source:http://sdcc17.ucsd.edu/~twan/LoneCulprit.htm        

    The new Mao biography by Jung Chang and Jon Halliday, Mao: the Unknown Story begins with the assertion that Mao was personally responsible for well over 70 million Chinese civilian deaths during peacetime.  This number has since been repeated innumerable times in the great wealth of literary reviews the book has received.  Chang and Halliday’s charge that Mao was responsible for over 70 million civilian deaths during peacetime is based on several factors, many of which prove overestimated, manipulated, and at times horribly exaggerated.  The importance of realizing these inaccuracies is more than nitpicking a book.  This is a blow to all real scholarship on Mao.  Critiques of PRC-sanctioned Mao history are important in discovering the truth behind this very important 20^th century leader, but by openly distorting history this book does a disservice to anyone seriously attempting to figure out the real history of Mao.  When reviewers lend credit to a book like this, the integrity of all Mao research is called into question.  The numbers of those dead proves an area of great deceit – and provides important insight into Chang and Halliday’s methodology.  Though the authors never provide an explanation of how they arrived at their 70 million figure, the actual breakdown from different parts of the narrative is as follows:

As one can see most of the deaths attributed to Mao resulted from the Great Leap Forward.  One should realize that a great deal of those killed in labor camps died as a result of the famine.  Chang and Halliday do not take this into account, and thus double count those deaths, as will be described later.

During the period known as the Great Leap Forward, Chang and Halliday calculate that 38 million people died because of starvation.  The math they offer is somewhat complex and is taken from a variety of sources.  While this might make one feel that the authors captured a more holistic view, the truth is that their picking and choosing of sources allows them to manipulate the numbers significantly.  The actual numbers they give are in the note on page 438, with the calculations breakdown as follows:

            These figures add up to what the authors promised: a big number that can be attributed to Mao.  While the actual extent of Mao’s personal responsibility for each death will not be argued here, the questionable method by which these numbers were achieved deserves attention.

            The scholarship on this period is very extensive, yet the authors chose to employ only parts of it.  The only English language text cited for this number is Dali Yang’s book: Calamity and Reform in China.  The only useful information offered in this book is a reproduction of numbers from Judith Banister’s book: China’s Changing Population.  Bannister’s study is generally highly regarded by the academic community despite its claim of the relatively large figure of possibly 30 million famine deaths.  Her math is not specifically mentioned in the book, but her numbers generally take the population statistics and death rates released by the Chinese government in the 1980s, which she then tries to correct for any inflation or deflation caused by politics.  With this her population estimates are lower and her death rate estimates are much higher than those of the PRC demographers.

            The Chang and Halliday death rates for 1960-1961 are roughly similar to estimates given by Banister (4.34 and 2.83 versus 4.46 and 2.30), but many other rates given do not match up (Bannister and PRC figures cited in Kane, 87).  For all other years Chang and Halliday use the official PRC numbers for their calculations.  This might seem odd at first considering that Banister puts higher death rates on these years, and the authors are always keen on increasing the numbers in anyway possible.  But the game Chang and Halliday play is much different.

            In choosing the numbers Chang and Halliday use the inflated population figures from the PRC (except for 1959 and 1960).  The death rates are where the number manipulation really comes into play.  For these numbers the author’s use famine era death rates similar to Banister for the years mentioned above, but for all other years they use the PRC estimates.  How do these affect their calculation?  By using the larger numbers (high estimates) for the famine years and smaller ones (PRC estimates) for the non-famine years the difference between the two (deaths presumed to be due to famine) becomes much larger.

They also choose to use the years immediately after the famine to calculate the average “normal year” death rates.  The difference is crucial since the weakest members of a population are most likely to die during the famine.  This effect is referenced by Xizhe Peng as a form of Darwinian selection[1].  This meant the old, young children, and people imprisoned in labor camps were going to die at much faster rates than the others, thus eliminating them as a source of normal deaths in post-famine years.  Once all these people were dead at the end of the famine, fewer were left to die, leading to very low death rates into the mid-1960s.  Calculating “normal” death rates from the years before the famine would be much more accurate.  Xizhe Peng uses the death rates from 1956 and 1957 to serve as the average, which rise up to 1.9 in his research (not 1, like Chang and Halliday use).  Though death rates were declining through the 1950s the average death rates between 1953 and 1956 is much larger than those for 1957 (according to the PRC) and isn’t surpassed by famine deaths in Banister’s calculation until 1960. By using the adjusted rates for famine years while using PRC post-famine rates for the average, Chang and Halliday were able to increase their number by at least a third.  If one uses only Banister’s numbers for the famine years and 1957 to represent a “normal” death rate (the lowest pre-famine number – so this assumption will lead to the highest percentage of unexpected deaths) then the years 1958-1961 only saw about 25 million excess deaths.  Other sources support this lower estimate.  Xizhe Peng’s calculations, first based on provincial death estimates of the hardest-hit areas used to generalize the rest of the country and followed by taking into account an increase of 10% in underreporting of deaths during the period, comes out with 23 million excess deaths[2].

While the famine was truly a catastrophe of biblical proportions, the kids of distortions introduced by Chang and Halliday’s interpretation of that horrid period only serves to disgrace and mock every one of those innocents who died during it.  The authors use this human disaster as a selling point for their book, a book whose purpose is to assassinate the image of Mao.  While Mao was by no means a nice man or a kind leader, these authors are discarding all the integrity and objective research of the academic community for a cheap shot at a dead dictator.

This is not the end of their numbers game.  Though a large number is traced to deaths in the Great Leap Forward famine, much of the rest of the total seems unaccounted for if not for a small footnote on page 325.  The note estimates the population of China’s prison and labor camps to be 10 million each year with a 10 per cent death toll per annum.  When calculated over the course of Mao’s rule (1949-1976) this equals 27 million deaths[3].  The methodology used to reach this enormous figure is remarkably primitive, and Chang and Halliday utterly neglect the findings of other scholars who have spent their entire careers studying the subject and writing exhaustive reports on the issue.

            The actual numbers used by the authors are borrowed from R. J. Rummel’s book, China’s Bloody Century.  The book’s methodology relies heavily on statistics compiled and interpreted by the author, whose list of sources is extensive but rather homogeneous. Almost all the pieces that Rummel cites for his evaluation of state-caused deaths in China are western secondary sources.  Several appear to be little more than ideologues working for the U.S. government.  The figures he provides are as voluminous as they are erratic.  During certain years the numbers given on the population of the prison camps alone go as low as one million while others reach heights of one hundred million.  Such ridiculously conflicting information should lead one to the conclusion that all is not right with some of the sources.

            The 10% death rate used by Chang and Halliday is also described in Rummel’s book, but it originates from Richard L. Walker’s The Human Cost of Communism in China, which was written for the US Senate.  Walker himself was an open conservative (his nickname is “Dixie Walker”), who served as U.S. ambassador to South Korea under Ronald Reagan.  His 1971 piece provides much, if not all, of Chang and Halliday’s math.  Rummel uses Walker’s estimate of an average labor camp population of ten million each year, but refuses to accept the death toll, which Chang and Halliday then proceed to use.  The immense rate of ten per cent of the labor camps dying each year is described in Rummel’s book, but not taken too seriously, as he points out that this number is an estimation of Soviet labor camp death rates using the assumption that China would have similar figures.  Rummel himself debunks this theory, making clear that there were great differences between Chinese and Russian labor camps.  The Chinese camps had much milder weather (as opposed to Siberia) and were much better organized so as to get the greatest amount of work out of the prisoners with as little death (loss of labor) as possible.  Rummel concludes that a one to ten per cent death rate depending on the year would be much more appropriate[4].

            Even a more accurate average death rate may not support Chang and Halliday’s argument as so many of these prisoners died during the famine brought on by the Great Leap Forward.  Like much of China’s population the prisoners suffered the same fate, perhaps even worse.  During any famine the people who generally die are those not strong enough to last it.  Most of those who died were too old, too young, or too undernourished to survive the lack of basic nutritional requirements.  This means that labor camp prisoners most likely suffered at much higher rates than the rest of the population.  Already half-starved and overworked these “class enemies” were not likely to get any favors from the government in terms of food.

Rummel explains that during the years of 1959-1961 the death rates skyrocket far higher than the rest of Mao’s rule.  This is not due to worsening labor camp conditions instituted by Mao, but because of a nationwide famine.  The prison guards themselves were also victims of the famine as they worked far from their families, unsure whether their relatives would survive on their meager government food rations[5].  Despite this, Chang and Halliday count all labor camp deaths the same, pursuing on with their sad race for numbers, refusing to acknowledge the role of the famine in their calculation, and indeed double counting those killed by the famine (already included in the aggregate calculations described above), and those killed by the famine while in labor camps.

            The actual death total that Rummel comes up with for the years 1949 through 1975 varies wildly.  The low-end estimate is only 625,000, while the high-end is 48,713,000.  This represents quite a range of opinion.  The mid-range estimate of 15,000,000 is a much more conservative and plausible figure (even agreed on by Laogai activist Harry Wu), but is still a difficult number to imagine.  The human rights activist and leading labor camp researcher and critic Harry Wu estimates that the total labor reform camp population for the first forty years of the PRC was 50 million people.  His more reliable figures for the post-Mao years suggest 500,000 sentenced each year in the post-Mao years, which would suggest about 7-8 million from Mao’s death to the publication of his study.  That would leave about 43 million in labor camps under Mao.[6]  By this calculation, Chang and Halliday’s figure of 27 million deaths represents a mortality rate of 63%.  That would seem rather implausible.

The total value of Rummel’s research using assumptions, estimations, and obvious fabrications is absolutely nothing.  These numbers are little more than the educated guesses of western scholars of a country they were not even allowed access to at the time.  This might not entirely matter, but one must ask why after reviewing all this Chang and Halliday decided to add another 12 million deaths on top of Rummel’s mid-range estimate.  His math was just as good, if not even more advanced, than theirs. Why are these numbers that important at all?  If no one knows the real answer, then why should one offer a wild guess at the expense of academic integrity?  Chang and Halliday’s approach requires a belief that there is a mathematical correlation to evil.  They seem to think that adding corpses to his rule makes Mao that much more evil a person.  But how evil is 27 million deaths?  Is it any more evil than 15 million, or 625,000?  Inflating numbers for shock value is a marketing tactic, one that has worked exceedingly well for this book.  It’s broad accusations, simplistic view of Chinese politics, and creation of numbers more wildly inflated than any ever heard of in the academic community has make it the most talked about Mao biography since The Private Life of Chairman Mao.  While statistical inflation to prove a point may be perfectly acceptable to some reviewers, let those of us who actually know something of Chinese history not forget that this same method, when executed by Communist cadres, was the very reason for much of the famine of the Great Leap Forward.