Coercion theory is one of the most fully developed bodies of theory in the social sciences, one that has advanced the field of national security by illuminating the logic that underlies threats, violence, and war. Coercion has a long history, of course, but its manifestation as a sustained point of focus in contemporary social science may arguably be traced to Thomas Crombie Schelling’s 1966 book, Arms and Influence.1 An economist by training, Schelling developed his early work at a time when debates over nuclear strategy dominated the landscape, although his work is applicable to all varieties of force.2 Over the past 50 years, scholars have embraced and built upon Schelling’s work, using it to shed light on an array of issues in defense and national security.3 If coercion theory is understood in the academy, however, it is less well understood by practitioners, especially those in the military. This is a problem for civil-military communication, and, more generally, for national strategy and military outcomes.
On those occasions when they encounter coercion theory, military practitioners are often instinctively wary of it. In general, they tend to be skeptical that theories produced by academics can help them understand war, which they believe is their dominion. After all, academics dwell in the realm of the abstract and the theoretical while military professionals dwell in the realm of the concrete and the real. Moreover, military professionals are not entirely comfortable with violence as a bargaining process. One does not, they believe, “bargain” with one’s enemies — one fights them. Nor do they find congenial the idea that coercion requires the cooperation of the enemy. Even if one explains that this is by no means happy cooperation, it rankles nonetheless because they (especially those in the U.S. military) believe they should own the initiative and maintain dominance across the full spectrum of conflict at all times.4
The word “coercion” itself sits uneasily with military professionals. It has overtones of blackmail and manipulation, which are anathema to their self-identity. In general, they also do not take readily to Schelling’s emphasis on threats. While they fully understand deterrence, they may draw back from the idea that they are in the business of “threatening” others (and sometimes making those threats credible by actions) in order to deter and compel. For Schelling, conflicts involving coercion unfold through a kind of violent communication about intentions and commitment. Understandably, few military officers see killing and dying as just a form of communication.5
Schelling’s phrase “brute force” receives no easier reception. Here, the problem is rather easy to understand since the phrase itself quickly conjures up images of indiscriminate and primal violence — a kind of warfighting that lies in direct opposition to the institutional identity of modern military professionals.
Military culture and identity thus prevent many practitioners from embracing a body of theory that offers them crucial insights into the nature and practice of their own profession. Understanding coercion is central to developing and implementing sound strategy. When practitioners (either military or civilian) remain innocent of or resistant to coercion theory, they fail to grasp the logic that animates their own decisions and strategies and to understand the ways that their enemies may resist and thwart them, even when those enemies are materially weaker. It also causes them to misunderstand the history of much of the U.S. military experience, especially since World War II. Most importantly, it causes practitioners to systematically overestimate their own chances for quick and low-cost victories.
Another stumbling block to the full use of coercion theory by practitioners is that scholars who write about the topic do not always use consistent terms, definitions, and categories. In some cases, the failure of contemporary scholars to invest sufficient time in Schelling’s original texts have resulted in errors that have muddied the theoretical waters. It does not help that Schelling himself wrote in an idiosyncratic way that is not readily grasped by those who do not have the luxury of devoting extensive periods of time to his work. Students in civilian programs of extended duration have this luxury. Students in rather more hurried professional military education programs do not.
This leaves room for substantial misunderstanding and miscommunication. For many years, for instance, the foundational doctrine of the Joint Chiefs of Staff has used the language of coercion theory — but in ways that have varied from problematic to simply wrong. While the authors of contemporary joint doctrine recognize the distinction between Schelling’s fundamental categories of “coercion” and “brute force,” they do not use these same terms, or trace the intellectual provenance of the ideas. Joint Publication 1, the Doctrine for the Armed Forces of the United States, explains that there are “two fundamental strategies” in the use of military force: “annihilation” and “erosion.”6 The first term corresponds roughly to Schelling’s “brute force,” and the second corresponds roughly to “coercion,” although the parallels in each case are imperfect. Annihilation seeks “to make the enemy helpless to resist us, by physically destroying his military capabilities. …It requires the enemy’s incapacitation as a viable military force.” Erosion, by contrast, seeks “to convince the enemy that accepting our terms will be less painful than continuing to aggress or resist.” Its goal is to erode “the enemy leadership’s or the enemy society’s political will.” With erosion, military force is employed “to raise the costs of resistance higher than the enemy is willing to pay.” The authors of the doctrine argue that erosion is used “in pursuit of limited political goals that we believe the enemy leadership will ultimately be willing to accept.”7
The latter sentence is especially problematic, revealing the root of several U.S. failures. A coercer may perceive that a contested stake is “limited,” but the state being coerced (i.e., the target state) may not see it that way at all. U.S. efforts to coerce the North Vietnamese in the 1960s were thwarted when the latter turned the tables and ultimately coerced the United States by raising the price of victory higher than the Americans were willing to pay. To civilian ears, “erosion” sounds vague and underspecified, while “annihilation” suggests something more dramatic than dispensing with the need for enemy cooperation. Both terms leave room for miscommunication.
Military culture and identity thus prevent many practitioners from embracing a body of theory that offers them crucial insights into the nature and practice of their own profession.
A new (as yet unpublished) version of Joint Publication 1.0 not only confuses Schelling’s categories, but uses his term “compellence” (which is a subset of coercion), to describe what Schelling called “brute force.” If adopted, this publication will flip the meaning of two of Schelling’s most important terms, putting the military’s doctrinal categories dramatically at odds with the civilian literature on coercion theory. Since doctrine ought to be a starting place for military thought and a mechanism for civil-military communication, this revision could have serious ramifications.8 On the other hand, the recent Joint Doctrine Note 2-19 (published in December 2019) uses the language of coercion theory accurately and is thus a welcome change. While a joint doctrine note does not supersede existing doctrine, it “facilitates information sharing on problems and potential solutions to support formal doctrine development and revision.”9 This new publication moves in the right direction.
To help address this ongoing terminological confusion, this essay seeks to clarify, systematize, and make readily accessible the language of coercion theory. Drawing on Schelling’s original texts, I explain the categories he used and the terms he developed. Throughout, I emphasize Schelling’s fundamental point: Coercion is difficult, even for actors who hold a preponderance of coercive leverage in a given situation. Schelling and those who have further developed his ideas worked hard to demonstrate that coercion is anything but simple, straightforward, or formulaic. It is not a silver bullet. Indeed, much of the motivation of Schelling’s 1966 effort was to explain the complexity of coercion and to provide insights into the challenges one should expect when employing it.
Finally, this essay seeks to draw the attention of practitioners in particular to a body of theoretical work that is exceptionally useful across the board — and not least in an era of renewed international competition. As the United States leaves behind its “long wars” and turns its attention back to near-peer competition and nuclear rivalry, the literature on coercion theory will help strategists understand and craft intelligent responses to current and future political challenges.10 Understanding the language of coercion theory will also help practitioners identify and distinguish the situations in which adversaries seek gains by faits accomplis and by working around red lines. The Joint Staff have made “deterrence theory” a special area of emphasis for professional military education in the immediate future,11 but the utility of such a move will rest on a shared understanding of terms and concepts among scholars and practitioners.
Threats, Influence, and Behavior
Schelling was interested in the ability of military power to “hurt” the enemy — to inflict pain or punishment — and the inherent “bargaining power” this confers. Coercion is about future pain, about structuring the enemy’s incentives so that he behaves in a particular way. It manipulates the power to hurt and involves making a threat to do something one has not yet done. The coercer forces another actor to calculate, to decide — based on his own interests and position — whether or not to resist the threat being made.
Observing human behavior, Schelling recognized that humans use threats constantly to shape the behavior of others. We do this for a range of reasons. Anyone who has raised a child has learned quickly how to influence that child’s choices: A parent may issue a threat in order to keep a child from harm, or to set boundaries to help prepare the child for civil interaction with others. As children grow older, the content of those behavior-influencing threats must change in order to reflect the child’s level of comprehension and new interests and the parent’s changing leverage over the child’s behavior.12
Similarly, if we wish to keep our homes safe from intruders, we may install a security system and then post a sign advertising it. A potential intruder is alerted to the negative consequences that will greet any attempt to enter without permission. This action is meant to deter — to prevent someone from taking an action he otherwise might take. But threats can be used to compel actions as well as deter them. In the film The Godfather, Don Corleone promises to influence the decisions of the head of a film studio, stating, “I’m gonna make him an offer he can’t refuse.” If the recipient of the threat refuses to accept the “offer” (which is actually a demand), then harm will follow. The coercive threat is designed to compel an individual to do something he would prefer not to do. If the threat derives from a source known to be willing and able to produce harm, then it is credible and must be taken seriously.
An actor being coerced (i.e., the target state) must assess its own willingness and ability to endure pain, as well as the credibility of the adversary’s threat. “The power to hurt,” Schelling explained, “is a kind of bargaining power, not easy to use, but used often.”13 Even great powers possessing high levels of coercive leverage over others find that target states can resist in unexpected ways, making the line between the application of power and the achievement of a desired outcome anything but direct and straightforward. By its nature, coercion requires a decision by the actor being coerced, thus placing the outcome in the actor’s hands. This is what makes coercion difficult and complex — and distinct from a more direct use of power that Schelling defined as “brute force,” wherein there is no need for a decision by the target state because power is imposed directly in such a way as to obviate choice.
Political actors use coercive threats all the time to protect themselves and to preserve and promote their interests. Schelling observed,
The bargaining power that comes from the physical harm a nation can do another nation is reflected in notions like deterrence, retaliation, reprisal, terrorism, and wars of nerve, nuclear blackmail, armistice and surrender, as well as in reciprocal efforts to restrain that harm in the treatment of prisoners, in the limitation of war, and in the regulation of armaments.14
The power to hurt confers bargaining power, Schelling insisted. The willingness to exploit it is diplomacy — “vicious diplomacy, but diplomacy.”15
Schelling explained that the use of “the power to hurt” operates like blackmail in that it exploits an enemy’s fears and needs. The power to hurt is usually most successful when it is held in reserve. Hostages, for instance, are taken and held for coercive purposes. Those taking the hostages seek to make another actor give up something — money, political prisoners, etc. But if they kill the hostage, the other actor no longer has an incentive to concede and coercive hostage-taking fails. Any coercive act that kills the hostage, as it were, reduces its own effectiveness. Hostages, Schelling argued, “represent the power to hurt in its purest form.”16
Deterrence, Compellence, and Brute Force: Definitions
In Schelling’s taxonomy, “coercion” is an overarching category encompassing both “deterrence” and “compellence.” The word “deterrence” was in common usage when he wrote Arms and Influence.17 The term “compellence” he coined himself, after rejecting several alternatives. Since 1966, it has become part of the lexicon of security studies.18 (Schelling admired, but chose not to select, the terms “dissuasion” and “persuasion” that J. David Singer had used several years earlier to describe a similar idea.)19
Deterrence involves a threat to keep an adversary “from starting something,” or “to prevent [an adversary] from action by fear of consequences.” Compellence is “a threat intended to make an adversary do something.”20 In deterrence, the punishment will be imposed if the adversary acts; in compellence, the punishment is usually imposed until the adversary acts.21 As noted, the central characteristic of both forms of coercion is that they depend, ultimately, on cooperation by the party receiving the threat. This is by no means friendly cooperation, but it is cooperation nonetheless. Compellence can be used in peacetime and in wartime, the former use being referred to generally as coercive diplomacy.22
Alexander Downes describes coercion as “the art of manipulating costs and benefits to affect the behavior of an actor.” Explaining its two forms, he writes, “Deterrence consists of threats of force designed to persuade a target to refrain from taking a particular action. Compellence, by contrast, utilizes force — or threats of force — to propel a target to take an action, or to stop taking an action it has already started.”23 The United States, he notes, is one of the most frequent users of compellent threats.24 Examples abound. Sometimes they involve the use (or threatened use) of U.S. troops, and sometimes they do not. But military power always stands in the background. In one notable example from 1956, President Dwight Eisenhower used economic and diplomatic threats to compel the British, French, and Israelis to cease the military operations they had begun in response to Egypt nationalizing the Suez Canal. More recently, the administration of President Donald Trump used a threat of economic sanctions to try to compel the Mexican government to more aggressively discourage population flows across the U.S. border.25
[T]he central characteristic of both forms of coercion is that they depend, ultimately, on cooperation by the party receiving the threat. This is by no means friendly cooperation, but it is cooperation nonetheless.
As a major power seeking to maintain the existing international structure, the United States possesses coercive leverage and uses it over other states. A sense of its own power, combined with a desire to use that power to solve complex problems with minimum trouble and expenditure, has inclined American decision-makers to look to coercion repeatedly. But coercion, especially compellence, is difficult, and provides no guarantee of success — not even to very powerful actors.26 Reviewing five empirical studies of coercion, Downes concludes that compellence succeeds only about 35 percent of the time.27
Karl Mueller adds further texture to the definition of deterrence: “causing someone not to do something because they expect or fear that they will be worse off if they do it than if they do not.” He stresses that deterrence “happens in the mind of the potential aggressor.”28 Moreover, he observes that because the future is uncertain and difficult to predict, states often make choices based on misperception — and those choices sometimes involve entering into ill-advised wars. 29
Robert Jervis’s work elaborates in detail how and why two actors may perceive (and respond to) the same situation very differently.30 He insists that decision-makers and analysts ask some key questions about coercion, including: Do the adversaries assess the stakes similarly? Do they view the credibility of threats the same way? Are both sides equally concerned about reputation? Jervis points out that cultural norms and expectations vary, perspectives differ, and domestic political imperatives may override other pressures. Opportunities for misperception and miscommunication abound.31
In the first pages of Arms and Influence, Schelling took pains to distinguish coercion from “brute force,” which, he argues, is used “without [reliance on] persuasion or intimidation.” Later in the book, he uses the phrase “forcible action” as a synonym, a preferable phrase because it does not carry so much linguistic baggage.32 With brute force/forcible action there is no need for the opponent’s cooperation. The actor simply takes what it wants. Schelling compared this to the way a tank or a bulldozer can simply “force its way, regardless of others’ interests.” Brute force, he explained, “is concerned with enemy strengths, not enemy interests.”33 Adding detail to the idea, Schelling wrote: “Forcibly a country can repel and expel, penetrate and occupy, seize, exterminate, disarm, and disable, confine, deny access, and directly frustrate intrusion or attack.” Brute force is directly measurable and is usually measured relative to an enemy’s strength, the one opposing the other.34 Unlike coercion, brute force/forcible action does not place the outcome in the hands of the adversary.35
Land power plays a particularly important role in the realm of brute force/forcible action. If an army can control the situation on the ground, it ultimately can dispense with seeking an adversary’s cooperation. If it is strong enough, an army can remove an existing government and replace it with one that is more congenial to the political authorities it serves, and then control the aftermath. This is demanding and costly, and is therefore typically reserved for extreme situations. Of course, armies have many roles short of using brute force/forcible action. Indeed, their existence affords decision-makers a critical tool for deterrence and compellence. However, their role as agents of brute force/forcible action is crucially important. It is thus vital for students of military power to understand the logical distinction between Schelling’s primary categories of coercion and brute force/forcible action, and the strengths and weaknesses of both. And it is equally important for civilian authorities to understand the ways in which different military instruments relate to them.
Methods of Coercion
Actors can deter by threat of punishment or by threat of denial. The meaning of the first is easy to discern: The coercing state threatens to impose pain on the target state for failure to comply with the coercer’s demand. This might involve an air strike on a location valued by the target state or a naval blockade to deny it crucial resources. Perhaps the most familiar version of deterrence since the advent of the Cold War is nuclear deterrence. Nuclear weapons certainly are not necessary to inflict punishment, but, as Schelling pointed out, no weapon has ever surpassed nuclear weapons for threatening severe pain.
An actor can also deter by (threat of) denial. This route seeks to dissuade an adversary by convincing him that any military campaign he may launch will fail militarily because the coercer will deny the ability to complete the action successfully.36 Richard K. Betts offers a definition of deterrence notable for its lucidity and conciseness:
Deterrence is a strategy for combining two competing goals: countering an enemy and avoiding war. Academics have explored countless variations on that theme, but the basic concept is quite simple: an enemy will not strike if it knows the defender can defeat the attack, or can inflict unacceptable damage in retaliation.
Here Betts refers first to deterrence by threat of denial (“can defeat the attack”) and then to deterrence by threat of punishment (“can inflict unacceptable damage in retaliation”).37 It should be noted here that deterrence and compellence can be used by small states as well as large states. It is an actor’s will and determination (to gain or hold a stake), rather than its raw power — defined in physical, military, or economic terms — that usually dictates the outcome in a coercive interaction.38
Deterrence by threat of denial and deterrence by threat of punishment are sometimes linked to one another. For instance, a state may imply to an enemy that if denial fails, it will resort to punishment. In the Cold War era, deterring a Soviet grab for Western Europe involved both threats of denial and threats of punishment. Because NATO had fewer ground forces than the Warsaw Pact, both denial and punishment rested heavily on U.S.-funded and controlled nuclear forces.39 Today, NATO ground forces once again are engaged in exercises designed to sharpen their conventional military skills, and thus to deter Russian action by threat of denial. Still standing in the background, though, is the potential threat of punishment posed by U.S. nuclear weapons.
It is an actor’s will and determination (to gain or hold a stake), rather than its raw power — defined in physical, military, or economic terms — that usually dictates the outcome in a coercive interaction.
Punishment and denial come into play in compellence as well. Airpower, particularly 21st-century airpower, is often viewed as a coercive tool of choice since it is easily scaled and tailored, and can be used by some air forces, including the U.S. Air Force, with high precision against discrete targets. Air strikes can, at once, inflict pain and signal an intention to inflict future pain. In a denial role, they can interdict military supplies and destroy key military infrastructure, preventing an adversary from fighting effectively.
For the practitioner, it is important to understand the fundamental logic being employed so that the strengths and weaknesses of the ways and means utilized will be clear, thus reducing the likelihood of frustration or surprise when the enemy seeks to thwart or resist those ways and means. In 1990, for instance, Operation Desert Shield sent U.S. troops to the Middle East to deter an Iraqi incursion into Saudi Arabia. In January 1991, air strikes against Iraq were used largely to compel Iraq to pull out of Kuwait, while air strikes against Kuwait served mainly to undermine Iraqi capabilities in that theater. If the Iraqis did not concede, then the United States and its allies were committed to an invasion of Kuwait designed to deny Iraq its gains, and, ultimately, to drive the Iraqi army from Kuwait through brute force/forcible action.40
Characteristics of Coercive Threats: Distinctions and Requirements
The two main categories of coercion — deterrence and compellence — are distinct in their nature and requirements. When an actor refrains from a behavior, one does not and cannot know the specific reason (or reasons) for that choice. Refraining from an action can be attributed to causes other than the specific deterrent threat. The enemy may never have intended to attack in the first place, for example, in which case the deterrent threat is not what prevented the attack. In fact, it is never clear whether the absence of an attack is due to an enemy giving in to a deterrent threat. This ambiguity enables enemies who have been deterred to save face.
Because compellence is active in ways that deterrence is not — the target state must perform an act rather than simply refrain from one — it is clear to all when compellence is successful. Moreover, the actor being compelled is usually being forced into some degree of humiliation. Robert Art explains that compellence requires the target state’s “overt submission.”41 The ability of a powerful state to compel a less powerful one is constrained and complicated by this fact. Coercers, Downes argues, “tend to underestimate the target’s concern for its reputation and thus offer too little compensation to obtain the target’s acquiescence.” He adds, “The target’s fear for its reputation and the challenger’s unwillingness to lower its demands or offer side payment to compensate the target for the damage to its reputation cause the target to resist threats from powerful states.”42 Thus, though the state being coerced may appear to be in a weaker position, this is not true in one important sense: The state being coerced ultimately makes a decision about whether or not to comply — and in that sense it holds the initiative.
As Schelling pointed out, when it comes to timing deterrence can be indefinite while compellence, by contrast, must be definite. Without a deadline, the adversary being compelled has no incentive to act: “If the action carries no deadline it is only a posture, or a ceremony, with no consequences.” But on the other hand, if too little time is given for compliance, then the coercer has put its adversary into an impossible position, virtually ensuring that compellence will fail. Schelling summed this up by stating, “Too little time and compliance becomes impossible; too much time and compliance becomes unnecessary.”43
To compel successfully, a coercer must convey specific information to the actor being coerced. If compliance is being demanded, then how much, and for how long? Moreover, the enemy must believe that the coercer will actually stop the pain if the target state concedes the stake. The promise to cease coercion if the enemy gives in must be believed, just as the threat to continue coercion if the target state withholds the stake must be. When it comes to deterrence, a promise not to attack if the enemy doesn’t invade is easier to believe — after all, there is no attack taking place right now. In deterrence, Schelling observed, “the objective is often communicated by the very preparations that make the threat credible.” By contrast, compellent threats “tend to communicate only the general direction of compliance, and are less likely to be self-limiting, less likely to communicate in the very design of the threat just what, or how much, is demanded.” Offering an example, Schelling focused on the Western military garrison in West Berlin during the Cold War, which, he argued, had an unmistakable deterrent purpose. Were it to intrude into East Berlin, though, “to induce Soviet or German Democratic Forces to give way,” there would be no “obvious interpretation” of “where and how much Soviet and East German forces ought to give way unless the adventure could be invested with some unmistakable goal or limitation — a possibility not easily realized.”44
The centrality of communication means that coercion is heavily dependent on knowledge, and thus on sophisticated intelligence.45 The coercer must understand the target state’s fears, vulnerabilities, and interests — as well as its willingness to endure pain on behalf of those interests. For these reasons, coercion is subject to cultural miscommunication while brute force is not. As Schelling explained,
To exploit a capacity for hurting and inflicting damage one needs to know what an adversary treasures and what scares him, and one needs the adversary to understand what behavior of his will cause the violence to be inflicted, and what will cause it to be withheld. The victim has to know what is wanted, and he may have to be assured of what is not wanted.46
Coercive action often begins with economic action — the freezing of assets, perhaps, or the imposition of sanctions. The goal is to force the target state (or actor) to choose between conceding the disputed stake or suffering future pain that making such a concession would avert. The target state must be convinced that if it resists it will suffer, but if it concedes it will not. If it suffers either way, or if it has already suffered all it can, then it will not concede and coercion will fail.
One can thus see the many formidable challenges facing a coercer. Precision of thought and language can matter greatly in compellence, while a degree of vagueness occasionally can be useful for deterrence.47 A nuanced understanding of the needs, fears, capabilities, interests, and will of the target state is essential. But the coercer must possess self-knowledge as well, including an understanding of the importance of the stake involved, and the likely commitment to it — by policymakers and by the domestic population — over time. And the coercer must be able to articulate the demand in ways the target state can comprehend and comply with. To understand all this is to understand the deeper meaning of Carl von Clausewitz’s insistence on the linkage between war and politics, and the need to recognize the relationship between the stake and the scale of effort required to achieve it. It is also to understand, beyond a superficial level, the meaning of Sun Tzu’s insistence on knowing one’s self, and knowing one’s enemy.
One should note here, too, that democracies engaging in coercion will face a challenge inherent in the structure of their system of governance: Communication is complicated by multiple power centers — built by design to check one another — and myriad interest groups. Indeed, bureaucratic (and organizational) models of decision-making are at the center of many scholars’ critiques of U.S. foreign policy, and deterrence in general.48
Communication by the coercer may be verbal, but it need not be. It can also be delivered through an action itself. Schelling argued:
Unhappily, the power to hurt is often communicated by some performance of it. Whether it is sheer terroristic violence to induce an irrational response, or cool premeditated violence to persuade somebody that you mean it and may do it again, it is not the pain and damage itself but its influence on somebody’s behavior that matters.49
In many instances verbal threats are backed up by actions to ensure that the message is being taken seriously. But the coercer must be sure that language and actions are aligned, clear, and suitably tailored to create the behavior being sought. If miscommunication occurs in any of these realms — due to carelessness, mixed messaging from different groups within the coercing state, or cultural obtuseness — then coercion is likely to fail. Military planners who understand this will have an elevated appreciation for the crucial importance of getting all the details right when they are engaged in the important work of operational design.50 And they will also have a heightened appreciation of the need for a fastidious commitment to inter-agency coordination.
Precision of thought and language can matter greatly in compellence, while a degree of vagueness occasionally can be useful for deterrence.
Credibility, which matters for deterrence and compellence, is neither unvarying nor permanent. Actors place different values on stakes. This means that adversaries will constantly try to calculate each other’s level of interest in and commitment to a given stake. A preponderance of strength does not imply successful coercion. Daniel Byman and Matthew Waxman have observed, “The United States failed to coerce North Vietnam; Russia failed to coerce Chechen guerillas to give up their struggle; Israel has pulled out of Lebanon. These instances … evince the importance of vital, if rather ineffable factors.” They add: “Will and credibility matter as much as, and often more than, the overall balance of forces. At times a coercer may have preponderant power in a general sense but lack specific means to influence an adversary.”51
In the case of deterrence, a nation’s willingness to defend its own sovereign territory is typically clear. Its willingness to defend another’s territory — or to risk drawing pain upon itself for the sake of another — is not as certain. Schelling explained:
To fight abroad is a military act, but to persuade enemies or allies that one would fight abroad, under circumstances of great cost and risk, requires more than a military capability. It requires projecting intentions. It requires having those intentions, even deliberately acquiring them, and communicating them persuasively to make other countries behave.52
During the Cold War, what came to be called “Mutually Assured Destruction” rested on the idea that a Soviet nuclear attack on the United States would be answered by a U.S. nuclear attack on the Soviet Union, and vice versa.53 Whether the United States would have risked a nuclear strike on its own territory in order to protect Paris or Bonn, however, was not nearly so clear. But U.S. statesmen, through words, policies, and actions, sought to convince their Soviet adversary that the threat was real. In other parts of the world, the United States made similar efforts. Describing the Formosa [Taiwan] Resolution passed by the U.S. Congress in 1955, Schelling observed that the resolution “was a ceremony to leave the Chinese and the Russians under no doubt that we [the United States] could not back down from the defense of Formosa without intolerable loss of prestige, reputation, and leadership.” He added, “We were not merely communicating an intention or obligation we already had, but actually enhancing the obligation in the process.”54
Airpower and Coercion
Naval forces have long been in a position to coerce enemies via “gunboat diplomacy,” as well as by threatening or imposing an economic blockade.55 But the advent of airpower quite literally added a whole new dimension to the possibilities for coercion. Three decades after Arms and Influence appeared, Robert Pape built on Schelling’s framework in his book Bombing to Win. He articulated and analyzed four types of coercion that could be carried out by airpower: punishment, denial, risk, and decapitation.56 The first two were clearly familiar to readers of Schelling. Pape’s risk category, however, has not generally been adopted in the literature because it is not sufficiently distinguished from punishment.57
Focusing on leadership as a center of gravity within an enemy state, decapitation seeks to disrupt an enemy’s will and ability to fight by attacking the state’s leader and communication assets.58 It rests on the idea that many warlike states are run by authoritarians — individuals who have highly personalized forms of governance that may be subjected to direct assault. Decapitation, which rose to prominence in the era of precision bombing, informed parts of the coercive air campaign over Iraq in the 1991 Gulf War.
In general, Pape’s work found that punishment, risk, and decapitation all have problems as methods of aerial coercion. Leaders attached to a particular stake are often willing to pass the pain on to their populations in order to protect the survival of the regime. In many instances, local coercive measures — i.e., acts taken on the ground by secret police or other privileged militia groups, such as the SS in World War II — may overwhelm the effects of more remote aerial coercion. Culture and tradition, as in the case of Japan during World War II, may dissuade a population from rising against its leadership, thus thwarting the effects of even overwhelming air strikes.59 Pape found that airpower is most effective as an instrument of denial, working to undermine an adversary’s ability to attain military aims.60
Pape’s argument — that aerial coercion is not simple or easy, and that punishment is less effective than most people expect — was important, and over the years has been influential. But Pape did not look in detail at the role of land power and the way it can work as a central element of denial and an essential component of brute force. His analysis, though important, captures only part of the picture. There is a need for further investigations into the way denial actually works in terms of the interplay between air, sea, and ground forces.
In general, air forces and navies can impose punishment and can aid denial in myriad ways. Navies can prevent an adversary from receiving crucial supplies needed to fight, while air forces can seek to interdict supplies, both strategically and operationally. At a basic level, ground forces in expeditionary campaigns cannot reach their destinations alone: They must be transported to the location where they will fight. Furthermore, they rely on their sister services for a steady supply of the equipment and materiel that allows them to fight. Navies thus seek sufficient control of the sea lanes to maintain routes for transportation and communications.
In land campaigns, ground forces rely on air forces to win and maintain enough control over air space to enable the land battle to be carried on successfully. Since World War I, no industrial nation has been able to fight a peer competitor successfully without the ability to largely control its own airspace and contest enemy airspace. Today, combat forces (of the ground, air, and sea) must also rely on cyber warriors to protect the many systems that enable communication, intelligence, navigation, kinetic action, and situational awareness. Likewise, they rely on space-based assets to enable and facilitate their functioning.61 Securing these systems — which poses an array of new and difficult challenges — has become a high priority for all states that conceive of themselves as major players in the international system. Because of this, these systems — as assets to be protected — now figure, in increasingly important ways, in discussions about deterrence in particular.
Escalation Dominance and the Role of Brute Force
A coercer, when setting out to influence another actor, must have a strategy for escalation in case its initial efforts fail. It must feel confident that it possesses what theorists call “escalation dominance.” Daniel Byman and Matthew Waxman have explained this as, “the ability to increase the threatened costs to the adversary while denying the adversary the opportunity to negate those costs or to counterescalate.” They add, “it is through the parties’ perceptions that the coercer can achieve the escalation dominance that enables coercive strategies to succeed … it requires a preponderance that is relevant to every form of possible escalation: no matter where the adversary chooses to increase pressure, the coercer is always able to overwhelm the adversary in that area.”62
What practitioners must understand is that escalation dominance is not just a matter of having better technology or more resources. War is a contest of wills as much as it is a contest of instruments and materiel. Once an actor has entered into coercive activity he must be prepared to go forward, matching the adversary’s resistance in determination as well as in capability. Again, this requires that the coercer have considerable insight into not only his own commitment to a stake, but his adversary’s as well. And it may require the coercer to climb the escalatory ladder longer than he would have predicted or preferred.
Most practitioners think of nuclear weapons as the pinnacle of the escalatory ladder, and this is certainly true. But coercers possessing nuclear weapons must also have maximum escalatory range in the conventional realm. This is because the stakes in a contest may be very important, but perhaps not important enough for the coercer to credibly contemplate the use of nuclear weapons. If an actor wishes to have dominance in conventional escalation, he must possess land power. The ability of ground armies to land on enemy soil, defeat the adversary’s forces (with the aid of air and naval power) and bodily remove the existing leadership — i.e., the threat of brute force/forcible action — provides a coercive tool that is without parallel in the conventional realm. Byman and Waxman explain, “The possible use of ground forces is a potent threat and, if credible, reinforces other instruments by highlighting the potential for escalation.”63 Regarding the role of land power in the 1999 Operation Allied Force in Kosovo, they argue that “the threat of NATO ground forces — though ambiguous — helped convince [Slobodan] Milosevic to meet NATO demands over Kosovo.”64
[A]ny state that wants to protect and preserve global interests must possess, and be prepared to use, sophisticated forms of expeditionary land power.
Land power is thus the ne plus ultra when it comes to coercion that is below the nuclear threshold.65 It can confer upon an actor the freedom to depart from the constraints and complications of coercion and move into the realm of brute force/forcible action, where the actor takes what he wants without seeking the adversary’s cooperation. This is what Brig. Gen. Huba Wass de Czege meant when he wrote of “vigorous campaigns to force conditions on the enemy regardless of his will.”66
An army is a powerful, indeed indispensable, tool in the tool box of a major power. Although neither coercion nor brute force is ever a silver bullet, when other coercive leverage has failed, the threat of a land invasion — even a vague threat — is sometimes enough to convince an adversary that the game is no longer worth the candle. And if a state with a powerful army (and sufficient resources to sustain it over time) is simply determined to win a stake, regardless of cost, it has the option to shift to brute force/forcible action. Thus, any state that wants to protect and preserve global interests must possess, and be prepared to use, sophisticated forms of expeditionary land power.67
Land power also can allow an actor to set and control the political terms of the post-hostilities phase, whether the circumstances involve enforcing the terms of a negotiated settlement or setting up a new regime after having removed the previous one. The political advantages accruing to a victor can be fully realized only if settlement terms can be enforced, or if a new (imposed) regime takes root in a form that is agreeable to the actor who did the imposing. And, at a basic level, a state that has taken down a previously existing government in a foreign land becomes responsible for the political resolution; for postwar justice (jus post bellum), including postwar political security and stability; and for the care and feeding of the domestic population until a new, functioning structure can be set up. This is an immense task, and an unavoidably expensive one. If an army is used simply to enforce negotiated settlement terms, its responsibilities will be lighter, but they will be significant nonetheless.68
While armies are powerful tools, their use is accompanied by some significant risks and drawbacks, even when the deployment is for something as seemingly straightforward as humanitarian assistance. Moving and using an army is costly in terms of time, treasure, and, sometimes, blood. The use of land power comes with strings attached that do not usually accompany discrete uses of air and naval power in independent coercive actions. An army’s presence on the ground is at once its greatest strength and its greatest weakness. Deploying an army is, first of all, obvious: It signals a commitment that cannot be shrugged off later without humiliation and, perhaps, costs to one’s credibility. The use of an army also does not guarantee success. A determined weaker enemy may be willing to enter an escalatory contest, upping the ante by turning to irregular methods and relying on time (and a high pain threshold) to hold out against a stronger force. Or it may turn to irregular methods once a conventional war has been fought, in order to shift the terms of surrender or alter the postwar political situation.
For all these reasons, an actor contemplating the use of land power must be prepared to commit to the possibility of a campaign that lasts years (or decades) rather than days and months.69 An adversary state will know all this, and will work hard to determine if a threat of land power is being made genuinely and credibly. Thus, those who would deploy armies must face up to the risks involved in doing so, and must be ruthlessly realistic about the demands and costs of such an undertaking.
Persuading decision-makers to fully consider these risks and make crucial calculations can be difficult, however. Political decision-makers generally seek to avoid acknowledging the potential costs of a land campaign — especially the complications of terminating a war and the requirements for transforming what was won by armies into durable and sustainable political gains.70 Politicians will avoid realistic cost estimates because they fear that domestic populations will not want to hear them.71 Meanwhile, military planners will gravitate to the operational details of opening and sustaining campaigns involving land power — they will not be drawn, whether due to natural interest or organizational culture, to the less immediate and perhaps less appealing details of the post-hostilities phases of war. And they may not want to acknowledge fully the asymmetric means and methods that might be employed by an enemy.
Both of these are considerable problems that undermine strategy and war planning. In the United States in particular, they have led to anger directed at the leaders of the U.S. Army and to claims that military professionals are poorly trained in the art of strategy. But the responsibility must be shared by military and civilian decision-makers since, in democracies, both are responsible for strategy.72
Above all, military professionals must be alert to the tendency of political decision-makers to assume that military force is easy to employ and that power equates to success. And they must realize that strategy will never be sound if decision-makers — both military and civilian — are not sufficiently attentive to the logic of the campaign, the strength of the enemy’s will, and the enemy’s likely countermoves. In addition to the use of force, strategists are obligated to face up to and prepare for what comes afterwards: the challenges of terminating war and the requirements of jus post bellum. If civilian communication with the military is poor, partial, or adversarial — or if civilians sidestep crucial issues of cost and commitment — strategy will suffer. Indeed, these problems can lead to the failure of a campaign or a war.73
The Ongoing Utility of Coercion Theory
National security practitioners need to have a strong grasp of coercion theory if they are to be effective strategists and warfighters. The value of deterrence in particular has been lost among those worrying about threats in realms where deterrence is difficult, such as terrorism and cyber attacks. But deterrence remains an invaluable asset for national security. In the 21st century, it must be updated and applied intelligently to a new landscape of threats and challenges. Yet this hardly means that all we know about deterrence from its long history is suddenly obsolete. If there is a tendency in the Defense Department to think that technology so changes the landscape of war that the past no longer applies, then it is a pernicious tendency that works to America’s disadvantage. As Clausewitz insisted, wars will change in character over time, but their essential nature does not.
As it emerges from the long wars that dominated the first two decades of the 21st century, the United States is thinking again about the possibilities of conflict with near-peer competitors and nuclear states. In both realms, coercion theory has a great deal to offer.74 Richard Betts, for instance, has argued the case for using carefully crafted deterrent threats to prevent Iran from using nuclear weapons against third parties. He argues for specific, tailored deterrence. A broad nuclear retaliatory threat (against the Iranian population) might not be believed, but a more specific threat focused on the regime itself — a threat the United States is perfectly capable of carrying out — might well deter Iran from using nuclear weapons in the future.75
It is equally important to ensure — to the greatest extent possible — that any attempts America makes to deter an adversary are not interpreted as offensive or provocative.
Betts urges clarity when making deterrent threats. He insists that, as a people and a nation, America ought to determine where its real interests are located, and then bolster them with credible threats to potential trespassers. Citing the example of Korea prior to 1950, and of Iraq in 1990, he observes that when America has communicated its interests and deterrent threats in vague language subject to misinterpretation, it has suffered for it. America should avoid ambiguity, mixed signals, and potential confusion — but he worries that this is precisely the situation in which the United States has placed itself recently regarding Taiwan.76
It is equally important to ensure — to the greatest extent possible — that any attempts America makes to deter an adversary are not interpreted as offensive or provocative. If, for instance, the country wishes to deter by threat of denial (perhaps by bolstering military capability through exercises), it must signal to a potential adversary that this is a defensive action. Both statesmen and military professionals must fully understand the implications of the “security dilemma”: Any steps the United States takes to bolster its own defense will be interpreted by an adversary as offensive or provocative, or both.77 This requires the United States, as it undergirds its defenses, to also communicate its intentions and offer reassurance to limit escalatory tendencies and arms races.
National security professionals also need to understand the ways in which military tools may be used in crisis scenarios. Many military instruments are versatile: They can be used to send strong signals that are not inherently escalatory. A good case in point was the use of naval ships to “quarantine” Cuba against the placement of further nuclear weapons in 1962. This line in the sand (or water) drawn by the U.S. Navy was a very clear signal, but was not inherently escalatory. It gave the Soviets the opportunity to withdraw without further inflaming the situation. No doubt it produced a tense and fraught scenario: The Soviets, if they chose not to challenge the U.S. ships, could not escape without some degree of humiliation. But the situation was not nearly so escalatory and unpredictable as an air strike on Cuban soil would have been.78
In a similar way, the Berlin airlift of 1948–49 outflanked the Soviet isolation of West Berlin, which lay within the post-World War II Soviet occupation zone, without automatically escalating the situation. The United States, Britain, and France held fast to their commitment to the occupants of the western zones of the city, using a mechanism that was innovative and ultimately effective. In both the Cuba and Berlin cases, the situation did not escalate automatically as a result of U.S. actions. The Soviets themselves would have had to take the responsibility for upping the ante further.79 In neither case, happily, were they willing to do so. Creative thinking, including nontraditional uses of military instruments, proved to be just what the situation required.
Such creative thinking, however, depends on a solid understanding of coercion theory. For strategists, this body of literature is crucial. It forces planners and decision-makers to think through the assumptions and the logic of their actions. And it pushes them away from the dangerous idea that material power or predominance guarantees victory in conflict. The promise of a quick return on a coercive action can be a dangerous siren song for decision-makers looking for a simple solution to a complex political problem. In any scenario involving potential conflict, military and intelligence professionals need to anticipate challenges and problems and convey them effectively and persuasively.
Finally, an understanding of coercion theory helps all students of strategy appreciate the timelessness of the writings of strategists like Sun Tzu and Clausewitz, who warned about the need to make careful assessments, not only of our enemy but of ourselves. We must understand whether the stake in a given contest is more valuable to us or to the enemy. And we must face, with honesty and sobriety, the likely cost of our choices in money, time, and blood.
Tami Davis Biddle is professor of history and national security at the U.S. Army War College in Carlisle, Pennsylvania. She writes frequently on airpower, grand strategy, and 20th- and 21st-century warfare. She is the author, recently, of Strategy and Grand Strategy: What Students and Practitioners Need to Know (Strategic Studies Institute, 2015), and “On the Crest of Fear: The V-Weapons, the Battle of the Bulge, and the End of War in Europe, 1944-1945,” in the Journal of Military History (January 2019).
This article reflects the views of the author. It does not necessarily represent the views of the U.S. Army or Department of Defense.
Acknowledgements: I am grateful to colleagues who offered ideas and helpful comments on earlier drafts of this essay, including Conrad Crane, Darrell Driver, Edward Kaplan, Michael Neiberg, Celestino Perez, James Powell, Marybeth Ulrich, and Doug Winton. Stephen Biddle has influenced my thinking on this topic for years.