A History of the World in 12 Maps

by Jerry Brotton

  Fig. 9 Depiction of the round heavens and square earth, from Zhang Huang, Tushu bian, 1613.

  From the sixth century onwards official histories regarded the huntian theory as dominant, although strains of all three recur throughout Chinese astronomy, cosmology and cosmography, and the theory was not without its own ambiguities. Although its metaphor of the earth as a ‘yolk’ at the centre of heavens suggests a spherical world, the theory was often starkly illustrated showing a square, flat earth encircled by the heavens; and even this assumption was not absolute. Chinese astronomy was already using armillary spheres (depicting the heavens as a sphere), whose surviving calculations, based on detailed observations, assume a circular earth to represent the cosmos. Nevertheless, a foundational belief that runs through these theories is the conviction, the first surviving record of which is first found in a mathematical work from the third century BC, that ‘Heaven is round, Earth is square’.20

  This belief was based on an even more basic principle that pervaded early Chinese culture, which organized terrestrial space according to the ‘nonary square’, ‘one of the great world-ordering discoveries or inventions’21 of ancient China. A nonary square is divided into nine equal squares, creating a three-by-three grid. Its origins remain obscure, ranging from the archaic observation of the shape of a turtle shell (with its round carapace covering the square plastron), to the more convincing explanation that the vast plains of northern China inspired a rectilinear way of understanding and dividing space.22 The Chinese celebration of the square was in direct contrast to the Greek philosophical (and geographical) ideal of the perfect circle. The nonary square also established the number nine as central to the classification of virtually every sphere of classical China: there were nine fields of heaven; nine avenues of the capital; nine divisions of the human body; nine orifices; nine viscera; nine wells in the realm of the dead, even nine branches of the Yellow River.

  These divisions originated in one of the most important foundational texts of classical Chinese culture, the ‘Yu Gong’, or ‘Tribute of Yu’ chapter of the Book of Documents, compiled some time between the fifth and third centuries BC, regarded as the oldest surviving book of Chinese geography. The book describes the legendary ruler Yu the Great, founder of the Xia dynasty in remote antiquity (c. 2000 BC). Yu is said to have brought order to the world following the great flood by organizing fields and channelling the rivers.23 Starting with the Yellow and Yangtze river basins, ‘Yu disposed the lands in order. Going along the mountains, he put the forests to use, felling the trees. He determined the high mountains and the great rivers.’24 The territory was demarcated into nine provinces (or ‘palaces’), described in nine land and nine river itineraries. The nine provinces are described as a three times three grid, with the length of each side of the nine squares measuring 1,000 li (one li was equal to approximately 400 metres).25

  As well as ordering the space of the known world according to the figure of nine, the ‘Yu Gong’ also offered a schematic division of the whole world into five concentric rectilinear zones, oriented according to the four cardinal directions based on the winds. This was a classic example of egocentric geography. Civilization resides at the very centre of the image, representing the royal domain. The degree of barbarism increases with each square outwards, from tributary rulers, the marches, the ‘allied’ barbarians, and finally the zone of cultureless savagery, which included Europe. Once again, the contrast between this scheme and its Graeco-Roman counterpart is striking. Although Western zonal maps are also rectilinear, they are based on latitudinal zones, and are not defined by a symbolic imperial centre as in the case of the ‘Yu Gong’.26

  The nonary square and its figure of nine enabled Chinese mapmakers to draw on a cosmological world view, and apply it to political administration and practical policy. At a symbolic level, the relationship between the circle and the square enabled scholars to recommend a particular way of running an empire. According to one Qin dynasty writer, ‘[w]hen the ruler grasps the round and his ministers keep to the square, so that round and square are not interchanged, his state prospers.’27 At a more practical administrative level, the nonary square also drew on the so-called jing tian, or ‘well-field’ system of agricultural cultivation. The Chinese character for ‘well’ (jing) closely resembles the three times three grid, and was used as the basis for the allocation of agricultural land. A group of eight families would be allocated equal allotments of land, leaving the ninth (central) portion to be collectively cultivated. This orderly division of space was regarded as a basic element of social cohesion and effective government. ‘Benevolent government must begin with demarcating boundaries,’ argued the Confucian scholar Mencius (fourth century BC). ‘Violent rulers are always sloppy in the demarcation of boundaries. Once boundaries are correctly demarcated, then the division of fields and the regulation of salaries can be fixed without exertion.’28

  In the surviving records, the earliest descriptions of maps (or tu) are similarly associated with issues of dynastic rule and its administration. One of the first written references comes from the Zhanguo, or ‘Period of the Warring States’, c. 403–221 BC, when regional states battled for dynastic supremacy. In the Shu jing (Book of Documents, dating from the early years of this period), the duke of Zhou is recorded as turning to a map to choose the kingdom’s capital city of Luoyi, today’s Luoyang in the Henan Province, 800 kilometres south-west of Beijing:

  I prognosticated about the region of the Li River north of the He; I then prognosticated about the region east of the Jian River, and west of the Chan River; but it was the region of Luo that was ordered [by the oracle]. Again I prognosticated about the region east of the Chan River; but again it was the region of Luo that was ordered. I have sent a messenger to come [to the king] and to bring a tu [chart or map] and to present the oracles.29

  The duke’s prognostications about the location of the dynasty’s capital were informed by political geography as well as providence. Following the pronouncements of the ‘Yu Gong’, the duke concentrates his attention on the agriculturally and politically pivotal areas of the Yellow and Yangtze river basins. Whatever the duke’s ‘map’ actually showed, it was clearly being used as a supplement to these pronouncements for finding locations for the Zhou kingdom’s new capital, in an attempt to unify newly conquered political space with the legendary geography of the ancient sages.

  The iconography of maps played its part in subsequent key moments in Chinese dynastic politics. The Period of the Warring States came to an end in 221 BC with the rise to power of the Qin dynasty, unifying the Chinese kingdom under one rule. But it was not without a struggle: in 227 BC, prior to his accession, the first Qin emperor was attacked by an assassin using a dagger wrapped in a silk map of territories coveted by the Qin.30 Nor was the dynasty necessarily secure: one third-century scholar advised the anti-Qin states that he had ‘examined a map [tu] of the empire, according to which, the territory of the princes is five times larger than that of the Qin . . . If the six states were to join forces, head west and attack Qin, Qin would be smashed.’31

  Alongside such explicitly political and symbolic functions, maps were also regarded as part of the administration of dynastic rule. ‘The laws are codified in maps and books,’ wrote the philosopher Han Feizi (d. 233 BC). They were, he claimed, ‘kept in government offices, and promulgated among the people’. Despite such claims, other scholars were more sceptical. The Confucian philosopher Xun Qing (d. 230 BC) claimed that state officials ‘preserve the laws and regulations, the weights and measures, the maps and books’. But unfortunately ‘they do not know their significance, but take care to preserve them, not daring to decrease or increase them’.32

  One of the earliest surviving maps of this period is an engraved bronze plate from the end of the fourth century BC found in the tomb of the King Cuo, ruler of the Zhongsan dynasty during the Period of the Warring States. It depicts a series of golden and silver inla
id rectangles and squares, interspersed with text, and is hardly recognizable as a map. It is in fact a mausoleum plan, or zhaoyu tu, offering a topography according to nonary principles of the carefully planned funerary rites of the Zhongsan ruler. The plate’s outer rectangles represent two walls, between which stand four square buildings. Within the third rectangle is a raised mound, on which sit five square sacrificial halls, designed to cover the tombs of the ruler and members of his family. The zhaoyu tu, which absorbs the classical measurements of the nine provinces and the well-field system, is the earliest surviving Chinese example of a map-like object from a bird’s-eye perspective. It is also drawn to scale: the plate’s notes provide dimensions and distances measured in terms of chi (a foot, approximately 25 centimetres), and bu (paces, equal to around 6 chi).33

  Why the map was in the tomb remains unclear. Traditionally, tombs contained precious objects imbued with arcane power, designed to convey ritual respects to ancestors.34 The inclusion of the map could be a memorialization by a relatively sophisticated political administration of the Zhongsan ruler’s control over his terrestrial space, just at the point when he enters the spiritual world of the afterlife.

  Both the Qin dynasty and its successor, the Han dynasty (206 BC–AD 220) utilized maps in the drive towards political, administrative and military centralization. Yudi tu, or ‘maps of the empire’, were still used as ritualistic and commemorative devices, including the diplomatic exchange of maps with neighbouring kingdoms (such as Korea), and in the confirmation of military victories or the subjugation of subject states. But they were also beginning to permeate the administration of imperial governance. The Zhou li (‘Ritual forms of Zhou’) offered an ideal of Han bureaucracy in which maps were central to policy making. They were vital to water conservancy projects; in taxation, mining and the demarcation of roads; and the settlement of boundary disputes, the delineation of fields and the assessment of livestock; in auditing the distribution of the population; in maintaining the accounts of government officials, and in sustaining the loyalty of feudal states and their fiefdoms. In a sign of growing civic awareness of the importance of maps, two officials were appointed to keep their ruler informed about geography. Both men travelled with the emperor wherever he went. The tuxun, or royal scout, explained maps, while the songxum, or travel guide, deciphered local records where disputes arose.35

  The best example of the Chinese approach to maps from this period is the work of Pei Xiu (AD 223–71). Pei Xiu is often referred to as the Chinese Ptolemy, primarily thanks to his establishment of what he called the six principles of mapmaking. Appointed as Minister of Works under the first emperor of the Jin dynasty (AD 265–420), Pei Xiu composed a study of ancient geography which drew on the ‘Tribute of Yu’ text, leading to the creation of his Yu gong diyu tu (‘Regional Maps of the “Yu Gong”’), since lost. Pei’s approach is recorded in the surviving Jin shu (‘History of the Jin’), which describes how he ‘made a critical study of ancient texts, rejected what was dubious, and classified, whenever he could, the ancient names which had disappeared’. The result was the eighteen-sheet Yu gong diyu tu, which he presented to the emperor, ‘who kept it in the secret archives’. In making his map, Pei followed his six principles. The first was fenlü, the ‘graduated divisions which are the means of determining the scale’. The second was zhunwang, a ‘rectangular grid (of parallel lines in two dimensions)’. The third was daoli, or ‘pacing out the sides of right-angled triangles, which is the way of fixing the lengths of derived distances’. The fourth was gaoxia, measuring ‘the high and the low’, the fifth was fangxie, measuring ‘right angles and acute angles’, and the sixth principle was yuzhi, measuring ‘curves and straight lines’.36

  To a Western reader, Pei Xiu’s six principles appear to offer the foundations for modern scientific cartography, with their emphasis on the need for a graticule, the use of a standard scale, and the calculation of distance, elevation and curvature using basic geometrical and mathematical calculations. This was as good as anything the Greeks or the Romans had to offer at the time – but in China it did not translate into the development of a recognizably modern science of mapmaking, partly because Pei was not exclusively interested in this kind of mapping. His work was an early example of what Sinologists call kaozheng, or evidential research, which involved textual scholarship that recovered the past, paying particular attention to ancient texts as a guide to the present. This kind of scholarship also describes Pei’s cartographic method. He acknowledged that his work involved ‘a critical study of ancient texts’, and his maps did not rely on any direct topographical measurements, but were based instead on reading textual sources. For Pei and the new Jin dynasty, the task was to superimpose a new, updated geography onto the authority of the classical text of the ‘Yu Gong’. The reverence for and continuity with the past meant that Pei tried to combine the new with the old, to validate the past and legitimate the present in a graphic picture (and written description) of dynastic continuity.37

  So powerful was this textual tradition that Pei’s later followers even pointed to the limitations of visual descriptions of physical geography. ‘On a map’, wrote the Tang scholar Jia Dan (AD 730–805), ‘one cannot completely draw these things; for reliability, one must depend on notes.’38 For Pei, the textual, classical tradition is even present when he appears to be writing about space. ‘When the principle of the rectangular grid is properly applied,’ he writes, ‘then the straight and the curved, the near and the far, can conceal nothing from us.’39 This is simultaneously both a justification of his new quantitative principles of mapmaking, and a celebration of the classical textual tradition of dynastic administration based on the nonary square.

  Like so many maps made before the accession of the Ming dynasty in 1368, Pei’s have not survived. One that has is the famous ‘Yu ji tu’, or ‘Map of the Tracks of Yu’, made under the Song dynasty (907–1276)and dated 1136, which draws on the legendary exploits of Yu the Great. Joseph Needham has called it ‘the most remarkable cartographic work of its age in any culture’, arguing that anyone looking at contemporary European mappaemundi ‘cannot but be amazed at the extent to which Chinese geography was at that time ahead of the West’.40 The map was engraved on a stone stela 80 centimetres square and stood in the courtyard of a prefectural school in Xi’an, the capital of today’s Shaanxi province. Like Pei’s six principles of mapmaking, the ‘Yu ji tu’ initially appears strikingly modern. Its depiction of China’s outline is in many cases remarkably accurate. It is also the first known Chinese map to use a cartographic grid to represent scale, as recommended by Pei. There are more than 5,000 squares, and the sides of each one represents 100 li (more than 50 kilometres). This gives the map an estimated scale of 1:4,500,000. But this grid is not the same as a Western graticule. The graticule plots a location through latitude and longitude relative to the rest of the globe’s surface; the Chinese grid has no such interest in projecting a spherical globe onto a flat surface but simply helps to calculate distance and local area.

  Fig. 10 ‘Yu ji tu’ (‘Map of the Tracks of Yu’), 1136.

  On the opposite side of the stela is another map, entitled ‘Hua yi tu’, or ‘Map of the Chinese and Foreign Lands’. This map obviously complements the ‘Yu ji tu’ in some way, but how? Its scope is much broader, marking more than 500 places, including the rivers, lakes and mountains of the nine provinces, as well as the Great Wall in the north-east. It also depicts ‘foreign lands’ on the empire’s borders (including Korea), listing more than 100 others on the copious notes running around its edges. But this map is also very different from the ‘Yu ji tu’. It lacks a grid, the coastline is extremely vague and often erroneous (especially in the crucial Liaodong peninsula), and its river-systems are inaccurate. To understand what is happening here, the viewer needs to walk back round the stela to examine the ‘Yu ji tu’ again.

  Fig. 11 ‘Hua yi tu’ (‘Map of the Chinese and Foreign Lands’)
, 1136.

  Just as striking as the grid of the ‘Yu ji tu’ is its network of rivers criss-crossing its surface, with the Yellow River to the north, the Yangtze to the south, and the Huai midway between the two. Central to the map’s toponymy are mountain names, but it also includes cities and provinces. The legend at the upper left suggests that, once again, textualism is as important to this map as quantitative measurement. It reads: ‘Names of mountains and rivers from the Yugong, names of provinces and prefectures from past and present, and mountain and river names and toponyms from past and present.’41 The ‘Yu ji tu’ represents contemporary geography by describing legendary times and places. It is marked by references to the foundational text of the ‘Yu Gong’ and its description of a mythical, unified China defined by rivers and mountains. For example, Yu is said to have guided the course of the Yellow River from a place called ‘Jishi’, which is reproduced on the map, even though thirteenth-century scholars knew that the river originated in the Kunlun mountain range in north-western China. The map retains ‘Yu Gong’ geography even where more recent Chinese mapmakers had shown this to be incorrect.