New Research on the Nanjing Incident

David Askew summarizes scholarship on what happened in Nanjing in December 1937.
December 31, 2004

David Askew
Ritsumeikan Asia Pacific University and Monash University
Japan Focus

1. Introduction

The Nanjing (or Nanking) Incident (also known as the Rape of Nanjing, the Nanjing Massacre and the Nanjing Atrocities) remains a highly controversial episode in Sino-Japanese relations.[1] Indeed, it remains so controversial, especially in Japan, that a neutral definition of the event, and even its name, has yet to be agreed upon.[2] However, most researchers would perhaps agree on the following. The Nanjing Incident refers to the killing and raping of large numbers of Chinese together with widespread looting and arson over a relatively short period of time (usually given as six to seven weeks) by the Japanese military prior to and following the capture of the city of Nanjing on 13 December 1937. Sadly for the historian, the Nanjing Incident is not only an important episode in Sino-Japanese relations, but is also emerging as a foundation stone in the construction of the modern Chinese national identity.[3] As a result, the historian's interest in and analysis of this event can be interpreted as an attack on the contemporary Chinese identity, while a refusal to accept the 'orthodox' position on Nanjing -- however defined -- can be construed as an attempt to deny the Chinese nation a legitimate voice in international society -- or, in Iris Chang's words, as a 'second rape'.[4] In the highly emotional and deeply politicised environment in which the history of Japanese imperialism is constructed, the temptation to vilify all who disagree with accepted orthodoxies has proved irresistible to those on both extremes of the debate -- extremes advocated by individuals who, here, will be labelled 'corpse minimisers' and 'corpse maximisers'.[5] On the one hand, to show too much scepticism is to risk being tarred as a nationalist revisionist or denier, an apologist for Japanese fascism and imperialism. On the other hand, any demonstrated interest in Nanjing can be viewed in some circles in Japan as 'Japan-bashing' or even anti-Japanese racism (in the case of foreign researchers) or 'self-flagellation' (in the case of Japanese). In this environment, the debate can become highly emotionally charged, and the historian's struggle to weigh the evidence can quickly fall victim to the demands of contemporary politics.

The importance of the Nanjing Incident to contemporary Sino-Japanese relations can hardly be overstated. Nanjing forms one of the core historical issues on which Japan and China cannot agree, and continues to bedevil the bilateral relationship. It contributes to, if not drives, the controversy over Japanese history textbooks. It certainly continues to poison Chinese opinion of Japan: for instance, one recent poll which received replies from over 100,000 young Chinese people showed that 83.9 per cent gave the 'great Nanjing massacre' as the issue they associated most with Japan.[6] Nanjing is also important in understanding contemporary domestic Japanese politics. The debate within Japan about Nanjing (and for that matter about textbooks) is also a debate about the legitimacy of the findings of the post-war military tribunals held in Nanjing and especially in Tokyo (the Tokyo Trial, or International Military Tribunal for the Far East). The broadly defined left in Japan is politically and ideologically committed to the tribunals and their findings, whereas those on the right reject the tribunals as illegitimate and the findings as 'victor's justice'. The debate in Japan is thus heavily influenced by a broader philosophical and ideological discussion of history and historiography, and in particular by a further debate over the legitimacy of the narrative on the history of pre-war and wartime Japan which emerged from the post-war military tribunals.

Nanjing is a topic that has attracted far more activists than historians, especially in the West, and especially on the web. It remains a hotly contested domestic and international political issue both in Japan and China. There are large organisations that seem to be involved solely in running anti-Japan and anti-Japanese campaigns about the Nanjing Incident; there are a number of academic associations, magazines and numerous websites devoted to Nanjing; and Iris Chang's polemic, The Rape of Nanking: The Forgotten Holocaust of World War II (1997), has enjoyed phenomenal sales.[7] Despite all the interest in Nanjing, however, the history of the incident remains a largely untold story. Indeed, one of the major problems with the public discourse on Nanjing in Japan, where the historical research is most developed, is that it has tended to collapse into largely meaningless semantics about whether the sum total of atrocities committed in and around Nanjing can be defined as a 'great massacre', or what the definition of 'Nanjing' is. Another problem is the obsession with numbers, where the moral and political implications of the discourse about, and events in, Nanjing are engulfed in a reductionism that focuses solely on the number of victims. There are, however, some encouraging signs that the situation is changing for the better. This paper will attempt to clarify the current state of research on this incident and identify future areas of research. First, however, the issue of semantics must be addressed.

2. Semantics

Any attempt to analyse either the historical events in and around Nanjing itself or the historiography of Nanjing must come to grips with, and attempt to clarify, the issue of definitions or semantics. It is only by acknowledging the semantics of the debate that the historian can hope to move away from the mutual vilification that characterises much of the dialogue between 'minimisers' and 'maximisers'. In this section, therefore, the various arguments about the major concepts that need to be defined will be summarised.

First, there is little consensus in the English-language discourse about what terminology is most appropriate when discussing the events in Nanjing in the winter of 1937-38. The Chinese term used almost universally is 'Nanjing datusha' (the great [as in large-scale] Nanjing massacre). The term 'datusha' or 'tusha' implies a systematic killing, as in an abattoir, and appeared at an early stage in the Chinese discourse on Nanking as a translation of H. J. Timperley's (1898-1954) word, 'slaughter'. In a reflection of the Chinese-language discourse, the debate in Japanese frequently uses the term 'Nankin daigyakusatsu' (literally 'the great Nanjing massacre', again 'great' as in 'large-scale') or 'Nankin gyakusatsu' (the Nanjing massacre). However, the most common term is 'Nankin jiken' (the Nanjing incident), which suggests a more neutral, objective tone, and helps distance the user from the more emotional connotations of 'massacre'.[8] This term will be used here.

Second, and far more importantly, there is no consensus in Japan about the definition of 'Nanjing'. Some of the 'corpse minimisers' define 'Nanjing' as the Safety Zone, a small area of a few square kilometers within the city walls. Others argue that since 'Nanjing' was a walled city, the area within the walls is an appropriate definition. Another group claims that a broader area that encompasses the suburbs of the city outside the walls should also be included (Xiaguan would therefore be viewed as part of 'Nanjing'). Yet others (the 'corpse maximisers') broaden the area to include the six xian (counties) surrounding the walled city of Nanjing. Finally, a smaller group of 'maximisers' broaden the area even further, some to Suzhou (190 kilometers away), others as far away as Shanghai (320 kilometers away). The size of 'Nanjing' can thus be defined as an area of only a few square kilometers or as an area of hundreds of square kilometers.

Of the various positions, the extremes that reduce Nanjing to a few square kilometers, or that exaggerate the size of the city out of all proportion, must be said to be definitions that have unfolded under the undue influence of political considerations. The temptation to define 'Nanjing' in very narrow terms is understandable; with a narrow enough definition, the massacre can be airbrushed from history. It is not surprising that no reputable historian has advocated this particular (mis)understanding of the geography of the Nanjing atrocities. At the same time, the larger 'Nanjing' is, the larger the population and hence the larger the potential death toll. What is surprising is that the second extreme is advocated not only by layperson 'maximisers', but also by some of the major Japanese historians working on Nanjing. Historians in Japan who advocate a large death toll (that falls within the 100,000 to 200,000 range) often base their arguments on a very broad definition of Nanjing, and fail to distinguish between deaths in battle and post-battle killings.

The problem with the existence of multiple definitions is that these differences are often not acknowledged, leading to debates that are largely meaningless, mutual accusations of refusing to accept known and established 'facts'. I am convinced that if a mutually acceptable definition could be agreed upon (or even several: 'Nanjing a', 'Nanjing b' and so on, for instance), many of the disagreements between historians would disappear.

A related (third) issue is that of the time-scale of the Nanjing Incident. Different definitions of the geography of 'Nanjing' inevitably produce different time-scales. The Japanese army rushed to Nanjing once Chinese resistance in Shanghai collapsed. The city itself fell after a week or so of fighting on 13 December, but troops did not enter the Safety Zone until 14 December. The larger the geographical definition of 'Nanjing', the earlier the Japanese entered 'Nanjing', and hence the earlier the atrocities began. No reputable historian argues that the events in and around 'Nanjing' began on 14 December, but some push the date back into November and even August.

A fourth major area of disagreement revolves around the word 'massacre'. Indeed, the discussion about whether or not there was a 'massacre' in 'Nanjing' relies, at least to a certain extent, on the definition of the term. Chinese people were without question killed. There is however no authorised and commonly accepted definition of 'massacre' for historians to call upon. There is no consensus about what type of death can or should be classified as part of a massacre, nor about how many deaths, or how many murders, make a massacre.

There were individual incidents where large numbers of Chinese people were killed in battle or captured and subsequently executed, but even here there is no consensus. On the one hand, some 'corpse maximisers' argue that the large numbers of Chinese military personnel who undoubtedly died in the fighting for Nanjing should be counted in any toll of the 'massacre'. On the other hand, some 'corpse minimisers' argue that the deaths in battle of Chinese military personnel and even 'legal' executions should be distinguished from massacres, and that only 'illegal' killings can be counted as 'massacres'. It could also be argued that the term 'massacre' is misleading in that it suggests an organised and/or planned effort, and thus distorts the fact that many of the murders were committed by individual Japanese army units on the battle-field, and were small-scale and sporadic.

In other words, a broad definition of 'massacre' counts all Chinese deaths as part of the Japanese 'massacre', and so includes in the final figure given for the 'massacre' of Chinese in and around Nanjing those Chinese soldiers who died in battle, even those soldiers who were shot by Chinese military units for abandoning their positions. In theory, a very narrow definition might demand proof that an organised, planned (and illegal) effort to kill civilians was ordered by the Japanese military, in which case the final figure would shrink to zero. In practice, however, the 'corpse minimisers' claim that only illegal executions of military personnel and civilians can be counted as part of any 'massacre'. They argue both that the execution of the 'plain-clothes soldiers' (bianyibing, ben'ihei) used by the Chinese military was legal and that most of the men in civilian clothes who were executed were in fact plain-clothes soldiers, and so conclude that the scale of the 'massacre' was relatively small. Here it is important to note that the first extreme counts all deaths as 'massacres', while the second counts only a small proportion of deaths as 'massacres'.

A fifth issue emerges when the deaths are counted. If there is no consensus about the definition of the term 'massacre', there is also little agreement about how to calculate a final death toll. On the one hand, it might be argued that each death must be confirmed by written documentation, or by genuine and reliable eyewitnesses to insist in other words on a narrowly forensic approach to counting deaths. This would serve deliberately or otherwise to minimise the final estimate. On the other hand, others uncritically accept all accounts as genuine, adding together all figures to produce a total that can be substantiated by nothing in the primary sources. Both extremes are problematic, but there is no consensus on which middle position to adopt. Oral history can also be treated in two different ways to demand solid corroboration for each account gained from oral history would be in effect to deny the victims of Japanese atrocities in and around Nanjing a voice, but at the same time to accept uncritically all accounts derived from oral history is problematic because many can easily be shown to be false. There is no happy answer to this dilemma. In determining which of the many various positions on Nanjing is legitimate, or more legitimate, it is necessary to examine carefully the sources that are given, to scrutinise the credentials of those informants who claim to have been in the city, and to compare the emerging narrative against as many primary materials as possible. 

3. Current Research

Academic research on the Nanjing Incident is mainly conducted in Chinese, English, and Japanese. Of the three language groups, Japanese has produced the most advanced research, with the debate in English lagging years if not decades behind. In both Japan and China, however, polemical writing has been driven by nationalism, and especially in China by state priorities. The discourse in Taiwan has produced research of a high quality, but the indigenisation of history that has occurred from the 1980s, together with the emergence of democracy and a new national identity, has had a negative impact on research on China, including Nanjing. The discourse in Taiwan has been further undermined by the inflow of research from mainland China.

3.1 Chinese-language Research

The most valuable Chinese language materials are the collections of various primary sources, including the recollections of many of the Chinese military personnel in Nanjing.[9] However, these collections show no evidence of any vigorous critical attempt to distinguish between valid and legitimate primary materials and other materials; photographs, for instance, which are known to be fabricated, or from different areas and different times, continue to be used to 'prove' Japanese guilt in the winter of 1937-38 in and around Nanjing.[10] Moreover, because of the limitations on free speech in mainland China, much of the secondary material merely parrots the government line of the day, and it would be difficult to describe the situation as a 'debate'. Thus, for instance, a group of researchers at Nanjing University in the 1960s condemned the members of the Western community who remained behind in Nanjing to run the humanitarian Safety Zone for turning a blind eye to the Japanese atrocities in the city, and 'misused' the primary sources to suggest that they co-operated in the Japanese slaughter of Chinese.[11] It is true that the Westerners in Nanjing did work with the Japanese, but it was a reluctant co-operation, and there is absolutely no evidence whatsoever that it extended to helping directly the Japanese kill anyone.

As Chinese concerns about 'American Imperialism' diminished, and as Japan became the target of official vitriol (partly at least because of the highly politicised and contentious issue of Japanese textbooks), Westerners came to be depicted as resistors rather than collaborators. In another work that is frequently based on a vivid imagination rather than primary sources and which demonstrates the influence of the Chinese literature on some sections of the English literature, Iris Chang claims that members of the international community jumped 'in front of cannons and machine guns to prevent the Japanese from firing' on unarmed civilians.[12] However, although there is not a shred of evidence that this happened the only documented case of a killing witnessed by any of the Westerners who remained in Nanjing after the journalists left on 15 and 16 December is that of a single man who was executed by Japanese soldiers the work of the international community is today highly lauded in all the literature on Nanjing and is one of the few areas about which all researchers of the Nanjing Incident can agree.[13]

Despite the fact that there seems to be little sign of internal debate in China, there are indications of an emerging discourse. Several Japanese works including Hata Ikuhiko's Nankin jiken -- 'Gyakusatsu' no kozo (The Nanjing incident: The structure of a 'massacre') (1986)have been translated into Chinese, so readers have access to non-official points of view.[14] In addition, the web provides a forum in which all points of view can be discussed, and the liberal world of free debate is open to those who can read and write English.

3.2 English-language Research

Although the research in Japanese remains superior to that in English and Chinese, this was not always the case. Surprisingly, perhaps, much of the primary material on Nanjing was originally written and published in English. The two central collections of primary materials -- given the discussion of the Chinese literature, it is ironic to note that both these works were products of GMD (Guomindang) propaganda -- consist of works published in English very soon after the incident itself: H. J. Timperley ed., What War Means: The Japanese Terror in China -- A Documentary Record (1938), and Hsu Shuhsi ed., Documents of the Nanking Safety Zone (1939). This head-start has not however been maintained. The first major monograph on Nanjing to be published in English after Hsu was the problematic work by Iris Chang, The Rape of Nanking ,(1997), a work that can only be described as frequently fabricated and/or fictitious. Following the publication of Chang, historians have at last started to write in English about this important event in Sino-Japanese history. Joshua A. Fogel's edited work, The Nanjing Massacre in History and Historiography (2000), is by any standard an impressive work, albeit one that focuses on the historiography rather than the history of Nanjing. Although flawed, both Honda Katsuichi, The Nanjing Massacre: A Japanese Journalist Confronts Japan's National Shame (1999) and Hua-ling Hu, American Goddess at the Rape of Nanking: The Courage of Minnie Vautrin(2000), are important contributions.[15] One of the latest in the long run of recent publications in English includes Masahiro Yamamoto, Nanking: Anatomy of an Atrocity (2000), a work that is easily the most objective historical account of Nanjing in the English-language literature to date.[16]

American Missionary Eyewitnesses to the Nanking Massacre, 1937 - 1938 (1997), Timothy Brook ed., Documents of the Rape of Nanking (1999), and Zhang Kaiyuan ed., Eyewitnesses to Massacre: American Missionaries Bear Witness to Japanese Atrocities in Nanjing (2001), are all collections of primary materials long unavailable in English. Finally, John Rabe's diary, Der gute Deutsche von Nanking (1997) -- translated into English as The Good Man of Nanking: The Diaries of John Rabe (1998) -- is a crucial piece of evidence.

3.3 Japanese Language Research

The Japanese language literature is even more impressive. Unlike the debate in English, Japanese researchers have been debating -- and truly debating -- the incident for decades rather than only the past few years, so the Japanese language materials can only be summarised here.[17] This debate has ebbed and flowed over the years, but has been steadily building up steam since the mid-1980s. If the post-war period from 1945 to 2003 is divided into three periods of 20 years each -- 1945-1964, 1965-1984, 1985-2004 -- and if we look at books published in Japanese with Nanjing in the title or sub-title, during the first period no books, during the second 17, and during the third over a hundred books were published. Moreover, the 1980s in particular saw a collapse in the perceived legitimacy of the Illusion School, with, for instance, Kaikosha, a support group of individuals from the pre-1945 Military Staff College, which had previously denied that massacres occurred in Nanjing, expressing regret for the killings on behalf of those directly involved.[18] Recent popular Japanese interest in the Nanjing Incident in particular has triggered a flood of books that could well be described as a publishing industry. I believe that this was stimulated by the publication in English of Iris Chang's book, together with the publication in Japanese of John Rabe's diary in 1997.[19] Moreover, a neo-nationalist political movement, the Japanese Society for History Textbook Reform,[20] together with the Association for the Advancement of a Liberal View of History,[21] has helped to foster an intellectual environment in which many Japanese reject interpretations of the colonial and wartime epochs its critics describe as the Tokyo Trial View of History. Chang's work in particular is unashamedly based on this view of history and, as is often the case with this particular historical ideology, is fatally flawed. The intellectual environment in Japan has changed to such a degree that Chang's work has found very little support there, even among the corpse maximisers, left-wing advocates of the validity of the Tokyo Trial, who argue that a 'great (or large-scale) massacre' did occur. The reception of Rabe's diary has been, in general, much more positive. Together, these two works have served to reopen the debate in Japan on the Nanjing Incident.

Although the best introductory work on Nanjing in any language probably remains Hata Ikuhiko's Nankin jiken (1986), recent work in Japan has moved far beyond what was possible in the mid-1980s, principally because so many primary sources have since been published. The debate has also moved firmly into the mainstream. Although lay authors continue to write on Nanjing, the early debate was largely between a journalist (Honda Katsuichi) and a free-lance writer (Suzuki Akira), but now is also a debate between large groups of academics.[22]

One of the blemishes of much of the research on Nanjing in English to date is that frequently it has been based on secondary materials. Indeed, one of the great differences between the research in Japan and that in the English-speaking world, and one of the great strengths of the Japanese-language literature, is that it has relied heavily on primary sources. Ironically, perhaps, a large amount of material originally written in English is in fact far more readily available today in Japanese than in English. For instance, the first volume, Amerika kankei shiryohen (American materials), of a two volume set, Nankin jiken shiryoshu (Materials on the Nanjing incident), edited by the Nankin jiken chosa kenkyukai (1992), contains 85 newspaper and magazine articles originally printed in English at the time of the Nanjing Incident but now readily available in English only to the researcher with access to a good library. In addition, this collection contains over 150 primary documents that shed much light on the events in Nanjing during the winter of 1937-38. Both Timperley and Hsu have long been available in Japanese. Rabe's diary appeared in Japanese before an English edition was published, and while a Japanese-language edition of Minnie Vautrin's diary exists, researchers are still waiting for an English-language edition.[23] Kasahara has in fact recently noted that '[n]ine different collections of historical materials on the massacre have been published [in Japan]. Rarely has so much documentation been compiled and published with regard to a single historical event'.[24]

4. Schools of Thought

Since the Japanese discourse on the Nanjing Incident is the most sophisticated, the following discussion about various schools, methodologies and sources will focus on the situation in Japan. Interpretations of the Nanjing Incident in Japan are usually divided into three schools of thought, defined by the number of people each school argues was massacred in Nanjing.[25] They are the Nanjing Incident as Illusion School (maboroshi-ha), which argues that at most several thousand were 'massacred' in Nanjing; the Middle-of-the-Road School (chukan-ha), which argues that between 13,000 (in the case of Itakura Yoshiaki) and 38,000-42,000 (in the case of Hata Ikuhiko) were massacred; and the Great Massacre School (daigyakusatsu-ha), which argues, in the words of one of its leading advocates, Kasahara Tokushi, that 'over 100,000, perhaps nearly 200,000 or even more', were killed in Nanjing.[26] (Note that the three schools have differing views on what constitutes a 'massacre'.) The English language debate does not have so great a range of opinion, although Masahiro Yamamoto clearly falls within the Middle-of-the-Road School, and Iris Chang even more clearly argues for a massacre on a far greater scale than any member of the Great Massacre School. Chinese language sources are closer to Iris Chang than any of the three Japanese groups.

An introduction to the three schools was recently provided by the conservative Japanese opinion magazine, Shokun!, which sent out a questionnaire to which almost every important researcher of the Nanjing Incident in Japan replied.[27] The questionnaire was sent to both academic and lay members of all three groups, and responses were received from 23 individuals: Ara Ken'ichi, Oi Mitsuru, Takaike Katsuhiko, Fujioka Nobukatsu, Fuji Nobuo, Watanabe Shoichi, Tanaka Masaaki, Matsumura Toshio and Kobayashi Yoshinori (all from the Illusion School), Suzuki Akira (not clear, but given here as a member of the Illusion School), Unemoto Masaki, Nakamura Akira, Okazaki Hisahiko, Sakurai Yoshiko, Tanabe Toshio and Hara Takeshi (all of whom Shokun! places in the Middle-of-the-Road School), and finally Eguchi Keiichi, Fujiwara Akira, Himeta Mitsuyoshi, Inoue Hisashi, Yoshida Yutaka, Kasahara Toshushi and Takasaki Ryuji (from the Great Massacre School). By any standard an impressive and comprehensive list, it includes almost every researcher actively working on the Nanjing Incident in Japan. The major omission, apart from Hata Ikuhiko and Higashinakano Shudo, who were involved elsewhere in the Shokun! project, is Honda Katsuichi (Hora Tomio and Itakura Yoshiaki have recently died, while Kitamura Minoru's book on Nanjing first appeared after this survey was published).

The researchers and writers were asked to reply to a number of questions, including how many Chinese each believes the Japanese killed illegally ('killed illegally' is a narrow definition of 'massacred') in Nanjing, how the Nanjing Incident should be defined in terms of both time and geography, whether the execution of soldiers who shed their uniforms and hid among the civilian population of Nanjing should be included in any count of a massacre, and whether the Japanese execution of plain-clothes soldiers was forbidden by international law.

The answers to the first question about the scale of Japanese atrocities in and around Nanjing are hardly surprising -- the various schools are after all defined by their views on the issue. Members of the Illusion School answered that the number was zero (Fuji Nobuo), almost zero, or, in the case of Watanabe, 40 to 50. The Middle-of-the-Road School, which is given a broader definition than the one I use, ranges from 'several thousand' (Nakamura and Unemoto) through about 10,000 (Okazaki, Sakurai, and Tanabe) to about 20,000 (Hara). I would place all but Hara in the Illusion School: the Middle-of-the-Road School as I understand it gives a death toll that ranges from Itakura's estimte of 20,000 to Hata's estimate of 38,000-42,000. The Great Massacre School ranges from at least 100,000 (Eguchi), more than 120,000 (10 suman), a figure that has become the orthodox position of this school and which is advocated by Himeta, Inoue, Kasahara and Yoshida, to the older orthodoxy, 200,000, which is still advocated by Fujiwara and Takasaki.

Table 1. The number of Chinese 'victims' in and around Nanjing: The three schools as defined by Shokun! and Askew

>Illusion School
Middle-of-the-Road School
Great Massacre School
Several thousand-20,000
100,000-200,000 plus
100,000-200,000 plus


The enormous differences in the various estimates of the scale of the Japanese atrocities in Nanjing are at least partly due to the differences in definition of concepts such as 'Nanjing' and 'massacre'. As has been mentioned above, the Illusion School's understanding of the temporal and geographical definition of the Nanjing Incident differs from that of the Great Massacre School. The majority of the Illusion School believes that the Nanjing Incident lasted for six weeks, from mid-December to late January (this definition -- that of the post-war war crimes trials -- also dominates the English-language literature, and has been literally set in concrete by the Chinese government). The Great Massacre School, however, gives mid-November to late January (Eguchi and Takasaki), six weeks (Fujiwara and Himeta), and 1 December, 4 December and mid-December to March (Inoue, Kasahara and Yoshida respectively). There is also a large variation in the geographical definition of Nanjing. Because their time frame has been pushed back so far, Eguchi and Takasaki appear to define Nanjing to include areas such as Suzhou, 190 kilometers away (occupied by the Shanghai Expeditionary Army on 19 November) and Jiaxing, which fell on the same day and which was even further away from Nanjing. Apart from Himeta, who defines Nanjing as the city and its suburbs, all other members of this school define Nanjing as the city and six surrounding xian (counties). Needless to say, by expanding the temporal and geographical definitions, it becomes possible to argue for a higher death toll; and by narrowing them to argue for a smaller one. As noted above, 'Nanjing' can thus be defined as a few square kilometres (some members of the Illusion School argue that Nanjing ought to be defined as the Safety Zone), or as several hundred square kilometers. One of the serious limits of the debate in Japan is that these differences are rarely if ever clearly articulated, making debate on the death toll in 'Nanjing' meaningless given completely different definitions.

There are also large differences regarding the soldiers who changed into civilian clothes and hid among the civilian population of Nanjing. Some claim that they should be viewed as regular soldiers, whereas others treat them as plain-clothes soldiers, civilians, or other. Here, the debate revolves around the issue of whether these soldiers should be viewed as legal combatants, illegal combatants, or non-combatants. Of the 16 members of the Illusion and Middle-of-the-Road Schools, 11 view such soldiers as plain-clothes soldiers and four as regular troops (in other words, 15 out of 16 view them as combatants, and 11 out of 16 view them as illegal combatants). Of the seven members of the Great Massacre School, one views such soldiers as regular troops (legal combatants), and six have replied 'other', giving their definition as defeated soldiers who had lost the will to fight (six out of seven view them as non-combatants). Needless to say, this difference has enormous implications for the debate about the legality of the executions of these soldiers, and hence the issue of whether the executions can be considered as part of a narrowly-defined 'massacre' of Chinese people.

Many members of the Illusion School in particular take a narrow, legalistic approach that distinguishes between legal combatants and non-combatants and the various rights granted to members of both groups by the laws and practices of war at the time, on the one hand, and illegal combatants who lack many of these rights, on the other. Here, the issue revolves around which of the three categories the soldiers in question belong to. In contrast, the Great Massacre School takes a moral and ethical position that views all deaths with abhorrence. The Middle-of-the-Road position as articulated by Hata is more nuanced. Hata argues that the executions may have been legal if carried out following appropriate procedures, including most importantly trial and sentencing in a court martial. Such courts were not convened.
There is in fact also a clear division in answer to the question whether the execution of these soldiers was legal: all members of the Great Massacre School declare that it was not; almost all others believe that it was.

This questionnaire provides the most detailed summary of the debate in Japanese circles about the Nanjing Incident that I am aware of. It was an impressive achievement to persuade so many researchers in Japan to reply, and to have made it possible to paint a picture of an emerging consensus about Nanjing in Japan. It is clear that many members of the Great Massacre School have begun to revise their figures for the scale of the killings quite dramatically downwards from 200,000 to 100,000-120,000 or more. It is also clear that the various schools share a very different set of assumptions about the temporal and geographical definitions of the Nanjing Incident. What would be of great interest would be to ask members of the Illusion School what they believe the death toll would be if the time span and circumference of 'Nanjing' were expanded, and at the same time to ask members of the Great Massacre School what the death toll would be if the definitions were narrowed. My own assumption is that the differences between the Middle-of-the-Road School member, Hata Ikuhiko, and the Great Massacre School member, Kasahara Tokushi, for instance, would disappear if this were done.

The Shokun! survey does not, however, provide more information on the schools themselves, or on their major characteristics. A summary of these characteristics will be attempted below.

The Illusion School mainly consists of conservative (and in some cases at least neo-nationalist) thinkers who are not professional historians, and of the three groups is easily the one with the largest number of lay members. It has, however, been given an enormous boost with the recent publication of 'Nankin gyakusatsu' no tettei kensho (A thorough investigation of the 'Nanjing massacre') (1998) by Higashinakano Shudo. This is one of the most important works on the Nanjing Incident as a whole to emerge since the publication of Hata Ikuhiko's authoritative Nankin jiken in 1986. Higashinakano's work makes a powerful and sophisticated (and, I believe, flawed) case for the Illusion School based on the type of solid, empirical research that the school to date has lacked. He has also helped to keep Nanjing in the public eye, writing constantly on the issue and playing a key role in founding the Nanjing Studies Association. Despite its many flaws, Higashinakano's work will continue to influence the debate in Japan for the foreseeable future.[28] Higashinakano has also teamed up with Fujioka Nobukatsu to publish a series of articles that mercilessly examine Iris Chang's work. These articles were subsequently brought together as 'Za x Reipu x obu x Nankin' no kenkyu -- Chugoku ni okeru 'johosen' no teguchi to senryaku (Research on The Rape of Nanking: China's methods and strategy in the 'information war') (1999). The Illusion School publishes through a number of small conservative publishers, frequently appears in the pages of right-wing magazines such as Sapio, Seiron, and Shokun! and has found support in the mainstream (albeit clearly conservative) newspaper, the Sankei shinbun. To the best of my knowledge, this school has no academic supporters in either the English-language or the Chinese-language discourse (although works by members have been translated into both languages).

A number of recent works, including the Rabe Diary, have tended to support the work of the Middle-of-the-Road School. The posthumous work by Itakura Yoshiaki, Honto wa ko datta Nankin jiken (The truth about the Nanjing incident) (1999), is an impressive summary of the work of an individual who devoted his life to researching the Nanjing Incident. It brings together much of the research that Itakura has done in the area, and will serve to bolster the Middle-of-the-Road School. Itakura also helped to edit one of the most important pieces of research on the Nanjing Incident, the three volume Nankin senshi (A history of the battle of Nanjing) (1993) work, which consists of an overview of the battle for Nanjing, a collection of diaries written by military personnel and official battle reports of the various Japanese military units that took part in the attack on Nanjing.[29] A recent individual who has joined the debate on Nanjing, Kitamura Minoru, sees himself as a member of this school (although he quite deliberately refuses to make any estimate of the death toll -- arguably a sensible option for Japanese researchers).[30] The authority on the Nanjing Incident, Hata Ikuhiko, is also a member of this school. Outside Japan, Masahiro Yamamoto clearly belongs to it, although his estimate of the total number of victims is a little high. To the extent that this school is defined as consisting of professional historians rather than ideologues (or myth-makers), and to the extent that it is defined as accepting the premise that the story of Nanjing can only be told through a reconstruction based on the primary documents, I would count many of the professional Western-based historians in this group too.[31] However, as long as the estimate of the number of victims remains the yardstick used to divide individual theorists into separate schools, and as long as many Western scholars refrain from making any such estimate, this may be premature.

Of the Illusion and Great Massacre Schools, the latter is clearly the more sophisticated, counting among its members a large number of academics who bring a great deal of authority to their findings. This school has been relatively quiet recently.[32] As even Kasahara (polemically) notes, '[i]n recent years more books questioning the massacre have been published [in Japan] than those confirming the facts of the incident'.[33] Iris Chang's work has clearly dealt the Great Massacre School a severe blow. Members of this school translated her book into Japanese but, through their publisher, the left-wing Kashiwa Shobo, had a public (and highly embarrassing) falling out with her when she refused her translators permission to correct the enormous number of mistakes her book is riddled with or to add translator's footnotes, and also when she objected to the publisher putting out a sister volume in which the mistakes would have been explained. In distancing themselves from Chang, and in explaining why their history differs from her myths, academic members of the Great Massacre Faction have found themselves in an unusual position. Rather than concentrating their criticisms on those who argue for a smaller death toll than that which they see as acceptable, they have found themselves criticising a work that argues for a larger death toll, and in doing so have to a certain extent blurred the clear lines that separated them from (or at least introduced some ambiguity in the relationship with) the members of the Middle-of-the-Road School. Despite is relative sophistication, some works of the Great Massacre School, especially those produced by lay members, can be said to share much with the Illusion School. Both can be highly ideological and dogmatic, both can be extremely violent in the language they use, and both can be more than careless with the historical facts and sources.[34]

Members of the Great Massacre School have recently published a volume that harshly criticises the work of the Illusion School.[35] In doing so, however, they merely reinforce the perception that they are no longer positively advancing new theories and interpretations, but merely fighting a defensive rearguard action. The works of this school are published by left-wing publishers such as Aoki Shoten and Otsuki Shoten, which serves to emphasise its increasing marginalisation. Kasahara Tokushi did publish Nankin jiken (The Nanjing incident) from the left-wing, but much more mainstream, Iwanami Shoten as recently as 1997. This work, however, inadvertently used a fabricated photograph, and Kasahara was forced to make an embarrassing and public apology (typically, Iris Chang used the same photograph in her work after it had been exposed in Japan as a fake).[36]

One of the great strengths of this school has been its continued efforts to bring together, translate and publish the primary sources on the Nanjing Incident. Moreover, as noted above, a large group within the school has begun to revise its numbers downwards. The result is that the differences between the estimates of the death toll within the Great Massacre School are greater than the differences between an Eguchi in the Great Massacre School and a Hata in the Middle-of-the-Road School: this suggests a narrowing of the gap between at least some of the professional historians in the two groups. Finally, it should be noted that the Great Massacre School is aging, and is not showing the steady influx of new blood that is so characteristic of the Illusion School.

5. A New Categorization

The three schools -- Illusion, Middle-of-the-Road, and Great Massacre Schools -- are well established in Japan, and this categorisation will therefore continue to be used when discussing the debate there. However, in analysing the debate both within and especially outside Japan, these categories are arguably misleading. I believe that a better way would be to divide the various positions between the serious 'historians' and the 'myth-makers'.[37]

Many members of the Illusion School actively construct mythologised narratives of the past that serve the political, ideological and emotional needs of neo-nationalists by minimising or dismissing all charges of Japanese atrocities. They are highly selective in utilising primary materials, restricting their gaze to those that reaffirm the master narrative of Japanese as victims of malicious Chinese slander. Nor are some above misusing primary materials and their use of oral history is confined to individuals who reinforce their views. Some members of the Great Massacre School have similarly distorted history, for example through fabricating stories in the case of several layperson authors (a well-known example is Azuma Shiro), or through the uncritical use of materials from the Tokyo Trial or Chinese-language materials (Hora Tomio and Honda Katsuichi spring to mind). However, the most important contributions to the understanding of the NI have come from professional historians of both the GMS and the Middle-of-the-Road School. The work of these historians underlines the integrity of the historiographical process of reconstructing history based on an informed and self-critical interpretation of the primary and to a certain extent the secondary materials. The focus of these historians on the primary materials, allows (and actually forces) members to change their minds and draw different conclusions as new sources emerge. The result has been a significant narrowing of differences between members of the two schools with respect, for example, to the numbers killed.

The advantage of a classification that looks at the orientation of the researcher towards either processes or outcomes is that when it is used to analyse the debate on Nanjing, it clarifies and highlights the similarities among some members of each of the three schools. It can also be used to a far greater extent in examining the debate in English. The new classification would divide researchers into three groups: right-wing myth-makers (many members of the Illusion School) or 'corpse minimisers', left-wing myth-makers (hard-line members of the Great Massacre School) or 'corpse maximisers', and finally historians (the Middle-of-the-Road School, plus some of the academic historians in the Great Massacre School). The first two groups are dominated by lay members, the last by academics. When examining the debate outside Japan, it is clear that much of the Chinese literature would fall within the left-wing myth-maker or corpse maximiser group, as would some of the literature published in English (Iris Chang, for instance). Much of the literature published in English, however, is the work of historians. 

6. Methodologies and Sources 

The individual methodologies used to discuss the Nanjing Incident have been summarised by Hata Ikuhiko according to the four methods by which he believes the number of victims in Nanjing can be counted: oral history, burial records, data sampling, and Japanese army field reports.[38] These will be briefly summarised.

Oral history has provided some important insights, but it must be emphasised that it is arguably the most problematic methodology in researching the incident. Those who rely mainly on Chinese oral sources produce one set of figures on the scale of the massacre and the brutality of the Japanese that cannot be substantiated by any other methodology, whereas some of those who rely solely on Japanese oral sources have denied that any massacre occurred, again a claim that cannot be substantiated. Given that the issues are so highly charged, distortions, deliberate or otherwise, must be anticipated. The historian will attempt to corroborate oral history where possible with written sources, and be cautious of using material that cannot be substantiated. The historian will also attempt to verify Japanese oral history with accounts from Chinese oral history, and vice-versa. At the same time, it must be acknowledged that the Chinese of Nanjing were frequently illiterate, and certainly lacked the opportunity under Japanese rule to set down on paper their experiences, so that the more skeptical one is of oral sources, the greater the degree to which the Chinese voice is denied. Given that the voice of the victims can in some cases only be reconstructed through interviews, oral history certainly has a role to play. Just as officially or documentary sources must be verified, so too must oral sources. Given the fact that the Nanjing Atrocities occurred over 60 years ago, the opportunities for new research in this area are rapidly fading.[39]

The second methodology is to examine the burial records. Although any such examination is doubtless an important step in any overall reconstruction of the events in Nanjing, this methodology also has its limits, the main one being that the lack of complete contempory records makes for much guess-work. I have attempted such an examination, juxtaposing the various primary sources against the burial records in order to shed light on their reliability.[40] Although these records are at best partial, an examination of the primary sources casts much light on the burial effort in and around Nanjing.

The third methodology is data sampling, of which there is only one case. This was L. S. C. Smythe, War Damage in the Nanking Area: December 1937 to March 1938 (1938). Smythe was a sociologist who conducted an extensive survey of Nanjing in early 1938 in the immediate aftermath of the Japanese atrocities. He was well qualified to conduct such a survey, having received his PhD in sociology from the University of Chicago, and had experience in conducting at least two similar surveys in 1931 and 1932. Smythe's survey was conducted in two areas: within the city walls of Nanjing and in the surrounding rural areas. In the City Survey, investigators surveyed every 50th inhabited house. The survey covered the whole of the city inside the walls, together with areas just outside some of the gates, and took place from 9 March to 2 April 1938, with some supplementary work from 19 to 23 April 1938. The Agricultural Survey was conducted over 2,438 square miles in 4.5 counties around Nanjing. These surveys produced much data that has yet to be properly analysed. No other survey was carried out in and around Nanjing so soon after the city fell. Surprisingly few authors have made extensive use of this piece of documentation.[41]

The final methodology, the examination of Japanese army field reports, also has its limits. The Japanese military was very rigorous with regard to some aspects of what it reported (how many rounds of ammunition were used on any particular day, for instance, or how many Japanese soldiers were killed), but at the same time individual units regularly inflated the number of enemy soldiers killed on the battlefield (an examination of the rounds of ammunition expended may in some cases shed some light on the true Chinese death toll). This methodology has been extensively utilised by Hata Ikuhiko, Masahiro Yamamoto, and the authors of Nankin senshi.

The above methodologies can be defined by the sources they use. The other primary sources that exist are the diaries, letters and other documents written by members of the three major groups in Nanjing: the 'bystanders', members of the international community in Nanjing, the Chinese 'victims', and the Japanese 'perpetrators'.[42] Hata does not believe that a close analysis of this set of sources can provide a means by which the number of victims in Nanjing can be counted. However, I am convinced that he is wrong. The various documents authored by members of the international community in particular provide a great deal of (reasonably objective) information, but again these have not been adequately utilised in the English language literature. Indeed to the best of my knowledge, I am the first to have identified precisely the membership of the Western community in Nanjing at the time in any language.[43] There are a number of accounts in Chinese that are said to have been written by Chinese individuals who were in Nanjing during the early occupation.[44] Some of these at least are clearly false in parts (reporting conversations with members of the International Committee who had left the city, for instance), and almost certainly were the products of Chinese government propaganda. More work needs to be done to identify what is genuine, and to make greater use of it in telling the story of occupied Nanjing as experienced by the Chinese residents of the city. The diaries of a large number of Chinese military personnel have been brought together and published in works such as Yuan Guomindang jiangling kang-Ri zhanzheng qinliji: Nanjing baoweizhan (Personal experiences of former Nationalist Party generals in the War of Resistance against Japan: The defence of Nanjing) (1987), and so for the first time it is now possible to review the Chinese military experience of fighting the Japanese.[45] None of this material is available in English, and Yamamoto and myself are perhaps the only researchers to have begun to use this treasure trove of information in reconstructing the history of Nanjing in English.[46] Many Japanese accounts only began to appear long after the event, and in many cases have to be treated with some caution; 'diaries' are not always products of the winter of 1937-38, for instance, but reconstructions written decades later with particular political objectives in mind.

A final source of information is the records of the Nanjing and Tokyo Trials (many of the burial records were in fact drawn up for the post-war trials of the Japanese held responsible for the atrocities committed in Nanjing).[47] These records again have to be treated with some caution. The perpetrators, the Japanese on trial, obviously had very strong motives for giving false testimony (some were facing the death penalty), but some aspects of the testimony of other witnesses can also easily be shown to be false. This can be explained perhaps by the long lapse of time between the events and the trials, although a desire for revenge cannot be completely ruled out. As a result, secondary materials based solely or mainly on the post-war military tribunals held in Nanjing and Tokyo have to be treated with some caution and skepticism (the work of Hora Tomio, for example, is a case in point). 

7. Recent Trends

The debate in Japan underwent a sea change as the full implications of John Rabe's diary were digested (Hata Ikuhiko among others speaks of the 'Rabe effect') and as Iris Chang's book was absorbed. Although the flood of publications continues, there are signs of an emerging consensus. Rabe has clearly destroyed much of the basis for the more extreme casualty estimates of the Great Massacre School, but also makes it absolutely clear that he was convinced that the Japanese army was responsible for looting, arson, rape and the execution of thousands of men identified as 'ex-soldiers'.[48] He has thus been most vigorously denounced by members of the Illusion School. However, it must be said that the greatest impact in the long term will probably be felt among the ranks of the Great Massacre School, members of which have already begun to revise their numbers downwards. For instance, in the recent English translation of his The Nanjing Massacre, the 'corpse maximiser' Honda Katsuichi has significantly reduced his estimate of the scale of the Japanese atrocities in and around Nanjing. As Frank Gibney notes in his introduction, Honda now believes that 'a bit over 100,000' is the true figure for the scale of the massacre during the Nanjing Incident.[49] Kasahara derives a similar figure based on Rabe's estimate of 50,000 to 60,000 for both civilians and soldiers, including soldiers killed in action, to which is then added a second figure of 80,000 soldiers (this assumes that 90,000 soldiers died, of whom 10,000 died in action, and 80,000 were executed).[50] In other words, at least some members of the Great Massacre School appear to have accepted Rabe's estimate, but use it for civilians only, despite the fact that Rabe clearly states that at least 30,000 of this estimate were soldiers killed in combat, and despite the fact that his estimate of the civilian death toll in an official report to the German Embassy was 'thousands'.[51] Although Honda's revised estimate is a product of the Rabe Diary, the text itself contains an earlier, pre-Rabe estimate. Honda here asserts that 'we need to treat as a single phenomenon the approximately three months from November through January of the assault on Nanjing' -- an assertion that matches his later arguments -- but then goes on to state that, once the time-frame (and geography) is thus broadened, 'we are dealing with too much time to say anything specific about the numbers of people killed, but no one can deny that the victims of the massacre numbered in the hundreds of thousands'.[52] The English translation of his work thus contains both the 'old' orthodox figure of 'hundreds of thousands' in the main text and the 'new' orthodox figure of 100,000 plus in the introduction.

As noted above, Iris Chang's work has energized the Illusion School, providing an easy target for a group to score endless goals in their attempts to demonstrate that the debate on Nanjing is biased, and based on willful ignorance if not deliberate fabrications. The School is now showing increasing signs of dominating the debate: in terms of quantity, if not quality, for instance, four of the five most recent works on Nanjing were published by members of the Illusion School, and only one by members of the Great Massacre School.

In addition to the impact of Rabe and Chang, a second trend in Japan is the internationalisation of the debate. Honda's work was the first to be translated into English, but was quickly followed by one of the major figures in the Illusion School, Tanaka Masaaki.[53] In another sign of internationalisation, one of the recent works on Nanjing was originally published in both English and Japanese.[54] Although both corpse minimisers and corpse maximisers are publishing in English, this volume is the first to publish translations of some of the historians.

A third trend is the increasing interest shown in the debate by Western academics who are aware of and well-versed in both the Japanese-language and Chinese-language literature. The work edited by Joshua Fogel is perhaps the best example of this, but others such as Timothy Brook and Bob Wakabayashi are also doing highly original research that is bound to change general perceptions of Nanjing. Wakabayashi, for instance, has recently published a paper on the story of the competition between two Japanese officers to see who could first kill (decapitate) a hundred Chinese with their swords.[55] This competition emerged as part of war-time Japanese mythology -- the two officers were said to have charged enemy (Chinese) machine-gun positions, where each decapitated over a hundred enemy soldiers in combat -- but in the post-war world was reworked into a competition to execute POWs, becoming a major part of the myth of Nanjing in both the English- and Chinese-language literature. Wakabayashi's paper is the best piece of academic research on this competition in any language. A related trend is, as noted above, the increasing number of edited volumes of primary materials published in English.

A fourth trend has been the recent attempts to shed light on aspects of Nanjing long ignored in the Japanese debate and its fixation on the number of victims. An outstanding example of this is Timothy Brook who, in an as yet unpublished paper, examines the first collaborationist regime established in Nanjing, the Autonomous Government Committee, and in particular one of its members, Jimmy Wang.[56] Elsewhere, Brook examines the Reformed Government (Weixin Zhengfu) that replaced the Self Government Committee.[57] The story of occupied Nanjing, and the links established between the Japanese rulers and Chinese ruled, has long been overlooked in the debate, and Brook's work enlarges our understanding of the event. I have written two papers on a related topic, the International Committee for the Nanking [Nanjing] Safety Zone and its experience of Japanese rule.[58] The discussion of the entire discourse on Nanjing and a comparison with the discourse on the holocaust -- in other words an analysis not of the history of Nanjing but of its historiography -- is again a relatively new theme that is providing new and fruitful insights. The work of Daqing Yang and Joshua Fogel here is especially sophisticated.[59] Kanemaru Yuichi has recently published ground-breaking research on the fate of many of the books and other cultural treasures in areas of Central China, including Nanjing.[60]

There has also been a recent attempt to overcome some of the limits of the mindsets that underlie much of the previous literature on Nanjing. For instance, one common (if unconscious) assumption that can be seen in much of the English-language literature on Nanjing is the notion of Chinese as feminised and Orientalised 'passive' non-actors. To give a single example, a large part of Nanjing was destroyed by fire during the early weeks of occupation. Despite the official Chinese scorched-earth policy, the well-known existence within the walls of Nanjing of large numbers of Chinese military personnel, the extensive looting of the city by Chinese as well as Japanese, and the fact that it was in the interest of the Japanese to maintain a viable urban centre once they had captured it (just as much as it was in the interest of the Chinese government to deny the Japanese this centre), this arson has long been implicitly, if not explicitly, attributed to the Japanese. While it is clear that much of the arson was indeed the work of Japanese soldiers, the possibility that some of it may have been the result of Chinese sabotage has been almost completely overlooked. An examination of whether a Chinese resistance movement existed within Nanjing also remains virgin territory.[61]

Researchers have been very reluctant to examine certain topics that will (I suspect) increasingly become the focus of attention. For instance, the basic assumption that the Japanese were all evil and the Chinese all innocent victims, while emotionally satisfying, does not permit a complete historical account. To reach a deeper understanding of the events in and around Nanjing, a number of disturbing questions will have to be asked. Was the Chinese decision to make a stand at Nanjing, despite the large numbers of civilians trapped within its walls, the correct one? What of the decision to fight to the death, to (in some cases) lock the doors of concrete emplacements to prevent Chinese soldiers from fleeing? Did the Chinese use of what were known as 'plain-clothes soldiers' (soldiers fighting in civilian clothes) contribute to the executions by Japanese of plain-clothes male civilians of weapons-carrying age? Did the Chinese military's decision to have at least some units change out of military uniform after Nanjing fell and hide among the civilian population contribute to such executions? Was the Japanese decision to execute men in civilian clothes found (in some cases at least) with weapons hiding among the civilian population legal? The English language literature here may well come into its own. Many Japanese clearly would be extremely reluctant to tackle these issues, and many of these questions will remain taboo in the Chinese-language discourse for the foreseeable future. To ask these questions is not to deny the atrocities that doubtless occurred in and around Nanjing, but merely to demonstrate that the historical discourse on Nanjing must acknowledge that the issue is more complex than a black-and-white, good-versus-evil position might otherwise suggest.

There is another set of issues that, to my mind, somewhat surprisingly remains unresearched and unanalysed. So much heat has been generated over the question of numbers, that the question of explaining the Nanjing Incident within the context of the early stages of the Sino-Japanese War has been overlooked. Was the experience of Nanjing qualitatively and/or quantitatively different to that of other Chinese cities that fell to the Japanese, and if so why?
There were not comparable massacres each time the Japanese took a city. So why at Nanjing? A possible factor is that Japanese high command learned some lessons from Nanjing and took steps to prevent it later. A possible answer is that Japanese military leaders deliberately used the massacre to terrorise Chinese in an attempt to end the war as quickly as possible. But another answer may be that far from emanating from the top commanders, there was simply a breakdown of the chain of command at Nanjing. Yet another approach would be to locate the Nanjing Incident against systematic terror (encapsulated by the term the 'three alls') against civilians conducted by Japan in the course of the war. I am not attempting to support one or another of these theories here, but to note my impression that, to date, too little attention has been paid to this entire set of large issues. And to wonder why.

8. Conclusion

In this paper, I have attempted to shed some light on the state of research of the Nanjing Incident today.

First, to put it mildly, Nanjing is a controversial topic. Although our understanding of the events of Nanjing does not even begin to approach our understanding of the holocaust, in both instances there have been tendencies to demonise anyone who budges from the orthodox position. The problem is that the orthodox position is completely different in China and Japan, and within Japan itself three distinct orthodoxies have emerged. Although there is real debate in Japan, no one there now accepts the figure of 300,000 victims as plausible, while in China the figure is set in concrete (both figuratively and literally) at the entrance of the Memorial for the Compatriot [Chinese] Victims of the Japanese Massacre in Nanjing. It would be unfortunate if the debate were to continue to run on parallel lines, never to come together to produce a deeper, more complete and transnational understanding of this historical event. How to overcome the obstacles, on the other hand, poses a dilemma. As long as much of the debate is dominated by ideologues, the sensible option for historians may well be to keep their heads down and research other topics.[62] That, however, cannot be a desirable outcome. Historians surely have an obligation to combat the tendency to use Nanjing as a weapon in contemporary ideological and international contests.

Second, too many Japanese authors are insensitive to the fact that Nanjing for better or for worse has become a central plank in the construction of the modern Chinese identity. To discuss Nanjing is to threaten this self-identity. Those who participate in the debate, therefore, need to show some sensitivity to it. I am not arguing that the Chinese orthodoxy needs to be accepted without question. Indeed, my own position is that it is more dangerous (at least in the long term) to found national identity on a lie than to discover the truth and live with it. However, some effort does need to be made (at both extremes of the debate) to avoid the use of violent and inflammatory language, and to show a much greater awareness of and sensitivity to the moral implications of historical inquiry.

Third, historians are obliged to examine calmly the primary materials and reconstruct the history of Nanjing on the basis of what those materials say. Some who write on Nanjing clearly want to absolve the Japanese of all blame, while others want to depict the Japanese as a uniquely brutal and ruthless race. Neither position should form the starting point of any discussion of the events in Nanjing -- although, of course, either might be the conclusion of any such examination. The publication of as many primary materials as possible is clearly an essential condition to this approach, so the discovery and publication of as much as possible must be encouraged.

Finally, a dialogue among historians working on the Nanjing Incident needs to be promoted. One such forum can be provided by the English language, where researchers from both Japan and China can debate with researchers from third-party countries such as the USA, Canada Britain and Australia. The problem with the Chinese and Japanese language discourses is that they are both so insular and the political environments are so charged. It is in the market of ideas and through constant debate (including, perhaps, the mediation provided by third-party historians), that the truth will be approached.


Anon., 1997, 'Nihon de renso "Nankin daigyakusatsu" 83% ... Chugoku kyosanto no kikanshi ga chosa' (83 per cent think of the 'great Nanjing massacre' when asked about Japan ... A [Chinese] Communist Party paper survey), Asahi shinbun, 17 February 1997 (evening edition), p. 2.

Anon., 1998, 'The Woman Who Wouldn't Forget', Reader's Digest, 1 September 1998, available at

Ara Ken'ichi, 1987/2002, 'Nankin jiken' Nihonjin 48-nin no shogen (The 'Nanjing incident': The testimony of 48 Japanese), Tokyo: Shogakkan Bunko.

Askew, David, 2000, 'Honda Katsuichi, The Nanjing Massacre: A Japanese Journalist Confronts Japan's National Shame', Asian Studies Association of Australia (Electronic Journal:, no. 10, August.

Askew, David, 2001a, 'The Nanjing Incident: An Examination of the Civilian Population', Sino-Japanese Studies, vol. 13, no. 2, March, pp. 2-20.

Askew, David, 2001b, 'The Japanese Debate on the Nanjing Incident: An Overview', Asian Studies Association of Australia (Electronic Journal:, no. 15, May.

Askew, David, 2002a, 'The International Committee for the Nanking Safety Zone: An Introduction', Sino-Japanese Studies, vol. 14, April, pp. 3-23.

Askew, David, 2002b 'The Nanjing Incident: Recent Research and Trends', Electronic Journal of Contemporary Japanese Studies (

Askew, David, 2002c, 'Kitamura Minoru, 'Nankin jiken' no tankyu -- Sono jitsuzo o motomete', Japanese Studies, vol. 22, no. 1, May, pp. 79-82.

Askew, David, 2002d, 'Nanking 1937-38: The Evidence from the Diary of John Rabe', Ritsumeikan Journal of Asia Pacific Studies, vol. 9, pp. 95-122.

Askew, David, 2003a, 'Defending Nanking: An Examination of the Capital Garrison Forces', Sino-Japanese Studies, vol. 15, April, pp. 148-73. Askew, David, 2003b, 'Buried Bodies, Hidden History: Towards an Estimate of the Scale of Japanese Atrocities in Nanjing', Ritsumeikan Journal of Asia Pacific Studies, vol. 13.

Askew, David, 2004, 'The Contested Past: History and Semantics in the Nanjing Debate', Ritsumeikan International Affairs, March (forthcoming).

Askew, David, forthcoming a, 'Westerners in Occupied Nanking: December 1937 to February 1938', in Bob Tadashi Wakabayashi (ed.), The Nanking Atrocity: Complicating the Picture (forthcoming).

Askew, David, forthcoming b, 'Population and Victimization Statistics', in Bob Tadashi Wakabayashi (ed.), The Nanking Atrocity: Complicating the Picture (forthcoming).

Askew, David, unpublished manuscript, 'The Military Death Toll in Nanking: An Examination of the Fate of the Capital Garrison Forces'.

Azuma Shiro san no Nankin saiban o sasaeru kai (ed.), 2001, Kagai to yurushi -- Nankin daigyakusatsu to Azuma Shiro saiban (Harm and forgiveness: The great Nanjing massacre and the Azuma Shiro trial), Tokyo: Gendai Shokan.

Brook, Timothy, 2001a, 'The Creation of the Reformed Government in Central China, 1938', in David P. Barrett and Larry N. Shyu (eds.), Chinese Collaboration with Japan, 1932-1945: The Limits of Accommodation, Stanford, California: Stanford University Press, 2001.

Brook, Timothy, 2001b, 'The Tokyo Judgement and the Rape of Nanking', The Journal of Asian Studies, vol. 60, no. 3, August, pp. 673-700.

Brook, Timothy, forthcoming, 'Sino-Japanese Collaboration during the Rape of Nanking (Jimmy Wang's Tour of Duty along the Border between Good and Evil)', in Bob Tadashi Wakabayashi (ed.), The Nanking Atrocity: Complicating the Picture (forthcoming).

Brook, Timothy, (ed.), 1999, Documents of the Rape of Nanking, Ann Arbor: University of Michigan Press.

Buruma, Ian, 1999, 'The Joys and Perils of Victimhood', New York Review of Books, April 8, pp. 4-9.

Chang, Iris, 1997, The Rape of Nanking: The Forgotten Holocaust of World War II, New York: Basic Books.

Cohen, Paul A., 1997, History in Three Keys: The Boxers as Event, Experience, and Myth, New York: Columbia University Press.

Eykholt, Mark, 2000, 'Aggression, Victimization, and Chinese Historiography of the Nanjing Massacre', in Joshua A. Fogel (ed.), The Nanjing Massacre in History and Historiography, Berkeley and Los Angeles, California: University of California Press, 2000.

Fogel, Joshua A., 1998, Review of The Rape of Nanking: The Forgotten Holocaust of World War II, by Iris Chang, Journal of Asian Studies, vol. 57, no. 3, August, pp. 818-19.

Fogel, Joshua A., 1999a, 'Correspondence: How Bad Was the Nanking Massacre?', Letters to the editor, The Los Angeles Times, 15 August.

Fogel, Joshua A., 1999b, 'Airisu Chan no egaku Nankin jiken no gonin to henken' (The misunderstandings and biases of the Nanjing Incident as depicted by Iris Chang), Sekai, no. 667, November, pp. 252-57.

Fogel, Joshua A., 2000, 'The Controversy over Iris Chang's Rape of Nanking', Japan Echo, vol. 27, no. 1, February, pp. 55-57.

Fogel, Joshua A. forthcoming, 'The Nanking Atrocity and Chinese Memory', in Bob Tadashi Wakabayashi (ed.), The Nanking Atrocity: Complicating the Picture (forthcoming).

Fogel, Joshua A., (ed.), 2000, The Nanjing Massacre in History and Historiography, Berkeley and Los Angeles, California: University of California Press.

Gao Xingzu et al., 1962, 'Riben giguozhuyi zai Nanjing datusha' (Japanese imperialism and the massacre in Nanjing), translated by Robert Gray at, last accessed 21 March 2001.

Gibney, Frank, 1999, 'Editor's Introduction', in Honda Katsuichi, The Nanjing Massacre: A Japanese Journalist Confronts Japan's National Shame, Armonk, New York: M. E. Sharp, 1999.

Hata Ikuhiko, 1986, Nankin jiken -- 'Gyakusatsu' no kozo (The Nanjing incident: The structure of a 'massacre'), Tokyo: Chuko Shinsho.

Hata Ikuhiko, 1993, 'Ronsoshi kara mita Nankin gyakusatsu jiken' (The Nanjing massacre as seen from a history of the debate [on Nanjing]), in Showashi no nazo o ou (The puzzles of Showa history), Tokyo: Bungei Shunju, vol. 1.

Hata Ikuhiko, 1995, Nanjing datusha zhenxiang: Riben jiaoshou de lunshu, Hong Kong: Shangwu Yinshudian. Translation of Hata (1986).

Hata Ikuhiko, 1998a, Gendaishi no soten (Issues of modern Japanese history), Tokyo: Bungei Shunju.

Hata Ikuhiko, 1998b, 'The Nanking Atrocities: Fact and Fable', Japan Echo, vol. 25, no. 4, August 1998, pp. 47-57.

Hata Ikuhiko, 1998/1999, 'Nankin gyakusatsu jiken -- Kazu no kosatsu' (The Nanjing massacre: Examining the numbers), in Hata Ikuhiko, Gendaishi no hikari to kage (Light and shadow: Essays on modern history), Tokyo: Gurafusha, 1999. Translation of Hata (1998b).

Hayase Toshiyuki, 1999, Shogun no shinjitsu -- Nankin jiken Matsui Iwane jinbutsuden (The general's truth: A biography of Matsui Iwane of the Nanjing incident), Tokyo: Kojinsha.

Hein, Laura and Mark Selden, eds., 2000, Censoring History: Citizenship and Memory in Japan, Germany and the United States, Armonk: M.E. Sharpe.

Higashinakano Shudo, 1998, 'Nankin gyakusatsu' no tettei kensho (A thorough investigation of the 'Nanjing massacre'), Tokyo: Tendensha.

Higashinakano Shudo, 2001, Quedi jianzheng 'Nanjing datusha' (A thorough investigation of the 'Nanjing massacre'), translated by Qiu Zhenrui, Taipei: Qianwei Chubanshe. Translation of Higashinakano (1998).

Higashinakano Shudo and Fujioka Nobukatsu, 1999, 'Za x Reipu x obu x Nankin' no kenkyu -- Chugoku ni okeru 'johosen' no teguchi to senryaku (Research on The Rape of Nanjing: China's methods and strategy in the 'information war'), Tokyo: Shodensha.

Higashinakano Sh�d� (ed.), 2002, Nankin 'gyakusatsu' kenky� no saizensen (Heisei 14-nen-ban) � Nihon 'Nankin' gakkai nenp� (The frontlines of research of the Nanjing 'massacre': The 2002 Japan Nanjing Research Association annual report), Tokyo: Tendensha.

Higashinakano Sh�d� (ed.), 2003a, 1937 Nankin k�ryakusen no shinjitsu � Shin shinry� hakkutsu (The truth of the 1937 battle for Nanjing: A new material unearthed), Tokyo: Sh�gakkan Bunko.

Higashinakano Sh�d� (ed.), 2003b, Nankin 'gyakusatsu' kenky� no saizensen (Heisei 15-nen-ban) � Nihon 'Nankin' gakkai nenp� (The frontlines of research of the Nanjing 'massacre': The 2003 Japan Nanjing Research Association annual report), Tokyo: Tendensha.

Hoffman, Michael, 2001, 'Internet Hate-Porn Novels Inflame Historical Passions', The Japan Times, June 3.

Honda Katsuichi, 1999, The Nanjing Massacre: A Japanese Journalist Confronts Japan's National Shame, Armonk, New York: M. E. Sharp.

Honda Katsuichi, Watanabe Harumi, and Hoshi T�ru, Nankin daigyakusatsu rekishi kaizanha no haiboku � Ri Sh�ei meiyo kison saiban kara mirai e (The defeat of the rewriters of the history of the great Nanjing massacre: From the Li Xiuying libel case to the future), Tokyo: Ky�ikushiry� Shuppankai.

Horo Tomio (ed.), 1973, Nankin jiken (The Nanjing incident), vol. 9 of Nicchu sensoshi shiryo henshu iinkai (ed.), Nicchu Sensoshi shiryo (Historical materials of the Sino-Japanese War), Tokyo: Kawade Shobo Shinsha.

Hora Tomio (ed.), 1985, Nicchu senso Nankin daizangyaku jiken shiryoshu (Materials on the great Nanjing atrocity of the Sino-Japanese War), vol. 2, Eibun shiryohen (English materials), Tokyo: Aoki Shoten.

Hora Tomio, 1987, Nanjing datusha (The great Nanjing massacre), translated by Mao Lianghong and Zhu Agen, Shanghai: Shanghai Yiwen Chubanshe.

Hora Tomio, Fujiwara Akira, and Honda Katsuichi (eds.), 1987, Nankin jiken o kangaeru (Thinking about the Nanjing incident), Tokyo: Otsuki Shoten.

Hora Tomio, Fujiwara Akira, and Honda Katsuichi (eds.), 1988, Nankin daigyakusatsu no genba e (To the site of the great massacre of Nanjing), Tokyo: Asahi Shinbusha.

Hora Tomio, Fujiwara Akira, and Honda Katsuichi (eds.), 1982, Nankin daigyakusatsu no kenkyu (Research of the great Nanjing massacre), Tokyo: Banseisha.

Hsu Shuhsi (ed.), 1939, Documents of the Nanking Safety Zone, Shanghai: Kelly & Walsh.

Hu, Hua-ling, 2000, American Goddess at the Rape of Nanking: The Courage of Minnie Vautrin, Carbondale: Southern Illinios University Press.

Igarashi Zennojo, 2000, Ketteiban Nankin jiken no shinjitsu (The truth of the Nanjing incident: The final word), Tokyo: Bungeisha.

Igarashi Zennojo, 2002, Nankin jiken no hansei to heiwa no kochiku ni tsuite (On regrets for the Nanjing incident and the creation of peace), Tokyo: Bungeisha.

Inoue Hisashi, 1987, 'Nankin jiken to itai maiso mondai' (The Nanjing incident and the issue of burials), in Hora Tomio, Fujiwara Akira, and Honda Katsuichi (eds.) Nankin jiken o kangaeru (Thinking about the Nanjing incident), Tokyo: Otsuki Shoten, 1987.

Inoue Hisashi, 1988, 'Itai maiso kara mita Nankin jiken giseishasu' (The number of victims in the Nanjing incident as seen from the burials of the dead), in Hora Tomio, Fujiwara Akira, and Honda Katsuichi (eds.), Nankin daigyakusatsu no genba e (To the Site of the Great Massacre of Nanjing), Tokyo: Asahi Shinbusha, 1988.

Ishida Yuji (ed.), 2001, Shiryo Doitsu Gaikokan no mita Nankin jiken (Materials: The Nanjing incident witnessed by German Diplomats), Tokyo: Otsuki Shoten.

Itakura Yoshiaki, 1986, 'Matsui Iwane Nikki no kaizan ni tsuite' (On the Fabrication of Matsui Iwane's Diary), Bungei Shunju, January, pp. 186-194.

Itakura Yoshiaki, 1999, Honto wa ko datta Nankin jiken (The truth about the Nanjing incident), Tokyo: Nihon Tosho Kankokai.

Kanemaru Yuichi, 2000, 'Chushi Kensetsu Shiryo Seibi Iinkai to sono Shuhen -- "Shina Jihen"-ki Nihon no tai-Chugoku chosa katsudo o meguru shusaku' (A Study of the 'Chushi Kensetsu Shiryo Seibi Iinkai' during the 'China Incident', 1937-1942), Ritsumeikan Keizaigaku, vol. 49, no. 5, December, pp. 93-127.

Kanemaru Yuichi, 2002, 'Kyokuron no keifu -- Nankin jiken-ki ni okeru tosho ryakudatsu mondai no kensho' (The genealogy of a distorted theory: An examination of the issue of looting at the time of the Nanjing incident), Ritsumeikan kokusai gengo bunka kenkyu, vol. 40, no. 2, pp. 123-38.

Kasahara Tokushi, 1992, 'Nankin boeisen to Chugokugun' (The battle to defend Nanjing and the Chinese army), in Hora Tomio, Fujiwara Akira, and Honda Katsuichi (eds.), Nankin daigyakusatsu no kenkyu (Research of the great Nanjing massacre), Tokyo: Banseisha, 1992.

Kasahara Tokushi, 1995, Nankin nanminku no hyakunichi -- Gyakusatsu o mita gaikokujin (The 100 days of the Nanjing refugee [safety] zone: Foreigners who witnessed the massacre), Tokyo: Iwanami Shoten.

Kasahara Tokushi, 1997, Nankin jiken (The Nanjing incident), Tokyo: Iwanami Shinsho.

Kasahara Tokushi, 1998, "Shinsho Nankin jiken no keisai shashin ni tsuite" (On the photograph used in my book Nankin jiken [Nanking incident]), Tosho, April. Available at

Kasahara Tokushi, 1999, Nankin jiken to sanko sakusen -- Mirai ni ikasu senso no kioku (The Nanjing incident and the three alls: Remembering the war for the future), Tokyo: Otsuki Shoten.

Kasahara Tokushi, 2001, book review of Joshua A. Fogel ed. (2000) and Timothy Brook ed. (1999), Monumenta Nipponica, vol. 56, no. 2, pp. 266-69.

Kasahara Tokushi, 2002, Nankin jiken to Nihonjin -- Senso no kioku o meguru nashonarizumu to gurobarizumu (The Nanjing incident and the Japanese: Nationalism and globalism in memory of the war), Tokyo: Kashiwa Shobo.

Katogawa Kotaro, '"Shogen ni yoru Nankin senshi" -- Sono sokatsuteki kosatsu' (A history of the battle of Nanjing from testimony: An overview), Kaiko, March.

Kitamura Minoru, 2001, 'Nankin jiken' no tankyo -- Sono jitsuzo o motomete (An enquiry into the 'Nanjing incident': The search for the true picture), Tokyo: Bunshun Shinsho.

Manne, Robert, 2002, 'Blind to Truth, and Blind to History', Sydney Morning Herald, 16 December 2.

Matsumura Toshio, 1998, 'Nankin gyakusatsu' e no daigimon (Serious doubts about the 'Nanjing massacre'), Tokyo: Tendensha.

Matsuo Ichiro, 2003, Puropagandasen 'Nankin jiken' -- Hiroku shashin de miru 'Nankin daigyakusatsu' no shinjitsu (The Nanjing incident as a propaganda war: The truth of the 'great Nanjing massacre' as seen through secret photographs), Tokyo: Kojinsha.

'Nanjing datusha' shiliao bianji weiyuanhui and Nanjing tushuguan (eds.), 1997, Qin-Hua Rijun Nanjing datusha shiliao (Source materials relating to the great massacre committed by the Japanese army in Nanjing), Nanjing: Jiangsu Guji Chubanshe.

Nankin jiken chosa kenkyukai (ed.), 1992, Nankin jiken shiryoshu (Materials on the Nanjing incident), vol. 1, Amerika kankei shiryohen (American materials), Tokyo: Aoki Shoten.

Nankin jiken chosa kenkyukai (ed.), 1999, Nankin daigyakustsu hiteiron 13 no uso (Thirteen lies by the deniers of the great Nanjing massacre), Tokyo: Kashiwa Shobo.

Nankin senshi henshu iinkai (ed.), 1993a, Nankin senshi (A history of the battle of Nanjing), expanded and revised edition, Tokyo: Kaikosha.

Nankin senshi henshu iinkai (ed.), 1993b, Nankin senshi shiryoshu 1 (A history of the battle of Nanjing: Collected materials, vol. 1), revised and expanded edition, Tokyo: Kaikosha.

Nankin senshi henshu iinkai (ed.), 1993c, Nankin senshi shiryoshu II (A history of the battle of Nanjing: Collected materials, vol. 2), revised and expanded edition, Tokyo: Kaikosha. Aoki Shoten.

Pritchard, R. John, and Sonia Magbanua Zaide, 1981, The Tokyo War Crimes Trial, New York: Garland Publishers.

Rabe, John, 1997a, Der gute Deutsche von Nanking, herausgegeben von Erwin Wickert, Stuttgart: Deutsche Verlags-Anstalt.

Rabe, John, 1997b, Nankin no shinjitsu (The truth of Nanjing), Tokyo: Kodansha. Translation of Rabe (1997a).

Rabe, John, 1998, The Good Man of Nanking: The Diaries of John Rabe, edited by Erwin Wickert, translated by John E. Woods, New York: A. A. Knopf.

Sandall, Roger, 2002, 'Mr Windschuttle Versus the Professors', The Australian, 23 December.

Shokun! 2001, 'Ketteiban 'Nankin jiken' saishin hokoku' (The final word: The latest report on the Nanjing incident), Shokun!, February, pp. 128-212.

Smalley, Martha Lund, (ed.), 1997, American Missionary Eyewitnesses to the Nanking Massacre, 1937 - 1938, New Haven: Yale Divinity School Library.

Smythe, L. S. C., 1938, War Damage in the Nanking Area: December 1937 to March 1938. Urban and Rural Surveys, by Dr. Lewis Strong Casey Smythe and Assistants, on behalf of the Nanking International Relief Committee, Compiled June 1938, Shanghai: Mercury Press.

Suzuki Akira, 1982, edited by Cheng Jiashan et al., Tusha (Massacre), Taipei: Yuancheng Chubanshe.

Suzuki Akira, 1999, Shin 'Nankin daigyakusatsu' no maboroshi (The illusion of the great Nanjing massacre: Further thoughts), Tokyo: Asuka Shinsha.

Takemoto Tadao and Ohara Yasuo, 2000, The Alleged 'Nanking Massacre': Japan's Rebuttal to China's Forged Claims. Saishin 'Nankin daigyakusatsu' -- Sekai ni uttaeru Nihon no enzai, Tokyo: Meiseisha.

Tanaka Masaaki, 2000, What Really Happened in Nanking: The Refutation of a Common Myth, Tokyo: Sekai Shuppan.

Tanaka Masaaki, 1987/2001, Nankin jiken no sokatsu -- Gyakusatsu hitei no ronkyo (The Nanjing incident summed up: The grounds of the denial of a massacre), Tokyo: Tendensha.

Timperley, H. J., (ed.), 1938a, What War Means: The Japanese Terror in China -- A Documentary Record, London: Victor Gollancz.

Timperley, H. J., (ed.), 1938b, The Japanese Terror in China, New York: Modern Age Books.

Tomisawa Shigenobu, 2003, Nankin jiken no kakushin -- Detabesu ni yoru jiken no kaimei (The core of the Nanjing incident: Solving the incident through a data base), Tokyo: Tendensha.

Unemoto Masaki, 1998, Shinso Nankin jiken -- Rabe nikki no kensho (The True Nanjing incident: An examination of the Rabe diary), Tokyo: Kenpakusha.

Vallace, Charles, 1998, 'America o yurugasu "Za x Reipu x obu x Nankin"' (The Rape of Nanking shocks America), Chuo Koron, August. Also available at

Vautrin, Minnie, 1999, Nankin jiken no hibi -- Minii x Botorin no nikki (Living the Nanjing incident: The diary of Minnie Vautrin), translated by Okada Ryonosuke and Ihara Yoko, Tokyo: Otsuki Shoten.

Wakabayashi, Bob Tadashi, 2000, 'The Nanking 100-Man Killing Contest Debate: War Guilt Amid Fabricated Illusions', Journal of Japanese Studies, vol. 26, no. 2, Summer, pp. 307-40.

Wakabayashi, Bob Tadashi, 2001, 'The Nanjing Massacre: Now You See It,...', Monumenta Nipponica, vol. 56, no. 4, Winter, pp. 521-44.

Wakabayashi, Bob Tadashi, (ed.), forthcoming, The Nanking Atrocity: Complicating the Picture.

Yamamoto, Masahiro, 2000, Nanking: Anatomy of an Atrocity, Westport, Conn.: Praeger.

Yang, Daqing, 2000, 'The Challenges of the Nanjing Massacre: Reflections on Historical Inquiry', in Joshua A. Fogel (ed.), The Nanjing Massacre in History and Historiography, Berkeley and Los Angeles, California: University of California Press, 2000.

Yang, Daqing, 2001, 'The Malleable and the Contested: The Nanjing Massacre in Postwar China and Japan', in T. Fujitani, Geoffrey M. White and Lisa Yoneyama (eds.), Perilous Memories: The Asia Pacific War(s), Durham and London: Duke University Press, 2001.

Yoshida, Takashi, 2000, 'A Battle over History: The Nanjing Massacre in Japan', in Joshua A. Fogel (ed.), The Nanjing Massacre in History and Historiography, Berkeley and Los Angeles, California: University of California Press, 2000.

Yoshimoto Sakae, 2001, Nankin daigyakusatsu no ouso -- Naze itsumade makaritoru ka (The lies of the great Nanjing massacre: Why do they continue to remain unchallenged?), Tokyo: Tokyo Tosho Shuppankai.

Zhang Kaiyuan (ed.), 2001, Eyewitnesses to Massacre: American Missionaries Bear Witness to Japanese Atrocities in Nanjing, Armonk, New York: M. E. Sharpe.

Zhongguo dier lishi dang'anguan and Nanjing shi dang'anguan (eds.), 1997, Qin-Hua Rijun Nanjing datusha dang'an (Archival documents relating to the great massacre committed by the Japanese army in Nanjing), Nanjing: Jiangsu Guji Chubanshe.

Zhongguo Guomindang zhongyang weiyuanhui dangshi weiyuanhui (ed.), 1987a, Geming wenxian (Revolutionary documents), vol. 108, Rijun zai-Hua baoxing -- Nanjing datusha (Atrocities committed by the Japanese Army in China: The great Nanjing massacre), vol. 1, Taipei: Zhongyang Wenwu Gongyingshe.

Zhongguo Guomindang zhongyang weiyuanhui dangshi weiyuanhui (ed.), 1987b, Geming wenxian (Revolutionary documents), vol. 109, Rijun zai-Hua baoxing -- Nanjing datusha (Atrocities committed by the Japanese Army in China: The great Nanjing massacre), vol. 2, Taipei: Zhongyang Wenwu Gongyingshe.

Zhongguo renmin zhengzhi xieshanghuiyi quanguo weiyuanhui wenshi ziliao yanjiu weiyuanhui 'Nanjing baoweizhan' bianshenzu (ed.), 1987, Yuan Guomindang jiangling kang-Ri zhanzheng qinliji: Nanjing baoweizhan (Personal experiences of former Nationalist Party generals in the War of Resistance against Japan: The defence of Nanjing), Beijing: Zhongguo Wenshi Chubanshe.

Zhongyang dang'anguan, Zhongguo dier lishi dang'anguan, and Jilinsheng shehui kexueyuan (eds.), 1995, Nanjing datusha (The great Nanjing massacre), Beijing: Zhonghua Shuju.

Zhu Chengshan et al. (eds.), 1998, Qin-Hua Rijun Nanjing datusha waiji renshi zhengyanji (A collection of the testimony of foreigners about the invading Japanese military's great Nanjing massacre), Nanjing: Jiangsu Renmin Chubanshe.


[1] My own views of Nanjing have evolved in part through a long-running e-mail conversation with Bob Wakabayashi. I would like to take this opportunity to acknowledge the impact Bob has had. I would also like to express my gratitude to A. G. Geddes and Peter Kirby. Both read through and commented on drafts of this paper. An early version of this paper was published as Askew (2002b), and is reprinted here with the permission of the editor of the Electronic Journal of Contemporary Japanese Studies.

[2] Even the terminology for the incident itself is controversial. Of the various terms that have been used, 'Nanjing Incident' is, I believe, the most neutral, and one that is particularly beneficial in combating the arguments of the deniers. While it is possible, after all, to deny that a 'great massacre' occurred (depending on one's definition of 'great' and 'massacre'), it is impossible to deny that an 'incident' occurred.

[3] On this point, see, for instance, Buruma (1999). The Chinese diaspora has played a major role in bringing the Nanjing Incident to international attention. There is no simple explanation for the fact that overseas Chinese in countries such as the USA have felt the need to place such an emphasis on the atrocities at Nanjing. The desire for a strong national identity is understandable, but it is unfortunate that victimhood should play such a crucial role in the construction of this identity. It is possible that a shared sense of outrage and victimhood helps create links between members of the community. It is perhaps even possible, as at least one American journalist has speculated, that within a system of preferential treatment, there are social, political and economic gains to be had from being defined as a member of an ethnic group that has suffered a large-scale atrocity. See Vallace (1998).

[4] Chang (1997: chapter 10).

[5] For the terms 'corpse minimiser' and "corpse maximiser', see Askew (2004), Manne (2002) and Sandall (2002).

[6] Anon. (1997). Also cited in Yang (2001: 50).

[7] For an example of the anti-Japan and anti-Japanese sites that exist, see Hate Japan (, cited in Hoffman (2001). For other organisations, see, for instance, the Chinese Alliance for Commemoration of the Sino-Japanese War Victims (, the Society for Studies of Japanese Aggression against China (Riben qin-Hua yanjiu xuehui), and the Alliance for Preserving the Truth of the Sino-Japanese War ( Japan boasts several associations, including the left-wing Nankin jiken chosa kenkyukai (The Nanjing incident survey and research association) and the right-wing Nankin jiken gakkai (The Nanjing incident studies association). For journals, see, in addition to the various publications of the Nankin jiken gakkai, including Higashinakano ed. (2002, 2003b), Kang-Ri zhanzheng yanjiu (The journal of studies of the Chinese War of Resistance against Japan) and Riben qin-Hua yanjiu (Studies of Japanese aggression against China), both of which frequently publish papers on Nanjing. One of the best sites on Nanjing (authored by a Japanese) is Nankin jiken shiroshu (Materials on the Nanjing incident) (, while another superb site is Masato Kajimoto's 'Online Documentary: The Nanking Atrocities' ( Also see Nankin daigyakusatsu wa uso da (The great Nanjing massacre is a lie!) ( and Nankin jiken shiro (Materials on the Nanjing incident) ( For major Chinese sites, see the WWW Memorial Hall of the Victims in the Nanjing Massacre (1937-1938) ( Also see The Nanjing Massacre ( and The Association for Victims of Japanese War Crimes ( For Iris Chang as a best seller, see for instance anon. (1998), an article in an issue of the Reader's Digest with a photograph of Chang on the cover. The article is available on Chang's 'official' home page (

[8] Even left-wing historians in Japan use 'incident', though often first attempt to please all by stating that the term 'Nanjing incident' is short for 'great Nanjing massacre incident' (Nankin daigyakusatsu jiken).

[9] For some of the better collections, see, for instance, Zhongguo renmin zhengzhi xieshanghuiyi quanguo weiyuanhui wenshi ziliao yanjiu weiyuanhui 'Nanjing baoweizhan' bianshenzu ed. (1987), Zhongyang dang'anguan, Zhongguo dier lishi dang'anguan, and Jilinsheng shehui kexueyuan eds. (1995), 'Nanjing datusha' shiliao bianji weiyuanhui and Nanjing tushuguan eds. (1997), and Zhongguo dier lishi dang'anguan and Nanjing shi dang'anguan eds. (1997). For a representative collection published in Taiwan, see the two volume set, Zhongguo Guomindang zhongyang weiyuanhui dangshi weiyuanhui ed. (1987a, 1987b). Some of the English language material has also been translated. See, for example, Zhu et al. eds. (1998).

[10] This can also be said about many of the photographs used by Iris Chang in her work.

[11] See Gao et al. (1962). The word 'misused' is from Eykholt (2000: 25). The 'imperialist and fascist' members of the Western community had been accused a decade previously of helping the Japanese to massacre innocent Chinese. See Xinhua yuebao, 1951, cited in Yang (2000: 178, note 93). Also see Eykholt (2000: 24-25).

[12] Chang (1997: 139). Chang has not misused her sources here: together with a number of other events depicted in her book, this is pure fiction. The issue of Iris Chang requires further research. Her book has been used in Japan by corpse minimisers to discredit all those who argue that atrocities took place. Given that Chang is so obviously wrong on important questions of fact, the conclusion drawn by some is that if Chang is wrong, then all who argue that atrocities took place are equally wrong. As rhetoric, this methodology is very effective, but clearly inappropriate. For a hard-hitting example, see Higashinakano and Fujioka (1999). The most significant reviews of Chang have been written by Joshua Fogel. See Fogel (1998, 1999a, 1999b, 2000).

[13] For my own views of the International Committee, see Askew (2002a, forthcoming a). Also see Kasahara (1995).

[14] For representative translations, see Hata (1995), Higashinakano (2001), Hora (1987), and Suzuki (1982).

[15] I have discussed what I see as the major flaws of Honda's work -- in particular his over-reliance on and uncritical acceptance of oral sources -- in Askew (2000).

[16] In this paper, I have endeavoured to give the author's names as they appear in the works I cite: the work cited here by Masahiro Yamamoto, for instance, gives his name in this order, rather than Yamamoto Masahiro.

[17] For an overview of the post-war debate in Japanese on Nanjing, see Yoshida (2000). Wakabayashi (2000) provides a very good introduction to the early debate.

[18] On this point, see Katogawa (1985: 18).

[19] Rabe's diary was published in Japanese as Nankin no shinjitsu (The truth of Nanjing). See Rabe (1997b).

[20] For the Japanese Society for History Textbook Reform, see their web-site (in English), The Japanese version of the same (the Atarashii rekishi kyokasho o tsukuru kai) is at For a critical discussion, see Hein and Selden eds. (2000).

[21] For the Association for the Advancement of a Liberal View of History, see and, in Japanese,

[22] For recent monographs alone, see, for example, Azuma Shiro san no Nankin saiban o sasaeru kai ed. (2001), Hayase (1999), Higashinakano (1998), Higashinakano and Fujioka (1999), Higashinakano ed. (2002, 2003a, 2003b), Honda, Watanabe, and Hoshi (2003), Igarashi (2000, 2002), Itakura (1999), Kasahara (1999, 2002), Kitamura (2001), Matsuo (2003), Nankin jiken chosa kenkyukai ed. (1999), Suzuki (1999), Takemoto and Ohara (2000), Tomisawa (2003), and finally Yoshimoto (2001). Reflecting the high interest in Nanjing in Japan, several works have recently been republished. See, for instance, Tanaka (1987/2001), a work originally published in 1987 and no longer easily available, and Ara (1987/2002), again originally published in 1987.

[23] Timperley ed. (1938a) and Hsu ed. (1939) are both available in translation in Hora ed. (1973, 1985). Rabe (1997b). Vautrin (1999). Note that parts of Vautrin have been published in Zhang ed. (2001).

[24] Kasahara (2001: 266-67). This tradition of publishing collections of primary materials in Japanese is still thriving: for recent works, see Minnie Vautrin, Nankin jiken no hibi -- Minii x Botorin no nikki (Living the Nanjing incident: The diary of Minnie Vautrin) (1999), and Ishida Yuji ed., Shiryo Doitsu gaikokan no mita Nankin jiken (Materials: The Nanjing incident witnessed by German diplomats) (2001).

[25] Hata (1993).

[26] Kasahara Tokushi, unpublished paper distributed at the Nanjing Incident Symposium held at Princeton University on 21 November 1997, cited in Hata (1998a: 12). This paper has also been printed in Japanese in Kasahara (1999: 238-48). See Kasahara (1999: 240).

[27] Shokun! (2001) For a review of this questionnaire, see Askew (2001b), upon which the following is based.

[28] Kasahara's latest work (Kasahara 2002: chapter 2) contains a section that attempts to highlight the flaws in Higashinako's work. A translation is available in Wakabayashi ed. (forthcoming).

[29] Nankin senshi henshu iinkai ed. (1993a, 1993b, 1993c)

[30] For my own views on Kitamura, see Askew (2002c). Note that he publishes to a large extent in the Illusion School press, and shows signs of an inclination to move closer to the Illusion School.

[31] This is not to deny the inherent role ideology plays in all the social sciences, including historiography. For a stimulating work on the historian's approach to reconstructing history as compared with those interested in ideology and myths, see Cohen (1997).

[32] Honda Katsuichi's Nankin e no michi (The road to Nanjing) has recently been published in English. It may be that the Great Massacre School, having arguably lost the intellectual debate in Japan over the Nanjing Incident, has decided to concentrate on English. See Honda (1999).

[33] Kasahara (2001: 266).

[34] Itakura Yoshiaki, among others, has identified a number of cases where members of both groups have fabricated materials and stories on the incident. For representative work, see him on the 'corpse maximiser' Sone Kazuo in Itakura (1999: 233-283). Also see him on the 'corpse minimiser' Tanaka Masaaki in Itakura (1986).

[35] Nankin jiken chosa kenkyukai ed. (1999).

[36] For Kasahara's apology, see Kasahara (1998). Iwanami Shoten subsequently reissued this work with a different photograph.

[37] The following owes much to Cohen (1997).

[38] Hata (1998b, 1998/1999).

[39] I have interviewed and am corresponding with one surviving member of the Nanjing Special Services Organisation that was entrusted with the task of overseeing the Chinese collaborationist regimes in Nanjing, and am also interviewing a surviving member of the Sixth Division, but doubt that there can be many more individuals who were there and who are still alive.

[40] See Askew (2003b).

[41] See Smythe (1938). I utilised this survey in establishing the size of the civilian population of Nanjing from December 1937 when the city fell through to February the next year. See Askew (2001a, forthcoming b).

[42] The terms 'bystander', 'victim' and 'perpetrator' have been borrowed from Yang (2000: 138).

[43] See Askew (2002a). Kasahara (1995) is a good introduction to the international community and utilises many of the various sources that are available.

[44] For a collection of these recollections, see for example 'Nanjing datusha' shiliao bianji weiyuanhui and Nanjing tushuguan eds. (1997).

[45] Zhongguo renmin zhengzhi xieshanghuiyi quanguo weiyuanhui wenshi ziliao yanjiu weiyuanhui 'Nanjing baoweizhan' bianshenzu ed. (1987).

[46] Yamamoto (2000) and Askew (2003a, unpublished manuscript). Also see Itakura, 'Nankin senshi to Nankin jiken no suryoteki haaku' (Nankin senshi [A history of the battle of Nanking] and a mathematical grasp of the Nanking incident) in Itakura (1999), Kasahara (1992), and Nankin senshi henshu iinkai ed. (1993a).

[47] For a stimulating and highly sophisticated discussion of the Tokyo Trial as it related to Nanjing, see Brook (2001b).

[48] For a more detailed discussion of Rabe, see Askew (2002d).

[49] Gibney (1999: xiii).

[50] See Kasahara (1999: 239). The assumption that 90,000 soldiers died must be viewed with some scepticism, and there certainly is no basis whatsoever for the claim that eight out of nine who died were executed.

[51] Letter from John Rabe, included in 'Reports from the German Foreign Office in China to the German Foreign Office in Berlin', cited in Pritchard and Zaide eds. (1987: vol. 2, 4594).

[52] Honda (1999: 285).

[53] See Tanaka (2000).

[54] See Takemoto and Ohara (2000).

[55] Wakabayashi (2000).

[56] Brook (forthcoming).

[57] Brook (2001a).

[58] Askew (2002a, forthcoming a).

[59] Yang (2000) and Fogel (forthcoming).

[60] See Kanemaru (2000, 2002).

[61] I have started to examine this possibility and tentatively concluded that there almost certainly was some form of organised resistance. See Askew (2002d).

[62] This point has been forcefully made to me by a number of Japanese historians.

David Askew is Associate Professor, Ritsumeikan Asia Pacific University and Monash University. This article was originally published by Japan Focus and is used by permission. Copyright Japan Focus, 2004. Click here for the original version, which includes diacriticals.


The views expressed above are those of the author and do not necessarily reflect those of AsiaMedia or the UCLA Asia Institute. You may comment on this and other AsiaMedia articles and essays at: