In her recent Foreign Affairs essay, The Taiwan Temptation: Why Beijing Might Resort to Force, Center Fellow Oriana Skylar Mastro argues that Chinese leaders now consider a military campaign to take Taiwan a real possibility and cautions that the United States cannot by itself alter Beijing’s calculus on Taiwan. The essay sparked a heated debate. In the September/October issue of Foreign Affairs, several scholars — Rachel Esplin Odell and Eric Heginbotham, Bonny Lin and David Sacks, and Kharis Templeman — provide counterarguments to Mastro's analysis and she responds to their criticism. Read her complete rebuttal below.

Sign up for APARC newsletters to receive our experts' analysis and commentary. 

Rachel Esplin Odell and Eric Heginbotham, Bonny Lin and David Sacks, and Kharis Templeman all argue that China is unlikely to attempt armed unification with Taiwan. Although I appreciate their perspectives, they do not present any new evidence that would make me reconsider my assessment that the risk of Chinese aggression across the Taiwan Strait is real and growing. To the contrary, they repeat many of the increasingly dangerous misperceptions that I sought to dispel in my original article—namely, that China does not have the military capabilities to pull off an amphibious invasion, that the economic costs of an invasion would be sufficient to deter Chinese President Xi Jinping, and that China can afford to wait indefinitely to achieve its most important national goal of unification. My critics assume that insofar as there are risks, they can be dealt with through relatively limited adjustments in U.S. policy and military posture — a position with which I still strongly disagree. 

Let’s take these arguments in order. My critics say that I have exaggerated China’s military capabilities and understated the difficulties of an invasion. But their assessments rely on outdated or largely irrelevant comparisons. Odell and Heginbotham, for instance, note that the United States needed more naval tonnage to capture Okinawa from Japan in 1945 than China has today. But this example is inapposite. Japan’s military was more than six million strong in 1945 and had been fighting for over a decade; Taiwan’s military consists of 88,000 personnel and two million reservists, of whom only 300,000 are required to complete even a five-week refresher training course. Tonnage, moreover, is not a useful metric. Modern navies have moved to lighter, more flexible fleets. Odell and Heginbotham point out that civilian ships were of only limited use in the Falklands War, but the United Kingdom used just 62 of them in that campaign. The People’s Armed Forces Maritime Militia has many thousands of ships and is closer to a naval force than a civilian one. If China were to mobilize all its naval vessels, including its new large amphibious transport ships and civilian ships, it could hypothetically carry hundreds of thousands of troops across the 80-mile-wide Taiwan Strait in a short period of time. Even if the United States had enough warning to optimally position its submarines, it does not have enough munitions to target such a large force. 

For their part, Lin and Sacks argue that to believe China can take Taiwan by force is to fall for a Chinese misinformation campaign. They warn that “analysts should not accept at face value China’s claim that it could easily win a fight against Taiwan.” But no one, not even the cockiest of People’s Liberation Army analysts, argues that a full-scale attack on Taiwan would be easy, only that the PLA could prevail at an acceptable cost. Moreover, my assessment of Chinese military capabilities is not based on Chinese discourse or the results of war games alone. Reams of unbiased and rigorous analysis—from the U.S. Department of Defense’s annual report to Congress on China’s military modernization to Congressional Research Service reports on Chinese naval modernization to hundreds of studies by think tanks and defense-affiliated organizations, such as the RAND Corporation—suggest that the PLA has made unparalleled advances in the past two decades and could take on the United States in certain scenarios. Indeed, Heginbotham himself argued in 2017 that “the balance of power between the United States and China may be approaching a series of tipping points, first in contingencies close to the Chinese coast (e.g., Taiwan).” 

I do not mean to suggest that a Chinese invasion would be a cakewalk. Taiwan could get some shots in, but it does not have the ability to defend itself. Luckily, the United States would, I believe, come to Taiwan’s aid and could still prevail in many scenarios. Taiwan is far from a lost cause. But ten years ago, the United States would have prevailed in any scenario. Because there are now some scenarios in which U.S. strategists think the United States could lose, it is not unfathomable to think that Chinese strategists have come to a similar conclusion. 

My critics also argue that economic considerations will deter Beijing. Should China attempt to use force to assert control over Taiwan, the international response would be severe enough to imperil Xi’s ambitious development goals. But as I argued in my original article, Chinese analysts have good reason to think the international response would be weak enough to tolerate. China could even reap economic benefits from controlling Taiwan, whose manufacturers accounted for more than 60 percent of global revenue from semiconductors last year. The United States is heavily reliant on Taiwanese semiconductors. Should China take Taiwan, it could conceivably deprive the United States of this technology and gain an economic and military advantage. 

But economic costs or benefits, while part of Beijing’s calculus, are unlikely to be the determining factor. Xi’s top priority is protecting China’s sovereignty and territorial integrity—as Beijing defines it. China’s Belt and Road Initiative, its militarization of the South China Sea, and its sanctions against countries that offend it, such as Australia or South Korea, all demonstrate that Chinese leaders are willing to subordinate economic considerations to considerations of power and prestige. In a speech marking the 100th anniversary of the Chinese Communist Party in July, Xi warned against foreign attempts to bully or oppress China, declaring that “anyone who dares try to do that will have their heads bashed bloody against the great wall of steel forged by over 1.4 billion Chinese people.” Those words should be taken seriously. 

Finally, my critics argue that China has no need to attempt to forcibly unify with Taiwan. Lin and Sacks think peaceful unification is working; Templeman believes China can wait indefinitely to resolve the issue. I disagree because I think unification is a top priority for the Chinese Communist Party and Taiwan will not give up its autonomy without a fight. 

A Chinese invasion is by no means imminent or inevitable, but Beijing is now seriously considering initiating a conflict to gain political control over Taiwan, whereas in the past the only scenario in which it would have used force was to prevent Taipei from declaring independence. I agree with Templeman that China is unlikely to invade in the next four years (although I think this is largely because China could benefit from more time to prepare, not because it fears U.S. President Joe Biden’s resolve), but his argument that China can wait indefinitely is logically and empirically flawed. As I argued in my original article, Xi has made numerous statements that suggest he wants to achieve unification during his reign. It would be unwise to dismiss these as mere rhetoric, since he has repeatedly voiced his intention to assert control over other territorial claims before doing exactly that — in the South China Sea, by building military infrastructure and conducting naval drills, and in Hong Kong, by imposing a harsh national security law last year.

Templeman argues that if China believes the United States is in decline, then it has every reason to wait on Taiwan. But in the eyes of Chinese strategists, American decline actually hastens the need for action. Power transition theory, which holds that war becomes more likely as the gap between a rising power and an established great power diminishes, is also studied in Beijing. And although U.S. strategists fret that a rising China, dissatisfied with the U.S.-led international order, will become aggressive and start a conflagration, Chinese strategists fear a different pathway to war. They worry that the United States, unable to accept its inevitable decline, will make a dangerous last-ditch effort to hold on to its unrivaled great-power status. By this logic, a declining United States is more dangerous than a stable, ascendant one. 

Lin and Sacks make a different argument for why Beijing does not need to attempt armed unification. They believe that Chinese leaders remain committed to their long-standing approach of limited coercion coupled with economic incentives showcasing the benefits of unification because that strategy is working. As evidence of Beijing’s progress, Lin and Sacks point to polling that shows the majority of people in Taiwan support the status quo, not independence. But it is an enormous leap from not supporting independence to desiring or conceding to unification. As Lin and Sacks themselves acknowledge, China has employed this strategy of limited coercion and economic inducements for decades, but Taiwan is no closer to being a part of mainland China. In a September 2020 poll conducted by National Chengchi University, only six percent of Taiwanese citizens preferred eventual or immediate unification. So although Lin and Sacks are correct that Beijing will likely continue with its carrot-and-stick approach, it will still need to put boots on the ground to gain full political control of Taiwan. 

My critics also raise concerns about some of the policy implications of my argument. Odell and Heginbotham warn against focusing too much on the credibility of the U.S. military threat when it comes to deterrence, rightly highlighting the equal importance of reassurance. They warn that changes in U.S. policy toward Taiwan could convince Beijing that the United States now supports Taiwanese independence — a misperception that could lead to war. But my argument is for a change in posture, not in policy: the United States should develop the force posture and operational plans to deny China its objective in Taiwan and then credibly reveal these new capabilities. It should not make dangerous policy changes that would risk provoking a Chinese military response. Indeed, I have argued elsewhere that even if a war breaks out over Taiwan and the United States wins, Washington should not demand Taiwan’s independence as one of the terms of peace. 

Templeman raises a separate concern: that highlighting the potential costs of defending Taiwan could bolster the case of those advocating that Washington abandon Taipei. If this were a serious worry, I would be the first to shift my work to more private channels. But those calling for the United States to reconsider its commitment to defend Taiwan are still in the minority, and the Biden administration has been clear that it would come to Taiwan’s aid in the event of an invasion.

Moreover, the reaction of the U.S. Department of Defense to the threat posed by China’s growing military power has been not to back down but to ramp up efforts to counter it. From new doctrines that enhance joint capabilities between the U.S. Air Force and the U.S. Navy to base-resilience initiatives to efforts to improve U.S. early warning systems in the region, the Pentagon is firing on all cylinders to ensure it can deter and, if necessary, defeat China in a wide range of conflict scenarios. U.S. Cyber Command, the U.S. Space Force, and the Department of Defense’s Joint Artificial Intelligence Center were all established partly to counter Chinese advantages in those organizations’ respective domains. If Lin and Sacks are correct that China exaggerates its capabilities to try to convince the United States to give up, Beijing has achieved the opposite.   

In the end, all my critics highlight an important truth: the situation across the Taiwan Strait has been relatively stable for 70 years because of the United States. Washington has managed to convince Beijing that armed unification would fail and that China would pay a hefty price for trying. But China is not the same country it was 70 years ago. Its rapid military modernization, spectacular economic ascent, and growing global influence have changed Beijing’s calculus on many issues. It has taken a more assertive approach to international institutions; built one of the world’s largest, most capable militaries; and extended its economic influence deep and far throughout the world. It would be wishful thinking to assume that China has not also changed its thinking on Taiwan.

Indeed, although my critics argue that China is unlikely to invade, they still recommend that Taiwan improve its defenses and that the United States enhance its military posture in the region — not exactly a vote of confidence in Beijing’s restraint. I had hoped to convince skeptics that China is now seriously considering armed unification, but at least our debate has yielded a consensus that more must be done in Taipei and Washington to enhance deterrence across the Taiwan Strait.

This essay was originally published in Foreign Affairs magazine.

Chinese leaders will continue to pay lip service to peaceful unification until the day the war breaks out, but their actions increasingly suggest that they have something else in mind. As tensions with the United States have heated up, China has accelerated its military operations in the vicinity of Taiwan, conducting 380 incursions into the island’s air defense identification zone in 2020 alone. In April of this year, China sent its largest-ever fleet, 25 fighters and bombers, into Taiwan’s air defense identification zone. Clearly, Xi is no longer trying to avoid escalation at all costs now that his military is capable of contesting the U.S. military presence in the region. Long gone are the days of the 1996 crisis over Taiwan, when the United States dispatched two aircraft carrier battle groups to sail near the strait and China backed off. Beijing did not like being deterred back then, and it spent the next 25 years modernizing its military so that it would not be so next time.  

Much of that modernization, including updates to hardware, organization, force structure, and training, was designed to enable the People’s Liberation Army to invade and occupy Taiwan. Xi expanded the military’s capabilities further, undertaking the most ambitious restructuring of the PLA since its founding, aimed specifically at enabling Chinese forces to conduct joint operations in which the air force, the navy, the army, and the strategic rocket force fight seamlessly together, whether during an amphibious landing, a blockade, or a missile attack—exactly the kinds of operations needed for armed unification. Xi urgently pushed these risky reforms, many unpopular with the military, to ensure that the PLA could fight and win wars by 2020.

The voices in Beijing arguing that it is time to use these newfound military capabilities against Taiwan have grown louder, a telling development in an era of greater censorship. Several retired military officers have argued publicly that the longer China waits, the harder it will be to take control of Taiwan. Articles in state-run news outlets and on popular websites have likewise urged China to act swiftly. And if public opinion polls are to be believed, the Chinese people agree that the time has come to resolve the Taiwan issue once and for all. According to a survey by the state-run Global Times, 70 percent of mainlanders strongly support using force to unify Taiwan with the mainland, and 37 percent think it would be best if the war occurred in three to five years. 

The Chinese analysts and officials I have spoken to have revealed similar sentiments. Even moderate voices have admitted that not only are calls for armed unification proliferating within the CCP but also they themselves have recommended military action to senior Chinese leadership. Others in Beijing dismiss concerns about a Chinese invasion as overblown, but in the same breath, they acknowledge that Xi is surrounded by military advisers who tell him with confidence that China can now regain Taiwan by force at an acceptable cost. 

Battle Ready

Unless the United States or Taiwan moves first to alter the status quo, Xi will likely consider initiating armed unification only if he is confident that his military can successfully gain control of the island. Can it? 

The answer is a matter of debate, and it depends on what it would take to compel Taiwan’s capitulation. Beijing is preparing for four main campaigns that its military planners believe could be necessary to take control of the island. The first consists of joint PLA missile and airstrikes to disarm Taiwanese targets—initially military and government, then civilian—and thereby force Taipei’s submission to Chinese demands. The second is a blockade operation in which China would attempt to cut the island off from the outside world with everything from naval raids to cyberattacks. The third involves missile and airstrikes against U.S. forces deployed nearby, with the aim of making it difficult for the United States to come to Taiwan’s aid in the initial stages of the conflict. The fourth and final campaign is an island landing effort in which China would launch an amphibious assault on Taiwan—perhaps taking its offshore islands first as part of a phased invasion or carpet bombing them as the navy, the army, and the air force focused on Taiwan proper. 

Among defense experts, there is little debate about China’s ability to pull off the first three of these campaigns—the joint strike, the blockade, and the counterintervention mission. Neither U.S. efforts to make its regional bases more resilient nor Taiwanese missile defense systems are any match for China’s ballistic and cruise missiles, which are the most advanced in the world. China could quickly destroy Taiwan’s key infrastructure, block its oil imports, and cut off its Internet access—and sustain such a blockade indefinitely. According to Lonnie Henley, a retired U.S. intelligence officer and China specialist, “U.S. forces could probably push through a trickle of relief supplies, but not much more.” And because China has such a sophisticated air defense system, the United States would have little hope of regaining air or naval superiority by attacking Chinese missile transporters, fighters, or ships. 

But China’s fourth and final campaign—an amphibious assault on the island itself—is far from guaranteed to succeed. According to a 2020 U.S. Department of Defense report, “China continues to build capabilities that would contribute to a full-scale invasion,” but “an attempt to invade Taiwan would likely strain China’s armed forces and invite international intervention.” The then commander of U.S. Indo-Pacific Command, Philip Davidson, said in March that China will have the ability to successfully invade Taiwan in six years. Other observers think it will take longer, perhaps until around 2030 or 2035. 

What everyone agrees is that China has made significant strides in its ability to conduct joint operations in recent years and that the United States needs adequate warning to mount a successful defense. As Beijing hones its spoofing and jamming technologies, it may be able to scramble U.S. early warning systems and thereby keep U.S. forces in the dark in the early hours of an attack. Xi’s military reforms have improved China’s cyberwarfare and electronic warfare capabilities, which could be trained on civilian, as well as military, targets. As Dan Coats, then the U.S. director of national intelligence, testified in 2019, Beijing is capable of offensive cyberattacks against the United States that would cause “localized, temporary disruptive effects on critical infrastructure.” China’s offensive weaponry, including ballistic and cruise missiles, could also destroy U.S. bases in the western Pacific in a matter of days.

In light of these enhanced capabilities, many U.S. experts worry that China could take control of Taiwan before the United States even had a chance to react. Recent war games conducted by the Pentagon and the RAND Corporation have shown that a military clash between the United States and China over Taiwan would likely result in a U.S. defeat, with China completing an all-out invasion in just days or weeks.

Ultimately, on the question of whether China will use force, Chinese leaders’ perceptions of their chances of victory will matter more than their actual chances of victory. For that reason, it is bad news that Chinese analysts and officials increasingly express confidence that the PLA is well prepared for a military confrontation with the United States over Taiwan. Although Chinese strategists acknowledge the United States’ general military superiority, many have come to believe that because China is closer to Taiwan and cares about it more, the local balance of power tips in Beijing’s favor. 

As U.S.-Chinese tensions have risen, China’s state-sponsored media outlets have grown more vocal in their praise for the country’s military capabilities. In April, the Global Times described an unnamed military expert saying that “the PLA exercises are not only warnings, but also show real capabilities and pragmatically practicing reunifying the island if it comes to that.” If China chooses to invade, the analyst added, the Taiwanese military “won’t stand a chance.”

Go Fast, Go Slow

Once China has the military capabilities to finally solve its Taiwan problem, Xi could find it politically untenable not to do so, given the heightened nationalism of both the CCP and the public. At this point, Beijing will likely work its way up to a large-scale military campaign, beginning with “gray zone” tactics, such as increased air and naval patrols, and continuing on to coercive diplomacy aimed at forcing Taipei to negotiate a political resolution. 

Psychological warfare will also be part of Beijing’s playbook. Chinese exercises around Taiwan not only help train the PLA but also wear down Taiwan’s military and demonstrate to the world that the United States cannot protect the island. The PLA wants to make its presence in the Taiwan Strait routine. The more common its activities there become, the harder it will be for the United States to determine when a Chinese attack is imminent, making it easier for the PLA to present the world with a fait accompli.

At the same time that it ramps up its military activities in the strait, China will continue its broader diplomatic campaign to eliminate international constraints on its ability to use force, privileging economic rights over political ones in its relations with other countries and within international bodies, downplaying human rights, and, above all, promoting the norms of sovereignty and noninterference in internal affairs. Its goal is to create the narrative that any use of force against Taiwan would be defensive and justified given Taipei’s and Washington’s provocations. All these coercive and diplomatic efforts will move China closer to unification, but they won’t get it all the way there. Taiwan is not some unoccupied atoll in the South China Sea that China can successfully claim so long as other countries do not respond militarily. China needs Taiwan’s complete capitulation, and that will likely require a significant show of force. 

If Beijing decides to initiate a campaign to forcibly bring Taiwan under Chinese sovereignty, it will try to calibrate its actions to discourage U.S. intervention. It might, for example, begin with low-cost military options, such as joint missile and airstrikes, and only escalate to a blockade, a seizure of offshore islands, and, finally, a full-blown invasion if its earlier actions fail to compel Taiwan to capitulate. Conducted slowly over the course of many months, such a gradual approach to armed unification would make it difficult for the United States to mount a strong response, especially if U.S. allies and partners in the region wish to avoid a war at all costs. A gradual, coercive approach would also force Washington to initiate direct hostilities between the two powers. And if China has not fired a shot at U.S. forces, the United States would find it harder to make the case at home and in Asian capitals for a U.S. military intervention to turn back a slow-motion Chinese invasion. An incremental approach would have domestic political benefits for Beijing, as well. If China received more international pushback than expected or became embroiled in a campaign against the United States that started to go badly, it would have more opportunities to pull back and claim “mission accomplished.”  

But China could decide to escalate much more rapidly if it concluded that the United States was likely to intervene militarily regardless of whether Beijing moved swiftly or gradually. Chinese military strategists believe that if they give the United States time to mobilize and amass firepower in the vicinity of the Taiwan Strait, China’s chances of victory will decrease substantially. As a result, they could decide to preemptively hit U.S. bases in the region, crippling Washington’s ability to respond.

In other words, U.S. deterrence—to the extent that it is based on a credible threat to intervene militarily to protect Taiwan—could actually incentivize an attack on U.S. forces once Beijing has decided to act. The more credible the American threat to intervene, the more likely China would be to hit U.S. forces in the region in its opening salvo. But if China thought the United States might stay out of the conflict, it would decline to attack U.S. forces in the region, since doing so would inevitably bring the United States into the war. 

Wishful Thinking

What might dissuade Xi from pursuing armed unification, if not U.S. military might? Most Western analysts believe that Xi’s devotion to his signature plan to achieve the “Chinese dream” of “national rejuvenation,” which requires him to maintain economic growth and improve China’s international standing, will deter him from using military force and risking derailing his agenda. They argue that the economic costs of a military campaign against Taiwan would be too high, that China would be left completely isolated internationally, and that Chinese occupation of the island would tie up Beijing for decades to come. 

But these arguments about the cost of armed unification are based more on American projections and wishful thinking than on fact. A protracted, high-intensity conflict would indeed be costly for China, but Chinese war planners have set out to avoid this scenario; China is unlikely to attack Taiwan unless it is confident that it can achieve a quick victory, ideally before the United States can even respond. 

Even if China found itself in a protracted war with the United States, however, Chinese leaders may believe they have social and economic advantages that would enable them to outlast the Americans. They see the Chinese people as more willing to make sacrifices for the cause of Taiwan than the American people. Some argue, too, that China’s large domestic market makes it less reliant on international trade than many other countries. (The more China economically decouples from the United States and the closer it gets to technological self-sufficiency, the greater this advantage will be.) Chinese leaders could also take comfort in their ability to quickly transition to an industrial wartime footing. The United States has no such ability to rapidly produce military equipment.

International isolation and coordinated punishment of Beijing might seem like a greater threat to Xi’s great Chinese experiment. Eight of China’s top ten trading partners are democracies, and nearly 60 percent of China’s exports go to the United States and its allies. If these countries responded to a Chinese assault on Taiwan by severing trade ties with China, the economic costs could threaten the developmental components of Xi’s rejuvenation plan.

But Chinese leaders have good reason to suspect that international isolation and opprobrium would be relatively mild. When China began to cultivate strategic partnerships in the mid-1990s, it required other countries and organizations, including the European Union, to sign long-term agreements to prioritize these relationships and proactively manage any tensions or disruptions. All these agreements mention trade, investment, economic cooperation, and working together in the United Nations. Most include provisions in support of Beijing’s position on Taiwan. (Since 1996, China has convinced more than a dozen countries to switch their diplomatic recognition to Beijing, leaving Taiwan with only 15 remaining allies.) In other words, many of China’s most important trading partners have already sent a strong signal that they will not let Taiwan derail their relationships with Beijing. 

Whether compelling airlines to take Taiwan off their maps or pressuring Paramount Pictures to remove the Taiwanese flag from the Top Gun hero Maverick’s jacket, China has largely succeeded in convincing many countries that Taiwan is an internal matter that they should stay out of. Australia has been cautious about expanding its military cooperation with the United States and reluctant even to consider joint contingency planning over Taiwan (although the tide seems to be shifting in Canberra). Opinion polls show that most Europeans value their economic ties with China and the United States roughly the same and don’t want to be caught in the middle. Southeast Asia feels similarly, with polls showing that the majority of policymakers and thought leaders from member states of the Association of Southeast Asian Nations believe the best approach to U.S.-Chinese sparring is for the association to “enhance its own resilience and unity to fend off their pressures.” One South Korean official put it more memorably in an interview with The Atlantic, comparing the need to pick sides in the U.S.-Chinese dispute to “asking a child whether you like your dad or your mom.” Such attitudes suggest that the United States would struggle to convince its allies to isolate China. And if the international reaction to Beijing’s crackdowns in Hong Kong and Xinjiang is any indication, the most China can expect after an invasion of Taiwan are some symbolic sanctions and words of criticism. 

The risk that a bloody insurgency in Taiwan will drag on for years and drain Beijing of resources is no more of a deterrent—and the idea that it would be says more about the United States’ scars from Afghanistan and Iraq than about likely scenarios for Taiwan. The PLA’s military textbooks assume the need for a significant campaign to consolidate power after its troops have landed and broken through Taiwan’s coastal defenses, but they do not express much concern about it. This may be because although the PLA has not fought a war since 1979, China has ample experience with internal repression and dedicates more resources to that mission than to its military. The People’s Armed Police boasts at least 1.5 million members, whose primary mission is suppressing opposition. Compared with the military task of invading and seizing Taiwan in the first place, occupying it probably looks like a piece of cake.

For all these reasons, Xi may believe he can regain control of Taiwan without jeopardizing his Chinese dream. It is telling that in the flood of commentary on Taiwan that has come out of China in recent months, few articles have mentioned the costs of war or the potential reaction from the international community. As one retired high-level military officer explained to me recently, China’s main concern isn’t the costs; it’s sovereignty. Chinese leaders will always fight for what is theirs. And if China defeats the United States along the way, it will become the new dominant power in the Asia-Pacific. The prospects are tantalizing. The worst-case scenario, moreover, is that the United States reacts more quickly and effectively than expected, forcing China to declare victory after limited gains and go home. Beijing would live to capture Taiwan another day. 

No Exit

These realities make it very difficult for the United States to alter China’s calculus on Taiwan. Richard Haass and David Sacks of the Council on Foreign Relations have argued in Foreign Affairs that the United States could improve cross-strait deterrence by ending its long-standing policy of “strategic ambiguity”—that is, declining to state specifically whether and how it would come to Taiwan’s defense. But the main problem is not U.S. resolve, since Chinese leaders already assume the United States will intervene. What matters to Xi and other top Chinese leaders is whether they think the PLA can prevail even in the face of U.S. intervention. For that reason, successful deterrence requires convincing China that the United States can prevent it from achieving its military objectives in Taiwan, a difficult undertaking that would come with its own downsides and potential risks. 

One way to convince Beijing would be to develop the capabilities to physically stop it from taking Taiwan—deterrence by denial. This would involve positioning missile launchers and armed drones near Taiwan and more long-range munitions, especially antiship weapons, in places such as Guam, Japan, and the Philippines. These weapons would help repel a Chinese amphibious and air assault in the initial stages of an attack. If Chinese leaders knew their forces could not physically make it across the strait, they would not consider trying unless Taiwan took the truly unacceptable step of declaring independence. 

The United States would also need to invest heavily in intelligence, surveillance, and reconnaissance in the region. The attractiveness of a full-on invasion from China’s perspective lies in the possibility of surprise: the United States may not be able to respond militarily until after Beijing has taken control of the island and the war is over. Leaving aside the operational challenges of such a response, it would be politically difficult for any U.S. president to authorize an attack on China when no shots were being fired at the time. 

An enhanced U.S. military and intelligence presence in the Indo-Pacific would be sufficient to deter most forms of armed unification, but it wouldn’t prevent China from using force altogether. Beijing could still try to use missile strikes to convince Taiwan to bend to its will. To deter all Chinese military aggression, the United States would therefore need to be prepared to destroy China’s missile batteries—which would involve U.S. strikes on the Chinese mainland. Even if U.S. intelligence capabilities improve, the United States would risk mistaking Chinese military exercises for preparations for an invasion—and igniting a war by mistake. China knows this and may conclude the United States would not take the chance. 

The most effective way to deter Chinese leaders from attacking Taiwan is also the most difficult: to convince them that armed unification would cost China its rejuvenation. And the United States cannot do this alone. Washington would need to persuade a large coalition of allies to commit to a coordinated economic, political, and military response to any Chinese aggression. And that, unfortunately, remains a remote possibility, since many countries are unwilling to risk their economic prospects, let alone a major-power war, in order to defend a small democratic island. 

Ultimately, then, there is no quick and easy fix to the escalating tensions across the strait. The only way the United States can ensure Taiwan’s security is to make an invasion impossible for Beijing or to convince Chinese leaders that using force will cause them to be pariahs. For the last 25 years, however, Beijing has sought to prevent Washington from doing either. Unfortunately for Taiwan, only now is the United States waking up to the new reality.

This interview by Bruce Goldman was originally published by the Stanford School of Medicine.

On May 13, the journal Science published a letter, signed by 18 scientists, stating that it was still unclear whether the virus that causes COVID-19 emerged naturally or was the result of a laboratory accident, but that neither cause could be ruled out. David Relman, MD, the Thomas C. and Joan M. Merigan Professor and professor of microbiology and immunology, spearheaded the effort.

Relman is no stranger to complicated microbial threat scenarios and illness of unclear origin. He has advised the U.S. government on emerging infectious diseases and potential biological threats. He served as vice chair of a National Academy of Sciences committee reviewing the FBI investigation of letters containing anthrax that were sent in 2001. Recently, he chaired another academy committee that assessed a cluster of poorly explained illnesses in U.S. embassy employees. He is a past president of the Infectious Diseases Society of America.

Stanford Medicine science writer Bruce Goldman asked Relman to explain what remains unknown about the coronavirus’s emergence, what we may learn and what’s at stake.

1. How might SARS-CoV-2, which causes COVID-19, have first infected humans?

Relman: We know very little about its origins. The virus’s closest known relatives were discovered in bats in Yunnan Province, China, yet the first known cases of COVID-19 were detected in Wuhan, about 1,000 miles away.

There are two general scenarios by which this virus could have made the jump to humans. First, the jump, or “spillover,” might have happened directly from an animal to a human, by means of an encounter that took place within, say, a bat-inhabited cave or mine, or closer to human dwellings — say, at an animal market. Or it could have happened indirectly, through a human encounter with some other animal to which the primary host, presumably a bat, had transmitted the virus.

Bats and other potential SARS-CoV-2 hosts are known to be shipped across China, including to Wuhan. But if there were any infected animals near or in Wuhan, they haven’t been publicly identified.

Maybe someone became infected after contact with an infected animal in or near Yunnan, and moved on to Wuhan. But then, because of the high transmissibility of this virus, you’d have expected to see other infected people at or near the site of this initial encounter, whether through similar animal exposure or because of transmission from this person.

2. What’s the other scenario?

Relman: SARS-CoV-2 could have spent some time in a laboratory before encountering humans. We know that some of the largest collections of bat coronaviruses in the world — and a vigorous research program involving the creation of “chimeric” bat coronaviruses by integrating unfamiliar coronavirus genomic sequences into other, known coronaviruses — are located in downtown Wuhan. And we know that laboratory accidents happen everywhere there are laboratories.

All scientists need to acknowledge a simple fact: Humans are fallible, and laboratory accidents happen — far more often than we care to admit. Several years ago, an investigative reporter uncovered evidence of hundreds of lab accidents across the United States involving dangerous, disease-causing microbes in academic institutions and government centers of excellence alike — including the Centers for Disease Control and Prevention and the National Institutes of Health.

SARS-CoV-2 might have been lurking in a sample collected from a bat or other infected animal, brought to a laboratory, perhaps stored in a freezer, then propagated in the laboratory as part of an effort to resurrect and study bat-associated viruses. The materials might have been discarded as a failed experiment. Or SARS-CoV-2 could have been created through commonly used laboratory techniques to study novel viruses, starting with closely related coronaviruses that have not yet been revealed to the public. Either way, SARS-CoV-2 could have easily infected an unsuspecting lab worker and then caused a mild or asymptomatic infection that was carried out of the laboratory.

3. Why is it important to understand SARS-CoV-2’s origins?

Relman: Some argue that we would be best served by focusing on countering the dire impacts of the pandemic and not diverting resources to ascertaining its origins. I agree that addressing the pandemic’s calamitous effects deserves high priority. But it’s possible and important for us to pursue both. Greater clarity about the origins will help guide efforts to prevent a next pandemic. Such prevention efforts would look very different depending on which of these scenarios proves to be the most likely.

Evidence favoring a natural spillover should prompt a wide variety of measures to minimize human contact with high-risk animal hosts. Evidence favoring a laboratory spillover should prompt intensified review and oversight of high-risk laboratory work and should strengthen efforts to improve laboratory safety. Both kinds of risk-mitigation efforts will be resource intensive, so it’s worth knowing which scenario is most likely.

4. What attempts at investigating SARS-CoV-2’s origin have been made so far, with what outcomes?

Relman: There’s a glaring paucity of data. The SARS-CoV-2 genome sequence, and those of a handful of not-so-closely-related bat coronaviruses, have been analyzed ad nauseam. But the near ancestors of SARS-CoV-2 remain missing in action. Absent that knowledge, it’s impossible to discern the origins of this virus from its genome sequence alone. SARS-CoV-2 hasn’t been reliably detected anywhere prior to the first reported cases of disease in humans in Wuhan at the end of 2019. The whole enterprise has been made even more difficult by the Chinese national authorities’ efforts to control and limit the release of public health records and data pertaining to laboratory research on coronaviruses.

In mid-2020, the World Health Organization organized an investigation into the origins of COVID-19, resulting in a fact-finding trip to Wuhan in January 2021. But the terms of reference laying out the purposes and structure of the visit made no mention of a possible laboratory-based scenario. Each investigating team member had to be individually approved by the Chinese government. And much of the data the investigators got to see was selected prior to the visit and aggregated and presented to the team by their hosts.

The recently released final report from the WHO concluded — despite the absence of dispositive evidence for either scenario — that a natural origin was “likely to very likely” and a laboratory accident “extremely unlikely.” The report dedicated only 4 of its 313 pages to the possibility of a laboratory scenario, much of it under a header entitled “conspiracy theories.” Multiple statements by one of the investigators lambasted any discussion of a laboratory origin as the work of dark conspiracy theorists. (Notably, that investigator — the only American selected to be on the team — has a pronounced conflict of interest.)

Given all this, it’s tough to give this WHO report much credibility. Its lack of objectivity and its failure to follow basic principles of scientific investigation are troubling. Fortunately, WHO’s director-general recognizes some of the shortcomings of the WHO effort and has called for a more robust investigation, as have the governments of the United States, 13 other countries and the European Union.

5. What’s key to an effective investigation of the virus’s origins?

Relman: A credible investigation should address all plausible scenarios in a deliberate manner, involve a wide variety of expertise and disciplines and follow the evidence. In order to critically evaluate other scientists’ conclusions, we must demand their original primary data and the exact methods they used — regardless of how we feel about the topic or about those whose conclusions we seek to assess. Prior assumptions or beliefs, in the absence of supporting evidence, must be set aside.

Investigators should not have any significant conflicts of interest in the outcome of the investigation, such as standing to gain or lose anything of value should the evidence point to any particular scenario.

There are myriad possible sources of valuable data and information, some of them still preserved and protected, that could make greater clarity about the origins feasible. For all of these forms of data and information, one needs proof of place and time of origin, and proof of provenance.

To understand the place and time of the first human cases, we need original records from clinical care facilities and public health institutions as well as archived clinical laboratory data and leftover clinical samples on which new analyses can be performed. One might expect to find samples of wildlife, records of animal die-offs and supply-chain documents.

Efforts to explore possible laboratory origins will require that all laboratories known to be working on coronaviruses, or collecting relevant animal or clinical samples, provide original records of experimental work, internal communications, all forms of data — especially all genetic-sequence data — and all viruses, both natural and recombinant. One might expect to find archived sequence databases and laboratory records.

Needless to say, the politicized nature of the origins issue will make a proper investigation very difficult to pull off. But this doesn’t mean that we shouldn’t try our best. Scientists are inquisitive, capable, clever, determined when motivated, and inclined to share their insights and findings. This should not be a finger-pointing exercise, nor an indictment of one country or an abdication of the important mission to discover biological threats in nature before they cause harm. Scientists are also committed to the pursuit of truth and knowledge. If we have the will, we can and will learn much more about where and how this pandemic arose.