MAPS OF HISTORY · China in Revolution · ALL CHAPTERS · CHAPTER 10
CHAPTER 10 · 1945–1947 · 1946
The Failed Peace and Manchuria
The map above is the world of 1945 — Surrender — the liberated areas.
BRIEF EXPLANATION — THE ESSENTIAL ARGUMENT
Japan’s surrender set off a race for Manchuria, because whoever accepted the Japanese surrender there would hold the territory afterward. Soviet forces overran Manchukuo, stripped its industrial equipment, and quietly allowed captured Japanese weapons to reach the Communists, while the United States moved Nationalist armies north to take the formal surrenders. Both Chinese parties understood that the northeast’s railways, ports and industry could decide the entire civil war.
Negotiations at Chungking and American mediation produced signed agreements but no trust between the parties. Nationalist troops won the cities and rail lines through 1946, yet Communist forces rebuilt across the countryside around them. Even the capture of Yan’an in 1947 carried little military weight, so that the government’s apparent high point actually concealed overextended supply lines and fast-eroding legitimacy.
GUIDED EXPLANATION
The postwar race was fundamentally about position, because accepting the Japanese surrender also meant occupying the territory a rival might otherwise take. Japan’s surrender opened the largest land-grab in Chinese history, and the map draws it as exactly that race. The two tan arrows from the north are August Storm (the ● at Harbin), for the Soviet Union entered the war on 9 August 1945 and overran Manchukuo in eleven days. It then stripped the region’s factories, a reparations commission later pricing the removals near nine hundred million dollars, and with studied ambiguity let captured Japanese arsenals leak to the Communist columns arriving on foot and by junk from Shandong (the red arrow across the gulf). The blue arrow up the coast is the other racer, as America sealifted and airlifted half a million Nationalist troops north to take the surrenders of a million Japanese, the greatest troop movement Washington ever performed for an ally and the clearest statement of whose China it preferred. In between, the hatching records what both parties already knew, that Manchuria, Japan’s industrial estate and the one region where a Chinese war could be won with factories, was contested from the hour it was created. Its countryside was turning red before its cities had finished changing flags.
The peace that everyone professed was negotiated twice and believed by no one. At Chungking (the ● there), Mao and Chiang toasted the republic and signed a communiqué of unity in October 1945, and General Marshall arrived to mediate and even conjured a ceasefire and a joint army on paper. The paper burned in Manchuria. Chiang, against American advice, committed his best armies to taking every Manchurian city, and at Siping (the ✕) in spring 1946 Lin Biao made the mistake of meeting them conventionally, lost 40,000 men, and retreated north across the Sungari. It looked decisive at the time. The map tells a different story, because the Nationalist tide filled the cities and rail lines of a region whose villages, distances and hinterland increasingly belonged to an enemy who recruited where he retreated, the same points-and-lines geometry that had swallowed Japan and was now inherited by the government. In March 1947 the government even took Yan’an (the ● in the loess hills), so that the red zone the map had carried since 1935 vanished, though Nationalist banners photographed well over an evacuated cave town of no military value. This was the high-water mark, and every plate of this atlas after it runs the other way.
DEEP EXPLANATION — EVIDENCE & INTERPRETATION
How much did Stalin arm Mao?
A durable claim in Nationalist and Cold War accounts is that the Soviet Union, having overrun Manchuria in August 1945, simply handed the region’s captured Japanese arsenals to the Communists and thereby manufactured their victory. The archival picture assembled by historians such as Odd Arne Westad and Steven Levine is more equivocal and more interesting. Real transfer did happen, as rifles, artillery and stocks from the Kwantung Army’s depots reached Communist columns, sometimes directly and more often through Soviet forces that conveniently “failed” to guard what they withdrew from. Yet Moscow also stripped the factories for its own reparations, signed a friendship treaty with Chiang’s government, and hedged against a Communist victory it did not expect. The Communist army in the northeast grew at least as much from local recruitment, absorbed puppet troops and land-reform mobilization as from Soviet largesse. The measured verdict is that Soviet policy tilted the board decisively without playing the pieces, an external intervention that shaped the outcome while leaving the winning to be earned.
“Who lost China?”: the question and its costs
The recrimination began before the war ended and outlived everyone who started it, and it repays dismantling as a case study in how foreign-policy myths are manufactured. The phrase “who lost China” smuggles in a premise, that China was America’s to keep or lose, which almost no serious historian now accepts, since a few hundred million dollars of aid and a Marshall mission were never going to reorganize a continental society against its own dynamics. Yet the myth carried enormous consequences. In the McCarthy years the State Department’s China specialists, John Service, John Paton Davies and John Carter Vincent, were driven out for having reported, accurately, that the Nationalists were losing, so that the government destroyed its own expertise at the moment it most needed it. Historians link this hollowing-out to two decades of non-recognition of the People’s Republic and to the misjudgments that helped carry the United States into Vietnam. The transferable test is to ask, of any “who lost X” debate, whether the questioner has assumed a control that never existed.
New here? Chapters 1–2 of every atlas are free to sample, and the WW2 atlas is free in full. One membership opens all thirteen — the Cartographer’s Circle.
THE SITES
ON THIS DAY
WHY IT HAPPENED
Manchuria as the decisive prize. Whoever held the northeast held the only industrial complex in China, ports and railways facing a friendly USSR, and grain surpluses to feed an army. Mao stripped his best cadres and 100,000 troops from other theaters and sent them there in autumn 1945 (“develop the north, defend the south”). Chiang, warned by Wedemeyer and Marshall that his logistics could not carry a Manchurian war, went anyway — legitimacy demanded the recovery of the region whose loss in 1931 had defined him. Both were arguably right about the stakes. Only one could afford the bet.
Soviet policy: two hands, one direction. Stalin signed a friendship treaty with Chiang’s government (August 1945), extracting Yalta’s concessions — Port Arthur, the railways, Mongolia’s independence confirmed. He simultaneously timed his Manchurian withdrawal (spring 1946) so that Lin Biao’s armies, quietly fed Japanese stocks, stood nearest the vacated cities. Moscow wanted a weak, divided China on its border and hedged toward whichever outcome produced one. It did not expect its client to win outright any more than Washington expected its own to lose.
Mediation against both parties’ convictions. Marshall’s mission failed not from want of skill but of material: both parties had spent twenty years learning that armies are the only guarantee (Chapter 5’s purge was the syllabus), both used truces to reposition, and neither would fold its forces into a state the other might capture. The ceasefires of 1946 arguably served Lin Biao’s recovery after Siping — a fact Nationalist historiography has never forgiven — and Chiang’s captures served his conviction that force would settle what talks could not. Coalition government was a third China nobody armed believed in.
Victory’s poisoned inheritance. The government returned to the coastal cities as a conqueror returns, not a liberator: carpetbagging takeover officials, currency conversion that expropriated everyone who had held puppet-issue money, collaborationist industry seized into crony hands, hyperinflation accelerating. In the year of its greatest prestige the state squandered the urban legitimacy it would need for the war it was choosing — while in the villages it recovered, the landlords came back with the flag.
THE TURN
August Storm, 9–20 August 1945. Eleven Soviet days created the civil war’s decisive arena and dealt its opening hands: Japanese arsenals within Communist reach, a stripped industrial base, a border sanctuary, and a region whose recovery Chiang could not decline and could not sustain. Everything at Liaoshen in 1948 — the sealed rail gate, the besieged garrisons at the end of 1,800-kilometer supply lines — descends from the geometry of these two weeks. Great-power interventions often decide civil wars before the civil wars properly begin. This is the era’s cleanest case.
WHAT IT CHANGED
The countryside is reorganized while the cities are besieged. From May 1946 the party converts wartime rent reduction into outright land redistribution across the northern base areas — violent, sweeping, and militarily brilliant: millions of families now hold title deeds that a Nationalist victory would cancel. The armies of 1948 march on that arithmetic. The porters of Huaihai push barrows for their own deeds.
Lin Biao builds an army out of a retreat. North of the Sungari, the defeated force of Siping is rebuilt into the Northeast Field Army: Japanese equipment, absorbed puppet troops, conscription from redistributed villages, winter offensives that bleed the garrisons. By 1948 it outnumbers and outguns the government’s best — the first Communist force in the era’s history to hold every material advantage.
America hedges into irrelevance. Marshall departs in January 1947 blaming both sides. Aid to Nanjing continues at levels large enough to implicate and small enough not to rescue. The “who lost China” recrimination is already being drafted. Its premise, that China was Washington’s to lose, is the first thing a careful student should question.
FIELD QUESTION — The “lost chance” debate: could the 1945–46 mediation have produced a coalition China — or was the civil war already decided?
The lost-chance case notes real material: both parties signed the October 1945 communiqué and the January 1946 ceasefire. War-weariness was universal. And a genuinely enforced standstill would have frozen the Communists out of Manchuria’s cities, arguably capping their ambitions. The skeptics. The archives opened since the 1990s, read by Odd Arne Westad, Chen Jian and others — are more persuasive: Mao’s internal directives treated negotiation as a phase of struggle (“talk, talk, fight, fight”). Chiang’s own diaries show identical convictions with opposite casting. And the structural fact stood that two Leninist party-armies had each concluded from 1927 that disarmament equals death. Marshall could suspend the shooting. He could not insure either party against the other’s victory, and coalition without insurance was surrender on an installment plan. The verdict most evidence supports: not a lost chance but a mirage both sides found briefly useful. The transferable test for any peace process: ask not whether the parties will sign, but what enforces the clause each one intends to break.
AN INTERESTING FACT
The Soviet removals from Manchuria were inventoried almost immediately: the Pauley reparations mission toured the northeast in 1946 and priced the stripped generators, machine tools, locomotives and whole factory lines at roughly 858 million US dollars (about 2 billion including deterioration) — Moscow called the equipment “war booty” taken from Japan, not from China. The commission’s photographs of concrete floors bolted for vanished machines circulated worldwide. Manchuria thus entered the civil war as the prize both armies bled for and an industrial shell — which made the human factors, conscription and grain and porters, weigh all the more in who finally held it.
THE DISPATCH
One short letter when a new atlas opens — and the printable study guide for China in Revolution is yours now, free.
NO TRACKING · YOUR ADDRESS IS USED FOR THE DISPATCH AND NOTHING ELSE · UNSUBSCRIBE ANYTIME