Analysis
Estimating Death and Persecution during the Cultural Revolution
Comparing gazetteers, internal investigations, local archives, and victim definitions.
What happened before the analysis
The Cultural Revolution, 1966–1976
From central mobilization, Red Guards, and factional conflict to military restoration, purges, rehabilitation, and memory management.
Read the documented chronologyContents
No national victim list followed the end of the Cultural Revolution
The Cultural Revolution ended in 1976. The CCP later rehabilitated large numbers of cases and investigated severe violence in some localities. The Central Committee never released a national list of deaths that families and researchers could verify person by person. Nor did it open a complete case database covering every province and county, the military, public security, revolutionary committees, and work units. Researchers face records that are dispersed and incomplete.
Some county gazetteers provide death totals for a factional battle, massacre, or purge. Some say only that an event caused casualties. Others omit the most sensitive local violence entirely. Post-Mao investigations gathered more material, but much of it remained for internal Party use. Family testimony and private victim lists can fill gaps in official accounts, but they have limited geographic coverage too.
Cultural Revolution death estimates are therefore not the sum of entries in one national table. Scholars collect as many local records as possible, identify places that report figures and places that do not, and estimate nationwide results while accounting for population, events, and sample bias. A credible total must state its evidence and method.
A study of 2,213 local annals estimates 1.1 to 1.6 million deaths
Andrew Walder conducted a nationwide study of rebellion and repression from 1966 through 1971 using historical narratives in 2,213 county and city annals. The study did more than add published figures. It compared gazetteer totals with internal investigation reports and used sample selection models to address selective reporting. [8]
The study estimates 1.1 to 1.6 million deaths from 1966 through 1971 and 22 to 30 million direct victims of some form of political persecution. The two ranges must not be added together. Death is the most severe outcome, while direct persecution is a broader category that includes people who survived. [8]
The research focuses on the first five years after the campaign began because nationwide insurgency, factional conflict, military intervention, and organized repression were concentrated in that period. Its numbers are not an official CCP total or a precise count based on completed individual registration. The range reflects missing reports in gazetteers, differences in internal material, and uncertainty in the statistical model.
That uncertainty has limits. The range from 1.1 to 1.6 million is not a guess picked at will. It is the result of different corrections applied to the same large body of local evidence. Presenting the midpoint as a certain death toll would be careless. Denying mass death because the estimate is not precise to the last person would be equally hostile to the evidence.
Persecution totals depend on which acts the study includes
Death is comparatively simple to define, but causes still require separation. Combat deaths in factional fighting, targeted mass killing, organized execution, death from torture or custody, and suicide under political pressure involve different perpetrators and different evidence. A single total can conceal the difference between actions by state organs, mass groups, and individuals.
Persecution is more complicated. A study may include detention, investigative confinement, assault, home raids, dismissal, forced labor, political labeling, and other direct punishment. Whether relatives affected by a family member's verdict count as direct victims depends on the definition. Adding every person sent to the countryside, deprived of education, or subjected to generalized political fear would destroy comparability.
Walder's estimate concerns direct political persecution, so 22 to 30 million must not be described as everyone affected by the Cultural Revolution. Nearly the whole society experienced the campaign, but that does not place every person in the same victim category. A strict definition does not excuse the CCP. It tells the reader what the number measures.
Authorities caused most casualties while rebuilding control
The most familiar image of Cultural Revolution violence is the Red Guards attacking teachers and homes in 1966. Those attacks were real and fatal. The nationwide gazetteer study nevertheless finds that the vast majority of casualties resulted from repression by authorities rather than insurgent action. [8]
That finding shifts the center of responsibility. After 1967, the military intervened in local conflicts, revolutionary committees took power, and campaigns such as the Cleansing of the Class Ranks followed. Victorious factions, military officers, restored cadres, and new organs of power could use organizational investigation, public security, and detention facilities against defeated groups. Violence did not end with the restoration of order. It moved into state institutions.
National death estimates therefore cannot be explained only through Red Guard brutality. The CCP first authorized attacks on enemies inside the Party and society. It then allowed the military and reconstructed authorities to decide which former rebels had become counterrevolutionaries. In factional fighting and state repression alike, the CCP created the political categories and field of action.
Guangxi shows how national averages conceal extreme local violence
Death rates and forms of persecution varied sharply among provinces and counties. Factional structure, military alignment, local leadership, ethnic relations, and the timing of purges all changed outcomes. A national range describes overall scale but cannot replace local history.
The regional civil war and subsequent repression in Guangxi form one of the most severe local cases. Researchers have used local material to reconstruct armed conflict, changes in power, and the punishment of defeated groups. Official inquiries conducted from 1983 through 1987 also left internal records far deeper than the account in an ordinary gazetteer. [10] [11]
The Guangxi evidence matters in two ways. It shows that local publications can omit casualties and responsibility visible only in internal inquiry. It also warns against treating an extreme region as the national average. Credible estimation must prevent silence in ordinary localities from producing an underestimate without mechanically projecting the worst case onto the whole country.
The CCP acknowledges disaster but withholds a verifiable national account
The CCP's 1981 historical resolution acknowledged that the Cultural Revolution caused severe disaster and large numbers of unjust cases. It assigned Mao primary responsibility. [1] The document supplied a political judgment but no national total of deaths or persecution and no case-level evidence supporting such a total.
The CCP possesses the material needed to improve precision: public security and military records, special case files, local post-Mao investigations, organizational penalties, and death registration. It decides which archives open, which material local histories omit, and which research can be published. The persistence of a numerical range reflects more than natural loss of historical records. It also reflects a regime refusing external verification.
That uncertainty is politically useful to the CCP. A lower number lets it present the catastrophe as a limited collection of local mistakes. A higher number can be dismissed as politicized or unsupported by complete archives. The archives are incomplete in public precisely because the accused party controls them.
An evidence range does not weaken responsibility for state violence
Auditing a death toll rigorously is not an effort to find a politically acceptable upper limit for the CCP. Every estimate should identify its years, geographic coverage, victim definition, sources, and method of extrapolation. Studies that count different harms should appear separately, not be added merely to produce a larger number.
The same rigor applies to responsibility. The large gazetteer study provides both a range and a finding that authorities caused most casualties. The CCP launched the movement, classified enemies, deployed the military, rebuilt organs of power, and controlled the later investigation and archives. The precision of the toll remains disputed. The institutional chain is much clearer.
Ending the dispute would require the CCP to open central and local archives, give families access to records, permit independent replication, and allow courts outside Party control to investigate individual deaths. The CCP refuses those conditions while demanding that the public accept only its authorized historical judgment. The main obstacle to counting the dead is no longer statistical technique. It is the fact that the party responsible for the catastrophe still monopolizes the evidence.
What the available sources establish
County-based research estimates roughly 1.1–1.6 million deaths and 22–30 million people subjected to political persecution, with regional and classificatory uncertainty.
Sources
Resolution on Certain Questions in the History of the PRC since 1949primary-recordUnavailable
Chronology of One Hundred Years of the CCPprimary-recordUnavailable
Library of Congress China Country Studygovernment-reportLive
CCP Central Committee May 16 Circularprimary-recordLive
Sixteen Points on the Great Proletarian Cultural Revolutionprimary-recordLive
Bombard the Headquarters—My Big-Character Posterprimary-recordLive
UK National Archives Resource on the Cultural Revolutiongovernment-reportLive
Rebellion and Repression in China, 1966–1971academic-researchRedirected
The Political Legacy of Violence during China's Cultural Revolutionacademic-researchRedirected
Anatomy of a Regional Civil War: Guangxi, 1967–1968academic-researchRedirected
Official Historical Inquiries into Guangxi's Cultural Revolutionacademic-researchRedirected
Economic Legacies of the Cultural Revolutionacademic-researchLive
The Early Cultural Revolution, 1966–1968academic-researchRedirected
Demobilization and Restoration in the Late Cultural Revolutionacademic-researchRedirected
Harvard Fairbank Center Archive on Cultural Revolution Big-Character Postersacademic-researchLive
Foreign Relations during the Cultural Revolutionacademic-researchRedirected
The Chinese Cultural Revolution in the Cambridge History of Communismacademic-researchRedirected