Tyranny of the Ladle

There is no doubt that the Great Leap Famine in China more than half a century ago was the worst man-made calamity of modern times. Between early 1958 and the spring of 1961, somewhere between 30 and 45 million people died. The more cautious figure is five million more than the population of Scandinavia, the higher one the current population of Spain. The famine was compressed geographically and temporally. Henan, Gansu, Anhui, Guizhou, Qinghai and Sichuan were the worst hit provinces. Most of the victims died between January and April 1960, when all the grain was gone and before the new crops were up. They had tried everything they could to save themselves. They stripped the bark from trees, and ate the roots, runners and leaves of crop plants, as well as insects, bran, cornstalks and rotten sweet potatoes. The rural population of North China had age-old ways of coping with food shortages. They were no strangers to drought, famine and flood; millions had died in an earlier man-made famine in 1938, when Chiang Kaishek decided to slow the Japanese advance by opening the dikes and changing the course of the Yellow River. In the Great Leap Famine, the Entomological Research Institute of China’s Academy of Sciences helpfully supplied lists of the protein and fat content of many insects: dried larva of the corn borer, dried fly maggot, dried dung beetle, termites, locusts and silkworm pupae, together with recipes for their preparation. Before reaching this point, people had eaten the green crops in the fields (chi qing), thereby reducing the harvest that the authorities could seize to negligible proportions. Quite apart from scouring the environment for everything and anything that could be eaten, they did what any starving people would do. They tried to flee, first to the cities, where the food supply was better. For the most part they were stopped on the roads and at transportation hubs by militiamen, who arrested and beat them. If they couldn’t escape, they raided public granaries and storerooms; they stole; they rustled livestock and fodder; they torched the homes of hated cadres; and they rioted.

Finally, they died along the roads, in the fields and at home, the oldest and youngest first and then many of the rest. In October 1960, at last, Mao Zedong was handed a frank report on mass starvation in Xinyang, and within a month investigative teams fanned out to the provinces to document the death toll. By the beginning of 1961, the policies of the Great Leap Forward had been rolled back and grain was being imported from the West.

In local oral memory the hunger was associated more with enforced communal dining in huge militarised communes than with national slogans such as ‘The Great Leap’, and thus ‘peasant time’ was divided into two epochs: ‘when we ate out of the big pot’ and ‘after eating out of the big pot’. In the middle of 1961, when the worst was over, Gansu Province’s first secretary wrote:

The masses deeply detest and loathe the communal kitchens. The masses say: ‘Make friends with a canteen manager and you’ll never want for buns and soup … A knife hangs over the rice ladle.’ The masses say: ‘The communal canteen is a dining hall (a place for getting food), a tribunal (a place where kitchen staff beat and scold people) and a bordello (where team leaders and managers hire the prettiest girls as kitchen staff and mess around with them).’

What he doesn’t mention, surely because it was by now all too well understood, was that after the first two or three months, the communal canteen was no longer a dining hall because there was virtually no food left.

How do we take the measure of such a disaster? Who was responsible? Ralph Thaxton’s Catastrophe and Contention in Rural China, Gail Hershatter’s The Gender of Memory and Chen Yixin’s essay ‘When Food Became Scarce’ all begin with individual experience.* Thaxton shows in meticulous detail how the crisis fractured the village of Da Fo in Henan Province, the desperate struggle to survive, and the abiding resentment among villagers that the cadres most culpable for their suffering were never punished. Hershatter’s volume is devoted to bringing to life women’s experience as it diverged from both men’s history and the national story. Her portraits, based on the lives of 72 women from four villages in Shaanxi Province, are a landmark in women’s history and the history of China; no future account of the famine can ignore them. Chen studies three villages in Anhui, another devastated province, seeking carefully to understand how some villages were effectively extinguished while others managed to limit the death toll.

Mao’s Great Famine by Frank Dikötter and Yang Jisheng’s Tombstone have more ambitious goals. Each aims to understand how the post-liberation party-state created the structures of power and information that made such a massive famine possible. Each describes the stages of the famine as it unfolded and its effects, province by province, based on newly available provincial and national documentary sources and on interviews. Inevitably, each arrives at an overall total of deaths. Dikötter’s estimate, and it is no more than that, of 45 million is larger by 10 million than the estimates of most experts, but it is not preposterous. Finally, and also inevitably, Dikötter and Yang are preoccupied with judging the degree to which all this blood is on Mao’s hands. What did he know and when did he know it? Who set the murderous machinery of grain requisition in motion? When its consequences became apparent, who tried to stop the machine and who simply ignored the toll and pushed ahead?

Dikötter’s answer to this question is implicit in the possessive case of his book’s title. When it comes to a more precise accounting of Mao’s role at each stage of the disaster, Dikötter is considerably more cagey, however, since the available archives leave many questions unanswered. But Mao’s Great Famine is surely the best and most comprehensive general history of the famine published to date. Dikötter is a reliable guide to the utopian origins of the Great Leap, to the export of agricultural commodities to pay for industrial investment at the height of the famine, to the competitive frenzy among provincial cadres to promise unattainable bumper harvests, to the mobilisation of untold millions of people in ill-advised drainage, dam, irrigation and iron foundry projects, to the agro-fantasy that close planting and deep (one-metre) ploughing would raise yields astronomically, to the evolving elite debates on the consequences for the rural population of gigantic communes, collective dining and grain requisitions.

The human costs are not scanted. Dikötter has amassed stories of sadistic cadres beating, humiliating and raping a desperate and exhausted peasantry, of parents abandoning children, and of instances of the last extremity of all great famines, cannibalism. The hundreds of horror stories he piles up have a numbing effect. One has the impression of a mass of disconnected, if horrific detail extracted from his trawl through the archives with very little in the way of historical or cultural context. There are no voices here, as there are in Chen, Hershatter and Thaxton, explaining, in their own words, what happened to them. Here, and quite inadvertently, Dikötter falls into the same sort of error as Mao, who declared in 1956 that China’s two advantages in development were ‘emptiness’ and ‘blankness’. Such flatness is possible only when looking down on the world from a great height. This attitude might explain Dikötter’s scolding tone towards the starving peasants who stole grain: ‘Even when it seemed that petty theft took place against a faceless state, somebody down the chain of distribution paid the price.’

The Chinese edition of Yang Jisheng’s Tombstone caused a sensation when it came out in Hong Kong in 2008 and Yang has been described as the Great Leap Famine’s Solzhenitsyn. Tombstone combines telling details with a rigorous historical reconstruction from oral and written sources, all assembled to allow us to understand how and why the famine occurred. The product of a lifetime of devoted research, Tombstone is a monument. As Yang explains, the book is intended first as a tombstone for his father, who died of starvation in 1959; ‘second … to erect a tombstone for the 36 million Chinese who died of starvation; and third … to erect a tombstone for the system that brought about the Great Famine’.

Yang was putting together his school’s Communist Youth League wall-newspaper in April 1959, when he was visited by a childhood friend from his village of Wanli who urged him to return home with some rice: his father was starving and didn’t ‘even have the strength to strip bark from the trees’. Rushing home, he found things

radically changed. The elm tree in front of our house had been reduced to a barkless trunk and even the roots had been dug up and stripped, leaving only a ragged hole in the earth. The pond was dry; neighbours said it had been drained to dredge for rank-tasting molluscs that had never been eaten in the past. There was no sound of dogs barking, no chickens running about; even the children who used to scamper through the lanes remained at home. Wanli was like a ghost town.

His father was lying in bed, not a grain of rice in the house, waiting to die. In a faint voice, he told Yang ‘to go quickly, go quickly back to school’. Yang fetched water, dug up a few tough peanut sprouts and boiled congee for his father, but he was too weak to swallow and died three days later.

Who killed Yang Jisheng’s father and millions of others is the question that preoccupies both Yang and Dikötter. Both indict Mao for his callous utopianism, his megalomania and the structural brutality of the system he was largely responsible for creating and perpetuating. Comparisons with Hitler and Stalin are as inevitable as they are pointless. While it seems clear that Mao did not intend the deaths of the famine’s victims in the way that Hitler did the deaths of the Jews after Wannsee, he presided over a violent state apparatus whose schemes murdered them as surely as if they had been lined up and shot.

Mao’s utopianism was, one might say, seriously overdetermined. Not only was he caught up in the socialist dream of forced-draft industrialisation pioneered by the Soviet Union but, as with the Russians, his index of progress was heavy industry – in particular the annual production of steel. He pushed to double the output of steel from 5.3 million tons to 10.7 million in a single year, in a bid to catch up with Britain. Despite the resistance to collectives, he relentlessly pushed for huge, militarised communes and collective dining halls, not simply because he thought they were more productive, but because they realised the promise of communism. Inspired by the text of Engels’s speech at Elberfeld, in which he extolled communes, and by the Chinese anarchist-utopian thinker Kang Youwei’s book, The Great Harmony, Mao declared in March 1958, at the beginning of the Great Leap:

In socialism, private property still exists, factions still exist. Families are the product of the last stage of primitive communism, and every last trace of them will be eliminated in the future … Historically, the family has been a production unit, a consumption unit, and a unit for giving birth to the next generation of the labour force and educating children. Now worker families are no longer production units … it is possible that in the future, the family will no longer be beneficial to the development of productivity.

In themselves, such ideas were hardly remarkable; they were shared by thousands of communists, socialists and anarchists throughout the world. It was only when a leader with near-dictatorial power became determined to impose them on a resistant population that they became potentially lethal. Mao was fond of calling his political philosophy ‘one part Marx and one part Qin Shi Huang’, the ambitious and brutal founder of the unified Qin state in 221 BCE. At a party conference in May 1958, Mao urged the emulation of an emperor who had advocated eliminating anyone who used history to criticise the present. When Marshal Lin Biao noted that Qin Shi Huang had burned books and buried scholars, Mao replied: ‘What was Qin Shi Huang? He only buried 460 scholars but we buried 46,000 scholars. During the suppression of counter-revolutionaries didn’t we kill some counter-revolutionary intellectuals? I’ve discussed this with advocates of democracy. “You call us Qin Shi Huang as an insult, but we’ve surpassed Qin Shi Huang a hundredfold.”’

As both Yang and Dikötter show, Mao lived up to this claim at the Lushan Conference in late July and early August 1959. Faced with widespread opposition to the communes and evidence of serious food shortages, most of the delegates favoured extending a tactical retreat that was already underway. It’s inconceivable that Mao, who had condoned a retreat from collectivisation in late 1958, was unaware of the massive problems it had created. The delegates’ misgivings were put deferentially by Peng Dehuai, Mao’s longtime companion in arms and defence minister, but Mao took criticism of the Great Leap as a direct challenge to his leadership and launched a blistering attack on Peng and his allies, branding them a ‘right-deviating, opportunist, anti-party clique’. Once Mao had declared Peng a party traitor, only a handful of brave officials dared risk disgrace by defending him. Yang, in a brilliant piece of detective work, demonstrates that leader after leader on record as agreeing with Peng’s misgivings now fell abjectly into line behind the Great Helmsman: the commodious hall of shame took in Liu Shaoqi, Zhou Enlai, Zhu De, Lin Biao, Tan Zhenlin, Chen Boda and Hu Qiaomu. Nothing better illustrates the fact that, at this fateful juncture, the party hierarchy was less the democratic centralist organisation of the Leninist ideal than a monarchy with courtiers who served at the sovereign’s arbitrary pleasure. Those who persisted in their dissent were purged, from the top all the way down to the township and village level.

When the food minister, anxious to please, promised an excellent harvest, Mao seized the opportunity to push for much larger grain requisitions to fuel industrialisation, even if it meant short rations in the countryside. ‘We can strive for a procurement of 55 billion kilos in the first step from passivity to taking the initiative … Tell the peasants to resume eating chaff and herbs for half the year, and after some hardship for one or two or three years, things will turn around. We should build up our reserves and consume less.’ The imperative of forced-draft industrialisation on the backs of the peasantry was never far from the leader’s mind.

Mao seems, especially in Yang’s scrupulous account, to vacillate between a willingness to believe the most improbable reports, on the one hand, and a suspicion that his ministers were afraid to disagree with him or give him bad news, on the other. Thus in August 1958, on an inspection tour in Hebei, he swallowed the wildly exaggerated harvest figures he was fed and said: ‘How can this county … eat so much grain? What will you do with the extra food?’ He foresaw the surplus being fed to pigs and said: ‘With more food available, we can plant less in the future, and people will only have to put in half a day’s labour.’ Again and again, Mao urged cadres to speak honestly, and when they occasionally did, so long as his leadership was not at stake, he praised them for their forthrightness. Yet he also lamented: ‘Why won’t they tell me the truth? Why?’ His propaganda chief could have told him the reason. ‘When you stand in the middle of the Circular Mound Altar in the Temple of Heaven and call out, echoes come back at you from every direction. But what you hear is still just your own voice.’ As Yang demonstrates, Mao was trapped by the system in an echo chamber of his own devising and, for the most part, these were the sounds he wanted to hear.

The speeches, statistics, claims and hyper-enthusiasms of the early Great Leap give the impression of a collective hysteria. How could so many people believe and propagate such preposterous theories and facts for so long? In retrospect, it was like a giant Ponzi scheme, bound sooner or later to come crashing to earth. Capitalist economies have long experienced speculative bubbles fuelled by the irrational exuberance of a bull market for one asset or another, from tulip mania in the 17th century to the South Sea Bubble, the dotcom bubble and, most recently, the housing bubble. In the case of the Great Leap Forward, the irrational exuberance had several sources. The first was the euphoria of a successful state-socialist revolution. The French, the Russian, the Vietnamese and the Khmer Rouge revolutions all experienced such moments of exuberant possibility. For a while, it is Year Zero, experimentation is rife and achievements seem limitless. The 20th-century socialist revolutions were heir to the high-modernist dreams of Western progress: heavy industry, inventions, massive public works, engineering projects, high productivity and material plenty. In the socialist context this promise was linked to collectivisation and five-year plans.

It wasn’t just magical thinking. Like any speculative bubble, it required some evidence to make it seem plausible. The party-state had substantially transformed China in the ten years between its conquest of power and 1958, and had left the chaos of the civil war and late Republican China behind. The first two five-year plans had achieved high levels of growth in industry and agriculture and the Chinese had fought the United States to a standstill in Korea. For the elite, then, the success of the Revolution, of socialist production and of a resurgent China helped nourish quasi-millenarian hopes. The positive feedback loop fuelling the bubble was amplified by the silencing, by purge, of cadres who were less than completely enthusiastic about the Great Leap. There is nothing particularly exotic about this. During the last stages of the American war in Vietnam, so many of the dissenting civil and military voices in Washington had resigned or been fired that Lyndon Johnson’s White House became an echo chamber for those who thought the war was going well.

Even for many of the famine’s ordinary victims there were reasons initially to suspend disbelief. Hershatter captures the atmosphere: ‘In an era when the collective imagination encompassed electric lights, multi-storey dwellings and socialised food preparation, merely factual reporting of crop output could be construed by the party-state, and by one’s neighbours, as a failure to strive for a future everyone wanted.’ Yang’s father was a believer. Around 1950, as part of the land reform he was given three mu – about half an acre – of land for growing rice. He was so happy to have the status of landowner that he took Yang along when he went to pay his agricultural tax. Three years later, the land was collectivised; he died of starvation a year after the gigantic communes were formed and collective dining enforced.

In 1959 the food gave out: the bubble had definitively burst. As rations were cut, what little enthusiasm remained at local level gave way to resistance, flight, pilfering, theft, the consumption of unripe crops, foot-dragging, misrepresentation, abject fear and sullen compliance. It’s unlikely that those who survived were ever again susceptible to promises of that kind. If we think of the earlier enthusiasm as an epidemic – and it is not clear how deeply it penetrated below cadre level – it was a deadly, ‘virgin-soil’ epidemic, which left those who were lucky enough to survive it with a lifetime of acquired immunity.

The huge communes and collective dining halls that characterised the Great Leap Forward did not last much more than three years before they were dismantled by a combination of fiat and desertion. Though not designed to extinguish lives, the Great Leap Forward had most certainly intended to eliminate the family as a unit of production, as a hearth, and as a ceremonial unit, to be replaced with large, modern, militarised and disciplined formations of labour.

Immense public works projects were an integral part of the Great Leap, the aim being to substitute labour for capital in building a modern productive infrastructure. Though most of the projects were misbegotten and environmentally harmful, they mobilised millions of people for months at a time on massive work sites or steel mills far from home. During an inspection tour in Hunan, Mao’s home province, in 1958, Peng Dehuai was slipped a note by a crippled veteran: ‘The grain is scattered, the potato leaves are withered, and the young and strong are sent to the steel mills, leaving women and children to bring in the harvest; how are we to survive the coming year? Please get word back on behalf of the people.’ Peng wondered later why there had not been more unrest, as there had been in Hungary, which he had visited before the Lushan Conference. He concluded that it was only the ‘decency’ of China’s workers and peasants that prevented the party from having to call out the Red Army.

With the creation of the communes, the Chinese people became subjects of the largest, most radical and most ill-considered social experiment in modern history. The communes were designed as social condensers, to replace most of the functions of the patriarchal family. All land, even the small family plots left after collectivisation, was confiscated. All kinds of petty trade were suppressed. Property, commerce and the working day were now in the hands of the commune and the cadres who ran it. In much of the countryside, the kitchen hearth itself was requisitioned. As a result of the campaign to smelt any available steel and catch up with Britain, villagers surrendered their scrap iron: pots, pans, implements, hinges, buckets, watering cans, old bicycle parts. In many cases the kitchen stove was dismantled too and, in the manic search for building material and wood for fuel, houses and shops were pulled down. Dikötter claims that the destruction represents ‘by far the greatest demolition of property in human history’, with between 30 and 40 per cent of all houses reduced to rubble.

Once again, this experiment aimed at achieving high-modern socialist goals by state-collectivist means. According to Felix Wemheuer’s unpublished manuscript, ‘Dining in Utopia: An Intellectual History of the Origins of the Chinese Dining Halls’, the August 1958 decree establishing the communes specified that they ‘should build public dining halls, infant nurseries, kindergartens, homes of happiness (xinfu yuan) for the elderly and public shower halls in order to establish a new collective life and raise the collective consciousness of peasants’. That the Bolsheviks, as urban cosmopolitans, were contemptuous of the peasantry was only to be expected: Trotsky described the countryside as ‘the Russia of icons and cockroaches’. But even Mao, coming to power behind a peasant army, was determined to turn the peasants into a proletariat, socialism’s only progressive class, by stripping them of their petty bourgeois property: land, trade, shops and finally the domus. Thus atomised, they could be reconstituted as military units – militias, battalions, brigades – and mustered into combat as units of production in the drive to modernise China.

With the commune as a total institution, a new kind of tyranny emerged that took on aspects of the prison camp. How well one ate, or whether one ate at all, was in the hands of the cadres who controlled the food. The ration, meal by meal, ladle by ladle, became an exquisite and implacable form of discipline and punishment. In some cases the cadres divided workers by productivity and calculated their rations accordingly. ‘Bad elements’ – who might be members of a rival faction, personal enemies, slackers – would often get nothing to eat. The tyranny of the ladle was a daily, calibrated and often arbitrary form of control over life and death. An inspection team in Neijiang county in Sichuan reported that 80 per cent of those who died had been deprived of food as a punishment. The starvation rations dominate the memory of survivors, as in this ‘Satire on the People’s Soup’:

In the mirror provided by the thin porridge
     soup one can see four eyes and two heads
As well as the reflections of houses and buildings.
By blowing at the porridge, one can make a large wave in a small bowl.

It was not just the shortage of food that enraged people but the sometimes fatal inequalities in its distribution. The leading cadres and officials ate well; in many communes there was a special closed dining area for team leaders and visiting officials. In a society not long removed from periodic famines, and where the symbolic meaning of food holds such a prominent place on ceremonial occasions, the contrast between the privileged stuffing themselves with pork dumplings while the rest ate chaff and wild greens provided bitter memories. One of the oldest slogans used in Chinese riots and rebellions driven by hunger was: ‘Equalise the food!’ Then, it had been directed largely at landlords and merchants; now it was largely aimed at cadres who fed themselves at the people’s expense. At the height of the famine, the privileged of the party-state could have been distinguished from commoners solely on the basis of body weight or red blood-cell count.

All of this was made possible by the household registration system, extended to the countryside in 1958, which riveted peasants to their place of residence. Movement was forbidden, though many did flee and many more tried and failed. As in the Soviet Union the rural population was denied the modest guarantees of housing, food rations and access to healthcare given to the urban workforce, and no effort was spared to prevent them from running away. Information was impounded to stop news of the famine spreading. While the peasants of Xinyang prefecture in Henan were starving in droves, the local party had the post office hold back more than 12,000 letters appealing for outside assistance. Death tolls, including deaths by beatings and suicide, were underreported by cadres fearful of inciting even greater panic and exposing themselves to retribution.

Stalin distrusted the peasantry; Mao did too, though with a great deal more firsthand knowledge. In 1933, Mikhail Sholokhov wrote to Stalin protesting against the collectivisation campaign’s grain seizures and predicting starvation. Stalin replied:

The esteemed grain-growers of your region [Kuban] and not only your region carried out a sit-down strike and would not have minded leaving the workers and the Red Army without bread. The fact that the sabotage was quiet and apparently harmless does not alter the fact that the esteemed grain-growers were basically waging a quiet war against Soviet power. A war by starvation, dear comrade Sholokhov.

Mao was more precise about the reason for his distrust. He believed peasants would underreport production even though the basic problem was the precise opposite. Cadres, out of intimidation or utopian enthusiasm, wildly overestimated yields and requisitions were correspondingly outrageous, leaving the peasantry to starve. Still, in the midst of the worst month of the famine, Mao said:

Everyone can see that there is a certain amount of strain in our relations with the peasants at this time over certain matters. An obvious phenomenon is that following the bumper harvest in 1958, part of the procurement of grain, cotton, oil crops … has still not been met. Moreover (apart from a minority of disaster areas) there has been widespread incidence of false reporting of output and private withholding throughout the country with unrest over claimed shortages of grain, oil, pork and vegetables.

There followed a relentless drive to uncover the hoarded grain and punish those responsible. Party organs denounced the feigned hunger and crop failures as a pretext for holding back grain. Grains – millet, wheat and rice – were the focus of attention: they had high value per weight and volume and could easily be transported and stored. Communes with root crops such as turnips and sweet potatoes came under less scrutiny and were likely to fare better.

By now desperate peasants as well as cadres were hiding what grain they could and underestimating the acreage the commune farmed in order to reduce the state’s take. The communes that were relatively successful in limiting the damage, Chen suggests in his careful comparison of villages in Anhui, were those where local cadres with strong social ties to the peasantry were in charge. Outsiders were more likely to risk starving the commune by stripping it of grain in order to curry favour with their demanding superiors.

When you map the points of convergence in Yang, Wemheuer and Dikötter’s accounts, it is hard to escape the conclusion that in the context of ‘building socialism’ the structural contradictions between industry and agriculture were decisive. He may have been the leader of a peasant revolution, but Mao and most of his colleagues were mesmerised by the promise of a modern, industrial, socialist and proletarian China. The peasantry represented its humiliating and poverty-stricken past; the industrial proletariat, steel mills and great earth-moving feats represented its future. As in Russia, socialist modernisation in a largely agrarian country required the systematic transfer of grain, investment and resources from the countryside to the city. Mao didn’t precisely intend to starve the peasantry, but whenever it came to deciding who would eat well or, in a pinch, who would eat at all, the farmers or the workers, the proletariat won every time. This is the argument Wemheuer makes so convincingly both for Soviet Russia and for China. Mao’s attack on Liang Shuming, a leader of the Rural Reconstruction Movement in the Republican era, is a case in point. Liang accused the party of abandoning the peasants who ‘lived in the ninth hell’ and lavishing resources on the workers who ‘lived in the ninth heaven’. Mao attacked Liang in ‘one of the most aggressive speeches … in his life’, as Wemheuer describes it, and ‘warned that the adjustment of urban and rural wages would lead to the ruin of industry and the decline of the state’.

Speaking at the height of the famine in Gansu, Mao ordered that grain procurement be raised to a third of the (exaggerated) harvest, judging that this was the limit beyond which revolt would break out. He urged seizing the grain quickly before it could be eaten or hidden and added: ‘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 policies of the party-state accentuated the structural contradiction. Between 1957 and 1960, the agricultural workforce dropped by about 33 million, while the rural non-agricultural workforce – most of them working on massive earth-moving projects – increased by more than 50 million and the urban population by nearly 20 million. Together, these 70 million new non-agricultural workers were building the future of China and so they would be fed first. In 1960, the options narrowed. Only half the planned transfers of grain were made. ‘A Food Ministry report on 12 July 1960 described a state of emergency in Beijing, Shanghai and Tianjin and in ten cities in Liaoning Province and portions of Jilin Province. Grain reserves in Beijing and Tianjin would last only four more days, those in Shanghai two, and in Liaoning six,’ Yang writes.

The city was not only the new China: it had also become the nerve centre of the regime and unrest there could threaten it in a way that famines in remote provinces never could. Nothing better captures the logic of modernisation combined with the fear of revolt than Zhou Enlai’s comments to the Central Committee in August 1961, by which time the dimensions of the famine were well known:

Right now it looks as if there are no problems in the cities, but it is hard to predict whether or not unexpected circumstances will develop. We must anticipate that it will become tighter (jin) in the cities, but if the cities are thrown into disorder, all sides will be influenced. If we do our work in the countryside well, serious problems can be entirely prevented, but the situation in the cities is still not clear now, so the problem is now that the cities and the countryside may not both be stable. But which side should be tighter? We have discussed this problem several times in the Central Committee and made reports to the Chairman, we must tighten the countryside to protect the cities. Would this be acceptable? … Could we dissolve the cities and all go back to the countryside, back to Yan’an? No, not a single person would approve this.

Zhou is given credit for having, in late 1960, ordered relief grain to be shipped to three of the most hard-hit provinces: Sichuan, Anhui and Hunan. He also played a major role in the later forced rustication of 20 million urban migrants. But millions had long since died. The dreams, the policies and the very survival of the party-state all pointed in a single direction, towards the decision, as Wemheuer writes, ‘to save the cities at the expense of higher death rates in the countryside’.