Did the Great Wall of China Work? It Was Never One Wall

On 22 May 1644 a Ming general named Wu Sangui opened the gates of the Great Wall at Shanhai Pass and let the Manchu army through from the north. A peasant army under Li Zicheng had taken Beijing on 25 April, and Wu had gone over to the Manchus against him. The Manchus crushed the rebels at the pass on 27 May, took Beijing on 5 June, and kept it.

The most famous failure of the Great Wall was not a breach. Nobody scaled it or knocked a hole in it. The man holding the keys turned them.

Whether that counts as the wall failing depends on what it was for, and on which wall is meant, just as the date Rome fell depends on which Rome you mean. There were several, raised centuries apart, and they do not add up to the barrier the name suggests.

Did the Great Wall of China work?

Partly, and not in the way the question implies. The walls slowed and channelled raids rather than sealing the frontier, and doubled as a customs and signalling system that forced traffic through a handful of guarded passes. Mongol forces crossed gaps in 1550, and in 1644 a Ming general opened the gates of Shanhai Pass himself.

The size of the disagreement shows up in the length. National Geographic Education gives the commonly cited figure as “somewhere between 4,000 and 5,500 kilometers (2,500 and 3,400 miles)”. China’s National Cultural Heritage Administration concluded in 2012 that all surviving Great Wall walls and trenches together span 21,196.18 km. Neither figure is wrong; they are counting different objects.

Three separate walls, fifteen centuries apart

Wall-building starts long before there is one empire to defend. Wikipedia’s construction history of the wall puts the technique in place by the Spring and Autumn period, between the 8th and 5th centuries BC, with Zheng, Chu, Qin, Wei, Zhao, Qi, Lu, Han, Yan, Zhongshan and Zhou all fortifying their own borders.

When Qin unified China in 221 BC, the First Emperor’s opening move was demolition. He ordered the destruction of the wall sections that divided his empire among the former states, then new walls to connect the remaining fortifications along the northern frontier.

Those Qin walls were rammed earth, raised by forced labour, and by 212 BC they ran from Gansu to the coast of southern Manchuria. Later dynasties rarely followed the routes Qin had used.

The wall in the photographs is roughly fifteen centuries younger. Ming builders working between 1368 and 1644 used brick, tile, lime and stone, and up to 25,000 watchtowers are estimated to have gone up on the wall.

What “the Great Wall” can mean	When	Built of
Walls of the rival states	8th to 5th centuries BC	rammed earth
Qin’s connected frontier line	from about 220 BC	rammed earth, forced labour
The Ming wall tourists visit	1368 to 1644	brick, tile, lime, stone

A quarter of the Ming wall is hills and rivers

The Ming wall has been surveyed on its own, and the result is 8,850 km (5,500 miles) made up of 6,259 km of actual wall, 359 km of trenches, and 2,232 km of natural defensive barriers such as hills and rivers.

So about a quarter of the official length is not masonry at all. Those 2,232 km are terrain, counted as defence because that is what the terrain was doing: where a gorge or a ridge held the line for nothing, nobody built.

The wall’s steady job was counting people

Defence was real, and it was intermittent. The administrative work happened every day. The wall’s listed purposes include border controls allowing control of immigration and emigration, the imposition of duties on goods transported along the Silk Road, and the regulation of trade.

A fortified line with a small number of gates does not stop movement so much as decide where movement happens. Merchants, envoys and migrants had to present themselves at a named pass, Jiayu in the west or Shanhai at the sea, where they could be counted, taxed, questioned or sent back.

Signal towers went up on hilltops for visibility, and warnings travelled the line as smoke or fire. A public structure can carry a folk explanation and an operating reason that barely overlap, which is also what happened to the twice-yearly clock change.

Raiders got through, and a trade deal ended the raids

On 1 September 1449 a Ming field army, often estimated at half a million men though the account of the Tumu Crisis says the real number may have been far smaller, was broken at the Tumu post station in Huailai County. The emperor was captured two days later. The Great Wall had not yet been built there, and the border was guarded only by patrols between fortified cities.

The masonry at that stretch came afterwards. Designs often take their final shape from one specific disaster, which is why passenger jets ended up with rounded window corners.

It did not settle the frontier. In 1550 Altan Khan crossed gaps in the Great Wall, besieged Beijing and set its suburbs on fire, and he had already raided northern China in 1529, 1530 and 1542 for livestock and produce that never included the finer goods he actually wanted, according to the record of his campaigns.

What he wanted was a market. He had been trying to trade horses for Chinese agricultural products and textiles, and his approaches were repeatedly rebuffed, his envoys sometimes killed. In 1571 the Ming gave him the title Shunyi prince and the trading rights he had spent decades asking for.

Altan Khan wanted to trade horses for Chinese cloth and grain. For four decades the Ming answered with masonry, and got raids, including the one that burned the suburbs of Beijing in 1550. In 1571 they answered with a market, and he took the title and the trading rights instead.

The myth is the continuity

Arthur Waldron’s The Great Wall of China: From History to Myth came from Cambridge University Press in 1990. The Internet Archive’s record calls it a scholarly study drawing on primary sources and the latest archaeology.

Publishers Weekly summarised the argument in 1990: China’s modern rulers have nurtured the popular myth that the Great Wall is a single, continuous barrier built in the third century BC, when most of what is called the Great Wall today was built during the Ming dynasty of 1368 to 1644. The same review says Ming rulers, fearful of invasion by Mongols and other nomads, “chose wall-building over trade or diplomatic relations”.

That makes the continuous-since-Qin wall a backwards-assembled idea rather than an ancient fact: separate campaigns, centuries and dynasties apart, gathered under one name and one founding date long after the last brick. Japan’s official founding date of 660 BC was fixed in a similar way, by a chronicle finished in AD 720, nearly 1,400 years after the accession it records. Beliefs about rules survive the same way, without anyone having issued them, which is roughly the position of the supposed ban on starting a sentence with a conjunction.

Not visible from space, and the claim is older than spaceflight

The wall’s other durable claim predates any way of testing it. Scientific American quotes a 1932 Ripley’s Believe It or Not cartoon calling the wall “the mightiest work of man, the only one that would be visible to the human eye from the moon”, nearly three decades before anyone reached orbit.

NASA’s caption on an International Space Station photograph of the wall says that “despite myths to the contrary, the wall isn’t visible from the moon, and is difficult or impossible to see from Earth orbit without the high-powered lenses used for this photo”. Jeffrey Hoffman, a former NASA astronaut, told Scientific American that he had spent a lot of time looking at the Earth from space and never saw the wall.

National Geographic Education agrees, and gives the reason: the wall looks a lot like the stone and soil that surround it. When the Chinese astronaut Yang Liwei returned from the Shenzhou 5 mission in 2003 and said he had not seen it, China’s Ministry of Education revised its elementary textbooks.

So the honest verdict is a frontier management system that worked best at management: it slowed cavalry, counted and taxed whoever came through a named gate, and relayed warnings by smoke. Sealing the border was never available, and the surveyors’ own arithmetic admits it.

On 22 May 1644 the masonry did exactly what it had been built to do, and it made no difference, because the man holding the gate had decided to open it. Strictly FYI.