More Missiles, More Players, More Escalation Risks

Table of Contents

The most likely pathways to nuclear escalation in the Indo-Pacific, as elsewhere, begin with the outbreak of conventional war where one or multiple nuclear-armed states are implicated. In Asia today, any major conventional war—be it on the Korean Peninsula or in the Taiwan Strait—will implicate nuclear-armed states and/or beneficiaries of U.S. extended deterrence (Australia, Japan, and South Korea). However, as described earlier, conventional and dual-capable missiles can be expected to play a prominent role early in such a conflict. While deliberate resort to nuclear first use remains a risk—particularly with North Korea—unintentional pathways to nuclear escalation stemming from the use of non-nuclear or dual-capable missiles are a prominent, yet underappreciated, risk in the region.

The core risk stems from the possibility that intense conventional military operations—particularly those involving the large-scale use of missiles against a range of military and national leadership targets—could be perceived by a nuclear-armed state as targeting its nuclear operations or capabilities even when the intention behind such an operation was more limited in scope. This risk is particularly acute given the growing emphasis by regional militaries on holding at risk nuclear force assets with conventional weapons. The growing pursuit of conventional counterforce strategies presents serious escalation risks that continue to be largely discounted by planners and policymakers.

Unintentional escalation risks encompass inadvertent and accidental escalation. The latter may arise when missiles malfunction—particularly when tensions are high in a crisis. Inadvertent escalation concerns scenarios where the effects of military operations are greater in scope than intended—or are perceived to be so. While early scholarship on inadvertent escalation focused on the idea that “intense conventional operations may cause nuclear escalation by threatening or destroying strategic nuclear forces,”¹ prominent contemporary analyses in the post–Cold War era emphasize the possibility for escalation through other means, including the entanglement of conventional and nuclear command, control, communication, and intelligence systems.² Both sets of inadvertent escalation risks are prominent in the Indo-Pacific today. These risks could manifest without the use of missiles in war, but regional planners and policymakers should pay particular attention to missiles, due to many of the same characteristics that raise their appeal for regional states.

A New Age of Conventional Counterforce?

In East Asia, the appeal of using precise conventional long-range strike systems to destroy or degrade adversaries’ nuclear capabilities—what might be termed conventional counterforce—has steadily risen in recent years. North Korea’s development of missile-mateable nuclear warheads and its increasingly sophisticated array of strike systems has primarily driven this trend. South Korea’s approach to coping with North Korea’s asymmetric nuclear capabilities emphasizes the use of conventional missiles to strike launchers, command and control systems, and other support infrastructure for Pyongyang’s nuclear forces. Similarly, the growing scope of North Korea’s capabilities has been frequently cited by Japanese proponents of longer-range missile capabilities; these proponents note that Japan should possess the capability to destroy North Korean launchers. North Korea’s advancing missile capabilities are a core justification for these capabilities in Tokyo’s updated 2022 National Security Strategy and National Defense Strategy. Seoul’s investments in enabling capabilities and missiles over decades leave it as the most capable pursuer of conventional counterforce strategies in the region today, but Japan is expected to devote substantial resources through the 2020s and into the 2030s to attain similar capabilities. Many of these capabilities would be adaptable for use against targets in China—particularly as the ranges of missile capabilities in both countries may increase over time. Japan’s 2022 National Defense Strategy underscores that counterstrike capabilities are “key to deterring invasion against Japan,” a threat that Tokyo does not perceive from North Korea but does to a much greater extent from China (especially in the East China Sea).³

Beyond military rationales for these capabilities, political factors are salient as well. In the absence of any diplomatic measures to restrain Pyongyang, for instance, leaders in both Japan and South Korea have sought to communicate to their respective publics that their militaries are planning and posturing to manage the consequences of any war with North Korea. In South Korea, in particular, the return of a conservative government in 2022 under Yoon, paired with poor diplomatic prospects, has resulted in a concerted and repeated emphasis on Seoul’s conventional counterforce options in the form of the Kill Chain and KMPR strategies.⁴ The Yoon administration has also demonstrated greater risk acceptance more generally. For instance, it responded to a violation of South Korean airspace by North Korean drones in December 2022 by sending drones into North Korean airspace—a response that was deemed a violation of the 1953 Korean War armistice by the United Nations Command.⁵ Yoon criticized the South Korean military’s failure to interdict the North Korean drones and blamed his predecessor’s policies for the incident.⁶

Plans for the strict conventional counterforce of nuclear weapons have not generally been a prominent feature of military planning through much of the nuclear age. There have been two prominent instances of practical planning and high-level political consideration of the use of non-nuclear weapons against nuclear-armed systems: the Cuban Missile Crisis, when certain advisers to U.S. president John F. Kennedy promoted the idea of a surprise aerial bombardment campaign against Soviet nuclear-capable missiles in Cuba, and inquiries by U.S. president Obama about the feasibility of comprehensively destroying North Korean nuclear targets with the exclusive use of non-nuclear U.S. capabilities.⁷ In both examples, despite the availability of vastly different strike platforms and enabling technologies, U.S. political decisionmakers opted against conventional counterforce plans, which they saw as being too risky. Other instances of such planning took place during the Cold War—notably, against China’s nascent nuclear force in the mid-1960s—but decisionmakers did not rule out the potential use of nuclear weapons.⁸ Similarly, while NATO and the Warsaw Pact each planned to hold at risk the other’s forward-deployed nuclear assets in Europe with conventional weapons during the Cold War, both sides attached importance to their regional nuclear systems for this task as well.⁹ Similarly, U.S. Navy anti-submarine warfare planning efforts against Soviet strategic ballistic missile submarines during the Cold War did not exclusively rely on conventional armaments.¹⁰

Political leaders contemplating a conventional counterforce strike—especially to preempt an adversary attack—will seek high assurances that military plans against another state’s nuclear forces will be highly likely to succeed in destroying all targets that could contribute to a massively damaging nuclear attack. Failing this, they will seek high assurances that whatever proportion of targets could not be destroyed could be addressed by active defenses, such as missile defenses, and that further damage could be mitigated by passive defenses, including civil defense. (The availability of missile defenses could lead decisionmakers to tolerate a lower probability of success for an initial strike, under the assumption that residual inbound missiles could be managed by such defense systems.) A final factor deserving of consideration is that the probability of successful preemption of nuclear forces would likely significantly decline in a conventional war already underway, when nuclear warheads and mobile launchers would be generated and dispersed.

Risks and Obstacles to Effective Conventional Counterforce

While many Japanese and South Korean defense planners privately recognize the limitations of conventional counterforce plans in blunting the totality—or even just a substantial part—of North Korea’s forces, they nevertheless point to the damage-limiting benefits of any counterforce strategy. Furthermore, South Korea, in particular, has long seen promise in threatening to hold at risk the North Korean leadership—specifically, Kim Jong Un, who is the sole known release authority for nuclear weapons in the country. Since the Park administration (2013–2017), Seoul has made explicit its intentions to kill Kim in retaliation for any North Korean nuclear attack. These intentions, however, coexist with Seoul’s separate-but-related plans to preempt North Korean missile launches (the Kill Chain strategy). Disambiguating the two strategies and assuring Kim that preemptive decapitation would not be part of South Korea’s warfighting approach has not been a prominent focus of Seoul’s messaging efforts. As a result, North Korea has behaved in ways that strongly imply that it perceives a preemptive decapitation strike as a primary threat to be deterred. For instance, to deter this perceived threat, North Korea updated its nuclear doctrine in September 2022 to explicitly threaten the “automatic and immediate” release of any and all nuclear weapons that would be available to the country’s military should Kim be killed or should its nuclear command and control systems otherwise be degraded (presumably through conventional operations, but possibly through U.S. nuclear strikes). While Kim’s implementation of such a dead-hand arrangement in practice will likely depend on advances in North Korean nuclear command and control systems, this step underscores the severity with which Pyongyang views Seoul’s leadership targeting plans.

For Seoul and Tokyo, apart from targeting North Korea’s nuclear command and control, which will remain tempting as long as the country does not move toward the delegation or pre-delegation of nuclear use authority, plans largely focus on employing precision strike systems against Pyongyang’s missile launchers. This approach borrows from post–Cold War doctrinal preferences in the United States, where the advent of precision-guided munitions and wars against technologically inferior adversaries prompted a greater interest in shooting the archer, metaphorically speaking.¹¹ The canonical case is that of the so-called Scud hunt in the 1991 Gulf War, when U.S. forces sought to find, fix, and finish Iraq’s Al-Hussein Scud-B-variant short-range ballistic missiles.¹² Partly due to intelligence limitations and partly due to active and passive deception efforts by Iraqi forces, U.S. and coalition forces faced substantial difficulties in completing real-time assessments of Iraq’s mobile Al-Hussein missiles. Fixed targets and supporting infrastructure for missile launchers were less of a problem, but Iraqi mobile missiles remained exceptionally survivable and were able to successfully carry out launches. The coalition eventually adopted high-intensity aerial sorties of F-15E and F-16L fighters over kill boxes—predefined geographic areas where mobile missile units were expected to be operating—but still failed to blunt Iraqi launches.¹³

Intelligence, surveillance, and reconnaissance platforms and technologies have since improved—and substantially so. For instance, Israeli military intelligence was considerably more successful in cuing Israeli Air Force assets against mobile missile launchers in the 2006 Israel-Hezbollah War.¹⁴ Paired with initial preemption against known munitions storage sites, the Israeli Air Force succeeded in destroying “most” 202-millimeter and 302-millimeter rocket artillery systems available to Hezbollah.¹⁵ Despite these general technology improvements, the scope of North Korea’s missile capabilities is far greater, in both qualitative and quantitative terms, than that of Iraq’s missile forces in 1991 or Hezbollah’s in 2006. (And North Korea’s terrain and hundreds of underground facilities countrywide are much more amenable to concealing missile units than the deserts of Iraq, for example.) Moreover, the costs of failure to coalition forces in the first Gulf War were limited due to the imprecise, conventional nature of the Al-Hussein missiles, whereas nuclear-capable North Korean missiles would inflict massive amounts of damage against military and civilian targets in a war, demanding greater prudence in military planning.

Partly in recognition of this challenge, South Korea has invested considerable analytical efforts to improve so-called counterbattery targeting—even before North Korea’s nuclear-capable missile forces began to grow. Seoul has invested in new counterbattery radars and sensors to allow it to find, fix, and finish North Korean artillery and missile launchers, but only after they have fired and revealed their locations.¹⁶ Because Seoul has long faced a substantial threat from North Korean artillery systems, its investments and expertise in counterbattery planning are long-standing. Seoul has broader plans to adopt space-based optical sensors to aid in the tracking of North Korean military forces, but even these systems will be limited in their ability to abet the targeting of mobile missiles.¹⁷ Japan, meanwhile, is less experienced in this area and has faced apparent difficulties in properly tracking and characterizing the trajectories of certain North Korean missile launches.¹⁸ While Japanese and South Korean defense officials and military planners largely understand the difficulties of targeting mobile missiles, which can launch from nonpredetermined sites (sometimes termed launch pads), some civilian leaders and politicians in both countries appear to believe that North Korea relies on fixed, known launch sites. Such a belief may have had a role in the development and sustainment of these plans, which have received high-level political sanction in both countries.

There is likely no feasible conventional strike option that, if enacted, could be certain to spare Japan and South Korea nuclear retaliation by North Korea.

Whatever efforts Seoul and Tokyo make to address the substantial demands of conventional counterforce targeting, it appears increasingly likely that the rate of qualitative refinement and quantitative growth in North Korea’s missile arsenal will allow it to remain several steps ahead of its regional adversaries. With its pursuit of more responsive solid-propellant missiles, off-road-capable transporter erector launchers, an ever-expansive network of underground facilities and missile drive-through shelters, and extensive use of camouflage and concealment, targeting North Korean mobile missiles will remain highly challenging and the task of comprehensive counterforce intractable. In other words, there is likely no feasible conventional strike option that, if enacted, could be certain to spare Japan and South Korea nuclear retaliation by North Korea. Even if North Korean ballistic missile operating areas were well-understood, and if fixed facilities associated with the maintenance of missile bases were known—two open questions given Pyongyang’s penchant for secrecy—the levels of confidence required to assure that most or all of North Korea’s nuclear-capable systems would be destroyed will likely prove unattainable. The United States and its allies would likely face the greatest odds of success in a bolt-from-the-blue strike, but in any sufficiently advanced crisis that may transgress into a full-scale war, North Korea would likely have raised alert levels and dispersed its launchers.

Given these expectations, military planners and political decisionmakers in Japan and South Korea should seek to better understand the scope of what their conventional counterforce strategies could achieve in practice and the consequences of failure. They should also study the pathways to unintentional escalation that will remain with conventional counterforce planning. The earlier-described challenge to successfully executing a conventional counterforce campaign against North Korea underscores the most likely path to large-scale nuclear use, that is, if the combined forces of the United States and its allies were unable to destroy all possible vectors of North Korean nuclear employment in a war with conventional means. Another pathway to nuclear escalation concerns Pyongyang’s own doctrinal preference for both nuclear and conventional preemption. Even as North Korea’s nuclear and missile forces grow large and sophisticated enough to be effectively invulnerable to a conventional “splendid first strike,” Pyongyang will continue to perceive a high premium on shooting first if it assesses that a major crisis may have the potential to lead to significant military action by the United States and its allies.

Though North Korea is not the first nuclear-armed state in a competitive dyad to seek nuclear weapons out of a recognition of its own conventional military weaknesses, Pyongyang’s contemporary conventional limitations compared to its adversaries are especially severe. North Korean infantry, mechanized forces, air forces, naval forces, and air defenses rely largely on obsolete technologies and platforms and, due to resource shortages, would face serious difficulties sustaining high-intensity warfare for more than a few weeks. In a 2021 public assessment of North Korea’s military capabilities, the U.S. Defense Intelligence Agency identified “logistics for sustained combat operations” as a key vulnerability facing the country’s armed forces.¹⁹ As a result, North Korea is explicit about its intentions to use nuclear weapons to “repel” the conventional military forces of Japan, South Korea, and the United States. By doing so, Pyongyang hopes to degrade the willingness and ability of the allies to carry out conventional military operations under favorable conditions. Certain capabilities, such as F-35A stealth fighters, are of particular concern to Pyongyang,²⁰ given the near-total inability of its air defense radars to detect and engage such a system. North Korean state media accounts of U.S.–South Korea military activities have expressed the view that F-35A fighters could be used “in a bid to mount a ‘preemptive attack’ on [North Korea].”²¹ These capabilities would be high priorities for North Korean preemption. North Korea would also seek to target South Korea’s mobile missile units, whose peacetime basing areas are generally known, and related command and control facilities early in a war to prevent their use against its launchers. The presence of strong preemptive incentives in both countries is deeply destabilizing and reflects that they both have largely succumbed to what Thomas Schelling once described as the “reciprocal fear of surprise attack.”²² Even without overtly planning for preemption of imminent attacks, merely possessing the capabilities, as Japan plans to do pursuant to its updated National Security Strategy, will carry destabilizing risks, potentially prompting a North Korean nuclear attack when either no attack or a conventional attack had been originally planned. For Japan and South Korea, some of these risks could be mitigated through the adoption of an exceptionally high intelligence standard for assessing that an attack is imminent.

Risks in a U.S.-China Conflict

While the risks stemming from conventional counterforce planning around North Korea are substantial, a potentially much greater medium-term risk in the Indo-Pacific pertains to possible nuclear escalation with China. In 2019, the U.S. Department of Defense’s annual report to lawmakers on Chinese military capabilities acknowledged, for the first time, that “adversary attacks against Chinese conventional missile forces-associated [command and control] centers could inadvertently degrade Chinese nuclear [command and control] and generate nuclear use-or-lose pressures.”²³ While the adversary went unnamed in the report, it was likely at the time that this referred to the United States. Increasingly, however, U.S. treaty allies—and even Taiwan—will be able to deliver equivalent effects using their own independently fielded and operated long-range strike capabilities. In a high-intensity conventional conflict in East Asia, it is likely that the United States would be implicated alongside its allies. An unaddressed and underappreciated escalation risk here concerns possible inadvertent strikes on Chinese nuclear-related command and control or other facilities by a U.S. ally that Beijing interprets as having originated from a U.S. launcher or platform. While China may not reasonably fear that Australia or Japan would attempt a conventional counterforce attack on its nuclear forces, it may fear such an attack by the United States. This has been a long-standing concern for Chinese officials and experts. Allied targeting could aim to hold at risk a wide array of Chinese capabilities, including Beijing’s theater-range, dual-capable missile systems and associated infrastructure. Allies could further aim to hold these capabilities at risk, either with the intention of limiting damage amid fears that China could resort to nuclear first use or simply blunting Beijing’s conventional strike capabilities.

Just as the United States has traditionally disfavored nuclear proliferation by its allies partly out of its interest in being the sole decider of when the nuclear threshold might be deliberately crossed, so too does it today have an interest—publicly unacknowledged so far—in ensuring that its allies do not inadvertently contribute to crisis escalation dynamics. To be sure, planners in the United States have readily imagined that allied strike and other capabilities could contribute, in a coordinated manner, to a dedicated offensive campaign. The 2019 Missile Defense Review, for instance, notes explicitly that the United States will “seek to integrate U.S., allied, and partner capabilities for . . . attack operations capable of striking the entire range of infrastructure supporting adversary offensive missile operations.”²⁴ The opposite—when an ally’s military operations inadvertently escalate beyond a threshold that the United States itself is not ready to cross—is less considered. Formally, the problem for the United States is one of allied entrapment. For some allies, escalatory action could be partly precipitated by fears of abandonment by the United States and thus deliberately pursued to catalyze and ensure U.S. involvement. For instance, North Korea’s development of ICBMs has raised the salience of decoupling in assessing the credibility of Washington’s extended deterrence reassurances to Seoul and Tokyo.²⁵

Both Chinese and North Korean nuclear forces continue to grow in ways that, if other relevant factors remain unaltered, should give their respective leaderships greater reason to withstand use-or-lose pressures in a crisis—particularly against conventional attacks. But this logic may not hold in a crisis and can hardly be relied on to mitigate the risk of nuclear escalation, which may depend more on subjective leadership perceptions that are susceptible to cognitive biases, for example. Given deterrence objectives in both countries, these problems may not be solvable, but the prospect of managing their consequences in potential crises would benefit from new institutional processes within U.S.-allied states to fully consider the risks of deliberate conventional counterforce strategies and the implications of inadvertent targeting in wartime. They will also depend on frank and honest consultations with the United States. But the United States is unlikely to replace its allies’ capabilities with offerings of its own to seize control of escalation through long-range conventional strikes. Allies’ investments in strike capabilities, as described previously in this report, are partly a means of enhancing self-reliance and potentially even hedging against longer-term concerns about the political reliability of the United States as an ally.²⁶

The Risks of Accidental Escalation

No weapon system, past or present, exhibits perfect reliability. The same goes for missiles of all types. Accidents involving missiles can and do occur, either due to human error or technical malfunction, and this has been true throughout the missile age. The possibility of accidents involving missiles presents an important pathway toward escalation, including possible nuclear escalation, stemming largely from the proliferation of non-nuclear missiles and non-surface-attack missiles, such as air defense missiles. These concerns are far from theoretical. In 2022 alone, the world witnessed multiple instances of accidents involving missiles. In March 2022, an Indian BrahMos supersonic land-attack cruise missile was unintentionally launched into Pakistani territory, where it made impact without killing anyone. Due to relatively calm background conditions between the two countries, Pakistan did not interpret the event as an intentional attack but appeared to properly assess the event as an accident, demanding an explanation from India.²⁷ New Delhi’s investigation later determined that the incident was due to operator error and reprimanded Indian Air Force officers involved in the accidental launch.²⁸ This event was the first of its kind between two territorially contiguous nuclear rivals. A similar incident took place in 2016 when a Taiwanese warship accidentally released a supersonic Hsiung Feng III anti-ship cruise missile westward toward China, striking a civilian fishing boat and killing its captain.²⁹ Once again, due to a generally calm state of affairs between China and Taiwan at the time, the incident did not escalate, with China’s Taiwan Affairs Office demanding a “responsible explanation” for the incident.

Accidents have also occurred under more tense background conditions. Two such cases took place in 2022, amid spiraling demonstrations of resolve between North Korea and South Korea. In October 2022, North Korea launched an intermediate-range ballistic missile over Japan and into the northern Pacific Ocean, the first time it carried out such a launch since September 2017. In response, South Korea and the United States carried out a joint missile exercise, each launching two ATACMS surface-to-surface missiles. In addition, South Korean forces launched one Hyunmoo-2 short-range ballistic missile, but this launch was not publicly announced like the ATACMS launches were. At around 11:30 p.m. on October 4, 2017, civilians near the South Korean city of Gangneung reported hearing a loud explosion and seeing a fire near the South Korean Air Force’s 18th Fighter Wing Base.³⁰ Social media footage of the event quickly went viral, with some people expressing concern that the event could have been the start of a North Korean attack. Tensions between the two countries had been simmering for weeks by the time of the incident, so that conclusion was not entirely unlikely. South Korean authorities later confirmed that a Hyunmoo-2 missile had been launched and had failed, resulting in the explosion.³¹ “Immediately after its launch,” one anonymous South Korean military official noted to the press, “the missile flew inland instead of toward the sea and abnormally landed on a golf course on the base at a location approximately 700 meters from a civilian residential area.”³² Given the northerly launch site for this missile, a booster failure at a higher altitude could have resulted in the missile transgressing the inter-Korean Military Demarcation Line and striking North Korean territory. South Korean authorities also noted that the missile’s warhead detached from the booster and did not explode; it is unclear if the lack of a detonation was due to the status of the missile’s safing, arming, fuzing, and firing system or due to chance.³³

Weeks later, North Korea carried out an unprecedentedly intense spate of missile launches, responding to aerial exercises by the United States and South Korea and launching more than twenty missiles, the most it ever launched in a single day.³⁴ One missile landed 57 kilometers off the South Korean coast, an unprecedentedly southeasterly splashdown point for any North Korean missile launch. While South Korean officials initially described the event as “intolerable” and perceived it to be an intentional provocation by North Korea,³⁵ recovery of the missile’s debris through a salvage mission revealed it to be an old, Soviet-era SA-5 air defense missile that had been launched in surface-attack mode.³⁶ SA-5 air defense missiles have prominently gone off course in other scenarios; in 2019, a Syrian-launched SA-5 aimed at an Israeli fighter aircraft landed in Cyprus, for instance.³⁷ A North Korean statement after the incident appeared to reject that it intentionally tried to strike that particular aimpoint in the Sea of Japan, underscoring that South Korea was “claiming” that North Korea had fired a missile near its territorial waters. (In that same statement, North Korea did claim that it intentionally launched cruise missiles 80 kilometers off of South Korea’s coast at an even more southeasterly aimpoint.³⁸) The SA-5 was launched alongside a diverse array of North Korean missiles, including newer, solid-propellant missiles and older, Scud-variant missiles. It cannot be ruled out that the unprecedented impact off South Korea’s coast was due to a technical malfunction; under other circumstances, such a missile could have veered off course more substantially and landed on South Korean territory.

A final accidental missile event in 2022 took place in eastern Poland, along the Ukrainian border, in November. In response to Russia’s large-scale launches of cruise missiles, Ukraine fired an unknown number of air defense interceptors. One such interceptor—a Soviet-origin S-300—crossed into Polish territory and struck a village, killing two people. In the immediate aftermath of the event and before the event had been properly analyzed by NATO, an anonymous senior U.S. intelligence official told the Associated Press that “Russian missiles crossed into NATO member Poland.”³⁹ This report, which was quickly disseminated worldwide due to the Associated Press’s wire services, prominently influenced perceptions, with analysts and even some officials from eastern NATO member states attributing the missile detonation in Poland to Russia and suggesting that it may have been deliberate. Senior Ukrainian officials, prior to NATO’s assessment that the missile as an errant interceptor, alleged that it was a “conspiracy theory” that the missile was an interceptor.⁴⁰ The event did not lead to broader NATO-Russia escalation due to a conclusive assessment by the alliance that the missile was not of Russian origin. But the early public confusion underscored the dangerous possibility of escalation. Here again, background conditions played an important role: observers appeared motivated to interpret the limited available facts early on as a deliberate Russian attack due to the ongoing war and perceptions that Russia’s leadership might seek to deliberately probe NATO’s thresholds for escalation.⁴¹ Prior to NATO reaching a conclusive assessment, Poland had intended to initiate Article IV consultations within NATO, which it later remanded, indicating that it did not view the event as a deliberate attack and saw no need to escalate.

These incidents—all from 2022 alone—underscore that the risk of unintended and accidental escalation stemming from the use, testing, and operation of missile forces is all too real. Because every missile will have an associated nonzero probability of failure in the course of ordinary use, the firing of missiles—for testing, demonstrative, or operational purposes—manifests some risk that the missiles will fail or otherwise behave in an undesired manner. While not all failures will lead to escalation, some types of failure are clearly riskier than others. The earlier examples all underscore, for instance, that background conditions are highly germane to how decisionmakers might interpret a given accident when only limited information is available. Pakistan was disinclined to view India’s accidental cruise missile launch as a deliberate attack due to the relatively calm state of bilateral relations, while certain NATO states appeared motivated to portray the Ukrainian S-300 misfire as a deliberate Russian attack due to the extremely poor state of relations between Moscow and NATO during the Russia-Ukraine war. The inter-Korean examples, by contrast, underscore the possibility of missile-related accidents catalyzing escalation despite a shared interest by both sides in demonstrating resolve without upending the broader status quo. An errant South Korean Hyunmoo-2 landing on North Korean territory or an errant North Korean SA-5 landing on South Korean territory in a broader tit-for-tat spiral between the two sides could well spark broader escalation, particularly given the preemptive incentives for both countries described in this chapter. While wars might not emerge out of times of peace purely by accident,⁴² the prospects for accidents to stoke crises in escalatory ways should not be easily dismissed.

The frequent brandishing of missile capabilities in the Indo-Pacific—in peacetime and crisis, from the Korean Peninsula to the Taiwan Strait—underscores the importance for decisionmakers to understand accidental escalation risks. Other plausible accidental pathways to escalation include debris from non-notified North Korean missile tests striking civilian aircraft or ships, or missiles failing in overflight of another state’s territory. Even if such incidents are unlikely to precipitate a decision to massively retaliate, they could prompt limited retaliatory actions that could lead to broader escalation. The latter is particularly concerning as missiles’ overflights of populated areas are becoming more frequent in Asia. In August 2022, China, for the first time, launched multiple DF-15 ballistic missiles over Taiwanese territory—including over densely populated urban areas.⁴³ North Korea has overflown Japan’s Tsugaru Strait in an apparent bid to carry out long-range missile tests while minimizing its overflight of more populated Japanese regions in Honshu and Hokkaido. While in both cases, the ballistic trajectories of the involved missiles ensure that RVs are well outside of the earth’s atmosphere, an unintended in-flight failure could result in a range of accidental outcomes that could prove escalatory. Failures are particularly possible in developmental testing: at least one North Korean intermediate-range ballistic missile failed in flight after a significant period of boost-phase flight and crashed into a populated area on the country’s own soil.⁴⁴ While legal definitions concerning the altitudinal limits of sovereign airspace and space are gray, states generally find missile overflight of their national territory objectionable.⁴⁵ Notably, however, Taiwan appeared to underplay China’s missile overflight of its territory in August 2022 by underscoring that the missiles flew “beyond the atmosphere.”⁴⁶

Persistent hotlines, while not without their shortcomings, are a somewhat classic solution, but they have seen limited implementation in East Asia.

Fortunately, some solutions may help manage what otherwise might be highly escalatory accidents. Persistent hotlines, while not without their shortcomings, are a somewhat classic solution, but they have seen limited implementation in East Asia. Inter-Korean hotline communications, for instance, tend to thrive when relations are good between the two countries and are otherwise unattended by the North Korean side when tensions are high—precisely when a hotline would be most useful. (The hotline between United Nations Command and North Korea’s Korean People’s Army has proven somewhat more resilient.) A similar dynamic appears to be at play in the U.S.-China relationship; attempts by senior U.S. officials to use existing crisis communications as a high-altitude Chinese surveillance balloon intruded into U.S. airspace in February 2023 were met with silence.⁴⁷ In April 2023, China and Japan established a new hotline to manage incidents at air and sea, but the efficacy of that hotline remains untested.⁴⁸

Psychological factors may also complicate the practical utility of hotlines—particularly in fast-moving crises. The ability to quickly communicate with one’s adversary to explain the accidental nature of a missile launch event or other technical malfunctions may be essential in shaping what could otherwise be a highly uncertain information environment where decisionmakers could be primed to interpret benign accidents as the start of a highly threatening preemptive attack. However, decisionmakers could also be disinclined to take at face value any assurances delivered by an adversary through a hotline. Uncertainty about intentions can further prompt certain psychological biases, such as confirmation bias, to lead an adversary to conclude that an accident is, for instance, deliberate and thereby beget an escalatory response.

Beyond hotlines, missile tracking technologies may help avert escalation. A robust capacity to understand and assess missile events can help prime decisionmakers toward averting escalation. But technology is unlikely to be a panacea given that proper and complete event assessments can take hours, as demonstrated by the NATO-Russia example above, and escalation could take place on shorter timescales (particularly in cases where strong incentives to preempt exist, such as on the Korean Peninsula).

Ambiguity, Speed, and Decisionmaking Under Pressure

Many of the physical characteristics of missiles are likely to contribute to nuclear escalation risks in the Indo-Pacific. Speed is the most familiar of these; ballistic missiles, since the German V-2 in 1944, have provided a prompt strike capability outclassing crewed air platforms substantially. This remains true in Asia, where even legacy, Scud-derived North Korean missiles have flight times measured in minutes to targets in South Korea. Intermediate-range North Korean missiles, such as the Hwasong-12, once launched could reach the U.S. territory of Guam in just around twenty minutes. More advanced payloads, such as HGVs, can exhibit similar speeds but are often slower to target than equivalent-range ballistic missiles. What renders HGVs notable is thus not their speed—missiles have exhibited hypersonic speeds in their terminal phase for decades—but their endoatmospheric maneuverability through much of the midcourse flight phase. For instance, while advanced postlaunch tracking of a ballistic missile may allow for a defender to reasonably extrapolate a possible target or a possible range of targets, doing so for longer-range HGVs may be less feasible. Cruise missiles present a similar set of challenges but are generally much more difficult to detect upon launch and track in flight than either HGVs or ballistic RVs.

With the exception of the United States, no regional state possesses a robust enough missile tracking and warning capability to maintain high-fidelity situational awareness about missile events after launches have been carried out. Russia’s missile tracking and warning capabilities in its Far East region are likely second to the United States, but evidence suggests substantial gaps in Moscow’s ability to accurately interpret missile events in the Pacific.⁴⁹ With Russian assistance, China started to develop a multimodal missile warning and tracking system in the latter half of the 2010s. Little is known about the fidelity of these capabilities, but the primary motivator of their development appears to be the goal of enabling a launch-on-warning posture for Chinese strategic forces.⁵⁰ As a result, China’s terrestrial and space-based missile tracking capabilities may be especially calibrated to detect strategic missile attacks from the United States. Even the United States—which possesses a diverse array of land-, sea-, and space-based sensors, occasionally supported by deployed air-based sensors—has mischaracterized missile launches. In early 2022, for instance, a North Korean test launch of a MaRV-equipped ballistic missile prompted the U.S. Federal Aviation Administration to ground air traffic on the West Coast amid concerns that the missile may have been due for the continental United States.⁵¹ Though this error was quickly rectified, similar false positive signals in a crisis could prove escalatory, for instance, if they were misperceived as potentially targeting national leadership. Though the United States is pursuing new sensors to adapt to the proliferation of longer-range HGVs, sensor error or analytical error in crises could prompt an escalatory response when one is unnecessary.⁵² The general lack of robust postlaunch trajectory, payload, and target characterization capabilities in the Indo-Pacific means that certain forms of ambiguity pertaining to missiles may be more pertinent prior to launch, as forces are being massed or otherwise prepared for use.

This points to a prominent source of inadvertent escalation risk concerning missile capabilities in East Asia today: the growing proliferation of dual-use nuclear-and-conventional-warhead-capable ground-launched missiles, primarily in China but also increasingly in North Korea. The primary danger stems from prelaunch warhead ambiguity, which could lead military planners to mischaracterize targets prior to an attack. An attack intended to destroy a given missile unit or launcher assessed as conventional that turns out to be deployed with a nuclear weapon could be misinterpreted by the target country as the start of a broader counterforce campaign.⁵³ As countries including Japan and South Korea posture their long-range strike capabilities to explicitly hold at risk North Korean missile launchers, their ability to properly characterize and discriminate nuclear-armed missile units from non-nuclear ones will have important implications for escalation management and control.

For instance, in a limited, conventional war, either Seoul or Tokyo may seek to carry out retaliatory conventional strikes on a specific North Korean unit that may have been implicated in a conventional missile launch against their territory. If the North Korean unit was operating a dual-capable system—possibly with nuclear warheads colocated nearby—Pyongyang may have reason to interpret such a strike as the start of a broader disarming counterforce attempt and respond with massive nuclear use, even if its military planners would otherwise anticipate attacks intended to attrite its conventional forces. As North Korea moves toward the deployment of tactical nuclear weapons, this problem is likely to be further amplified as nuclear warheads may be stored and colocated with a greater number of missile units. Similar concerns exist with the potential targeting of Chinese missiles systems, such as the DF-26, but most plausible scenarios where U.S. allies are carrying out deep strikes on Chinese soil likely involve a larger-scale regional conventional war where missiles may have been broadly used by both sides against a range of targets. China’s adoption of dual-capable missiles may not be driven entirely by deliberate deterrence considerations but also by technical and bureaucratic factors.⁵⁴ Similar factors may be at play in North Korean decisionmaking concerning dual-capable launchers. Neither country, however, may see reason to abandon this practice.

Regional military and policy establishments generally underrate the possibility that unintentional escalation pathways could have a bearing on crises and overrate their ability to control the full extent of escalation.

The full range of unintentional escalation pathways described earlier deserve serious consideration by regional military planners and political decisionmakers. Regional military and policy establishments generally underrate the possibility that unintentional escalation pathways could have a bearing on crises and overrate their ability to control the full extent of escalation. For Australia and Japan, both of whom are comparatively less experienced with the challenges of operating long-range strike systems, escalation concerns should be an integral part of developing new operational plans and procedures for the use of missiles in times of conflict. For the United States, it will be especially important to begin incorporating detailed, scenario-based discussions of escalation dynamics in policy and military consultative forums with Indo-Pacific allies. Washington and its partners may have divergent beliefs about the salience of specific escalation risks, which may remain undiscovered until a crisis presents itself. Addressing such divergences with frank and open exchanges during relative peacetime can ensure that military operations can proceed with the least amount of added risk should a conflict present itself.

Notes

¹ Barry R. Posen, “Inadvertent Nuclear War?: Escalation and NATO’s Northern Flank,” International Security 7, no. 2 (1982): 28–54, https://doi.org/10.2307/2538432; italics in original.

² James M. Acton, “Escalation Through Entanglement: How the Vulnerability of Command-and-Control Systems Raises the Risks of an Inadvertent Nuclear War,” International Security 43, no. 1 (August 1, 2018): 56–99, https://doi.org/10.1162/isec_a_00320.

³ “National Defense Strategy (Provisional Translation),” Ministry of Defense of Japan, December 16, 2022, 13, https://www.mod.go.jp/j/approach/agenda/guideline/strategy/pdf/strategy_en.pdf.

⁴ Josh Smith, “Analysis: South Korea Doubles Down on Risky ‘Kill Chain’ Plans to Counter North Korea Nuclear Threat,” Reuters, July 26, 2022, https://www.reuters.com/world/skorea-doubles-down-risky-kill-chain-plans-counter-nkorea-nuclear-threat-2022-07-26.

⁵ Shreyas Reddy, “UN Command Says Both Koreas Violated Armistice With Drone Intrusions,” NK News, January 26, 2023, https://www.nknews.org/2023/01/un-command-says-both-koreas-violated-armistice-with-drone-intrusions.

⁶ Hyonhee Shin, “South Korea’s Yoon Slams Response to North Drones, Vows to Create Drone Unit,” Reuters, December 27, 2022, https://www.reuters.com/world/asia-pacific/south-koreas-yoon-vows-advance-creation-military-drone-unit-2022-12-27.

⁷ Joshua H. Pollack, Cristina Varriale, and Tom Plant, “The Changing Role of Allied Conventional Precision-Strike Capabilities in Nuclear Decision Making,” Nonproliferation Review 27, no. 1–3 (January 2, 2020): 26–27, https://doi.org/10.1080/10736700.2020.2003561.

⁸ Jim Mann, “U.S. Considered ’64 Bombing to Keep China Nuclear-Free,” Los Angeles Times, September 27, 1998, https://www.latimes.com/archives/la-xpm-1998-sep-27-mn-26986-story.html.

⁹ NATO, in particular, sought to offset its local conventional inferiority in Central Europe by resorting to early nuclear use. See “MC 14/3(Final) Overall Strategic Concept for the Defense of the North Atlantic Treaty Organization Area,” NATO, January 16, 1968, https://www.nato.int/docu/stratdoc/eng/a680116a.pdf.

¹⁰ Owen R. Cote, Jr., The Third Battle: Innovation in the U.S. Navy’s Silent Cold War Struggle With Soviet Submarines: Naval War College Newport Papers 16 (Naval War College Press, 2012), https://apps.dtic.mil/sti/pdfs/ADA421957.pdf.

¹¹ Guillaume Lasconjarias and Jeffery A. Larsen, eds., “NATO’s Response to Hybrid Threats,” NATO Defense College, Forum Papers Series no. 24, 2015, 59, https://www.files.ethz.ch/isn/195405/fp_24.pdf.

¹² William Rosenau, “Coalition Scud-Hunting in Iraq, 1991,” in Special Operations Forces and Elusive Enemy Ground Targets, 1st ed., Lessons from Vietnam and the Persian Gulf War (RAND Corporation, 2001), 29–44, http://www.jstor.org/stable/10.7249/mr1408af.9.

¹³ James W. MacGregor, “Bringing the Box Into Doctrine: Joint Doctrine and the Kill Box,” School of Advanced Military Studies, May 26, 2004, https://apps.dtic.mil/sti/citations/ADA429320.

¹⁴ Uzi Rubin, “The Rocket Campaign Against Israel During the 2006 Lebanon War,” Mideast Security and Policy Studies 71 (Bar-Ilan University, June 2007), 19–21, https://besacenter.org/wp-content/uploads/2007/06/MSPS71.pdf.

¹⁵ Rubin, “The Rocket Campaign Against Israel,” 20.

¹⁶ Lee Chi-dong, “S. Korea Develops Counter-Battery Radar Against N. Korea,” Yonhap, April 24, 2017, https://en.yna.co.kr/view/AEN20170424004651315.

¹⁷ Kelsey Davenport, “South Korea to Pursue Military Satellites,” Arms Control Today, September 2020, https://www.armscontrol.org/act/2020-09/news/south-korea-pursue-military-satellites.

¹⁸ Jesse Johnson, “North Korean ICBM Launch That Triggered Alert May Have Failed In-Flight,” Japan Times, November 3, 2022, https://www.japantimes.co.jp/news/2022/11/03/national/north-korea-missile-j-alert. For a contrary perspective, see Markus Schiller, “Why Japan’s Ability to Detect North Korean Missiles May Be Better Than It Seems,” NK PRO, November 18, 2022, https://www.nknews.org/pro/why-japans-ability-to-detect-north-korean-missiles-may-be-better-than-it-seems.

¹⁹ “North Korea Military Power,” U.S. Defense Intelligence Agency, 2021, https://www.dia.mil/Portals/110/Documents/News/North_Korea_Military_Power.pdf.

²⁰ John Hudson, “North Korean Saber-Rattling Dims Euphoria of Trump’s DMZ Meeting,” Washington Post, July 26, 2019, https://www.washingtonpost.com/world/national-security/north-koreas-saber-rattling-dims-euphoria-of-trumps-dmz-meeting/2019/07/26/87c71596-b861-487f-afe1-ec45c04fbb8f_story.html.

²¹ “CPRC Reunification and Agitation Department Blames South Korea-U.S. Joint Military Exercises,” KCNA Watch, August 18, 2019, https://kcnawatch.org/newstream/1565332202-699518050/cprc-reunification-and-agitation-department-blames-south-korea-u-s-joint-military-exercises/.

²² Thomas C. Schelling, “The Reciprocal Fear of Surprise Attack,” RAND Corporation, January 1958, https://www.rand.org/pubs/papers/P1342.html.

²³ “Military and Security Developments Involving the People’s Republic of China,” U.S. Department of Defense, 2019, 66, https://media.defense.gov/2019/May/02/2002127082/-1/-1/1/2019_CHINA_MILITARY_POWER_REPORT.pdf.

²⁴Missile Defense Review, U.S. Department of Defense, 2019, XI, https://www.defense.gov/Portals/1/Interactive/2018/11-2019-Missile-Defense-Review/The%202019%20MDR_Executive%20Summary.pdf.

²⁵ Mira Rapp-Hooper, “Decoupling Is Back in Asia: A 1960s Playbook Won’t Solve These Problems,” War on the Rocks, November 20, 2019, https://warontherocks.com/2019/11/decoupling-is-back-in-asia-a-1960s-playbook-wont-solve-these-problems-2.

²⁶ Experts in South Korea expressed this view in interviews with the author in 2021 and 2022. Some Japanese experts shared a similar assessment with the author in 2019.

²⁷ Amina Afzal, “The Luck Factor: Preventing Nuclear War in South Asia,” South Asian Voices, March 25, 2022, https://southasianvoices.org/the-luck-factor-preventing-nuclear-war-in-south-asia.

²⁸ “India Sacks Officers for Accidentally Firing Missile Into Pakistan,” BBC News, August 24, 2022, https://www.bbc.com/news/world-asia-india-62655736.

²⁹ Kay Guerrero, “Taiwan Accidentally Launches Missile Toward China, Kills Fishing Boat Captain,” CNN, July 1, 2016, https://www.cnn.com/2016/07/01/asia/taiwan-fires-missile-on-china/index.html.

³⁰ “Missile Misfires, Creates Fireball on Golf Course,” Joongang Daily, October 5, 2022, https://koreajoongangdaily.joins.com/2022/10/05/national/defense/Korea-South-Korea-Hyunmoo2/20221005184305952.html.

³¹ Jeongmin Kim, “South Korean Missile Fails During Combined Weapons Test With US Forces: Seoul,” NK News, October 4, 2022, https://www.nknews.org/2022/10/south-korean-missile-fails-during-combined-weapons-test-with-us-forces.

³² “Missile Misfires, Creates Fireball on Golf Course,” Joongang Daily.

³³ Kim, “South Korean Missile Fails During Combined Weapons Test With US Forces.”

³⁴ Colin Zwirko, “North Korea Fired Cruise Missiles Near ROK Coast, New ICBM: State Media,” NK News, November 7, 2022, https://www.nknews.org/2022/11/north-korea-fired-cruise-missiles-near-rok-coast-new-icbm-state-media.

³⁵ Yi Wonju, “N. Korea’s 1st Firing of Missile Into Area Near S. Korean Territorial Waters ‘Intolerable’: Seoul Military,” Yonhap, November 2, 2022, https://en.yna.co.kr/view/AEN20221102004951325.

³⁶ Josh Smith, “Analysis: North Korea’s Other Missiles: Salvaged Debris Shines Light on Aging Air Defences,” Reuters, November 11, 2022, https://www.reuters.com/world/asia-pacific/north-koreas-other-missiles-salvaged-debris-shines-light-aging-air-defences-2022-11-11.

³⁷ Michele Kambas and Daren Butler, “Errant Missile From Syria-Israel Clash Lands on Cyprus,” Reuters, July 1, 2019, https://www.reuters.com/article/uk-cyprus-crash-idUKKCN1TW15Y.

³⁸ Ankit Panda, “North Korea’s Unprecedented Missile Launches Provide a Sneak Peek at Nuclear War,” NK PRO, November 7, 2022, https://www.nknews.org/pro/north-koreas-unprecedented-missile-launches-provide-a-sneak-peek-at-nuclear-war.

³⁹ “A Senior U.S. Intelligence Official Says Russian Missiles Crossed Into NATO Member Poland, Killing Two People,” Associated Press, November 15, 2022, https://apnews.com/article/nato-ap-news-alert-europe-poland-government-and-politics-ba48101fd25c86e68e57dc56fe2adf80.

⁴⁰ Dmytro Kuleba (@DmytroKuleba), “Russia now promotes a conspiracy theory that it was allegedly a missile of Ukrainian air defense that fell on the Polish theory. Which is not true. No one should buy Russian propaganda or amplify its messages. This lesson should have been long learnt since the downing of #MH17.,” Twitter post, November 15, 2022, 4:35 p.m., https://twitter.com/DmytroKuleba/status/1592632386751434752.

⁴¹ Two such statements were issued, one by the Slovakian minister of defense and one by the Latvian deputy prime minister, on social media prior to NATO’s final assessment of the event. See Artis Pabriks (@Pabriks), “My condolences to our Polish brothers in arms. Criminal Russian regime fired missiles which target not only Ukrainian civilians but also landed on NATO territory in Poland. Latvia fully stands with Polish friends and condemns this crime.,” Twitter post, November 15, 2022, 1:41 p.m., https://twitter.com/Pabriks/status/1592588505393541122; and Jaro Nad (@JaroNad), “Very concerned by Russian missiles dropping in Poland.Russia must explain what happened.Senseless attacks on infrastructure must stop immediately.Russia's recklessness is getting out of hand. Will be in close contact w/@mblaszczak and Allies to coordinate response @Slovakia_NATO,” Twitter post, November 15, 2022, 2:46 p.m., https://twitter.com/JaroNad/status/1592605004149366784.

⁴² Evan Luard, War in International Society: A Study in International Sociology, First Edition (New Haven: Yale University Press, 1987), 232–233.

⁴³ “China Sends Missiles Flying Over Taiwan,” The Economist, August 4, 2022, https://www.economist.com/china/2022/08/04/china-sends-missiles-flying-over-taiwan.

⁴⁴ Ankit Panda and Dave Schmerler, “When a North Korean Missile Accidentally Hit a North Korean City,” The Diplomat, January 3, 2018, https://thediplomat.com/2018/01/when-a-north-korean-missile-accidentally-hit-a-north-korean-city.

⁴⁵ Jonathan C. McDowell, “The Edge of Space: Revisiting the Karman Line,” Acta Astronautica 151 (October 1, 2018): 668–677, https://doi.org/10.1016/j.actaastro.2018.07.003.

⁴⁶ Chris Buckley and Amy Chang Chien, “Chinese Missiles Strike Seas Off Taiwan, and Some Land Near Japan,” New York Times, August 4, 2022, https://web.archive.org/web/20220804173323/https://www.nytimes.com/2022/08/03/world/asia/taiwan-china-military-exercises.html.

⁴⁷ Associated Press, “China Failing to Answer U.S. Crisis Line Call During Balloon Incident Highlights ‘Dangerous’ Communications Gap,” CBS News, February 10, 2023, https://www.cbsnews.com/news/china-spy-balloon-beijing-us-crisis-phone-hot-line-dangerous-communications-gap.

⁴⁸ Reiko Miki, “Japan and China Set Up Hotline to Prevent Maritime, Air Clashes,” Nikkei Asia, April 1, 2023, https://asia.nikkei.com/Politics/Defense/Japan-and-China-set-up-hotline-to-prevent-maritime-air-clashes.

⁴⁹ Kelsey Atherton, “Russia’s Radar Shortcomings Are a US Problem Now,” C4ISRNet, December 14, 2018, https://www.c4isrnet.com/intel-geoint/sensors/2018/01/02/russias-radar-shortcomings-are-a-us-problem-now; and Joshua Pollack, “Russia’s EW Is Worse Than You Thought,” Arms Control Wonk, April 14, 2009, https://www.armscontrolwonk.com/archive/502257/russia-north-korea-worse-than-you-thought.

⁵⁰ “Military and Security Developments Involving the People’s Republic of China,” U.S. Department of Defense, 2022, 99. See also Stephen Clark, “China Launches Military Satellite Toward Geostationary Orbit,” Spaceflight Now, February 7, 2021, https://spaceflightnow.com/2021/02/07/china-launches-military-satellite-toward-geostationary-orbit.

⁵¹ Oren Liebermann, Katie Bo Lillis, and Barbara Starr, “Early Warning Systems First Suggested North Korean Missile Could Hit US, Causing Temporary Scramble,” CNN, January 14, 2022, https://www.cnn.com/2022/01/13/politics/north-korean-missile-faa-grounded-planes/index.html.

⁵² Lolita C. Baldor, “US Developing Satellite System to Track Hypersonic Weapons,” U.S. Space Development Agency, July 18, 2022, https://www.sda.mil/us-developing-satellite-system-to-track-hypersonic-weapons.

⁵³ For a comprehensive treatment of this matter, see Acton, “Is It a Nuke?”

⁵⁴ David C. Logan, “Are They Reading Schelling in Beijing? The Dimensions, Drivers, and Risks of Nuclear-Conventional Entanglement in China,” Journal of Strategic Studies 46, no. 1 (January 2, 2023): 5–55, https://doi.org/10.1080/01402390.2020.1844671.