Phase I
Qing frontier and Japanese colonisation
Taiwan entered the Chinese imperial system late and left it early.
Phase I
Taiwan entered the Chinese imperial system late and left it early.
Phase II
Japan's defeat transferred Taiwan to Republic of China administration in 1945 — an arrangement the ROC treated as restored sovereignty and which later peace treaties left conspicuously unresolved.
Phase III
The Korean War transformed Taiwan's position: the US Seventh Fleet entered the strait in 1950 and Taiwan became a Cold War security commitment, formalised in the 1954 Mutual Defense Treaty.
Phase IV
American recognition of Beijing in January 1979 could have ended Taiwan's security position; instead Congress rebuilt it in a different form.
Phase V
Martial law ended in July 1987 after 38 years, and Taiwan democratised rapidly: opposition parties legalised, parliament fully re-elected, and in March 1996 the first direct presidential election.
Phase VI
The two decades after the missile crisis ran in opposite directions at once.
Phase VII
Under Xi Jinping the PLA underwent its deepest reorganisation since the 1950s, creating theatre commands — the Eastern Theater Command holding the Taiwan direction — and building out naval, air, missile and information capabilities at a pace that changed regional planning assumptions.
Phase VIII
Tsai Ing-wen's re-election in January 2020 confirmed the political trend, and PLA activity in Taiwan's air defence identification zone grew into a routine feature rather than an episodic signal.
Phase IX
Lai Ching-te won the January 2024 presidential election, giving the DPP an unprecedented third consecutive term — but simultaneously losing the legislative majority to a Kuomintang and Taiwan People's Party bloc, producing a divided government in which defence budgets became contested domestic politics.
Phase X
The current phase is defined less by escalation than by uncertainty on the deterrence side of the ledger.
Taiwan entered the Chinese imperial system late and left it early. The Qing annexed the island in 1683 and governed it for two centuries as a peripheral frontier, only making it a full province in 1885 — a decade before ceding it to Japan in perpetuity under the Treaty of Shimonoseki after the First Sino-Japanese War. Fifty years of Japanese colonial rule followed: infrastructure, public health, industrial development and education alongside political subordination and, in the war years, forced assimilation. That half-century is the reason Taiwan's twentieth century diverged from the mainland's before 1949 — and a reason Taiwanese identity has never mapped neatly onto Chinese nationalism.
Taiwan's separation from mainland governance began in 1895, not 1949 — a fact both claimants' narratives tend to compress, and one that underlies the identity question at the centre of the dispute.
Japan's defeat transferred Taiwan to Republic of China administration in 1945 — an arrangement the ROC treated as restored sovereignty and which later peace treaties left conspicuously unresolved. Relations between the incoming Nationalist administration and the local population deteriorated quickly, culminating in the 28 February 1947 uprising and a violent crackdown that killed thousands and became the founding trauma of modern Taiwanese politics. As the Chinese Civil War turned, Chiang Kai-shek's government withdrew to Taiwan in 1949 with its army, its gold reserves and its claim to be the government of all China — while the Communist victory established the People's Republic in Beijing.
The dispute was born unresolved and stayed that way: an unfinished civil war, not an invasion of one state by another. Every legal argument on all sides still traces back to that unfinished quality.
The Korean War transformed Taiwan's position: the US Seventh Fleet entered the strait in 1950 and Taiwan became a Cold War security commitment, formalised in the 1954 Mutual Defense Treaty. Two crises followed over the offshore islands — the First Taiwan Strait Crisis of 1954–55 and the intense artillery confrontation over Kinmen in 1958 — both of which stopped short of general war and both of which established the pattern of testing at the edges rather than at the centre. Politically the position eroded: the UN seat passed to Beijing in 1971, and Washington's recognition followed the PRC in 1979, ending the defence treaty.
The offshore-island crises set the template still in use: pressure applied to the periphery, calibrated to test commitment without triggering war. Kinmen and Matsu remain the places where that testing happens.
American recognition of Beijing in January 1979 could have ended Taiwan's security position; instead Congress rebuilt it in a different form. The Taiwan Relations Act of April 1979 committed the United States to provide Taiwan with arms of a defensive character and to regard any non-peaceful resolution as a matter of grave concern — without committing to defend it. Combined with the Three Communiqués and later the Six Assurances, this produced the arrangement that has governed ever since: unofficial relations, continued arms sales, and a deliberate refusal to say what the US would do. Taiwan meanwhile remained under martial law, which had run since 1949.
Ambiguity was designed to deter Beijing from attacking and Taipei from declaring independence — restraining both. Its weakness is that it depends entirely on being believed, which makes credibility itself a variable that can be spent.
Martial law ended in July 1987 after 38 years, and Taiwan democratised rapidly: opposition parties legalised, parliament fully re-elected, and in March 1996 the first direct presidential election. Democratisation changed the dispute's nature — Taiwan's future became something its population voted on, and a distinctly Taiwanese identity grew steadily at the expense of an exclusively Chinese one. Beijing responded to the 1995–96 election period with missile tests and exercises in the strait; the United States deployed two carrier battle groups. The crisis passed, and left two lessons: Beijing learned the limits of its 1990s military, and drew from it the modernisation programme that followed.
1996 is the hinge: democratisation made the status question popular rather than dynastic, and the crisis convinced Beijing that coercing Taiwan required a military it did not yet have. Both consequences define the present.
The two decades after the missile crisis ran in opposite directions at once. Economically the strait integrated deeply: Taiwanese manufacturers moved production to the mainland at scale, cross-strait trade and investment grew into the hundreds of billions, and the 2010 ECFA framework formalised the relationship. Politically the gap widened: the DPP won the presidency in 2000, ending half a century of Nationalist rule, and identity surveys showed a steady, one-directional drift toward exclusively Taiwanese self-identification. Beijing's 2005 Anti-Secession Law set out a declared legal basis for non-peaceful means. Interdependence, the theory went, would produce convergence; it did not.
This period is the strongest available evidence that economic integration does not produce political convergence — the assumption that underwrote a generation of engagement policy, tested and falsified.
Under Xi Jinping the PLA underwent its deepest reorganisation since the 1950s, creating theatre commands — the Eastern Theater Command holding the Taiwan direction — and building out naval, air, missile and information capabilities at a pace that changed regional planning assumptions. Politically, Beijing hardened: pressure on Taiwan's remaining diplomatic partners, exclusion from international bodies, and after the DPP's 2016 return to the presidency, the suspension of official cross-strait dialogue. The 2014 Sunflower Movement had already shown the limits of Taiwanese public appetite for closer integration, and Hong Kong's 2019–20 crackdown removed "one country, two systems" as a credible offer.
Beijing built the capability to coerce and simultaneously destroyed its most attractive political offer. The result is a strategy that leans increasingly on pressure because persuasion has stopped working.
Tsai Ing-wen's re-election in January 2020 confirmed the political trend, and PLA activity in Taiwan's air defence identification zone grew into a routine feature rather than an episodic signal. The August 2022 visit of US House Speaker Nancy Pelosi triggered the largest exercises in decades — live-fire areas encircling the island, missiles overflying it, and, critically, the effective abandonment of the strait median line as a restraint. What had been an unwritten rule since the 1950s became a line crossed routinely. Further exercises followed political events in 2023 and 2024. The significance was less the drills than the ratchet: each round established a new normal that was not given back.
2022 converted exceptional signalling into baseline activity. That ratchet — new normals that are never rolled back — is the core mechanism of grey-zone pressure and the reason "no change" reporting can still mean deterioration.
Lai Ching-te won the January 2024 presidential election, giving the DPP an unprecedented third consecutive term — but simultaneously losing the legislative majority to a Kuomintang and Taiwan People's Party bloc, producing a divided government in which defence budgets became contested domestic politics. Beijing, which regards Lai with particular hostility, responded to his May inauguration with major exercises and roughly doubled detected air activity around the island. A separate instrument became prominent in February 2024: China Coast Guard patrols in waters around Kinmen, asserting law-enforcement jurisdiction where Taiwan administers. Undersea cable damage incidents near Matsu and elsewhere added a civil-infrastructure dimension.
Pressure diversified rather than simply intensified: military, law-enforcement, informational and infrastructure instruments applied together, each individually below any threshold that would compel a response.
The current phase is defined less by escalation than by uncertainty on the deterrence side of the ledger. Detected PLA air activity has run closer to its pre-2024 baseline since early 2026 — read by analysts as a recalibration of coercion rather than a relaxation of aims — while coast guard presence and information pressure continued — and widened geographically, with Beijing publicly describing patrols east of Taiwan as routine law enforcement in July 2026. Taiwan's special defence budget passed in May 2026 at NT$780bn, well below the NT$1.25tn the executive sought, after months of gridlock in the opposition-held legislature, which removed roughly NT$470bn of domestic uncrewed-systems funding while leaving foreign procurement intact. In Washington the pattern ran both ways: the largest package ever notified for Taiwan was entered in December 2025 and the delivery backlog fell through the first half of 2026, while a further $14bn package was left unapproved and publicly framed as leverage in dealings with Beijing. Japan, having signalled possible involvement in a contingency, absorbed sustained Chinese economic pressure of its own, while US access arrangements in northern Luzon deepened. Deterrence in this phase is being tested through budgets, signals and credibility rather than through force.
The variable that moved most in 2026 was not Chinese capability but allied credibility and Taiwanese political consensus. Deterrence is a belief held by the other side — which makes budgets and signals part of the military balance, not a substitute for it.