FOR NEARLY twenty years following the end of the Cold War, military confrontation between the United States and the Russian Federation seemed implausible. Even during periods of tension, as during the Kosovo crisis in the late 1990s, few believed that disagreement between Washington and Moscow could lead to a serious crisis, no less war. Before the first decade of the new century had passed, however, Russian officials were accusing the United States of working to isolate Russia. Such apprehensions have mounted steadily in Russia in the years since. At the same time, Russian behavior, including interventions in Ukraine and Syria, military posturing and harassment in Europe, and interference in Western elections, has led many in the United States to conclude that, while a U.S.-Russian conflict is by no means inevitable, the risk of such a confrontation is growing.

Even as U.S.-Russian tensions have risen, fundamental shifts in the military-technological environment threaten to erode strategic stability between the two nations. In the coming years, both sides’ extensive dependence on information technology, coupled with likely perceptions of lower risk for the use of “nonkinetic” and nonlethal attacks, are creating new incentives to use cyber and/or counterspace weapons early in a crisis or conflict. At the same time, the advent of novel cyber, counterspace, precision-strike, missile-defense and autonomous military systems could cause one or both nations to lose confidence in their nuclear second-strike capabilities—thereby eroding the stability afforded by mutually assured destruction.

WASHINGTON AND Moscow hold divergent—and, in some cases, directly contradictory—views of the international security environment. Whereas Russian leaders see their country as behaving in a fundamentally defensive manner, the United States and NATO perceive a revanchist actor intent on reimposing its will in eastern Europe and Central Asia.

The Russian government rates the United States and NATO as the most serious threats to Russian national security. From its standpoint, the United States is intent on remaining the world’s sole hegemon, and, as such, is unwilling to tolerate a strong Russia that enjoys its own sphere of influence. Moscow has spoken out against American and European efforts to encircle the Russian Federation by integrating former Soviet republics into Western institutions like NATO and the European Union. Russian officials further condemn the United States’ purported use of “color revolutions”—or, as Moscow would characterize them, the sponsorship of coups d’état under a guise of democracy promotion—to install proxies in Russia’s “near abroad.” In addition, Russian analysts assert that the United States and NATO are using a variety of political, economic and informational tools to penetrate and disrupt Russian society itself. In response to these perceived threats, Russia has undertaken a major military modernization effort and used more aggressive rhetoric and military operations, economic coercion and inducements, and information operations to counter alleged U.S. and NATO expansionism.

The view from Washington and Europe differs markedly. In the assessment of the United States and NATO, the Kremlin appears intent on restoring a buffer zone of compliant or client states throughout its self-declared “near abroad.” Western officials condemn Russia’s violations of international law and norms by using force against Ukraine, changing borders in Europe through violence, violating arms-control agreements and seeking to undermine Western democratic elections. They are equally concerned by Russian military modernization, which, coupled with intensified military exercises, activities and bellicose rhetoric, is seen as a direct threat to NATO security. The United States and its allies have responded to perceived Russian aggression by strengthening NATO’s deterrent posture, as well as holding increasingly frank discussions about NATO’s central role in defeating a Russian attack.

U.S.-Russian relations in the coming years will take one of three forms: strategic rapprochement, intensified military competition or managed competition. Although a long-term rapprochement cannot be ruled out, and indeed is a valuable (very) long-term goal, striving for such an outcome—or even another attempted “reset” (or “re-reset”)—in the near term would likely lead to quick disappointment. This reality does not eliminate room for the pursuit of common interests on issues such as nuclear nonproliferation, counterterrorism and counternarcotics. Yet both the United States and Russia must act on the understanding that there is a real potential for political disputes to lead to crisis, and for crisis to lead to conflict.

At the same time, Russia’s nuclear forces will, for the foreseeable future, allow it to destroy the United States as a functioning society. Hence, as distasteful as “working with” Russia may seem, the alternative of full-throated confrontation would pose unacceptable and unnecessary risks to the United States. That said, one should not brim over with unbridled optimism. Russian leaders are engaged in continuing efforts to undermine America’s alliances, democratic processes and global role. A change in this strategic approach appears highly unlikely, and as a result U.S.-Russia competition is the likeliest path short of outright confrontation. The challenge, then, is charting a balanced path ahead that recognizes the real competition and potential for conflict, while allowing for prudent cooperation and improvement in the relationship where possible.

THE UNITED States and Russia are each pursuing a range of advanced military technologies in order to bolster their respective conventional military postures in Europe (and in the western Pacific, in the case of the United States). Priority investments for both sides include novel cyber, counterspace, long-range nonnuclear-strike, missile-defense and autonomous weapons systems. Russian strategists writing in Military Thought, the in-house journal of the Russian General Staff, rightly note that these technologies are likely to significantly increase the pace of military engagements. Uncertainty about these systems’ effects will also increase risk of miscalculation or misunderstanding. These factors, both singularly and in interaction with one another, could lead to “slippery slopes” of rapid escalation from crisis to conflict, especially in cyberspace and outer space.

The U.S. and Russian militaries are increasingly dependent on networked information technology (IT), and both have embarked on ambitious offensive cyber programs. Because of the frailty of cyberweapons—once a weapon is revealed in detail, the adversary can fashion effective defenses—there exists a tremendous premium on secrecy regarding both states’ cyber tool kits. As a result, a great deal of uncertainty surrounds each side’s capabilities. Even so, a 2013 Defense Science Board report conveyed a sense of cyberweapons’ potential impact: “The benefits to an attacker using cyber exploits are potentially spectacular.” Similar dynamics obtain in the space domain. The United States relies heavily on vulnerable military satellites for a host of critical military functions. The Russian Federation does so too, but to a far lesser extent. At the same time, both side possess inherent antisatellite (ASAT) capabilities in their ballistic-missile defense interceptors, and are developing other ASAT weapons, such as co-orbital satellites, missiles, directed-energy weapons and cyberattacks.

Five specific factors could lead to rapid and unintended escalation in cyberspace and outer space. First, the United States and Russia will likely face strong incentives to use cyber and counterspace attacks early in a future crisis or conflict. These capabilities could offer—or be seen to offer—an opportunity to rapidly degrade an adversary’s information systems at the outset of a conflict, conferring a significant advantage to the first mover. Indeed, Russian strategists have been clear in their assessment that seizing the initiative in the information domain will be key to victory in future wars. In addition, Washington and Moscow may perceive the use of cyber and nonkinetic counterspace weapons as less escalatory than that of kinetic weapons, since they can be employed in nonlethal, non-physically-destructive and reversible ways. This assessment is also reflected in Russian doctrinal writings. Russian strategists characterize both ASAT and cyberweapons as optimal tools for deterrence, since they can be used to destroy enemy infrastructure without inflicting heavy civilian casualties. Such a belief could further lower the threshold for early cyber and counterspace attacks.

Second, cyber and counterspace attacks intended to be highly discriminative against military targets may cascade to affect critical infrastructure essential to a broader society and economy (e.g., electrical grids or commercial satellites). If this occurred, an attack intended to be precise and limited to military targets could instead result in the widespread loss of electrical power, water or other essential services, with resulting economic disruption and potential loss of life. The attacked side could feel compelled to respond at least in kind. Alternatively, a tit-for-tat cycle may occur, as one side might believe it could gain coercive advantage by intentionally demonstrating its ability to pose risks to the other side’s critical infrastructure through a combination of cyber, counterspace and perhaps sabotage attacks. Prominent Russian strategists endorse this view, arguing that attacks on socioeconomic targets can frighten an enemy population into abandoning its war effort against Russia. Yet such countervalue strikes could lead to major conflict and even, in an extreme scenario, nuclear war.

Third, attacks intended to target nonnuclear systems (including but not limited to cyber and space attacks) could inadvertently impinge on nuclear systems, and be misread as a much more escalatory move. For instance, some assets in outer space support both conventional and nuclear missions—particularly in the case of the United States—and both theater and strategic missions. In addition, many terrestrial elements of U.S. command, control and communications, as well as long-range strike systems, are dual-use, and there may be co-location of conventional and nuclear systems by one or both sides. Cyber or counterspace attacks on these systems could therefore implicate nuclear systems, raising the potential for inadvertent escalation.

Fourth, to the extent that the attacker’s initial cyber, space and precision-strike attacks were successful in negating a portion of the other side’s military, the attacked side would fear further debilitating attacks, and could worry that it must use or lose its strategic-level attack capabilities, including not only cyber and space, but also long-range strike capabilities. In the extreme, it may feel its conventional capabilities so weakened that it would consider the use of nuclear weapons. By the same token, nuclear forces use IT and space assets for warning and communications. As a result, a cyber or space attack could put nuclear use-or-lose considerations into play early in a crisis.

Russian strategists are especially cognizant of these dynamics and recognize that a drawn-out conflict will likely force both sides up the escalation ladder. To avoid this outcome, a number of Russian military theorists argue for the use of defensive preemptive strikes—especially using nonnuclear capabilities—on enemy military and/or socioeconomic targets. Some suggest that these strikes could degrade adversary power-projection capabilities, such that Russia could avoid being forced into a situation where it had to use or lose its strategic-level assets in the first place. Others say—as previously mentioned—that tailored preemptive strikes could deter aggression by communicating to target policymakers and publics alike that the costs of attacking or escalating a military confrontation with Russia would outweigh the benefits.

Fifth, inadvertent escalation could result from a misattributed attack or third-party false-flag operation. Chance errors in a key system, in the midst of crisis, such as an internal fault in a side’s command-and-control system or one induced by natural causes (e.g., a solar flare or an electrical surge), could be construed by one side to be an intentional act by the other. In addition, the diffusion of offensive cyber capabilities could allow smaller powers or nonstate actors to provoke a conflict—for instance, by conducting a “false flag” digital operation designed to trigger a crisis. Alternatively, once a conflict has begun, they could use their own capabilities to expand the scope or scale of the conflict.

The possibility of escalation to large-scale war stemming, even inadvertently, from lower-order conflicts or tensions has long been appreciated in the context of the United States and Russia. The contention here, however, is that technological advances, their integration into military postures and doctrines on both sides, and the often-unanticipated ways in which such integrations may interact are together heightening the possibility of inadvertent, rapid and dramatic escalation in the event of crisis or conflict between the United States and Russia.

STRATEGIC STABILITY between the United States and Russia has long rested on each side’s confidence that it could absorb even an all-out nuclear first strike by the other side and then unleash a devastating nuclear second strike. That confidence, however, is being tested by the deployment of new military systems. As these capabilities mature, each side is likely to have growing fears that the other side might employ these capabilities (with or without also using nuclear weapons) in a first strike to attempt to negate its nuclear second-strike capabilities, thereby obviating mutually assured destruction.

Cyber weapons could be used against vulnerable nuclear weapons, delivery systems, and command, control and communications (NC3). The potential vulnerability of these systems, particularly as both countries’ offensive cyber capabilities mature, may exacerbate each side’s fears about the vulnerability of its nuclear deterrent to the other side’s potential preemptive attack. For instance, if a cyberattack on NC3 could delay the other side from giving an order to execute a nuclear strike for even thirty minutes, it could potentially negate the other side’s ability to launch its intercontinental ballistic missiles (ICBM) under attack, increasing the risk to such ICBM and narrowing the potential response options for the victim’s leadership. In a future crisis, in which one side believed that the other was able and willing to stage such an attack, it could perceive itself as having extremely little time to make a decision and might employ cyber or other capabilities (e.g. nonnuclear or nuclear weapons) preemptively, or more extensively than otherwise might be the case.

Counterspace attacks also pose serious escalation risks given space systems’ relevance to nuclear operations, especially for the United States. There appears at this time little prospect that either side could substantially impact the other’s second-strike capabilities through counterspace attacks, and unclassified reporting suggests that neither the United States nor Russia has a robust counterspace capability. Moreover, even if it were able to disrupt or destroy key elements of the other side’s space architecture, each side has ground-based radars to support early warning and substantial terrestrial and/or airborne communications to support secure communications. However, as ASAT technologies improve—for instance, with the deployment of space-based interceptors or directed-energy systems—the risk of an adversary using counterspace operations to disable critical NC3 systems, intentionally or otherwise, will grow commensurately.

Long-range nonnuclear weapons could also pose a threat to strategic stability. Neither side has yet deployed conventional-prompt global-strike (CPGS) capabilities—either conventional weapons on long-range ballistic missiles or hypersonic cruise missiles—that could realistically threaten to disarm an opponent’s strategic deterrent or decapitate its NC3. However, nonnuclear precision strike appears likely to become an increasingly severe problem over time, for two reasons. First, there is concern that the launch of a CPGS missile could be mistaken for the launch of a nuclear-tipped missile, leading the side that fears attack to launch nuclear-tipped missiles in response. Second, the United States and Russia could develop and deploy sufficient numbers of highly capable CPGS weapons to imperil the strategic nuclear deterrent of the other side. Many nuclear delivery platforms, such as road- and rail-based ICBM launchers, might readily be destroyed by conventional forces if they could be effectively targeted. Moreover, future CPGS systems, some Russian and other analysts believe, might ultimately be able to destroy even more defended targets, such as hardened ICBM silos. Missile defenses could “mop up” residual second-strike forces, in turn. The counterforce threat posed by a combination of long-range nonnuclear strike and improved missile defenses is a priority concern for Russian (and Chinese) officials.

As with long-range nonnuclear weapons, neither the United States nor Russia has sufficiently capable or extensive missile-defense systems to deny the other side from being able to conduct a devastating nuclear attack, including in a second strike. Three possible future developments in missile defenses could, however, undermine strategic stability. The first is the deployment of large numbers of kinetic-kill or nuclear-tipped interceptors with the sensor capability, burnout velocity and other features required to engage CPGS and SLBM. The second is the deployment of space-based kinetic-kill interceptors. Third is the deployment of directed-energy systems for missile defense, which appear increasingly plausible as advances are made in solid-state lasers. Any of these developments could seriously threaten the viability of a nation’s second-strike capability.

Finally, the advent of autonomous systems and artificial intelligence may allow states to more reliably target adversary SSBNs and mobile ICBMs. Inasmuch as these systems form the backbone of the U.S. and Russia nuclear forces, respectively, such a breakthrough would pose a severe threat to one or both sides’ nuclear deterrents. Ultimately, however, it is very difficult in an unclassified article to assess the plausibility of developments in strategic antisubmarine warfare or the ability to target mobile ICBMs. It is possible that advances in big-data analytics, for example, yield a breakthrough in antisubmarine warfare and/or time-critical targeting of mobile missiles. Even in this case, however, it is one thing to locate a system, for instance in the middle of the Atlantic Ocean or the Siberian forest. It is another thing to be able to deliver a sufficiently destructive and accurate weapon against the targeted system before it is able to fire or conceal itself.

STABILIZING U.S.-Russian relations requires actions along each of these three pathways, conducted in parallel. Shaping and managing the overall relationship is fundamentally important. But whatever the course of U.S.-Russian relations in the future, there will remain a possibility (one, we argue, that is growing over time) of sliding into crisis and even armed conflict. Moreover, if a crisis or conflict does occur, there is a possibility (also growing over time) that escalation to strategic attack could occur. The following recommendations seek to manage these risks by helping shape the ongoing debate regarding U.S.-Russian relations and guide actions affecting U.S. nuclear posture, ballistic-missile defenses, cyber deterrence and space resilience. The recommendations also address the American role in NATO and NATO-Russia relations, both of which are of critical importance to all three pathways.

To safeguard American interests in the face of Russian actions, the Trump administration should begin by articulating a clear policy on Russia, in close coordination with Congress and NATO allies. In the absence of a coherent American approach, Russian leaders are less likely to cooperate on common interests, since Russian advocates of cooperation will wonder whether the United States will reverse itself and make them appear naive. Russian leaders are also less likely to be deterred, as advocates of a more aggressive approach can argue credibly that Russia should take advantage of a window of incoherence in Washington. And of fundamental importance, in the absence of a clear U.S. policy, Russian leaders are more likely to miscalculate how the United States will respond in a crisis—and, if a crisis does occur, be more likely to miscommunicate.

A clear U.S. policy toward Russia should include sanctions in response to Russia’s annexation of Crimea, its continued military intervention in Ukraine, and its meddling in U.S. and European elections. The absence of painful and sustained costs would prompt Putin and his leadership coterie to conclude that they have little to fear from Washington and its allies, as long as there is the thinnest patina of deniability. At the same time, however, the United States should clarify for Moscow what actions it can take (or avoid taking) over a given period of time to achieve relief from sanctions. Unconditional—or poorly conditioned—sanctions would leave Russia with little incentive to alter threatening behavior.

In addition, the United States should respond with military deployments to Russia’s violation of the Intermediate-Range Nuclear Forces (INF) Treaty. This should include strengthening America’s nonnuclear long-range strike capabilities in Europe and supporting its partners’ efforts to do so as well. Washington should also work with NATO allies to continue to improve missile defenses in Europe. It should make clear that EPAA deployment, along with other missile-defense capabilities, will have a role in deterring Russia’s use of missiles in Europe, while reaffirming that EPAA deployments in Romania and Poland will still be unable to engage Russian ICBMs aimed at the United States. The U.S. response to Russia’s INF violation should further include the deployment of a follow-on to the Tomahawk Land Attack Missile–Nuclear (TLAM-N), built with stealth features based on the Long-Range Standoff (LRSO) nuclear cruise missile. This follow-on to TLAM-N would fill a deterrence gap by adding a survivable and credible theater nuclear deterrent that complements dual-capable fighter-bombers (potentially vulnerable both to preemptive attacks on air bases and to advanced air defenses) and dual-capable long-range bombers (the use of which, if in response to theater use of nuclear weapons by Russia, would require the United States to be the first to engage in homeland-to-homeland nuclear strikes).

Finally, the United States should continue to develop areas of cooperation with Russia. Washington will need Russian support for (or abstention on) further U.N. Security Council–imposed sanctions on North Korea, and likely other threats to international peace and security that come before the council in the future. The two might productively cooperate in some areas of the Arctic, in civilian space activities, in diplomatic negotiations over Syria’s future and in the Global Initiative to Combat Nuclear Terrorism, which Russia and the United States cochair. Even as American leaders continue to press a modest, positive agenda, they should put greater priority on deterring bad behavior and avoiding a slide toward crisis and conflict.

THE UNITED States should take additional steps to mitigate the potential for attacks in cyberspace and outer space to trigger rapid, uncontrolled escalation. One of the first steps should be to define its desired rules of the road for cyberspace and outer space, not only in peacetime, but also in crisis and conflict. It should then seek consensus with key allies and partners, with whom a common understanding of preferred guidelines for offensive cyber and outer-space activities remains lacking. Armed with an allied consensus, Washington should test the degree to which arriving at a common view with Moscow (and, likely separately, Beijing) is possible. Even if the United States and Russia fail to reach a common view, well-prepared bilateral discussions regarding rules of the road in cyberspace and outer space would help clarify where various actions might fall on the escalation ladder, thereby reducing the risk that either side unwittingly takes actions viewed by the other as extremely threatening.

Next, following a framework offered in a Defense Science Board Task Force report on cyber deterrence, DOD should bolster cyber and space resilience for critical military capabilities in three ways. First, it should ensure the cyber and space resilience of the nuclear triad, and the “thin line” of NC3 systems that supports it even in a nuclear exchange. Second, the department should ensure the essential cyber and space resilience to support a select but substantial subset of nonnuclear long-range strike capabilities, such as the new B-21 bomber and JASSM-ER and attack submarines equipped with conventional Tomahawk cruise missiles. Having punishing nonnuclear strike options available for response even after withstanding the other side’s best efforts at cyber and space attacks would significantly decrease the incentive to conduct such an attack, without requiring the president to escalate to a nuclear response. Third, DOD should ensure that select offensive cyber (and, if applicable, offensive counterspace) capabilities are highly resistant to both cyber and counterspace attacks, so that the United States may respond in kind to an attack limited to cyberspace and outer space.

The United States should also improve the digital resilience of its critical infrastructure. A focused national effort sustained over a period of many years could fundamentally reduce the cyber vulnerability of at least the most essential U.S. critical infrastructure, including the electrical grid, key elements of the financial sector, water and wastewater systems, and the electoral system. There will be no quick fixes, but with strong leadership from both the public and private sectors, the United States could substantially reduce the digital vulnerabilities of select portions of its critical infrastructure over the next ten to twenty years. By addressing these vulnerabilities, Washington can reduce Russia’s incentives to attack and the potential escalatory impact if it does so.

In addition—and critically—the United States should reopen diplomatic and military lines of communication with Russia. These channels are crucial for reducing the risks of miscommunication and avoidable conflict. Notwithstanding how difficult U.S.-Russian relations are today, the United States should work to reopen channels of communication, including diplomatic as well as military-to-military ones. Some initial steps—for example, tactical deconfliction in Syria—have taken place already, but much more is needed.

JUST AS the integration of a range of new technologies is undermining strategic stability, so too is an integrated program necessary to buttress strategic stability between the United States and Russia in the coming years and decades. This program must consider changes in both nuclear and nonnuclear systems, and in both nuclear and nonnuclear strategies.

As a first step in such a program, the United States should adopt a “triad-plus” strategic force structure. This means proceeding with the Columbia-class strategic submarine modernization program, B-21 dual-capable bomber program, and LRSO missile program. It also means that, rather than pursuing a one-for-one replacement of Minuteman III ICBMs in underground silos, the United States should develop a replacement ICBM that is significantly lighter than the Minuteman III and deploy perhaps two or three hundred of the missiles in silos. The United States should also initiate a mobile ICBM research-and- development program, including prototypes, so that the United States can shift weight to a mobile ICBM force in the event of a Russian breakthrough in antisubmarine warfare (ASW). Remanufacturing a stealthy version of the TLAM-N would also allow the United States to hedge against Russian ASW improvements.

At the same time, the Department of Defense should address vulnerabilities in NC3 systems and review launch-under-attack postures. DOD should invest first in ensuring that its nuclear forces and NC3 are highly resilient to a top-tier cyber adversary. Next, both the U.S. and Russian leadership must understand the reality that their NC3 systems could suffer some degradations in crisis or conflict—some of which may not be due to attacks by the other side; a third party might attempt a false-flag operation. Accidents and acts of nature can also cause service disruptions of some systems. Both sides should ensure that their planning and exercises account for such events. Finally, the U.S. and Russian postures to be prepared to launch ICBMs under attack deserve careful review, so as to minimize the risk of launching on false warning. How both sides can adjust their postures to this effect remains difficult to say. It is important to note, however, that the more the United States hedges through other means (e.g., TLAM-N and mobile ICBMs) in the future, the less pressure there will be to launch ICBMs under the warning of attack.

As the United States proceeds with a program of nuclear modernization, it should also develop and deploy nonnuclear hypersonic weapons, tailored to defeat major improvements in the air-defense systems of potential U.S. adversaries. The United States should aim for a “sweet spot” for nonnuclear hypersonic weapons in terms of military effectiveness (high), cost (relatively low), potential for high-volume strikes (significant) and impact on strategic stability (low). Medium-range ballistic missiles (with and without boost-glide vehicles) and hypersonic cruise missiles, launched from heavy bombers and/or attack submarines, could fall in this sweet spot. The systems would have a low impact on strategic stability because their infrared and other signatures would be substantially different and distinguishable from those of U.S. nuclear-delivery systems (including SLBMs and nuclear-tipped cruise missiles), attacks on one nation would not require overflight of others, and these systems would either lack the range to attack deep into Russia (the case for submarine-based medium-range systems) or would be unable to do so in volume without creating a massive detectable signature (the case for bomber-based systems).

The United States should also invest in its missile-defense architecture. To start, as North Korea improves its nuclear-tipped ICBM capabilities, the United States should continue to grow its missile defenses. Second, the United States should continue to press forward with directed-energy systems for defensive purposes. Priority should be given to helping to deal with the immediate challenge of negating North Korean long-range and medium-range missiles, likely by deploying directed-energy systems on manned or unmanned aircraft. Third, the United States should forswear space-based missile defense interceptors and directed-energy systems, strongly urge Russia to do the same and pursue a bilateral agreement with Russia (and, separately, a bilateral agreement with China) to this end. Because of the massive and immediate threat that such systems would pose to American satellites and their essential support to warfighting—and potentially to early warning and secure communications satellites that are vital to the U.S. nuclear deterrent—any Russian deployment of space-based missile defense interceptors or lasers would pose an immediate and unacceptable threat.

Finally, the United States should regularize strategic-stability talks with Russia and seek to extend the New START treaty by five years. The U.S.-Russian meeting in Finland in September 2017 was an important first step toward a regular Track 1 dialogue on strategic stability issues in the coming years. The American and Russian governments should sustain these sorts of efforts. At the same time, however, in view of the volatility of U.S.-Russian relations at this time, and recognizing that circumstances may delay or derail Track 1 efforts, both parties should pursue Track 2 and Track 1.5 dialogues on strategic stability. Extending New START would serve strategic stability through its verification provisions, which provide transparency and predictability, thereby reducing the propensity of each side to rely on worst-case assessments. It would not currently be helpful to press for further reductions in force levels, as having some extra margin above the bare minimum force levels that each side thinks it needs will help buffer the impact of new military capabilities as they are deployed.

THE UNITED States and Russia have reentered a period of serious tension that shows no sign of abating. Relations between the two sides appear likely to remain tense, if not hostile—at least through the medium term—and may involve considerable turbulence. Bluntly put, serious disagreement and even outright conflict are possible. Exacerbating this geopolitical reality, emerging new military capabilities—cyber, space, missile defense, long-range strike and (cutting through all) autonomous systems—are increasing uncertainties associated with strategic stability. Unless measures are taken to cushion the consequences of these military trends, conflict may become more probable and escalation more dramatic and severe than they need to be—all in an era when both crisis and conflict are more plausible than they were just ten years ago. If adopted by the Trump administration and its successors, a fresh American approach to U.S.-Russian relations will protect American and allied interests. If the U.S. approach is also articulated clearly and is consistent over time, it may well reduce the risk of crisis or conflict arising from Russian miscalculation.

James N. Miller is president of Adaptive Strategies, LLC, a senior fellow at Harvard Kennedy School’s Belfer Center and served as under secretary of defense for policy. Richard Fontaine is president of the Center for a New American Security. Alexander Velez-Green is a research associate at the center.

Image: Reuters