Conflict across the Taiwan Strait could disrupt East Asia’s extensive trade links, sever global production chains, generate serious shocks to regional economies, upend Asia’s security architecture, and, potentially, escalate into a catastrophic superpower war. Many regional states — including U.S. allies —  are beginning to seriously consider how they would respond to a potential use of force by the People’s Republic of China (PRC). But analytic attention has focused overwhelmingly on the worst-case scenario of a PRC invasion, largely ignoring more likely contingencies calculated to stay below the threshold of lethal force. It is in this “gray zone” that the PRC has made strategic advances in the East and South China Seas in recent years.

This paper argues that, compared with an invasion or blockade of Taiwan’s main island, an operation to capture one or more offshore islands currently controlled by the Republic of China (ROC or Taiwan) would offer Beijing considerable advantages. In an immediate tactical sense, it would offer Beijing greater flexibility and escalation control, lower risk of civilian casualties, and less likelihood of sparking a strong Taiwanese response or U.S. intervention. Strategically, such an operation could open up an array of options for further probes, faits accomplis, information gathering, and coercive pressure on ROC forces—and, in the case of the Penghu (Pescadores) Islands, substantial opportunities for enhanced surveillance, reconnaissance, and logistical support for a future invasion of the main island. Domestically, in contrast with a bloody and potentially catastrophic all-out invasion or a blockade that would risk conflict with the United States, outlying island seizure could offer Beijing a low-risk yet highly symbolic rallying point in a period of likely economic struggles and rising social dissatisfaction.

Policymakers and strategists on all sides of politics in Taiwan, the United States and elsewhere need to carefully consider how they would respond to such contingencies, in order to enable an effective international response.

Tensions over Taiwan have been a matter of regional importance to East Asia since the evacuation of Chiang Kai-shek’s Republic of China (ROC) to the island then known as Formosa in 1949. For a brief few months in 1950, the Harry S. Truman administration maintained that it would not intervene if Mao Zedong’s Communist forces attacked. There is little reason to doubt that Truman was sincere when he stated on January 5 of that year that “the United States Government will not provide military aid or advice to Chinese forces on Formosa.” Less than six months later, however, North Korea’s attack on South Korea triggered a dramatic about-face: “The occupation of Formosa by Communist forces,” Truman declared on June 27, “would be a direct threat to the security of the Pacific area and to United States forces” in the region.

Truman’s stated commitment to defending Taiwan successfully deterred an invasion by the PRC. Yet, over the following decade, China and the United States would come to the brink of war at least three times, with nuclear weapons even considered on the U.S. side. On each occasion, the venue of escalation was not Taiwan itself, but rather the small islands on the other side of the Taiwan Strait, close to mainland China, that the ROC government continues to occupy today (see map). The heavily fortified outpost of Kinmen (Quemoy), which lies within sight of the mainland Chinese city of Xiamen, and the Matsu Islands, just off the coast of Fuzhou, were targets of heavy PRC shelling in 1954– 1955 and 1958. The PRC also mounted a successful but now largely forgotten amphibious assault on the Yijiangshan Islands off the coast of Taizhou in 1955. Shortly afterward, U.S. and ROC forces evacuated the population of nearly 30,000 in the neighboring Tachen archipelago, leaving China’s People’s Liberation Army (PLA) to take control of 29 islands unopposed.

In 2022, Russia’s invasion of Ukraine and China’s belligerent response to U.S. House Speaker Nancy Pelosi’s visit to Taipei fueled a surge of analysis and commentary about potential Chinese blockades or invasions of Taiwan. Yet a wide consensus exists that a PRC assault on Taiwan would face severe challenges that are not present in the Ukraine case — starting with the logistics of mounting an invasion across the 80-mile Taiwan Strait, the significant likelihood of direct U.S. military intervention, and the prospect of massive economic sanctions. While it is important to consider worst-case scenarios, evidence from the Chinese Communist Party (CCP) Congress in 2022 augurs against an invasion or all-out blockade in the short term. Meanwhile, a more likely set of contingencies has received relatively little attention: PRC seizure of one or more of the islands currently controlled by the ROC. Besides Kinmen, Matsu, and the tiny Wuchiu Islands off the Fujian coast, the ROC occupies the Pratas (Dongsha) Islands and Itu Aba (Taiping) Island in the South China Sea and the Penghu (Pescadores) Islands in the Taiwan Strait. As this paper details, each scenario offers the PRC a distinct set of tactical, strategic, and political costs and benefits — but none comes close to the risks of an outright invasion or blockade.

This paper sketches profiles of six potential targets for PRC seizures of ROC-held islands. Along the Fujian coast of mainland China, the ROC-held Kinmen and Matsu Islands featured prominently in U.S.-China conflict planning during the twentieth century, but today, the lesser-known Wuchiu Islands that lie between the two could offer the PRC a more appealing target. In the South China Sea, the isolated atoll of Pratas (Dongsha) Island off Guangzhou offers potential reconnaissance value for Beijing, while Itu Aba (Taiping) is the largest naturally formed island in the Spratly archipelago, to which Vietnam and the Philippines also lay claim. The Penghu (Pescadores) archipelago, located on the Taiwan side of the median line of the Taiwan Strait, offers the PRC a potential staging post that could mitigate the challenges of invading Taiwan’s main island, but it is the closest and most integrated with Taiwan, raising the risks of a forceful ROC response, potentially backed by the United States.

After profiling these six potential targets, the paper considers likely ROC, United States, regional, and global reactions, highlighting the difficulty of reversing a fait accompli and mobilizing effective multilateral policy responses to such contingencies. Costs to the PRC could include elevated Taiwanese threat perceptions, U.S. economic sanctions, and enhanced U.S.-ROC diplomatic and military cooperation. Yet CCP leaders may believe that many of these consequences are already foregone conclusions, regardless of their conduct. The key task for Taipei and Washington’s diplomacy will be to convince Beijing otherwise.

The coast of Fujian Province, directly opposite Taiwan, is a complex geography of semi-submerged mountains, including more than 1,500 coastal islands and islets. Many of these features are formed from granite, enabling determined defenders to blast and build solid defenses to survive artillery and bombings. Taipei’s presence on several dozen of these islands dates back to Chiang Kai-shek’s attempts to blockade his CCP enemies by interdicting ships headed for mainland Chinese ports. Outposts on Kinmen, Matsu, and Wuchiu were also intended to deny the PRC unfettered use of key harbors that could serve as launching sites for invasion operations against Taiwan. But as Taiwanese identity grows stronger, the resolve to hold on to such outposts has declined: the ROC still officially administers them as part of its nominal Fuchien (Fujian) Province, i.e. part of the mainland. The U.S. commitment to defend these islands, meanwhile, is less robust than its “strategic ambiguity” with regard to its promise to defend Taiwan; past American commitments to the ROC have notably excluded them. While the nearest potential target, Kinmen, has a significant population and is heavily guarded, a quick occupying action against the Wuchius or uninhabited features in the Matsu Islands offer the PRC options to make discernible tactical advances in the Strait with minimal risk.

The ROC stronghold of Kinmen consists of 15 islands totaling around 150 square kilometers, located just offshore from the Chinese city of Xiamen. The main island of Kinmen has a resident population of around 60,000 civilians, a number that has been increasing as a result of programs to attract internal immigrants and the steady demilitarization of the island.¹ The number of ROC troops stationed in Kinmen fell from a high of 120,000 during the Cold War to around 5,000 by the mid-2010s, but even this small number could constitute a substantial force to overcome because of the island’s geographical and historical defensive advantages. Chinese military strategists are believed to regard subduing Kinmen as a precondition for the PLA successfully “reunifying” Taiwan by force.

Kinmen was the venue for some of the most dramatic confrontations between Beijing and Taipei during the Cold War. ROC soldiers there suffered heavy shelling during the First and Second Taiwan Crises in the 1950s, and the PLA continued to shell the islands on alternate days until 1979. But the island’s hard granite geology enabled the Kuomintang (KMT) military to build an extensive set of tunnels that, while now primarily tourist attractions, could offer protection from airborne attack. The PLA attempted amphibious operations to seize the main Kinmen Island in 1949, and nearby Tungting (Dongding) Island in 1958, but was repelled by ROC defensive emplacements. In September 2022, in the aftermath of China’s response to Pelosi’s visit, ROC soldiers stationed on Lion Islet, a small rock adjacent to Kinmen, shot down a civilian drone that they said had come from the PRC. This incident illustrates the ongoing tensions despite the military drawdown on the Taiwanese side.

Kinmen had no direct political-administrative connection with Taiwan prior to the ROC government’s evacuation to Taiwan. Control by a government based in Taipei is a historical novelty — to this day, the island and its appended islets and rocks are legally part of the ROC’s Fuchien (Fujian) Province, rather than Taiwan. While debates rage between Chinese and Taiwanese nationalists over when — and, indeed, whether — Taiwan may have rightfully been a part of China in history, the status of the Kinmen Islands has no such ambiguity; their status as Chinese territory is not contested by either side. For the United States during the Cold War, Kinmen lay outside the U.S.-declared defense perimeter, except insofar as it was deemed essential to the defense of Taiwan. These historical and administrative intricacies ostensibly work to the PRC’s advantage.

However, there would be major risks for the PRC in attempting to seize Kinmen. Despite lying just off the Fujian coast, Kinmen and its surrounding islands are extremely well fortified, with protruding geography and rocky geology enabling deep entrenchment of hardened ROC positions. Thus, any attempt to physically remove ROC forces from the area would likely entail fierce fighting and severe casualties. Second, despite the islands’ formal administrative status within the ROC as part of Fujian rather than Taiwan, Kinmen has acquired a symbolic status there as the front line of defense against PRC aggression. The history of successfully repelling PRC amphibious landing attempts at Kinmen is relatively well known in Taiwan. In addition, the fate of Kinmen’s significant civilian population could become a ready rallying cause in Taiwan if the PRC attacked. Even assuming the United States would not intervene militarily to prevent a PRC take-over of Kinmen, it would be difficult for the PRC to keep in the “gray zone,” The scale and violence of the PRC operation would likely entail major economic sanctions from the U.S. That said, America’s allies and partners in the region would be unlikely to participate actively in any sanctions response to a violent or nonviolent PRC takeover of Kinmen. Nor could the European Union be counted on to actively support sanctions in the short to medium term, given the bitter costs of imposing sanctions on Russia in 2022–2023 over its invasion of Ukraine. Still, unless PRC-U.S. trading relations wither to a fraction of their current size, the PRC is highly unlikely to calculate that the gains of a bloody Kinmen assault would outweigh the costs.

The PRC has more appealing options than Kinmen if it seeks to seize ROC-occupied features along the Fujian coast. One is the little-known Wuchiu (Wuqiu) Islands, which also sit administratively under the ROC’s Kinmen County, around 130 kilometers northeast of Kinmen, near the PRC city of Quanzhou. The Wuchius are far smaller at only 1.2 square kilometers, with far less military significance and a population of fewer than 400, along with several hundred soldiers. While Kinmen’s military significance in preventing a takeover of Taiwan was hotly contested in the 1950s, the Wuchius were declared expendable: a 1954 U.S. National Security Council briefing casually described the Wuchius as “isolated and unimportant.” The islands have suffered from ongoing underinvestment in infrastructure, weak transport links, and a struggling economy due to declining fish stocks. The ROC Navy deployed around 20 modern remote-controlled XTR-102 guns to the islands in 2019 — further decreasing the need to physically staff these defenses.

Wuchiu’s combination of low strategic significance, geographic proximity to the PRC coast and isolation from other ROC outposts, small population and relatively low political profile within Taiwan could make a PLA seizure of one or both of the islands appealing. Doing so would advance the PLA’s geographical presence, albeit incrementally, and offer opportunities to gain crucial experience in modern amphibious operations. As a symbolic move, it would have the capacity to rally or pacify nationalist sentiments inside China, while in Taiwan, political will to resist would be relatively lower than for other, more symbolically invested or populated targets.

During the 1996 Taiwan Strait Crisis, Taiwanese observers flagged the Wuchiu Islands as a possible target for a PRC attack. Since that time, intensive mainland Chinese fishing activities have been reported around the islands in many years, giving the PRC ample opportunities for intelligence-gathering in the area and planning for a surprise attack. Among inhabited Taiwan-held islands with a permanent civilian population, the Wuchius would be among the least risky options for a PRC seizure operation; a 2003 U.S. assessment reportedly found that the ROC Marine Corps there would be ineffective in the event of invasion. However, as detailed next, the Matsu Islands to the north present uninhabited options.

Further north, beyond the Taiwan Strait, the Matsu Islands consist of 36 islands and rocks in the East China Sea, located around 10 kilometers offshore from the Fujianese provincial capital of Fuzhou. Since the 1950s, English-language policy discourse has frequently rolled the group together with Kinmen (“Quemoy and Matsu”), but crucial differences exist in their geographical, administrative, and human circumstances. The two ROC-held island groups share a similar granite geology, history of heavy ROC military presence and PRC bombardments, and a contemporary economic profile based on new tourism industries. However, the Matsu archipelago has only a fraction of Kinmen’s land area and population — around 20 percent in total — and its features are far more dispersed, offering the PRC much more flexible options for quick and bloodless armed seizures.

Measuring around 10 square kilometers, Matsu’s main island of Nangan is substantially smaller even than Lesser Kinmen Island. Just over half of Matsu’s estimated 12,000 civilians live there, while around 1,200–1,500 live in each of the outlying Beigan, Dongyin, and Juguang townships, which represent the northern, eastern, and southern subgroups of the Matsu Islands, respectively (see map). Nangan and Beigan have airports, a major attraction for the PLA. However, each of the outlying townships also contains numerous uninhabited islands and islets, potentially offering the PRC the option to rapidly and bloodlessly seize a foothold in the archipelago. As early as 1956, the U.S. Central Intelligence Agency reported that the group’s outlying Juguang (“White Dogs”) and Dongyin (“Yinshans”) islets were both “highly vulnerable to Communist strikes” and “of little military value” to the ROC.

Seizure of an uninhabited feature in the Matsu group would leave Taipei with the undesirable choice of either initiating the use of force to evict PRC occupiers from an islet of little inherent value, or tolerating a new PLA presence in the area. Given the Matsus’ relatively strong name recognition among domestic and international audiences, a PRC island seizure there could prompt economic sanctions from the United States and its closest allies. Yet the benefits could still outweigh the costs for the PRC: an amphibious operation to establish and sustain even a small presence there would provide valuable operational experience, as well as information-gathering opportunities to underpin possible further advancements, and the chance to gauge ROC responses military and political responses. Beijing could then plan further steps to incrementally increase pressure on the ROC presence in the Matsu Islands, ranging from harassment and maritime militia “presence operations” up to a blockade of one or more of the archipelago’s inhabited islands.

It is far from clear that Taiwan would resist an island seizure in the Matsus. The United States has never made any commitment to defend ROC control there, and their strategic value likely pales in comparison with Kinmen. The Matsus are located outside the Taiwan Strait and belong administratively to the ROC’s Fuchien Province. As a historical legacy of Chiang Kai-shek’s plans to eventually retake the mainland, some in the independence-leaning Democratic Progressive Party see the islands as not merely lacking in strategic value, but also an inconvenient connection to the mainland. Just as Mao Zedong and Zhou Enlai recognized the value of the ROC’s Kinmen and Matsu outposts to the cause of unification, some Taiwanese nationalists regard the ROC’s occupation of the Matsus as ultimately unhelpful to the cause of independence. In such scenarios, the PRC would have opportunities to pair its real-world advancements with information operations to promote acquiescence among its staunchest opponents in Taiwan.

Recent commentary and analysis discussing possible PRC seizures of ROC-held off-shore islands have focused on two features in the South China Sea: Pratas and Itu Aba. A Chinese move there could offer some advantages. The large, isolated atoll of Pratas provides the only airstrip within a several-hundred-kilometer radius, and arguably it constitutes an important node for sea control out to the Bashi Strait. Itu Aba, meanwhile, is the largest naturally formed island in the Spratly archipelago, and it sits at the center of the PRC’s narrative supporting its expansive claims in the South China Sea. On the whole, however, the PRC has more to lose from a move against Taiwan in the South China Sea than to gain — particularly at Itu Aba, whose seizure by PRC forces would assuredly alarm Southeast Asian claimants occupying other nearby disputed features.

The northernmost Taiwanese outpost in the South China Sea is Pratas (Dongsha) Island, with a land area of around 1 square kilometer perched atop a shallow, circular mid-ocean atoll measuring about 25 kilometers across. Pratas is isolated, situated just over 250 kilometers from the coast of Guangdong and nearly 450 kilometers from Kaohsiung in Taiwan, to which it belongs administratively. Since 2000, it has been garrisoned by around 200 personnel from Taiwan’s Coast Guard Administration, whose sustained presence is facilitated by a 1.5-kilometer airstrip. In May 2020, rumors swirled about PLA training drills involving Pratas as the imagined target, and China held naval drills between Pratas and the Guangdong coast that September. Taiwan’s Coast Guard reportedly detected PRC drones in the area in April 2022, warning that their entry into Taiwan’s territorial airspace could prompt ROC forces to open fire.

One factor favoring Pratas as a target is the absence of civilian residents. Taipei designated the atoll a marine park in 2007, allowing scientists to visit, but it has remained off-limits to tourists and residents. Some analysts have posited that the feature may have significance for anti-submarine warfare — indeed, a PLA Navy sub reportedly malfunctioned while operating nearby 2005 — while others emphasize its location on Taiwan’s southern approaches and between the Chinese mainland and the strategic Bashi Strait. At a minimum, Pratas’s relative isolation means that it would stand to benefit the PRC’s maritime domain awareness. PLA-affiliated commentators have warned against potential U.S. military use of the feature, a concern that might explain the PRC’s apparent targeting of anti-submarine warfare helicopters in the area in April 2022.

However, such advantages are unlikely to outweigh the major downside risks for Beijing. First, given the island’s sandy, low-lying geology, seizing control would be easier than keeping it. Further, a takeover would not likely be bloodless, since Taipei has been strengthening its armed presence there. Besides a reported 500 marines, military-trained Coast Guard personnel have been equipped since 2021 with shoulder-launched, armor-piercing rockets that would trouble an amphibious assault force. English-language reporting has often characterized PLA Air Force entries into Taiwan’s Air Defense Identification Zone as “near” the Pratas Islands, but the PLA flight paths released by Taiwan’s Defense Ministry suggest little inclination to approach the atoll. In November 2021, Taiwanese official Chen Ming-tong explicitly tamped down speculation about a potential Pratas operation.

A second potential South China Sea island target is Itu Aba (Taiping) Island, in the northwestern Spratly Islands. A palmfringed, sandy island of only 0.5 square kilometers and virtually no elevation, Itu Aba, like Pratas, is tiny and hard to defend in comparison with the rocky islets to the north. Its significance lies primarily in its status as the largest naturally formed island in the sprawling Spratly archipelago, and in the fact that Vietnam and the Philippines both claim it as their sovereign territory. A move on Taiping could therefore be presented domestically as taking land from Taiwan while strengthening Beijing’s position in the broader South China Sea dispute vis-à-vis Southeast Asian rival claimants.

But key disadvantages associated with an operation to seize Pratas Island also apply to Itu Aba. The ROC maintains similar defensive deployments to those at Pratas, meaning that a seizure would likely require some level of violence, and a successful occupation force would be exposed and vulnerable. Any such operation would alarm Southeast Asian claimants that occupy nearby features.

The political disincentives to military seizure are perhaps even more important, for the ROC’s occupation of Taiping forms a central element of the historical narrative by which the PRC has sought to justify its expansive claims in the South China Sea. In particular, the PRC has claimed that the ROC “resumed” sovereignty over the islands in 1946, before the Southeast Asian claimant states were present. Taipei and Beijing even found rare agreement on the 2016 Award in the South China Sea Arbitration case brought by the Philippines against China under the United Nations Convention on the Law of the Sea (UNCLOS), when both rejected the tribunal’s ruling that none of the Spratly Islands, including Taiping, are entitled to an exclusive economic zone or continental shelf.

The ROC’s positions on Itu Aba have helped the PRC add respectability to its South China Sea claims in the face of their systematic legal dismantling by the arbitration panel. Under Ma Ying-jeou’s KMT administration (2008–2016), Taiwan actively sought to persuade the international community that Taiping Island met the criteria for an exclusive economic zone and continental shelf under UNCLOS, with Ma even drinking water from a well on the island in an attempt to demonstrate its habitability. A Chinese seizure of the island would effectively eliminate the rare possibility of working with future ROC administrations over the issue, perhaps to jointly develop resources in the Spratlys. Both sides of politics in Taiwan appear committed to maintaining the ROC positions from which the PRC’s South China Sea claims are derived. President Tsai Ing-wen has restated the claim of ROC sovereignty over “the South China Sea islands and their relevant waters,” and said the 2016 arbitral ruling “seriously undermines our country’s rights in the South China Sea.” Such bipartisan alignment in Taiwan greatly reduces the potential benefits of using force to take control of Itu Aba.

There would be little material benefit to the PRC from seizing Taiping, as it already controls nearly 13 square kilometers of land in the Spratlys — 26 times the land area of Taiping — thanks to its campaign of island-building in the 2010s. Seizing Itu Aba could also entail a high risk of meeting Taiwanese resistance. Unlike Kinmen, Matsu, and Wuchiu, there is a history of Japanese colonial occupation of both Pratas and Itu Aba via Taiwan that could underpin a future independent Taiwanese claim to those features. The choice of Kaoh-siung City to administer the South China Sea outposts — as Japan’s colonial authority had done—suggests the ongoing relevance of such a linkage. Thus, there are a multitude of political and military reasons why the PRC is not likely to seek to seize Itu Aba — or Pratas — from Taiwan.

The most consequential target for a PRC island seizure would be the Penghu archipelago, located about 40 kilometers off Taiwan’s southwest coast, and home to one of the best ports in the area at Magong Island. Part of the Qing dynasty’s Taiwan Prefecture (a province from 1887), the Penghus are known in English as the Pescadores. The archipelago includes a total of 90 islands and rocks that add up to a total land area of more than 140 square kilometers, a relatively stable civilian population of more than 100,000, and an economy that delivers Taiwan’s seventh-highest regional per capita GDP. The ROC’s Penghu Defense Command is a capable corps-level unit of 8,000 troops, equipped with tanks, long-range radar, and anti-ship and anti-air-craft missiles. Lying so close to Taiwan, an all-out assault on the Penghus would likely be viewed as a prelude to an invasion of Taiwan itself. Yet the PLA may have options below this threshold: only around half of the 90 islands are inhabited, including 16 reportedly with ROC troops present, and the archipelago’s low-lying geography — in contrast with Kinmen and Matsu — would make many individual features relatively simple to land on.

Analyses of PLA military strategy for Taiwan generally concur that occupation, or at least neutralization, of the Penghus would be central to any plan to cross the Strait. Seizing the archipelago as a whole would present major challenges to the PLA because of the significant civilian population and Taiwan’s defensive forces. But, rather than seeking to attack or dislodge Taiwan’s forces in the area, China might instead send a small company of troops — or, applying a template from the South and East China Sea disputes, maritime militia or patriotic activists — to land on a small island at the western edge of the archipelago. Such an operation would deliberately minimize, and possibly avoid, contact with the ROC forces stationed in the area, placing the onus on Taipei to respond.

Even if it were designed to avoid confrontation, a Penghu island occupation would be a risky operation for the PLA. Taiwanese forces might well move swiftly to evict or arrest intruders. However, choosing the level of force to employ — and the circumstances in which it could be used — would present a serious dilemma for Taipei. Authorizing the use of lethal force to implement an arrest or eviction of armed intruders, especially of unclear institutional identity, would risk sparking a public crisis that could provide the PRC with a pretext with which to justify further military reinforcement of the new position.

A successful PRC occupation of any feature in the Penghus would assuredly prompt ROC countermeasures, though the extent to which the United States would support Taipei in any subsequent escalation is less clear. On one hand, unlike the island groups discussed earlier, the Penghus were explicitly mentioned in the United States’ 1954 treaty commitments to the ROC, and in the 1979 Taiwan Relations Act.

On the other hand, it may be difficult for the United States to mount a national interest argument for committing to back a Taiwanese forceful eviction of Chinese presence from a small islet there, especially if the occupation had been quick and bloodless. Doing so would risk being perceived as escalating and pushing the region toward a general war. A more likely U.S. response would be some level of economic sanctions against the PRC — but even Washington might struggle domestically to mobilize a full-scale trade embargo over a tiny unoccupied islet. Meanwhile, the PRC would retain and normalize the new foothold, reconnaissance outpost, and potential future staging post.

PLA propaganda has clearly signaled the military’s interest in Penghu. Yet, given the archipelago’s proximity to Taiwan — as well as the islands’ long-term administrative status as part of Taiwan — a Chinese island seizure operation in the Penghus would likely prompt ROC resistance. Still, with dozens of uninhabited islands to choose from, a Penghu operation would be preferable to an attack on Kinmen, the populated islands in the Matsu archipelago, or Taiwan’s main island.

If successful, it would represent an outsized material advancement toward control of Taiwan in a way that seizure of other island groups discussed earlier would not. Even a tenuous presence in the archipelago would provide the PLA with enhanced intelligence and reconnaissance in the area to underpin an eventual confrontation of ROC forces there, as well as a potential staging post for any attack on the main island.

Compared with an outright invasion or full-press blockade of Taiwan’s main island, “gray-zone” seizure of an ROC-held outlying island is arguably a much more likely cause of a crisis in the Taiwan Strait in the 2020s. The PRC’s intent to “reunify” Taiwan is clear, but the absence of serious invasion attempts in recent decades attests to Beijing’s professional respect for the ROC military and caution at the prospect of escalation and U.S. intervention. Absent a major authoritative reappraisal of American power and resolve, the United States’ prospective involvement will likely continue to deter the PRC from taking any actions as clear-cut as a cross-Strait invasion. Yet, as this paper has laid out, Beijing has an array of more viable options to advance its position below that threshold. And unlike invasions or blockades of Taiwan, the PRC has successfully seized offshore islands from the ROC before, such as the Yijiangshan and Dachen Islands in 1955. Policymakers and strategists on all sides of politics in Taiwan, the United States and elsewhere need to carefully consider how they might respond to such contingencies.

Mobilizing an effective international response to such an operation would be far more difficult than it was following Vladimir Putin’s unambiguous invasion of Ukraine in February 2022. Whereas Russia rolled columns of tanks across clearly demarcated boundaries in a manner recognizable as an invasion to audiences the world over, China could send state or nonstate actors across short stretches of water to bloodlessly occupy islands that Taiwan itself regards administratively as Chinese territory — and located far closer to China than to Taiwan on a map. Such prospects should be particularly concerning given that control of many of these offshore islands may be necessary steps toward any eventual PRC military takeover of Taiwan. The PLA’s limited operational experience, particularly in amphibious operations, suggests that it is likely to engage in probes well before any invasion of the island of Taiwan.

The progression of the South China Sea dispute since the PRC established a foothold in the Spratly Islands in December 1987 suggests that — despite well-documented deficiencies in lower-level coordination, particularly before the Xi Jinping era — Beijing is capable of managing strategic surges in assertive gray-zone activities to advance its position in key maritime and territorial disputes without triggering conflict. Given this history, concerns about the possibility of an eventual PRC invasion of Taiwan should also be significantly focused on potential gray-zone operations against outlying islands in the short and medium term.

The foregoing analysis suggests three policy takeaways. First, for Taiwanese leaders, the
likelihood of limited gray-zone operations implies a need for continued investment in maritime domain awareness to monitor movements and detect micro-level changes that could constitute warning signs. They should also ensure credible plans are in place to destroy any important infrastructure on targeted islands should they fall. It is equally important to make sure that the PRC is aware of this, in order to minimize the chances that Beijing overestimates the perceived material benefits of proceeding with such an island seizure operation.

Second, policymakers in all countries need to be attuned to the possibility that the PRC’s domestic politics may incentivize such actions more so than in the past. The scenarios described here are all highly symbolic but relatively controllable, making them potentially useful targets for diversionary conflict. With the Chinese economy slowing down, Xi Jinping’s incentives to resort to externally directed aggression to rally public support or redirect public attention may be increasing. Diversionary war arguments have found little supporting evidence from China over the past few decades; when Beijing has been faced with internal crises in the past, it has more often sought to retrench rather than escalate internationally. In the wake of the Tiananmen crackdown in June 1989, the PRC’s assertive advances toward control of the South China Sea temporarily ceased. Similarly, during the Asian financial crisis of the late 1990s, the PRC opted for a relatively low-key foreign policy. Yet today’s circumstances may be different: the PRC economy of the 2020s is more developed; its citizens hold higher baseline expectations for living standards; its society is certainly more informationized; and its information ecosystem is more nationalistic.

Third, Taiwan’s government and aligned governments with an interest in maintaining the status quo should conduct detailed working-level discussions of possible economic responses to such operations. This should start with key regional partners such as Japan, South Korea, Singapore, Australia, and the Philippines and build out to include India, Indonesia, Canada, the United Kingdom, and the European Union. Vietnam is unlikely to play any active part in pushing back, but it would likely offer tacit support given its constant concern about PRC advancements in the South China Sea. While a region-wide united front against aggression may not be possible, it is vital that such collective responses as can be arranged are clearly visible to Beijing.

1. Kinmen County is nominally home to more than 140,000, though it is estimated that around half are at most part-time residents.