Taiwan's Unfinished Civil War: How 75 Years of Ambiguity Created a Nuclear Flashpoint

The Republic of China does not control China. It has not done so since October 1, 1949, when Mao Zedong stood in Tiananmen Square and proclaimed the People's Republic. Yet the Republic of China still exists, squeezed onto an island 180 kilometres from the mainland, maintaining embassies in 12 countries, and fielding a military that lists 'retaking the mainland' in its constitutional mandate. The People's Republic, meanwhile, claims sovereignty over an island it has never governed, whose 23.5 million people carry passports Beijing does not issue, pay taxes to a treasury Beijing does not control, and vote in elections Beijing considers illegitimate. The United States insists it recognises Beijing's government as the sole legal authority over China while simultaneously arming Taiwan and pledging—perhaps—to defend it. This is not policy. It is collective delusion dressed as diplomacy.

For 75 years, this arrangement has preserved peace through deliberate ambiguity. But ambiguity is a wasting asset. Xi Jinping has made unification a legacy priority, the Democratic Progressive Party has won three consecutive presidential elections in Taiwan, and the United States has begun treating the island as a de facto ally. The result is the most dangerous geopolitical standoff on the planet—one that could draw the world's two largest economies into a war neither wants but both are preparing for.

The Retreat That Never Ended

When Chiang Kai-shek's Kuomintang forces fled to Taiwan in 1949, they brought two million mainlanders, the national treasury, and the conceit that their exile was temporary. For decades, the KMT maintained that it was the legitimate government of all China, a position the United Nations accepted until 1971, when the General Assembly expelled Taipei and seated Beijing. The Republic of China lost its international legal standing but kept its island.

Taiwan's transformation from authoritarian garrison state to vibrant democracy—martial law ended in 1987, direct presidential elections began in 1996—was supposed to make the problem easier. Instead, it made it insoluble. As Taiwan democratised, its people developed a distinct identity. Polls by National Chengchi University show that 60.8% of Taiwanese now identify as exclusively Taiwanese, not Chinese—a figure that has doubled since 1992. Meanwhile, Beijing's position has hardened. The 2005 Anti-Secession Law authorises the use of force if Taiwan moves toward formal independence, and Xi has repeatedly stated that unification 'cannot be passed down from generation to generation.'

Xi's Escalation Timeline

Since Xi took power in 2012, Beijing has moved from rhetorical pressure to military encirclement. In 2016, the People's Liberation Army began circumnavigation exercises around Taiwan. By 2022, those exercises had become rehearsals for blockade. After then-Speaker Nancy Pelosi visited Taipei in August 2022, the PLA launched its largest-ever drills, firing ballistic missiles over the island and simulating a sea-and-air blockade that lasted five days. The frequency has only increased: Taiwan's Ministry of National Defence recorded 1,727 PLA aircraft incursions into its air defence identification zone in 2023, a 79% increase from 2022.

The PLA is not sabre-rattling. It is normalising the operational environment for invasion. Admiral John Aquilino, former head of U.S. Indo-Pacific Command, testified to Congress in March 2023 that China could be ready to invade by 2027, a timeline Xi has reportedly instructed the PLA to meet. Taiwan's own defence ministry assessed in its 2024 quadrennial review that a 'comprehensive blockade' is now Beijing's most likely opening move—strangling the island economically before firing a shot.

The Semiconductor Shield

Taiwan's best defence is not its 165,000 active-duty troops or its ageing fleet of F-16s. It is Taiwan Semiconductor Manufacturing Company. TSMC manufactures 92% of the world's most advanced chips—the sub-5-nanometre processors that power everything from iPhones to F-35 fighter jets. A Chinese invasion would not merely threaten 23.5 million people; it would sever the jugular of the global economy.

This gives Taiwan leverage—what former Taiwanese digital minister Audrey Tang has called 'the silicon shield.' But it is a double-edged weapon. If TSMC's fabs were destroyed in a conflict, the global shortage would make the 2021 chip crisis look trivial. The Biden administration understood this and passed the CHIPS Act in 2022, allocating $52 billion to rebuild domestic semiconductor manufacturing. TSMC is now constructing two fabs in Arizona. The first, scheduled for 2025, will produce 4-nanometre chips; the second, planned for 2027, will target 3-nanometre. But even at full capacity, they will supply only a fraction of global demand. Taiwan's silicon shield is thinning.

Washington's Useful Ambiguity

The United States has no formal defence treaty with Taiwan. It severed diplomatic relations in 1979 when it recognised the People's Republic under the One China Policy. But that same year, Congress passed the Taiwan Relations Act, committing the U.S. to provide Taiwan with 'defence articles and services in such quantity as may be necessary to enable Taiwan to maintain a sufficient self-defence capability.' The law does not say the U.S. will defend Taiwan. It does not say it will not.

This 'strategic ambiguity' was designed to deter both sides: Beijing from invading, Taipei from declaring independence. For decades it worked. But the Trump and Biden administrations have eroded it. In 2020, the U.S. sold Taiwan $5.1 billion in weapons, including Harpoon missile systems. In 2021, President Biden said—three times—that the U.S. would defend Taiwan if attacked, statements the White House walked back each time. In 2023, the U.S. approved a $345 million arms sale. Senior U.S. officials now visit Taipei openly. Members of Congress call for abandoning ambiguity entirely.

The problem is that clarity cuts both ways. If Washington explicitly commits to Taiwan's defence, it boxes itself into a war with a nuclear-armed peer. If it explicitly rules out intervention, it invites Beijing to act. Strategic ambiguity preserved peace because no side could be sure what the other would do. That uncertainty is evaporating.

The DPP's Dangerous Victory

In January 2024, Taiwan re-elected the Democratic Progressive Party for an unprecedented third term, choosing Lai Ching-te as president. Lai, a former physician and Tainan mayor, is more hawkish than his predecessor. He has called himself a 'pragmatic worker for Taiwan independence,' a phrase that sent tremors through Beijing. The KMT accused him of provoking China; Beijing's Taiwan Affairs Office called him a 'destroyer of peace.' Lai won with 40% of the vote in a three-way race, carrying younger voters who have never known a Taiwan deferential to Beijing.

The DPP's win reflected a generational shift. Taiwanese under 40, who came of age after democratisation, overwhelmingly reject unification. A 2023 survey by the Election Study Center found that only 1.3% of Taiwanese support unification 'as soon as possible,' while 5.1% support immediate independence. The plurality—28.6%—favour indefinite maintenance of the status quo. But 'status quo' is itself a fiction. Taiwan functions as an independent state in every way except name. It has its own currency, military, and passport. It elects its leaders in free elections. The only thing it lacks is international recognition—and Beijing's acquiescence.

What Is to Be Done

The current trajectory leads to war. Not inevitably, but probabilistically. Xi has staked his legitimacy on unification. Taiwan has staked its identity on self-determination. The United States has staked its credibility on preventing Chinese hegemony in Asia. These positions are irreconcilable.

The least bad option is to rebuild ambiguity—not as a permanent solution, but as a mechanism to buy time. That means Washington must stop treating Taiwan as a quasi-ally and return to the Taiwan Relations Act's careful language. It means Taipei must avoid rhetorical provocations that give Beijing a pretext without surrendering any substantive sovereignty. And it means Beijing must recognise that time is not necessarily on its side: the longer Taiwan remains free and prosperous, the more its people will resist unification, but military action would destroy the very thing Xi seeks to reclaim.

None of this will happen. Washington is locked in strategic competition with Beijing and sees Taiwan as leverage. Taipei's voters have repeatedly chosen leaders who reject Beijing's terms. And Xi has made unification a legacy goal, which means the 2027 deadline is not an intelligence estimate—it is a political commitment.

The alternative to managed ambiguity is not clarity. It is war—one that would kill tens of thousands, crater the global economy, and risk nuclear exchange between great powers. The tragedy is that all three governments understand this. They are choosing their positions anyway.

The Republic of China has not controlled the mainland for 75 years. The People's Republic has never controlled Taiwan. And the United States has built a policy on pretending both claims are somehow compatible. That pretence is collapsing. When it does, the world will discover whether deterrence can survive without ambiguity—or whether the unfinished civil war of 1949 will finally demand its conclusion.