Jump to ContentJump to Main Navigation
Picturing Technology in ChinaFrom Earliest Times to the Nineteenth Century$

Peter J. Golas

Print publication date: 2015

Print ISBN-13: 9789888208159

Published to Hong Kong Scholarship Online: September 2017

DOI: 10.5790/hongkong/9789888208159.001.0001

Show Summary Details
Page of

PRINTED FROM HONG KONG SCHOLARSHIP ONLINE (www.hongkong.universitypressscholarship.com). (c) Copyright Hong Kong University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in HKSO for personal use (for details see www.hongkong.universitypressscholarship.com/page/privacy-policy). Subscriber: null; date: 25 September 2018

Han to Tang

Han to Tang

Realism on the Rise

(p.13) Two Han to Tang
Picturing Technology in China

Peter J. Golas

Hong Kong University Press

Abstract and Keywords

Chapter 2 deals with the period mainly from the third to the tenth centuries when the continuing paucity of surviving illustrations of technical subjects obliges us to examine general developments in painting in order to tease out what they might be able to add to our knowledge of technical depictions at this time. We shall see that, during these centuries, Chinese painting remained generally committed to realistic narrative representations, with many artists devoted to creating greater verisimilitude in their paintings. But just as the prevailing aesthetic values were encouraging greater realism in painting and drawing, the almost exclusive use of the Chinese brush for visual representations may have impeded the development of certain representational techniques. Moreover, while the invention of woodblock printing toward the end of this period made possible a much wider reproduction and circulation of illustrations, it was a technology that also contained in itself the potential to inhibit advances in illustration techniques.

Keywords:   China, Scientific illustrations, Drawing techniques, Technical development, Literati, Science, Wood block printing, Agriculture, Industry

What we can say about graphic illustrations of technology in the centuries from the end of the Han to the end of the Tang is severely limited by a dearth of surviving drawings or paintings containing elements of technological interest.1 All too often we can only speculate even about very important topics. For example, the increasing availability of paper in China from at least the fourth century CE2 must have led to a greater production and circulation not only of texts but also of illustrations, some of which surely included technological subjects. On the other hand, at least by the seventh to the ninth centuries during the Tang, literary evidence indicates that the government was already wary of the spread of information that might be considered a threat to the political authorities. While astronomy and astrology as well as divination especially fit into this category, it would be surprising if the same attitudes did not sometimes apply also to technologies such as the making and use of weapons.3

The literary evidence also refers on occasion to drawings that dealt with specialized subjects, some of which may have been technological. The bibliographic chapters of the official histories are a case in point. They include many works now lost that had the word tu 圖‎ or yang 樣‎ (drawing) in their titles, usually indicating that they included visual depictions of some kind (pictures, diagrams, etc.) When writing the Illustrated Standards for Machines (Qi zhun tu 器準圖‎) in the early sixth century, Xindu Fang 信都芳‎ benefitted from being able to draw on a very large library of writings and illustrations belonging to his patron, an imperial prince who was intrigued by all kinds of scientific apparatuses.4 Even earlier, there is considerable evidence for architectural drawings, none of which seem to have survived.5

(p.14) Literary accounts can also suggest developments that may have encouraged the making of drawings that contained what we might regard as technological elements. For example, it appears that the post-Han centuries witnessed considerable interest in mechanical devices or toys, a subject to which we shall return shortly. Given the complexity of these automata, it is likely that some use must have been made of working drawings in their construction. Also, toward the end of this period, we have an important document from the Jiu Tang shu 舊唐書‎ (Old history of the Tang dynasty) relating that the court issued a drawing (yang 樣‎) of a standard model of chain pump (shui che 水車‎) and ordered the authorities of the capital area to have a number of the machines built to be used by the people for irrigation.6 What is especially suggestive here is that it was apparently taken for granted that the illustration was sufficiently accurate and detailed to serve as a construction guide. But in this case as in others, while the texts sometimes provide enough information to enable us to imagine what the devices themselves may have looked like, the lack of a single surviving drawing means we can only speculate about the appearance and techniques of the drawings.

In any case, it seems safe to conclude that the necessity of producing each drawing individually made illustrations of any kind relatively uncommon during those centuries that preceded the invention and spread of woodblock printing. It is true that there appeared at this time a number of techniques for making the process of copying more accurate (always a big problem, greater even for illustrations than for texts)7 and perhaps somewhat more efficient. But the graphic skills required and copying costs still must have impeded any extensive production and circulation of illustrations.

Aesthetics and Realism to the Fore

During the centuries from the Han to the Tang, moral concerns played an important role in Chinese painting. John Hay would even argue that “almost all painting [down to the seventh century] was essentially narrative in the service of ethical precepts, Confucian and Buddhist alike.”8 Ancient beliefs in the magical potential of paintings also still survived. In the fourth century, one of the first great Chinese painters, Gu Kaizhi 顧愷之‎ (c. 344–406), was said to have moved a woman by painting her portrait and then pricking the heart area with a needle.9 As late as the thirteenth century, there was a belief that the replacement of the (p.15) Pictures of Tilling and Weaving10 on the walls of the imperial palace contributed to political troubles that followed.11

After the Han, however, aesthetic criteria loomed increasingly large in the assessment of paintings. This stimulated improvements in painting technique, including techniques that would enable artists to reproduce more exactly what they saw in the world around them.12 Already in the Han, painters had been experimenting with techniques such as modeling strokes and hatching to convey a better sense of contours and volumes.13 These experiments continued in the succeeding centuries. In the sixth century, the artist Zhang Sengyao 張僧繇‎ (502–56), apparently borrowing a technique from India, won praise for the realism of his trompe l’oeil flowers painted on a Buddhist temple.14

The availability of new techniques, however, did not lead inevitably to their wide adoption. On the contrary, techniques seemingly full of promise for further development were used in a rather halfhearted fashion by only a few artists who seldom showed much interest in exploring their further possibilities. Illusionistic shading provides a good example. Shading has frequently been thought of as an import along with other techniques and styles of Buddhist art. A lacquer painting from the Later Han, however, shows it to have been known in China well before the case can be made for a Buddhist influence (Fig. 2.1).15 (p.16)

Han to TangRealism on the Rise

2.1 Detail of a lacquer painting of paragons of filial piety on a basket-work box from Lolang, Korea, c. 100 CE

Shading was also one of the techniques employed in the remarkable paintings of the late sixth century tomb of Lou Rui (婁睿‎) in Shanxi in which realistic portrayal reached heights not seen in earlier painting.16 It continued in use into the Tang and is not infrequently to be seen even in later paintings, both colored and monochrome.17 A most remarkable painting of bamboos (Fig. 2.2) shows what Chinese painters were able to achieve when they pursued the possibilities offered by shading.18 By and large, however, the Chinese seem never to have done much to develop the technique beyond a very basic level.19 Of special importance for our story, neither shading nor modeling strokes found their way, except rarely, into the line (p.17)

drawings that provide most of our examples of graphic portrayals of technology.

Han to TangRealism on the Rise

2.2 Anonymous tenth-century(?) painting of Bamboo and Old Tree Growing by Rocks

Apart from specific techniques, however, it was the overall Chinese approach to drawing, in these centuries as well as later, that was most important for limiting technological painting. Overwhelmingly, drawing in China was what the Chinese called baimiao 白描‎ (plain drawing) or outline ink monochromatic drawings. Wen Fong’s description of baimiao draws an important contrast:

Plain drawing is not drawing at all in the Western sense of the word, since it does not build form with multiple modeling strokes. Rather, it is monochromatic ink painting with a single brushline, a modulated ink outline that captures and reveals the essence of an object spontaneously and completely, without correction.20

A moment’s reflection suggests how antithetical such a “spontaneous” approach could be to drawings with a technological purpose where one is not interested in the “essence” of a machine but rather in a depiction that reveals accurately and in some detail what the machine looks like or how it is constructed or how it works.

Scale Drawing and Perspective

The Han and post-Han period also witnessed the emergence of drawings to scale, especially in astronomical calculations and the construction of astronomical instruments as well as in the making of maps. The first explicit reference to a scale drawing seems to occur in The Arithmetical Classic of the Gnomon and the Circular Paths of the Heavens (Zhou bi suan jing 周髀算經‎) from about 100 BCE.21 We shall return to this in Chapter 3 when we discuss Su (p.18) Song’s clock tower. In the case of mapmaking, which was usually seen in China as closely related to painting, the use of scaled distances (though not a clearly measured surface on which to draw them) may have begun as early as the third or fourth century BCE.22

Between the Han and the Tang, the spread of Buddhism in China led to the painting of many large-scale murals. Sarah E. Fraser has examined in detail the kinds of preliminary drawings used in the production of these murals.23 One might expect that the difficulty of moving from smaller design to large mural would have encouraged the production of scale drawings or at least reduced-scale sketches24 which, perhaps in conjunction with a grid overlay, could then have been systematically enlarged to provide the patterns for the final mural.25 Nevertheless, among the surviving materials (admittedly very limited), we have no examples of scale drawings made for this purpose.26 In the same way, one might have expected scale drawings to have been used in the production of the massive Buddhist icon sculptures so characteristic of this period. Nevertheless, here too we have neither surviving examples or even textual references suggesting the use of scale drawings.

In the area of building technology, by contrast, a piece of evidence from as early as the end of the fourth century BCE might be taken to indicate not only the use of scale drawings but even scale models. It consists of a unique artifact, the Zhaoyutu 兆域圖‎ (Picture of the omen area), a ninety-six by forty-eight centimeter bronze plaque found under the tomb of a feudal ruler of the kingdom of Zhongshan in Hebei that gives a bird’s-eye view of the ruler’s tomb complex (Fig. 2.3).27 This roughly scale portrayal in plan of the site and its buildings makes clear that both the concept of a site plan and the notion of scale drawing must have had some currency even by the fourth century BCE. On the other hand, one can question


Han to TangRealism on the Rise

2.3 Bronze plaque showing plan of the Zhongshan mausoleum (late 4th century BCE) together with a reconstruction substituting modern characters for the archaic forms of the original. The inscriptions provide dimensions for the various elements of the complex.

the extent to which the Chinese at this time had techniques sufficiently precise to enable them to avoid inaccuracies in measurement that could lead to mistakes if the drawings were overly relied on for building.28 This, surely, is also at least part of the reason why the plaque contains inscriptions giving the dimensions of all the architectural features.29 Interestingly, (p.20) in contrast to the norm for later Chinese portrayals of technological objects where measurements were more often presented not with the images but in an accompanying text, the measurements here are closely related to the elements to which they apply.

How Chinese artists attempted to create a sense of receding distance in their images also influenced their portrayal of technical subjects. Early Chinese painters, like painters in other ancient cultures such as classical Greece, did not create a sense of real space in their paintings (it probably never occurred to them as something they ought to do).30 Cahill speaks of their constructing a composition by juxtaposing forms without integrating them. “Between and around [the forms] was void; space had no existence except as that which separated one image from another.”31 Actually, this breaking up of space into cells32 or “compartmentalized units” (paralleled in Chinese views of time)33 can be seen as another example of the modular thinking to which the Chinese were so prone, what we might call a kind of digital rather than analog way of perceiving.34

During the period from the Han to the Tang, however, some Chinese artists did begin to experiment with various methods to convey not only a stronger feeling of real space35 but also a sense of receding distance in their paintings. Perhaps the most common early means of suggesting distance, at least in pictures presenting broad vistas, was to vary the size and/or height of objects, so that objects drawn relatively smaller or placed higher were understood to be deeper in the scene and therefore farther away from the viewer.36 Another early development that some artists had mastered at least by the first century BCE, was the use of perspective where parallel, diagonal lines strike off from the plane of the picture.37 Very effective, especially in landscape paintings, was the use of atmospheric projection in which the manipulation of colors, tints and washes created the sense of distance. Finally, as (p.21) early as the sixth century, Chinese painters were even experimenting with vanishing point perspective.38 A very interesting example is the famous depiction of a Buddhist paradise in a Dunhuang fresco from about 700, a careful examination of which reveals a “modular” approach to perspective in which several separate vanishing points are used.39

By the beginning of the Tang, then, Chinese painters were increasingly able to depict a tolerably convincing if not visually accurate world of space and nature.40 In later centuries, culminating in the Five Dynasties and Song periods, the best artists learned to create, in Cahill’s term, a “truly coherent spatial vision.”41 Something of this idea comes through in the comments of Li Zhi 李廌‎ (late 11th–early 12th centuries) on the architectural painting of the tenth-century painter Guo Zhongshu 郭忠恕‎ (d. 997): “Roof beams, girders, pillars, and rafters are shown with open spaces between, through which one might move. Railings, lintels, windows and doorways look as if they could really be passed through, or opened and shut.”42 Nevertheless, unlike Western painters since the Renaissance, Chinese painters never developed a notion of space as a measurable geometrical entity with the circle as the ideal form and the triangle as the master of measurements. They therefore never developed mathematical rules to guide the painter in organizing space and in creating more precise spatial relations.43 As Robert Maeda notes, the Chinese painters were thus able to maintain a much greater sense of spatial dynamism than Western painters, but at the expense of “pictorial orderliness.”44

Moreover, little of the increasing ability to present a convincing sense of space was applied to the illustrations in the books that, at the end of this period, were being produced in ever greater numbers by means of woodblock printing. Since the majority of our surviving illustrations of technology come from these book illustrations, we shall have occasion to return to this point below as we try to determine what accounted for the particular character (p.22) of Chinese book illustrations and how this influenced the way the Chinese went about picturing technology.

The Dominance of Brush and Line

We have suggested above that the first Chinese use of basic drawing tools such as the straightedge or ruler, the compass and the square may date back to as early as the Neolithic. By the post-Han centuries, however, there arose a clear prejudice against the use of such tools for painting, at least any painting expected to be taken seriously as “fine” art. At the heart of this prejudice was the strong emphasis, also dating back to very early times, on the ability of paintings to convey a sense of life, movement, vigor, vitality. Even the combining of geometrical ornamentation with totally non-geometric portrayals of vegetation, fauna, and mythological creatures, for example on bronze vessels, served to introduce a remarkable exuberance to what otherwise could have been rather formal and static decorative designs as one sees, for example, in even the anthropomorphic subjects on ancient Greek vases (Plates 3 (a) and 3 (b)).

Han to TangRealism on the RiseHan to TangRealism on the Rise

Plate 3 (a) Bronze vessel, perhaps depicting a shaman and his alter ego, from Shang China (c. 1600–1050 BCE); (b) Ancient Greek vase depicting the myth of Hercules, as the fourth of his twelve labors, bringing the Erymanthian Boar back to Eurystheus. The exuberance of the decoration on the bronze contrasts with the rather static placidity of the vase decorations.

This strong aesthetic predilection for vitality and spontaneity in paintings does much to explain the importance of the brush in China not only for writing but also for painting.45 Throughout Chinese history, nearly all writing, drawing and painting has been done with a brush.46 Evidence especially from early pottery decoration shows that some kind of brushlike writing implement using flexible plant material was in use at least by 2000 BCE but quite possibly much earlier.47 Given that both animal hair brushes as well as dried ink have been discovered in Warring States tombs, we can be confident that writing and painting with brushes was widely practiced by the middle of the first millennium BCE.48 The almost universal use of the brush as an instrument for writing and painting was firmly established in the Han dynasty; it was never seriously challenged down to the twentieth century.

Much effort went into producing ever better and more varied writing and painting brushes until they became, in the words of James Cahill, “perhaps the most versatile and (p.23) responsive drawing implement devised by man.”49 The Chinese writing and painting brush consisted of a cluster of animal hairs arranged to form roughly a cone shape and fixed into the end of a bamboo tube.50 Softer hairs suited to absorbing ink or pigment formed the outer layers of the tip while stiffer hairs in the center provided resilience.

Long experience over centuries led to many improvements in the fabrication of brushes. By at least the third or fourth century BCE, brushmakers had discovered how to control the flexibility of brushes by arranging the hair at the tip in a series of concentric layers.51 Still more flexibility was achieved by either treating the core with wax, which produced a stiff and springy brush whose resilient tip was especially suited to producing lines of varying thickness, or by using softer hairs and leaving them un-waxed, resulting in a brush that could produce lines of more even width.52 Thick, soft-haired brushes were also produced for applying the “washes” that played such an important role in Chinese painting.53 Around the end of the Tang or beginning of the Song, a final major advance occurred that set the pattern for brushmaking that has predominated down to the present. Jean François Billeter seems to have been the first to note this advance and to summarize it:

[T]he improved technique consists in combining … longer and shorter hairs of the brush tip in such a way that the ink, instead of collecting in a single large circular reservoir, is retained in thousands of minute pockets distributed throughout a homogeneous tuft of hairs, wherever a shorter hair is packed between longer hairs. In this way the ink is more evenly distributed and its cohesive action is perfectly uniform.54

Considerable experimentation also led to the identification of many kinds of animal hair suitable for use in brushes, including rabbit and goat (long the most popular), fox, deer, sheep, and marten. Even mouse whiskers and the hair of human babies were tried. Marten and rabbit hair came to be preferred for fine line work.55 For marten hair, the best is considered to be that which comes from male wild martens (much better than that of domestic martens) hunted in the autumn.56


Han to TangRealism on the Rise

2.4 Detail from Zhou Wenju, In the Palace, showing effective use of calligraphy strokes in painting

Of all the standards for evaluating the quality of Chinese drawings and paintings, none was more important than the quality of the brushwork. As neatly summed up by Cahill, “the line drawn by a brush remains the central fact of Chinese painting throughout its history.”57 In contrast to European painters, Chinese painters never downplayed the importance of line in painting in order to concentrate on color or texture or surface contours or creating a sense of mass.58 Already in the late Zhou, it appears that artisan painters who produced wall and screen paintings for the royal court were divided into two groups: those who delineated shapes using fine lines and those who then filled in the colors. The first group, according to Silbergeld, seems to have enjoyed a higher status.59 In any case, down to the Tang, this “outline and color mode” predominated in Chinese painting.60 In Song times and later, the “tendency to work primarily in terms of line” was further reinforced by an ever-closer association of painting with calligraphy and by the rise of scholar-amateur painting as the most highly regarded kind of painting (not only in scholarly circles, but to a considerable extent among all those who took an interest in fine painting).61 An excellent example is the painting entitled In the Palace (Fig. 2.4), a twelfth-century outline copy (baimiao) of an earlier painting by Zhou (p.25) Wenju 周文矩‎ (active c. 940–97).62 It provides a particularly good example of how calligraphic brushstrokes could effectively be used in painting.63

This example alerts us to the fact that the Chinese brush, with all its advantages, was by no means equally well suited for all writing and drawing tasks. To some degree, the desire to draw different kinds of lines, as discussed above, was met by the creation of different kinds of brushes, each having its particular strength(s). But the lines on which Chinese artists focused in their paintings were not necessarily the same kind of lines that make for a good illustration of a mechanical apparatus. As Chiang Yee has noted:

Writing lines [i.e., the standard strokes of calligraphy] are used very extensively in Chinese paintings, which indeed are built up of them. Even solid objects usually have their contours outlined … [but] they seldom run continuously round the object. Except in delicate or elaborate work we prefer to use a few simple and carefully constructed strokes to convey the essentials of the form.64

It is not that the Chinese did not produce some brushes that were better than others for making outline drawings. And a painting master such as Wang Zhenpeng 王振鵬‎ (active c. 1280–1329) was said to be able to depict buildings in such meticulous detail that they could serve effectively as guides to construct the buildings (Plate 4).65 The point is that the Chinese writing and painting brush is much better as an instrument for aesthetic expression than for producing images marked by precision and detail. For example, it was always relatively difficult to get a line of consistent width when running a brush along a ruler.66 Given the general disparagement by the Chinese of painting and sketching that made use of straightedges, this particular limitation of the brush was of little importance to the Chinese artist.67 For accurate depictions of implements and mechanisms, however, lines just the opposite of those described by Chiang Yee are more appropriate, and here the exclusive use of the brush to the exclusion of any hard-tipped drawing instruments would have been a significant impediment.68

Han to TangRealism on the Rise

Plate 4 Detail of a fourteenth-century(?) copy of Wang Zhenpeng’s Dragon Boat Regatt a on Jinming Lake

(p.26) Another characteristic that limited the effectiveness of the Chinese brush for the drawing of technical images was the requirement that its ink supply be constantly replenished. The ink itself was typically prepared at each use by grinding an inkcake made of compressed soot (usually pine) on some kind of suitable stone using a bit of water to make it liquid. This made spur-of-the-moment, on-the-scene sketching inconvenient, especially outdoors.69 Not for nothing were many Chinese painters in the habit of observing landscapes as intensely as possible and then returning to their studios to paint the scenes from memory. This may have worked well enough for landscapes where absolute fidelity to the original scene was not considered important. But in the case of machines of any complexity, it was hardly likely to be an effective substitute for careful sketching while looking at the machine. Nothing in the artists’ background would have trained them to remember accurately technical details.

The nature of the brush also influenced the subjects Chinese artists chose to paint. Besides the aesthetic concern to create paintings full of life, there was also a preference for subjects that lent themselves especially well to depiction by brush and ink, subjects that included certain trees such as pines, bamboos and willows,70 strange rocks, towering mountains and soft southern landscapes. Even certain seasons: the tenth-century painter, Li Cheng 李成‎, was partial to wintry landscapes where the contrasts of the black-and-white scenery “served as ideal vehicles for displaying the extraordinary vitality of his brushwork.”71 The painting produced by literati from Song times onward set the standards for what constituted high art in painting. One result was a narrowing repertoire of subject matter that was rich in symbolic significance but that could also be satisfactorily represented by painters well-trained in calligraphy but generally not deeply interested in developing their painterly skills.72 By the Ming, with the increased availability of painting manuals or copybooks, even the actual way of portraying these subjects tended to become highly standardized.73 We can see these tendencies fully at play in the illustrations of China’s greatest technological (p.27) compendium, the 1637 Tian gong kai wu 天工開物‎ of Song Yingxing 宋應星‎ (discussed in Chapter 5), where the artists, for example, frequently introduced (certainly to no technological purpose) their beloved, highly conventional motifs such as banana trees or Lake Tai rocks to outdoor scenes, and delicate decorative screens to indoor workshops.74 Inanimate objects, by contrast, offered much less attraction, especially when they were difficult to draw.

Finally, what one might call the Chinese idolatry of the brush made Chinese artists remarkably impervious to the possibilities of other kinds of drawing implements. The Chinese brush may indeed be uniquely versatile but it certainly cannot match the range of possibilities offered by the variety of drawing instruments available to an artist in the time of, say, Leonardo da Vinci: not only the small pointed brush that had been in use since Roman times75 but also silver-or alloy-tipped rods for metalpoint drawing, quills and reeds, chalks and charcoal.76 In versatility, no single Western drawing instrument could rival the Chinese brush and ink. But used separately or in combination, the Western tools and materials offered a range of possibilities undreamed of in China.77 The availability of all of these possibilities was itself a powerful spur to experimentation and creativity in drawing, and led European draftsmen from the late Middle Ages on to develop a whole range of skills appropriate both to the requirements of precision and accuracy in technical drawings as well as to the aesthetic demands in what have since come to be known as “master drawings.”

Models, Automata, and Technological Drawing

Pre-Han and Han burial remains have yielded up many models of everyday objects, including implements, utensils and machines. They provide valuable information on the technology in use when they were made. Most of these were simple models that could undoubtedly have been constructed without any need for design plans simply by scaling down the dimensions of the original.78 One wonders, however, about more complex examples well represented by the spectacular bronze chariots with their four horses excavated from the tomb of the First Emperor of Qin (Fig. 2.5).79 Half the size of real chariots, these models were built with extreme precision; one was made up of 3,462 individual parts, surely one of (p.28)

Han to TangRealism on the Rise

2.5 Two four-horse chariots found in the necropolis of the First Emperor of Qin. Though greatly damaged by the earth that had collapsed on them, they have since been fully restored.

the most complex models ever constructed in the premodern world. Could such a model have been built without any plans or working drawings? Unfortunately, there is no evidence from early mortuary remains or from other sources pointing to the existence of such drawings. One might also ask if it is reasonable to expect, again in the absence of any evidence, that the Chinese were capable at this time of producing working drawings that would have been of real use in planning and constructing such a complex model. Perhaps this model was achieved by systematically scaling down the parts of a full-sized chariot and relying on something in the way of assembly drawings or diagrams.

From early times, the Chinese used models of various kinds and materials (e.g., wood, clay, papier-maché, wax) for designing buildings and other structures or for aiding in their construction.80 Tomb reliefs and models from the Warring States and Han periods give us a good picture of their appearance, including the complexity they sometimes displayed (Fig. 2.6). Where the making of architectural models was aided by the use of drawings of some kind,81 they might have served as the predecessors of a rich history of architectural drawing to which we shall return below.82 (p.29)

Han to TangRealism on the Rise

2.6 Fortified manor house dated 76 CE. Within the four towers and two pavilions are two buildings of two rooms each. Inside are also figures engaged in agricultural and domestic activities.

Another use of models may have been in the designing and construction of complex astronomical instruments. It is difficult to believe that a complicated machine such as Zhang Heng’s water-powered armillary sphere would have been built in bronze without the use of one or more wooden prototypes.83 Zhang was reputed to be skillful at painting/drawing (shan hua 善畫‎)84 and evidence from the titles of books that have not survived suggests strongly that there were also drawings made in connection with the armillary sphere. In the time of Emperor Yang of Sui (r. 605–617), there was at the imperial court a major project to recreate objects from the past that ran a gamut from clothes and many kinds of vehicles to even a seismoscope for detecting earthquakes. The process often seems to have consisted of looking up illustrations in old books, building models on the basis of the illustrations, and then having workers refer to the models when building the full-size versions.85

(p.30) A final and perhaps the greatest stimulus to drawings with technological content in the post-Han centuries were the mechanisms constructed to amuse princes and their courtiers in the many feudal courts of the period.86 Such toys or “automata” had made their appearance in China before the Han, though the extremely sparse sources hardly support even a guess as to when or to what extent. The somewhat richer evidence from the post-Han centuries, however, reveals considerable ingenuity applied to producing automata of various kinds. Especially popular were doll-like figures that could bow, pour wine, play musical instruments, and dance. The interest in amusing mechanisms flowed over into at least somewhat more functional areas, leading to the invention of “self-moving” carts, doors that could open and shut automatically, several versions of south-pointing carriages (a cart with a standing figure that, by means of gearing, always pointed to the south whatever the turns of the cart),87 and hodometers (vehicles able to measure distance traveled).88 The importance attached to these mechanical devices is suggested by the fact that quite a number of their inventors or improvers are known to us by name, an exception to the anonymity so often surrounding inventors and craftsmen in early times.

Unfortunately, our sources almost always limit themselves to describing what the automata could do and tell us nothing about how they did it, i.e., any details of their mechanical construction.89 This was due no doubt in part to the fact that the inventors of the mechanisms had good reason to keep them secret. Many of them used their secret knowledge and technical skills to win or maintain the favor of emperors or other patrons.90

As suggested above, it is difficult to imagine that these mechanisms, some of them quite complex, could have been designed and constructed without at least some recourse to sketches and perhaps even construction drawings of some kind.91 That nothing of this kind has survived is hardly surprising. While the wider availability of paper encouraged more “writing” of all kinds, texts and illustrations, its fragility assured that virtually all of this writing would fail to survive.92 With the exception of the considerable number of (p.31) documents and books from the cache found in a Dunhuang cave in western China at the beginning of this century (very few of them predating the seventh century),93 nearly all of the manuscript materials from these centuries have long since perished. To be sure, much of the textual material in these manuscripts has survived either because the texts were printed once woodblock printing came into wide use or because they were cited by later writers who still had access to the originals. But this is much less true for illustrations.

What is more surprising, however, is the silence of the Chinese literature not only in this period but in later periods as well on the subject of, to use Graham Hollister-Short’s term, sketches as “vehicles for ideas.”94 Ma Jun 馬鈞‎, whose puppet theater we have just referred to (see note 89), presents an interesting example. We know a great deal about him, thanks to an essay written by his friend Fu Xuan 傅玄‎.95 But it seems that Ma’s mechanical genius was matched only by his inability to explain his work. Thus we read in the essay: “[Ma’s] powers of exposition fell far behind his mechanical ingenuity, and I doubt if he could express half of what he knew” and “Mr. Ma’s gifts are all of the mind and not of the tongue.” In connection with his work on a south-pointing carriage, Fu comments: “But again it was almost impossible [for him] to describe (the principle of it) in words.”96 After several statements of this sort, one is left puzzled why the idea of drawings is never raised. It seems that, at least among those relatively rare figures who combined an education in the highly word-oriented culture of traditional China with a serious interest also in technology, using drawings to work out mechanical ideas was seldom if ever attempted.

(p.32) The Advent of Woodblock Printing

Before the invention of woodblock printing, the Chinese made use of a variety of methods for reproducing manuscript texts and illustrations. These included free hand copies (by far the most common method), tracing, grid techniques, and the use of a kind of stenciling, the pounce method, in which powder made from chalk or some other material was tamped through small perforations made along the major outlines of an original design, leaving behind guide points for redrawing the design (Plate 5).97

Han to TangRealism on the Rise

Plate 5 Image of the Buddha with pounce perforations, ninth or tenth century, Dunhuang.

Courtesy of the Bibliothèque Nationale de France.

All of these methods suffered from more or less serious disadvantages. They were slow and tedious, hence relatively costly. Moreover, the very tediousness encouraged frequent errors. In the copying of written text, the need to preserve meaning served as a check against mistakes, although sometimes a very imperfect one. It obviously did not work very well with copyists of limited literacy. The problem could be even greater with illustrations, however. Illustrations of course possessed their own kind of “meaning,” but it was of a visual kind that worked much less effectively as a check on accuracy. Add to this the much lesser effectiveness of collating in the reproduction of images as well as the use of conventions in the original that might be unfamiliar to the copyist and we can understand how, as Richard Smith has pointed out, a “copy” of an illustration or a painting could range from a tracing to a free interpretation in the manner of the original.98 Actually, close copies made by tracing seem to have been the exception; most often, the “copies” were to a greater or lesser extent revised renderings.99

Illustrations of technology were also prone to certain factors beyond sheer carelessness that could introduce distortions and mistakes into copies. Common among them was a less than full understanding on the draftsman’s part of how a machine of some complexity actually worked.100 This problem could be exacerbated if the original used conventions unfamiliar to the copyist.101

(p.33) By at least the fifth or sixth centuries, and quite possibly even earlier, the Chinese had devised a technique that greatly facilitated the duplication of texts. It consisted of inscribing texts either in intaglio or in relief and usually on stone. Thereupon, copies (rubbings, dab prints, ink squeezes) could be made by inking the surface, covering it with a piece of paper that was then patted to facilitate transfer of the image to the paper, and then carefully removing the paper.102 As early as the seventh century, carved blocks of wood were being substituted for the stone carvings and carvers were beginning to discover what kinds of wood103 as well as what carving tools104 worked best for different purposes. Now not only text material but also illustrations could be rapidly reproduced in large numbers and at relatively low cost. As the process came into wider use, books became available in unprecedented numbers throughout China. The quality of the printing also gradually improved so that, even by the ninth century, Chinese woodcuts had reached a high level of artistic and technical maturity.105

The ability to produce a large number of exact copies by means of woodblock printing seems to have been first exploited on a large scale by Chinese Buddhists.106 During the centuries from late Han to the Tang, Buddhism provided a two-fold stimulus to woodblock printing. On the one hand, the missionary zeal of the Buddhists encouraged the reproduction of large numbers of Buddhist sacred texts and religious pictures.107 Secondly, and this is especially important for illustrations, the “religious discipline” of the Buddhists meant considerable emphasis was placed on getting right a complex iconography.108 These two stimuli came together in the pounce or stencil method. Used in the painting of the famous cave murals at Dunhuang and elsewhere, it helped achieve a high degree of iconographic accuracy while at the same time making possible speedier completion of complex murals since the same design could be used repeatedly.109 (p.34)

Han to TangRealism on the Rise

2.7 Earliest portrayal (c. 950) of a “fire lance” making use of a non-explosive gunpowder mix. Buddhist painted banner from Dunhuang.

It is not easy to assess the extent to which the pounce method may have been a stimulus to woodblock printing. Certainly, the two techniques are intrinsically very different. Nevertheless, both had as their goal easier and more accurate reproduction, if in very different media. More important, however, is that both techniques in responding to the desire for iconographic accuracy could get the details right to a degree that was previously rare in any kind of Chinese art and that ran counter to the general neglect of detail that was so typical of much early Chinese art. How far such capabilities had developed in woodblock printing even by the ninth century can be seen in the frontispiece of the world’s earliest surviving printed book (in scroll format), the Diamond Sutra of 868 (Plate 6).

Han to TangRealism on the Rise

Plate 6 Detail of the ninth-century Diamond Sutra, a printed scroll 17 1/2 feet long, discovered in the Dunhuang cave complex in 1907. It portrays the Buddha, his disciple Sabhuti, and numerous other figures.

© The British Library Board, OR. 8210/P.2, Frontispiece and text.

The extent to which some of this new attention to detail might have appeared also in pictures of technological subjects is impossible to say. The surviving art of this period, most of it Buddhist, rarely portrays technological subjects beyond buildings. Even more rarely do we encounter an image that is significantly informative, such as the tenth-century portrayal of a fire lance (Fig. 2.7).110 Moreover, there is little either in Buddhist or in non-Buddhist writings during the first three or four centuries of woodblock printing to suggest a significant body of illustrations having to do with technology that may have existed but simply did not survive.111 It may well have been that this was a period when illustrations simply were (p.35) not commonly looked to for this kind of information.112 But as Chinese technology registered unprecedented advances in the Song, woodblock-printed books with illustrations of technological subjects came to play a significant role in that growth. It was already true then, as it has been since, that much of technological understanding could not be transmitted except by means of illustrations.113 This makes it all the more ironic that, at a certain point, a kind of inertia to which the technology was particularly susceptible set in so that, instead of a vehicle for spreading new knowledge, woodblock printing came often to embalm technology that was obsolescent if not obsolete. We shall return to this point later, especially in connection with Wang Zhen’s Agricultural Treatise. (p.36)


(2.) Tsuen-hsuin Tsien, “Chemistry and Chemical Technology: Paper and Printing,” Science and Civilisation in China (SCC), ed. Joseph Needham, vol. 5, part 9 (Cambridge: Cambridge University Press, 1985), 43; Frances Wood, Chinese Illustration (London: The British Library, 1985), 10.

(3.) Alexei Volkov, “Geometrical Diagrams,” in Bray et al., Graphics and Text, 429 and 432–34.

(4.) Wu Jiming, History of Chinese Drawing, 29; Needham and Wang, SCC 4:2, 35. The Chinese reference to this library speaks, probably with more than a little exaggeration, of its more than 10,000 items, many of which contained illustrations. By comparison, it is interesting that Cambridge University’s library had only 122 volumes in 1424. Norma Levarie, The Art and History of Books (New Castle, DE and London: Oak Knoll Press and The British Library, 1995), 67.

(5.) Wu Jiming, History of Chinese Drawing, 65 ff.; Lawrence Sickman and Alexander Soper, The Art and Architecture of China (Baltimore: Penguin Books, 1956), 243; Chen Tongbin 陳同濱‎ et al. (eds.), Zhongguo gudai jianzhu da tudian 中國古代建築大圖典‎ [Illustrations of ancient Chinese architecture] (Beijing: Jinri Zhongguo chubanshe, 1996), vol. 2, 1185.

(6.) Needham and Wang, SCC 4:2, 347. I feel quite confident that yang here means a drawing rather than a model.

(7.) Liu Keming 劉克明‎ and Gong Shixing 龔世星‎, “Zhongguo gudai gongcheng tuxue de ruogan chengjiu” 中國古代工程圖學的若干成就‎ [Some achievements of ancient Chinese engineering graphics], Gongcheng tuxue ji jisuanji tuxue [Journal of Engineering Graphics and Computer Drafting] 1(1992): 79.

(8.) John Hay, “‘Along the River during Winter’s First Snow’: A Tenth-Century Handscroll and Early Chinese Narrative.” Burlington Magazine 114 (May 1972), 298.

(10.) This work is discussed in the next chapter and in Chapter 6.

(11.) Pelliot, “A propos du Keng Tche T’ou,” 95. For further examples of supposed magical powers of paintings, see Golas, “Emergence of Technical Drawing,” 54n8; Liu Heping, “Water Mill,” 584; James Cahill, “Confucian Elements in the Theory of Painting,” in Wright, The Confucian Persuasion, 117.

(12.) Susan Bush, The Chinese Literati on Painting: Su Shih (1037–1101) to Tung Ch’i-ch’ang (1555–1636) (Cambridge, MA: Harvard University Press (Harvard-Yenching Institute), 1971), 13–14 and 17–18.

(13.) Mary H. Fong, “The Technique of ‘Chiaroscuro’ in Chinese Painting from Han through T’ang,” Artibus Asiae XXXVIII 2/3 (1976), 111. For examples of the use of modeling strokes, see Jan Fontein and Wu Tung, Han and T’ang Murals Discovered in Tombs in the People’s Republic of China and Copied by Contemporary Chinese Painters (Boston: Museum of Fine Arts, 1976), 29, Nos. 7 & 8; 43, No. 20 (ram at lower left); 74–75, Nos. 78 & 82; 113, No. 133; and passim; Wu, “Origins of Chinese Painting,” 37, Fig. 27. For an example of hatching, see Fontein and Wu, Han and T’ang Murals, 99, No. 120. The use of modeling strokes probably developed quite naturally out of the strokes used to depict folds in clothing, a technique already in common use in Han paintings. Hatching, however, may have been more difficult for Chinese artists to accept since, in direct contrast to the importance Chinese painters attached to individual (often calligraphic) lines in painting, hatching was a process that minimized the attention to the individual line.

(14.) Hubert Delahaye, “Du peu d’effet de la peinture occidentale en Chine aux XVIIe et XVIIIe siècles,” in Catherine Jami and Hubert Delahaye, L’Europe en Chine: Interactions scientifiques, religieuses et culturelles aux XVIIe et XVIIIe siècles (Paris: College de France: Institut des Hautes Études Chinoises, 1993), 245. For trompe l’oeil in late Northern Song, see Robert J. Maeda, “Spatial Enclosures: The Idea of Interior Space in Chinese Painting,” Oriental Art 3 (Winter 1985–86), 381 and many other examples from the Baisha tombs in the report of the excavations; Su Bai 宿白‎, Baisha Song mu 白沙宋墓‎ [The Song tombs at Baisha] (Beijing: Wenwu chubanshe, 1957).

(15.) Fontein and Wu, Han and T’ang Murals, 20 and 28; 29, No. 7; 44, No. 22; 86 and 89, No. 107. This conclusion is further reinforced by the thoroughly Confucian subject matter (paragons of filial piety) of the painting. See also Fong, “Technique of ‘Chiaroscuro,’” 115; Sickman and Soper, Art and Architecture, 64, 67; Sherman Lee, A History of Far Eastern Art (Englewood Cliffs: Prentice Hall and New York: Harry N. Abrams, 1964), 141–42; Cahill, Chinese Painting, 15, with a good illustration on p. 17. This calls into question Sullivan’s contention that shading was always regarded as a “foreign” technique, appropriate only to foreign subjects. Michael Sullivan, The Meeting of Eastern and Western Art from the Sixteenth Century to the Present Day, 2nd ed. (Berkeley and Los Angeles: University of California Press, 1989), 54.

(17.) See, for example, Zhao Mengjian’s Narcissus from the thirteenth century (Fong, Beyond Representation, 306–7, Pl. 68) and Li Gan’s Bamboo and Rocks from 1318 (Fong, Beyond Representation, 390, Pl. 86).

(18.) For another example, this one from the Ming, see Dai Jin’s 戴進‎ Bamboo, Chrysanthemum, and Garden Rock in Richard M. Barnhart, Painters of the Great Ming: The Imperial Court and the Che School (Dallas: Dallas Museum of Art, 1993), 185, cat. 50.

(19.) Cahill, Chinese Painting, 15. It is particularly interesting to realize that “shadowing” in the sense of a shadow actually cast by an object in sunlight seems, with a single exception, to be entirely absent in the Chinese landscape paintings that have survived. See Michael Sullivan, Symbols of Eternity: the Art of Landscape Painting in China (Stanford: Stanford University Press, 1979), 8; Anita Chung, Drawing Boundaries: Architectural Images in Qing China (Honolulu: University of Hawai‘i Press, 2004), 72, Fig. 2.8. Sullivan sums up the case nicely: “Chinese artists, if they used shading at all, used it very sparingly to make an object look more solid, while avoiding cast shadows and a single light source.” Sullivan, Meeting of Eastern and Western Art, 62, Caption to Fig. 44.

(20.) Fong, Beyond Representation, 35 and 60–61. Hence, most of the drawings we shall be examining have much more in common with comic books or manga than with fine art drawings as they emerged in the Renaissance Europe.

(21.) Wu, History of Chinese Drawing, 111; Cordell D. K. Yee, “Reinterpreting Traditional Chinese Geographical Maps,” in J. B. Harley and David Woodward, A History of Cartography, vol.2 book 2 (Chicago: Chicago University Press, 1994), 42. The earliest explicit reference to an architectural scale drawing may date from only the seventh century. Xiao Zhenshi, A History of Building Technology in China (Taipei: Po-yuan, 1993), vol. 3, 890.

(22.) Nancy Steinhardt, “Chinese Cartography and Calligraphy,” Oriental Art 43.1 (1997), 10, 11. For actual scales of early maps, see Yee, “Reinterpreting,” 41–46, and for ideas of Han intellectuals on the practice, ibid., 109–10. Yee nicely makes an important point well kept in mind when considering the use of “scale” in Chinese painting (and therefore also in other kinds of illustration): “In Chinese painting, … pictorial scale tends to dominate natural scale; that is, the size of an object depicted was determined by needs of design and not rules of geometric perspective. Foreground features might be diminished to avoid obstruction and overemphasis, while distant objects might be enlarged to act as counterpoint to the middle distance and foreground.” Yee, “Cartography in China,” 144.

(24.) Jerome Silbergeld uses the term “reduced-scale sketches” for those preliminary drawings that were not true drawings to scale. Silbergeld, Chinese Painting Style, 7.

(25.) Silbergeld, Chinese Painting Style, 7–8; Tseng Yu-ho Ecke, “A Reconsideration of Ch’uan-mo-i-hsieh, the Sixth Principle of Hsieh Ho,” Proceedings of the International Symposium on Chinese Painting (Taipei: National Palace Museum, 1972), 317. In Europe, it is not until the Renaissance that such grid drawings (necessarily drawn to scale) came to be regularly used for painting murals. Francis Ames-Lewis, Drawing in Early Renaissance Italy (New Haven and London: Yale University Press, 1981), 26.

(27.) Yee, “Reinterpreting,” 36–37; Wolfgang Behr, “Placed into the Right Position–Etymological Notes on Tu 圖‎ and Congeners,” in Bray et al., Graphics and Text, 117–19. Nancy Steinhardt (ed.), Chinese Architecture (New Haven and London: Yale University Press, 2002), 29, Fig. 1.19 presents a color photo of the plaque.

(28.) The ancient Greeks, for example, relied less on drawings in planning and executing their buildings and, instead, aimed for “precision in the widespread repetition of architectural components …” See Tom Porter, How Architects Visualize (New York: Van Nostrand, 1979), 2. Interestingly, the “repetition of architectural components” nicely describes the modular techniques of construction used in most Chinese buildings, a point to which we shall return below when we discuss the Building Standards, traditional China’s preeminent manual of building construction.

(29.) Yee, “Reinterpreting,” 36 Fig. 3.1 caption; Wu Hong, “Picturing or Diagramming the Universe,” in Bray et al., Graphics and Text, 195; Behr, “Placed into the Right Position,” 117.

(30.) Our modern consciousness of space influences what we look for in paintings. As Alfred Crosby nicely summarizes it: “For us today, things exist in space like vegetables in an aspic salad. The vegetables may be the chief items of interest, but the aspic is undeniably there occupying the area among the items of interest. We do not deny the aspic because it is transparent, and we rarely ignore space even if it is empty.” Alfred W. Crosby, The Measure of Reality: Quantification and Western Society, 1250–1600 (Cambridge: Cambridge University Press, 1997), 170. For the Greeks, see William M. Ivins, Jr., Art and Geometry: A Study in Space Intuitions (Cambridge, MA: Harvard University Press, 1946), 40ff.

(31.) Cahill, Chinese Painting, 12. For much the same kind of assessment, referring to Greek painting, See Ivins, Art and Geometry, 31.

(34.) Compare Cordell Yee’s characterization of the Chinese view of space as being “particulate.” Yee, “Cartography in China,” 145.

(35.) Contrast the two Dunhuang illustrations from the sixth and seventh centuries reproduced in Maeda, “Spatial Enclosures,” 375, Fig. 9 and 378, Fig. 12, the second displaying a “much more illusionistically viable space.”

(36.) Benjamin March pointed out many years ago that this technique was not normally used for short distances such as a group of buildings, i.e., the distances that would be more typical in technological illustrations; March, “Linear Perspective,” 117.

(37.) Powers, Art and Political Expression, 25; see also Wu, “Origins of Chinese Painting,” 31. The distinctive characteristic of isometric or parallel projections is that lines parallel in fact are also parallel in the drawing; see Needham, Wang and Lu, SCC 4:3, 114, Fig. 778.

(38.) Maeda, “Spatial Enclosures,” 375–76; Michael Sullivan, Chinese Landscape Painting. Vol. 2: The Sui and T’ang Dynasties (Berkeley: University of California Press, 1989), 91–94. It is also interesting that the frequently reproduced frontispiece from the earliest surviving woodblock-printed book (the Diamond Sutra, from 868) shows the Buddha in a three-quarter perspective view in place of the flat, full-front view commonly used; see Fig. 2.10.

(40.) For yet another example of the progress that was being made, compare Maeda, “Spatial Enclosures,” 375, Fig. 9 and 378, Fig. 12, together with Maeda’s accompanying discussion.

(41.) Cahill, Chinese Painting, 29. Nevertheless, Chinese painters even down to the Ming and Qing often continued to exhibit a preference for “pictorial” or “variable” scale where the size of objects depicted was influenced mainly by the requirements of design (avoiding for example overemphasis on certain objects or the obstruction of one object by another) rather than any rules of geometric perspective; Yee, “Cartography in China,” 144.

(42.) Robert J. Maeda, “Chieh-hua: Ruled-line Painting,” Ars Orientalis 10 (1975), 126.

(43.) This contrast between Western and Chinese approaches to space is nicely elaborated in Delahaye, “Du peu d’effet de la peinture occidentale,” 248–49.

(44.) Maeda, “Spatial Enclosures,” 370. March also points out that “there is greater freedom of design possible to a painter who is not bound by the rigid necessities of the single viewing point, vanishing points and other paraphernalia of Western perspective techniques.” Benjamin March, “Linear Perspective in Chinese Painting,” Eastern Art 3 (1931), 137.

(45.) Much has been written about the importance of the brush in Chinese painting and calligraphy. Among the works I have found most useful are: Jean François Billeter, The Chinese Art of Writing (New York: Skira/Rizzoli, 1990); Chiang Yee, Chinese Calligraphy: An Introduction to its Aesthetic and Technique, 2nd ed. (London: Methuen, 1954); Lothar Ledderose, “Chinese Calligraphy: Its Aesthetic Dimension and Social Function,” Orientations 17.10 (Oct. 1986), 35–50; Yolande Escande, “Perspectives et limites des recherches récentes sur la calligraphie et la peinture,” Revue Bibliographique de Sinologie 14 (1996), 219–42.

(46.) Of course, the early Chinese sometimes incised hard materials such as bone and stone; the oracle bones are the most striking example.

(47.) Sören Edgren et al., Chinese Rare Books in American Collections (New York: China Institute in America, 1984), 127; Thorp and Vinograd, Chinese Art and Culture, 38, 90. For a Yangshao pot decoration suggesting the use of a brush-like implement, see Wu Jiming, History of Chinese Drawing, 14, Fig. 11. Wu would date the appearance of crude brushes much earlier, perhaps even as early as 5000 BCE. However, no surviving brushes predate the Warring States period.

(50.) Now essential reading on the construction of Chinese brushes is the revisionist account by Billeter in The Chinese Art of Writing, 51–53. Billeter argues for a major advance in the making of Chinese brushes around the end of the Tang or the beginning of the Song.

(51.) Billeter, The Chinese Art of Writing, 51; Tsien Tsuen-hsuin, Written on Bamboo and Silk: The Beginnings of Chinese Books and Inscriptions (Chicago: Chicago University Press, 1962), 161.

(52.) Silbergeld, Chinese Painting Style, 5. Such brushes were especially suited for the line drawings that made up the great majority of Chinese book illustrations. See also Chung, Drawing Boundaries, 9–10.

(54.) Billeter, The Chinese Art of Writing, 52. See also the further description that shows why, remarkably, “the quantity of ink released at any given moment is always exactly proportional to the width of the stroke” (pp. 52–53).

(55.) Chiang Yee, The Chinese Eye: An Interpretation of Chinese Painting (Bloomington and London: Indiana University Press, 1970), 193–97. At present, hairs from goats, rabbits and martens (lang 狼‎, an animal related to the weasel, but usually translated “wolf”) are the most commonly used.

(56.) Billeter, The Chinese Art of Writing, 53. The extreme variety of hairs used in Chinese brushes is but one more among endless examples of a highly developed Chinese sensitivity to the technological advantages and disadvantages of various natural materials. This probably traces back in large part to the extreme importance in Chinese technology of agriculture and the efforts over the centuries to maximize agricultural production as well as to the extensive exploitation of a wide variety of minerals in her underappreciated mining industry. See Peter J. Golas, “Chemistry and Chemical Technology: Mining,” in Science and Civilisation in China (SCC), ed. Joseph Needham, vol. 5 part 13 (Cambridge: Cambridge University Press, 1999).

(57.) Cahill, Chinese Painting, 11. What is not often stressed when comparing Chinese and Western painting values, however, is that European painters of the late Middle Ages and the Renaissance were also keenly concerned with line. They simply held quite different ideas about what made for a good line in painting and drawing. The Chinese ideas were almost exclusively based on aesthetic concerns focusing on the brushstroke while European artists were much influenced by the importance of geometry at this time. Jean François Billeter points out that Western artists often tended to think of line as a “geometric abstraction which helps us to grasp space in intellectual terms.” Lines that did this had to be “thin, uncluttered, free of any variations or apparent irregularities.” Billeter, The Chinese Art of Writing, 46.

(58.) A 1603 collection of reproductions of paintings by great painters of the past produced by Gu Bing 顧炳‎, rested on the assumption that monochrome woodblock prints could provide adequate reproductions even of paintings in which color had figured prominently. Craig Clunas, Pictures and Visuality in Early Modern China (Princeton: Princeton University Press, 1997), 138 ff., esp. 143.

(60.) James Cahill, “Approaches to Chinese Painting: Part II,” in Yang, Barnhart et al., Three Thousand Years, 9.

(63.) The copy also illustrates the absence of color typical in baimiao paintings (which we shall discuss further in a moment). The original had actually been in full color. Fong, Beyond Representation, 35.

(65.) Fong, Beyond Representation, 397. This is not quite the exaggeration it might appear to be at first reading. A painting could indeed serve this function in major part because of the modular techniques of construction in Chinese architecture that we shall discuss in Chapter 3 when considering the Building Standards.

(66.) Gisbert Combaz, “La Peinture chinoise vue par un peintre occidental,” Mélanges chinois et bouddhiques 6 (1938–39), 56. The kind of result you are likely to get can be seen in Billeter, The Chinese Art of Writing, 45, illustration on lower right. On the same page, Billeter makes the important point that the Chinese did not think of the brush as an instrument to draw evenly thick or thin lines as we think of them but rather to produce “forms of another sort.” On the other hand, as we have seen, Jerome Silbergeld points to a distinction between waxed core and unwaxed core brushes with the latter producing a more even line of a kind that was used in ruled-line (jiehua) painting. Silbergeld, Chinese Painting Style, 5.

(67.) We shall return to this in our discussion of “ruled-line” painting in Chapter 3.

(68.) The Chinese also never attempted to use the flexibility of the brush to draw different kinds of lines to distinguish, for example, visible and invisible parts of a machine as Western draftsmen came to use broken and solid lines.

(69.) This is not to say that no Chinese artists ever made sketches from life. For some examples, see Cahill, The Painter’s Practice, 88; Sullivan, Symbols of Eternity, 8–9; Liu, “Water Mill,” 593n91. But even in these rare cases, the sketches seem to have been highly abbreviated. Finished paintings, even of landscapes, were virtually always a studio product. Compare Cahill, The Painter’s Practice, 88 and Liu Heping, “Painting and Commerce in Northern Song Dynasty China,” 143n68.

(73.) Craig Clunas, Art in China (Oxford and New York: Oxford University Press, 1997), 177, 191; Anne E. McLaren, Chinese Popular Culture and Ming Chantefables (Leiden: Brill, 1998), 64. This Chinese form of standardized painting did not, to be sure, completely close the door to creativity. As Lothar Ledderose points out, using Zheng Xie 鄭燮‎ (1693–1765) as an example: “Painters like Zheng Xie strive to emulate nature in two respects. They produce large, almost limitless quantities of works and are enabled to do so by module systems of compositions, motifs, and brushstrokes. But, they also imbue every single work with its own unique and inimitable shape, as nature does in its prodigious invention of forms. A lifetime devoted to training his aesthetic sensibilities [rather than his artistic skills?] enabled the artist to approximate the power of nature. For the Chinese literati painter, modular systems and individuality are but two sides of the same coin. Its name is creativity.” Ledderose, Ten Thousand Things, 213.

(75.) Henry Petroski, The Pencil: A History of Design and Circumstances (New York: Knopf, 1989), 7.

(76.) Watrous, Craft of Old-Master Drawings. The lead pencil had apparently not yet been invented; there is no written or pictorial evidence for its existence before 1564. Petroski, The Pencil, 7.

(77.) To take the goose-quill pen as a single example: it “can be cut in many different ways to produce different shapes, and varying degrees of fineness, of the nib. Because it is an organic material, the quill is flexible and malleable … If finely cut, the quill responds to slight changes of pressure or of speed of movement to leave thickened or tonally strengthened lines.” In an experienced hand, the quill will glide over the paper surface with smooth fluency. Ames-Lewis, Drawing in Early Renaissance Italy, 46.

(78.) For references to portrayals of many of these models, see Golas, “Emergence of Technical Drawing,” 30.

(79.) Shaanxi sheng Qin yong kaogudui 陝西省秦俑考古隊‎and Qin Shihuang bingmayong bowuguan 秦始皇兵馬俑博物館‎ (eds.), Qin ling er hao tong che ma 秦陵二號銅車馬‎ [A bronze chariot and horses from the Necropolis of Qin] (Xi’an: Kaogu yu wenwu, 1983), 1–2, 65; Ledderose, Ten Thousand Things, 56, 3.9.

(80.) Klass Ruitenbeek, Carpentry and Building In Late Imperial China: a Study of the Fifteenth-Century Carpenter Manual Lu Ban jing (Leiden: Brill, 1993), 49–50, 169; Needham, Wang and Lu, SCC 4:3, 105–6; Xiao, History of Building Technology, 3, 890–91; Adam Schall, Historica Relatio (Tientsin: Hautes Études, 1942), 90–91. By contrast, architectural drawings and models are remarkably absent in early Europe; Needham, Wang and Lu, SCC 4:3, 107, citing L. F. Salzman.

(81.) The earliest use of drawings to portray buildings could well go back to the beginnings of the Zhou dynasty (Wu Jiming, History of Chinese Drawing, 65–6). Wu suggests that “plans” of some sort must have been regularly used to get rulers’ approval for the appearance of the palaces they ordered built. We have specific evidence in Sima Qian’s Records of the Historian showing this was certainly true in the case of the Qin First Emperor. Ibid., 67–68; Xiao, History of Building Technology, 3, 888.

(82.) In our discussion of one of the masterworks of Song technological writing, the Building Standards (See Chapter 3), we note Needham’s suggestion that it may have been in the drawing of buildings that Chinese technological drawing reached its zenith, at least insofar as precision was concerned. Needham, Wang and Lu, SCC 4:3, 106–7.

(83.) Almost a millennium later, Su Song and his collaborators on the great Northern Song clocktower received government authorization to build the final version only after completing a working wooden model for approval by the emperor. Joseph Needham, Wang Ling and Derek J. de Solla Price, Heavenly Clockwork: the Great Astronomical Clocks of Medieval China (Cambridge: Cambridge University Press, 1986), 9, 20, 21.

(85.) Wu, History of Chinese Drawing, 30. Only rarely does there seem to be any hint in the Chinese sources, at this time or later, of the complications that could be posed by the transition from even carefully constructed scale models to functioning, full-scale instruments or machines; see for one example, Needham, Wang and de Solla Price, Heavenly Clockwork, 128n8. On the limitations of models as indicators of how the full-scale machine will work, see Arnold Pacey, The Maze of Ingenuity: Ideas and Idealism in the Development of Technology (Cambridge, MA: MIT Press, 1992), 95; and Friedrich Klemm, A History of Western Technology, trans. Dorothea Waley Singer (London: Allen and Unwin, 1959), 162–64.

(89.) Needham and Wang, SCC 4:2, 160, 162. An interesting exception to the silence on mechanisms is the statement that a mechanical puppet theater made by the famous third-century engineer Ma Jun 馬鈞‎ was powered by a horizontal water-wheel, though we have no further information on how it worked; SCC 4:2, 158.

(90.) Examples include the just-mentioned Ma Jun; Xie Fei 解飛‎ who devised a large sandalwood cart with animated religious figures for the Emperor Shi Hu 石虎‎ of Later Zhao in the mid-fourth century; and Huang Gun 黃袞‎ who constructed one-third life-size boats with animated figures for the Emperor Yang of the Sui dynasty in the early seventh century. Needham and Wang, SCC 4:2, 158, 159, 160.

(91.) For a good idea of the complexity that could be involved, one can turn to just a few of the proposed reconstructions for south-pointing carriages; see Needham and Wang, SCC 4:2, Figs. 528, 529, 531, and 533.

(92.) The earliest piece of paper with writing on it so far found in China dates from c. 110 CE; Tsien, SCC 5:1, 42, Fig. 1061. It was during the Eastern Jin (317–420) that paper largely if not completely supplanted bamboo or wooden tablets as the material on which books were written. Silk would continue to be used, but it was too expensive for writing on in any but very special circumstances. Tsien, SCC 5:1, 43; Wood, Chinese Illustration, 10. See also Tsien, SCC 5:1, 30 on the measured pace with which newer writing materials replaced older kinds.

(94.) Graham Hollister-Short, “On the Origins of the Suction Lift Pump,” History of Technology 15 (1993), 69.

(95.) The full translation of this essay appears in Needham and Wang, SCC 4:2, 39–42 and 158, and Joseph Needham and Robin D. S. Gates, “Chemistry and Chemical Technology: Military Technology: Missiles and Sieges,” in Science and Civilisation in China (SCC), ed. Joseph Needham, vol. 5, part 6 (Cambridge: Cambridge University Press, 1971), 200.

(96.) Needham’s take on these statements is not entirely convincing: “The fact that Ma Chün (Ma Jun) could not … explain his ideas to the supercilious scholars of the Chin (Jin) court only meant that he was neither a philosopher nor an orator, it did not mean that he could not explain to his own artisans exactly what he wanted made in the world of gear-wheels and link-work.” Joseph Needham, The Grand Titration: Science and Society in East and West (Toronto: University of Toronto Press, 1969), 39. Though he is making an argument against exaggerating the idea that the Chinese classical written language inhibited the advance of science and technology in China, the question here is whether Ma was simply not very adept in the classical language or, what seems more likely, whether his verbal abilities in general were limited, whether using the classical language or ordinary spoken Chinese. For a more social take on these descriptions of Ma, see Barbieri-Low, Artisans, 201.

(98.) Richard J. Smith, China’s Cultural Heritage: the Qing Dynasty 1644–1912, 2nd ed. (Boulder: Westview Press, 1994), 204 and 209. Indeed, the “original” served often as a stimulus rather than a pattern. Chou Ju-hsi, “Painting Theory in Eighteenth-Century China,” in Willard Peterson et al., The Power of Culture (Hong Kong: Chinese University Press, 1994), 324.

(99.) What Ivins notes about copyists in Europe has at least some relevance for China: “When it came to copying a picture, … the copyist felt under no obligation to be faithful to either the peculiar forms or the linear syntax of the earlier draughtsman he thought he was copying.” Ivins, Prints and Visual Communication, 61. See also Arnold Pacey, Technology in World Civilization: A Thousand-year History (Cambridge, MA: MIT Press, 1991), 27 and, for some excellent European examples from as late as the sixteenth century, showing how even trained copyists could bungle the portrayal of relatively simple mechanical elements, Eugene S. Ferguson, Engineering and the Mind’s Eye (Cambridge, MA: MIT Press, 1992), 107–13.

(100.) One sees this with particular clarity in the efforts of Chinese artists to copy Western technological illustrations as late as the seventeenth and eighteenth centuries. For examples, see Golas, “Technical Representation in China: Tools and Techniques of the Trade,” EASTM 20 (2003), 18–20, 26–27; Needham and Wang, SCC 4:2, 212–14, Figs. 465–68; 548, Fig. 679 and Pl. CCLXVII, Fig. 680.

(101.) Francesca Bray, “Biology and Biological Technology: Agriculture,” in Science and Civilisation in China (SCC), ed. Joseph Needham, vol. 6, part 2 (Cambridge: Cambridge University Press, 1984), 63.

(102.) The spread of this process as well the following woodblock printing process went hand in hand with the increasing availability in these centuries of suitable kinds of paper.

(103.) For example, plankwood could produce only limited tonal gradations compared with end-grain wood. Chia, “Text and Tu in Context,” 243n4.

(104.) Carvers eventually came to recognize at least sixteen different cutting techniques that could be used with the quan dao 拳刀‎, a standard cutting knife; David Barker, Traditional Techniques in Contemporary Chinese Printmaking (Honolulu: University of Hawai’i Press, 2005), 26–27.

(106.) There is one minor qualification to be made here concerning the word “exact”: woodblocks did show the effects of wear and tear over time and use. On occasion, this could result in information being lost. For a striking example, see Golas, “Emergence of Technical Drawing,” 45.

(107.) Lucille Chia, “Mashaben: Commercial Publishing in Jianyang from the Song to the Ming,” in Smith and von Glahn, Song-Yuan-Ming Transition, 286. Personally copying religious works or subsidizing their reproduction was for Buddhists a major means of accumulating spiritual merit. The Diamond Sutra to which we shall refer again in a moment was subsidized by a certain Wang Jie as a meritorious act “on behalf of his [presumably dead] parents.” Wood, Chinese Illustration, 11.

(111.) This contrasts with architectural drawings, none of which have survived but which we know from references in written sources were produced and used with some frequency. Wu Jiming, History of Chinese Drawing, 73.

(113.) Often enough, even illustrations were able to convey technological knowledge only imperfectly (Basalla, Evolution of Technology, 83–84). We shall see shortly one very good example in the weeding illustrations of the Pictures of Tilling and Weaving which, by themselves, give us virtually no information on how this task is performed, mostly under water.