Defeat Is Possible
If the United States is to have a reasonable hope of winning a war, it needs to think very seriously about what it would be like to lose. For several years, analysts have been sounding the alarm that the United States and its allies might not prevail in a high-level conflict with a near-peer adversary. While Russia and China fall short of the United States in overall military power, they enjoy local overmatch in key theaters that might allow them to defeat U.S. forces. In 2019, analyst David Ochmanek of the RAND Corporation remarked that “In our games, when we fight Russia and China, blue gets its ass handed to it.” In November 2018, the National Defense Strategy Commission found that “If the United States had to fight Russia in a Baltic contingency or China in a war over Taiwan … Americans could face a decisive military defeat … Put bluntly, the U.S. military could lose the next state-versus-state war it fights.” These findings suggest that, in a pitched battle with a near-peer adversary such as China, American forces may be defeated even if its commanders don’t make any mistakes. Unfortunately, there exists a longstanding taboo in American strategic culture against the contemplation of defeat. To the extent that the possibility U.S. forces might lose on the battlefield is acknowledged at all, prescriptions for dealing with it fail to account for all the ways in which this defeat might transpire. If defeat is to be prevented, U.S. strategy and planning may need to think about all the different forms defeat might take so as to be ready for alternative kinds of conflicts and concepts of operations.
Thinking about defeat is not merely taboo in U.S. strategic culture. It is illegal in some cases. In August 1958, Sen. Richard Russell expressed outrage that he had heard on the radio “that some person or persons holding office in the Department of Defense have entered into contracts with various institutions to conduct studies to determine when and how, and in what circumstances, the United States would surrender to its enemies in the event of war.” Russell proposed an amendment to the supplemental appropriations bill then under consideration that “no part of the funds appropriated in this or any other act shall be used to pay” for studies of this kind. While the Eisenhower administration (which protested that Russell misrepresented the studies he was condemning) and some senators pushed back against the amendment, it ultimately passed with 88 votes for and only two against.
The Department of Defense and the RAND Corporation protested that the study that sparked this controversy, Strategic Surrender: The Politics of Victory and Defeat, was not really about the prospect of the United States surrendering to its communist enemies at all. Instead, its author Paul Kecskemeti argued that the policy of “unconditional surrender” that the United States and its allies had insisted upon during World War II had been counterproductive and had prolonged conflict with the Axis powers. What little discussion the study contained of a prospective U.S. surrender was limited to the final chapter, which examined the implications of surrender in the nuclear age. That merely hinting at the possibility that U.S. surrender might be possible elicited a law prohibiting any federal funding of research on the topic exemplifies the American allergy to thinking seriously about defeat. In an era when it appeared that a major war would end in a mutual annihilation rather than surrender, this tendency was perhaps excusable. But, in the present, when near-peer adversaries are increasingly capable of defeating U.S. conventional forces on a theater level, U.S. decision-makers can no longer afford to pretend that defeat is not a real possibility. And, so long as policymakers do not take losing seriously, they are unlikely to take the difficult steps needed to prevent such a defeat.
There is a natural tendency to avoid thinking about how America might lose by imagining how U.S. forces might win instead. Recent debates in U.S. defense policy circles about how to address the possibility of near-peer conflict have focused on denial strategies that would prevent Chinese or Russian territorial grabs from succeeding in the first place. Robert Work contends that the capability to “Sink 350 Chinese navy and coast guard vessels in the first 72 hours of a war, or destroy 2,400 Russian armored vehicles” would have this effect. Work also suggests that, with some relatively affordable investments (about $8 billion per year for three years), the U.S. military could actualize such capabilities in the near term.
But as Evan Braden Montgomery of the Center of Strategic and Budgetary Assessments argued in War on the Rocks last year, conceptualizing the problem in narrow operational terms could prove very risky. Montgomery pointed out rightly that doing so might undermine both the deterrence of potential aggression and allies’ confidence in U.S. security assurances, as foreign observers might doubt the United States will to act with the swiftness essential for destroying invading ships or armor before the invader attained its objectives under the assumptions of the denial strategy. Should an adversary invasion appear underway, it may become necessary to launch the campaign against it as quickly as possible. But, with feints, provocation, and deception, the adversary may obscure whether an invasion has begun, sowing uncertainty and undermining U.S. decision-makers’ willingness to take the fateful step of attacking Russian or Chinese forces. If adversary leaders come to doubt U.S. resolve, then deterrence could fail. If strategic partners come to doubt it, they may feel compelled to appease their neighbors rather than entrust their security to Washington.
Along with imperiling deterrence and assurance, a denial strategy framed in narrow operational terms may also increase the risks of defeat should conflict occur. The most effective deterrent force is not necessarily the same thing as the most effective war-fighting force, as the military capabilities that appear most impressive to either adversaries or allies may not be the same as those that prove most effective on the battlefield once conflict begins. Even if planners make the optimistic assumption that the capability to sink Chinese ships and destroy Russian armor can be made a reality at modest cost, as argued by Work and others, this does not mean that the associated denial strategy is a sure bet in case of war. Firstly, one or more necessary enablers of the capabilities may prove more technically challenging than anticipated or even impossible. Secondly, even if the requisite capabilities could be realized quickly and cheaply in the abstract, this does not mean they necessarily will be in practice. The history of U.S. defense procurement in the post-Cold War period is littered with examples of technologically ambitious systems that consumed huge development budgets without becoming operational realities. Implementation of the new capabilities may simply be flawed or delayed for all-too-familiar mundane reasons. The denial strategy also hinges on presidential resolve, as Montgomery pointed out. Even if the capability works, if the United States fails to use it promptly, the effect could be the same as if it didn’t exist. Finally, defining the problem in narrow operational terms simply invites the adversary to devise an alternative mode of attack. Explicitly signaling that U.S. leaders think of the problem this way makes it makes it clear for the other side how to begin planning accordingly.
Because of these considerations, the national security establishment may need to contemplate some kinds of near-peer conflict that it would rather not think about if it is to minimize the chances of defeat. If a near-peer conflict is not terminated on terms favorable to the United States and its allies in its opening phase, then there are two basic possibilities. The first of these is that the adversary managed to accomplish some or all of its goals in the opening phase of the war, but that the United States and its allies have refused to give up the fight. This presumably means a transition to a protracted war, possibly one of attrition. Even if the status quo ante could not be restored, Washington decision-makers might still feel it necessary to pursue such a conflict. Rationales for this could include a desire to shore up the faith of other allies in U.S. security guarantees and to impose costs on adversaries to deter them from further aggression. Unfortunately, U.S. strategy has not planned seriously for protracted near-peer conflict since the early Cold War. Being prepared for this contingency could demand considerable preparation and significant opportunity costs, but these may prove the best defense investment budget planners could make in the case of high-level conflict.
The second way that a protracted near-peer conflict might begin would be if the United States and its partners successfully parried an initial assault but the adversary refused to retire, initiating a protracted war on their own terms instead. While it seems like Western analysts have neglected this scenario, it seems all too plausible in some of the contingencies of interest (such as a Chinese invasion of Taiwan). Such a campaign might allow an aggressor to attain their goals even if these eluded them at first. For example, if a U.S. denial campaign depended upon an inventory of some hard-to-replenish resource (such as precision-guided munitions) and these assets were depleted in the opening phase of the war, the adversary might be able to regroup and mount another more successful attack before the United States could reconstitute its denial capability. Moreover, the failed initial attempt would also provide learning opportunities that might be exploited to neutralize U.S. advantages in the rematch. And, even if an adversary campaign of attrition fails to achieve its objectives, the United States and its partners might have to deal with its consequences.
It is much more unpleasant to envision losing than winning — but this does nothing to change the fact that defeat is an increasingly plausible possibility in a war with Russia or China. Brad Roberts has argued forcefully that the United States should have clearer “theories of victory,” but, in addition to “theories of victory,” defense planners need to have “theories of defeat” in order to turn those theories into self-negating prophecies. In order to forestall defeat, the Pentagon may need to envision how it could lose. Defense intellectuals could contemplate all the diverse ways U.S. forces might be defeated instead of one or two specific ways in which they would prefer to win. At the very least, planners could begin formulating contingency plans to continue the fight should the opening phase of a near-peer conflict fail to go as desired. An essential first step could be to start taking the prospect of protracted near-peer conflict seriously. Whether or not U.S. policymakers want such a conflict, one may be imposed upon them — and at present, America is woefully underprepared for it.
Edward Geist is a policy researcher at the nonprofit, nonpartisan RAND Corporation. Geist received a Smith Richardson Strategy and Policy Fellowship to write a book on artificial intelligence and nuclear warfare.