For three years, the rhythm of Washington’s China conversation has been set by a single image. An armada crossing the Strait. Missiles arcing toward Taipei. Marines clawing onto a Pacific beach in some violent inversion of Iwo Jima. The Davidson Window. The 2027 deadline. The CSIS wargames in which both sides bleed for a month and the world economy seizes up overnight.

It is a vivid picture. It is also, almost certainly, the wrong one.

Xi Jinping has watched what happened to Vladimir Putin. He has watched what happens when an autocrat who has not lost a fight in twenty years stakes his legacy on a 72-hour war that becomes a four-year quagmire. He has read the same casualty estimates American admirals have. He knows the geography of the Strait better than any Pentagon analyst, because his predecessors have been studying it since 1949.

And he has, in front of him, a far more attractive option than invasion. A path that avoids the catastrophic risks of an amphibious assault, exploits weaknesses already open in Taiwan’s politics, and delivers the strategic outcome Beijing wants without firing the war’s opening shot.

That path is coercion. Blockade in slow motion. Customs enforcement at gunpoint. A quarantine dressed in legal language. The strangulation of an island whose own legislature is helping tighten the rope.

The question that has framed every China-Taiwan conversation for half a decade — will Xi invade? — is the wrong question. The right one is: what does Xi need to do to make Taiwan accept a settlement without invading? Almost everything Beijing has done since 2022 answers that question. The world has not yet noticed.

The Bottom Line

For readers who want the argument before the evidence:

A full-scale amphibious invasion is the least likely path to reunification. It is also the most expensive, most uncertain, and most career-ending option Xi has on the table.

A coercion-blockade campaign is the most likely path. It has already begun, in slow motion, through naval drills, undersea cable cuts, and customs harassment that would be acts of war if anyone in Washington was prepared to call them that.

Taiwan’s biggest vulnerability isn’t its beaches. It’s its parliament. The KMT-controlled Legislative Yuan has spent two years gutting the defense budget. No invasion is needed if the island can’t fund its own defense.

The peaking-power thesis — that a declining China will lash out before its window closes — is real but misread. Xi’s closing window doesn’t push him toward war. It pushes him toward a forced settlement before 2035.

Most probable timeline: gray-zone escalation through 2026, a major blockade or quarantine event between 2027 and 2030, political endgame by the mid-2030s.

If you want the reasoning, keep going.

The Invasion Fantasy

Amphibious assault is the hardest operation a modern military can attempt. The Allies needed two years of preparation, total air superiority, and a continent-sized staging ground to pull off Normandy across 33 kilometres of channel. Taiwan sits 130 kilometres from the Chinese coast, ringed by mountains on its eastern shore and by dense urban sprawl on its western one. There are roughly fourteen beaches across the entire island where a force of any size could land. Taiwan has known this since the day Chiang Kai-shek arrived, and has spent seven decades fortifying accordingly.

The PLA has not fought a real war since its 1979 disaster against Vietnam, in which a much larger Chinese force was bloodied by Vietnamese militia. Its officer corps, top to bottom, has zero combat experience. Its joint command structure was rebuilt only in 2016 and has never been tested under fire. Even the most generous estimates of Chinese amphibious lift capacity put the first wave at around 25,000 troops. They would land against a defending force of 169,000 active personnel and a notional 2.3 million reservists, fighting on home terrain, with American precision munitions in inventory and seventy years of fortification under their feet.

This is before the typhoon season. Before the question of whether Chinese pilots can actually fly the missions their war plans demand. Before the matter of how a force that has never moved at scale across open water keeps a beachhead supplied for the months an island campaign would take.

Xi Jinping is many things. Stupid is not one of them. He has watched what a far smaller, far better-prepared Russian military did to itself in Ukraine. He has seen what happens when authoritarian intelligence services tell their leader what he wants to hear, and the leader believes them. The lesson Beijing took from Kyiv was not “invade faster.” It was “don’t invade at all if there’s another way.”

There is another way. The remainder of this piece is about what that way looks like, why it works, and why almost no one in the West is preparing for it.

The Peaking Power Problem

Hal Brands and Michael Beckley have built the most influential argument for the opposite view. Their thesis, sketched in Danger Zone and refined across a half-decade of essays, runs as follows. China is a peaking power. Peaking powers, historically, strike before they decline. Demographic collapse plus economic stagnation plus encirclement by American alliances gives Xi a closing window in which the costs of military action only grow. Therefore, expect war.

The thesis deserves engagement, because parts of it are correct.

China is peaking. The working-age population shrank by five million last year. The property sector is a slow-motion balance-sheet crisis that has wiped out an estimated $18 trillion in household wealth. Local government debt has crossed 100% of GDP on most credible estimates. By 2035, China will have more pensioners than workers. By 2049 — the centenary deadline Xi has set for national rejuvenation — the demographic math will have eaten whatever GDP runway remains.

But the conclusion does not follow from the premises. Peaking powers don’t always lash out. Sometimes they consolidate. Imperial Britain peaked around 1900 and managed forty more years of strategic relevance through alliance-building and managed retreat. The Soviet Union peaked in the early 1970s and chose stagnation over war for two more decades, until its system collapsed under its own weight without firing a shot at NATO.

The peaking-power logic, applied carefully to China, doesn’t push Xi toward an invasion he is likely to lose. It pushes him toward a settlement he is likely to win — before American semiconductor controls bite deeper, before demographic decline becomes irreversible, before Taiwan’s own asymmetric deterrent matures. The window is closing. Xi needs to act inside it. But the action that fits both his constraints and his ambitions is coercion, not conquest.

The Brands-Beckley framework gets the urgency right and the verb wrong.

What Operation Sindoor Did and Did Not Show

In the spring of 2025, the brief India-Pakistan flare-up known as Operation Sindoor became instant content for the geopolitics commentariat. Chinese-supplied PL-15 missiles and J-10C fighters, fielded by the Pakistan Air Force, performed poorly against Indian Rafales operating with Israeli electronic-warfare suites and Russian S-400 air defenses. The takeaway, repeated for weeks across Twitter and Substack, was that Chinese weapons don’t work.

This is the wrong lesson, and leaning on it is dangerous.

Pakistan operates Chinese export variants. Export weapons are downgraded — different software, different seekers, different electronic warfare suites. The PLA itself flies a generation of fighters and fires missile variants that have never been exported to anyone. Inferring PLA capability from PAF performance is like judging the United States Air Force by what F-16s built for Egyptian export can do against a near-peer adversary. It is a category error.

What Sindoor did expose is more useful. Doctrinal gaps in how Chinese-trained air forces fight integrated battle. Integration problems between Chinese-built platforms and non-Chinese kill chains. An ecosystem-wide weakness in real-time targeting against an adversary who could see them as well as they could see him.

None of this is nothing. It suggests the PLA’s untested combined-arms doctrine has the same disease the Russian military had on the eve of Ukraine. It looks impressive on a parade ground in Beijing. It has not been tested against an enemy that shoots back.

Xi is not blind to this. The honest reading of Sindoor in Beijing is not “our weapons failed.” It is “our doctrine is fragile, and we should not bet the regime on a war we have not fought.” That reading pushes Xi away from kinetic options. It does not push him away from ambition.

The Real War Is a Blockade

This is where serious China analysts have been pointing for two years, and where the public conversation has not yet caught up. Lonnie Henley, the former senior intelligence officer who spent his career on Chinese military affairs, has been the loudest voice. Bonny Lin at CSIS. Oriana Skylar Mastro at Stanford. Their work, read together, sketches a scenario the Pentagon takes seriously and the public has barely noticed.

The PLA Navy does not need to land troops on Taiwan. It needs to convince the world that landing is unnecessary.

The shape of a coercion campaign runs roughly like this. China announces, unilaterally, a customs inspection regime covering all maritime traffic to Taiwan. Beijing claims the right to board vessels, inspect cargo, and turn back ships carrying military-related material. The legal architecture is borrowed from China’s existing position that Taiwan is a domestic matter — under Chinese law the regime is not a blockade but a customs operation.

Simultaneously, a no-fly zone is enforced through long-range surface-to-air missiles based on Pingtan Island, supplemented by carrier-based aviation operating east of Taiwan. Civilian aviation insurance collapses within hours. Submarine communication cables — which Taiwan depends on for almost all its bandwidth and which have been cut repeatedly in the last two years under suspicious circumstances — are severed. Energy imports are choked. Taiwan has roughly eleven days of natural gas reserves on the island. Eleven.

Insurance markets respond before any shot is fired. Lloyd’s pulls coverage on Taiwan-bound vessels. Container traffic into Kaohsiung collapses by 70% in a week. Semiconductor supply chains seize up globally. The Taiwanese stock market loses a third of its value in three days.

Taiwan now has a choice. Fight a war it cannot win alone. Or negotiate.

The United States has its own choice, and it is harder than the public conversation has admitted. Break the blockade with American warships, committing to combat in a scenario where China has not technically invaded anything? Or accept a fait accompli and trust that some negotiated arrangement will hold?

Every American president since Truman has faced this question only as a hypothetical. The next one may have to answer it for real. There is no good answer. That is precisely the point of the strategy.

Taiwan’s Real Problem Is in Taipei, Not Beijing

The most underreported strategic story of the last two years is that Taiwan’s own legislature has been actively dismantling its deterrent.

The KMT-TPP opposition coalition, which has controlled the Legislative Yuan since the January 2024 elections, has cut, frozen, or restricted significant portions of the defense budget submitted by President Lai Ching-te’s DPP government. The cuts have hit the exact porcupine strategy capabilities the Pentagon has been begging Taipei to invest in. Submarine programs slowed. Drone procurement gutted. Asymmetric warfare investments restricted or delayed.

Public polling tells the rest of the story. Among Taiwanese under 30, surveys consistently show under half saying they would fight if China invaded. A generation that has known only de facto independence does not feel a visceral commitment to defending the formal sovereignty their grandparents fled the mainland to preserve.

Beijing reads all of this. Carefully. Because it solves a problem.

Why invade an island when its own parliament is doing your work for you? Why pay the price of an amphibious assault when the political will to resist one is being hollowed out by Taiwan’s own internal divisions? Why force a confrontation now, when waiting six more years yields a Taiwan that is poorer in deterrent, more politically fragmented, and less psychologically prepared?

This is what coercion looks like in the long game. Don’t break the will to resist with bombs. Hollow it out with patience.

The America Question

The wildcard in every Taiwan scenario is not Chinese capability. It is American resolve.

The second Trump administration has made one thing unmistakably clear. Alliances are transactions. Taiwan has been told publicly that it isn’t paying enough for its own defense, that its semiconductor industry is stolen American capacity, that tariffs will be deployed against allies as readily as adversaries. The strategic ambiguity that has held the Strait quiet for fifty years now has a new and unstable variable.

Beijing reads American politics with more sophistication than American commentators credit. The window between now and 2028 — when American attention is divided, when allied trust in Washington is at its lowest point since Suez, when the dollar’s weaponization has accelerated yuan-clearing networks across the Global South — is the most permissive geopolitical environment China has had since the Korean armistice.

If a coercion campaign is going to begin in earnest, this is when it begins. The structural conditions will not get better for Beijing. Trump is not forever. The next administration, of either party, will likely re-anchor Pacific deterrence with the urgency of a country that has just watched its credibility tested. Xi has perhaps a thirty-six-month window in which the American president is the friendliest he is going to get.

Coincidentally, this is also the window in which Xi will be 73 to 75 years old, in the back half of his rule, with closing economic options at home and a legacy still unwritten on the Taiwan question.

The Prediction

2025 to 2026. Sustained gray-zone escalation. Median-line crossings normalized to the point that Western media stops covering them. Cable cuts treated as accidents with rising frequency. Joint Sword drills become annual rehearsals for a quarantine scenario, each iteration more ambitious. The KMT continues to constrain Taiwan’s defense budget. Public will erodes further. Beijing builds the legal, diplomatic, and operational architecture for a quarantine without yet imposing one.

2027 to 2030. The most dangerous window. The probability of a major coercive event — a declared customs enforcement regime, a multi-week PLA Navy quarantine of a Taiwanese port, a simulated decapitation strike that becomes real — is, on this analysis, above 50%. The probability of a full kinetic invasion remains below 15%. The trigger is most likely a manufactured crisis around a Taiwanese election the DPP loses badly, or a perceived provocation that gives Beijing political cover to move.

2030 to 2035. The endgame. Either Taiwan has accepted some form of interim arrangement under sustained coercive pressure — a neutralized status, restrictions on weapons procurement, a cross-strait economic framework that locks the island into Chinese supply chains — or the relationship has hardened into a permanent armed standoff that quietly normalizes Taiwanese de facto independence under a Chinese veto. The first outcome is more likely than the second.

The war for Taiwan, in other words, has already started. It just doesn’t look like a war.

What This Analysis Doesn’t Know

Honest analysis names its uncertainty.

Xi’s health. He is 72. A succession crisis in Beijing rewrites every timeline. A weakened Xi may lash out to consolidate his legacy. A confident successor with a longer horizon may calm the entire Taiwan question for a decade.

A Taiwanese black swan. A political crisis. An assassination attempt. A PLA aircraft shot down by Taiwanese forces. Any of these can collapse a coercion campaign into a kinetic one in seventy-two hours.

The post-Trump American election. A 2028 administration that re-anchors Pacific deterrence changes Beijing’s calculus before the most dangerous window fully opens.

An economic accident. A Chinese banking crisis severe enough to consume regime attention. A Korean Peninsula flare-up that pulls Chinese attention north.

And finally — every major war in the last century surprised the analysts who studied it most closely. Few predicted the speed of the Ukraine invasion. Almost none predicted October 7. There is no reason to think this analysis is exempt from the same humility.

The probability distribution sketched above is held with perhaps 60% confidence overall. The mode of that distribution — coercion not invasion — is held with closer to 80%. Anyone who claims more is selling something.

The Wrong Question, the Right Answer

For half a decade, the question framing every China-Taiwan conversation has been: will Xi invade? The wargames are built around it. The Pentagon’s procurement priorities are anchored to it. The public commentary cycles back to it every time PLA aircraft cross the median line.

It is the wrong question.

The right one is the one Xi is actually asking. What does it take to make Taiwan accept a settlement without invading? How do I close the deal my predecessors couldn’t close, on terms my successors will inherit, without betting my regime on a beach landing the world’s best amphibious militaries would hesitate to attempt?

Almost everything Beijing has done since 2022 answers that question. The gray-zone pressure. The cable cuts. The trade restrictions. The targeted economic punishment of Taiwanese politicians who defy Beijing. The patient erosion of the island’s political will, year after year, while the West watches the wrong horizon.

The war for Taiwan will not be a Pacific D-Day. It will be a slow strangulation, dressed in the language of customs enforcement, fought in shipping lanes and undersea cables and parliamentary committees in Taipei. By the time the world recognises it as a war, the war will be over.

“That is the war China is fighting. The West isn’t preparing for it. And on current trajectories, China is going to win it.”