← explore connections

One story leads to another

Read history

Discover the events behind the dates. Start with the landmarks, then follow the stories that catch your attention.

771 stories · Page 54 of 65 · Research and review status

AI-generated illustrations · Each story shows its review status.

August 1, 1927

Nanchang uprising

A leaked plan moved the attack forward by two hours. More than 20,000 troops rose before dawn; after more than four hours of fighting, the insurgents controlled the provincial capital by dawn.

AI-generated illustration for Nanchang uprising View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

2 a.m., 1 August 1927

More than 20,000 National Revolutionary Army troops under Communist leadership or influence rose in Nanchang, Jiangxi, at about 2 a.m. on 1 August 1927. An official Chinese military history says the start was advanced from 4 a.m. after an officer disclosed the plan.

A Chinese Communist Party Front Committee headed by Zhou Enlai directed the operation. He Long, Ye Ting, Zhu De and Liu Bocheng were among the commanders.

Taken in more than four hours, given up on 3 August

After more than four hours of fighting, the insurgents controlled Nanchang by dawn.

They began withdrawing south on 3 August, and government forces recovered the city within days. One scholarly account puts about 21,000 leaving Nanchang and roughly 1,000 remaining by the time the column reached Guangdong; the southward expedition was largely defeated by late September and early October.

The army that did not yet carry the name

This was a mutiny by existing National Revolutionary Army formations under Communist leadership or influence, not a new army assembled that morning. The force was not called the Chinese People's Liberation Army in 1927.

That name came with the manifesto of October 1947, and a November 1948 order standardized it in unit designations. The CCP formally designated 1 August as the anniversary of its armed forces in 1933.

Researched 23 Aug 2026 6 sources 2 audit passes, 1 issue open

How this was checked

Researched from the web into a fact sheet, rewritten from that sheet, then audited against it by a different model. How the pipeline works →

What the sources leave uncertain

  • August 1 is securely established as the uprising's date and is officially commemorated as the birth date of the CCP-led people's army.
  • Calling the force the “People's Liberation Army” on August 1, 1927 is retrospective. The Communist forces then remained units of the National Revolutionary Army; “Chinese People's Liberation Army” appeared much later, with the October 1947 manifesto marking an army-wide renaming and formal uniform unit designations following in 1948.

Checked against

March 12, 1930

Salt March

The first day's destination was not the sea. It was Aslali, thirteen miles off, where Gandhi turned the march into a village-by-village organizing campaign.

AI-generated illustration for Salt March View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

6:30 a.m., 12 March 1930

Mohandas Gandhi left Sabarmati Ashram near Ahmedabad at 6:30 a.m. on 12 March 1930 with 78 selected satyagrahis, walking toward the coastal village of Dandi. Counting Gandhi himself, the original party numbered 79.

On 2 March he had warned Viceroy Irwin that he would begin civil disobedience unless the government addressed eleven demands, among them abolition of the salt tax and monopoly. No concession came.

Thirteen miles to Aslali

The core group covered 13 miles that first day. At Aslali, Gandhi addressed a crowd and urged people to take part through disciplined sacrifice rather than passive sympathy.

The Statesman, as cited by IGNOU, reported 100,000 people lining the route on the first day.

The distance, and the law still intact

The full planned route ran approximately 241 miles; the familiar figure of 200 is materially too low. The marchers reached Dandi on 5 April.

No salt law had been broken on 12 March. Gandhi picked up salt at the coast the next morning, 6 April.

Researched 23 Aug 2026 6 sources 2 audit passes

How this was checked

Researched from the web into a fact sheet, rewritten from that sheet, then audited against it by a different model. How the pipeline works →

What the sources leave uncertain

  • March 12, 1930, the 6:30 a.m. departure and the initial group of 78 companions are strongly documented.
  • Sources round the full route to 240 or 241 miles and convert it variously to about 385–390 kilometres. “200 miles” is materially too low.
  • The walking party reached Dandi on April 5. Gandhi broke the salt law the next morning, April 6, so sources differ over whether the “march” ended on arrival or with the culminating act.
  • Gandhi heritage sources conflict on whether he picked up the salt at 6:30 or 8:30 a.m. on April 6; that timing does not affect the March 12 event.

Checked against

May 11, 1931

Creditanstalt

Officials announced the rescue in the same breath as the disaster. It still produced a run that took more than 30 percent of the bank's liabilities in two days.

AI-generated illustration for Creditanstalt View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

11 May 1931: the number goes public

Austria disclosed publicly on 11 May 1931 that Vienna's Creditanstalt, the country's largest bank, had uncovered a 140-million-schilling deficit while closing its 1930 accounts. That was about 7.5 percent of the balance sheet and roughly 85 percent of the bank's equity.

Creditanstalt had told the Austrian government and the Austrian National Bank privately on 8 May. A weekend of secret emergency meetings followed before anything was said publicly.

Not a closure

Authorities announced a financial-reorganization program at the same time as the loss, so the bank did not simply close its doors or enter ordinary liquidation.

The government took the largest share of the recapitalization, alongside the central bank and the S. M. von Rothschild banking house, a major shareholder.

A run it survived on borrowed cash

Within two days the bank lost more than 30 percent of its total liabilities to withdrawals. It kept paying because the central bank supplied the liquidity.

Creditanstalt held about 1.2 billion schillings in assets against a federal budget of roughly 1.8 billion. Confidence weakened at other Austrian institutions as well, beginning a flight of capital and a drain on the central bank's foreign-exchange reserves.

Researched 23 Aug 2026 6 sources 2 audit passes, 1 issue open

How this was checked

Researched from the web into a fact sheet, rewritten from that sheet, then audited against it by a different model. How the pipeline works →

What the sources leave uncertain

  • Contemporary and scholarly sources commonly call this a failure or collapse, but “went bankrupt” is misleading if understood as a May 11 closure or liquidation: a rescue plan was announced simultaneously, the bank stayed open, and withdrawals were honored with central-bank assistance.
  • The later international crisis had multiple causes and transmission channels. Creditanstalt was an important trigger, but saying this single announcement directly “caused a worldwide financial meltdown” overstates what can be established for May 11 itself.

Checked against

April 7, 1933

Law for the Restoration of the Professional Civil Service

A statute titled as a bureaucratic “restoration” authorized dismissals even where existing law supplied no grounds at all.

AI-generated illustration for Law for the Restoration of the Professional Civil Service View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

7 April 1933, in Berlin

The Reich government enacted and promulgated the Law for the Restoration of the Professional Civil Service on 7 April 1933. Hitler, Interior Minister Wilhelm Frick and Finance Minister Lutz Graf Schwerin von Krosigk signed it.

It reached the Reich, state, municipal, public-corporation and equivalent civil services, and authorized dismissals even when existing law supplied no grounds.

Articles 3 and 4

Article 3 ordered the retirement of civil servants classed as “non-Aryan.” Article 4 authorized dismissal of officials whose political conduct supposedly did not guarantee unconditional support for the Nazi state.

The original text carried exemptions that Hindenburg had insisted on: officials appointed by 1 August 1914, First World War front-line veterans, and people whose father or son had died in the war.

How it was actually applied

The statute did not dismiss every Jewish official at a stroke. An implementing regulation of 11 April defined “non-Aryan” ancestry through Jewish parents or grandparents and required affected officials to document their ancestry or an exemption.

In Prussia's justice administration, 420 of 6,560 officials were removed under Articles 3, 4 and 6 — 128 classified as Jewish, 100 as politically unreliable, and 192 for administrative simplification.

Researched 23 Aug 2026 5 sources 2 audit passes

How this was checked

May 5, 1933

Karl Guthe Jansky

A front-page story grew out of a telephone company's hunt for annoying static — under a headline reassuring readers there was “No Evidence of Interstellar Signalling.”

AI-generated illustration for Karl Guthe Jansky View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

5 May 1933, above the fold

The New York Times of 5 May 1933 put Karl Jansky's work on its front page above the fold, under the headline “New Radio Waves Traced to the Centre of the Milky Way.” The same headline carried an assurance that there was “No Evidence of Interstellar Signalling.”

The report publicized a Bell Telephone Laboratories announcement about mysterious radio static apparently arriving from the direction of the Milky Way's centre. What the paper did was turn an already presented scientific finding into prominent national news.

Not the announcement — the amplification

The science was not new that morning. Jansky had already presented “Electrical Disturbances Apparently of Extraterrestrial Origin” in April to a meeting of the US National Committee of the International Scientific Radio Union in Washington, and Bell Labs issued publicity after it.

Follow-up coverage came within days. A scholarly history reports that WJZ and the NBC Blue Network carried a live broadcast of the Holmdel “star noise” on 6 May, the evening after the front page ran.

What he had actually measured

Jansky's work at Bell Labs had begun as an investigation of noise limiting long-distance shortwave telephone service. Observing at 20.5 MHz — about 14.6 metres — on a rotating directional antenna array, he found that the signal's direction kept sidereal time, repeating every 23 hours 56.06 minutes, which separated it from a solar source.

The date marks the newspaper report, not the first detection; the decisive observations were collected during 1932. Jansky's paper gave an approximate direction — right ascension 18 hours, declination about −10 degrees, with substantial uncertainty in declination — and said association with a particular region was not yet definite.

Researched 23 Aug 2026 6 sources 2 audit passes

How this was checked

June 30, 1934

Night of the Long Knives

Despite the singular name it ran three days — and it opened with Hitler walking into a Bavarian resort hotel at dawn to arrest the sleeping SA leadership.

AI-generated illustration for Night of the Long Knives View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

Dawn, 30 June 1934, at Bad Wiessee

Early on 30 June 1934, Hitler and an SS-police party surprised Ernst Röhm and the assembled senior SA leaders at the Hanselbauer hotel in Bad Wiessee, arrested them and transported many to Munich's Stadelheim prison.

Six senior SA officers were shot at Stadelheim that day. Killing orders were carried out in Berlin, Dachau and elsewhere by the SS, the Gestapo and Göring's police apparatus.

Not only the SA

The victims reached well past the stormtroopers: former chancellor Kurt von Schleicher and his wife, the former Nazi rival Gregor Strasser, the Catholic leader Erich Klausener and Gustav Ritter von Kahr.

Röhm survived the first day. Hitler decided his fate on 1 July, and the Dachau commandant Theodor Eicke and Michael Lippert shot him in his Stadelheim cell after he refused to kill himself.

A law written after the fact

Hitler ordered the operation stopped in the early hours of 2 July. The total killed remains uncertain because killings were concealed and surviving lists are incomplete; the German Historical Museum estimates about 200.

On 3 July the cabinet enacted a one-article law retroactively declaring the measures of 30 June and 1–2 July lawful “state self-defense.”

Researched 23 Aug 2026 6 sources 2 audit passes

How this was checked

Researched from the web into a fact sheet, rewritten from that sheet, then audited against it by a different model. How the pipeline works →

What the sources leave uncertain

  • The dates and main sequence are well established, but the total number murdered remains uncertain because killings were concealed and surviving lists are incomplete.
  • “At least 85” is a commonly repeated documented minimum; the German Historical Museum estimates about 200, and some scholarship gives roughly 150–200. A precise final total is not supportable.

Checked against

August 14, 1935

Social Security Act

Roosevelt signed with five pens, and Frances Perkins was the only woman in the room. Nobody got a retirement check that day — under the original schedule, nobody would until 1942.

AI-generated illustration for Social Security Act View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

14 August 1935, about half past three

President Franklin D. Roosevelt signed H.R. 7260 at the White House at approximately 3:30 p.m. on 14 August 1935, making it Public Law 74-271. Congress had approved the final conference report and sent him the enrolled bill five days earlier.

He used five pens. Frances Perkins, the Secretary of Labor who had chaired the Committee on Economic Security, attended; the Social Security Administration identifies her as the ceremony's only woman.

Eleven titles

The act created federal old-age benefits, a federal-state unemployment-compensation framework, grants for old-age assistance and for dependent children, public-health and welfare grants, and a Social Security Board.

In his signing statement Roosevelt described the law as an incomplete “cornerstone” and said it would give at least some protection to 30 million citizens.

Nothing paid out, nothing collected

The old-age insurance taxes and benefits did not begin at the signing. Payroll taxes were scheduled for 1937 — 1 percent of covered wages from employees and 1 percent from employers, on up to $3,000 annually — and monthly old-age benefits for 1942.

About half the jobs in the economy fell outside contributory coverage, agricultural and domestic workers among the excluded groups.

Researched 23 Aug 2026 8 sources 2 audit passes

How this was checked

Researched from the web into a fact sheet, rewritten from that sheet, then audited against it by a different model. How the pipeline works →

What the sources leave uncertain

  • August 14 is confirmed by the enrolled act, SSA chronology, signing statement, and contemporary imagery.
  • A National Archives narrative sentence incorrectly says August 15, but its document citation and transcript identify August 14; the independent SSA records resolve the discrepancy.

Checked against

November 23, 1935

Sulfanilamide

By a later account, the decisive test happened because four infected mice were left over and Bovet suggested trying the plain colourless fragment of the dyes.

AI-generated illustration for Sulfanilamide View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

A two-page report, 23 November 1935

A two-page communication associated with the Société de Biologie session of 23 November 1935 reported the activity of p-aminophenylsulfonamide — sulfanilamide, laboratory code 1162F — against experimental streptococcal infections in mice and rabbits.

The authors were Jacques Tréfouël, Thérèse Tréfouël, Federico Nitti and Daniel Bovet, working in Ernest Fourneau's medicinal-chemistry laboratory at the Institut Pasteur in Paris. Older citations nearly erase Thérèse Tréfouël by listing her only as "Mme J. Tréfouël."

They reported a tolerated dose of 1,000 mg/kg without observed harmful effects.

The colour was never the point

They found that the simple, colourless sulfanilamide molecule protected infected animals with activity comparable to the more complex azo dye Prontosil.

Their structural argument suggested that Prontosil-like dyes were split in the body to release sulfanilamide. That antibacterial activity did not depend on the colored dye structure opened the way to systematic development of sulfonamide drugs — and undercut the commercial advantage of Prontosil's patented structure.

Direct detection of sulfanilamide in blood and urine after Prontosil administration came later, from A. T. Fuller in 1937.

A report date, not a eureka date

The 23 November date is well supported as the date of the Société de Biologie session or issue associated with the communication, which ran to two pages at 756–758 of volume 120.

It is not securely documented as the calendar day on which the decisive laboratory experiment occurred, so it should be read as the report date.

The finding was also not the first synthesis of sulfanilamide: Paul Gelmo had made the compound decades earlier without recognising any antibacterial use.

Researched 23 Aug 2026 7 sources 2 audit passes Traditional date

How this was checked

Researched from the web into a fact sheet, rewritten from that sheet, then audited against it by a different model. How the pipeline works →

Traditional date. This date is the conventional one, not one fixed by a contemporary record.

What the sources leave uncertain

  • November 23 is well supported as the date of the Société de Biologie session or issue associated with the two-page communication.
  • It is not securely documented as the calendar day on which the decisive laboratory experiment occurred. The event should therefore be presented as the report or communication date, not a precisely dated eureka moment.
  • The 1935 work inferred that Prontosil was cleaved to an active sulfanilamide component; Fuller directly demonstrated sulfanilamide in the blood and urine after Prontosil administration in 1937.

Checked against

1935

Daventry Experiment

The transmitter was a BBC station already broadcasting from Daventry. The receiver had been built to study the ionosphere, then carried into a field to look for a bomber.

AI-generated illustration for Daventry Experiment View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

26 February 1935: a bomber flies through a borrowed beam

On 26 February 1935, Robert Watson-Watt and Arnold Wilkins staged a feasibility demonstration in a field near Weedon, Northamptonshire, using transmissions from the BBC Empire short-wave station at Daventry — 10 kilowatts at a wavelength of 49 metres.

An RAF Handley Page Heyford bomber flew repeatedly through the beam at a reported 5,000 feet while the ground team watched a cathode-ray display for fluctuations caused by radio energy reflected from the aircraft.

The display responded while the bomber was approximately eight miles away.

Repurposed apparatus, not a radar set

The receiver had originally been used by the National Physical Laboratory for ionospheric measurements. The demonstration repurposed existing apparatus and a broadcasting transmitter rather than employing a completed operational radar set.

It revealed the aircraft's presence through interference between the direct and reflected continuous-wave signals. It did not yet measure a target's range and bearing the way the later pulsed system would.

The acronym "radar" came later; the British work was then described as radio detection, or RDF.

What the Air Ministry took away

The Air Ministry had rejected a radio "death ray" as impractical. Watson-Watt then submitted a 12 February memorandum proposing aircraft detection by reflected radio waves, and officials required a convincing demonstration before committing to development.

A. P. Rowe attended for the Committee for the Scientific Survey of Air Defence and was convinced that an aircraft could be detected by reflected radio signals. Secret government development work followed, moving to Orfordness and later Bawdsey, where pulsed aircraft-detection systems were produced.

Scholarly histories describe radar as a case of simultaneous invention across several countries, and distinguish Daventry as a decisive British feasibility demonstration rather than the invention of radar.

Researched 23 Aug 2026 8 sources 2 audit passes Date corrected

How this was checked

April 12, 1937

Power Jets WU

Whittle closed the fuel valve and the engine kept accelerating, shrieking, its combustion casing glowing red.

AI-generated illustration for Power Jets WU View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

12 April 1937, at the Rugby works

Frank Whittle and Power Jets began running the complete experimental WU turbojet at the British Thomson-Houston works in Rugby on 12 April 1937. Whittle was 29.

The run was alarming rather than smooth. The engine accelerated beyond his control, the combustion-chamber casing showed red-hot patches, and closing the fuel valve did not immediately stop the rise.

The uncontrolled acceleration stopped at about 8,000 rpm and then subsided.

Fuel in the bottom of the chamber

Investigators determined that fuel left pooled in the combustion chamber by earlier pump tests had ignited, causing the runaway.

Financial constraints had forced Whittle to test a complete engine without a full programme of separate component trials.

Another attempt the following evening reportedly produced the same behaviour; modifications were needed before controlled running was achieved, and development continued through the summer.

The disputed first — and an engine that never flew

12 April is the best-supported date, given in Whittle's own James Clayton Lecture and in the Rugby sources, though a National Academy of Engineering biography prints 12 March 1937 without explaining the discrepancy. The place is not in question. The larger dispute is the claim that the run was unquestionably the world's first operating turbojet.

The Smithsonian and the US National Academies place Hans von Ohain's hydrogen-fuelled HeS 1 run in early 1937 or March 1937, potentially before Whittle, while a Cambridge chronology based on Ernst Heinkel's memoir dates that German run to September. Sources also distinguish an experimental hydrogen demonstrator from a liquid-fuelled practical turbojet.

The WU was an experimental bench engine and never powered an aircraft; the later Power Jets W.1 powered the Gloster E.28/39 on Britain's first jet-aircraft flight in 1941.

Researched 23 Aug 2026 7 sources 3 audit passes, 1 issue open Date disputed

How this was checked

Researched from the web into a fact sheet, rewritten from that sheet, then audited against it by a different model. How the pipeline works →

Date disputed. Sources disagree about when this happened.

What the sources leave uncertain

  • The date and place of Whittle's WU run are strongly attested. The disputed element is the claim that it was unquestionably the world's first operating turbojet.
  • The Smithsonian and US National Academies place Hans von Ohain's hydrogen-fueled HeS 1 run in early 1937 or March 1937, potentially before Whittle. A Cambridge chronology based on Ernst Heinkel's memoir dates that German run to September. Different sources also distinguish between an experimental hydrogen demonstrator and a liquid-fueled practical turbojet.
  • The WU was an experimental bench engine, not a flight-qualified engine intended to be installed in an aircraft.

Checked against

August 5, 1937

NKVD Order No. 00447

A person's fate could already be encoded beside their name on an NKVD list: first category meant a bullet, second category meant a camp.

AI-generated illustration for NKVD Order No. 00447 View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

5 August 1937: arrests from lists prepared in advance

On 5 August 1937 the NKVD began implementing secret Operational Order No. 00447 across most Soviet republics, territories and regions.

Regional NKVD offices made arrests from precompiled lists. A person's assigned category largely predetermined the outcome: first category meant execution by shooting, second category meant imprisonment in camps or prisons.

The start was geographically staggered — the order set 10 August for four Central Asian republics and 15 August for the Far Eastern and Krasnoyarsk territories and the East Siberian region.

Who the order named, and who decided

The operation targeted broadly defined groups: former kulaks, alleged criminals, clergy, former White officers, former officials, non-Bolshevik political activists, returned emigrants and other people labelled anti-Soviet or socially dangerous.

Sentences came from regional three-person troikas, usually chaired by the NKVD regional chief and joined by party secretaries and prosecutors.

The proceedings were accelerated and extrajudicial: the accused was absent, and so was any defence lawyer.

The quotas, and how far past them it ran

Stalin had ordered regional leaders on 2 July to estimate how many former kulaks and alleged criminals should be executed or sent to camps. Nikolai Yezhov converted that into Order No. 00447 on 30 July, the Politburo approved it on 31 July, and the order prescribed the 5 August start.

Regional party and NKVD leaders soon requested larger quotas. An operation originally intended to last four months continued until November 1938; the initial regional quotas had covered 269,100 people, of whom 76,000 were assigned to execution.

Surviving NKVD statistics produce approximately 767,000 sentences and 387,000 executions under Operation 00447; scholars who adjust for incomplete documentation estimate roughly 820,000 condemned, including 437,000–445,000 executed.

Researched 23 Aug 2026 4 sources 2 audit passes

How this was checked

Researched from the web into a fact sheet, rewritten from that sheet, then audited against it by a different model. How the pipeline works →

What the sources leave uncertain

  • August 5 is the explicit start date for most of the USSR, not every region. Four Central Asian republics began August 10, while designated eastern regions began August 15.
  • Surviving NKVD statistics produce approximately 767,000 sentences and 387,000 executions. Scholars who adjust for incomplete documentation estimate roughly 820,000 sentences and 437,000–445,000 executions.

Checked against

December 1, 1937

Battle of Nanking

Tokyo's order to attack Nanjing arrived after Japanese soldiers were already racing toward it.

AI-generated illustration for Battle of Nanking View full image ↗
AI-generated illustration · Ideogram 4 (fp8) · 2026-08-27
How was this image made?

The creative brief sent to the image model. These are instructions, not a record of what actually happened.

Original prompt & settings (JSON) ↗

1 December 1937: Orders No. 7 and No. 8

On 1 December 1937 the orders formalising a full-scale assault on Nanjing, the Chinese capital, were issued. How much was decided that day is itself contested: a second chronology places the approval several days earlier, in late November, and treats 1 December as the day Hayao Tada, the Army General Staff's deputy chief of staff, delivered the already-approved order in Shanghai.

Army General Staff Order No. 7 converted the provisional Central China Area Army into a formal battle formation; Order No. 8 directed that army to assault Nanjing with naval support.

The operational plan had the Tenth Army beginning its next movement on 3 December and the Shanghai Expeditionary Army on 5 December, with units regrouping before a coordinated assault.

The armies were already past it

Frontline forces neither waited for the authorisation nor observed its staged timetable. The Tenth Army had taken Guangde on 30 November, and the Shanghai Expeditionary Army was already beyond its intended line.

They occupied Danyang on 2 December — five days ahead of the formal schedule — reached the city's defences, began the general assault on 10 December and captured Nanjing on 13 December.

On the Chinese side, 1 December was a day of unravelling rather than of orders. Chinese forces began withdrawing on the Jiangyin front; poor communications sent the 112th Division out prematurely and helped turn the 103rd Division's retreat toward Nanjing into a rout.

The historian Akira Fujiwara describes the advance as out of control from the start.

Which day the battle begins

1 December is secure as the date of Orders No. 7 and No. 8. What is disputed is whether an administrative and operational step — and one that may have ratified a decision already taken — should be called the beginning of the Battle of Nanjing.

Some military scholarship dates the actual attack on Nanjing to 8–13 December, while other chronologies use 9–13 December for the final battle at the city, or 10 December for the all-out assault.

The Nanjing Atrocity is sometimes given a broader 1 December starting boundary to cover crimes committed during the approach, but the mass violence inside the captured city followed its fall on 13 December.

Researched 23 Aug 2026 4 sources 2 audit passes, 1 issue open Date disputed

How this was checked

Researched from the web into a fact sheet, rewritten from that sheet, then audited against it by a different model. How the pipeline works →

Date disputed. Sources disagree about when this happened.

What the sources leave uncertain

  • December 1 is secure as the date of Japanese Army General Staff Orders No. 7 and No. 8. What is disputed is whether that administrative and operational decision should be called the beginning of the Battle of Nanjing.
  • Japanese units were already advancing and fighting toward Nanjing before December 1. Some military scholarship dates the actual attack on Nanjing to December 8–13, while other chronologies use December 9–13 for the final battle at the city or December 10 for the all-out assault.
  • The Nanjing Atrocity is sometimes given a broader December 1 starting boundary to include crimes committed during the approach, but the mass violence inside the captured city followed its fall on December 13.

Checked against