The Nanking Safety Zone: Twenty-Two Westerners and a Quarter of a Million Refugees

Members of the International Committee for the Nanking Safety Zone, the group of Western residents who ran a refuge of about 3.86 square kilometres for some 200,000–250,000 people. Photo: Unknown photographer, Memorial Hall of the Victims in Nanjing Massacre collection, PD China via Wikimedia Commons.

The Rape of Nanking · Chapter 5 of 10

The Nanking Safety Zone

Twenty-two Westerners who stayed, hundreds of Chinese staff and volunteers, about 3.9 square kilometres of the city, and a quarter of a million people who had nowhere else to go.

22 November 1937 – 18 February 1938Xinjiekou · Gulou · Ginling · Xiaofenqiao~28 min read

When the Japanese army entered Nanjing on 13 December 1937, most of the people still in the city were crowded into a district in its north-western quarter that a committee of foreign residents had declared a Safety Zone three weeks earlier. The zone was about 3.9 square kilometres (1.5 square miles), roughly a fourteenth of the area inside the Ming walls. It held the University of Nanking and its hospital, Ginling College, the American embassy, several government ministries and the new residential district where diplomats and officials had lived. By the committee’s estimate some 250,000 people sheltered inside it, in 25 or 26 organised camps and in private houses, while the city around them was looted, burned and turned into a killing ground.

The zone did not keep the Japanese army out. Soldiers entered it from the first day to search for Chinese soldiers who had shed their uniforms, and they took away thousands of men, raped women in the camps and looted the houses. What the zone did was to give the civilian population a place where foreigners were present, food was distributed, a hospital still operated and every crime that came to the committee’s attention was written down and sent to the Japanese embassy. Most historians agree that it saved many thousands of lives. It also produced the largest body of contemporary evidence about the massacre.

This chapter tells the zone’s story from its model in Shanghai to its dissolution in February 1938: the people who ran it, Western and Chinese; its boundaries, camps and food; the protest letters and case reports; the Japanese refusal to recognise it; and the rescues outside the walls at the Jiangnan Cement Works and Qixia Temple. Minnie Vautrin’s Ginling College camp is the subject of Chapter 4, and John Rabe and the German thread of the story are followed further in Chapter 6.

A note on evidence. Almost everything here rests on contemporary documents written inside the zone: the committee’s letters and case reports (printed by Hsü Shuhsi in 1939 and reprinted in Timothy Brook’s Documents on the Rape of Nanking, 1999), the diaries of Rabe, Vautrin and others, the letters of the missionaries now at Yale Divinity Library, and Lewis Smythe’s 1938 survey. Population and camp numbers are the committee’s own estimates, made in impossible conditions; the historian David Askew’s studies (2001, 2002) are the most careful modern examination of them and are followed where they differ from older accounts.

The Shanghai Model: Father Jacquinot’s Zone

The idea came from Shanghai. On 9 November 1937, as the Chinese army was retreating from the city after three months of fighting, the French Jesuit Robert Jacquinot de Besange opened a neutral refugee zone of about one square kilometre in the northern part of Shanghai’s old Chinese city, Nanshi. With the agreement of both the Chinese and Japanese authorities, it sheltered about 250,000 people at the height of the fighting, according to reports of 1938, and remained in operation until 1940. The International Committee of the Red Cross later cited the “Jacquinot Zone” as a precedent when the Fourth Geneva Convention of 1949 provided for safety zones for civilians (Jacquinot de Besange).

In Nanjing, the idea was taken up after Shanghai fell in mid-November. The proposal is associated with Hang Liwu, chairman of the board of the University of Nanking, and with a group of American missionaries: W. Plumer Mills of the Presbyterian mission, the historian Miner Searle Bates and the sociologist Lewis Smythe, both of the University of Nanking, who met on 17 November to draw up a plan. On 22 November a meeting of foreign residents formed the International Committee for the Nanking Safety Zone and elected the German businessman John Rabe, head of the Siemens office in Nanjing, as its chairman. A German, and a member of the Nazi Party, was thought more likely to be heard by the Japanese, Germany’s partners in the Anti-Comintern Pact.

The committee then sought the consent of both sides. It cabled its proposal to the Japanese authorities through the American consul general in Shanghai on 24 November, asking that the zone be spared in return for the removal of all soldiers and military installations; Father Jacquinot acted as an intermediary in Shanghai. On 27 November the committee met Chiang Kai-shek and General Tang Shengzhi, the commander of the Nanjing garrison, and on 1 December the mayor, Ma Chaojun, handed the committee administrative responsibility for the zone, together with police, rice, flour, salt and money (Facing History; Hsü, 1939). The Japanese reply never amounted to recognition: the zone would be respected, in effect, only as far as military necessity allowed.

Rabe also tried to use his German connections at the highest level. He telegraphed Adolf Hitler, asking him to urge the Japanese government to respect the zone; the telegram went unanswered (Deutsche Biographie). In the last days before the siege, the committee stocked the zone with what it could gather, posted its flags, and opened the camps on 8 December, when the first refugees arrived from the suburbs outside the walls.

Boundaries: About 3.9 Square Kilometres

The zone’s borders followed streets. On the east it ran along Zhongshan Road from the great circle at Xinjiekou north to the Drum Tower (Gulou), then north-west along Zhongshan North Road to Shanxi Road Plaza. On the north it ran west from Shanxi Road Plaza to Xikang Road; on the west, south along Xikang Road and then south-east to the junction of Shanghai Road and Hanzhong Road; on the south, east along Hanzhong Road back to Xinjiekou. The area enclosed was about 3.86 square kilometres, the figure most sources now give; older accounts round it to “about 4 square kilometres” or “about 2 square miles.”

The district was chosen because it lay away from the main defensive lines on the eastern and southern sides of the city, contained many foreign properties and institutions that might deter shelling and had space to house refugees: college campuses, schools, ministry buildings and the new residential area, many of whose houses had been vacated by officials who left for the interior. Inside it the committee divided the population into districts under Chinese heads, organised a police force and posted its flag at the corners.

The Committee and the Twenty-Two

The International Committee had about fifteen members at its founding, Americans, Germans, British and a Dane. Four British businessmen from the trading and oil companies and several others left before the city fell, as their firms and embassies evacuated staff upriver. By 13 December, according to Askew’s reconstruction from the committee’s own lists, 22 Western residents remained in the city, plus five journalists who left on 15 and 16 December (Askew, 2002, Table 3). Askew notes that the true number of foreign residents may have been slightly higher, about 24 to 26.

Five journalists also stayed through the fall: F. Tillman Durdin of the New York Times, Archibald Steele of the Chicago Daily News, C. Yates McDaniel of the Associated Press, Leslie C. Smith of Reuters and the Paramount newsreel cameraman Arthur Menken. They left on 15 and 16 December, and their dispatches gave the outside world its first reports of the mass executions. After that the committee’s members were the only Western witnesses in the city until the diplomats returned in January, and their letters, diaries and Magee’s film became the outside world’s main sources.

Fourteen of the 22 were Americans, almost all missionaries or staff of mission institutions: at the University Hospital, the surgeon Robert O. Wilson, the physician C. S. Trimmer, the hospital administrator James McCallum, the laboratory technician Grace Bauer and the nurse Iva Hynds; at the University of Nanking, Searle Bates, Charles Riggs and Lewis Smythe; at Ginling, Minnie Vautrin; W. Plumer Mills of the Northern Presbyterian mission; Hubert Sone of the Nanking Theological Seminary; George Fitch, secretary of the YMCA; and John Magee and Ernest Forster of the American Church Mission (Episcopal). Five were Germans: Rabe of Siemens, Eduard Sperling of the Shanghai Insurance Company, Christian Kröger of Carlowitz and Co., R. Hempel of the North Hotel and A. Zautig of Kiessling and Bader. Rupert Hatz, a mechanic, was Austrian; Cola Podshivoloff of Sandgren’s Electric and the mechanic A. Zial were White Russians.

The committee’s officers reflected its mixture. Rabe was chairman. Smythe was secretary and wrote most of its letters. Kröger became treasurer on 29 November. Mills was vice-chairman and succeeded Rabe as chairman in February 1938. Fitch served as the zone’s director, Sperling headed the zone police, and Hang Liwu served as a deputy until 28 November; he later escorted Palace Museum treasures to Hankou. A separate International Red Cross Committee of Nanking, chaired by John Magee with Forster as secretary and Kröger as treasurer, ran medical relief, with Wilson, McCallum and Vautrin among its members.

The Chinese Who Ran the Zone

The Western names are the ones that survived in English, but the zone could not have functioned without its Chinese staff, who outnumbered the foreigners many times over and ran most of its daily operations. The committee’s secretariat was headed by Tang Zhongmu. The housing commission, the largest, was chaired by Wang Ding with Charles Riggs as deputy and divided the zone into eight districts plus a new district, each under a Chinese head. The sanitation commission under Shen Yushu began with about 400 workers. The food commission was directed by Han Xianglin, Rabe’s Chinese colleague at Siemens, with the Red Swastika Society leader Tao Xisan among its members (Wikipedia, citing Chinese archival studies; Hsü, 1939).

These men faced dangers the foreigners did not. On 15 December six sanitation workers were shot in the street. On 17 December Japanese troops took away 50 uniformed police and 45 volunteer police from the zone, and 40 more uniformed police the next day; most were presumed killed. On 28 December the head of the zone’s fourth district, Wang Xinglong, was seized and killed by Japanese soldiers. Half the Chinese staff left Nanjing once conditions eased in early 1938, and many never recorded their experiences, so that, as Chinese historians of the zone note, the work of many of them cannot now be verified in detail.

Xu Chuanyin, vice-chairman of the Nanjing branch of the Red Swastika Society, a Buddhist-Daoist charity, worked in the zone’s housing administration; his society’s burial teams collected tens of thousands of bodies in the months after the fall. He was among the Chinese witnesses who testified in person before the Tokyo tribunal in 1946.

Refugees: How Many People?

Nanjing had about a million inhabitants before the war. By December most who could leave had left, by river, rail and road, for the interior. How many remained is one of the central questions about the massacre, because every estimate of the death toll has to fit inside it.

The committee’s own figure, used in its letters to the Japanese, was that about 200,000 people were in the city when it fell, and that most of them, eventually some 250,000, were inside the zone, with not more than about 10,000 outside it. Askew, working through the committee’s documents, the Japanese registration and Smythe’s survey, puts the civilian population of the city in late December and early January at 200,000 to 250,000, with a best estimate of about 224,500. The Japanese registration of December and January counted about 160,000 people, excluding children under ten and some older women; Smythe’s survey of spring 1938 gave 221,150 (Askew, 2001). The camps themselves held fewer than the zone: about 50,000 people on 17 December, rising to nearly 70,000 by 21 December, with the rest in houses, compounds and public buildings.

Twenty-Five Camps

The committee organised the refugees into 25 camps by Fitch’s count, 26 by others, in buildings it controlled or requisitioned: university and school campuses, ministries, clubs and missions. Some were run by the committee’s staff, some by their institutions’ own foreign or Chinese staff and a few by the refugees themselves. The largest were on the mission campuses, where the foreigners lived and could intervene at night.

Rabe’s own compound at Xiaofenqiao No. 1 was among the smallest and best known. He had built an air-raid shelter in the garden during the autumn bombing, and when the city fell neighbours, employees and strangers crowded into the garden and the house. He spread a large Nazi flag in the garden so that Japanese aircraft would recognise the property as German; on the ground, he used his standing as a German and a party member to order soldiers out. The camp counted 602 people on 4 January 1938.

Rice, Porridge and Coal: Feeding the Zone

Food was the committee’s most constant problem. When the mayor transferred responsibility on 1 December, the city promised the committee 20,000 sacks of rice and 10,000 of flour, with salt and cash; transport problems meant that much of it never arrived before the siege, and the committee’s records show it holding 9,067 bags of rice, 1,000 bags of flour and 350 bags of salt, with cash grants of about 100,000 to 120,000 dollars over time. Some stocks outside the zone were seized by the Japanese army after the fall (Wikipedia, citing Chinese studies of the committee’s accounts; Hsü, 1939).

The food commission sold rice cheaply to those who could pay and distributed it free to the destitute, and the camps ran porridge kitchens. Allocations were small and fixed: the records show the Army School camp drawing ten bags of rice a day, Ginling twelve, the Hankou Road school four, the Bible Teachers Training School five every two days. The University of Nanking Middle School gave ten bags a day free to the poor and let the better-off buy two. Coal for cooking was as scarce as rice. Food and fuel are among the most frequent subjects of the committee’s letters to the embassy through January.

The University Hospital, on Zhongshan Road at the zone’s eastern edge, was the main hospital still working; the committee kept three hospitals open in all (Facing History). Wilson, Trimmer, McCallum, Bauer, Hynds and their Chinese nurses treated civilians with bullet, bayonet and burn wounds, and survivors of rape and of mass executions. Wilson’s letters to his family, written in these weeks, are among the most exact contemporary descriptions of the injuries inflicted (Brook, 1999). Magee filmed some of the patients there.

Sixty-Nine Letters and Four Hundred Cases

From the first day, the committee did what it could do best: it wrote. On 14 December Rabe sent the first protest letter to the Japanese commander. From then until 19 February, Smythe wrote 69 letters to the Japanese embassy on the committee’s behalf, reporting crimes, asking for military police and food, and proposing practical measures to restore order (Yale Divinity Library, Smythe). Attached to many of them were numbered case reports: short, dated descriptions of individual incidents, giving place, time and, where known, the names of witnesses. By early February the committee had forwarded about 450 such cases, according to standard accounts. The IMTFE noted that Smythe “filed two protests a day for the first six weeks.”

The embassy’s officials received the letters and, the committee members believed, passed them on; they were sympathetic in person and powerless in practice against the army. In 1939 the Chinese political scientist Hsü Shuhsi published the letters and case reports in Shanghai as Documents of the Nanking Safety Zone, and they became evidence at the Tokyo tribunal. They remain the single most important contemporary record of what happened inside the zone.

Smythe, the Secretary of the Int. Committee for the Safety Zone, filed two protests a day for the first six weeks.

— Judgment of the International Military Tribunal for the Far East, ch. VIII, 1948 (text)

Smythe also produced the zone’s most systematic piece of evidence. Between March and June 1938, with about 20 student investigators of the Nanking International Relief Committee, he surveyed one household in fifty in the city and a sample of farm families in the surrounding counties, recording deaths, injuries, abductions and losses. Published in June 1938 as War Damage in the Nanking Area, it found, for the city, 2,400 civilians killed by soldiers’ violence and 4,200 taken away, most of whom were never seen again, and 26,870 killed by violence in the rural districts. Smythe stressed that the survey counted only families that survived to be interviewed and excluded soldiers; its use and limits are discussed in Chapter 3.

Disarmed Soldiers and the Searches

The zone’s founders had promised both sides that it would contain no soldiers. On 12 and 13 December that promise collapsed with the Chinese defence. As Tang Shengzhi’s withdrawal order failed and the crossings at Xiaguan jammed (see Chapter 2), thousands of Chinese soldiers threw away their weapons and uniforms and fled into the zone in civilian clothes. The committee had no means of keeping them out, and hoped that disarmed men would be treated as prisoners of war. They were not.

The Japanese army’s answer was a series of sweeps through the zone in the days after the fall. Units went through the camps and houses picking out men they judged to be soldiers and marched them away in columns. Many were civilians. The army’s own records give a sense of the scale: the 7th Infantry Regiment, which swept the zone, reported that it captured and executed 6,670 people there between 13 and 24 December (Askew, 2002, citing Hata Ikuhiko). It is a perpetrator record, compiled for a unit report, and it covers one regiment in one sector over twelve days. The IMTFE later found that more than 20,000 men of military age were marched out of the city and killed in this way.

Late in December the army began its registration of the population. At some of the camps, men were told that those who admitted past military service would be spared and put to work; according to later accounts, more than 200 who came forward were taken away and executed. The killings that followed the searches are described in Chapter 3.

The Japanese Army and the Zone

The Japanese authorities never formally recognised the Safety Zone. The army regarded it as an obstacle: a district full of Chinese men of military age, some of them former soldiers who had thrown away their uniforms, protected by foreigners who could not be fully controlled. At least one Japanese newspaper had argued before the assault that the zone would hinder it. From 14 December units entered at will. They searched the camps for soldiers and took away men by the hundred, many of whom were executed; they took women; they looted the houses of the new residential area, including foreign property; and they set fires.

Individual foreigners stood in the way when they could. Rabe, Sperling, Kröger, Fitch and others repeatedly ordered soldiers out of houses and compounds and were threatened for it; on 22 December Kröger was assaulted while trying to help a wounded man. The committee’s letters show a steady pattern of promises from the embassy, a short improvement, and renewed violence. The appointment of a few military police to the camps helped at times; at others the guards themselves were among the offenders.

The occupiers also set up their own Chinese administration. A Nanjing Self-Government Committee was inaugurated under Japanese sponsorship on 1 January 1938. On 28 January a Japanese special-service official, accompanied by a representative of that committee, ordered the camp leaders to send the refugees home, and two days later the authorities set a deadline of 4 February, threatening forced eviction and banning trade inside the zone. Many refugees left; some came back after finding their homes destroyed or after being assaulted at home. Ginling, under Vautrin, was allowed to stay open for women who had nowhere safe to go.

Outside the Walls: Jiangnan Cement and Qixia Temple

The zone was not the only refuge. About 20 kilometres north-east of the city, near Qixia, the Jiangnan Cement Works, a new plant being built by the Danish engineering firm F. L. Smidth, was left in the care of two foreigners: Bernhard Arp Sindberg, a 26-year-old Dane hired as a caretaker who arrived on 2 December 1937, and Karl Günther, a German engineer. As the front passed, they flew the Danish and German flags, neutral and friendly in Japanese eyes, over the works and opened the grounds to the people of the surrounding villages. They set up a makeshift clinic. Estimates of how many people sheltered there range from about 6,000 to 20,000, and the numbers changed over the months; no exact count survives (Sindberg; Harmsen, 2024).

Sindberg photographed what he saw. His album, with his own captions, is among the few contemporary visual records of the countryside after the advance. After about three months the Japanese authorities lost patience with him and he was sent to Shanghai. Nearby, the Buddhist monks of Qixia Temple also sheltered thousands of refugees in the temple compound through the winter.

The Other Faces of the Zone

Behind the officers stood a small group who did most of the night work: answering calls to houses where soldiers had broken in, escorting women, driving rice to the camps and writing everything down. Many of them left letters or diaries that are now at Yale Divinity Library and in Brook’s and Zhang Kaiyuan’s collections. Eduard Sperling ran the zone police and, as Rabe’s diary records, repeatedly confronted soldiers in the act. James McCallum, the hospital administrator, kept a diary from 19 December 1937 to 15 January 1938 that was entered in evidence at the Tokyo tribunal (Yale, McCallum). Ernest Forster, newly transferred to Nanjing in November 1937, worked with Magee at St. Paul’s Church and wrote detailed letters (Yale, Forster). Charles Riggs, an agricultural engineer, was deputy head of housing and was slapped by a Japanese soldier alongside the American diplomat John Allison on 26 January 1938.

The End of the Zone

By February 1938 the occupation authorities wanted the zone gone. Its existence was an embarrassment and a rival to the Japanese-sponsored administration, and its foreigners were witnesses. American and German diplomats had returned in January (the Americans on 6 January, the Germans on 9 January), which gave the committee new channels but also reduced the case for its separate authority. On 18 February the International Committee for the Nanking Safety Zone renamed itself the Nanking International Relief Committee, a body for relief only, and notified the embassies. Smythe’s last letter in the series is dated 19 February.

Rabe left for Shanghai and Germany later that month; accounts give 23 February or a few days later. Mills became chairman of the relief committee, and Hubert Sone its administrative director after Fitch left. The last camps closed in late May and June 1938. The relief committee continued smaller programmes into 1939 and was formally dissolved in 1940.

Most of the foreigners stayed in China. Smythe moved west to Chengdu with the University of Nanking in 1938; Mills was interned after Pearl Harbor; Vautrin remained at Ginling until 1940. Several of them, Wilson, Bates and Magee among them, testified in Tokyo in 1946, and their documents became the backbone of the prosecution’s case, as Chapter 8 describes.

1. 9 Nov 1937Father Jacquinot opens the Nanshi refugee zone in Shanghai.
2. 17 NovMills, Bates and Smythe draft a plan for a Nanjing zone.
3. 22 NovInternational Committee formed; John Rabe chairman.
4. 24 NovProposal cabled to the Japanese via the US consul general in Shanghai.
5. 27 NovCommittee meets Chiang Kai-shek and Tang Shengzhi.
6. 1 DecMayor Ma Chaojun hands administration of the zone to the committee.
7. 8 DecRefugees begin to arrive in the camps.
8. 13 DecNanjing falls.
9. 14 DecRabe’s first protest; Smythe’s series of 69 letters begins.
10. 17–18 DecJapanese troops seize about 135 zone police.
11. 22 DecKröger assaulted while helping a wounded man.
12. 28 DecDistrict head Wang Xinglong killed.
13. 1 Jan 1938Japanese-sponsored Self-Government Committee inaugurated.
14. 4 Jan602 people in Rabe’s garden.
15. 6 and 9 JanAmerican, then German, diplomats return.
16. 28 JanCamp leaders ordered to send refugees home.
17. 4 FebDeadline for refugees to leave the camps.
18. 18 FebCommittee renamed Nanking International Relief Committee.
19. 19 FebSmythe’s last letter in the series.
20. late FebRabe leaves Nanjing.
21. Mar–JunSmythe’s War Damage survey.
22. May–JunLast camps close.

Judging the Zone

How many lives did the zone save? The question cannot be answered precisely, and the claim sometimes made, that the committee “saved 200,000 to 250,000 lives,” equates the number of people who sheltered in the zone with the number who would otherwise have died. That is not a claim any historian can test. What can reasonably be said is that the danger inside the zone, though terrible, appears from every contemporary account to have been lower than in the parts of the city and countryside where no foreigners were present; that the camps kept tens of thousands of people fed through the winter; and that the presence of witnesses restrained at least some soldiers on at least some nights.

The committee’s figures have also been misused. Some Japanese writers who deny the massacre have pointed to its population estimates, which rose from about 200,000 in December to about 250,000 in January, to argue that no mass killing can have taken place. Historians have answered that these were rough estimates made by people with no means of counting, for the zone rather than the whole city, at different dates; that refugees were returning from the countryside in January; and that such figures say nothing about the soldiers and men taken from the zone, or about the killings outside the walls. The committee itself never presented them as a census.

The zone’s second legacy is the record. Without the committee’s letters, Smythe’s survey, the hospital records, Magee’s film and the diaries of Rabe, Vautrin, McCallum and the others, the massacre would rest far more heavily on postwar testimony and on Japanese military records, many of which were destroyed in August 1945. Their documents were used at the Tokyo tribunal, they were inscribed on UNESCO’s Memory of the World register in 2015 as part of the “Documents of Nanjing Massacre,” and they are the foundation on which every serious estimate in this collection rests.