Democracy and Class Struggle will publish a new series of articles critiquing the works of the neo liberal Hayekian bourgeois propagandist Frank Dikotter who should win the Communism with its Mask Off Joseph Goebbels Award for emulating Goebbels and taking the famine in China to the same heights that Josef Goebbels did with the Ukrainian Famine in 1930's for Nazi Anti Soviet Propaganda.
Mao’s Great Famine is a sensational account of China during the Great Leap Forward. It argues the death toll in the Great Leap Forward was at least 45 million. It also claims that 2.5 million of these died due to violence. Most bizarrely, it makes the claim that 30-40% of all homes in China were demolished during the Great Leap Forward.
This book depends largely on quotations from documents found in local Chinese Communist Party archives. Dikotter treats these documents as authentic and their content as correct, without a great deal of analysis of the question. Without some acquaintance with the documents Dikotter’s book is based on, a proper review is not possible.
It is not possible for this evidence and by extension, Dikotter’s book, to be reviewed properly without examining and authenticating these documents. I am only able to give my initial comments here, therefore, rather than a more final assessment of the book.
I would ask readers of this book to heed a general warning about all evidence given by the Chinese Communist Party in the post-Mao era concerning the Great Leap Forward. As I said in my article ‘Did Mao Really Kill Millions in the Great Leap Forward?’, there was a sustained campaign by the Chinese government after Mao’s death to create a negative historical verdict about the Great Leap Forward and the Cultural Revolution.
Therefore statistics and documents, relating to these periods, compiled in the post-1976 era should not simply be taken at face value. They need to be authenticated and corroborated. However, another point needs to be remembered too-it should not be just assumed that because a government document has been found in an archive, even from before 1976, that its content must be true.
From the late 1950s a big struggle in the Communist Party took place between the right-wing and the left-wing, this went on right until 1976. For long periods of time the right were in the ascendancy in different areas and in the central government itself, even before Mao’s death. Reports drawn up by different factions in this struggle may well contain large doses of ‘political truth’.
China has gone through massive turmoil since 1949, this has included complete reversals in political line by the Communist Party and related radical changes in the Party’s verdict on historical events. It would be wrong to assume any historically contentious document in a Chinese Communist Party archive is genuine without properly determining its authenticity and that it is what it purports to be. In addition reports of historical events in archival documents need to be corroborated from other sources such as mutually supporting witness accounts and physical evidence.
The first step would be to assess how well Dikotter has interpreted the archival sources he cites in his work. This is likely to be a problem. According to Dikotter, he had to sign a contract promising not to lend the documents he found to anyone else or let them be copied, as a condition of access to the archives.
I have reproduced Dikotter’s email statement regarding this in an appendix to these comments. Only those judged by the authorities to be professional historians can get access to the archives. As there are only a limited number of professional historians in the world with an interest in the Great Leap Forward, it may be quite some time before we get a second opinion on these records. So, in the main, all we have for now is Dikotter’s interpretation of what he saw.
However, Dikotter did let two correspondents of mine have an informal look at one crucial document in his office. This is Mao’s speech on 25 March 1959. On p.134 of Mao’s Great Famine, Dikotter quotes Mao as saying during the Great Leap Forward ‘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.’
When the page is read it is clear that Dikotter's interpretation of Mao's comments is absolutely wrong. The document actually shows Mao strenuously advocating mass popular involvement in the policy-making process and trying to prevent the imposition of an excessive workload on the people.
By pointing us to a document where Mao is making these comments in private, where he would have had no incentive to mislead his audience, Dikotter has in fact done a great service to the historical image of Mao. The Mao we see in this document is a champion of mass democratic rights.
My correspondents saw the document and confirmed the document contains meeting notes of Mao’s speech in a meeting in Shanghai on March 25, 1959. It does not list who attended the meeting. The file runs to 8 pages long in its entirety and the quotation (which they translate as “If there is not enough to eat, all will be starved. Rather, let half of population die so the other half can have their fill.”) is toward the end of the meeting notes. The entire meeting covered quite a lot of issues, mainly agrarian issues and food supply. My correspondents remember the following points from the document.
(1) Mao urged other officials to set the grain collection quotas as early as possible. He is recorded as saying “If grain collection does not exceed 1/3 of the harvest, peasants will not rebel.” However, this must be seen in the context of a lot of other comments. He also said ”Set the quota earlier so peasants can be relieved. Even if peasants want to give us their surplus, we will not accept it [because the quota has been set]. It is better to leave more grain to peasants.” Mao asked local officials not to set the goal too high like before. It must be stressed Mao seemed to be setting limits here, not minimum standards. Presumably in areas of food shortage, the quotas could have been set lower.
(2) Although the new quotas are still very possibly too high, the main tone of the speech, according to my correspondents, is to protect the peasants’ interests, stabilise people’s life at the time of food shortage, further expand mass participation in decision making, etc.
(3) The quotation “If there is not enough to eat…” , in the view of my correspondents, is said in a rhetoric tone, not as a serious command. My correspondents felt the quote is a kind of isolated comment within the speech, with no obvious connection to what is being said above and below.
However next to this quote is one sentence about not pursuing the Great Leap Forward in all areas, which I imagine may be related.
Given my correspondents could not take the document away to study it, we cannot say anything conclusive about its authenticity or indeed its overall content. The quotes above should not be regarded as fully verbatim for the same reason. If the whole document is authentic, we must wonder why Mao made such a comment about half of the people dying, especially when he had been so adamant at the Wuchang Conference, a few months earlier, that there should be no deaths at all due to the Great Leap Forward. Mao is quoted as saying at the Wuchang Conference:
‘In this kind of situation, I think if we do [all these things simultaneously] half of China’s population unquestionably will die; and if it’s not a half, it’ll be a third or ten percent, a death toll of 50 million?… If with a death toll of 50 million, you didn’t lose your jobs, I at least should lose mine; [whether I would lose my] head would be open to question. Anhui wants to do so many things, it’s quite all right to do a lot, but make it a principle to have no deaths.’ (1).
The quote suggests another possibility about Mao’s alleged comment about half the people dying so the other half could eat their fill in Shanghai. It might be regarded as a sarcastic statement, as when Mao said at Wuchang ‘Half of China might have to die…’, when he was warning others about over-ambitious economic plans. We must also remember that these were minutes of a meeting, and presumably not something written by Mao himself. It could well have been that Mao was simply repeating the kind of warning that he made at Wuchang and the statement was not fully minuted, leading to a misleading impression.
Dikotter’s approach in simply quoting this alleged statement, without taking an overview of all Mao’s statements on the issue and other statements made in the document is one-sided to say the least.
Some might argue, that setting any grain quota at all, at a time of food shortage was in some sense a ‘crime’. However, this would be very simplistic. For one thing, food had to be redistributed to areas most in need.
Han Dongping, a Professor at Warren Wilson College in the USA did some research into the effects of the famine in the Great Leap Forward in Jimo county in Shandong. On the subject of famine relief he noted that:
‘In 1960, six southeast provinces donated 215,000 kilos of grain, 650,000 kilos of dried vegetables and large quantities of winter clothes to Jimo County…In the same year, Qingdao municipal government provided Jimo County with …. 125,000 kilos grain, and over half of the households in Jimo County benefited…In November of 1960, a Shanghai municipal delegate brought to Jimo 60,000 kilos of grain, 650,000 kilos dried sweet potatoes and other relief materials.. In 1961, Shandong provincial government donated 15,000 tons of grain to Jimo and provided 200 grams of grain per villagers each day before the next harvest.’(2).
Han Dongping’s evidence comes from interviews with local farmers and local official records that he studied.
It is true that not all of the grain collected from rural areas was redistributed to famine areas.
However, urban dwellers had to be fed. Some food exports were necessary to buy the raw materials and machinery needed to prevent industry and transportation collapsing. A collapse of the transportation system and the urban economy would have just made famine relief and recovery harder while creating more hunger in the urban areas.
If the record of Mao’s speech is authentic, it may be that Mao believed that some reduction in quotas would be enough to allow a fairly high quota for famine relief and for the needs of the cities and industry, without the quotas themselves leading to more hunger. (Whether this was objectively speaking correct is a question beyond the scope of these comments.)
Any statements made here about the other documents Dikotter quotes from have to be even more tentative, as these were not viewed by my correspondents. However, I believe that something can still be said about how Dikotter evaluates them. Dikotter (p.328) writes that investigation teams fanned out over the country from October 1960, to investigate the behaviour of provincial leaders during the Great Leap Forward. These investigations led to the removals of many provincial leaders.
The rightists, Deng Xiaoping, Zhou Enlai, Peng Zhen and Liu Shaoqi made investigations at this time.
We must remember that Liu Shaoqi was in charge of the day to day running of the government by 1960 and would have overseen the process of ‘investigation’.
The majority of the allegations Dikotter makes regarding atrocities committed during the Great Leap Forward, such as beatings and torture, appear to be taken from these documents. This leads to two problems. Firstly, although Dikotter lifts a great many anecdotes and statistics from these documents, his direct quotes from them tend to be rather brief.
In addition, Dikotter does not tell us very much about the particular document he is quoting from at any one time. This makes any kind of evaluation of them, very difficult. Rather worryingly, some of the documents Dikotter quotes from were bought in ‘chaotic flea markets’ (page 345). He says that he only quoted ‘very few’ of these but we really need to know which of his quotations do come from the documents he acquired in this way. In addition there would be nothing to stop Dikotter putting these documents on the internet or circulating copies.
Also, assuming that the evidence Dikotter cites of atrocities are taken from the results of investigations, what we have are indictments used in a series of political struggles, with Liu Shaoqi ultimately presiding over the whole process of the investigation. The purpose of these indictments, according to Dikotter, seems to have been to get rid of local leaders blamed for implementing Mao’s line in an over-zealous manner.
One possible thesis is that the intended effect of these removals would have been to oust more left-wing leaders in favour of more right-wing leaders, which would increase Liu Shaoqi’s power base.
Overall, a stream of reports from the investigation teams to the centre documenting atrocities would clearly strengthen Liu and the political line he represented, while weakening Mao.
Of course, bourgeois authors tend to argue that Liu Shaoqi only became a rightest, once he saw that the Great Leap Forward was a failure. But this begs the question somewhat. Could it not be that when he saw problems with the Great Leap Forward occurring he sensed that he could use this in a competition for power with Mao?
Could not encouraging investigation teams to exaggerate the failures of the Great Leap Forward have been part of his strategy? Dikotter should at least consider such possibilities but he does not.
Dikotter presents documentary accounts that he believes show evidence for very serious crimes. He claims that mass violence was used against the population in the Great Leap Forward by local officials and their militias. This charge simply cannot be upheld without corroboration. Such allegations could only be proved if they were backed up by a sufficient quantity of mutually corroborative witness statements and by forensic evidence such as mass grave evidence.
Without such evidence it is not possible to ‘convict’ any individual or a political regime of mass murder or genocide. Indeed the lack of such evidence, at least of a sufficient quantity of documented witness evidence, would give good reason for doubting the archival evidence. It must also be said that some of the stories that have emerged from purported Party documents in the past have been outlandish in the extreme.
For example, Jasper Becker unearthed a ‘party record’ that claimed a Party Secretary in Qisi, Henan had boiled 100 children to make fertilizer. He quoted this in his book Hungry Ghosts This prompted Berlusconi’s infamous jibe at a political rally in 2006 about the Chinese ‘boiling babies for fertiliser’ that led to censure from the Chinese government.
Having said all this, the existence of the local party documents Dikotter has found is a matter of some interest and it must be hoped that the current onerous conditions on access and reproduction will be eased in the future. If nothing else, they may help illustrate the line of thinking and the different world-views of the two lines of the Chinese Communist Party during the Great Leap Forward.
Another positive feature of the book is the way Dikotter puts his own ideological cards on the table when he states in his preface that:
‘In a far more general way, as the modern world struggles to find a balance between freedom and regulation, the catastrophe unleashed at the time [of the Great Leap Forward] stands as a reminder of how profoundly misplaced is the idea of state planning as an antidote to chaos.’ (p.xii).
All historians can and should strive for objectivity. However, history can never be an exact science, so it is always very useful to know the political leanings of any historian when evaluating their work. Dikotter’s honesty about his right-wing ideological framework is genuinely refreshing.
However, little positive can be said about the aspect of his work the reviewers have got most excited about-his Great Leap Forward ‘statistics’. The national figures Dikotter tries to come up with for deaths by hunger and violence and figures for the destruction of housing are frankly of little value.
Dikotter wants to establish a new ‘headline’ figure for Great Leap Forward deaths of 45 million. To understand how Dikotter tries to do this it must be understood that he is discussing two separate sets of documentary evidence concerning the death toll.
One is an estimate of 32 million excess deaths by Cao Shuji, who bases his figure on a survey of reports drawn up by local Communist Party branches into Great Leap Forward deaths.
These reports were produced in 1979, when the Party line had swung decisively and finally against the principles of the Great Leap Forward. The second set of documentary evidence consists of the documents in the local Party archives that Dikotter himself has discovered, that were discussed above-the reports of the investigation teams sent by the central government to investigate the provinces from 1960-62. Dikotter calculates that the excess death tolls he has found in the local Party archives, compiled from 1960-62 tend to be 50% higher than those in the reports Cao Shuji cites, which were compiled in 1979.
Therefore Dikotter decides the death toll must have been 45 million. Dikotter favours the reports he has found from 1960-62 from the investigation teams over the reports given by the local branches of the Communist Party in 1979 because he believes the latter would have given more conservative figures for deaths, as they were trying to hide things.
This reasoning is not very convincing. Why would there have been any remaining reason for local Party officials to try to hide the figures in 1979, if the Great Leap Forward deaths had supposedly been investigated nearly twenty years before by the central government? Moreover, the Party as a whole was in no way trying to hide Great Leap Forward deaths in 1979.
As I said in ‘Did Mao Really Kill Millions in the Great Leap Forward?, senior Party leaders openly attacked the Great Leap Forward. Marshal Ye Jian ying made a speech about disasters in the Great Leap Forward in 1979. A Party resolution talked of ‘serious losses to our country and our people between 1959 and 1961’. Local party organisations would certainly have been aware of the new line when compiling their 1979 reports and would have known that they were expected to go along with the new political line on the Great Leap Forward. If the choice really was between endorsing a figure of 32 million and a figure of 45 million, then Dikotter’s book really gives us no real reason for choosing one figure above another.
However, the main problem is the reliability of any death rate estimate at all for China from 1958-1961. This point is illustrated by Dikotter’s rather selective faith in the available demographic data for the Great Leap Forward. Dikotter decides to round up Liu Shaoqi’s baseline figure of 0.8% deaths a year before the Great Leap Forward to 1% (p. 329). Dikotter decides that any number of deaths above this figure in the Great Leap Forward, were excess deaths.
By extrapolating the excess deaths he finds on a local level to the country as a whole he calculates the 45 million deaths figure. Dikotter’s figures seem to be based on figures gathered by the local investigation teams. ‘Liu Shaoqi’s 1%’ is more or less the ‘baseline’ death rate figure in the demographic figures released in the early 1980s by the Deng Xiaoping regime.
The Deng Xiaoping figures show up to 16.5 million excess deaths, due to increases from the baseline figure in 1959-1961 (with a high of 2.5% in 1960). The death rate figures were presented at a public academic gathering in China in 1981. ‘Liu’s 1%’ might be seen as corroboration of the Deng Xiaoping figures that emerged more than twenty years later.
But Dikotter cannot accept the Deng Xiaoping figures in full, as the Deng Xiaoping figures for 1959-1961 imply a death toll a lot smaller than the one he is proposing. So Dikotter has a problem. If Dikotter believes Liu knew a true death rate figure for years prior to 1960, he must also believe a comprehensive system of death registration was in place during the Great Leap Forward, despite the fact that the records of these death registrations seem to have been hidden from all eyes ever since.
Therefore Dikotter must accept the baseline figure of 1% given by Liu Shaoqi but then assume that the death registration information that shows 45 million deaths for 1958-1961 has been deliberately hidden by the Chinese authorities.
But if the Chinese authorities are in the business of hiding and manipulating population figures about this period, why does Dikotter insist that the 1% baseline figure must be true? What are his grounds for saying that the one figure should be accepted, while the other Deng Xiaoping figures must have been falsified? It might be said that the other documents Dikotter has found corroborate the 45 million figure but this is not really so.
As Dikotter book illustrates, Liu stated openly in a speech that 0.8% should be used as the baseline when making calculations of excess deaths that occurred in the Great Leap Forward. Therefore the figures that appear in the documents Dikotter has seen in the archives, cannot be used as any kind of corroboration as they were probably drawn up following Liu’s instructions. The investigation teams most likely came up with a total number of deaths in a province-by whatever means-and then subtracted ‘Liu’s 1%’ to get a figure for excess deaths.
As I showed in ‘Did Mao Really Kill Millions…’ it is not clear at all where death rate figures for the Great Leap Forward came from. This makes any of the figures given for the Great Leap Forward death toll, from the 16.5 million figure, to the estimates of local investigation teams, to Dikotter’s 45 million, mere speculation. Judith Banister, a leading western demographer of the Great Leap Forward period, expresses severe doubts about death rate figures for China in the 1950s and 1960s (3). It does not seem that there was anything like a comprehensive national death registration scheme at this time. People who have examined local population records for the Great Leap Forward seem to have found records of population changes but I have not seen examples of locally kept death toll figures (4). Could it be that when asked for death toll figures, local officials simply offered some variant on population change figures, having nothing else at their disposal? But a population change figure for a given locality provides absolutely no guide to the number of deaths in the Great Leap Forward. This was a time of massive movements of population as workers migrated from their villages to the towns or to construction sites or left their locality to find food when famine struck. If you just take figures for every area where population decreased and assume that this was part of a national death trend, then you might come up with a figure of 45 million. Overall, then, it must be asked where the local investigation teams got all their death rate figures from, given the lack of comprehensive death registration.
I think there is a much more sensible account of what happened in the Great Leap Forward, than the apocalyptic version given by the Jasper Becker, Jung Chang and Frank Dikotter side of the debate.
Firstly, we should, like Banister, accept that the 1% baseline figure is too low. Banister discusses the official figure of 10.8 deaths per thousand in 1957, given in 1981. She writes:
‘This is an unrealistic claim. Of course, the PRC made great strides in mortality reduction in the 1950’s. As of 1957, the patriotic public health campaigns had reduced the level of filth and the number of disease-carrying pests. A large proportion of China’s midwives had received instruction in modern midwifery. There were many epidemic-control stations monitoring infectious diseases and specialized centers attacking particular diseases….Yet underlying health conditions in China remained poor…This population might have achieved a crude death rate below 20 per thousand by 1957, but not nearly so low as the official death rate of 10.8.’ (5).
We can perhaps speculate a little about why Deng Xiaoping figures give such a low death rate figure for 1957 and such a high rate for 1960. Liu Shaoqi announced his ‘1%’ (or 0.8%) in a speech in his home town, just as he was starting his political campaign against the line of the Great Leap Forward.
Once Deng Xiaoping consolidated his power after the death of Mao he asked for statistics to be put together concerning Great Leap Forward deaths. The workers compiling the statistics would have known of Liu’s 1% baseline figure and this became their own baseline for the figures released in 1981.
This is not necessarily a matter of complete conspiracy. Maybe Liu said the figure was 1% because this had been the figure the Party had wanted to give in the late 1950s to celebrate its achievements. Maybe given the euphoria of the time the Party thought they actually had achieved such a low figure. However, maybe Liu just invented it in 1960 because it made the death rate figures look worse than they were, thus undermining his political rival Mao.
When it came to compiling population statistics in the late 1970s, perhaps researchers, finding nothing else to work on, came back to Liu’s low figure and decided to adopt a figure that was roughly equivalent. It was after all a figure that had been endorsed by the Head of State at the time. Demographers can make assumptions on thinner grounds than this when faced with a paucity of hard evidence. Once the 1% is accepted as a baseline, demographers have the problem of trying to come up with a series of birth rate and death rate figures that in some way correlates with the census figures of 1953 and 1964. The obvious solution is to push all the deaths which you cannot account for, given the 1% baseline rule, into the famine periods. Thus the researchers came up with the death rate figures in 1981 which gave rise to the 16.5 million death toll statistic.
Writers on the Great Leap Forward are routinely taking these figures and extrapolating from them to reach even higher figures, which they then give the status of fact. Without some idea of how the 1981 death rate figures were actually calculated, they are of little use for such purposes. Once you adopt a sceptical attitude to the Deng Xiaoping figures, other death rate figures such as Dikotter’s and Banister’s begin to look unconvincing too (6).
Dikotter’s figures for deaths by violence and home demolitions are certainly the weakest part of the book. Dikotter states that 2.5 million people died of violence during the Great Leap Forward. His evidence again comes from the ‘investigation teams’. The figure appears to come from an extrapolation from figures given for one region (Xinyang) and two counties (see page 297-8). Dikotter tells us that as ‘rough approximation’ 30-40% of all houses were turned to rubble in China in the Great Leap Forward. Dikotter’s source for this astounding figure is, Liu Shaoqi again, who apparently claimed that 40% of all houses in Hunan had been destroyed (p.169). The other main source is a figure that 45-70% of homes in ‘the most affected counties’ of Sichuan were demolished (p. 170). Even if both these reports were completely true, one could hardly extrapolate from these two figures and say that 30-40% of homes in the whole of China were destroyed. These were just two provinces and we do not even have an estimate for the total number of home demolitions in Sichuan, just those for the allegedly most affected counties.
The question we have to ask about the figure for home demolitions is, again, where is the witness evidence? Of course the media in China is fairly stringently censored. But especially in the last three decades millions of people have travelled into and out of China. If 40% of all homes had been demolished in the whole of China in the Great Leap Forward, would not this fact have come out before now?
Other somewhat strange claims in Dikotter’s book would also bear further analysis, no doubt. He writes of the Ming Tombs (Shisanling) Reservoir, that was built in 1958. Dikotter states (page 30) : ‘As the reservoir was built in the wrong location, it dried up and was abandoned after a few years.’
Anyone who was been there recently will testify that it is actually rather full of water. The fact is that Dikotter just assumes the whole project must have been a total failure because it was carried out during the Great Leap Forward. Such errors illustrate the need for rather more even-handed historians to go over the evidence that Dikotter has presented in more detail than I am able to do here.
Overall, Dikotter’s book is, on the face of it, unconvincing. His claims are just too exaggerated and his analysis of the veracity of his sources is just too underdeveloped. It is part of a trend towards ‘death toll inflation’ which sees the numbers of those allegedly killed by Mao increase year after year as ‘new historical evidence’ is published.
Deng Xiaoping released figures that gave rise to the 16.5 million death toll. Judith Banister raised this to 30 million. Now, Dikotter has taken Banister’s 30 million and raised it to 45 million. But this of course is only ‘a minimum’, some historians put the figure at 50 to 60 million, Dikotter tells us (page 333). But as the death rate totals inflate, it will get harder and harder to fit in all these excess deaths between the figures provided by the two censuses of 1953 and 1964, unless the death toll in the non-Great Leap Forward years is pushed down to a ridiculous level. This will not bother Dikotter as he seems to be a sceptic about all the Chinese demographic data. This position is a perfectly acceptable one to take but where will it leave ‘Liu’s 1%’ baseline on which all Dikotter’s figures are based?
You cannot state that a death rate figure is credible when you believe that all the available population figures are completely false.
The death rate is a percentage of the population after all. Presumably at some point the death rate figures will have been thrown out too. When we get to 60 million, there will be no real reason left not to allow the death figure to rise ceaselessly up towards the 100 million mark and beyond.
Of course, there is a real story about the Great Leap Forward buried under all the nonsensical ‘death toll figures’. Certainly, that story includes the tragedy of the famine that occurred in China in the Great Leap Forward. The story must include the fact that the deaths that occurred were due to policy errors, as well as the very adverse natural conditions of the time. However, it is also a story of a nation surrounded by adversaries, desperately trying to pull itself out of the economic backwardness that had repeatedly condemned it to famine in the past.
(1) R. MacFarquhar, T. Cheek and E. Wu (eds) The Secret Speeches of Chairman Mao: From the Hundred Flowers to the Great Leap Forward, p.494-5. Harvard University Press.
(2) Han Dongping, ‘The Great Leap Famine, the Cultural Revolution and Post- Mao Rural Reform: the Lessons of Rural Development in Contemporary China.’’ http://www.chinastudygroup.org/article/26. 2003.
(3) Banister, J. (1979), China’s Changing Population Stanford University Press, p.87-8.
(4) See for example, Endicott (1988) Red Earth Revolution. In a Sichuan Village, p.55-6 and Han Dongping (2003).
(5) Banister, J. (1987), p.80-81.
(6) See Banister, J. (1987), p.114-119. Banister’s own figure of 30 million deaths is just a variation on the Deng Xiaoping figure. Banister gets her figure by using a significantly higher figure for the number of births between the censuses of 1953 and 1964, than the official figure, given by the Deng Xiaoping regime. The higher rate of births, combined with the census figures imply a higher rate of deaths, than the official figures show, otherwise the 1964 population figure would have been greater. Banister then apportions the extra death according to proportions derived from the death rate figures for this period released by the Deng Xiaoping regime
SOURCE AND MORE UPDATED VERSION