The U.S. Department of Deterrence
Madeleine Albright once asked Colin Powell, “What’s the point of having this superb military you’re always talking about if we can’t use it?” Nobody ever asked that question about nuclear weapons. Not using them is the whole point.
Deterrence thinking, far from being in decline, is now more fashionable than ever. Deterring China in particular calls for all hands on deck. The notion of “integrated deterrence” is thus a key pilar of the current U.S. National Security Strategy. While buzzwords may change, the basic idea is sure to feature in future strategies as well.
Integrating all forces, in all domains, in all regions, with all allies, for all kinds of conflict is obviously a daunting task. The fact that it may also be impossible has not stopped the U.S. military from trying. Indeed, the U.S. Department of Defense is struggling to become, in effect, the Department of Deterrence.
“Defense” is already a euphemism. The U.S. military is specialized for war. Its services and combatant commands are specialized for fighting in the land, sea, air, space, and cyber domains. But deterrence is a political function. National leaders ought to make difficult choices about what they really want, how much they are willing to pay for it, and where they are willing to compromise. Yet “deterrence” has become a euphemism as well, covering up hard tradeoffs about national interests.
We question whether the same forces that are specialized for winning modern wars can also be used to prevent them, or vice versa. As we argue in a new book on deterrence, forces that excel at “winning” are not always ideal for “warning.” Other policy tools are even better for “watching” in secrecy or even “working” with others. These are very different political activities, enabled by different specialized organizations.
In the first part of this essay, we discuss the enduring aspiration to integrate strategy. The second part likens integrated deterrence to combined arms warfare at the grand strategic level and highlights some important disanalogies. In the third we focus on the tension between military forces specialized for “winning” modern wars and nuclear forces specialized for “warning” about them. Finally, we discuss the importance of managing the strategic tradeoffs of deterrence in practice.
Aspiration vs. Implementation
Integrated deterrence is not new. Emerging threats are often accompanied by calls for reinvigorating deterrence with a new adjective. The George W. Bush administration sought “tailored deterrence” to mobilize nuanced capabilities for regional problems of counter-terrorism and counter-proliferation. The Obama administration worried about “cross-domain deterrence” in response to challenges in space and cyberspace, dramatized by the test of a Chinese anti-satellite weapon and a barrage of Russian cyber-attacks against Estonia in 2007. Turning cross-domain deterrence on its head, the Trump administration’s Nuclear Posture Review aimed to deter “an unprecedented range and mix of threats, including major conventional, chemical, biological, nuclear, space, and cyber threats, and violent non-state actors.” The special operations community even floated the notion of “comprehensive deterrence” to counter low-intensity threats as well in today’s complex strategic environment.
Not to be outdone, China developed a concept of “integrated strategic deterrence” to combine conventional, nuclear, and informational tools to offset American strengths. And to counter the looming threat of China, the Biden administration responded with “integrated deterrence.” The phrase first appeared in a speech by U.S. Secretary of Defense Lloyd Austin at a change of command ceremony at U.S. Indo-Pacific Command on April 30, 2021. Austin called for U.S. personnel to “rise above the old stovepipes” because “deterrence now demands far more coordination, innovation, and cooperation from us all.” The 2022 National Security Strategy later defined integrated deterrence as “the seamless combination of capabilities to convince potential adversaries that the costs of their hostile activities outweigh their benefits” and “working seamlessly across warfighting domains, theaters, the spectrum of conflict, all instruments of U.S. national power, and our network of Alliances and partnerships.”
The enduring aspiration for a seamless strategy belies the fact that there are many seams in practice. The notion of “integrated deterrence” takes longstanding efforts to integrate “joint operations” and “whole of government” policies to an unprecedented extreme. The Department of Defense was established in 1949 to combine the old Departments of War and the Navy, along with a new Department of the Air Force. Each service had different ideas about strategy. While the Army prepared to fight and win in wartime, the Navy patrolled the waves in peacetime, and the Air Force championed a new doctrine of strategic bombing, in which the transition from peace to war might occur without warning. Strategic Air Command, the forerunner to U.S. Strategic Command, grappled with the novel strategic problems of nuclear weapons. Subsequent technological innovation in space and cyberspace has culminated in new functional organizations like U.S. Space Force and Cyber Command.
Every new warfighting domain increases the integration challenge, no matter what buzzword is used to describe it. Getting the Department of Defense’s many services and combatant commands to sing from the same sheet of music has always been a hard problem. Wrangling the fractious components of the department has defied several ambitious secretaries such as Robert McNamara and Donald Rumsfeld. The post-9/11 challenges of counter-terrorism and counter-insurgency piled on additional challenges of integrating with the intelligence community, Department of State, and law enforcement agencies as well. Chinese military modernization, combined with more aggressive foreign policy under Xi Jinping, now puts a premium on both preparing for and avoiding war with China. Integrated deterrence, from this perspective, is mainly a bureaucratic problem of getting parochial organizations to work together for the common good.
Operations vs. Strategy
It is tempting to think about integrated deterrence, accordingly, as a straightforward application of proven operational concepts, but at the strategic level instead. In combined arms warfare, for instance, infantry provide flexibility and reconnaissance, tanks provide mobile fires and protection, artillery and aircraft deliver massed fires, and helicopters provide battlefield mobility. In “joint” or “multidomain” operations, likewise, navies project power on land, air forces shape the battlefield, everyone relies on satellite communications and intelligence, and mutual cooperation across all domains creates warfighting synergies. The intelligent orchestration of diverse specialties makes the whole team much stronger than the sum of its parts.
Implementing the modern system of force employment requires extensive training, refined doctrine, and high-quality troops. These luxuries are not available to every nation. Even for the U.S. military, achieving joint command and control was (and remains) a long, difficult institutional saga. But at the end of the day, integrated joint operations have enabled U.S. forces to prevail against less integrated foes, at least in battles if not in wars. Surely integrated deterrence might deliver similar benefits?
Operational integration across domains is challenging in practice. No doubt, integration at the strategic or grand strategic level is even more challenging. Yet integrated deterrence aspires to coordinate not only the entire Department of Defense but also agencies across the rest of the U.S. government with different administrative cultures and political constituencies — as well as U.S. coalition partners, sovereign nations with their own peculiar interests and defense politics.
The bureaucratic hurdles of implementation are formidable, perhaps insurmountable. Yet the strategic assumptions of integrated deterrence may be even more problematic. Operational integration, after all, serves the common ends of war, but deterrence serves many different political ends besides. Deterrence aims to produce strategic stability, of course, but appeasement might accomplish that, too. Deterrence thus also aims to protect or improve the status quo, and to do so without breaking the bank or compromising military effectiveness. What leader would not want to get their way, at low cost, without war, yet still win a war if deterrence fails? The problem is that these desirable goals cannot be accomplished at the same time, or to the same degree.
Winning vs. Warning
The 2022 National Security Strategy assumes that “Integrated deterrence is enabled by combat-credible forces prepared to fight and win, as needed, and backstopped by a safe, secure, and effective nuclear deterrent.” The same forces that are capable of “winning” in combat are also supposed to improve the “warning” needed for deterrence. This is often true. Military effectiveness is the foundation of deterrence by denial, which some consider to be more effective than punishment, especially in conventional deterrence.
But what happens when the means of winning are at odds with the means of warning? What happens when combat-capable forces are not combat-credible? Many of the hallmarks of modern military power — multidomain maneuver forces, automated weaponry, secret cyber and space operations — entail distinct liabilities for deterrence. It is possible for the U.S. military to be combat-capable without being combat-credible.
Naval forces, for instance, are very useful for projecting power across the world’s oceans. But their very mobility makes it hard for states to credibly commit to sending the fleet to any one location, or keeping it there, at any given point in time. Reliance on offshore balancing to deter China, for instance, raises questions about whether the United States will be willing to bet the U.S. Navy on a crisis over Taiwan. Flexible options — highly desirable for global power projection and maneuver — weaken deterrence commitments.
Automation produces a related problem. Drones help to lower the cost of intervention abroad and improve force protection. But it becomes difficult for adversaries to judge the resolve of an attacker that, quite literally, has no skin in the game. Unmanned options — highly desirable for limiting cost and risk in war — undermine costly signaling.
Many special technical operations in space and cyberspace are highly classified. But deterrence signals must be revealed to their target, and the revelation must be specific enough to be credible. It is not possible to reveal unique capabilities that can be easily countered, since the act of communicating a threat disarms it. Worse, secret weapons and sensitive methods create private information, which raises the risk of war. Secret options — highly desirable for seizing the advantage in war—undermine strategic stability.
It is telling that U.S. Cyber Command spent most of its first decade arguing that deterrence is not appropriate for cyberspace. Instead it adopted a doctrine of “persistent engagement” and “defending forward” to actively contest threat actors. This contrasts starkly with U.S. Strategic Command, where deterrence is the raison d’être for nuclear weapons. Strategic Command is successful when its capabilities are never used, but Cyber Command is successful when it is capabilities are used “persistently.” But now cyber warriors find themselves in the awkward position of explaining how cyber contributes to “integrated deterrence.” An underappreciated feature of nuclear forces is their relative transparency. Because there is no perfect defense against concerted nuclear attack, and even one nuclear bomb can incinerate a city, states are paradoxically free to share information about their nuclear capabilities with one another. This in turn leads nuclear weapons to be unusually effective in achieving deterrence. The ability to brandish nuclear capabilities without diminishing their power to hurt is one of the most consequential but misunderstood attributes of nuclear security.
Secrecy is the opposite of transparency. As Gen. John Hyten said, “You can’t deter people if everything you have is in the black.” Worse, secret advantages can undermine the deterrent benefits of transparency. Consider the use of offensive cyber operations to target nuclear command-and-control systems. This provides an invaluable “left of launch” counterforce option in case deterrence fails. But cyber exploits must be carefully concealed to be viable. Private information about cyber advantages in turn increases the chance that deterrence will fail, as wargames have demonstrated. Sometimes integration creates synergy, as in combined arms warfare, but integration can also produce painful tradeoffs.
As much as integration sounds appealing in abstract, and policy coordination is mandatory, the heterogeneity of intent and action between “winning” and “warning” — to say nothing of other objectives — requires specialization, rather than integration. Indeed, deterrence began here. The advent of the nuclear era dramatized the difference between “winning” a war and “warning” about one. As Bernard Brodie famously observed shortly after atomic bombs were used by the United States against Japan, nuclear weapons are more useful for deterrence than warfighting. Avoiding nuclear war while retaining political influence became the essential dilemma of the Cold War, and remains so today, even or especially when conventional force is being used or contemplated.
Managing Deterrence in Practice
The vast literature on deterrence emerged in response to a very specific military specialization — nuclear weapons. Scholarship on deterrence continues to burgeon in the 21st century. Ironically, technological specialization itself never found an explicit articulation in nuclear deterrence theory. Strategists should not assume that other forms of military specialization will improve deterrence as well. The problem is not only that deterrence depends on perceptions and beliefs, but also that deterrence is a relationship between many means and many ends.
Specialized services and combatant commands are not just repositories of operational expertise or management nodes. They are also specialized instruments for different strategic goals. Some military instruments — for example, ground forces and nuclear weapons — have specific advantages for the classical deterrence goals of improving credibility and preventing war. Other military capabilities — such as stealthy and maneuverable air and maritime forces — sacrifice credibility in return for enhanced warfighting. Emerging technologies in space and cyberspace are particularly well suited to intelligence collection and deception, which differ in subtle ways from traditional strategies of deterrence and defense. Non-military tools like cultural soft power or economic inducements, in turn, may be better suited for stabilizing strategies that eschew conflict altogether (i.e., accommodation or disarmament).
If we conceive of deterrence as containing various “ingredients,” then competent practitioners are faced with difficult decisions about how to combine these ingredients in order to produce “recipes” designed to conform with their respective national tolerances for cost, performance, risk, and reward. The trade-offs involved help to explain patterns of conflict and restraint that seem otherwise surprising. Given the multiple political ends of deterrence, the recipes used may often be incompatible or inconsistent.
Of course, strategic deterrence, or U.S. acumen in multidomain warfare, are hardly a waste of time. Quite the opposite. They ensure that the United States does not have to fight wars in places and in ways that are most costly, risky, or where defeat is least acceptable, such as on America’s own shores. Obvious military superiority is an excellent deterrent: Countries tend to avoid contests they don’t think they can win, or where the costs or risks of fighting appear excessive. But deterrence does not render potential adversaries impotent — the enemy still gets a vote.
A basic misconception about deterrence is that it somehow prevents an enemy from taking any action. A more accurate way to think about deterrence is that it is the process of steering the behavior of others, much as defensive obstacles channelize an enemy in war. Deterrence is a mobile buffet — doing well is an invitation for an adversary to innovate or otherwise disrupt the status quo. Indeed, the more effectively and extensively that a nation practices deterrence, the more likely it is that the contests it experiences are those for which it is least prepared.
The rock, paper, scissors logic of strategic interaction requires considering the adversary’s response to your actions, and your response to their response, as well as accepting tradeoffs between alternatives. This implies that we should think about integrated deterrence more dynamically. In practice this means decomposing deterrence into digestible components, at least initially. If one cannot win all the time everywhere, then picking where to win — and by extension when and what to lose — is the consummate art of grand strategy. Getting deterrence right may ultimately be less about distilling an abstract ideal recipe of military threats and assurances and more about discerning what a society really values.
To sum up, integrating deterrence is not simply an organizational problem, but also a political process. Decisions about force posture and structure tend to shape how and what a nation can and cannot deter. These political choices tend to become “baked into” a nation’s capabilities for considerable periods of time. Geopolitical competitors, in turn, are motivated to offset strategic investments by posing new challenges in different areas. Put simply, any strategy of integrated deterrence is destined for disintegration.
Deterrence is vital and essential for U.S. grand strategy. It is also inevitably complex and frustrating. In the best of all worlds, one should not have to choose between winning and warning, between defense and deterrence. But few would advocate designing national security on this “best case” basis. Integrated deterrence has yet to grapple with these tradeoffs, assuming them to be limited or non-existent. When one relaxes such assumptions to consider tradeoffs such as those between deterring and defending (or deceiving and disarming), we again face dilemmas in the implementation of this grand strategic vision. This requires a clear sense of our own national priorities and of the limitations of available tools.
Erik Gartzke is Professor of Political Science and Director of the Center for Peace and Security Studies (cPASS) at the University of California, San Diego. He is the author, with Jon R. Lindsay, of Elements of Deterrence: Strategy, Technology, and Complexity in Global Politics. He received a Ph.D. in political science from the University of Iowa.
Jon R. Lindsay is an Associate Professor in the School of Cybersecurity and Privacy and the Sam Nunn School of International Affairs at the Georgia Institute of Technology (Georgia Tech). He is the author of Information Technology and Military Power and other books. He holds a Ph.D. in political science from the Massachusetts Institute of Technology.