The Strategy of Denial: American Defense in an Age of Great Power Conflict

by Elbridge A. Colby

Importantly, this condition does not require that the fait accompli be quick, cheap, or easy, though these attributes naturally increase its allure. Executing a fait accompli can take time and be hard and costly. The essence of the fait accompli, rather, is the attacker’s ability to persuade the victim and its allies and partners not to try to regain the seized territory. It is thus largely about the resolve of the defenders—it exploits the reality that the defender and its allies and partners will typically only go so far to retake the territory in question. Success simply requires that the attacker be able to seize the territory in question and that the victim and its allies and partners be unwilling to suffer the costs and run the risks needed to reverse the gains. Second, the attacker must be able to hold the seized territory or consolidate its control over it.

Making such a strategy work against an anti-hegemonic coalition, however, requires not just any form of fait accompli. Simply seizing and holding some arbitrary part of a victim state’s territory is not enough. The goal is disaffiliation of the target from the coalition, and inducing a state to disaffiliate from an alliance or coalition will almost certainly impinge on its core goods. The whole point of membership in such an alliance—and in the coalition—is precisely to protect the state’s core goods, especially its political autonomy. Getting the state to give up its core goods will therefore require a degree of leverage that can plausibly come only from holding its key territory. The fait accompli in this context must therefore involve seizing and holding the victim state’s key territory.

The utility of the fait accompli—and by extension the importance to Beijing of having the ability to conquer its neighbors—is a critical point that is often confused in popular discussion. Many argue that Beijing does not want to create a territorial empire in Asia.69 This very well may be true, but it misses the point from a military planning point of view. Even if it does not want to directly control or annex other states, China does aspire to be the predominant power in Asia, and that means overcoming any anti-hegemonic coalition or preventing one from forming. The best way for it to do so is by using a focused and sequential strategy to compel states to leave the coalition. Yet as we have seen, the punishment approach is unlikely to give China the leverage it needs to achieve this end. Rather, Beijing’s better course is to use the fait accompli against coalition members, conquering their key territory in order to compel their disaffiliation from the coalition. Therefore, from a practical military point of view, China will want to be able to conquer substantial parts of Asia. This is because only the ability to seize and hold the territory of its neighbors will give Beijing the degree of power needed to force key regional states to disaffiliate from alliance with the United States and from the anti-hegemonic coalition more broadly.

Thus, although China may possibly pursue territorial expansion, it is not necessary for it to do so to gain regional hegemony and could well backfire, as Germany’s taking of Alsace and Lorraine in the peace settlement following the Franco-Prussian War contributed to French hostility under the Third Republic.70 Rather, Beijing’s core demand of any targeted state would more probably be that it disaffiliate from an alliance with the United States, if applicable, and the anti-hegemonic coalition.

Given that Beijing would profit most from subordinating a US ally within the anti-hegemonic coalition, China’s best, most consequential strategy for establishing hegemony would be to employ the fait accompli to conquer the key territory of a contiguous or proximate state or entity that is part of the anti-hegemonic coalition, ideally one that has a security commitment from the United States. This would necessitate defeating the victim’s military forces and conquering its key territory in order to compel its surrender.

China enjoys local military superiority over Taiwan to a greater extent than it does relative to any other member of the anti-hegemonic coalition.71 Consequently, and in addition to the reasons discussed previously for why Taiwan is China’s most attractive initial target, China’s most net gainful strategy is to attempt a fait accompli against Taiwan first. Beijing would very likely need to seize much of the main island of Taiwan itself, particularly Taipei and adjoining areas, to bring about Taiwan’s government’s capitulation.

The most attractive military strategy for the United States is dominance. In such a model, the United States could readily, decisively, and at relatively low cost defeat a Chinese assault on an ally, with no good way for Beijing to escalate.

But even though US military dominance over China is certainly desirable, it is simply no longer attainable, particularly with respect to Beijing’s ability to apply the focused and sequential strategy against the anti-hegemonic coalition.

An alluring alternative for the United States, rather than directly contest a fait accompli attempt, is to rely on the imposition of cost—especially through horizontal escalation, or widening the war and imposing costs on China beyond the immediate area of battle—in

Horizontal escalation would, for instance, seek to compel Beijing to give up on Taiwan or the Philippines by imposing costs on its interests farther afield, perhaps outside Asia. Yet China’s best way to gain global preeminence is first to establish regional hegemony in Asia.

The Best US Strategy: Denial Defense

In simpler terms, success for China is to subordinate the targeted state; defeat is to fail to do so. Success for the alliance, and thus for the anti-hegemonic coalition, is to keep the targeted state in the fold. As long as the defense is strong enough to keep the ally on side and affiliated with the coalition, then it is succeeding in its core strategic purpose.

Denial Option 1: Denying the Attacker’s Ability to Seize Key Territory

The defenders do not necessarily have to eliminate the entire first wave of invasion forces in order to prevent the attacker from seizing key territory. Nor do they necessarily have to destroy fully all the waves that follow. This is because denying an attacker’s ability to seize key territory is not an all-or-nothing proposition.

But the need for an attacker to have something approaching naval and air dominance before undertaking an invasion by sea is even more acute today, under what has been termed the “mature precision strike regime.”25 This phrase refers to the great advances in modern militaries’ ability to strike precisely at targets, including moving targets, at greater ranges and under more conditions. These trends further reduce the opportunities for concealment or cover available to sea and airborne forces, accentuating the defensive advantages against invasion afforded by the maritime domain.

To execute a fait accompli against a US ally in the coalition, and assuming the United States and other defenders were well prepared, China would very likely need to invade with a large force in a commensurate number of transport ships and aircraft. To defeat the Chinese fait accompli, the United States and other defenders would need only to prevent enough of those forces from reaching their destination so that whatever forces did land could be prevented from seizing the defender’s key territory.

Denial Option 2: Denying the Attacker’s Ability to Hold Seized Territory

There may, however, be instances in which an attacker is able to seize part or all of the target’s key territory.

it is one thing to seize territory, which only requires moving into territory a defender has left unprotected or been ejected from, and quite another to hold that territory in the face of counterattack, which requires preparing defenses adequate to resist such an assault. The history of conflict is full of assaults in which an attacker pushed defenders back but then could not consolidate its hold on the position;

But a defense in depth might also be useful for Taiwan, even though its key territory is more exposed to Chinese invasion. In the event of Chinese assault, Taiwan’s defenders might give up parts of Taiwan’s key territory in places such as Taipei and Kaohsiung and seek to draw Chinese forces into costly and difficult battles in urban, mountainous, or heavily forested areas.

That said, a defense in depth would be much more likely to succeed for Taiwan if it were coupled with the first denial option and the defenders could weaken and reduce the invasion force before it reached Taiwan, disrupt and degrade cross-strait lines of communication for Chinese forces once they had landed, or both.

Let us suppose, then, that a denial defense of some kind has succeeded—whether by stopping the invasion before it arrived on Taiwan or by preventing those Chinese forces that did land on the island from consolidating their hold. China’s invasion would have been defeated, but its ability to wage war would not have been. The United States and its allies and partners would therefore need to be ready for the possibility of a longer, broader war and be prepared to end that war on favorable terms.

If the United States forswore the ability to attack targets on the Chinese mainland that were materially involved in the war, it would gravely weaken its ability to defend Taiwan; treating them as off-limits would also only raise questions about US seriousness and resolve. This is not to say that these strikes need to be indiscriminate or even expansive. To the contrary, any strikes on the Chinese mainland should follow a clearly elucidated logic. For instance, the United States could make clear by both its official statements and the deeds of its forces that the defenders would strike mainland targets only if they were directly engaged in or supporting the campaign against Taiwan or were within a designated geographical area around the Taiwan battle area.1

this is almost certainly why China has developed the world’s largest and most formidable air defense network over the mainland, something that would be unnecessary and wasteful if Beijing deemed even focused and limited mainland strikes intolerable provocations demanding dramatic—especially strategic nuclear—escalation.

That said, what is good for the goose is good for the gander. If China has to accept the harder implications of such rule sets, so do the United States and the other defenders. If China agreed to a rule set permitting attacks on directly engaged forces, this definition would likely include at least some air and naval bases in the United States as well as cyber and space assets. The defenders should therefore propose and seek acceptance only of rule sets that are advantageous and whose implications they can live with.

Chinese economic coercion in this context would be more likely to catalyze the defenders’ resolve than lead to their capitulation. Beijing would in effect be seeking to reverse a local military defeat by turning the conflict into a societal contest of wills, a competition in pain tolerance. It would thereby make the conflict about much more than the disposition of Taiwan. The war would become about the security of the allied societies themselves, and their fates would now be much more clearly tied to the solidity of the alliance and the coalition’s ability to resist Chinese coercion.

Indeed, analysts frequently worry that China would go nuclear rather than lose a war over Taiwan.5 But it is very hard to see how vertical escalation—whether to high nonnuclear or nuclear levels—would not, like horizontal escalation, fail.

By definition, China’s strikes would not be coupled with a plausible way of attacking and subjugating allied states, which would have already won the local battle. They would thus be cruel but feckless.

Moreover, the defenders would have significant options to impose severe costs on China in ways that, because China had struck first, would seem highly justifiable. The United States alone has multiple options for selective responses using its long-range strike, cyber, and other capabilities.

If China is willing to use nuclear weapons and the United States is not, Beijing will dominate over whatever interests are at stake—whether about Taiwan’s fate, that of another US ally, or free American access to Asia more broadly. Moreover, given the repercussions of such a conflict for the anti-hegemonic coalition as a whole, the stakes at issue would likely be much broader and deeper than just about Taiwan. Conversely, a state willing and able to use its nuclear weapons effectively can show its opponent the inutility of or high costs associated with employing nuclear weapons against it, thereby almost certainly making such nuclear employment against it less likely in the future. This does not mean, of course, that a state must immediately resort to Armageddon in the face of any nuclear use by an adversary. But that state must have some way of responding effectively with its own nuclear forces—or it will be dominated.

Such a strategy would be based on the idea that China cares more about Taiwan’s disposition than the United States does and is thus more willing to bear pain in a brinkmanship contest. But things would not be so simple. First, such a contest would surely be perceived by Americans and others as about more than just Taiwan. If Taiwan were the only thing at stake, then China might indeed care more. But a Chinese attempt to use nuclear weapons against the United States to turn Beijing’s local defeat into victory would inevitably have wider implications. Furthermore, a China that used nuclear weapons in such circumstances would very likely seem to be far more dangerous than other states had previously anticipated. Allowing it to triumph in these conditions could only augur worse things in the future.

a Chinese brinkmanship strategy would not be very credible, as it would almost certainly end in a bluff. A series of tailored nuclear strikes against select US targets—sharp knife thrusts—would be one thing, but actually resorting to large-scale nuclear use against the United States would be another. Following through on such a course to its end would be genuinely crazy, since China would then lose everything in a major nuclear war over Taiwan—a highly valued but still decidedly partial interest. This madman approach might work if the stakes were low for the United States, but as we have seen, they would not be low.

Another alternative for China is a protracted war in which it could seek to regenerate the ability to mount another invasion attempt while improving the conditions for this subsequent effort. In this model, Beijing could work to rebuild its invasion force while seeking to erode the ability and resolve of Taiwan’s allies and partners to effectively defend the island against the next assault. It could try to do this by selectively escalating or expanding the conflict in ways that played to its advantages, such as by using targeted strikes or economic leverage to pressure key defenders or supporters. The logic of such an approach would be for China to apply pressure on Taiwan and the other defenders by continually degrading their defenses and resolve while regenerating its forces to prepare them to create or exploit a window of vulnerability in Taiwan’s defense.12

If Beijing chose this course, the United States and other involved states could seek to terminate the war, but—it is important to emphasize—they would not need to. They could accept a protracted war as the new normal if they preferred to avoid the considerable risks of trying to coerce China’s agreement to ending the war.

Such a protracted war can continue at a relatively high level of intensity if the sides direct their industrial bases toward sustaining such a conflict. Or it might become rather desultory. It is even plausible that trade and other nonbelligerent interactions between the defenders and China could resume amid such a protracted war. Alternatively, the war might persist at varying levels of intensity.

It must be emphasized that, in attempting to coerce China to terminate the war, the defenders would need to demand no more of China than that Beijing accept its failure to subordinate Taiwan and cease hostile efforts to subordinate it or any other US ally within the coalition. They would not need to require that China do something more fundamental, like change its government or give up territory. Moreover, they would not need China to relent formally; Beijing might meet the defenders’ demands informally. There has been no war on the Korean Peninsula since 1953, despite the two sides’ only having concluded an armistice; nor did Japan and the Soviet Union ever conclude a peace treaty after the Second World War, yet they have not fought since 1945. Both have been tolerable situations for the United States and its allies.

A strategy for war termination should thus enable careful escalation management yet also fail gracefully if it does not work. It should not only press China but offer China’s decision-makers reason to avoid a cataclysmic course and provide opportunities for de-escalation if things start to get out of hand. In other words, it should push China toward concession while giving it incentives to act with restraint. At the same time, a war termination strategy should allow for iteration and the prospect of error while avoiding prompting China to respond with massively destructive nuclear attacks. Such an effort would be an inherently experimental activity.

The logic of such strikes would be to make clear to China and important onlookers that China was bringing this harm on itself through its unreasonable recalcitrance, that the United States was not inflicting pain without good reason, and that China could end the infliction of such pain by recognizing the evident reality of its defeat over Taiwan—a limited equity well below the existential level. China’s leadership, in this model, would be the main driver of the harm brought on China, and they could turn off that suffering at any time by taking the limited, reasonable step of acknowledging a partial defeat.

Would this bring about the end of the CCP? If so, that would be existential.

First, a cost imposition approach would be better suited to keeping the war limited, especially as the conflict neared or even crossed the nuclear threshold, as it very well might. A cost imposition approach is meant to hurt the opponent enough to persuade it to relent, whereas approaches emphasizing denial require neutralizing a military force’s ability to do something. The problem with applying a purely denial approach is that it would raise the question of what the United States would seek to deny China’s ability to do.

The most intuitive way for the United States and its partners to employ a denial approach against China would be to reduce its ability to reconstitute a capability to attack Taiwan or other members of the anti-hegemonic coalition. Up to a point, this would make sense as part of a war termination strategy.

True denial of China’s ability to reconstitute its capability to attack Taiwan or another US ally in the coalition could therefore turn into an effort to destroy a much broader fraction—if not the entirety—of the Chinese military and industrial base.

Even if the United States and its allies intended only to destroy and degrade Chinese military power rather than conquer the country and forcibly change its government, Chinese leaders might fear the latter, making cataclysmic war more likely. Such an outcome would defeat the purpose of the defenders’ strategy—to keep the costs and risks incurred in the war matched with the interests at stake. Any victory of this sort would be pyrrhic—ash in the victors’ mouths.

By contrast, a strategy employing both denial and cost imposition would be more tailorable and controllable, and more likely to fail gracefully. Such a campaign would be less prone to spiral into a catastrophic conflict against a China possessed of survivable nuclear forces. Imposing costs would simply require the infliction of harm rather than eliminating China’s ability to take action, although the former might sometimes entail the latter. Thus targets could be selected to mitigate Chinese fears that the defenders’ attacks represented a prelude to invasion, forcible change of government, or conquest.

In other words, the defenders would demand of China only that it accept things as they had developed—that it swallow the results of a successful denial strategy. They would not need to compel China to give up anything more, such as territory or political concessions, to achieve their goals. The defenders would rely for war termination on measured doses of denial, augmented by cost imposition, particularly as denial reached a point at which it might provoke significant Chinese nuclear use.

In summary, then, the optimal strategy for the United States and other defenders of a threatened state in the anti-hegemonic coalition would be a combination of denial and cost imposition: denial defense against the fait accompli, with cost imposition layered on that effective denial to persuade China to accept the defenders’ preferred rule sets for limited war and as an option for inducing China to accept local defeat and agree to terminate the war. Denial would defeat China’s theory of victory against the vulnerable state; cost imposition layered atop denial would induce it to accept the defenders’ preferred rule sets and accept local defeat. But we must be clear: making a denial defense strategy work will not be easy or cheap. It is hard to overstate the scale and sophistication of the resources Beijing can bring to bear to subordinate Taiwan.

Even with the substantial changes in US forces needed to optimize such a posture, it will be tremendously difficult to deny a determined assault by China while keeping such a war limited in ways that correlate the demands of a denial defense with the resolve of Americans and the populaces of other participating nations. And without such changes, it may be impossible.

In a recapture approach, the United States and any potentially participating confederates would face a choice analogous to Beijing’s in its initial conquest: they could employ a punishment strategy to seek to make China give up the captured ally, or they could rely primarily on brute force to seize it back. A sufficient recapture approach does not necessarily require freeing every piece of seized territory; rather, it means liberating the captured state’s key territory to ensure that the ally can be restored as an independent state contributing to the coalition. In the context of Taiwan, this would very likely require freeing the main island—but not, for instance, the offshore islands of Quemoy and Matsu.

A punishment approach would mean a larger and more violent war because the engaged allies and partners would have to impose sufficient costs on China to induce it to give up the subordinated state. Given that Beijing would have manifold reasons to resist disgorging such a prized gain, and with so much riding on the resolution of the conflict, these costs would have to be very high even to prompt Beijing to consider giving up the held state. Such a punishment approach would be very unlikely to work, however, for the same reasons that it would very likely not work to deny China’s acquisition of the target in the first place. As in that instance, horizontal escalation alone would be unlikely to be effective because China is unlikely to possess anything beyond its borders that the coalition could threaten that is as important to it as prevailing in a war critical to its establishment of hegemony in Asia.

Consequently, in the event of the failure of a denial defense, the United States and other involved states would very likely need to seize back the conquered territory directly. This would require an invasion of the seized state, and because of the location of US allies in Asia, this would almost certainly involve an amphibious assault. As previously laid out, a successful amphibious invasion in these circumstances requires air and maritime dominance or something approaching it. Obtaining this dominance—if feasible at all—would very likely require a much larger, riskier, and costlier war effort than the more localized and constrained war envisioned for a focused denial defense.

The largest Western Allied counteroffensives in the Second World War did not take place until 1944, almost three years after the United States entered the conflict—and even longer after the United States, then the world’s largest industrial power, began to ramp up military production as the Arsenal of Democracy.6 And modern military weaponry can take considerably longer to field than that of the Second World War. Producing individual missiles under current circumstances can take years; production could be accelerated, but it is not clear by how much, especially since the demand would vastly outstrip current production capacity.7 Moreover, unlike in the Second World War, the United States would not enjoy a decisive advantage in industrial capacity; it is no longer unquestionably the world’s premier industrial state—indeed, that moniker may go to China.8