The Great Leap Forward and the Great Chinese Famine
The state raised the grain figures on paper, then requisitioned the harvest that was never there
Open
You think this began with a bad harvest.
It began with a steel target. In May of nineteen fifty eight, Mao Zedong told the party that China would overtake Britain in steel output within fifteen years. Then he cut the timeline to three.
The nineteen fifty seven target of five point three five million tonnes became ten point seven million tonnes. About one million backyard furnaces went up across the countryside. Peasants melted their woks, their hoes, their door hinges into iron.
A tonne from one of those furnaces cost roughly twice a tonne from a modern mill, and most of what came out of them was brittle and unusable. The villagers who fed those fires were pulled off the harvest to do it.
Scene 1The commune
Autumn of nineteen fifty eight, in a newly formed people's commune.
Seven hundred fifty thousand collectives have just been merged into about twenty five thousand of these, each one averaging five thousand households. By the end of that year, something like six hundred million rural Chinese have been folded into communes like it. Private plots are gone. Private kitchens are gone. Even the cooking pot goes to the commune now.
For a few months, in the mess hall, the food is free and there is as much of it as anyone wants. Cadres tell people to eat their fill, because the collective will always provide.
That same year, the state also declares war on sparrows. Villagers bang pots until the birds drop from exhaustion in the air. With the sparrows gone, locusts and insects multiply and strip the fields bare.
Scene 2The ledger
The numbers on paper stop matching the grain in the ground. Local cadres compete to report the largest harvest, and are afraid to report the smallest.
Some claim ten thousand catties per mu, on land that yields a few hundred. The state sets its procurement quota against those invented figures.
When the real harvest cannot meet the fictional one, work teams go from house to house. They search homes, dig up floors, and beat peasants to find grain that does not exist.
In March of nineteen fifty nine, Mao Zedong told a party conference in Shanghai what the shortage meant to him. The remark survives as quoted in Dikotter's history of the famine, not as a transcript.
· Remark at a Shanghai party conference · March 1959
“When there is not enough to eat, people starve to death. It is better to let half of the people die so that the other half can eat their fill.”
The famine ran two more years after that remark. Grain requisitioning continued at the same pace, from villages that had already given up their reserves.
Scene 3Xinyang
Xinyang prefecture, Henan province. On January the thirteenth, nineteen sixty one, the district reports its own death count for the year before: one million seventy thousand three hundred twenty one people, dead between November of nineteen fifty nine and October of nineteen sixty, thirteen point two percent of everyone who lived there.
Investigators sent in afterward find the granaries still holding grain. They also find evidence of cannibalism, and a deliberate cover up.
Scene 4Lushan
July the fourteenth, nineteen fifty nine, a party conference at the mountain resort of Lushan. Defense Minister Peng Dehuai hands Mao a private letter, cataloguing the waste and the exaggeration he has seen touring his home province.
On July the twenty third, Mao circulates that letter to the whole conference. Peng is denounced as the head of an anti party clique, alongside Huang Kecheng, Zhang Wentian, and Zhou Xiaozhou, and stripped of his post. The Anti Right Deviation campaign that follows purges officials who told the truth, and the same policies that are killing people run for two more years.
Scene 5Yang Jisheng's father
Nineteen fifty nine. A message reaches a young reporter at the Xinhua News Agency: his father is starving, back home in the Hubei countryside.
Yang Jisheng goes home. He finds the old man too weak to eat. His father dies days later.
For decades Yang believes the official account of natural disasters. Then, with the access of a senior journalist, he spends years inside provincial archives and party records.
What Yang Jisheng found became a book, published in Hong Kong in two thousand eight. He titled it Tombstone, and said why.
· Tombstone, The Great Chinese Famine 1958-1962 · 2008
“I named the book Tombstone. It is a tombstone for my father who died of hunger in nineteen fifty nine, for the thirty six million Chinese who died of hunger, for the system that caused their deaths, and perhaps for myself for writing this book.”
The book was banned on the mainland. Its estimate of thirty six million dead sits inside a much wider argument that is still not settled.
Five accounts
Five accounts, and the argument here is not whether people died. It is about why, who is to blame, and how many. The lowest estimate on record is eleven point five million. The highest is forty five million. Both numbers describe the same famine.
victorOfficial Party / State Narrative
The Party's Own Account
The party's case rests on two words: natural disaster. Drought and flood struck large parts of the country, and after Moscow withdrew its advisers in nineteen sixty, the debt owed to the Soviet Union added its own pressure. On this account the leadership made errors of enthusiasm, not a crime of policy.
In nineteen eighty one, under Deng Xiaoping, the party did revise its own story. It retired the phrase Three Years of Natural Disasters for Three Years of Difficulty, and it named Mao responsible for leftist errors, while preserving his standing.
“Comrade Mao Zedong's contributions to the Chinese revolution far outweigh his mistakes. His merits are primary and his errors secondary.”
That resolution is still the party's last word. The exact death toll remains a state secret, and county mortality records stay classified. The figure most often attached to the party's own account, when one is given at all, sits near fifteen million.
Read the full account
Narrative
The Communist Party's account of the catastrophe was fixed at the moment it began and has been managed ever since. In the early 1960s the years 1959 to 1961 were named the Three Years of Natural Disasters. The framing rested on two external causes: a run of drought and flood that struck large parts of the country, and the pressure to repay debts to the Soviet Union after Moscow withdrew its advisers and technical aid in 1960 following the Sino-Soviet split. In this telling the leadership made errors of enthusiasm, but the deaths were a tragedy of nature and of a hostile ally, not a crime of policy.
The party never denied that hardship occurred, and it did adjust the account over time. The 1981 Resolution on Certain Questions in the History of Our Party, adopted under Deng Xiaoping, acknowledged that the Great Leap Forward and the Anti-Right Deviation campaign had caused serious losses, and quietly retired the phrase Three Years of Natural Disasters in favor of the Three Years of Difficulty. The resolution assigned Mao chief responsibility for the leftist errors of the period while preserving his overall standing, summarized in the formula that his contributions were primary and his mistakes secondary.
The state has treated the specific death toll as a state secret and the higher figures as hostile propaganda. Detailed county mortality records were classified. Independent estimates in the tens of millions were dismissed as exaggerations by foreign scholars and dissidents seeking to discredit the socialist system and the party's legitimacy. Yang Jisheng's Tombstone, published in Hong Kong in 2008, was banned in the mainland, and its author was barred from traveling abroad to accept journalism prizes. Textbooks describe a period of economic difficulty and correction, not a famine of forty million.
The narrative's function is continuity. To concede a deliberate, ideology-driven famine of tens of millions would place the founding leader and the party's method at the center of the deadliest peacetime disaster in Chinese history. The euphemism, the classification, and the ban are not separate acts. They are one long project to keep the catastrophe describable only in terms the state chooses.
Arguments
- The famine resulted from natural disasters and the burden of repaying Soviet debt after 1960
- The 1981 resolution acknowledged serious policy errors while preserving Mao's overall standing
- Death-toll figures in the tens of millions are exaggerations advanced by hostile foreign scholarship
- Detailed mortality records are matters of state security, not public history
Sources
- Resolution on Certain Questions in the History of Our Party Since the Founding of the People's Republic of China
- A century of the Communist Party of China, the Great Leap Forward
academicThe Man-Made Verdict / Historians
The Historians' Verdict
The historians who worked the archives reject both official excuses. Weather records do not explain a catastrophe of this size and spread. And the debt explanation collapses against the trade figures: China exported grain throughout, four point one six million tonnes in nineteen fifty nine alone, the deadliest year.
One county level study found that nineteen sixty excess deaths would have been about twelve percent lower, had the nineteen fifty nine exports simply matched the nineteen fifty seven level.
Frank Dikotter's provincial archive research reframed the famine as more than starvation.
“Coercion, terror and systematic violence were the foundation of the Great Leap Forward.”
He documented that many deaths came from beatings and deliberate deprivation, not hunger alone. People caught stealing a handful of grain were buried alive or worked to death. The lie about the harvest, in this account, became the warrant to seize the food.
Read the full account
Narrative
Yang Jisheng was a young reporter for the Xinhua News Agency in 1959 when a message reached him that his father was starving. He went home to the Hubei countryside and found the old man too weak to eat. His father died days later. For decades Yang believed the official line about natural disasters. Then, as a senior journalist with access that few others had, he spent years inside provincial archives and party records. What he found became Tombstone, published in Hong Kong in 2008: a documentary reconstruction arguing that the famine was manufactured by policy and lies, and estimating roughly 36 million dead.
The historians who have worked the archives converge on a single verdict against the two official excuses. Yang traveled repeatedly to the national meteorological bureau and compiled the weather records; the drought-and-flood explanation does not fit a catastrophe of this magnitude and geographic spread. The Soviet-debt explanation collapses against the trade figures: China was a net grain exporter throughout, shipping 4.16 million tonnes abroad in 1959 alone, the deadliest year. One county-level econometric study found that 1960 excess deaths would have been about 12 percent lower had 1959 exports merely matched the 1957 level. The grain existed. The state moved it away from the mouths that needed it.
Frank Dikotter's Mao's Great Famine (2010), built from provincial and county archives opened in the 2000s, pushed the estimate to at least 45 million and reframed the event as one of mass violence, not just starvation. He documented that a large share of deaths were not from hunger alone but from beatings, torture, and deliberate deprivation inflicted by cadres enforcing quotas. People who stole a handful of grain were buried alive, mutilated, or worked to death. The famine, in this reading, was enforced.
The mechanism the historians describe is a closed loop of ideology and falsified numbers. The pressure to report ever-larger harvests produced fantastical yield claims. Procurement quotas were set against the fantasy. When the real crop could not satisfy the fictional one, the shortfall was treated as concealment by peasants rather than error by the state, which licensed the search teams and the violence. The lie about the harvest became the warrant to seize the food. This is why the historians reject the word accident. A drought does not file a false report. A flood does not set a quota. The catastrophe was administered.
Arguments
- Weather records do not account for a famine of this scale, and China exported grain throughout
- Inflated yield reports set procurement quotas that requisitioned grain that did not exist
- A large share of deaths came from beatings and deliberate deprivation, not hunger alone
- The estimates cluster from about 36 million (Yang) to at least 45 million (Dikotter)
Sources
- Tombstone, The Great Chinese Famine 1958-1962
- Mao's Great Famine, The History of China's Most Devastating Catastrophe, 1958-1962
- Hungry Ghosts: Mao's Secret Famine
vanquishedThe Peasants Who Starved
The Peasants Who Starved
The commune took the cooking pot first, and for one season the mess hall gave people more food than they had ever eaten. By the spring of nineteen fifty nine it was serving thin gruel, then leaves, then nothing, and no household had a private pot or a private plot left to fall back on.
The grain that was harvested left the village anyway. When the sacks a village could produce fell short of the inflated quota, the shortfall was called hoarding.
An account from Xinyang prefecture, unnamed in the record, described what the search teams found there.
“The granaries were full and the people were dying at the doors.”
Villagers caught gleaning fallen grain, or trying to flee, were labeled thieves. A family's survival could depend on the temper of one local official.
Read the full account
Narrative
The commune took the cooking pot first. When the mess halls opened in the autumn of 1958, peasants were told that private kitchens were finished and that the collective would feed everyone. For a season the food was free and the portions were large, and cadres urged people to eat their fill because communism had arrived. By the spring of 1959 the mess halls were serving thin gruel, then leaves, then nothing. Households had no pots, no private plots, and no reserves, because all of it had been surrendered to the commune that could no longer feed them.
The grain that was harvested left the village. Procurement teams arrived with the quota figures derived from the inflated reports, and when the sacks the village could produce fell short, the shortfall was called hoarding. In Xinyang prefecture in Henan, one of the worst-hit areas, the district later reported that 1,070,321 people died between November 1959 and October 1960, more than one in eight residents. Investigators sent afterward found that the granaries had held grain the whole time. They also found evidence of cannibalism, and of local officials who had beaten peasants to death to stop word of the deaths from reaching Beijing, and who blocked the roads so the starving could not leave to beg.
The violence at the village level is central to the survivors' account. Cadres who enforced the quotas held near-absolute power over food, and therefore over life. Peasants caught gleaning fallen grain, digging up seed, or fleeing were labeled thieves and enemies. Punishments recorded in the archives include being tied up and frozen, denied the mess-hall ration, buried alive, and having ears or noses cut off. A family's survival could depend on the temper of a single sector chief.
These deaths were largely undocumented as individuals. Rural households did not register every death, and the state had reasons not to count. The people who starved left oral testimony gathered decades later, and gravestones that in many villages do not exist. Yang Jisheng titled his book Tombstone in part for this absence: the tens of millions who died in the countryside were buried in unmarked ground, in a silence the state preferred, their number itself becoming the object of a dispute conducted by everyone except them.
Arguments
- The commune abolished private kitchens and reserves, leaving households with no fallback
- Grain was requisitioned and exported while granaries held stock the starving could not reach
- Cadres wielded lethal power over rations and punished gleaning and flight as crimes
- The rural dead were largely uncounted, their number contested by others
Sources
- Tombstone, The Great Chinese Famine 1958-1962
- Mao's Great Famine, The History of China's Most Devastating Catastrophe, 1958-1962
- Agency and Famine in China's Sichuan Province, 1958-1962
revisionistThe Contested Toll / Demographers and Revisionists
The Contested Toll
Almost no one was counted at the time, so every figure is a reconstruction, built on an assumed normal death rate subtracted from what the records show later. A different baseline moves the number by tens of millions.
The demographers converge near thirty million. Ansley Coale found about sixteen point five million. Basil Ashton and colleagues found about thirty million. Judith Banister found about the same. Cao Shuji's county gazetteer study reached thirty two point five million. Former official Chen Yizi's own investigation concluded about forty three million. Yang Jisheng and Frank Dikotter go higher again.
The economist Utsa Patnaik disputes the whole exercise.
“The estimate of thirty million deaths depends on a benchmark death rate lower than that prevailing in India at the same time.”
She and others put the excess closer to eleven point five million. Mainstream historians answer that county archives independently confirm mass death, whatever baseline is chosen.
Read the full account
Narrative
The single most disputed fact about the famine is how many died, because almost no one was counted at the time. Every estimate is a reconstruction, built by subtracting an assumed normal death rate from the deaths that appear in later census and registration data. The answer depends entirely on the baseline chosen, the reliability of the underlying registration, and how missing births are handled. This is why honest scholars produce numbers that differ by a factor of four, and why bad-faith actors on both sides can find a figure to suit them.
The mainstream demographic work sits near 30 million excess deaths. Ansley Coale (1981) estimated about 16.5 million; Basil Ashton and colleagues (1984) about 30 million; Judith Banister (1987) about 30 million, with roughly 29.9 million abnormal deaths. Chinese scholars using more granular records pushed higher: Cao Shuji (2005), compiling county and city gazetteers, reached 32.5 million; the former official Chen Yizi's investigation concluded about 43 million. Yang Jisheng's archival work gave roughly 36 million, and Frank Dikotter's, at least 45 million. These figures are not arbitrary. They rest on the collapse of registered population between 1958 and 1961 and on the sharp fall in births.
A revisionist current disputes the whole exercise. The economist Utsa Patnaik argues that the widely cited 30 million rests on comparing famine-year mortality against an unusually low pre-famine death rate of about 12 per thousand, a rate China had only just achieved and one lower than contemporary India's. Measured against a more conservative baseline, she and others put the excess nearer 11.5 million. Joseph Ball contends the death-rate data released under Deng Xiaoping cannot be trusted precisely because Deng was waging a political campaign against Mao's legacy. The demographer Sun Jingxian has argued that much of the apparent population loss reflects double-counting and the correction of household-registration errors as people moved, not deaths.
Mainstream historians answer that the revisionist baseline is itself a choice that minimizes, that county-level archives and gazetteers independently corroborate mass death regardless of the national registration figures, and that Sun's migration argument cannot explain the physical evidence of famine in the archives. The dispute is not settled, and the platform does not pretend it is. What is established is that a catastrophic famine occurred and that the state concealed its scale. What remains genuinely contested, and politically loaded on every side, is the final number, which ranges across an interval wide enough to contain the entire population of several nations.
Arguments
- Every estimate is a reconstruction whose result depends heavily on the chosen baseline mortality
- Mainstream demographers (Coale, Ashton, Banister) cluster near 30 million; Cao and Dikotter go higher
- Revisionists (Patnaik, Ball, Sun Jingxian) argue the baseline was too low or the data unreliable, yielding as few as 11.5 million
- The famine itself is established fact; the exact toll is honestly and heatedly contested
Sources
- Famine in China, 1958-61 (Population and Development Review)
- China's Changing Population
- On Measuring Famine Deaths: Different Criteria for Socialism and Capitalism?
vanquishedThe Purged Reformers
The Purged Reformers
Peng Dehuai had commanded Chinese forces in Korea. When he toured his home province and saw the mess halls failing, he wrote to Mao rather than staying silent.
In a line attributed to him, on the reasoning behind that letter:
“Putting politics in command is no substitute for economic principles.”
Mao read the letter as a challenge, not a warning, and inverted the conference into a purge. The correction the letter asked for became an escalation, and the famine ran two more years past Lushan.
Liu Shaoqi delivered, in nineteen sixty two, the reckoning Peng had been destroyed for attempting three years earlier.
Read the full account
Narrative
In July 1959 the leadership gathered at the mountain resort of Lushan, ostensibly to cool the fever of the Leap. Peng Dehuai, the Defense Minister and a marshal of the revolution who had commanded Chinese forces in Korea, had toured his home province of Hunan and seen the mess halls failing and the fields neglected. On July 14 he handed Mao a private letter. It praised the general line but catalogued the waste of the backyard furnaces, the exaggeration in the reports, and the petty-bourgeois fanaticism that had substituted enthusiasm for economic sense.
Mao read it as a challenge to his authority. On July 23 he circulated the letter to the whole conference and demanded that the delegates take sides. Officials who had come to Lushan expecting a correction of the Leap watched the agenda invert. Peng was denounced as the head of an anti-party clique, together with Huang Kecheng, Zhang Wentian, and Zhou Xiaozhou. He was stripped of his position as Defense Minister. The conference then launched an Anti-Right Deviation campaign that reached down through the ranks, punishing cadres who had reported shortfalls or doubted the targets, and rehabilitating the radical policies. The correction became an escalation, and the famine deepened for two more years.
The lesson that internal criticism was fatal held until the deaths could no longer be denied. In January 1962, at a conference of seven thousand cadres, head of state Liu Shaoqi delivered the reckoning Peng had been destroyed for attempting. He told the assembly that in Hunan the peasants said the disaster was three parts natural and seven parts man-made. Mao, present, absorbed it in silence and resentment. He retreated from day-to-day management but not from the conviction that his enemies inside the party had used the famine against him.
The reformers did not survive that conviction. In 1966 Mao launched the Cultural Revolution, in part to overturn the verdict of the seven thousand cadres and to destroy the men who had delivered it. Liu Shaoqi, denounced as the leading capitalist-roader, was beaten, denied medicine, and died in a cell in 1969, his death kept secret for years. Peng Dehuai, dragged back from internal exile, was interrogated and abused until he died in 1974. The two officials who had tried to name the catastrophe were themselves erased, and the naming was deferred for a generation.
Arguments
- Peng Dehuai's July 1959 letter documented the waste and exaggeration of the Leap
- Mao inverted the Lushan conference into a purge and an Anti-Right Deviation campaign that intensified the policies
- Liu Shaoqi's 1962 seven-parts-man-made verdict was the reckoning Peng had been purged for attempting
- Both men were destroyed in the Cultural Revolution Mao launched partly to reverse that verdict
Sources
- The Tragedy of Lin Biao and Peng Dehuai (studies of the Lushan Conference)
- Tombstone, The Great Chinese Famine 1958-1962
- Peng Dehuai, memoirs of a Chinese marshal
The turn
What each account cannot see from where it stands.
Official Party / State Narrative
The party calls the deaths a natural disaster while it keeps the county mortality records classified, as if nature needed sealing away.
From the record
Inflated grain reports and quota-based requisition of nonexistent harvestsGrain exports continued through the peak famine yearsThe purge of Peng Dehuai and the intensification of the policies after LushanThe scale of excess mortality documented in classified county recordsThe Man-Made Verdict / Historians
The historians, so careful about the mechanism, say less about the peasants who enforced the quotas out of their own fear, not just cruelty.
From the record
The genuine severity of localized drought and flood in some regionsMethodological disputes over the demographic baseline used to compute excess deathsThe role of lower cadres acting under duress and fear of denunciationThe peasants' account has the least reason to lie, and it says nothing about the debates in Beijing that set the quota in the first place.
From the record
The national policy debates and leadership struggles in BeijingStatistical and demographic framing of the catastropheComparative and geopolitical context of the famineThe Contested Toll / Demographers and Revisionists
The revisionists, so careful about the baseline, spend little time on the archival evidence that the low estimates have to explain away.
From the record
The lived experience behind the statisticsThe political motives that can drive minimizing as well as maximizing the tollThe corroborating physical and archival evidence that constrains the low estimatesAnd the purged reformers' account, about two men punished for telling the truth, says almost nothing about the tens of millions who never got the chance to speak at all.
From the record
The broader rural experience of the famine below the leadership levelThe demographic dispute over the total number of deadThe international and Sino-Soviet dimensions of the period
In January of nineteen sixty two, at a conference of seven thousand cadres, head of state Liu Shaoqi told the party what he had heard in Hunan.
· Speech to the Seven Thousand Cadres Conference, Beijing · January 1962
“In Hunan, the people say that three parts were natural calamity and seven parts were man-made disaster.”
Mao heard it as an accusation, not an admission. Four years later he launched the Cultural Revolution. Both Peng Dehuai and Liu Shaoqi died in its custody.
Close
Tombstone was published in Hong Kong in two thousand eight. It has never been printed on the mainland.
When Yang Jisheng was invited abroad to accept a journalism prize for writing it, the government would not let him travel.
The record
By the numbers
- Span
- 1958-1962peak famine 1959-1961
- Killed
- 11.5-45Mdisputed; revisionist low ~11.5 million, Beijing ~15…
- Displaced
- ~600Mrural Chinese reorganized into 25,000 people's communes…
- Place
- China
Key figures
- Mao ZedongChairman of the Communist Party of China; architect of the Great Leap Forward1893 to 1976
- Peng DehuaiDefense Minister who criticized the Leap at Lushan in July 1959 and was purged1898 to 1974
- Liu ShaoqiHead of state who told the 1962 cadres conference the famine was seven parts man-made; died in Cultural Revolution custody1898 to 1969
- Yang JishengFormer Xinhua journalist and Yanhuang Chunqiu editor; author of Tombstone, whose father starved in 19591940 to ?
- Frank DikotterHistorian whose Mao's Great Famine used provincial archives to estimate at least 45 million dead1961 to ?
- Deng XiaopingParamount leader under whom the 1981 party resolution recast the famine as Three Years of Difficulty1904 to 1997
Primary sources
· Speech to the Seven Thousand Cadres Conference, Beijing · January 1962
“In Hunan, the people say that three parts were natural calamity and seven parts were man-made disaster.”
· Tombstone, The Great Chinese Famine 1958-1962 · 2008
“I named the book Tombstone. It is a tombstone for my father who died of hunger in 1959, for the 36 million Chinese who died of hunger, for the system that caused their deaths, and perhaps for myself for writing this book.”
· Mao's Great Famine, The History of China's Most Devastating Catastrophe · 2010
“Coercion, terror and systematic violence were the foundation of the Great Leap Forward.”
· Remark at a Shanghai party conference, quoted in Dikotter, Mao's Great Famine · March 1959
“When there is not enough to eat, people starve to death. It is better to let half of the people die so that the other half can eat their fill.”
Threads
- The Chinese Cultural RevolutionStudents beat their teachers to death with the textbooks they’d been taught from.
- The Korean WarThe US dropped more bombs on Korea than it used in the entire Pacific war. The armistice was signed in 1953. No peace treaty has ever followed.
- The Chinese Civil War and the Founding of the People's RepublicThe peasant army that won the civil war in 1949 put Mao in the unchallenged power that let him order the Leap nine years later. The same promise of land that carried the countryside to the Communists became the commune that took the harvest away