The debate behind Xi Can Purge His Generals, But He Cannot Purge Doubt

Advocate A’s weakest move is treating every exposure of corruption as a net indicator of military weakness, without comparing that condition to the counterfactual: a PLA in which corrupt procurement networks, Rocket Force cliques, and senior patrons remain hidden and protected. The fact that Wei Fenghe and Li Shangfu were compromised is damaging, but A assumes the purge itself is the disease rather than part of the treatment. A stronger inference is narrower: corruption existed in critical institutions, and Xi judged it dangerous enough to impose extraordinary discipline before a crisis. That is not the same as showing that the PLA is currently unable to coerce Taiwan or obey orders under pressure.

The first counterpoint is that the purges are embedded in a long-running institutional project, not an isolated spasm. Since the 2015–16 reforms, Xi has tried to dismantle older military-region fiefdoms, centralize authority under the CMC, and assign theater commands responsibility for joint operations. That matters because Taiwan is primarily an Eastern Theater Command and CMC-directed problem, not a prestige portfolio for individual defense ministers. Removing officers who profited from procurement or built patronage networks can therefore strengthen the very chain of command A says is at risk, especially in an authoritarian system where political obedience is a component of operational reliability.

Second, A relies heavily on the Pentagon’s concern that corruption “may” have disrupted modernization, but the same U.S. defense reporting still portrays the PLA as rapidly improving, not collapsing. The 2024 China Military Power Report states that China continued expanding nuclear, naval, air, cyber, space, and conventional missile capabilities, and U.S. officials tied the 2027 modernization milestone to making the PLA a more credible instrument for Taiwan contingencies. If purges were exposing terminal weakness, we would expect visible strategic paralysis. Instead, the evidence is of a force still expanding capabilities while its top leadership is being politically scrubbed.

Third, actual behavior around Taiwan cuts against the collapse frame. After the Rocket Force and Equipment Development Department scandals became public, the PLA still conducted repeated joint drills around Taiwan, including Joint Sword-2024A/B, Strait Thunder-2025A, and Justice Mission 2025. AP reporting on the December 2025 exercises says the PLA used air, naval, rocket, and other forces and claimed the drills tested integrated joint operations. Those exercises do not prove the PLA could successfully invade Taiwan, but they do show that purges have not prevented Beijing from sustaining coercive multi-service operations.

The more complete reading is therefore mixed but unfavorable to A’s headline. Yes, corruption is a real readiness liability. But Xi’s response is also a readiness signal: he is replacing unreliable elites, warning the rest, and hardening centralized command before a contingency. Turbulence during consolidation can look like weakness while producing a more obedient war machine afterward.

A’s rebuttal concedes the decisive ground: the question is not whether the PLA has corruption or disruption, but whether the purge is better interpreted as terminal weakness or as coercive consolidation. A explicitly accepts that purges can improve political reliability, that the PLA is not paralyzed, and that modernization continues. Those concessions narrow A’s position to a timing claim about near- and medium-term friction. That is plausible, but it no longer supports the stronger headline implication that the purge “signals weakness inside China’s war machine” in the strategically relevant sense. It signals that Xi found weakness and is using centralized coercion to remove it before a crisis.

The specific weakness in A’s rebuttal is that it treats senior vacancies and removals as if they are independently probative of operational incapacity, while discounting why those removals occurred and how the post-2015 command system is designed to absorb them. China’s own reform formula places overarching authority in the CMC, operations in theater commands, and force-building in services; that means defense ministers and procurement chiefs matter, but Taiwan operations are not personally dependent on Li Shangfu or Wei Fenghe. A is right that their corruption is serious, especially given Rocket Force and procurement links. But the logical implication is not that Xi has lost the war machine; it is that he is preventing compromised procurement and promotion networks from becoming veto points in wartime.

A’s best new evidence is the IISS/Reuters readiness warning. I accept it as evidence of transitional cost. But even that formulation is organizational, conditional, and vacancy-focused: until positions are filled, command structure suffers. That does not defeat B’s claim because consolidation campaigns often impose short-run disruption to produce a more obedient hierarchy. The Chinese Defense Ministry’s October 2025 expulsion list is revealing here: the targets included He Weidong, Miao Hua, a former deputy chief of the CMC joint operations command center, the former Eastern Theater Command commander, the Rocket Force commander, and political commissars. That is exactly the kind of patronage and loyalty architecture Xi would need to scrub if he feared hesitation, factionalism, or procurement fraud in a Taiwan contingency.

Finally, A still has not answered the behavioral evidence. After the purge wave began, U.S. defense reporting still described steady PLA modernization toward the 2027 milestone and continued development of missiles, naval operations, air integration, space, cyber, and nuclear forces. Large drills do not prove invasion success, but they disprove the idea that the purge has produced paralysis. The more rigorous conclusion is therefore not “collapse exposed.” It is “corruption exposed, command being recentralized, and readiness possibly degraded during transition.” For a Taiwan scenario, a temporarily disrupted but increasingly loyal and centrally disciplined PLA may be more dangerous than a larger but factionalized force whose senior officers cannot be trusted to execute Xi’s orders.

A’s strongest case is that the purge is not a normal personnel rotation but a pattern reaching the Rocket Force, weapons procurement, theater commands, political work, and the Central Military Commission; those are precisely the institutions China would need for a fast, joint Taiwan contingency. A can also fairly argue that independent assessments such as Reuters’ reporting on IISS’s Military Balance 2026 identify real command-structure deficiencies and likely readiness costs, so China’s order of battle should not be equated with usable combat power.

I still think B’s position is stronger because A’s best evidence proves the existence of embedded weakness, not that the purge is best interpreted as exposing collapse rather than consolidating the force. The strategic question is comparative: is Xi worse off after removing compromised senior officers, or would he be worse off entering a Taiwan crisis with Rocket Force, procurement, personnel, and political-work networks still protected by patrons? The available evidence supports the second view. AP reports that Wei Fenghe and Li Shangfu were sentenced for bribery on May 7, 2026, and that Li had missile and procurement experience; that is damaging, but it also shows Xi is willing to punish exactly the senior figures whose corruption could distort missiles, weapons acquisition, and promotion incentives.

A is right that the purge is broad, but breadth can support the consolidation thesis. The October 2025 Defense Ministry list included He Weidong, Miao Hua, a former deputy chief of the CMC joint operations command center, a former Eastern Theater Command commander, the former Rocket Force commander, and senior political commissars. Those are not random removals; they are command, loyalty, personnel, and operational nodes. If Xi’s aim is to make the PLA obey centralized orders during a Taiwan confrontation, those are precisely the nodes he would discipline.

Nor does the institutional structure make individual defense ministers decisive. China’s post-2015 reform principle assigns overarching leadership to the CMC, operations to theater commands, and force-building to services. That does not eliminate disruption, but it means the system was deliberately redesigned away from old military-region fiefdoms and toward centralized joint command. Purges inside that architecture can be understood as enforcing the reform rather than invalidating it.

Finally, the behavioral record still matters. The U.S. Defense Department says corruption could slow progress toward 2027, but it also continues to describe PLA modernization across missiles, naval operations, air integration, space, cyber, and nuclear forces. AP reported that the PLA’s Justice Mission 2025 exercises around Taiwan tested integrated joint operations. Exercises do not prove invasion competence, but they disprove paralysis. Therefore the best reading remains: corruption and transitional readiness costs are real, yet Xi is hardening political reliability and operational accountability before a crisis, which may make the PLA more obedient and strategically dangerous, not less.