CN-I-03: The CCDI and the Discipline-Inspection State — The Party's Sword from Wang Qishan's Campaign to the National Supervisory Commission (2012–2026)

Status: DRAFTWords: 10,462

1. Key Takeaways

The Central Commission for Discipline Inspection (CCDI, 中央纪律检查委员会) is the Party's organ for disciplining the Party, and under Xi Jinping it became the most feared institution in China and the load-bearing instrument of his rule. The CCDI is elected by each Party Congress in parallel with the Central Committee, and its mandate is internal: it enforces the Party's own rules — which run far wider than the criminal law, covering political loyalty, factional activity, lifestyle, and "work discipline" — against the Party's roughly 99 million members [TBD-VERIFY: CCP membership ~99.2 million at end-2023, per Organization Department annual release]. The anti-corruption campaign Xi launched in late 2012 transformed it from a chronically weak watchdog, captured by the local party committees it nominally supervised, into a vertically commanded enforcement machine whose investigations have reached every province, every ministry, every major state enterprise and bank, the military high command, and — for the first time in PRC history — a former member of the Politburo Standing Committee.
The institution's pre-2012 history is an arc of creation, destruction, re-creation, and structural weakness. A Central Discipline Inspection Commission was established in November 1949 under Zhu De, replaced in 1955 by a Central Control Commission under Dong Biwu, and abolished outright in the Cultural Revolution — when "discipline" was enforced instead by mob struggle and the Central Case Examination Group, an experience that hangs over all subsequent institutional design (CN-J-03). The Third Plenum of December 1978 re-established the CCDI under Chen Yun, whose successors (Qiao Shi, Wei Jianxing, Wu Guanzheng, He Guoqiang) presided over an apparatus hobbled by the "dual leadership" (双重领导) system: each local discipline commission answered both to the higher-level commission and to the party committee at its own level — meaning, in practice, that the provincial or municipal party secretary controlled the budget, personnel, and case decisions of the very body charged with watching him. Corruption metastasised through the 1990s and 2000s precisely in the gap this design created.
Wang Qishan's tenure as CCDI secretary (2012–2017) was the institutional revolution: he converted a captured watchdog into the engine of the most sustained elite purge in post-Mao history. Three reforms did the work. First, verticalisation: the November 2013 Third Plenum decision made case work answerable primarily to the higher-level commission and gave higher-level commissions control of nominations for lower-level discipline secretaries — ending the dual-leadership capture in all but name. Second, the central inspection teams (中央巡视组, xunshi): roving delegations dispatched to provinces, SOEs, banks, universities, and ministries achieved, for the first time, full coverage of all provincial units and major central SOEs within a single congress cycle, generating the case leads for many of the biggest falls. Third, full-coverage stationing (派驻全覆盖): resident CCDI offices embedded in every central Party and state organ. The CCDI thereby became the centre's eyes inside every node of the architecture mapped in CN-I-01.
The campaign's scale has no precedent in the history of Leninist parties: millions disciplined, hundreds of senior officials felled, and no end-point ever declared. By the eve of the 19th Congress the CCDI reported over 1.5 million Party members punished since late 2012 [TBD-VERIFY: 1.537 million in the official 2012–2017 series]; by the 20th Congress the cumulative figure approached 4.7 million investigated [TBD-VERIFY: CCDI's October 2022 report cited ~4.65 million cases filed since 2012]; and the annual totals rose rather than fell thereafter, with the 2024 figure reported above 800,000 punished in a single year — a campaign-era record [TBD-VERIFY: CCDI-NSC 2024 annual statistics, reported at ~889,000 disciplined]. The "tigers" (老虎) include Zhou Yongkang (the first PSC-rank conviction, life sentence June 2015), Bo Xilai (life, September 2013), Ling Jihua (life, July 2016), Sun Zhengcai (life, May 2018), and the military tigers Xu Caihou and Guo Boxiong treated in CN-I-02 §4 — alongside hundreds of officials at vice-ministerial rank and above [TBD-VERIFY: counts of centrally-managed cadres investigated range from ~440 (2012–2022) upward depending on counting rules].
The 2018 creation of the National Supervisory Commission (NSC, 国家监察委员会) constitutionalised the discipline apparatus and extended it beyond the Party to everyone exercising public power. The 11 March 2018 constitutional amendment — the same package that abolished the presidential term limit (CN-K-01) — created a new category of state organ ranking above the Supreme People's Court and Procuratorate, and the Supervision Law of 20 March 2018 gave it jurisdiction over six categories of public personnel: civil servants, SOE managers, public-hospital and public-university staff, village officials, and others performing public duties — a roughly threefold expansion of supervised persons in the pilot jurisdictions [TBD-VERIFY: the Beijing pilot reported supervision targets rising from ~210,000 to ~997,000]. The NSC shares offices, personnel, and leadership with the CCDI under the "one institution, two nameplates" device: it is the same body wearing a state hat. The extra-legal shuanggui (双规) detention was replaced by statutory liuzhi (留置) — read by defenders as the legalisation of a long-lawless practice, and by critics as the constitutional entrenchment of incommunicado detention without counsel.
The campaign is simultaneously a genuine anti-corruption drive and an instrument of power consolidation, and the corpus holds both readings rather than choosing. The factional arithmetic is unmistakable: the Zhou Yongkang network (the petroleum system, the Sichuan apparatus, the political-legal organs), the Ling Jihua circle, and the Sun Zhengcai removal cleared every organised alternative to Xi's authority, and "political discipline" — loyalty to the centre and to Xi as its core — now ranks first among the Party's six disciplines. Yet the quantitative scholarship complicates the pure-purge reading: Lorentzen and Lu's analysis found that targets were predicted by corruption indicators, not merely by factional ties, and the campaign's reach into "flies" — the hundreds of thousands of township-level extortionists, clinic directors, and village accountants disciplined annually — has no factional rationale. The honest formulation: the campaign is real and weaponised; its genuine popular legitimacy and its elite-political function are the same instrument viewed from below and from above.
The campaign's signature governance side-effect is bureaucratic paralysis — the "lying flat" official — and the centre has spent a decade fighting the deterrence it created. When the price of initiative can be a liuzhi cell, officials rationally choose inaction (不作为, "non-performance"; 躺平, "lying flat"). The collapse of the banquet-and-gift economy after the December 2012 Eight-Point Regulations (八项规定) — crashing premium liquor prices, luxury retail, and Macau gaming revenue — demonstrated the apparatus's behavioural power; the same power froze discretionary decision-making in finance bureaus and project approvals. The centre's counter-doctrine — the "three distinguishes" (三个区分开来) fault-tolerance rules meant to protect bold-but-honest officials, and repeated exhortations to "dare to act" (敢于担当) — is an official acknowledgment that the discipline state over-deterred [TBD-VERIFY: the specific corpus of 担当/容错 directives, 2016–2025]. Whether the trade was worth it is one of the document's open questions.
The discipline state never ended; it serialised. The 2012–2017 storm gave way not to normalisation but to rolling sectoral sweeps: the political-legal apparatus "rectification" of 2020–2022 (Sun Lijun, Fu Zhenghua); the finance sweep of 2021–2025 that reached the chairmen of major state banks and a PBoC deputy governor, and executed Huarong's Lai Xiaomin (January 2021); the 2023 healthcare and pharmaceutical sweep that detained hospital directors by the hundred; the football and sports-administration cases (national coach Li Tie's 20-year sentence; CFA president Chen Xuyuan's life sentence) [TBD-VERIFY: sentence details]; and the 2023–2025 military purge of the Rocket Force, two defence ministers, and serving CMC members (CN-I-02 §5). Xi's own formulation — the struggle is "always on the road" (永远在路上) — is institutional doctrine: the campaign is not a phase of the system but a permanent organ function.
The deepest problem the CCDI embodies is the self-supervision paradox: in a system where the Party leads everything, only the Party can police the Party — and who, then, polices the police? The CCDI answers to the Central Committee and, in practice, to the General Secretary; no court reviews its detentions, no press investigates it freely, no opposition audits it. The apparatus has had to invent recursive layers — an internal supervision office for discipline officials created in 2014, and a 2023 "education and rectification" campaign aimed at the discipline-inspection corps itself [TBD-VERIFY: the 2023 纪检监察干部队伍教育整顿 campaign and its case totals] — which reproduce the same structure one level down. The "who watches the watchdog" question is not a Western gotcha but the system's own acknowledged design problem, and it connects directly to the succession question: an apparatus this powerful, loyal to one man, is itself a stake in any transition (CN-O-01 §2).
Comparatively, the CCDI-NSC is the most consequential answer any Leninist party has given to the discipline problem that destroyed its Soviet predecessor — and the jury on its long-run effect is out. The Soviet party-control organs decayed into irrelevance and Brezhnev-era corruption was never seriously confronted; Vietnam's parallel "blazing furnace" campaign shows the model travelling within the Leninist family; Singapore's CPIB and Hong Kong's ICAC represent the rival, rule-of-law route to clean government that China explicitly declined. The scholarship's central finding (Carothers, Manion) is that authoritarian anti-corruption succeeds when the top leader is both willing and institutionally unconstrained — which is exactly the configuration that makes such campaigns inseparable from personal power. The CCDI's centrality is therefore the clearest single window onto the system's nature: a party-state that, rejecting external accountability on principle, has built the largest internal-accountability machine in history, and staked its integrity on the machine's master.

2. The Institution — Architecture, History, and the Wang Qishan Transformation

2.1 What the CCDI Is

The Central Commission for Discipline Inspection occupies a distinctive place in the architecture mapped in CN-I-01: it is the only major Party organ elected directly by the National Congress alongside the Central Committee itself, a formal parity that signals its constitutional role as the Party's conscience standing apart from — though never above — the Party's executive line. Chapter VIII of the Party Constitution charges it with upholding the Party Constitution and rules, enforcing discipline, and combating corruption. Its secretary has, since the 14th Congress (1992), conventionally held a seat on the Politburo Standing Committee — Wang Qishan (2012–2017), Zhao Leji (2017–2022), Li Xi (since October 2022) [TBD-VERIFY: Li Xi's continuation as CCDI secretary as of 2026] — which places the discipline portfolio permanently inside the innermost cabinet.

The crucial conceptual point, developed throughout Ling Li's scholarship, is that Party discipline (纪律, jilü) is a separate and wider normative order than state law. The Party's disciplinary regulations — comprehensively revised in 2015, 2018, and December 2023 — codify "six disciplines" (六项纪律): political discipline, organisational discipline, integrity discipline, mass-work discipline, work discipline, and lifestyle discipline. Only a fraction of this maps onto the criminal code. A cadre can be destroyed for "improper discussion of central policy" (妄议中央), for forming cliques, for extravagant banqueting, for adultery, or for "disloyalty and dishonesty to the Party" — none of which is a crime. The 2015 revision deliberately stripped legal-offence language out of the disciplinary regulations to enforce the doctrine of "discipline before the law" (纪在法前): the Party judges its members by its own, stricter standard first, and hands them to the state's courts only afterwards, if at all. The standard sequence in a major case — disciplinary investigation, expulsion from the Party (双开, "double expulsion" from Party and office), then transfer to prosecutors — enacts this priority visibly: by the time a tiger reaches a courtroom, his political annihilation is complete and the verdict is a formality.

2.2 The Arc from 1949: Creation, Destruction, Re-creation

The institution's history before 2012 is essential context, because the Xi-era transformation answered weaknesses that were three decades old, and the institution's Cultural Revolution abolition explains the system's insistence on organised discipline.

The first Central Discipline Inspection Commission was established in November 1949, weeks after the PRC's founding, under Zhu De; it was replaced in 1955 by a Central Control Commission under Dong Biwu, modelled more closely on the Soviet party-control organs. Both were marginal bodies. In the Cultural Revolution the Control Commission was abolished altogether (1969), and "discipline" passed to the Central Case Examination Group and to struggle sessions — discipline as mob politics, directed by the leader against the apparatus. That experience (treated as contested memory in CN-J-03) is the negative founding of the modern institution: when the Third Plenum of the 11th Central Committee re-established the CCDI in December 1978, with the immense prestige of Chen Yun as First Secretary, its first great task was rehabilitation — reversing the Cultural Revolution's verdicts — and its design intent was that discipline should henceforth run through rules and organs, never again through campaigns against the organisation itself. The irony that the CCDI would become, after 2012, the engine of the largest campaign-style purge since Mao is not lost on the institution's scholars.

Chen Yun's successors — Qiao Shi (1987–1992), Wei Jianxing (1992–2002), Wu Guanzheng (2002–2007), He Guoqiang (2007–2012) — ran an apparatus that grew steadily in formal powers and steadily lagged the corruption it confronted. The reform era's marketisation created what Andrew Wedeman called the "double paradox": corruption intensifying alongside rapid growth, with the state periodically executing a banker or a provincial chief (Hu Changqing, 2000; Cheng Kejie, the NPC vice-chairman executed in 2000) without denting the underlying political economy. High-profile pre-2012 cases — Chen Xitong (Beijing party secretary, felled 1995), Chen Liangyu (Shanghai party secretary, felled 2006), Lai Changxing's Yuanhua smuggling empire in Xiamen — followed a pattern the field read as factional: the watchdog bit whom the leadership pointed it at, and the leadership pointed it, by an unwritten rule, at no one above the Politburo Standing Committee. The maxim 刑不上常委 — "punishment does not reach the Standing Committee" — was the system's informal immunity clause.

2.3 The Dual-Leadership Weakness

The structural reason for the watchdog's weakness was the dual leadership system (双重领导体制). Every local discipline inspection commission answered to two masters: the discipline commission one level up (the professional line, 条) and the party committee at its own level (the territorial block, 块) — the same tiaotiao kuaikuai geometry diagnosed in CN-I-01 §4. But the block dominated: the local party committee controlled the discipline commission's budget, offices, careers, and — decisively — the decision whether a case against a local official could be opened at all. A municipal discipline secretary investigating the municipal party secretary's allies was investigating the man who wrote his evaluation. The predictable result, documented in Xuezhi Guo's China Quarterly anatomy, was that discipline commissions caught flies in other people's jurisdictions and went blind at home; major cases happened only when the centre intervened from above, ad hoc.

The apparatus's signature investigative instrument compounded the problem of legality with a problem of accountability. Shuanggui (双规, "double designation") — formalised in the CCDI's 1994 case-work regulations, with a state-side twin (shuangzhi) under the 1990 administrative-supervision rules — empowered investigators to order a Party member to present himself at "a designated time and place" and remain there for questioning. In practice this meant incommunicado detention, typically in commandeered hotels, party schools, or unmarked facilities, without arrest, charge, lawyer, judicial supervision, or fixed duration. Flora Sapio's foundational study located shuanggui in a "zone of lawlessness" deliberately maintained: because the subject was held under Party rules rather than state law, none of the Criminal Procedure Law's protections applied. The system's defenders argued it was indispensable — corrupt officials commanded the local police and courts, so only an extra-judicial channel could break protection networks. Its documented costs ran from coerced confessions to deaths in custody: the case of Yu Qiyi, a Wenzhou state-enterprise engineer drowned during interrogation in April 2013 (six investigators were later convicted), became the emblematic documented fatality [TBD-VERIFY: Human Rights Watch's 2016 report catalogued at least 11 shuanggui deaths 2010–2016 from public sources; the true figure is unknowable].

2.4 The Wang Qishan Transformation, 2012–2017

What changed after November 2012 was not the CCDI's formal mandate but its command structure, its coverage, and the political will behind it. Xi Jinping signalled the campaign within weeks of taking office — the Politburo's Eight-Point Regulations on official austerity (4 December 2012), then the January 2013 CCDI plenary speech vowing to cage power "in the cage of regulations" and to strike "tigers and flies together" (老虎苍蝇一起打). The instrument was the CCDI under Wang Qishan, the troubleshooting "fire brigade chief" of the elite (treated biographically in the planned H-CCDI series), whose tenure remade the institution in three structural moves.

First, verticalisation — the effective end of dual leadership. The Third Plenum decision of November 2013 ordered that case work (查办腐败案件) be led primarily by the higher-level commission: discovery of a problem and the opening of an investigation must be reported up the discipline line, not cleared with the local party committee; and nominations for lower-level discipline secretaries passed to the control of the higher-level commission together with the organisation departments. Formally, dual leadership survives in the Party Constitution; operationally, the reform inverted the dominance of block over line. The local watchdog now answers, on everything that matters, to Beijing's watchdog. This is why the corpus speaks of a discipline-inspection state: a single national hierarchy of supervision, commanded from the centre, running parallel to and unaccountable to every territorial government.

Second, the weaponisation of the central inspection system (xunshi, 巡视). Roving central inspection teams had existed since 2003, producing polite reports. Wang turned them into the campaign's reconnaissance arm: teams of senior (often retired, hence unbribable-by-promotion) officials descending on a province, ministry, SOE, bank, or university for weeks, taking petitions, interviewing cadres one by one, auditing books, and reporting directly to the central leadership — bypassing every local structure. Innovations sharpened the blade: "one authorisation per assignment" (一次一授权) so no team developed a standing relationship with its targets; specialised inspections (专项巡视) aimed at a single suspect organisation; unannounced "look-back" re-inspections (回头看) to catch institutions that had relaxed. The 18th Central Committee cycle achieved, for the first time, full inspection coverage of all 31 provincial units, the major central SOEs and financial institutions, and the centrally administered universities [TBD-VERIFY: the official claim of 100% coverage across 277 inspected entities in the 18th CC cycle]. Inspection leads generated many of the era's biggest cases — the Shanxi "systemic collapse" of 2014, in which most of the provincial leadership fell, and the lead chain to Sun Zhengcai among them [TBD-VERIFY: attribution of specific case origins to xunshi leads].

Third, full-coverage stationing (派驻全覆盖). Resident discipline offices had existed in some ministries; by 2015–2016 the CCDI completed their installation in every central Party and state organ — including, pointedly, the Party's own central departments (Organization, Propaganda, United Front, the General Office) which had never before hosted a resident watchdog. The resident offices answer to the CCDI, not to their host. Combined with the verticalised local commissions and the roving inspections, the Party centre acquired, for the first time, a standing surveillance presence inside every node of its own apparatus.

Wang's tenure also supplied the campaign's doctrine. His formulations became canon: "trust cannot replace supervision" (信任不能代替监督); the strategic sequencing of "not daring to be corrupt, not being able to be corrupt, not wanting to be corrupt" (不敢腐、不能腐、不想腐) — deterrence first, institutional closure second, ideological conversion third; and the "four forms of discipline supervision" (四种形态), which deliberately push the apparatus's centre of gravity down the severity scale, making criticism-and-education the most common disposition and criminal transfer the rarest — discipline as ambient, continuous management of the entire membership rather than occasional decapitation. By the 19th Congress the CCDI was no longer a watchdog at all in the liberal sense of the word. It was, as this document's title proposes, the core organ of a discipline-inspection state — and the 2018 reform (§4) would write that state into the constitution.

3. The Campaign's Record, 2012–2026 — Scale, Tigers, Phases, and the Effectiveness Question

3.1 The Arithmetic

The campaign's scale is best grasped through the CCDI's own annual statistics, which the commission publishes with bureaucratic pride and which, even discounted for definitional elasticity, describe an enforcement operation without precedent in any ruling party's history. By the eve of the 19th Congress (October 2017), the apparatus reported more than 1.5 million Party members and officials punished since the 18th Congress [TBD-VERIFY: the commonly cited official figure is 1.537 million disciplined November 2012–October 2017, with ~58,000 transferred to judicial organs]. By the 20th Congress (October 2022) the cumulative count of cases filed since 2012 approached 4.65 million, involving roughly 4.7 million people [TBD-VERIFY: CCDI work report to the 20th Congress]. And the curve did not flatten: annual dispositions reached approximately 610,000 in 2023 and approximately 889,000 in 2024 — each a record — with 2025 reported in the same range [TBD-VERIFY: CCDI-NSC annual statistics for 2023–2025; the 2024 figure of ~889,000 was widely reported from the January 2025 release]. At the elite tier, the count of "centrally managed cadres" (中管干部, roughly vice-ministerial/provincial rank and above) placed under investigation runs in the hundreds: about 440 across 2012–2022 by common tallies, with the annual rate accelerating in the third term — 45 in 2023 and over 50 in 2024, the highest yearly totals of the entire campaign [TBD-VERIFY: all elite-tier counts; tallies differ by whether military officers and retirees are included].

Two features of this arithmetic deserve emphasis. First, the overwhelming majority of cases are flies: township cadres, village accountants, clinic and school administrators, traffic police — the petty-extortion layer where ordinary citizens actually encounter corruption. The CCDI's reporting consistently highlights "corruption at the masses' doorstep" (群众身边的腐败), and the post-2018 supervision system's extension to village officials (§4) pushed the apparatus's reach below the Party's own membership floor. Second, the rising annual totals a decade into the campaign are themselves an interpretive puzzle the corpus flags without resolving: they can be read as evidence that corruption regenerates faster than enforcement (the pessimist reading), that detection capacity keeps improving (the institutionalist reading), or that the apparatus, like any bureaucracy evaluated on output, has learned to produce cases (the public-choice reading).

3.2 The Tigers

The campaign's elite inventory broke every protective convention of post-Mao politics in sequence.

Bo Xilai — Politburo member and Chongqing party secretary — fell in March 2012, before Xi's accession, in the Wang Lijun/Neil Heywood scandal (the prelude treated in CN-B-01); his life sentence (September 2013) for bribery, embezzlement, and abuse of power, delivered after a part-transcribed trial in Jinan, established the template of the political trial conducted in legal form.
Zhou Yongkang — retired PSC member, former security czar and Political-Legal Affairs Commission chief — was the rule-breaking case: his investigation (formally announced 29 July 2014), expulsion (December 2014), and life sentence (11 June 2015, closed trial in Tianjin) ended the 刑不上常委 immunity convention. The case dismantled an entire interlocking network: the petroleum system (Jiang Jiemin, Su Shulin), the Sichuan apparatus (Li Chuncheng), the security services (Ma Jian), and the "secretaries gang" (秘书帮) of aides.
Xu Caihou and Guo Boxiong — the former CMC vice-chairmen whose destruction (2014–2016) is treated in CN-I-02 §4 — extended the campaign into the military's highest uniformed offices, where rank itself had been for sale.
Ling Jihua — Hu Jintao's chief aide, whose fall began with his son's fatal Ferrari crash (March 2012) and the cover-up that followed — was investigated from December 2014 and sentenced to life in July 2016; his "Xishan Society" network of Shanxi-connected officials fell with him, contributing to the 2014 "systemic collapse" (塌方式腐败) of the Shanxi provincial leadership.
Sun Zhengcai — Politburo member, Chongqing secretary, and by common reading a potential successor-generation figure — was removed in July 2017, weeks before the 19th Congress, and sentenced to life in May 2018; the official charges included bribery, and the accompanying political vocabulary ("conspiratorial activities," ambitions against the Party centre) marked the campaign's open fusion of corruption and loyalty offences. Official rhetoric retrospectively grouped Zhou, Bo, Ling, Xu, and Sun as plotters whose "political conspiracy activities" (政治阴谋活动) had threatened Party unity [TBD-VERIFY: the canonical official formulation, used in 19th Congress-era CCDI communiqués and the Always on the Road documentary cycle].

The post-2017 tigers confirmed that no office conferred immunity: Interpol's serving president Meng Hongwei (detained on a 2018 visit home, 13½ years); justice minister Fu Zhenghua and vice-minister of public security Sun Lijun (the 2020–2022 political-legal sweep, both receiving suspended death sentences); and, in the third term, the serving foreign minister Qin Gang (removed July 2023, never formally charged publicly) and the defence ministers Li Shangfu and Wei Fenghe (expelled June 2024) — the military cases cross-referenced to CN-I-02 §5 [TBD-VERIFY: Qin Gang's case disposition remains officially unexplained as of 2026].

3.3 The Phases: Storm, Institutionalisation, Serial Sweeps

The campaign's fourteen years sort into three phases. The 2012–2017 storm was the Wang Qishan era described in §2.4: maximum velocity, the destruction of the great factional networks, the structural reforms. The 2018–2022 institutionalisation converted campaign into system: the NSC's creation (§4), the Supervision Law, the proliferation of liuzhi facilities, the routinisation of inspection cycles, and the discipline regulations' progressive rewriting around political loyalty under Zhao Leji's lower-profile secretaryship. The third-term serialisation (2022–2026), under Li Xi, abandoned any implication of an end-state in favour of rolling sectoral sweeps:

Political-legal rectification (2020–2022): a self-described "blade turned inward" (刀刃向内) campaign across police, courts, procuratorates, and state security — the apparatus Zhou Yongkang had built — felling Sun Lijun's "clique" and provincial police chiefs by the dozen.
Finance (2021–2025): a sweep across state banks, regulators, asset managers, and the central bank — Huarong chairman Lai Xiaomin (executed January 2021, an extraordinary penalty signalling the sector's turn), Bank of China chairman Liu Liange, China Life's Wang Bin, Everbright's Li Xiaopeng and Tang Shuangning, China Merchants Bank's Tian Huiyu, and PBoC deputy governor Fan Yifei among the most senior [TBD-VERIFY: individual dispositions]. The sweep coincided with — and enforced — the "common prosperity" repricing of the financial elite treated in CN-D-01.
Healthcare and pharmaceuticals (2023–): a centrally coordinated sweep of hospital procurement and pharmaceutical kickbacks that reportedly placed well over a hundred hospital directors and party secretaries under investigation in its first months [TBD-VERIFY: figures of ~160–180+ hospital directors investigated in 2023 circulated in Chinese financial media], visibly chilling medical-conference activity and device procurement.
Sports and football (2022–2024): the destruction of the football administration — former national coach Li Tie (20 years), CFA president Chen Xuyuan (life), and sport-ministry deputy Du Zhaocai among more than a dozen convicted [TBD-VERIFY: sentences].
Grain, tobacco, energy, and agriculture (2021–2026): rolling sweeps of the grain-storage system, the tobacco monopoly, and the energy SOEs; agriculture minister Tang Renjian's conviction (suspended death sentence, 2025) [TBD-VERIFY] kept serving State Council ministers within the campaign's reach.
The military (2023–2026): the Rocket Force decapitation, the equipment-procurement investigations, and the expulsions reaching serving CMC members — treated in full in CN-I-02 §5, but institutionally significant here because the cases were processed through the same CCDI-centred machinery, confirming the discipline apparatus's jurisdiction over the gun.

3.4 The Effectiveness Debates

Has it worked? The question decomposes into three sub-questions on which the literature divides.

Did corruption fall? Measurement is the first casualty: corruption is observed only when prosecuted, so enforcement statistics conflate the disease with the cure. The indirect indicators mostly point one way for petty and transactional corruption: the banquet economy's collapse (§6), survey evidence of sharply reduced gift-and-bribe expectations in routine dealings, and China's gradual climb in Transparency International's perception index from the high-90s ranks toward the mid-60s [TBD-VERIFY: CPI rank trajectory ~rank 80 (2013) to ~65–76 (2020s)]. For elite corruption the verdict is murkier: Minxin Pei argues the underlying crony-capitalist political economy — power monetised through family networks, as the 2012 New York Times and Bloomberg exposés of elite family wealth documented — was disciplined, not dismantled, and that the campaign conspicuously never produced public asset-disclosure rules for senior officials' families, the reform every clean-government literature recommends first.

Was it a purge or a clean-up? Yuen Yuen Ang's typology sharpens the question: the campaign demonstrably crushed "speed money" and petty theft, but "access money" — the elite exchange of power for wealth that Ang argues fuelled China's growth model — is precisely what the biggest cases involved, and its prosecution tracked political geography. The quantitative work splits the difference: Lorentzen and Lu found that investigated officials were predicted by corruption-proxies (land transactions, construction booms) and not solely by factional distance from Xi, supporting a real-enforcement reading; network studies of patron-client ties find protection effects — officials tied to incumbent Politburo patrons faced lower investigation hazard — supporting a selective-enforcement reading [TBD-VERIFY: attribution of the patronage-protection finding, associated with work by Junyan Jiang and co-authors]. The corpus's position is the both/and stated in §1: the campaign is simultaneously real and political, and its two functions are not separable in principle, because in a Leninist system disloyalty is a discipline violation.

What did deterrence cost? The campaign's least-contested empirical legacy is bureaucratic risk-aversion. With initiative carrying career-ending tail risk and the old incentive currency (gray income, banquets, promotion-for-patronage) confiscated, a substantial fraction of the official class rationally retreated into "lying flat" (躺平) inaction — approving nothing, signing nothing, innovating nothing. The phenomenon is officially acknowledged: the centre coined the discipline offence of "non-performance and slow performance" (不作为、慢作为) — punishing inaction with the same apparatus that caused it — and promulgated the "three distinguishes" (三个区分开来) fault-tolerance doctrine, instructing discipline organs to distinguish honest reform errors from venality and to protect officials who "dare to act" (敢于担当) [TBD-VERIFY: the doctrine's canonical statements, attributed to Xi from 2016 and elaborated in 2018 and 2022 CCDI documents]. Economists have connected discipline-driven paralysis to slowed local investment execution and to the local-government behaviour patterns treated in CN-D-03. The deterrence-versus-paralysis trade is the campaign's standing governance dilemma: the fear that makes officials honest is the same fear that makes them inert, and fourteen years in, the centre is still adjusting the dial rather than resolving the tension.

4. The 2018 Reform — The National Supervisory Commission and the Constitutionalisation of Discipline

4.1 The Constitutional Moment

On 11 March 2018 the 13th National People's Congress adopted a constitutional amendment package by a vote of 2,958 to 2. Western coverage focused overwhelmingly on one clause — the deletion of the presidential two-term limit, treated in CN-K-01 — but the package's largest structural change was the insertion of an entirely new Section 7 into Chapter III of the State Constitution, creating supervisory commissions (监察委员会) as a fourth branch of the state apparatus, listed alongside and ranking in protocol above the Supreme People's Court and the Supreme People's Procuratorate. Nine days later the same NPC session adopted the Supervision Law (《监察法》, 20 March 2018), the new organ's statutory charter. The simultaneity with the term-limit removal was not incidental: both changes constitutionalised the Xi-era configuration of power — the indefinite leader and the permanent discipline apparatus — in a single legislative stroke, and the corpus treats them as one constitutional event with two faces.

The reform had been piloted from November 2016 in Beijing, Shanxi, and Zhejiang — three jurisdictions chosen, observers noted, for their political reliability and, in Shanxi's case, as the site of the collapse the apparatus was proudest of repairing. The pilots merged each jurisdiction's discipline commission with the government's supervision department and the procuratorate's anti-graft and anti-dereliction bureaus into a single supervisory commission. The national rollout in early 2018 replicated the merger at every level, transferring tens of thousands of procuratorate investigators into the new commissions [TBD-VERIFY: figures of ~44,000 procuratorate anti-corruption personnel transferred circulated in official accounts].

4.2 The State Wrapper on the Party Organ

The National Supervisory Commission's defining design feature is that it is not a new institution. It is the CCDI wearing a state nameplate — the fullest application of the "one institution, two nameplates" (一个机构两块牌子) device described in CN-I-01 §13. The NSC and CCDI share premises, personnel, case files, and leadership: the CCDI secretary outranks the NSC director, who has conventionally been the CCDI's deputy secretary (Yang Xiaodu 2018–2023, Liu Jinguo from 2023 [TBD-VERIFY: current NSC director as of 2026]). Formally, the NSC is created by and answerable to the NPC; in reality no NPC organ supervises its case work, and the Supervision Law's drafting was run through the CCDI itself rather than the NPC's usual legislative channels — a provenance the law's academic critics, including PRC constitutional scholars during the 2017 comment period, did not fail to note [TBD-VERIFY: the public comment controversy, including objections from PRC law professors to the draft's lawyer-access exclusion].

The functional meaning of the state wrapper is jurisdictional. As a Party organ the CCDI could discipline only Party members. The NSC's writ runs to all persons exercising public power — the Supervision Law's six enumerated categories: civil servants and those managed as such; persons in organisations authorised by law to manage public affairs; managers of state-owned enterprises; managers in public education, research, culture, healthcare, and sport institutions; managers of grassroots mass autonomous organisations (village and residents' committees); and a catch-all of "other persons performing public duties." The expansion roughly tripled the supervised population where measured — the Beijing pilot reported supervision targets rising from about 210,000 to about 997,000 [TBD-VERIFY: pilot coverage figures] — and brought the non-Party hospital director, the village accountant, and the university administrator inside an apparatus from which Party membership had previously been the gate. Among the categories swept in, the village-committee clause matters most for daily governance: the "flies at the doorstep" enforcement of §3.1 runs on it.

4.3 Liuzhi — The Legalisation of the Extra-Legal, or Its Entrenchment

The Supervision Law abolished shuanggui and created liuzhi (留置, "retention in custody"): detention in dedicated facilities for up to three months, extendable once to six, with family notification within twenty-four hours "except where it would impede the investigation," interrogations recorded, and detention conditions standardised. What it conspicuously did not provide is access to a lawyer at any point during liuzhi, judicial authorisation or review of the detention, or application of the Criminal Procedure Law — the detainee enters the criminal-justice system, with its (limited) protections, only when the commission transfers the case to the procuratorate, by which point the investigation is complete and a confession nearly always in hand. The parallel 2018 Criminal Procedure Law amendment fitted the receiving end to the pipeline, extending the "leniency for admission of guilt and acceptance of punishment" (认罪认罚从宽) system that makes the post-transfer trial an act of ratification.

Two readings of liuzhi contend, and both are stated honestly. The legalisation reading — the official account, with some support among PRC legal scholars — holds that the reform brought a four-decade-old lawless practice under statute: fixed time limits where shuanggui had none; purpose-built, camera-monitored facilities where shuanggui used hotel rooms (the setting of most documented deaths); recorded interrogations; family notification as default. On this account the 2018 reform is the single largest rule-of-law gain of the Xi era, achieved by the paradoxical route of constitutionalising the institution that had most flagrantly operated outside law. The entrenchment reading — Human Rights Watch, most foreign legal scholarship, and the quieter PRC critics — holds that legalising a practice is not the same as constraining it: liuzhi preserved the operative features that made shuanggui coercive (incommunicado custody, no counsel, no court), extended them from Party members who had notionally volunteered for Party rules to all public employees who had not, and removed the procuratorate's independent investigation as even a formal check by absorbing it. The December 2024 Supervision Law amendments, which refined the measure's procedures and reportedly extended maximum durations for major cases, are read by each side as confirmation [TBD-VERIFY: the 2024 amendment's content — reported provisions included extending liuzhi to a further period for especially complex cases and adding lesser compulsory measures; verify against the adopted text]. The documentation problem is itself part of the institution: no independent monitoring of liuzhi facilities exists, detainee accounts surface only in fragments, and the system's interior is known mainly through its outputs — confessions, convictions, and the occasional announced death [TBD-VERIFY: documented liuzhi-era deaths in custody; isolated cases have been reported in Chinese media but no systematic count exists].

5. The Political Functions — Consolidation Instrument, Governance Machinery, and the Discipline-State Concept

5.1 The Power-Consolidation Function

Read politically, the campaign's first decade performed a function familiar from the comparative literature on authoritarian consolidation: it destroyed every organised network capable of constraining the new leader. The sequence is legible in the §3.2 inventory. The Zhou Yongkang case dismantled the security-petroleum-Sichuan complex — the only institutional bloc with coercive resources outside the military chain. The Ling Jihua case decapitated the outgoing patron's office network. The Xu Caihou and Guo Boxiong cases broke the military patronage machine (CN-I-02 §4). The Sun Zhengcai case, on the eve of the 19th Congress, eliminated the most plausible successor-generation rival and warned the rest. By 2017 the factional balance-of-power politics that had structured the Jiang-Hu decades was simply gone — an outcome CN-C-01 documents at the congress level and CN-O-01 §2 extends into the succession analysis: a system purged of alternative networks is also a system with no organised mechanism for replacing its leader.

After consolidation, the same machinery converted to loyalty enforcement. The discipline regulations' 2015, 2018, and 2023 revisions progressively moved political offences to the front: "political discipline" heads the six disciplines, and its enumerated violations — improper discussion of central policy (妄议中央), failure to implement the centre's decisions, "two-faced" behaviour (两面人), cliques and factions, disloyalty to the Party — criminalise, in Party-normative terms, precisely the behaviours of elite politics. The "two safeguards" (两个维护) — safeguarding Xi's core status and the centre's authority, treated doctrinally in CN-K-05's companion analysis of the thought apparatus — are enforced as discipline, audited in every inspection, and cited in nearly every senior expulsion notice of the third term. The CCDI thus polices not a boundary between honesty and theft but a boundary between loyalty and everything else; corruption charges are the lingua franca in which loyalty verdicts are pronounced, which is why the system's falls are simultaneously unfalsifiable (every official has discipline exposure) and informative (the choice of whom to fall is the message).

5.2 The Governance Function

It would nonetheless be a misreading — the mirror image of taking the official account at face value — to treat the apparatus as only a purge machine. The discipline-inspection state is also the centre's principal compliance machinery: the instrument by which Beijing makes its writ run through the fragmented, principal-agent-riddled hierarchy described in CN-I-01 §4 and §8.

The mechanism is the conversion of policy priorities into discipline categories. When poverty alleviation became the signature first-term target, the CCDI ran dedicated inspections of poverty-relief funds and disciplined tens of thousands of cadres for embezzling or falsifying them [TBD-VERIFY: annual totals for poverty-alleviation discipline cases, ~80,000+ in peak years per CCDI releases]. When environmental targets rose, the parallel central environmental-protection inspections (a xunshi-modelled system run through the environment ministry but backed by discipline consequences) made pollution a career event for the first time. During zero-COVID (CN-K-02), discipline organs punished thousands of officials for outbreak "failures" in both directions — first for insufficient rigour, then, after December 2022, for excessive rigidity. The 2023-revised regulations added "formalism and bureaucratism" (形式主义、官僚主义) — fake compliance, paperwork theatre, hollow implementation — as an enforcement priority, an attempt to use the discipline state against the evasive behaviours the discipline state itself incentivises. In Pierre Landry's terms (CN-I-01 §8), the nomenklatura controls who rises; the discipline apparatus controls who falls — and the pairing of promotion tournament and investigation risk is the complete incentive system of the Chinese official class.

5.3 The Discipline-State Concept and the Self-Supervision Paradox

The deepest analytical claim this document makes is that the CCDI-NSC is not an agency of the system but a constitutive answer to the system's core problem. A party that leads everything, acknowledges no external sovereign above itself — no independent court, no free press, no opposition, no electorate — must generate all of its accountability internally. "Self-revolution" (自我革命), Xi's term of art elevated in the third term as the Party's answer to the "historical cycle" (历史周期率) question — Huang Yanpei's famous 1945 challenge to Mao about dynastic rise and decay, to which Mao answered "democracy" and Xi has answered "self-revolution" — is the official name for this wager: that a ruling organisation can be its own permanent auditor [TBD-VERIFY: the Xi "second answer" formulation, articulated around the 2021 centenary and the Sixth Plenum].

The wager's structural flaw is the self-supervision paradox. The CCDI answers upward — to the Central Committee formally, to the General Secretary actually. Nothing audits it from outside. The apparatus's own answer has been recursion: a dedicated internal-supervision office for discipline officials (created 2014, after cases of CCDI investigators selling case information); the 2017 documentary To Forge Iron, One Must Be Strong Oneself, devoted entirely to corrupt discipline officials; and the 2023 nationwide "education and rectification" (教育整顿) of the discipline-inspection and supervision corps, which reportedly placed thousands of the watchers under investigation [TBD-VERIFY: 2023 rectification case totals; senior internal falls include the state-security discipline chief Liu Yanping (2022) and CCDI-system officials at vice-ministerial rank]. Each recursive layer reproduces the paradox one level down; the regress terminates only at the person of the leader, who is supervised by no one. This is why the "who watches the CCDI" question and the succession question are the same question (CN-O-01): the discipline state is coherent exactly as long as its apex is singular, and an apparatus built to destroy any official the centre designates would be, in any contested transition, the most valuable and most dangerous prize in the system. Wang Qishan's trajectory is the precedent elite actors will have studied: the campaign's architect was himself eased to the ceremonial vice-presidency in 2018 and into retirement by 2023, his former aides not immune from later investigation [TBD-VERIFY: dispositions of Wang Qishan-associated figures, including the Dong Hong case (his former inspection-team deputy, sentenced 2022)] — a demonstration that in the discipline state, even the sword's wielder holds it at the leader's pleasure.

6. The Human and Systemic Texture — Detention, Fear, Spectacle, and Popularity

6.1 Inside the Investigation

The investigation experience, reconstructed from detainee accounts, defence-lawyer testimony, and the Human Rights Watch documentation, follows a recognisable script. The subject is taken — often from a meeting he was summoned to, the colleagues' first sign being an empty chair — and disappears into liuzhi (formerly shuanggui) custody: a padded, camera-monitored room, lights on continuously, two rotating minders present even during sleep and toilet use (a suicide-prevention protocol that doubles as total surveillance), fixed daily routines, and interrogation sessions that detainee accounts describe as built on exhaustion, repetition, isolation, and the manipulation of hope — the case officers controlling the only information the subject receives about his family, his patrons, and his prospects [TBD-VERIFY: detention-condition details synthesised from HRW (2016), trial testimony in investigator-abuse cases, and journalistic reconstructions; conditions in post-2018 purpose-built facilities are less documented than the shuanggui-era record]. The system's output is the confession: Chinese discipline jurisprudence, like the imperial magistracy before it, treats the written self-criticism and confession (交代) as the case's centrepiece, and the "leniency for confession" structure of both Party regulation and the 2018 Criminal Procedure Law makes resistance nearly irrational once detention begins. Defence lawyers enter only at the transfer stage, when the file is closed; acquittals in transferred cases are effectively unknown [TBD-VERIFY: conviction rates in supervision-transferred prosecutions, reported near 100%].

The era's grimmest statistic is the wave of unnatural deaths of officials — suicides recorded or rumoured among officials under investigation, awaiting it, or fearing it. Chinese media in the campaign's first years tallied dozens of official suicides annually, a visible multiple of the pre-2012 baseline, before such reporting was itself curtailed [TBD-VERIFY: tallies of official suicides 2013–2016 — figures of 120+ in some years circulated in mainland media compilations; no authoritative series exists]. The deaths sit at the system's moral fault line: the official account treats them as guilt's confession ("death to escape punishment," which also conveniently terminates the case and can shield the family's assets [TBD-VERIFY: the case-termination-upon-death rule and its asset consequences]); the critical account treats them as the index of what custody, or its anticipation, does to human beings processed without counsel or court.

6.2 The Chilling and the Reset of Official Life

Below the detention layer, the campaign transformed the official class's everyday culture more thoroughly than any post-Mao policy. The Eight-Point Regulations and the discipline apparatus's lifestyle enforcement killed the banquet economy within two years: Kweichow Moutai's retail price collapsed from its 2012 peak, premium restaurants closed or pivoted to mass dining, gift-card and mooncake markets shrank, and Macau's gaming revenue — the laundering and entertainment annex of mainland officialdom — fell by more than a third in 2015 [TBD-VERIFY: Macau GGR −34.3% in 2015; Moutai price trajectory]. The deeper change was relational: the dense exchange economy of banquets, red envelopes, festival gifts, and favours through which Chinese officialdom had transacted for decades became legible as discipline risk, and officials responded by transacting less — with businesspeople, with each other, with anyone. Entrepreneurs reported a paradox the political-economy literature has chewed on since: officials became cleaner and less useful, no longer extorting but also no longer facilitating — the "lying flat" problem of §3.4 experienced from the private sector's side.

The campaign also rewrote the official career's risk-return profile. The pre-2012 bargain — modest salary, substantial gray income, near-zero downside — became modest salary, near-zero gray income, catastrophic tail risk. Predictions that talent would flee the civil service proved wrong in the aggregate: civil-service examination applications rose through the 2020s to record multiples of available posts — driven less by the bargain's appeal than by the private economy's post-2021 turbulence (CN-D-01, CN-D-03) — though anecdotal and survey evidence suggests selection effects within the applicant pool, toward security-seekers and away from the entrepreneurial cadre type the promotion tournament was designed to harvest [TBD-VERIFY: civil-service application figures, ~3.4 million registrants for the 2024 national exam; the selection-effect claim is interpretive].

6.3 The Popular Dimension

The campaign's domestic popularity is among its best-documented features and the one external accounts most often underweight. The Harvard Ash Center's longitudinal surveys found satisfaction with central-government anti-corruption performance rising from roughly a third of respondents in 2011 to around two-thirds by 2016 — the steepest improvement in any governance domain measured [TBD-VERIFY: Cunningham, Saich and Turiel (2020) figures of 35.5% (2011) to 71.5% (2016) approval on corruption-fighting]. The popularity is not mysterious: the flies-level enforcement of §3.1 addressed the corruption ordinary citizens actually experience — the clinic envelope, the school placement fee, the village land account — and the tigers-level spectacle satisfied a deep folk narrative of the upright centre punishing wicked magnates, a narrative with centuries of opera and dynastic-history precedent behind it.

The apparatus cultivates participation. The CCDI's website and app run tip lines that channel millions of petitions and reports; whistleblowing rewards and "doorstep corruption" hotlines extend the intake to the village level; and the xunshi teams' arrival in a province is publicised precisely so that petitioners will queue. The campaign is also produced as spectacle: the terse fall announcement on the CCDI website (the formula "suspected of serious violations of discipline and law, currently under disciplinary review and supervisory investigation" — 涉嫌严重违纪违法, the phrase by which China learns a career has ended); the staged perp-walk of the documentary confessions in Always on the Road and Zero Tolerance, where former Politburo-adjacent figures in regulation dark jackets weep on camera; the inventoried grotesquerie of the searches — Gu Junshan's gold boat, the tonnes of cash requiring counting machines that burned out in the Wei Pengyuan coal-official case [TBD-VERIFY: the burned-out counting machines detail, widely reported in 2014 coverage of the Wei Pengyuan cash seizure of ~200 million yuan]. Social media amplifies and occasionally initiates: the pre-campaign falls of "Watch Brother" Yang Dacai (identified by netizens via his luxury watches, 2012) prefigured a participatory dynamic the apparatus has since channelled into its own intake systems while suppressing its uncontrolled forms — independent anti-corruption activism, notably the New Citizens' Movement's asset-disclosure campaigners, was prosecuted in the same years the official campaign accelerated, the clearest possible statement that the discipline state claims a monopoly on accountability.

7. Comparative Perspective — The Party-Discipline Genre and What the CCDI Reveals

7.1 The Leninist Family

Party-discipline organs are a defining feature of the Leninist genus, and the comparative record is mostly a record of failure — which is what makes the CCDI's trajectory analytically interesting. The Soviet prototype — the Central Control Commission of the 1920s, later the Party Control Committee, alongside the state-side Rabkrin that Lenin himself despaired of — decayed in precisely the way the pre-2012 CCDI was decaying: captured by the apparatus it supervised, reduced to processing appeals and minor cases, structurally incapable of touching the nomenklatura elite whose corruption became, by the Brezhnev era, a system-defining fact. Andropov's brief disciplinary revival (1982–1984) and Gorbachev's anti-corruption prosecutions came too late and read, in retrospect, as symptoms rather than cures. The CCP, whose leadership studies Soviet collapse the way other establishments study scripture, drew the explicit lesson that uncorrected elite corruption delegitimises Leninist rule terminally — Xi's early speeches on the Soviet fall paired the corruption diagnosis with the "nobody was man enough to resist" line about the CPSU's dissolution [TBD-VERIFY: Xi's December 2012 Guangdong remarks on the Soviet collapse, reported via internal-speech leaks]. The Xi-era CCDI is best read as the deliberate construction of the discipline organ the Soviet party never had: vertically commanded, elite-capable, permanently funded with the leader's personal authority.

The nearest living relative is Vietnam. Nguyễn Phú Trọng's "blazing furnace" (đốt lò) campaign, run from 2016 through the VCP's Central Inspection Commission and Central Steering Committee on Anti-Corruption, tracked the Chinese template with a lag: thousands of officials disciplined, Politburo members prosecuted (Đinh La Thăng, 2018), and — exceeding even the Chinese campaign's formal reach — two serving state presidents (Nguyễn Xuân Phúc, 2023; Võ Văn Thưởng, 2024) and a National Assembly chair forced from office in anti-corruption-coded resignations [TBD-VERIFY: the Vietnamese campaign's case totals and the characterisation of the leadership resignations]. The parallel is doubly instructive: it shows the discipline-state model travelling within the Leninist family as a system response rather than a Xi idiosyncrasy, and it reproduces the same interpretive dispute — clean-up versus succession politics — that attends the Chinese original, with Tô Lâm's rise from security minister through the campaign to General Secretary (2024) supplying the Vietnamese version of the weaponisation reading.

7.2 The Rule-of-Law Alternative

The standing counter-model is the Singapore CPIB / Hong Kong ICAC route, the East Asian demonstration that clean government can be produced without a Leninist discipline organ. The Corrupt Practices Investigation Bureau (Singapore, 1952) and the Independent Commission Against Corruption (Hong Kong, 1974) share surface features with the CCDI-NSC — independence from the agencies they police, answerability to the apex of government, formidable investigation powers — but differ on the axis that matters: their cases are tried in independent courts under ordinary criminal law, their targets have counsel from arrest, their jurisdictions are defined by statute rather than by the ruling organisation's internal rules, and they are flanked by complementary institutions (competitive official salaries, asset-disclosure regimes, a free or semi-free press in the Hong Kong case) that the Chinese system has declined. Melanie Manion's Corruption by Design drew the canonical contrast a decade before the Xi campaign: Hong Kong converted a campaign moment into institutions that made integrity self-enforcing, while mainland anti-corruption remained trapped in the campaign cycle — enforcement waves that alter behaviour while they last and require permanent repetition because they change incentives rather than structures. The Xi-era answer to Manion is, in effect, that the campaign has been made the institution: permanence by serialisation (§3.3) rather than by the rule-of-law lock-in China rejects on principle (CN-I-01 §12). Whether a permanent campaign is a stable equilibrium or a contradiction in terms is the open question the comparison sharpens.

7.3 The Scholarship and the System-Revealing Question

The authoritarian-anti-corruption literature organises the CCDI's record around two findings. First, the loyalty-integrity trade-off (Egorov and Sonin's formalisation): autocrats prefer loyal agents to competent or honest ones, because the competent and honest are dangerous, and tolerated corruption functions as both compensation and leash — the kompromat equilibrium. Authoritarian anti-corruption campaigns are therefore self-limiting: pressed to completion they would destroy the loyalty machine. Christopher Carothers's comparative study qualifies the pessimism: authoritarian corruption control can succeed durably — Singapore, Taiwan and South Korea in their authoritarian phases, arguably post-2012 China — but only where the top leader is personally committed and institutionally unconstrained, which entangles clean government with concentrated power by construction. Second, the campaign-versus-institution debate (Manion, Wedeman, Ang): campaigns suppress; institutions prevent; China has built the largest suppression apparatus in history while withholding the prevention package (asset disclosure, free press, independent adjudication) because the prevention package is, in Leninist terms, regime-threatening. Both findings converge on the same structural sentence: in this system, anti-corruption capacity and personal power are the same capacity.

That convergence is what the CCDI's centrality ultimately reveals about the system, and it closes the loop with CN-I-01's architecture analysis. A party-state that vests all supervision internally must build an organ like the CCDI or rot like the CPSU; an organ like the CCDI, vertically commanded and exempt from external check, is usable only by — and therefore strengthens only — the apex; and an apex strengthened by such an organ faces ever-weaker institutional restraint on its own errors. The discipline-inspection state is thus the purest expression of the Leninist adaptation the China corpus documents: the system solving its genuinely existential corruption problem with an instrument that deepens its other existential problem, the unconstrained and unsuccessable centre.

8. Conclusion — The Permanent Campaign and Its Open Questions

Between 2012 and 2026 the Central Commission for Discipline Inspection travelled further than any other institution in the Chinese system: from a captured watchdog, structurally blinded by dual leadership and operating its signature instrument outside any law, to a constitutionalised fourth branch of the state with jurisdiction over every person exercising public power in China. The Wang Qishan reforms — verticalisation, the weaponised xunshi system, full-coverage stationing — supplied the machinery; the 2018 National Supervisory Commission supplied the constitutional wrapper; and fourteen years of rolling enforcement, from Zhou Yongkang to the hospital directors and Rocket Force generals of the mid-2020s, supplied the demonstration that no office, sector, or rank stands outside the apparatus's reach. The institution this document names the discipline-inspection state is the Xi era's most consequential institutional creation after the recentralised commission system itself (CN-I-01 §4), and the two are halves of one design: the commissions concentrate decision; the discipline state enforces compliance.

The document has held its disputes open, per the corpus's tone discipline. The campaign is a genuine assault on a corruption problem the leadership plausibly judged existential, and the instrument by which a leader destroyed every rival network and entrenched personal rule — the readings are not alternatives but descriptions of the same machine from below and from above. The 2018 reform legalised a lawless practice and constitutionalised detention without counsel — both statements are accurate. The campaign earned genuine popular legitimacy at the flies level and froze the official class into defensive inaction the centre has spent a decade exhorting against. The corpus declines to net these out.

Three questions carry forward. First, sustainability: a permanent campaign whose annual case totals still rise in its fourteenth year has either failed to change underlying incentives or has redefined success as perpetual enforcement — Manion's campaign-versus-institution critique remains unanswered on its own terms. Second, the paralysis trade: whether a bureaucracy governed by investigation risk can deliver the initiative the system's economic predicaments (CN-D-01, CN-D-03) demand is the discipline state's standing collision with the developmental state. Third, and weightiest, the succession entanglement (CN-O-01): the apparatus answers, through the CCDI secretary, to the General Secretary alone; it holds files on every consequential official in China; and it has no doctrine for the day its principal is no longer singular. The self-supervision paradox the institution embodies — the Party as its own permanent auditor, audited by no one — is resolved in practice by the person of the leader, which is to say it is not resolved at all. The CCDI's history from 1949 already contains the warning case: an organ created to discipline the Party by rule was once dissolved entirely so that discipline could be exercised by a leader against the organisation itself. The discipline-inspection state of 2026 is that history's deliberate inversion — and its ultimate test, like the system's, is the transition it has not yet had to survive.