There are two basic ways to deter an enemy.1 One is to threaten to hurt him if he attacks you or your allies (deterrence by punishment). This form of deterrence depends on fear that the defender will inflict a level of pain that exceeds whatever gains the attacker hoped to achieve through aggression. For deterrence by punishment to work, the defender’s threat has to be credible. He has to possess sufficient lethal capabilities to carry through the threat. His weapons have to be known to be capable of reaching the attacker, evading or overcoming his defenses, and either defeating his forces, devastating his population or both. It also has to be clear that the defender is deeply attached to the object he is defending and what forms of behavior will prompt retaliation. The fuzzier the trigger mechanism, the more room there is for ambiguity, and the weaker the deterrent becomes.
A second way to deter an enemy is to make it physically difficult for him to achieve his objective (deterrence by denial). This form of deterrence also depends on fear, but of costs that will be inflicted during the act of aggression, in the place that it occurs. It seeks to make aggression unprofitable by rendering the target harder to take, harder to keep, or both. To work, the defender has to have sufficient lethal capabilities in or near the likely site of aggression to demonstrate that victory will be either be impossible or difficult to attain. The defender’s capabilities have to be known to be able to inflict substantial pain, not in counter-attack but in defense. This differentiates deterrence by denial from the concept of “tripwires,” which are small forces placed in harm’s way to activate retaliation rather than to inflict pain. There is little room for misinterpretation about the trigger mechanisms in deterrence by denial, since by definition it is activated by physical contact with an invading enemy. Deterrence by denial has been favored by small states, but it can also be used in extended form by a Great Power to protect key pieces of terrain or weaker allies, either as a standalone strategy or in tandem with punishment.
For most of the modern era, America’s extended deterrence has been based on deterrence by punishment. This form of deterrence was well matched, both to the tools at America’s disposal and to the nature of the objects it needed to defend. Where Britain in its heyday possessed deterrence mechanisms (the Royal Navy and, later, strategic bombers) that were limited in their ability to reach rivals deep within the Eurasian heartland, the United States has faced no such geographic restrictions in threatening punishment. With the advent of nuclear weapons, it possessed the means to reach and devastate any state threatening its homeland or allies. Indeed, so vast was America’s military edge that it hardly needed to invoke nuclear force to achieve an effective extended deterrent; its “overmatch” in conventional arms was sufficiently wide to deter most challengers. This allowed the United States to secure far-flung alliances across the rimlands of Eurasia while retaining the flexibility to shift attention back and forth from crisis points without undermining its overall strategic position.
Why It’s Getting Harder to Punish
It was generally assumed that the underlying logic of security through fear of retaliation, which worked well enough during the Cold War, would be equally valid against whatever assortment of competitors eventually emerged after the Cold War to challenge U.S. primacy. But this is not proving to be the case. America’s international threat environment is evolving in unexpected ways that pose significant challenges to traditional extended deterrence based on punishment. Three changes in particular are likely to erode its effectiveness in the coming years. First, there is the sheer number of competitors that the United States must deter. Rather than managing a contest with one, large ideological opponent, the United States must cope with several powerful rivals simultaneously, two of which possess large nuclear arsenals, all of which have significant conventional militaries, and at least one of which is growing rapidly in relative economic and political power. This reduces America’s overall military edge and muddies the underlying power relationships upon which much of the superstructure of U.S. extended deterrence has rested.
Second, America’s rivals are becoming better armed. In the words of one senior Pentagon official, Russia and China are “fielding very advanced capabilities at an extremely rapid pace.”2 They are investing in many of the types of technology—long-range missiles, stealth, next-generation fighter aircraft, and electronic jamming—in which the United States has long maintained a commanding lead. Both China’s build-up of Anti-Access Area Denial (A2AD) weapons and Russia’s “escalate-to-deescalate” tactical nuclear doctrine are examples of efforts to deprive the United States of escalation dominance in future contests. While they have virtually no effect on U.S. nuclear deterrence, these changes erode the effectiveness of deterrence by punishment by reducing the certainty that America could manage escalation and successfully impose high costs for acts of aggression. They increase the presumed reliance of the United States on nuclear deterrence for managing even small confrontations—a recourse that America’s competitors rightly assume it will be loath to use in an era of public aversion to casualties (both inflicting and suffering them).
Third and most importantly, America’s rivals are developing tactics for evading retaliatory deterrence.3 Russia and China have introduced limited-war techniques designed to avoid trigger mechanisms of extended deterrence—for Russia, “jab and grab” land incursions; for China, the creeping militarization of maritime zones. Both techniques operate below the threshold of deterrence by punishment and seek to create territorial faits accomplis that lower the costs of revisionism. Such methods pose serious problems for deterrence by punishment, which is predicated on aggression being identifiable and therefore punishable. In a limited war setting, punishment quickly morphs into compellance—not just dissuading an enemy but dislodging him and forcing a withdrawal from his limited, stealthy conquest. As Thomas Schelling has argued, compellance (“a threat intended to make an adversary do something”) is inherently harder than deterrence (“a threat to keep him from starting something”).4 Limited war shifts the psychological burden of conflict—fear of retaliation—away from the aggressor and places it on the shoulders of the defender—fear of escalation. It puts the defender in the position of perpetually under-responding to ambiguous provocations (and thereby losing control of strategically vital spaces by default) or over-responding (and risking war).
Diversifying Deterrence Options
So far, America’s response to erosions in its military position has been mainly technological: to bolster its competitive advantages against rivals and thereby shore up the credibility of existing deterrence methods. Attempts at reassuring frontline allies through the deployment of tripwires (most recently, in Eastern Europe); the push to refocus U.S. strategic nuclear doctrine on deterring major war; and some (though not all) of the components of the Pentagon’s third offset are all examples of how the United States is “doubling down” on punishment. These efforts are likely to yield positive results in bolstering key components of retaliation and ensuring that deterrence by punishment remains the backbone of America’s extended deterrent for the foreseeable future. However, as America’s lead in advanced weaponry narrows amid prolonged cuts in defense spending, the United States may face limits to how far it can go in using technology to deter its rivals. Unlike its past responses to the problem of competitor “catch-up” (for example, Eisenhower’s “New Look” and the Offset Strategy of the 1970s), the United States will need to look for other sources of comparative advantage with which to maintain its primacy.
One such advantage is geography. America’s possession of extensive security relationships with states along the rimlands of Eurasia offers natural strategic advantages that have not been systematically exploited in post-Cold War strategy. It was hard to employ such states effectively in past U.S. extended deterrence systems, since the very act of threatening punishment to rivals with America’s vast arsenal dis-incentivized self-help among its allies (the famous “free-rider” problem). More often than not, allies were viewed as liabilities, either because they required acts of assurance to maintain or because, if over-assured, they might themselves become sources of provocation (the “entrapment” problem). But as the global geopolitical environment turns predatory, frontline states now have a powerful incentive to do more for themselves in defense: self-preservation. Thus motivated, many are arming and aligning with similarly placed neighbors in an effort to contain large powers.
These dynamics create an opportunity for the United States. America’s extensive frontline alliances offer a natural means of pursuing a strategy of deterrence by denial, both as an end in itself for containing growing rivals, and as a compliment to deterrence by punishment. Historically, Great Powers have often employed alliances with states located near rivals in this way. Prior to the nuclear era, denial was a more common way to achieve extended deterrence, since the tools for projecting decisive military force were less reliable. When facing rivals that were numerous or far away, it was often more effective to make it physically harder for them to expand rather than by threatening to punish them. Such denial strategies were especially common among maritime or status quo powers attempting to raise the costs of revisionism in key strategic zones. Since they worked with the momentum of the small states’ desire to remain independent, deterrence by denial had the added benefit of being cost-effective, particularly at moments of power transition.
Three Ways to Deny
Generally speaking, there are three ways that a Great Power can achieve extended deterrence by denial. One is to make an ally or piece of territory harder to take. This usually involves placing defensive attributes of some kind in the place that is likely to be the object of the revisionist power’s desire—either using the forces of the patron or, far better, those of the targeted ally. The crucial difference between this approach and the “trip-wires” used for deterrence by punishment is that forces deployed for denial are intended not to die and trigger punishment but to live and inflict pain on the attacker. The most common way to ensure this is to help the frontline ally beef up its defensive properties. The classic example is 18th-century Britain’s policy of providing subsidies and arms to Europe’s smaller states to deter the expansion of continental rivals. Another, bolder method is to help the ally acquire offensive capabilities.3 This requires the ally to be large enough in size and capabilities to attempt a conventional deterrent of its own. France’s alliances de revers of the 17th and 18th centuries, for example, created an extended deterrence based on denial by building a bulwark of aggressive mid-sized frontier states (Sweden, Poland, and Turkey). Through subsidies, advisers, and shared technology, France encouraged these states to adopt eastward-facing offensive postures (Charles XII’s famous exploits were largely funded by France) to keep Russia on the defensive and impede its expansion into Europe proper.
A second form of deterrence by denial is to make the revisionist’s coveted object harder to keep. This is usually the best option when the ally in question is too weak to mount a credible defense but possesses sufficient willpower to make it indigestible for an attacker. Since the ultimate goal of revisionist powers is to achieve quick, easy grabs, this strategy seeks to make them prolonged and costly—what could be called the “bitter pill” strategy. Sixteenth-century Switzerland is one such example; another is 20th-century Finland. Both used scrappy defensive techniques and small but well-trained forces to advertise indigestibility to predators. In extended form, this approach can involve a Great Power providing weapons of a type or number that enable an otherwise indefensible state to be capable of waging guerilla war against an attacker and outlasting occupation. Britain used such an approach with the Netherlands in the early 18th century by funding mercenary armies and encouraging the flooding of Holland’s fields; in the early 19th-century, it supplied weapons and fortifications to Portugal to resist French encroachments.
The third form of deterrence by denial is to make the ally or territory in question stronger industrially than the attacker. Unlike the methods above, which focus on militarily penalizing an attack, this form of deterrence is long-term and mainly economic. Since a major goal of revisionist aggression is to break alliances, denying them their objective can be aided by strengthening the ties between the target country and its patron. Abundant evidence shows that extended deterrence is most effective when the military relationship between two states is undergirded by economic ties, particularly in strategic industries. A study of 54 cases of deterrence between 1900 and 1980 found that defensive alliances characterized by close political ties and even small amounts of trade succeeded in deterring aggression seven times out of eight, compared to much lower success rates in alliances based purely on military relations.5 France famously sought to build up the resiliency of its CEE alliances not just by aiding and advising their militaries but by providing investment to build up strategic industries and railways, and to fuel economic growth.
Whether by short- or long-term methods, the goal of these strategies is to deter a revision of the overall system by deterring the conquest of specific places. Where deterrence by punishment leaves certain frontline terrain unguarded and thereby involves an assumed sacrifice—both of tripwire troops and the soil of the target state—deterrence by denial seeks to make the conquest of the target an altogether unattractive prospect. The two forms of deterrence are not mutually exclusive; indeed, combining them can strengthen the credibility of both. Having the ability to punish while cultivating the local means to resist creates a virtuous cycle, communicating to the ally that self-defense is not hopeless and to the revisionist that he may have to pay twofold for whatever gains his aggression may yield.
Alliances as Denial Tools
America has abundant opportunities to inter-weave denial methods into its wider punishment-based system. Rimland alliances offer natural tools for managing multiple, large Eurasian competitors. In both Europe and Asia, the United States possesses alliances with small states with the motivation to oppose local hegemons. A U.S. strategy to “activate” these alliances as instruments for discouraging attempts at control of important real estate could include all three of the forms of denial mentioned above. It would use the military geography of frontline allies to make them and nearby real estate harder to take. America’s archipelagic Asian allies can employ naval mines, submarines, and an increasingly advanced array of missiles to turn the region’s narrow waterways into clogging mechanisms for impeding Chinese naval expansion. Similarly, in Eastern Europe, allies in the Baltic-Black Sea corridor can be turned into “bitter pills” bristling with anti-tank and anti-ship missiles. Mid-sized states (like Poland, Finland and Romania) can be armed with offensive weapons like the AGM-158 JASSM and Tomahawk missiles to keep revisionists off balance and diverting budgets to defensive capabilities.6 Tiny but determined allies like the Baltic States can be made harder to hold in an invasion by, for example, offering pre-set U.S. packages of nasty weapons like landmines that are automatically released on the first sign of aggressive behavior. In both regions, a strategy to instigate and organize dense networks of intra- and inter-regional cooperation in strategic industries and R&D would thicken the “hide” of allied frontiers on a longer-term basis.
Incorporating denial into the U.S. extended deterrence system would offer important advantages for the United States. Strategically, it would focus scarce resources to the places where conflict is most likely to occur. Technologically, it would play to many of America’s emerging areas of competitive strength, including in defensive weapons like surface, air, and naval missiles favored in the third offset, while prompting greater clarity in Pentagon thinking on the as-yet underexamined role for alliances in the offset strategy. In defensive terms, it would provide tools that, should deterrence fail, are more easily employed to fight and win a conflict than those used for punishment.7 Geographically, denial would play to America’s latent, underdeveloped advantage: its wide range of frontline allies with the predisposition and location to disrupt rivals’ offensive moves. Organizing these states for stronger self-defense would raise the visible costs of revision without necessarily adding commensurate defense burdens for the United States. By systematically bolstering the denial capabilities of frontline allies, the United States could concentrate its own resources on higher-tier punishment weapons. Strengthening frontline resistance would also reinforce and clarify the trigger mechanisms for deterrence by punishment and help to make extended deterrence as a whole more solvent against hybrid-war threats.
Injecting greater emphasis on denial into U.S. strategy could also have a political benefit for America’s alliances. Traditionally, frontline states have had an aversion to patron relationships that are overly reliant on punishment, especially when the patron in question is a maritime power. Unlike retaliation, which can seem remote and beyond control to the state being defended, denial is immediate and in the obvious self-preservation interests of the ally, which therefore has strong incentives to deny the enemy the immediate and narrow objective of conquest. Whereas relying primarily on deterrence by punishment from the patron may weaken the alliance, leading to fears of abandonment and being merely a great-power bargaining chip, joint investment by the patron and ally in denial strengthens the political bonds of the alliance.
Most importantly, a greater focus on denial could help to shift the psychological burden of 21st-century conflict back where it belongs: on the shoulders of states that wish to rearrange the international order. Whereas limited-war techniques enable revisionists to believe they can avoid triggering retaliation and thereby get away with an easy victory, denial signals that they will pay a steep price for aggression at the place it occurs, ranging from a sharp rebuff to a war of attrition. From Eastern Europe to the Western Pacific, the goal should be the same: to limit options for easy revisions and to increase the immediate cost and difficulty of grabbing and holding territory. Building up such mechanisms will help the United States avoid the predicament of holding together through compellance what it could not through deterrence. The goal should be to restore a healthy sense of fear in would-be predators. Doing this now, while the century is still young and revisionists are still mulling over their options, will be a far cheaper policy in the long run than waiting for deterrence by punishment to fail and then trying to regain lost ground through coercion.
1Glenn Snyder first drew the distinction between deterrence by punishment and deterrence by denial in 1961. See Glenn Snyder, Deterrence and Defense: Toward a Theory of National Security (Princeton Univ. Press, 1961), pp. 14-6. For a recent discussion see Michael S. Gerson, “Conventional Deterrence in the Second Nuclear Age,” Parameters, 2009.
2See “The Third U.S. Offset Strategy and its Implications for Partners and Allies,” comments delivered by Deputy Secretary of Defense Bob Work, Willard Hotel, Washington, D.C., Wednesday, January 28, 2015.
3For a detailed analysis of this problem, see Jakub Grygiel and Wess Mitchell, “Limited War is Back,” National Interest, winter 2014.
4For a discussion of the differences between deterrence and compellance, see Thomas Schelling, The Strategy of Conflict, Harvard University Press, 1963, p. 195.
5Paul Huth and Bruce Russett, “What Makes Deterrence Work? Cases from 1900 to 1980,” World Politics (July 1984), pp. 518, 516.
6See Jakub Grygiel, “Offensive Capabilities and Small States,” forthcoming.
7See Gerson, p. 38. If deterrence fails, denial “offers control rather than continuing coercion…with punishment, the [adversary] is left to decide how much more to take. With denial, the choice is removed.”