Vol. 27 No. 24 · 15 December 2005

Search by issue:

A Question of Sources

Andrew Nathan’s general criticisms of our book, Mao: The Unknown Story, rest largely on misrepresentations and distortion, especially of our use of sources (LRB, 17 November). The following are some key examples.

1. Nathan asserts that the sources we cite do not say the Chinese Communist Party was founded in 1920, but ‘merely confirm that early Communist cells’ were formed in that year. In fact, describing how the Party was set up by the Russians (‘born’) in 1920, both the (authoritative) sources we cite use the term Chinese Communist Party for the organisation as it was in that year. Nathan claims we ‘think [the date] important because Mao wasn’t in Shanghai in 1920’. On the contrary, we explicitly say that Mao was in Shanghai in 1920, a fact which is well known.

2. Nathan accuses us of ‘distorting’ our sources regarding the battle of Tucheng on the Long March, a battle initiated by Mao. According to him, the sources we cite say Tucheng was ‘a victory’. But none of our seven sources calls it a victory. Among the sources we cite is Mao himself, who twice described it as a ‘defeat’.

3. Nathan claims that we base our interpretation of how the Red Army was allowed by Chiang Kai-shek to escape at the beginning of the Long March on ‘one fugitive piece of evidence’. In fact, a glance at our notes would show 26 written sources, including Chiang’s orders for troop deployments when he knew the Reds were escaping.

4. Nathan distorts our use of the evidence showing that what has been touted as the most famous battle on the Long March, at the Dadu Bridge, never took place. He asserts that our ‘key piece of evidence is an interview’ with a 93-year-old. He ignores our seven written sources, including a contemporary Red Army publication, and Nationalist cables unequivocally showing that there were no troops at the bridge. He also chooses to ignore our statement that ‘the strongest evidence … is that there were no battle casualties.’

5. Nathan charges us with making ‘false claims’ about Mao having planned the border war with India in late 1962 ‘for some time’. The official Chinese history of the war shows Mao doing exactly this (from May 1962). He also writes that ‘according to their own source’ – singular, implying we have only one source – Khrushchev did not horse-trade with Mao in the lead-up to the Cuban Missile Crisis. We cite five written sources, the key one being Zhang Dequn (not Liu Xiao), who details the horse-trading.

6. According to Nathan, we have only one, uncheckable (‘unpublished’) source for a speech about coups by Lin Biao at the start of the Cultural Revolution in 1966. In fact, we cite three written sources, including the standard English-language source. The one classified source we cite is given a full reference.

7. Nathan says we make a ‘string of assertions’ about Mao’s treatment of Liu Shaoqi in detention in 1967-69, alleging that what we say is based solely on two interviews. In fact, our main source is an authoritative written one, by Huang Zheng. We ascribe to each interviewee only one very specific piece of information.

8. Nathan claims that we ‘misreport’ our source on the issue of whether a meeting between Peng Zhen, the mayor of Beijing, and the disgraced ex-defence minister Peng Dehuai in 1966 was held ‘in secret’. Nathan asserts that this was not the case, yet himself acknowledges that the source we cite states that the man who arranged the meeting (Sichuan boss Li Jingquan) agreed not to report it to Beijing. By Chinese Communist standards, this was an act of the utmost secrecy, as it involved visiting Mao’s top detainee (Peng Dehuai) without Mao’s knowledge. Nothing could be more secret. Nathan’s rendering of this episode shows a failure to understand how the Chinese Communist system worked.

9. Nathan faults us for ‘tak[ing] what Mao says literally, even his well-known outrageous statements that famine and nuclear warfare were no big deal’. He suggests we are missing ‘humour or irony’ in Mao’s remarks. It is dismaying that Nathan believes it right to brush off such vital evidence from the horse’s mouth. These statements represented Mao’s policies, which led to the deaths of tens of millions of Chinese.

10. Nathan claims that many of our discoveries ‘come from sources that cannot be checked’. This simply is not true. A survey of our notes should show that the vast majority of our major discoveries are based on documentary sources, and can be checked – provided one does a little work. To prove his ‘uncheckable’ point, Nathan cites two unpublished documents: the findings about the poisoning of Mao’s rival Wang Ming, and a conversation Mao had with Japanese Communists about Indonesia. In the first case, perhaps Nathan could look at Wang Ming’s papers in Russia, where we pursued the leads which got us to the document. And in the second case, as we state in our text, the source was the Japanese Communist Party, whose contact details are in the public domain.

Jung Chang & Jon Halliday
c/o Gillon Aitken, London SW10

Andrew Nathan writes: Most of Jung Chang and Jon Halliday’s complaints fall into two overlapping categories: I did not check enough sources; I misinterpreted what they or their sources said.

Chang and Halliday’s method of citation makes it necessary for the reader to check multiple sources in order to track down the basis for any single assertion. There were many passages in their book which I had doubts about that I could not check because the sources were anonymous, unpublished, or simply too hard to get. It’s true that I did not visit the Wang Ming papers in Russia or telephone the Japanese Communist Party. Is Chang and Halliday’s invitation to do this a fair substitute for citations to the documents they used – author, title, date, and where seen? I limited my published criticisms to those for which I was able to get hold of what appeared to be all the sources.

If a book’s citation system does not tell a reader which sources are cited in support of which claim, it is not doing its job. It is in order to avoid infinitely receding evidentiary horizons of the sort indicated in Chang and Halliday’s reply that serious scholarship lays out its train of evidence and assertion in systematic, replicable fashion. If it takes 26 separate sources to support a single assertion, then one is dealing with circumstantial evidence. Of these 26 items, which one, two or three unequivocally support the improbable claim that Chiang let the Reds escape intentionally? The 93-year-old who saw no battle at the Dadu Bridge is the key piece of evidence for this theory, while the other evidence is circumstantial. Mao’s ambassador to Moscow was the only eye-witness to the alleged Khrushchev horse-trade with Mao, and his account refutes rather than supports the claim that such a horse-trade took place. In the passage on Liu Shaoqi’s imprisonment and death, my point was not that it didn’t happen, but that three specific assertions are keyed in the notes exclusively to interviews, and the book contains no critical discussion of the authors’ interview methodology.

To save space, I sometimes paraphrased Chang and Halliday or their sources, but it does not matter for the substance of the issues. Thus, they are right to say that one of their sources refers to the ‘birth’ of the ‘Party’ in 1920, but the author goes on to make clear that what he is talking about is the formation of Party cells prior to the Party’s founding congress. Chang and Halliday do say that Mao was in Shanghai in 1920 but they go on to say that ‘the Party was founded … after Mao had left.’ In short, the substance of the issue remains the same: Mao was present at the founding congress of the CCP just as historians have always said, but unsurprisingly, he was not present at all the activities that preceded it.

The two sources that Chang and Halliday link to their three-paragraph account of the battle of Tucheng (the other five they mention are not keyed to those paragraphs) do not use my word ‘victory’ or their word ‘defeat’ but – as one would expect from PRC sources, whatever the truth of the matter – narrate a heroic battle by the Reds. Chang and Halliday cite selectively from these accounts to portray a disaster. The battle may indeed have been a disaster but that is not the account given by these sources.

Whether a meeting revealed to the local Party secretary was held in secret is a matter of interpretation. I gave my readers the relevant facts, which Chang and Halliday did not. In any case, they do not address my larger criticism, which is that they, as writers who claim to ‘understand how the Chinese Communist system worked’, should not have speculated that the meeting between the two Pengs involved discussion of possible military action against Mao. It is inconceivable that it could have done so. And their speculation that it did is explicitly contradicted by their source.

I apologise for not having noticed that they cite Ying-mao Kau’s 1975 work for their Lin Biao quote. The substance of my criticism remains, that here and elsewhere they cite a three-volume unpublished work in Chinese for which they do not give what I would call ‘a full reference’ – that is, they do not say where they saw this classified source.

As to how we read Mao’s quotations, the dictator said many beautiful and idealistic things which the authors do not take as ‘vital evidence from the horse’s mouth’. I agree that his policies count, and that they were disastrous. But a main argument of Chang and Halliday’s book is that his intentions also count. In assessing these, for Mao as for other historical subjects, we have to contend with the fact that actors in history avail themselves of the same opportunities that we do for irony, humour and indirection.


‘When I lived and worked in Moscow in the mid-1970s,’ Tony Neville writes, ‘everyone was expected to inform: failure to do so was a crime’ (Letters, 17 November). I was born in Leningrad in 1939, and lived and worked in the USSR until my emigration in 1979. Throughout that time, neither I nor others, to my knowledge, expected to inform or be informed on. Certainly, we were aware of the KGB and so on, but within the frame of that awareness we led a pretty full life. Drafted as a physician into the Soviet navy, I was dishonourably discharged in 1970 for ‘reading and spreading anti-Soviet propaganda’. Not because I was brave, but, precisely, because I was oblivious. True, during my ‘trial of honour’ by my fellow officers I became aware of a network of informers in the military. But quite a few ‘seditious’ things that I had, quite frivolously, done or said during my five-year service hadn’t been reported. Of about twenty officers on the ship only two did inform, and those who didn’t were not charged. I was surprised that some of those who took to the podium condemned the trial itself, not me.

Valentin Lyubarsky
Brooklyn, New York

Neil Forster (Letters, 1 December) remarks that Parveen Sharif’s communication to her brother, which it seems is what was relied on by the prosecution as encouraging him in his terrorist project, ‘was a purely private communication’, as if somehow that makes it all right. Leaving aside exactly what a ‘purely private’ email might be (for example, under the Regulation of Investigatory Powers Act and the Communications Act), is it right or even possible to distinguish between private encouragement and encouragement in public? Can Forster be saying that private communications should always be off limits where possible criminal offences are concerned?

Simon Cockshutt
London W5


It was John A. Costello, not de Valera, who declared Ireland a republic in 1948: an editorial slip-up in Matthew Kelly’s review of Charles Townshend’s Easter 1916 (LRB, 1 December).

Editors, ‘London Review’

Read anywhere with the London Review of Books app, available now from the App Store for Apple devices, Google Play for Android devices and Amazon for your Kindle Fire.

Sign up to our newsletter

For highlights from the latest issue, our archive and the blog, as well as news, events and exclusive promotions.

Newsletter Preferences