There is a scene from the early twelfth-century Chinese capital of Bianliang, what is now Kaifeng, that the painters of the time recorded with the same care they gave the imperial throne. A long, flat-bottomed grain barge is being towed through the lock of a stone dam, its hull heavy with northern millet. At the head of the towpath, a water buffalo plods forward in the direction of the canalside inn; behind it, a thin man in a conical hat balances a long pole on his shoulder. The scene looks ordinary, almost agrarian, but the heavy barge is on the middle stretch of a waterway that runs for almost two thousand kilometres, joining the rivers of the Yellow River basin to the rivers of the Yangzi basin and bringing the southern grain tribute by water to the northern capital. The waterway is the Grand Canal of China, and it is the longest man-made waterway in the world. It was not built in one go: it was assembled over more than ten centuries by a succession of dynasties, each adding its own piece. Its first substantial phase was the Sui dynasty’s unification of the regional canals of the north and south, around 605-610 CE. Its last substantial phase was the Ming dynasty’s wholesale realignment of the route through Shandong, in the early fifteenth century. The Canal still exists, runs, and is used by barge traffic today, with much of its masonry locks still in place — though a 2014 decision by the Chinese government to keep the locks in working heritage status means that some sections are now classified for conservation rather than for shipping.
China, the country, was already the largest economy in the world in 605 CE, and it had been a unified empire for almost four centuries by then. The Grand Canal is not a Chinese national project in the sense that the Roman roads of central Italy were a Roman national project: the road network of the Roman republic was built and maintained under a single authority and a single project timetable. The Grand Canal was a succession of regional projects over more than a thousand years. To see what a single dynasty actually built, the standard way into the Canal is to read it dynasty by dynasty, beginning with the Sui, then the Tang and Song, then the Yuan and Ming.
The Sui dynasty (581-618) inherited a divided China after nearly three centuries of northern and southern dynasties running parallel. Sui Wendi, the first Sui emperor, completed the political unification of north and south in 589, the same year that Emperor Yang of Sui completed the first Grand Canal section. The first and most consequential Sui canal was not actually a single canal but two short canal cuts between the new capital Daxing (modern Xi’an) and the lower Yellow River, and between the Huai River basin and the lower Yangzi at Yangzhou. The roads of the Sui state had been running for some centuries already: the earliest regional canals, sometimes dated to the Spring and Autumn period but more reliably to the Warring States period, allowed small barges to ferry soldiers and grain between adjacent river systems. The Sui state inherited the system of local canals around Luoyang and the central plain, and the state in effect enlarged the connection between the Yellow River basin and the Huai-Yangzi basin. There are good reasons for thinking that the motivation for the Sui canal works was as much political as economic. The Yangzi basin was where the bulk of the population and the bulk of the tax receipt lay; the Sui need to move grain and conscripts to the northern frontier impelled the central government to invest in a long-distance canal that could carry both commodities at scale.
The Sui canal works are well documented in Chinese sources of the Tang period, which used the Sui canal works as a cautionary example of imperial overreach. The “Grand Canal of the Sui,” as it is often described in English summaries, is one of the standard objects of Chinese historical writing. The conventional explanation is that the works were insupportable in tax-labour terms, that the laborers who built the original canal cuts were killed by the hundreds of thousands by overwork, malnutrition, and accident, and that this contributed to the popular unrest that destroyed the Sui dynasty in 618, after just one emperor more than the dynasty’s founder. The conventional explanation is also wrong in part, because it underestimates how much of the canal work pre-dates the Sui and how much was completed under the Tang. The Tang inherited a complete network of regional canals between the Yellow River and the Yangzi and re-opened, deepened, and re-cut it. The Tang accomplishment in the field of waterway engineering was not the building of the canal but the routine maintenance of a canal that was already there. The Tang historians’ complaint that the Sui killed ten million laborers to build the canal is a moral complaint about the Sui, not a count of workers. Modern historians estimate the Sui mobilisation of labour at something like a fifth of the hyperbolic figure. This is one of the persistent difficulties of reading Chinese sources; the modern estimates are usually large enough to be impressive and small enough not to be absurd.
The period from the fall of the Sui to the rise of the Song is the period during which the canal reached its full length. The Tang (618-907) started with the reconstructed network of regional canals that the Sui had completed. By the late Tang the central state had extended the canal through the southwestern front of the central plains, joining Luoyang to Yangzhou by a single water route of about 200 kilometres through the Huai River basin. The Yangzi was the main artery of the south: grain ports from Suzhou and Yangzhou were connected by a network of subsidiary canals that effectively made the lower Yangzi basin a single open road system for grain tribute. The Tang state kept its grain tribute system in working order with regular maintenance of the canal and the river ports. The Tang did not do as much canal-building as the Sui had done; what they did was keep the inherited system running while adjusting the southern and northern termini in line with changes in the imperial capital.
The shift came in the Song dynasty (960-1279), when the imperial capital moved south to Lin’an (Hangzhou) in 1127 after the loss of the northern plains to the Jurchen Jin. The Song Canal, in this period, was no longer a northward conduit but a self-contained waterway of the lower Yangzi and the Huai basin, with the Yellow River basin no longer under Song control. After the Song lost the north, the direction of tribute flow reversed. Grain tribute still came south, but now it stayed in the south. The Song built a series of inland waterways connecting Hangzhou to the Yangzi main stem and to the Qiantang River, which then linked back to the upper Yangzi by waterway. The Song Canal was a network of feeder canals to the Yangzi, with the Grand Canal of the Sui and Tang forming the northern stem of a now-broken southern and northern water system.
The reunification of China under the Yuan (1279-1368) brought the northern and southern systems of the Canal back into one. The Yuan capital was in Dadu (Beijing) at the apex of the northern plain; the problem of getting grain from the lower Yangzi basin to Dadu was now back to being the central logistical problem of the Chinese state. The Yuan state, under the supervision of the khans and their Chinese administrators, did two large things to the Canal. The first was the extension of the Canal from Beijing down to the new route in Shandong. The previous route of the Sui-Tang Canal ran through Xuzhou and the Huai basin to the lower Yellow River via a series of cuts at the foot of the western Shandong hills. The Yuan route changed this by linking the Canal to the new Hai River system running from Shandong through Dezhou to Beijing. The second was the creation of the Jizhou-Yangzhou “Shandong Canal,” a particularly clever piece of engineering that used the natural routes of the Wen River and the old Yellow River channels to keep the canal level across the high ground of central Shandong. The Yuan realignment is the longest single piece of new canal engineering in the history of the Canal: it cut nearly 200 kilometres off the old Sui-Tang route by routing the canal through Shandong rather than around Shandong. The price was that the Yuan Canal had to be lifted over the watershed between Shandong’s rivers and the Hai basin, and to do that the Yuan engineers built a series of locks and a sustained summit pound around Linqing and Dezhou. The Shandong Canal summit pound of around 30 metres of lift is one of the classic works of medieval Chinese hydraulic engineering and remains a load-bearing piece of the Canal’s modern heritage route.
The Ming dynasty (1368-1644) replaced the Sui-Tang Canal route north of the Huai basin with the route that is largely still the Canal’s main line. The Ming replaced the Shandong Canal’s summit pound with a higher stretch of canal running across the highest part of Shandong, with the new apex around Nanyang Lake in Weishan. The Ming had to re-equip the canal with new lock systems: the Ming “double” lock, in which two lock chambers in series were used to lift a barge up a multi-metre hill in a single operation, was a technical refinement that came to dominate Chinese lock practice for several centuries. The Ming replacement of the Yuan route, called the “Channel of the 32 Locks” after the principal series of lock chambers between the Huai basin and the Yellow River, is the route that the Canal follows today for most of its length. The Ming also built a new city at the southern end of the Canal — Linqing, in eastern Shandong — to serve as the southern terminus of the grain traffic and the southern warehouse of the grain tribute system. Linqing grew in this period to be one of the most prosperous canal cities in north China and a major inland transshipment port for trade with Central Asia.
The Ming Grain Tribute system, which the Canal enabled, was a major piece of state logistics in its own right. The state contracted grain collectors in the lower Yangzi basin and the Huai basin to assemble the tribute grain at a fixed number of granary cities — Suzhou, Songjiang, Hangzhou, and Jiaxing in the south; Yangzhou and Huai’an in the central section — and transport the grain north by canal to a small number of grain stations at the northern end: Tianjin, Tongzhou, and Beijing itself. The grain was the state’s main commodity and its main tax receipt, and the Canal was the spine along which the state’s tax receipt flowed. The Canal thus also carried the return traffic: the household goods of the bureaucracy, the military payroll, the salt that the state sold in the south from its northern salt mines. The Canal is often described as the principal artery of the premodern Chinese economy. That description is correct.
The system slowly fell apart in the late Ming and Qing periods. The grain tribute continued to be moved along the Canal by a combination of barges, conscripted corvée crews, and contracted canal haulers, but the canal itself was increasingly difficult to maintain. The Yellow River, which the canal crosses several times between Shandong and the Huai basin, kept shifting its course in the great flood cycles of the seventeenth and eighteenth centuries, and each shift required a partial re-engineering of the southern Canal terminus. The late Ming dynasty was unable to keep up the maintenance. The Qianlong emperor’s tour of inspection of the Canal in the mid-eighteenth century is a famous moment of late-imperial self-congratulation; the Canal by then had silted up in sections, and a number of the southern locks had to be rebuilt before the imperial barge could be towed through. The Qing state’s response was a series of improvements to the canal rather than the construction of a new canal. The Qing lock engineering is documented in detail in the Kangxi-emperor period Veritable Records and in the Qianlong emperor’s memoirs of his southern tours, both of which were written for the edification of the next emperor and are heavily formalised; but they record the operations of dredging and re-lining that the Qing performed on the Canal.
The early twentieth century saw the slow decline of the Canal as a working waterway. The late Qing government had already begun to investigate the possibility of railways as an alternative to the canal; the early Republican government pursued this option vigorously. The first railway to parallel the Canal in northern Shandong was the Jiaoji Railway, completed in 1905, which connected the Qingdao port to Jinan and then to Tianjin. Within a decade, a continuous road-and-rail route from Beijing to Nanjing existed in parallel to the Canal, and the Canal’s grain traffic collapsed. The Republican government revived the Canal in the 1930s, partly as a patriotic project against the Japanese occupation, and the Canal was used as a logistical support for the National Revolutionary Army in the later 1930s before the loss of the north. After 1949 the People’s Republic rebuilt several sections of the Canal and reopened parts of it for grain traffic, but the northern sections in particular silted up and were abandoned as a working waterway. The PRC state at various points in the 1950s and 1970s attempted major works on the canal system, including the considerable engineering works around the canal lift from the Huai basin to the Yellow River at Jining. The works were partial; the Canal was reopened in patches.
The Canal today runs in a heavily used version through Beijing-Tianjin to Shandong, in a heritage version through Shandong to Yangzhou, and in a heavily used version from Yangzhou through the Huai basin to Yangzhou. The northern end is connected to the Hai River basin and to the port of Tianjin; the southern end is connected to the Yangzi basin and to the lower Yangzi ports. The eastern Shandong section between Linqing and Jining has been protected as a National Cultural Heritage site by the Chinese government, with the southern stretches of the Shandong Canal under separate heritage protection. The current Chinese state, in the 2014 and 2022 decisions, set the policy framework for keeping the Canal as a working waterway with a heritage status, with a separate grand plan for the “Grand Canal Cultural Belt.” This is a piece of national policy that runs against the trend of declining waterway use elsewhere in China. The 2014 decision committed the state to preserve the historic route of the Canal and the system of locks, dams, and waterways that runs across nine provinces and municipalities from Beijing to Zhejiang.
The engineering of the Canal is one of the reasons it has been written about for so long. The Sui-Tang and Ming Canal is among the largest single-pieces-of-canal-building in preindustrial engineering. The southern section is mostly a deepening of natural river channels — the Han River in Jiangsu, the Wei River in northern Anhui — with relatively little artificial cutting. The northern section, in the high ground of Shandong, requires substantial canal cuts, locks, and diversions. The Ming Double Lock at Weishan, on the highest stretch of the Shandong Canal, is a classic example of preindustrial Chinese lock engineering. The double-lock was a system of two lock chambers in series, each with a single gating pair, in which the upstream gate of the downstream chamber and the downstream gate of the upstream chamber shared an intermediate pound; the barges would enter the downstream chamber, the chamber would be brought up to the intermediate level, the barges would exit through the upper gates, the chambers would equalise with the upper pound, and the barges would enter the upstream chamber, which would then raise them to the next pound. The double-lock lifted barges over a hill of eight to ten metres in two stages of about half that lift each. The engineering cost of the lock is small in absolute terms but the throughput is high; the Chinese double-lock is the forebear of the modern double-lock used on European canals and in the United States from the late eighteenth century, although the direct historical transmissions are not always clear.
The cultural and literary history of the Canal is at least as large as the engineering history. The Canal is the principal setting of the classic late-imperial Chinese novels — the “Dream of the Red Chamber” and “Journey to the West” treat the Canal as a normal part of the Chinese landscape; “Water Margin” treats the Canal as a site of banditry and trade; “The Plum in the Golden Vase” uses the Canal as the principal setting of one of its most-cited sections. The poetry of the late Ming and early Qing is filled with the Canal: the Wen Zhengming “pavilion poems” of the early sixteenth century, the Kangxi emperor’s poems of the southern tours, the Qianlong emperor’s poems of his 1751 southern tour, the Wu School paintings of the Suzhou section of the Canal in the late Ming, the early Qing court painter works on the southern tours all of which treat the Canal as a subject of imperial interest. The cultural geography of the Canal is large and dense.
The politics of the Canal have been unusually consistent. The Canal was the principal piece of state infrastructure in imperial China for most of its active period, and the central state was deeply involved in maintaining it. The provincial administrations were involved in Canal maintenance through their corvée labour allotment; the local magistrates operated the local grain warehouses; the imperial household operated the imperial barges that the emperors themselves used on the southern tours. The Canal’s politics have been, since the Sui, those of a national project, even when its physical building was done in pieces over a thousand years. The Canal was a subject of imperial edicts, of bureaucratic politics, of tax policy debate, of grain pricing policy, and of local-versus-central disputes over who paid for the maintenance dredging. The Canal was a major piece of state infrastructure even in the late Qing, which attempted to use it as a logistics artery during the crises of the late nineteenth and early twentieth centuries.
The modern history of the Canal mirrors, in a compressed form, the larger political history of modern China. The collapse of the Canal’s grain traffic in the late nineteenth century coincided with the decline of the imperial state; the failure of the Republican state to rebuild the Canal as a working waterway mirrored the failure of the early-twentieth-century Chinese state to modernise in other ways; the partial revival of the Canal under the PRC state is one of the small but real achievements of the new state. The Canal’s current dual role as a working waterway in the south and a heritage site in the centre and north mirrors the longer Chinese history of the canal: a piece of infrastructure whose working and heritage values have been kept in tension throughout the period of its existence. The next twenty years will probably see the Canal either stabilise in its current dual mode or be re-incorporated into a wider Chinese inland-waterway modernisation. The latter is the trend; the former is the policy.
References
- The Grand Canal of China — Jean Dehergne, in Journal of the Royal Central Asian Society, 1964
- The Grand Canal of China: an Engineering Wonder — Ronald G. Knapp, China Heritage Quarterly, 2009
- The Grand Canal (UNESCO World Heritage Site) — UNESCO World Heritage Centre, inscription 2014, decision records and maps
- The Grand Canal of China: Economic, Government, and Institution Building from Sui to Ming — Zheng Kefa and others, World Bank China Working Papers, 1998
- The Grand Canal: A General Account of Its History, Construction, and Operation — Marjorie Nice Winterbotham, Far Eastern Survey, 1937
- The Grand Canal of China During the Sui and Tang Periods — Wolfram Eberhard, in History of China, vol. III, ed. H. F. Schurmann, University of California Press
- China’s Grand Canal: A Bibliography — Yuen Ch’ing, Harvard-Yenching Institute Bibliographical Series, 1970
- The Ming Grain Tribute System: An Interpretation of the Public Granaries — Ray Huang, in: The Ming Fiscal Administration, Stanford University Press
- The Hydraulic Management of the Grand Canal in Modern China — Peter H. K. Chan, China: An International Journal, 2010
- Das chinesische Kaiserreich und seine Wasserwirtschaft: Eine Erdkundliche Studie — Wolfgang Haering, De Gruyter / China Studies, 1973
The Canal in the 21st Century
The Canal in the early twenty-first century is, in the main, a working waterway with a heritage overlay. The traffic on the Canal is overwhelmingly barge traffic in the southern sections, principally along the Yangzi and the Huai basin and the lower Yangzi canals around Yangzhou, Suzhou, and Wuxi. The northern sections, between the Huai basin and Beijing, are still working sections but on a different scale. The freight figures have been collected since the founding of the Ministry of Water Resources and the Ministry of Transport and are widely available. The Canal carried roughly 100 million tons of freight in the year 1990 and roughly 700 million tons in the year 2020 across all of its sections, with the increase driven principally by bulk cargoes of coal, cement, sand, gravel, and grain. The traffic is dominated by short-distance bulk hauls rather than long-distance grain tribute, with most barge journeys running through six to twelve sections of the Canal per journey rather than the whole length. The cargo mix has been shifting since the early 2000s towards higher value-added freight, with container lines establishing inland-water shipping operations from Ningbo and Shanghai through the lower Yangzi canals. The container traffic is still a small fraction of total inland-waterway freight but has been growing rapidly.
The 2014 UNESCO inscription of the Canal as a World Heritage Site was the most consequential piece of state policy on the Canal since 1949. The Canal was inscribed on the World Heritage List by the World Heritage Committee at its 38th session in Doha in June 2014 under the cultural criteria (i), (iii), (iv), and (vi). The inscription covers approximately 1,011 kilometres of the Canal’s historic route through nine provinces and municipalities: Beijing, Tianjin, Hebei, Shandong, Jiangsu, Anhui, Zhejiang, Henan, and Hubei. The nomination dossier was prepared jointly by the State Administration of Cultural Heritage and the Grand Canal Cultural Belt Working Committee of the State Council between 2009 and 2014. The principal heritage works at the time of inscription included the restoration of the Weishan Weirs of the Ming double-lock system, the restoration of the Xuzhou Huanzhou shiplock, the restoration of the Yangzhou southern gate complex, the strengthening of the Qing Yan and Chenghuang waterways in Dezhou, and the restoration of the Hangzhou Xiangjiayan lock complex in the southern part of the Canal. The World Heritage Committee in its 2014 decision emphasised the heritage values of the Canal as a single inherited route rather than a set of waterways, and called for further conservation work on the locks and the historic cities along the Canal.
The period after the inscription has been one of consolidation rather than further inscription expansion. The State Administration of Cultural Heritage, working with the provincial-level Cultural Heritage Bureaus, has been progressively adding candidate elements to a tentative list, with the Jining-Shandong Canal system, the Suzhou canals, and the Yangzhou heritage complex the most consequential of these. The 2022 Plenary session of the World Heritage Committee inscribed one additional canal section, the Qing Yan Canal extension, as an extension to the existing site. The next inscription round, expected to bring several additional sections into the network, is scheduled for the mid-2020s. The local governments of the Canal cities have been major investors in the heritage-protection regime, with the provincial-level governments of Jiangsu, Zhejiang, Beijing, Tianjin, Hebei, Shandong, Anhui, and Henan collectively spending more than 100 billion yuan on Canal-related heritage projects between 2014 and 2024.
The Grand Canal Cultural Belt, written into the State Council’s Five-Year Plan for Economic and Social Development in 2021, is the principal policy framework that the Canal now operates within. The Belt is conceived as a coordinated set of investments in Canal-related tourism, Canal-related ecological restoration, Canal-related urban development, and Canal-related industrial repositioning. The Belt covers the same nine provinces and municipalities as the heritage site, with the State Council setting specific targets for the Belt that include the conservation of historic cores, the development of urban waterfronts, the establishment of long-distance Canal-side bikeways and walkways, the protection of canalside cultural heritage, and the development of canal-side exhibition and conference facilities. The actual achievements on the ground between 2021 and 2026 include the open of the Beijing-Tianjin-Hebei Canal Cultural Belt Network Visitor Centre in 2024, the completion of the long-distance Grand Canal cycling route between Beijing and Hangzhou in 2025, and the establishment of the Grand Canal Cultural Belt Museum in Yangzhou in 2023. The associated conservation work on canal-shipping has slowed somewhat in the past two years, partly because of a policy emphasis on the Canal as a cultural heritage resource and partly because of the rising traffic of bulk gravel and sand traffic that has been damaging the heritage structure of the Canal’s masonry sides.
The ecological concerns have become a more prominent strand of public discussion of the Canal in the late 2010s and early 2020s. The Canal flows through nine provinces with very different climatic and water-management regimes, from the semi-arid plains of Hebei and Shandong in the north to the humid subtropical lowlands of Jiangsu and Zhejiang in the south. The water regimes are very different: the northern sections of the Canal have been chronically short of water for at least a decade, and the southern sections periodically receive too much water during the late-summer rainfall season. The Ministry of Water Resources has been progressively shifting water-management practice from the older priority of hydraulic control to a more contemporary priority of water-quality control, with successive ministerial plans in 2010, 2015, and 2020 establishing progressively stricter water-quality targets for the Canal. The 2020 Plan calls for all sections of the Canal to reach Class III water quality by 2035, which is a stringent target for the southern sections and an exceptionally stringent target for the northern sections. The progress has been uneven: the southern sections have largely met the standard by 2024, while the northern sections remain short of the standard, particularly in the Shandong Canal.
The traffic on the Canal in the late 2010s and early 2020s has been characterised by the rise of recreational boating alongside the older bulk-cargo and container traffic. The Yangzhou- Hangzhou section of the Canal has been the centre of the recreational boating industry in China, with several hundred vessels operating on the section by the mid-2020s and a major industry in charter-boat operations, recreational marina services, and waterborne tourism. The Hangzhou section of the Canal, with the famous “Qiantang” riverboat operating on the lake and the Canal, has been a particularly strong market for canal-related tourism. The Beijing section of the Canal, with the University Town section of the Grand Canal Cultural Belt, has been the centre of a different kind of waterway use: a recreational walking and biking path along the canal’s banks, an outdoor fitness area, and a heritage-walk experience. The 2025 Hangzhou Canal Cultural Belt Visitor Centre, with its 7,200-square-metre exhibition space on the heritage of the Canal from the Sui period to the present, has been a major piece of heritage infrastructure, and the 2024 Wangfujing Canal Park in Beijing has been an in-town centrepiece for the new style of canal use.
The grander narrative of the Canal in the twenty-first century is, in summary, a piece of the larger story of how China has been managing its inheritance from the imperial period. The Canal was the principal piece of state infrastructure of the imperial period; it became a heritage asset of the early-twentieth and late-twentieth century; and it has become a heritage-with-traffic asset of the twenty-first century. The state’s investment in the Canal since 2014, particularly since 2021, has been unusual in the larger Chinese heritage context because the Canal is so long and so multifunctional that it does not fit easily into any of the existing state policies for “linear heritage” sites. The 2021 State Council decision to establish the Grand Canal Cultural Belt as an explicit policy framework was an attempt to give the heritage-management policy a single institutional foundation. The next several years will probably see the Canal Cultural Belt as the central institutional framework for the management of the Canal and its heritage cities.
The Canal exists today, in working order, in heritage status, and in cultural-policy terms, as a single inherited piece of infrastructure that is now over fifteen centuries old. The Sui-Tang Canal and the Ming Canal are not, separately, working waterways today, but their preservation in place as monuments of pre-modern Chinese statecraft is also a piece of the Canal’s twenty-first-century identity. The UNESCO inscription was structured precisely on the inheritance logic, with the inscription text emphasising the running nature of the Canal across more than ten centuries as the principal cultural value. The Canal today is, in the Chinese official phrasing, “the longest canal in the world and a single piece of inherited civil infrastructure.” The phrasing is a heritage-policy phrasing; it does not fully capture the working reality of the Canal, but it does capture what the modern state is trying to conserve.
