[Global NK Commentary] A Critique of South Korea’s Entrapment and Abandonment Discourse Amid U.S.-China Competition

■ Read the original article on Global NK Zoom&Connect

I. Introduction

As U.S.-China competition intensifies, the concepts of “entrapment and abandonment” are frequently invoked in South Korea’s security discourse. When someone points out the risk of entrapment in relation to strengthening ROK-U.S.-Japan security cooperation or a contingency in the Taiwan Strait, a common retort is, “If we don’t contribute enough to the alliance, there’s also the risk of abandonment.” Such discussions typically go nowhere, ending with the platitude that “we must strike a good balance between the two.” This conclusion is difficult to refute, but it also fails to yield any meaningful policy implications.

The threat from North Korea is invariably summoned in this discursive framework. The logic is that countering North Korea’s nuclear weapons necessitates strengthening alliance cohesion and extended deterrence, and that a passive stance toward the alliance invites abandonment. This paper, however, argues that these two risks do not coexist symmetrically. While the North Korean threat is a relatively specific risk that the ROK-U.S. alliance has long “managed,” entrapment in great power conflict is a risk of a different dimension in both scale and lethality, one that could make South Korea the initial target of a conflict.

This paper therefore critiques the “notion of a mechanical balance” that treats the risks of entrapment and abandonment as equivalent. The relative magnitude of the two risks is not a constant but a variable, and which one dominates depends on the given structure and circumstances. Consequently, any argument about the risk of either entrapment or abandonment must specify the concrete actors, structural context, causal pathways, and timeline. To simply claim that “the opposite risk also exists” without these specifics is not a balanced proposal but rather a recipe for unproductive stalemate or a deferral of judgment.

Entrapment is an ongoing risk that can materialize for military and geopolitical reasons alone. Abandonment exists on a different timeline, emerging as a plausible risk only after the structure of entrapment has matured. Overlooking this difference justifies dangerous policy prescriptions, such as the “argument for preemptive contributions,” which posits that “to avoid being abandoned, we must cooperate more and give more now.” This prescription uses the possibility of an ill-defined abandonment to compel acceptance of the real, present danger of entrapment. Furthermore, it risks hastening the danger of abandonment, which would only become possible after entrapment has occurred.

II. Are Entrapment and Abandonment Symmetrical Risks? A Theoretical Reexamination

The notion of a mechanical balance finds little support even within its theoretical lineage. Glenn Snyder specified the conditions that determine whether the risk of entrapment or abandonment is dominant:[1] dependence on the alliance, the explicitness of the commitment, the scope of shared interests, the existence of alternative partners, and the polarity of the system. In fact, no theory of entrapment and abandonment suggests that a balance must always be maintained simply because both risks coexist. Rather, the focus is on identifying which variable becomes more decisively activated in a given structure and situation. Identifying the dilemma is the starting point of analysis, not its conclusion.

Victor Cha’s theory of quasi-alliances specifies this point in the context of ROK-U.S.-Japan relations.[2] According to him, the direction of ROK-Japan cooperation depended on which risk the two countries came to fear more dominantly as U.S. commitments changed. In this theory, the weight of risk tilts to one side depending on the situation; a state of equivalence is not assumed. And the determining factor is the U.S. variable, rather than the domestic politics or bilateral relations of South Korea and Japan. The fact that security cooperation between the two countries remains at an unprecedented level—despite shifts in their domestic political landscapes and at a time when the United States is emphasizing a comprehensive alliance with them more than ever—clearly illustrates this structural determinism.

The notion of a mechanical balance not only homogenizes the different pathways and timelines of risk but also leads to the dangerous policy prescription of preemptive contributions. This prescription fails to specify the pathways and mechanisms of the very abandonment it seeks to prevent. It also bears the internal explanatory burden of simultaneously positing the United States as an actor that could abandon an ally based on its interests and as an actor that can be bound by a partner’s contributions. Deferring a more detailed argument for later, it is sufficient here to establish that the key determinant of potential abandonment is not the level of South Korea’s contributions but the degree to which the United States needs South Korea’s assets.

The risk of entrapment that South Korea currently faces stems from structural changes that have been gradually unfolding in the geopolitical and military domains as the U.S.-China strategic competition has progressed. The risk of abandonment, however, does not lie on the same path. On the contrary, the more intense the U.S.-China competition becomes, the more the United States will value South Korea’s geopolitical utility. Furthermore, considering America’s current manufacturing capabilities, South Korea’s capacities in semiconductors, shipbuilding, and batteries are not something the U.S. can easily give up. In other words, the risk of abandonment only becomes plausible after the United States has substantially transferred the capabilities held by South Korea into its own “endogenous” capacity (the “internalization of industrial capabilities”). Thus, abandonment is not a symmetrical counterpart to entrapment on the same path and at the same time; it is more likely to be a consequence of the entrapment structure.

How, then, do the notions of mechanical balance and preemptive contributions distort South Korea’s discourse? The core issue lies in treating the relationship between entrapment and abandonment, which major theorists like Snyder presented as conditional, as if it were an unconditional law. The linkage that “distancing oneself to avoid entrapment increases the risk of abandonment” holds true only under specific conditions. The center of gravity for whether those conditions are met is the structural factor of how much the United States needs South Korea’s assets, rather than South Korea’s actions. As long as U.S. need remains, any increase in the risk of abandonment will be limited even if South Korea tries to create distance. Conversely, if that need disappears, it is more likely that even the closest alignment will have limited effect in reducing the risk of abandonment.

The notion of a mechanical balance strips away this conditionality, and the argument for preemptive contributions builds upon it the conclusion that if South Korea increases its contributions, the United States cannot abandon it. This constitutes a dual flaw: the error of mistaking a conditional relationship for an unconditional law, and the leap of building a policy upon that flawed premise.

III. The U.S. Variable: Shifting Historical Dynamics of ROK-U.S.-Japan Trilateral Cooperation

During the Cold War, the United States often called for ROK-U.S.-Japan trilateral cooperation. However, in periods when U.S. commitment and engagement were strengthening, both South Korea and Japan tended to avoid such cooperation. The two countries only grew closer when the United States scaled back its security commitments.[3] For example, the Nixon Doctrine and the partial withdrawal of U.S. Forces Korea (USFK) in the 1970s, as well as Carter’s attempted withdrawal, made the risk of abandonment the dominant concern in both countries. During this time, South Korea and Japan tilted toward cooperation despite their mutual animosity. Conversely, during periods when U.S. commitment and engagement were robust, both countries—especially Japan—took the lead in rejecting calls for trilateral cooperation.

Japan’s strategic stance had a clear geopolitical backdrop. During this period, Japan faced no imminent threats. The bulk of Soviet forces were deployed on the European front, and both North Korea and China were economically and militarily weak. Most importantly, as long as the Korean Peninsula functioned as a buffer zone in the U.S.-Soviet confrontation, Japan could stay off the front lines, delegate its security to the United States, and focus on its economy. The fact that Japan enjoyed its greatest-ever economic boom while keeping its defense spending below 1% of GDP was a product of this structure.

For Japan, trilateral cooperation, which implied the institutionalization of security cooperation with South Korea, was an excessive choice that increased risks without offering benefits. It not only raised the probability of being entangled in a conflict on the Korean Peninsula but also meant entangling itself more deeply in the confrontation with the nuclear-armed Soviet Union, America’s true target. A similar pattern persisted in the early post-Cold War era. With the Soviet Union gone and China still poor, Japan had no incentive to comply with repeated U.S. demands to contribute more to regional security.

However, as China’s rise became apparent after the end of the Cold War, Japan’s calculations began to change. Japan started to clearly perceive China’s belligerence during the 1995–1996 Taiwan Strait Crisis, and confrontations over the Diaoyu/Senkaku Islands made the prospect of a direct clash with China over territorial sovereignty and hegemony in the East China Sea a reality. In 2010, China’s GDP surpassed Japan’s, and today the gap is more than fourfold. Compounded by the memory of its own past aggression against China, Japan came to perceive itself as exposed to the threat from China.

Crucially, unlike the Soviet Union during the Cold War, China’s expansion no longer passes through the Korean Peninsula. China directly confronts the United States and Japan via its own maritime access routes in the South and East China Seas. Now, it is Japan itself that is exposed on the front lines of great power conflict. The very logic that led Japan to avoid trilateral cooperation during the Cold War has been inverted into a logic that demands it. In other words, Japan has come to pursue the institutionalization of ROK-U.S.-Japan security cooperation to shift the initial target in a contingency from Japan to South Korea, thereby reshaping South Korea into a buffer zone for both the United States and Japan in a new context.

The implication is clear. The current push to deepen trilateral cooperation is not happening because South Korea and Japan’s fears of abandonment have grown due to weakening U.S. engagement, as was the case in the 1970s. The United States is currently strengthening its institutional and military cohesion by codifying its security commitments at Camp David in August 2023, establishing a trilateral secretariat, and enhancing trilateral military cooperation. Thus, the dynamic at play is not the resolution of ROK-Japan anxiety over abandonment due to weakening U.S. engagement, but the deepening of an entrapment structure centered on a strengthened U.S. strategy toward China. This structure is being maintained despite prolonged wars in Europe and the Middle East, suggesting that the priority of the U.S. strategy against China has not wavered, even with its forces dispersed across multiple fronts.

The shift in America’s primary threat perception is evident not only in its approach to ROK-Japan relations but also in its handling of the North Korea issue. When the Soviet Union, its main adversary, disappeared after the Cold War, the United States needed to secure a new justification for the presence of USFK, and the North Korean nuclear threat filled that role. Since North Korea revealed its nuclear development intentions in the early 1990s, the United States has managed this threat while gradually reorienting the ROK-U.S. alliance toward China. The joint demand for missile defense in 1999 was a concrete signal of this shift. Through the Bush, Obama, and Trump administrations, strengthening trilateral cooperation was consistently pursued under the guise of responding to the North Korean threat. Even the Obama administration’s “strategic patience” toward North Korea, pursued within the broader framework of the “pivot to Asia,” served as a means to seriously lay the groundwork for trilateral cooperation. By the Trump administration, it was openly acknowledged that trilateral cooperation was aimed at China, a bipartisan consensus that has continued through the Biden era and is expected to persist into a second Trump term.

It is noteworthy that trilateral cooperation is not necessary to counter the North Korean nuclear threat itself. South Korea’s GDP is dozens of times larger than North Korea’s, and its defense budget alone exceeds the size of the North’s economy. The North Korean nuclear issue is a matter to be addressed through the ROK-U.S. alliance’s extended deterrence or South Korea’s own capabilities; it is not an issue that demands a ROK-U.S.-Japan trilateral framework. The fact that the North Korean threat has been repeatedly invoked as a justification for “strengthening” trilateral cooperation and extended deterrence suggests that its true function has been to serve as a mechanism for binding South Korea to America’s strategy toward China and its broader regional strategy. As will be discussed later, this point prompts a reconsideration of the conventional wisdom that reads concerns about extended deterrence as a sign of abandonment.

IV. The Present Nature of Entrapment Risk and the Subsequent Nature of Abandonment Risk

1. The Ongoing Risk of Entrapment

Entrapment is not a mere discursive assumption but a structural change that is actually in progress. Developments since the end of the Cold War, particularly since the rise of China, bear this out. The U.S. National Defense Strategy (NDS) released in January 2026 set four major tasks: defending the homeland and the Western Hemisphere, deterring China, expanding allied burden-sharing, and mobilizing the defense industrial base.[4]

In the Indo-Pacific, it declared the establishment of a “strong denial defense” along the first island chain. Regarding the Korean Peninsula, it emphasized that South Korea has the capability to bear the “primary responsibility” for deterring North Korea with “decisive but more limited” U.S. support. If this means the nature of USFK is shifting from deterring North Korea to deterring China, then the operation of forces stationed on the peninsula is increasingly likely to be linked to contingencies beyond it. Some in South Korea liken this to a “New Acheson Line” and read it as a signal of abandonment, but this analogy needs reexamination. The Acheson Line itself can be interpreted as part of a larger strategic design, contrary to the prevailing view. As this issue requires a separate discussion, it will not be addressed here.

Statements from the military leadership have also confirmed this trend. In a keynote speech at the AUSA Land Forces of the Pacific Symposium in Hawaii in May 2025, USFK Commander Brunson likened South Korea to an “aircraft carrier floating in front of China.” He stated that USFK is not solely focused on defeating North Korea but also on regional operations as part of the broader Indo-Pacific strategy. He also mentioned that USFK could play a significant role in supporting regional operations to overcome the “tyranny of distance.”[5] In a joint press conference following the 57th ROK-U.S. Security Consultative Meeting in November 2025, U.S. Secretary of War Hegseth, when asked if USFK could be used in operations in the Taiwan Strait or the East China Sea, responded to the effect that it was necessary to enhance the flexibility of USFK to be able to respond to other regional contingencies.[6]

This is no longer limited to statements by key figures. In 2026, there was an incident where U.S. fighter jets scrambled from South Korea toward the West Sea without prior coordination with Seoul, coming to the brink of a military friction with China. The deployment of forces launched from South Korean territory into a third-party conflict, regardless of South Korea’s will, and the resulting exposure of South Korea to tensions, is the textbook definition of entrapment.

The institutionalization of ROK-U.S.-Japan trilateral security cooperation, which progressed to an unprecedented level under the previous South Korean administration, has also entrenched this structure. The commitment to consult from Camp David, the Trilateral Security Cooperation Framework memorandum signed in Tokyo in July 2024, and regularized joint exercises are all part of an effort to institutionalize trilateral cooperation beyond changes in government.

Institutionalization is not inherently evil, and it can be pursued if necessary. However, one must question the verification and consultation processes that determine whether it aligns with our national interest. The factors mentioned above shorten the causal path to South Korea’s entrapment. We have already observed the codification in strategic documents, statements from military leaders, the actual diversion of forces, and institutional entrenchment. Each step makes the next more probable. And this institutionalization raises the costs of exit. Once consultation procedures and joint plans are in place, it becomes harder for South Korea to opt out of specific issues, which in turn increases the likelihood of being entrapped in extra-regional operations demanded by the United States.

The risk of entrapment is not confined to the military dimension. As will be discussed later, the United States seeks to compensate for the limitations of its own defense industrial base with the manufacturing capabilities of its allies. If South Korea takes on the role of America’s “forward arsenal” in a contingency, this also increases the probability of it becoming a priority target for China and Russia in a U.S.-China conflict.

Costs can be imposed even before a military clash. China, which accounts for over 20% of South Korea’s trade, controls the supply chains for key inputs like critical battery materials and rare earth elements. The possibility that Russia could increase the level of its military technology transfers to North Korea cannot be ruled out either. Entrapment, therefore, materializes not only in the form of a direct military strike but also through the activation of economic and technological vulnerabilities.

Thus, the “risk of entrapment” can be specified with its structural context, actors, incentives, pathways, and timeline. Now, we must ask: can the “risk of abandonment” be specified in the same way?

2. The Conditions for Abandonment Risk: Geopolitical and Industrial Value

In the U.S. strategy toward China, South Korea’s value is twofold. First, industrially, South Korea possesses capabilities that the United States has effectively lost: top-tier semiconductor production, large-scale commercial and naval vessel construction, and battery manufacturing. For instance, U.S. experts and officials openly acknowledge that the help of South Korea’s shipbuilding industry is crucial to counter China’s maritime expansion. Naturally, the United States cannot afford to discard such essential assets during a strategic competition with China. Second, geopolitically, South Korea is a forward military base on China’s doorstep.

This industrial value is set against the backdrop of the structural limitations of the U.S. defense industrial base. For example, with stockpiles of precision interceptors like the Patriot and THAAD depleted due to support for Ukraine and interventions in the Middle East, a weakened manufacturing base means there is a significant lead time from ordering a missile to its delivery. Even with increased budgets, there is an absolute shortage of skilled labor on the factory floor, compounded by a reliance on China for the rare earth elements essential for advanced weapons manufacturing. There is a pervasive sense of crisis within the U.S. Department of Defense that in a full-scale war, the country could be defeated not by combat but by the “limits of its supply chain.” The NDS task of mobilizing the defense industrial base, the recommendation to increase the defense budget to the mid-4% range of GDP, and large-scale investments in mass production of low-cost weapons and non-traditional defense companies are all products of this sense of crisis.

Two reports released by CSIS on July 6 and 7, 2026, respectively, concisely show how the United States views South Korea in this context. The July 6 report, which reviewed the progress of the U.S. defense industrial base’s transition to a wartime footing, treated industrial cooperation with allies as a key task. It singled out Hanwha Ocean’s acquisition of the Philly Shipyard and its $5 billion investment to convert it into a naval vessel construction facility as a prime example of allied industrial cooperation. The U.S. demand for South Korea’s shipbuilding mass production capacity is so desperate that President Trump even mentioned the possibility of bypassing the Burns-Tollefson Act, a domestic law that has prohibited the U.S. Navy from procuring foreign-built ships. In short, at this point in time, the United States has no incentive to forgo South Korea’s industrial capabilities; on the contrary, the dominant incentive is to secure and utilize them.

Fundamentally, the U.S. security architecture in East Asia was designed with the Korean Peninsula as a geopolitical constant, not a variable. This is a matter of structure, not policy preference, and thus is not something that changes with a new administration. Conventionally, Japan, not South Korea, is known as the cornerstone of U.S. strategy in East Asia. Indeed, Japan’s industrial base and system of bases have long supported U.S. military operations throughout the Western Pacific. However, for this structure to be built, there was a prerequisite that Japan itself not be on the front line. A front-line state is required to have a large standing army and a corresponding mobilization system. The Yoshida Doctrine would have been fundamentally impossible during the Cold War, when Japan was directly exposed to the line of confrontation between two nuclear powers. The security structure of East Asia, with Japan as a rear base, could only be established because the Korean Peninsula played that front-line role.

This is not a post-hoc interpretation. It was a structure that was repeatedly acknowledged by U.S. and Japanese policymakers during the Cold War. The structure of the Korean Peninsula as a forward base and buffer zone was the most fundamental condition for ROK-U.S.-Japan security cooperation. Since the cabinet of Shigeru Yoshida, Japan’s position has been consistent: it could not cooperate with the United States unless U.S. forces were stationed on the Korean Peninsula to guarantee its security. This was based on the recognition that Japan must not be on the front line of the U.S.-Soviet confrontation. America’s calculation was no different. To keep Japan tied to its camp for the purpose of checking the Soviet Union, it had to place the Korean Peninsula as a buffer zone instead of putting Japan on the front line.

In short, South Korea’s buffer role was a prerequisite for both the United States and Japan to establish security cooperation. Article 4 of the Sato-Nixon Joint Communiqué of November 1969 also codified that the security of the Republic of Korea was essential to Japan’s own security. This is the so-called “Korea Clause.” It is no coincidence that this clause appeared as part of the negotiations for the reversion of Okinawa. While returning administrative rights over Okinawa, the United States had to preserve its operational discretion over its largest base in the Far East. To do so, it needed Japan’s assurance that it would not restrict the use of the base in a Korean contingency. Indeed, in August 1972, Nixon warned Tanaka that if Japan made it difficult to use the bases, he would have no choice but to withdraw U.S. forces from Korea. This is a scene that directly confirms the fact that the commitment to the Korean Peninsula and access to bases in Japan were mutually conditional.

The differences between the geopolitical conditions of the Cold War and today must also be read accurately. The Cold War Soviet Union was a continental power centered on its land forces and effectively had no year-round ice-free ports in the Pacific. As a result, for Soviet power to be projected, it had to pass through the land axis of the Korean Peninsula as a matter of military strategy. In other words, the Korean Peninsula was both a buffer zone and a chokepoint blocking the only land axis for Soviet expansion.

At that time, the primary adversary was the Soviet Union for the United States and North Korea for South Korea, so their threat perceptions were not identical. However, the direction and nature of the threat were the same. That is, both the Soviet Union and North Korea were land-based threats coming from the north. The direction the United States faced to block the Soviet Union overlapped with the direction South Korea faced to block North Korea. Therefore, the ROK and the U.S. did not experience alignment problems arising from a mismatch in primary threat perceptions throughout the Cold War.

Today’s adversary, however, is not the Soviet Union but China, and China does not share the Soviet Union’s geopolitical conditions. China possesses the world’s largest fleet and a powerful air force, and it has its own maritime access routes in the East and South China Seas. In other words, unlike for the Soviet Union, passing through the Korean Peninsula is not essential for Chinese power to be projected abroad. The relatively aligned direction and nature of the threat for the ROK and the U.S. during the Cold War have fundamentally changed, and this is the root cause of today’s alignment problem.

However, the fact that threat perceptions are fundamentally misaligned is a separate issue from whether the strategic value of the Korean Peninsula has changed. This is because the alignment problem arises from the difference in what the United States and South Korea view as the primary threat, but the strategic value of the Korean Peninsula is not determined by that alone.

In fact, the “irreplaceability” of the Korean Peninsula has not disappeared; only its context has changed. During the Cold War, its value was in geographic denial, blocking the southward advance of the Soviet Union. Today, it is a base for the capabilities the United States mobilizes against the threat of China, and it holds vital value for Japan as well, which must also consider a direct conflict.

Viewed in this context, a U.S. abandonment of South Korea is not a decision to give up on South Korea alone. It is tantamount to a decision to abandon Japan in succession, because Japan cannot confront China on its own. In 2025, China’s defense budget was approximately $336 billion by SIPRI estimates, while Japan’s was $62.2 billion—a gap of more than five times. The reason for the gap is not a lack of will but the size of their economies. Even if Japan were to double its spending ratio, the gap would remain large. Added to this are China’s nuclear arsenal, Japan’s non-nuclear status, and demographic structures.

Therefore, if the United States were to abandon South Korea, Japan would be left with only two probable options: exiting the U.S.-led order or pursuing independent nuclear armament. The former could mean adjusting relations with China or adopting equidistant neutrality. Either way, it would collapse America’s position in the Western Pacific. The latter would dismantle the non-proliferation order that the United States has maintained for 70 years. Either outcome would overwhelmingly exceed any costs saved by abandoning South Korea. The costs of abandonment do not stop at a single target; they spread throughout the entire system.

This argument has two implications. First, from the U.S. perspective, South Korea’s value is systemic, not merely bilateral. The theory of preemptive contribution primarily focuses on the bilateral dimension, asking what South Korea must give the United States to prevent abandonment. However, the Korean Peninsula’s fundamental value stems above all from its geopolitical position within the regional security architecture. Second, this geopolitical value is not monolithic. Whereas its Cold War-era value of denial was physically fixed, its contemporary value of alliance cohesion depends significantly on capabilities, perceptions, and the institutionalization based upon them.

However, this does not justify the claim that ‘abandonment’ is a realistic risk. Even if the value of cohesion partly depends on perception, that perception is ultimately underpinned by the Korean Peninsula’s position in the security architecture and its own capabilities. It is worth noting, however, that this value is not automatically given in the way that Cold War-era geography was. The perception of capabilities and the mutual trust based on it are areas that require management. On the other hand, perceptions not based on structure and facts, and institutionalization built on such perceptions, can themselves become risk factors. And this is precisely why the ‘theory of preemptive contribution’ cannot be supported. As will be discussed later, the way to protect South Korea’s value lies in something closer to the opposite: precise management of the relationship.

In addition to its non-transferable positional value, South Korea currently possesses significant capability-based value. The problem is that industrial capabilities can be moved. It is this transferability that creates a tangible problem of abandonment. This is because current U.S. policy fundamentally seeks to both lock in South Korea and transfer what it possesses to the United States. And this policy trend has been consistent, not limited to a single administration. The Biden administration, through the 2022 CHIPS and Science Act and the Inflation Reduction Act, has drawn South Korea’s advanced manufacturing capabilities into the United States by conditioning subsidies and market access. Under a structure where production outside North America is excluded from subsidies and receiving U.S. support restricts production in China, massive investments in the U.S. by Samsung Electronics for its Taylor foundry and by the three major battery companies were effectively coerced. The Trump administration merely changed the form of pressure. Instead of institutionalized conditions, it used the blunt threat of tariffs to extract $350 billion in investment and shipbuilding cooperation in the 2025 tariff negotiations. The means differed in sophistication, but all were coercive, and the direction was singular. The two administrations were continuous in their goal of transferring South Korean capabilities to the United States.

A July 7 report from CSIS shows how the discourse justifying this transfer is constructed. The report posits that South Korea’s record-high export performance in 2025 ($709.7 billion) conceals more than it reveals. Citing the sharp decline in the trade surplus with China (from $99.7 billion in 2018 to $19.6 billion in 2025) and a demographic structure with a total fertility rate of 0.75, it warns that a country clinging to its existing export model will be gradually marginalized and become substitutable. While formally acknowledging South Korea’s strengths across four areas—AI, semiconductors, and telecommunications; batteries and precision manufacturing; robotics and machine tools; and critical minerals and energy—it urges the integration of these capabilities into the U.S.-led technology and security ecosystem. This structure, where a diagnosis of crisis and a prescription of integration are presented as a package, suggests that the logic of preemptive contribution was being formalized in U.S. discourse before it appeared in domestic South Korean discourse.

The goal is clear: to build manufacturing capacity within the United States using South Korean capital, technology, and labor. In 2026, this was followed by the implementation of the Special Act on Investment in the U.S., the establishment of the Korea-U.S. Strategic Investment Corporation, and the opening of the ROK-U.S. Shipbuilding Cooperation Center in Washington. Joint R&D, technology exchange, and workforce development were officially stipulated as the content of this cooperation. In the semiconductor sector, large-scale investments by Samsung Electronics and SK Hynix are replicating advanced production capabilities within the United States.

Structurally, this can be described as a process where the elements that constituted South Korea’s ‘irreplaceability’ are being transferred to the United States over time. These elements are production facilities, process knowledge, and skilled labor. Viewed differently, the cooperative language of ‘technology exchange’ and ‘workforce development’ signifies the transfer of tacit knowledge accumulated over decades. The problem is that once this process is complete, semiconductors can be produced and warships can be built within the United States without South Korea.

Conceptualizing this process as ‘reshoring’ distorts its essence. This is because it is not a process of American companies bringing back production they had moved overseas. Samsung Electronics’ investment in the U.S. is not the return of an American firm but a new greenfield investment by a Korean company. Hanwha’s acquisition of Philly Shipyard is not a repatriation of assets but an acquisition by foreign capital. Crucially, the objects of transfer include something the United States has never possessed: process knowledge that South Korea accumulated independently. One cannot bring back what one never had. Nor is ‘onshoring’ an appropriate term. The term is deceptive in that it only speaks to what the United States gains, not what its counterpart (South Korea) loses.

The current phase is a process of capability ‘internalization,’ in which the United States is moving manufacturing functions it has relied on from an external supplier, South Korea, into its own territory and technological base, transforming them into its own capabilities. The completion of internalization means the elimination of dependence on the external supplier and, consequently, the supplier becoming redundant. Internalization originally refers to a firm absorbing external functions within its own boundaries, but the key in this context is control, not ownership. Facilities in the U.S. owned by Korean companies also function as de facto U.S. jurisdictional assets through export controls, extraterritorial jurisdiction, and subsidy conditionalities. The separation of ownership and location, and the effective control by the host country, are the hallmarks of this internalization.

This is the crucial point this paper seeks to make. The material condition for the risk of abandonment is not the mere fact that factories are being built in the U.S., but the relational shift in which South Korea’s irreplaceability is being eroded. The domestic term ‘industrial hollowing-out’ captures the result but misses the cause. Hollowing-out implies a spontaneous outflow due to the loss of cost competitiveness. However, the issue here is that this is the product of coercive negotiations wielded by the United States against its ally for the sake of maintaining hegemony and competing with China.

For abandonment to become a realistic risk for South Korea through the aforementioned pathway, at least three conditions must be met from the U.S. perspective. First, a substitute for South Korea’s industrial value must be secured. Second, regional strategy must be realigned to the point of forgoing South Korea’s positional value, or that positional value must be securable through other means. Third, the overall cost of ‘maintaining’ South Korea must exceed its remaining value.

Currently, none of these conditions have been met. However, as the internalization of U.S. industry progresses, the first condition is gradually being fulfilled. Once the production base in the U.S. matures and the transfer of technology and labor is complete, the industrial incentive to hold onto South Korea will disappear, and the U.S. calculation will converge on a geopolitical cost-benefit analysis. However, as previously discussed, the second condition is systemic and difficult to meet in the short term.

The fulfillment of this first condition is already proceeding in an observable form. In the semiconductor sector, while relying on South Korean HBM (High Bandwidth Memory), the U.S. is intensively fostering its own company, Micron. Once the next-generation HBM production base linked to Japanese government subsidies becomes fully operational, the irreplaceability of South Korean memory will be correspondingly eroded. The structure of exports to the U.S. is also indicative. There are analyses showing that a significant portion of the recent increase in exports to the U.S. consists of materials and equipment for constructing local factories there. This implies that the current export boom may paradoxically be a process in which South Korea is building, at its own expense, the very U.S. production base that will one day replace its exports of finished products. If the transfer of smart automation processes and robotics technology is added to this, a path opens for U.S. production to become self-sufficient without a large-scale migration of skilled labor. In short, internalization is not a hypothesis but a clear, ongoing trend.

To summarize, the likely scenario for South Korea after the internalization of U.S. industry is not the risk of full-scale abandonment but of partial instrumentalization. In other words, once its industrial value disappears, South Korea will remain valuable to the U.S. not as an ally to be protected, but as a location to be secured and utilized. That is, U.S. forces would remain stationed, but their mission would not be the defense of South Korea. South Korea would become not a protected entity, but a utilized base. This is arguably more difficult to handle than abandonment. The U.S. pursuit of internalizing industrial capabilities while transferring responsibility for deterring North Korea to South Korea and enhancing its own strategic flexibility points exactly in this direction. It is about maintaining the location and absorbing our capabilities while offloading the strategic costs and burdens.

Therefore, for South Korea, the risk of abandonment is not simply the symmetrical opposite of entrapment but rather a consequence of the entrapment structure. If entrapment is a present problem, abandonment is a conditional future, and it is more likely to materialize in the form of ‘instrumentalization’ than full-scale abandonment. This could lead to a far more devastating outcome than simple abandonment, and policies based on the theory of preemptive contribution create or accelerate these very risk conditions.

3. The Dual Vulnerability of the Theory of Preemptive Contribution

The preceding reconstruction warns of the dangers that the ‘theory of mechanical balance’ between entrapment and abandonment and the ‘theory of preemptive contribution,’ both widely circulated in domestic discourse, could bring. The ‘theory of preemptive contribution’ expresses concern about the risk of abandonment without specifying what it entails. Beyond identifying the actor as the United States, there is no specific explanation of why, through what pathway, when, or to what extent. Preparedness for an unspecified risk cannot have its effectiveness evaluated. Since there is no standard to judge what and how much would be sufficient, this argument can always lead to another leap of logic: ‘it is still not enough.’

Let us examine in detail the more fundamental problem foreshadowed in Chapter II: the internal inconsistency of this argument. The theory of preemptive contribution posits the United States in two contradictory ways. On the one hand, the U.S. is an actor that can abandon an ally based on a calculation of its interests. Otherwise, there would be no reason to worry about abandonment. At the same time, the U.S. is an actor bound by the extent of its ally’s contributions. However, this ignores the fact that for an actor driven by interests, contributions already received can become sunk costs. Above all, the United States does not consider only its bilateral relationship with South Korea. If the U.S. were to enter a stage of considering abandonment for some reason, it might not consider what South Korea has given, but only what more it can give in the future within a broader framework.

More fundamentally, there are two additional problems. First is the unfalsifiability of this claim. The preventive effect of contributions cannot be disproven. If abandonment does not occur, it can be attributed to the contributions; if it does, it can be blamed on insufficient contributions. Second, while this unscientific thinking is a problem in itself, the content of the ‘effective contributions’ proposed by the theory is an even more serious issue. Most of this content consists of elements of entrapment, deeply related to out-of-area operational cooperation and the deepening of combinedness. This is because the very definition of strengthening combinedness is an act aimed at maximizing effectiveness against a common enemy. Ultimately, it is tantamount to arguing for accepting specific, present entrapment to prevent an unspecified, future abandonment. And (partial) agreement to the strategic flexibility demanded by such an argument, or the strengthening of the combined system, is difficult to retract even when the risk of entrapment becomes tangible. In the worst-case scenario, the price South Korea would pay is becoming a party to a great power conflict.

The prescription of the theory of preemptive contribution does not operate only in the geopolitical sphere. The demand to contribute reliably when the U.S. needs it was also the justifying logic for the $350 billion investment package. However, if abandonment comes after ‘internalization,’ this contribution betrays its own purpose. The more deeply South Korea contributes and the more it transfers, the more it is not proving its irreplaceability but rather building its own substitute on U.S. soil at its own expense.

Consequently, the theory of preemptive contribution generates losses in two directions. Geopolitically, it makes one accept a specific, present risk in the name of an unspecified, future risk. Industrially, it increases the likelihood of accelerating the completion of the very conditions for the risk it sought to prevent through ‘payment.’ This prescription is not a balance between the two risks but something closer to their simultaneous maximization.

V. Anticipated Counterarguments

Counterarguments to this paper’s position are also possible. First is the counterargument that concerns about the credibility of extended deterrence demonstrate the real risk of abandonment. This is indeed a realistic concern that has continuously driven discussions in South Korea about developing its own nuclear weapons.

However, before responding to this counterargument, the nature of extended deterrence must be clarified. All authority over nuclear weapons, from planning and operation to final use, constitutionally belongs to the exclusive national command authority of the nuclear-weapon state. Even if an ally participates in consultative bodies and receives some shared information, the authority itself is not divided. Ultimately, extended deterrence is structurally a domain of ‘expectation,’ not a sharing of authority, and the credibility of that expectation always entails unfalsifiability. Therefore, the rhetoric of ‘strengthening the relationship of expectation’ is a very awkward logical connection.

Because anxiety about an unverifiable expectation does not resolve itself, it is easily exploited as a pretext for endless lock-in. This unverifiability, as will be seen later, defines the very nature of the discourse surrounding extended deterrence. Acknowledging this fundamental limitation is the starting point for the following discussion.

Based on this principle, the conventional wisdom that concerns about the credibility of extended deterrence show the real risk of abandonment should actually be reversed. Contrary to the common belief that the credibility problem of extended deterrence is evidence of abandonment, it is more akin to another channel through which the United States locks in South Korea. The more South Korea fears North Korea’s nuclear weapons, the deeper its dependence on the U.S. nuclear umbrella becomes, and that dependence in turn functions as a mechanism to tie South Korea to U.S. regional strategy or to present it with a bill.

However, as examined earlier, the North Korean nuclear threat itself—the justification for that umbrella—is not purely exogenous. In the process of reorienting the U.S.-ROK alliance toward China after the Cold War, the United States has ‘managed’ the North Korean threat while also using it as a justification for its China and regional strategies. Therefore, South Korea’s anxiety surrounding extended deterrence is not simply a ‘precursor to abandonment.’ Rather, its character as a strategic mechanism for keeping South Korea bound within an entrapment structure is more fundamental. In other words, as long as the anxiety stemming from extended deterrence does not reach an extreme, it functions less as a signal of abandonment and more as a useful driver for efficiently deepening South Korea’s ‘voluntary’ entrapment.

This is compounded by another problem: a lack of specificity. The worst-case scenario envisioned by extended deterrence—North Korea’s use of nuclear weapons against the South—can only be rationally posited in an extreme situation where the regime’s survival is at stake. Nevertheless, South Korea’s preemptive demands have focused less on specific judgments about the probability of such use and more on measures that can be labeled as ‘strengthening’ extended deterrence. Examples include the establishment of new consultative bodies or the regular deployment of strategic assets. However, it is inherently unclear how much such measures actually ‘strengthen’ the credibility of extended deterrence. There is no evidence to suggest that the frequency of consultative meetings or the visibility of asset deployments is directly proportional to the effectiveness of deterrence. Ultimately, it is a structure of repeated demands without any standard for what must be obtained, and to what extent, for extended deterrence to be ‘strengthened’ to a sufficiently credible level. This absence of a standard works to the advantage of the United States. As long as there is no standard, the U.S. can present a bill in return that is unrestricted in category and scale.

Of course, if the United States were to actually withdraw its nuclear umbrella, that would be a clear case of abandonment. To say that the provision of extended deterrence is a form of lock-in does not mean its withdrawal is not abandonment. However, in the current phase, the moves surrounding extended deterrence are not toward withdrawal but toward superficial ‘strengthening’ and, on that pretext, institutionalization at various levels. The Washington Declaration is a prime example. In other words, the currently observed trend is not a sign of the umbrella being folded, but a deepening of institutional lock-in for various forms of entrapment, mediated by measures like the Nuclear Consultative Group and strategic asset deployments, whose effectiveness is unproven (and unprovable). Therefore, the counterargument that cites concerns about extended deterrence as evidence of abandonment ignores or overlooks how entrapment is actually strengthened by those very concerns.

However, a distinction must be made between discussing the credibility of extended deterrence as a policy issue and using the rhetoric that “because extended deterrence is uncertain, there is a risk of abandonment” to offset the discussion of entrapment. But this distinction is not because the former satisfies this paper’s discipline (specifying the actor, pathway, and timeline). Quite the opposite. As seen earlier, the authority to use nuclear weapons is, in principle, not shared with allies, and its credibility belongs to a realm of expectation that is inherently unverifiable and unspecifiable. In this respect, extended deterrence is an exceptional topic to which the discipline demanded by this paper cannot, in principle, be applied.

The problem lies precisely here. Because it cannot be specified, it cannot be falsified; and because it cannot be falsified, extended deterrence can be endlessly invoked as a catch-all for the risk of abandonment. The moment we classify extended deterrence as a problem of abandonment, we mistake the concept’s fundamental limitations for a legitimate concern about the logic of lock-in designed by the United States.

Second is the counterargument that reads the 2026 NDS’s transfer of primary responsibility for deterring North Korea to South Korea as a sign of abandonment. However, it is more accurate to read this as a redistribution of burdens. Its purpose is not to abandon South Korea but to shift U.S. military forces to missions focused on China. It is part of a restructuring of the entrapment framework rather than abandonment, and even this fact, which appears to be strong evidence for the abandonment theory, is written in the grammar of entrapment upon closer inspection. In particular, some argue that the Trump administration’s remarks about reducing U.S. Forces Korea and its pressure on defense costs constituted a risk of abandonment. However, that pressure was not a prelude to dissolving the alliance but a negotiating lever. Increased defense cost-sharing and the $350 billion investment package were extracted in return. The outcome of pressure that invoked abandonment was not disengagement but greater lock-in and deeper internalization.

VI. Conclusion

1. Disciplining the Discourse and Managing the Risk of Entrapment

For a claim of risk—be it entrapment or abandonment—to be recognized as a debatable argument, it must specify the actor, incentives, pathway, and timeline. An argument that posits an opposing risk without specifying the actor and pathway is not an argument but an evasion of one. The conditioned, mechanical balancing in the discourse itself creates another danger: it obscures real-world imbalances and leads to an uncritical treatment of ongoing structural changes. Mechanical balance does not mean neutrality. The rhetoric of balance atop a tilted reality is merely a reflection of that tilt.

Since the two risks operate on different timelines and in different domains, the responses must also be separate. Managing entrapment is a present-day, geopolitical, and military task. Above all, South Korea must adhere to the principle that it cannot become the front line of U.S.-China conflict. As a corollary, prior consent and consultation for the exercise of strategic flexibility by U.S. Forces Korea must be institutionally mandated.

Here, the issue of troop reduction needs to be addressed head-on. Conventional wisdom has read a reduction of U.S. Forces Korea as a sign of abandonment and has made preventing it a top security priority. However, as seen earlier, the nature of U.S. Forces Korea today is changing. As long as its mission is being redefined from deterring North Korea to deterring China, maintaining troop levels means maintaining the entrapment structure. Under this condition, a reduction could be an opportunity to mitigate entrapment, not a case of abandonment. Therefore, if the United States insists on out-of-area deployments, it may be better for South Korea to reduce the risk of being dragged into the heart of a great power conflict, even if it means accepting a partial reduction of U.S. forces. The equation that reduction equals a defense vacuum holds only when U.S. Forces Korea is dedicated to defending against North Korea. Conversely, the equation that reduction equals entrapment mitigation also requires specifying the mission composition of the remaining forces, so efforts toward this must be made in parallel.

The basis for this judgment lies in the asymmetry of the two risks. The threat from North Korea is a risk of relatively specific scale and character that the ROK-U.S. alliance has managed for a long time. In contrast, entrapment in a great power conflict has a high probability of turning South Korea into the first target in a conflict between great powers, and the damage could reach a level from which recovery cannot be guaranteed. Accepting an irrecoverable risk for the sake of a manageable one is not a rational trade-off. In this respect, the risk of entrapment in a great power conflict, at least in terms of its potential scale and lethality, is an issue that should be prioritized over the threat from North Korea. In the same vein, for activities in sensitive airspace like the West Sea, information sharing must be mandated from the planning stage. In short, the contest is not over the depth of the commitment, but its conditions.

2. Preparing for the Risk of Abandonment and Its Policy Implications

Criticizing the overuse of the ‘risk of abandonment’ is not to deny the possibility of abandonment. Rather, it is to insist that its precise pathway must be identified. Preparing for abandonment is a task for the future. Its substance is the management of the pace of U.S. industrial ‘internalization.’ If abandonment is conditioned on the completion of internalization, then true preparation is not ‘giving more now’ but protecting South Korea’s ‘irreplaceability.’ Mechanisms for principal recovery and profit sharing in U.S. investments must be made effective, and the speed and scope of technology and personnel transfers must be managed. The most upstream capabilities in design and process must be kept within the country. Efforts must also be made in parallel to renew irreplaceability itself by shifting to next-generation capabilities that the U.S. cannot substitute. South Korea’s safety and prosperity lie in the opposite direction of the prescription that preemptively demands virtually unconditional contributions by invoking the ‘risk of abandonment.’

Mixing the two tracks—abandoning the management of entrapment due to anxiety about abandonment, and accelerating internalization to prove ‘trust’ as an ally within the entrapment structure, thereby increasing the possibility of abandonment—creates the worst possible combination. Regrettably, this is the point that the ‘theory of mechanical balance’ and the ‘theory of preemptive contribution’ are likely to reach in practice. ‘Balance’ is a word that itself has positive connotations. But balance is something you speak of after reading the scales, not something you declare without even looking at what is on them. What is needed now is not a balance between entrapment and abandonment, but concrete explanatory power for each concept. ■

[1] Glenn H. Snyder, "The Security Dilemma in Alliance Politics," World Politics, vol. 36, no. 4 (1984), pp. 461-495.

[2] Victor D. Cha, Alignment Despite Antagonism: The United States-Korea-Japan Security Triangle (Stanford, CA: Stanford University Press, 1999).

[3] Jaewoo Jun, "Underlying Dynamics for US-Japan-South Korea Trilateral Security Cooperation," ISPI Commentary, Istituto per gli Studi di Politica Internazionale, 8 April 2024, https://www.ispionline.it/en/publication/underlying-dynamics-for-us-japan-south-korea-trilateral-security-cooperation-169126.

[4] U.S. Department of War, 2026 National Defense Strategy (Washington, D.C.: Department of War, 23 January 2026), https://media.defense.gov/2026/Jan/23/2003864773/-1/-1/0/2026-NATIONAL-DEFENSE-STRATEGY.PDF.

[5] Xavier T. Brunson, Commander, U.S. Forces Korea, keynote speech at the Association of the United States Army (AUSA) Land Forces of the Pacific (LANPAC) Symposium, Honolulu, Hawaii, May 15, 2025 (local time).

[6] Joint press conference by Minister of National Defense Ahn Gyu-baek and U.S. Secretary of War Pete Hegseth immediately following the 57th ROK-U.S. Security Consultative Meeting (SCM), Ministry of National Defense complex, Yongsan, Seoul, November 4, 2025. Related report: “Hegseth: ‘Actively Support South Korea’s Construction of Nuclear Submarines’…Strengthening ROK-U.S. Defense Industry Cooperation,” *Hankook Ilbo*, November 4, 2025.

■ Jeon Jaewoo_Research Fellow, Korea Institute for Defense Analyses.

■ Managed and Edited by: Lee Sangjun_Research Fellow, EAI; Oh Inhwan_Senior Research Fellow, EAI

    Inquiries: 02 2277 1683 (ext. 211) | leesj@eai.or.kr