February 26, 2026
On December 29th, 2025, China conducted the largest single-day military action ever recorded against Taiwan.
100 aircraft. 90 crossed the median line — the informal boundary dividing the Taiwan Strait that China had largely respected for decades. 13 warships. 14 coast guard ships. The following day, 27 rockets were fired from China's Fujian province. 10 landed inside Taiwan's 24-nautical-mile contiguous zone — the closest Chinese live-fire to Taiwan ever recorded.
No invasion. No war. Just a demonstration.
Then, four weeks later, Xi Jinping fired his most trusted general.
THE 2027 QUESTION.
In 2022, Xi Jinping gave the People's Liberation Army a directive: be capable of conducting a military operation to take Taiwan by 2027 — the PLA's centenary year. Former CIA Director William Burns confirmed it. The U.S. Department of Defense confirmed it in its December 2025 Annual Report to Congress.
What 2027 means is contested. It is a capability target, not a confirmed invasion order. Xi has publicly denied setting a Taiwan invasion deadline. 83 percent of China experts surveyed by CSIS reject the idea that China plans kinetic action by 2027 specifically.
But here's what is not contested: Xi set the clock. The PLA has been building toward it. And as of January 24th, 2026, the general responsible for telling Xi whether that clock was realistic is gone.
(U.S. Department of Defense Annual Report, December 23, 2025)
THE PURGE.
General Zhang Youxia was Xi's closest military ally. They had known each other since childhood — both "princelings," sons of former Communist Party leaders. Zhang was the second-highest ranking military official in China, Vice Chairman of the Central Military Commission. He was one of the last combat veterans in China's senior military ranks, having fought in border skirmishes against Vietnam in the 1970s and 80s.
On January 24th, 2026, he was placed under investigation for "serious disciplinary and legal violations." General Liu Zhenli, Chief of Staff of the CMC Joint Staff Department — the officer responsible for coordinating China's command, control, communications, and intelligence capabilities — was arrested the same day.
The Central Military Commission, which normally has seven members, now has two: Xi Jinping, and Zhang Shengmin — a career political commissar with virtually no operational command experience. Xi has purged five of the six uniformed members of the CMC since 2022.
(U.S.-China Economic and Security Review Commission, February 4, 2026)
This is the second time since the Cultural Revolution that a serving CMC member has been removed while still in office. The last time the CMC was this hollowed out was during a period of profound institutional chaos.
WHY ZHANG FELL.
The official framing — corruption — is the standard instrument for political removal in China. The actual reason, confirmed across multiple independent analyses of PLA Daily coverage and party documents, is more specific.
Zhang Youxia disagreed with Xi Jinping about the 2027 timeline. Not publicly. But in planning documents, in training priorities, in the way he structured the PLA's five-year development plan. Zhang wanted 2035 for full joint operational capability. Xi demanded 2027. Zhang's plan reflected what a combat veteran who understood warfare assessed as achievable. Xi's plan reflected what a political leader who has never commanded troops in combat demanded.
As Jamestown Foundation's analysis put it: "Even if newly appointed senior commanders declare the PLA's centenary goals met in 2027, its joint operational capability is unlikely to have been substantively achieved — much as Zhang Youxia, a competent and experienced commander, appears to have assessed."
In January 2026, the final annual training cycle before 2027 began. The gap between Xi's demand and Zhang's execution became impossible to paper over. The man who would not tell Xi what he wanted to hear was removed.
(Jamestown Foundation, February 2026)
WHAT WAS LOST.
Drew Thompson, former China specialist at the U.S. Department of Defense, said Zhang was "the one active-duty PLA officer who could give Xi the best, most objective advice about PLA military capabilities including the PLA's shortcomings, and crucially the human cost of military conflict."
"He could assess U.S. and Taiwan military capabilities objectively and explain to Xi Jinping what the military risks and costs of an operation to take Taiwan would be."
Then Thompson added: "A sycophant with no combat experience has to tell Xi what Xi wants to hear. I worry about the consequences of someone other than Zhang Youxia providing Xi Jinping with military advice."
(Globe and Mail, January 2026)
The U.S.-China Economic and Security Review Commission noted that replacements "could be more incentivized to implement Xi's policies regarding Taiwan and heighten the risk of miscalculation." The Taipei Times cited analysts warning that without Zhang, the PLA was "more likely to believe a self-constructed narrative that they were ready for a Taiwan adventure."
Xi Jinping now sits at the head of a 2 million-strong military, approaching a deadline he set himself, advised exclusively by people who have no incentive to tell him the truth.
THE STRATEGY NOBODY'S CALLING A STRATEGY.
The December 29th demonstration wasn't practice for an invasion. The framework that China's military planners have actually developed doesn't rely on decisive battle. It relies on paralysis.
The goal isn't to seize Taiwan in a weekend. It's to make the cost of defending Taiwan so ambiguous, so exhausting, and so risk-laden that the decision to respond gets delayed until the moment for response has passed. Fracture alliances. Reshape what "unacceptable" means. Make every actor — the U.S., Japan, Australia, Taiwan itself — run the calculation on whether this particular moment is the moment to commit.
China doesn't need to win a war. It needs to win the decision-making process before the war starts.
(CSIS China Power Project, 2025-2026)
The joint training gaps Zhang Youxia was trying to address — the ones Xi fired him for being too slow on — are real. An amphibious assault on Taiwan remains extraordinarily difficult. China's military has not fought a war in 45 years. Its equipment is modern. Its doctrine is untested. No one, including the Chinese themselves, knows how it would perform under fire.
But the paralysis strategy doesn't require winning an amphibious assault. It requires making the decision to contest one feel impossible.
THE COMPOUNDING PROBLEM.
This piece has run alongside three others in this series — the Iran negotiations, the OpenClaw exposure, and Volt Typhoon pre-positioned inside U.S. infrastructure. None of those were written to point here. But they point here.
The scenario that U.S. planners aren't publicly war-gaming: Iran goes kinetic. China reads the window — U.S. attention divided, carrier assets committed, USTRANSCOM managing a Middle East conflict. Volt Typhoon activates pre-positioned access across U.S. utilities. North Korea, reading the same window, moves against South Korea.
USTRANSCOM — the command that manages all U.S. military logistics globally — is headquartered in St. Louis. DISA Global Operations, NGA, Scott Air Force Base. All within 20 minutes of each other. All managing simultaneous crises.
The compounding problem isn't any one of these threats. It's all of them running at the same time, against infrastructure that researchers have already said will never be fully cleaned, advised by commanders who have every incentive to tell leadership what it wants to hear.
Xi Jinping isn't the only one in that situation.
WHAT 2027 ACTUALLY MEANS.
The 2027 deadline is approaching. The joint training model required for the PLA's centenary goal had not been finalized as of early 2026. The man who might have told Xi the truth about what that means was arrested in January. His replacement is a discipline officer.
Whether Xi moves on Taiwan in 2027 or not — and most serious analysts believe he won't, not yet — the clock he set is running. The capability is being built regardless of the internal chaos. The December 29th demonstration happened regardless of the purge. The paralysis strategy doesn't require a functional CMC to execute.
The Rubicon on Taiwan may not be a single dramatic crossing. It may be a series of demonstrations, each one slightly further than the last, until the line has been crossed so many times that no one remembers exactly where it was.
That's been the strategy all along.
This is the fifth piece in The Rubicon. Three pieces remain before the finale. The series continues.