet used, so that an enemy may avoid it by giving in. Brute force, on the other hand, means overcoming the resistance of the enemy and seizing what you want directly. In the first case, it’s necessary to consider the interests of the enemy and how the promise of suffering can be used to make conceding preferable to resistance based on these interests. Brute force does not need to consider enemy interests, but only enemy strength, whether it can physically resist the seizure of whatever object is at dispute. “Coercion requires finding a bargain, arranging for him to be better off doing what we want—worse off not doing what we want—when he takes the threatened penalty into account.” This is very similar to Clausewitz’s characterization of war as a whole. Clausewitz does not draw a clear distinction between the two methods on the basis that even use of brute force requires coercion to conclude a war. After all, while force can seize an object, it cannot force an enemy to accept its loss. This acceptance—however temporary—is what allows a war to be concluded.
Manipulation of Risk focuses on coercion rather than brute force on the basis that nuclear weapons cannot take something nor hold territory, but only be used as weapons of extermination. Unless extermination is itself the goal, their main utility must be in threat of their use. Therefore, any nuclear strategy must be focused not around the use of nuclear weapons and the futile effort to win a nuclear war, but around using the threat of nuclear war to extract concessions. “Any affirmative action, any collaboration, almost anything but physical exclusion, expulsion, or extermination, requires that an opponent or a victim do something, even if only to stop or get out. The threat of pain and damage may make him want to do it, and anything he can do is potentially susceptible to inducement. Brute force can only accomplish what requires no collaboration.” Manipulation of Risk means raising the danger of nuclear conflict until your opponent would rather collaborate than endure it. While the risk is mutual, or even unequal, that does not eliminate its utility. The example of a pedestrian walking heedlessly across a road causing drivers to stop demonstrates the principle. Undoubtedly the pedestrian would be worse off if struck, but it is nevertheless the driver who ultimately yields.
A Nuclear Revolution?
Manipulation of Risk seeks to answer why nuclear weapons have been revolutionary to international affairs and why they are not merely a tool of war but necessitate a new strategy. Schelling argues that what is novel about nuclear weapons is not their capacity for destruction.
“It is not true that for the first time in history man has the capability to destroy a large fraction, even the major part, of the human race. Japan was defenseless by August 1945. With a combination of bombing and blockade, eventually invasion, and if necessary the deliberate spread of disease, the United States could probably have exterminated the population of the Japanese islands without nuclear weapons. It would have been a gruesome, expensive, and mortifying campaign; it would have taken time and demanded persistence. But we had the economic and technical capacity to do it; and, together with the Russians or without them, we could have done the same in many populous parts of the world. Against defenseless people there is not much that nuclear weapons can do that cannot be done with an ice pick. And it would not have strained our Gross National Product to do it with ice picks.”
This course of action was not so much as suggested in 1945. Yet nuclear weapons promise something remarkably similar. In total nuclear war, it is very likely that the greater part of humanity will perish, either directly or through second-order effects, as assuredly as if the United States had made the effort to do the job with ice picks. Schelling poses the question of why this kind of total violence was unthinkable in the conventional context but is accepted in the nuclear age.
“It is imaginable that we might destroy 200,000,000 Russians in a war of the present, though not 80,000,000 Japanese in a war of the past. It is not only imaginable, it is imagined. It is imaginable because it could be done ‘in a moment, in the twinkling of an eye, at the last trumpet.’ This is a difference between nuclear weapons and bayonets. It is not in the number of people they can eventually kill but in the speed with which it can be done, in the centralization of decision.”
It is this that is the main distinction between conventional and nuclear conflict. In a conventional conflict, when faced with a helpless enemy, the victor may choose how much violence and death to inflict upon them. Even as it is in the process of killing its enemies, it reserves the right and maintains the ability to reduce its aims, having either thought better of it or attained its desired objective. While a state may completely destroy another as Rome did Carthage, at any point during the gruesome process of bayoneting the victor may call a halt, either out of humanity (when the bloodlust cools) or because further destruction serves no purpose but requires time and effort. This is not the case when it comes to nuclear weapons. Once the weapons are in the air, there is no opportunity for reconsidering or moderating the damage done or for the victim to concede something that averts their destruction.
Thus, the change in war brought about by the nuclear age is not the level of destruction, but that it can be unleashed instantaneously, without recall. Without nuclear weapons, killing millions not only required the means, but a sustained desire to carry it out. In conventional conflict, the killing of civilians continues only until its purpose is served, which is only rarely extermination. By contrast, the starting point of nuclear weapons is extermination. Conventional warfare has a sliding scale of carnage to utilize, whereas the nuclear age reduces destruction to a binary.
Unprecedented Vulnerability
Schelling identifies a second way in which nuclear weapons change the traditional formula of war. Conventionally, terrorizing an enemy’s civilian population requires access, which can only be gained by defeating their military forces. As such, when a country chooses to enter a war, it may be assured that its population is secure so long as it loses no territory to the enemy. Even if some of its territory is occupied, the rest of the country is secure. If its ability to defend its civilian population comes into question, it has the option to sue for peace and make concessions to protect it. Throughout the war, the safety of civilians is contingent on both denial and negotiation. An occupied city may be sacked in the instinct carnage of its capture, but its population is rarely systematically targeted. The city and its people are more useful as a point of negotiation—the aim is coercive, and so the violence is more effective kept in reserve. From this conventional paradigm, pure killing has few opportunities and very little direct usefulness.
While it is true that in the Second World War, both sides conducted terror bombing of the other, these were only possible so long as one had an air force capable of reaching the enemy’s homeland. As the Germans lost the war in the air, bombing raids on allied territory became less frequent as bombings of German cities became ubiquitous. Bombing was possible when the air was contested. But victory proved a shield against the burning of one’s cities. Not so with nuclear weapons.
“Victory is no longer a prerequisite for hurting the enemy. And it is no assurance against being terribly hurt. One need not wait until he has won the war before inflicting ‘unendurable’ damages on his enemy. One need not wait until he has lost the war. There was a time when the assurance of victory—false or genuine assurance—could make national leaders not just willing but sometimes enthusiastic about war. Not now.”
This is why it is common to say “there is no winning a nuclear war.” Not because it is literally impossible to emerge from a nuclear war with possession of the objects in contention, but because victory cannot protect you. Your armies may be advancing on the defenseless capital of the enemy, your air force may be striking targets with impunity, and the enemy’s fleet be entirely captured or destroyed, but no amount of success will impact the defeated enemy’s ability to inflict unimaginable suffering on you. This is very different from conventional war, where the victor has very little to fear from the victim. An army cannot sack cities until the army opposing it is defeated; there is no equivalent way to oppose or deny a nuclear strike.
“It is not ‘overkill’ that is new; the American army surely had enough 30 caliber bullets to kill everybody in the world in 1945, or if it did not it could have bought them without any strain. What is new is plain ‘kill’—the idea that major war might be just a contest in the killing of countries, or not even a contest but just two parallel exercises in devastation.”
To rephrase, that the world has enough weaponry to destroy the population several times over is nothing new and is not what makes nuclear arms revolutionary. It was not that killing was impossible, but that it previously served no purpose. In order to reach the point that one could inflict death en masse upon the enemy, one was positioned to extract anything else that might be desired. With an enemy’s armies defeated, there was no need to resort to bayonets and 30 caliber rounds to force compliance. Nuclear weapons allow states to kill one another’s peoples without achieving victory militarily. Rather than using victory in battle and the implicit threat of continued suffering to obtain concessions, nuclear weapons allow states to inflict suffering directly.
Schelling argues that war has changed to deemphasize contests of force-of armies, navies, and air forces clashing for superiority, in favor of a conflict of terror that is disconnected from success or failure in the conventional realm. Nuclear weapons allow states to skip the process of defeating one another and start with the killing. And they allow a state to do so with a single decision that is executed all at once. Thus, “If people would rather fight a dirty war than lose a clean one, the war will be between nations and not just between governments. If people have an influence on whether the war is continued or on the terms of a truce, making the war hurt people serves a purpose. It is a dirty purpose, but war itself is often about something dirty…To hope that they would be fought cleanly with no violence to people would be a little like hoping for a clean race riot.”
From this, the logic of nuclear war is clear. Nations and the people that compose them would rather resort to a contest of killing than place itself at the mercy of its enemy. Were there no circumstances that any country would use nuclear weapons, they would be irrelevant. However, since there is both the means and will to carry out a killing-contest, nuclear weapons give each state the option to nullify the conventional balance by escalation. The consequence is that states can no longer be induced to war with the promise of victory because no matter their battlefield success, an enemy cannot be rendered so helpless as to be unable to give the order to begin nuclear war, which is a disaster of sufficient scale to nullify any gains made by force of arms.
This is a principle shared with theorists who hold with the strategy of Pure MAD. In their view, this danger is sufficient to preclude war between the great powers. While the strategy of Manipulation of Risk agrees that nuclear weapons have made it impossible for one nuclear power to win a war against another, it argues that the very danger of nuclear war can be used to compel or deter a rival. Pure MAD holds that no one would rationally start a nuclear war, and therefore there is stability, whether recognized or not. Manipulation of Risk agrees, but says a nuclear war does not need to be started rationally or even deliberately. The speed and irrevocability of the decision create the danger that a moment’s misperception or irrationality lead directly to an exchange. It is the use of this risk to apply pressure that constitutes the strategy of Manipulation of Risk.
Brinkmanship
To illustrate, Schelling proposes the hypothetical situation that nuclear threats were inherently believable, ie, once made, they were binding and irrevocable. “If we could really make it believed that we would launch general war for every minor infraction of any code of etiquette that we wanted to publish for the Soviet bloc, and if there were high probability that the leaders in the Kremlin knew where their interests lay and would not destroy their own country out of sheer obstinacy, we could threaten anything we wanted to. We could lay down the rules and announce that if they broke any one of them we would inflict the nuclear equivalent of the Wrath of God. The fact that the flood would engulf us, too, is relevant to whether or not the Russians would believe us; but if we could make them believe us, the fact that we would suffer too might provide them little consolation.”
Schelling argues that successful use of brinkmanship does not require any kind of superiority in terms of nuclear arms. If we said to the Soviets “if you don’t withdraw from Poland, we will begin a nuclear exchange” and our word was unbreakable, the Soviets would have to choose between giving in and enduring a nuclear exchange that may as well be the Wrath of God. That both sides would be annihilated is no consolation to the Russians—the ability to retaliate and destroy the other does not prevent or even mitigate one’s own destruction. What matters is that the Russians are the only ones able to avoid the conflagration. In this scenario, with the Americans magically bound to fulfill the promise of their threat, the Russians have the ultimate choice.
Schelling—by this example—demonstrates the utility in ceding the initiative in a crisis, to force the other side to be the last to act. When the strategy of coercion is based on threatening mutual disaster, it is incredibly valuable to foist upon the enemy the onus to make the final decision to accept disaster or avert it.
“This is why Gandhi could stop trains by encouraging his followers to lie down on the tracks, and why construction-site integrationists could stop trucks and bulldozers by the same tactic; if a bulldozer can stop more quickly than a prostrate man can get out of its way, the threat becomes fully credible at the point when only the operator of the bulldozer can avert the bloodshed.”
This is well and good in the realm of theory and for illustrating a principle, but in reality, there is no way for the Soviets to have understood an American threat of nuclear war as binding. A state may break its word at any time, particularly when the alternative would oblige it to accept nuclear war, the Wrath of God. “It would be hard to keep the Soviets from expecting that we would think it over once more and find a way to give them what my children call ‘one more chance.’ Just saying so won't do it… What we have to do is to get ourselves into a position where we cannot fail to react as we said we would—where we just cannot help it—or where we would be obliged by some overwhelming cost of not reacting in the manner we had declared.”
It is this that is the central challenge in the strategy of Manipulation of Risk, the shifting of the onus to avoid disaster onto the other side. Brinkmanship or the Manipulation of Risk is fundamentally the art of commitment. It relies on demonstrating to the other side that you can no longer back down because of some unacceptable costs it involves and so they had better back down before things get out of hand. A mechanism can be reduced to “I can’t back down, therefore you must or everyone suffers the Wrath of God.” This can, of course, be said, but that is far from sufficient for it to be believed. Public statements have some effect in that they place a state’s credibility on the line. But loss of the credibility of words can be more easily borne than nuclear war. It is frequently necessary to demonstrate commitment by concrete action.
“At law there is a doctrine of the ‘last clear chance.’ It recognizes that, in the events leading up to an accident, there was some point prior to which either party could avert collision, some point after which neither could, and very likely a period between when one party could still control events but the other was helpless to turn aside or stop. The one that had the ‘last clear chance’ to avert collision is held responsible. In strategy when both parties abhor collision the advantage goes often to the one who arranges the status quo in his favor and leaves to the other the ‘last clear chance’ to stop or turn aside.”
In the case of the Cuban Missile Crisis, the United States was able to demonstrate its commitment through the implementation of a blockade. Removing the blockade or allowing Soviet ships to pass would have incurred a significant loss of face and credibility. This forced the choice onto the Soviets as to whether they were willing to give up or accept the collision that the Americans had backed themselves into. As Schelling puts it, “Often we must maneuver into a position where we no longer have much choice left. This is the old business of burning bridges… The commitment process on which all American overseas deterrence depends—and on which all confidence within the alliance depends—is a process of surrendering and destroying options that we might have been expected to find too attractive in an emergency.” By backing oneself into a corner, the danger of disaster increases, but the onus is placed on your opponent to act to reduce the danger they too are in. They may themselves also destroy attractive options, but in doing so they must decide how much risk of disaster they are willing to bear over this particular issue. If you have acted first, this thorny question is foisted on your opponent. In this way, the objective is to act first in ensuring your opponent is forced to act last, and have the last clear chance fall to them.
Compellence
Manipulation of Risk is therefore much easier to use in deterrence than in compellence. It is much easier to contrive a situation where you are in the way and your opponent is the only one with a chance to avoid the collision and ensuing disaster. However, to attempt to compel an opponent requires creating a danger that you cannot then remove and persists until they comply. “The threat that compels rather than deters often requires that the punishment be administered until the other acts, rather than if he acts.” This is far more difficult to contrive than punishment contingent on a certain action. As Schelling puts it, it is clearer to say no “no further” than “go back.” The former has little nuance, whereas the latter invites questions, such as “how far?” This opacity is undesirable in a crisis, as the strategy relies on an opponent having a clear idea as to what they must do to remove the risk of disaster. To illustrate the point, Schelling puts forward the complete conceptual framework for successful compellence.
“The ideal compellent action would be one that, once initiated, causes minimal harm if compliance is forthcoming and great harm if compliance is not forthcoming… and cannot be stopped by the party that started it but automatically stops upon compliance, with all this fully understood by the adversary… His is then the ‘last clear chance’ to avert the harm or catastrophe; and it would not even matter which of the two most feared the consequences as long as the adversary knew that only he, by complying, could avert them.”
In the concept of Manipulation of Risk, the harm that is forthcoming is the shared risk of nuclear war. It is not necessarily actual, immediate harm, but the risk of complete catastrophe that has the compellent effect. In utilizing this strategy, you give your opponent the choice between enduring the risk of total disaster or conceding the issue and rescuing the both of you from catastrophe. Successful utilization of the strategy requires putting the onus on your opponent to make the choice between disaster and concession. If they believe you can resolve the situation, they will be inclined to wait for you to do so. Thus, in using this strategy, a state needs to deliberately remove options by which it might escape, acting first to bind itself, leaving the “last clear chance” to the enemy.
This strategy does not involve threatening to initiate a nuclear exchange, as such a threat would be unbelievable. Instead, in Schelling’s words, “It involves setting afoot an activity that may get out of hand, initiating a process that carries some risk of unintended disaster. The risk is intended, but not the disaster.” The threat or harm imposed is the risk that circumstances may contrive so that neither side is capable of preventing a nuclear war. Schelling makes an argument similar to Clausewitz’s characterization of the “fog of war” to explain how a nuclear exchange might occur accidentally. “Violence, especially war, is a confused and uncertain activity, highly unpredictable, depending on decisions made by fallible human beings organized into imperfect governments, depending on fallible communications and warning systems and on the untested performance of people and equipment.” It is not necessary for it to be likely that a breakdown in this system (or through Clausewitz’s “friction”) will cause a nuclear launch to occur for it to have value as a threat. So long as the possibility exists and we remain aware of what factors may make one more likely to occur, the specific probability remaining unknown is not particularly consequential. When the consequences are Armageddon, any increase in risk is severely unappealing. The uncertainty involved is therefore a sufficient threat.
It can be hard to imagine a series of circumstances in which superpowers would be unable to prevent a nuclear war. A cold view would assume that states would settle things rather than run a serious risk of annihilation. However, states may overestimate their ability to back down and find themselves to be losing control over the pace of the confrontation. Schelling illustrates the danger that unintentional escalation poses. “One never quite knows what exits will begin to look cowardly to oneself or to the bystanders or to one's adversary. It would be possible to get into a situation in which either side felt that to yield now would create such an asymmetrical situation, would be such a gratuitous act of surrender, that whoever backed down could not persuade anybody that he wouldn't yield again tomorrow and the day after.”
By engaging in brinkmanship, states court not so much the direct danger of nuclear war, but of accidentally crossing an unseen point-of-no-return which ultimately leads to one. The uncertainty makes the threat credible. “Whether it is better to be red than dead is hardly worth arguing about; it is not a choice that has arisen for us or has seemed about to arise in the nuclear era. The questions that do arise involve degrees of risk— what risk is worth taking, and how to evaluate the risk in-volved in a course of action. The perils that countries face are not as straightforward as suicide, but more like Russian roulette.” The strategy of Manipulation of Risk means creating a game of Russian roulette that your opponent has to pay a price to leave.
An Accidental Apocalypse
It is the elements of chance and uncertainty that make the strategy of Manipulation of Risk credible. It does not require a government to coolly-decide that fighting a war likely to destroy the bulk of its population is the right decision. It requires only that such a war may occur beyond the capabilities of rational heads to prevent. “It is the essence of a crisis that the participants are not fully in control of events; they take steps and make decisions that raise or lower the danger, but in a realm of risk and uncertainty… We often talk as though a ‘deterrent threat’ was a credible threat to launch a disastrous war coolly and deliberately in response to some enemy transgression… They do not need to depend on a willingness to commit anything like suicide in the face of a challenge. A response that carries some risk of war can be plausible, even reasonable, at a time when a final, ultimate decision to have a general war would be implausible or unreasonable. A country can threaten to stumble into a war even if it cannot credibly threaten to invite one.”
This is the foundation of the strategy. It is not necessary for anyone to choose nuclear war and the mutual annihilation it entails for the conflagration to occur nevertheless. Both sides can create a risk that—regardless of their intentions—a nuclear exchange will occur. Not that they will choose to do so, but that they will be unable to prevent it. Manipulation or brinkmanship refers to using this danger to extract concessions from a rival.
Schelling uses the example of a modified chess game to demonstrate how risk can be used as a compellent and how nuclear weapons have changed the calculus from conventional war. “A chess game can end in win, lose, or draw. Let's change the game by adding a fourth outcome called ‘disaster.’ If ‘disaster’ occurs, a heavy fine is levied on both players, so that each is worse off than if he had simply lost the game.” In this scenario, if “disaster” occurs automatically when a special rule is broken, then the rule functions as a law. Neither player will ever violate it. “Prohibitive penalties imposed on deliberate actions are equivalent to ordinary rules.” In nuclear terms, this is as though there were a tripwire placed across a frontier that automatically initiated an exchange. Assuming this were able to be done irrevocably and without danger of error, this kind of tripwire would preclude war. It would make nuclear devastation inherent and aggression tantamount to suicide and would therefore function as though crossing that frontier was physically impossible. In reality, however, there are few actions that would engender an inherent nuclear response. There are, however, many that would create a risk of one through the aforementioned mechanisms of uncertainty and uncontrolled escalation. To account for this, a further modification to the game is needed.
“Let us not have disaster occur automatically… Instead, when that occurs, the referee rolls a die. If an ace comes up the game is over and both players are scored with disaster, but if any other number appears the play goes on. If [the rule is still being violated] the dice are rolled again, and so on… In this way uncertainty imports tactics of intimidation into the game. One can incur a moderate probability of disaster, sharing it with his adversary, as a deterrent or compellent device, where one could not take, or persuasively threaten to take, a deliberate last clear step into certain disaster.”
This translates clearly into the logic of nuclear strategy. Choosing to use nuclear weapons is impossible to credibly threaten under the shadow of MAD. But introducing and increasing the possibility of disaster until one side would prefer to give in rather than continue to bear the risk is credible because the threat is of unintentional escalation, rather than a deliberate last step to suicidal behavior. You do not need to convince your opponent that you are willing to commit suicide, merely that your willingness to endure the risk of death is greater than they are comfortable enduring for the price you demand.
While it was never possible for the United States to get the Russians to remove missiles from Cuba by threatening to begin a nuclear war, it could implement a blockade of Cuba, forcing the Soviets to either endure the increased danger of accidental escalation derived from running the blockade or choosing to instead back down. In the Cuban Missile Crisis, the Soviet Union had to choose what cost it would rather bear: removing their missiles from Cuba or accepting a greater chance of disaster.
In the example of the chess match that Schelling uses, one player could deliberately break the rule to place pressure on the other opponent to come to a negotiated settlement. They might be able to convince the other to accept a draw rather than wait until disaster comes from their continued violation of the rule. The other player could refuse, but it would then become not a conventional chess match, but a test of will and commitment. The same is true in a crisis between nuclear states. By raising the risk of nuclear conflagration, states may pressure one another. Nuclear weapons are not used like other weapons, but constitute latent violence, the threat of which is no less potent because it harms both parties. By creating situations in which the risk of disaster is contingent on our opponent doing or not-doing things, nuclear weapons may be used to serve both a compellent and deterrent purpose.
Conclusion
I have chosen to quote extensively from Schelling because, unlike many nuclear theorists, he puts matters in language that is both vivid and plain. By identifying the factors that make nuclear weapons revolutionary to the conduct of both war and international relations, Schelling uses game theory to demonstrate how the risk of disaster may be used to gain an advantage. Schelling rejects Nuclear Superiority, accepting that MAD is inescapable. However, he demonstrates that MAD is insufficient to prevent competition between nuclear powers. A strategy of Pure MAD relies on the kind of certainty that Schelling argues is absent in a real crisis. If it were a matter of absolute fact that crossing the Inner German Border would cause an American countervalue strike on the Soviet Union, NATO may as well have relied on Pure MAD during the Cold War. However, the inability to convincingly promise to coldly take a suicidal course of action made this strategy infeasible.
Counterforce Warfighting can be viewed through Schelling’s lens of Manipulation of Risk as a course of action that applies maximum pressure for both sides to come to a resolution because of the incredibly high risk of uncontrolled escalation in a counterforce exchange. Manipulation of Risk re-contextualizes an escalation ladder not as places at which to have dominance, but as varying degrees of risk. This is significant because a state need not believe it has an advantage at the next level of escalation to pursue it. In the framework of Manipulation of Risk, the important factor of each level is not which side has an advantage, but the pressure it puts on both sides to resolve the crisis before escalation becomes uncontrolled. For example, even if a state is conventionally outmatched, it may still escalate to open conventional warfare in the hope that its opponent will prefer to come to a settlement rather than endure the risk of escalation that fighting a conventional war would engender, even if that war was ultimately victorious. As in the example of the modified chess match, it becomes a contest of will.
Ultimately, the strategy of Manipulation of Risk distinguishes itself in being a method of utilizing the conditions of MAD to compel or deter. It argues that nuclear weapons have changed the face of war by making annihilation near-instant, and having the victor and vanquished equally vulnerable to destruction. What’s more, it recognizes that under conditions of MAD, starting a nuclear war is a suicidal act. To do so deliberately is scarcely believable. But the role of chance and uncertainty in war and crisis make it possible that it will happen nonetheless. From these observations comes the conclusion that uncertainty and the danger it entails are central to both deterrence and compellence. By manipulating the level of risk both parties are exposed to and by removing options of escape, one side in a crisis may force the other to concede in exchange for nothing more than the removal of the mutual danger.