emperor must strive to deserve and maintain, it also simply refers to the sky or the heavens. Unaccompanied by any discussion of divine symbolism, therefore, the “heavenly point” in these ceiling paintings merely indicates the highest point in a view that opens up toward the sky. The use of linear perspective in The Study of Vision therefore carries neither European nor Christian metaphors, regardless of the Jesuit origins of the subject in China. Despite the best efforts of the Jesuits and persistent scholarly attempts to link linear perspective in The Study of Vision with the connotations of Jesuit perspective, any deeper metaphorical significance for perspective or the vanishing point in The Study of Vision is uniquely Chinese.
It is entirely possible that Nian personally experienced the effects of such ceiling paintings, and that these images in the seventh section reflect not only the Perspectiva but also the appearance of now-lost paintings in Beijing’s Jesuit churches. Jean-Baptiste
du Halde recorded secondhand that Giovanni Gherardini had painted quadratura inside the French Jesuit North Church in 1703, including Pozzo’s trademark false dome:
The ceiling is all painted: it is divided into three parts; the middle part represents a dome, all open, of rich architecture. It has marble columns [that] support a range of arcades on top of which there is a nice balustrade. The columns themselves constitute a finely drawn balustrade with nicely placed vases of flowers. High above among the clouds over a group of angels, the Heavenly Father is holding the terrestrial globe in his hands. The Chinese cannot believe that all this has been painted on one plane, and cannot be persuaded that the columns are not straight as they seem to be. The light that comes through the arcades and the balustrades is so wisely painted that one can easily be deceived. . . . The altarpiece is painted too: both sides of it represent the continuation of the architecture of the church in perspective. It was amusing to see the Chinese visit that part of the church which seemed as if it was behind the altar: when they arrived at it they stopped, then stepped back a little, then forward again and put their hands on it to find out that there were really no reliefs or hollows.63
Gherardini’s paintings at the French North Church offered the first known example of Pozzo’s trademark motif in Beijing, as well as the earliest example of quadratura and the fact that Chinese outside the court were deceived by such paintings and read them as real, permeable spaces. Between 1715 and 1728, Castiglione also painted Pozzo’s false dome in the Portuguese East Church (Dongtang), and may have done so in the South Church (Nantang) as well, giving Nian multiple opportunities to experience the sort of overhead illusions he included in his treatise.64
The Jesuit churches in Beijing welcomed the public to visit. Ferdinand Verbiest recorded that nearly all officials came to Beijing at least once in their lives, and visited both the churches and the Jesuit residence: “moved by the fame of European products and by the desire to see them . . . with great admiration they contemplate long and intensely the paintings and the other European objects which we intentionally exhibit there, especially those things that reveal some work of rare skill.”65 Given the discrepancy in the elite’s ability to marvel at such paintings even as they criticized the artist’s technical skills required to produce them, the rhetoric inherent in this statement remains to be explored. But considering Nian’s connections with particular Jesuits on both the French and Portuguese missions, as well as his high official position and his time spent in Beijing, he would have had multiple opportunities to experience the illusionistic false domes in the Jesuit churches. Nian would also have been able to see Gherardini’s quadratura in the North Church and Castiglione’s quadratura on all four walls of the South Church, which included episodes from the life of the first Christian Roman emperor, Constantine I (r. 306–337). The literatus, civil official, poet, and amateur painter Yao Yuanzhi (1773–1852) described the detailed fictive architectural spaces created by Castiglione’s paintings on the east and
west walls: wholly believable rooms with curtains or blinds, bookshelves full of books, objects in the room casting shadows, and sunlight shining on the floor through an open window in the paintings. He concluded by noting that “inside the room are several tables that are distantly visible, so orderly it is as if you can walk in [to the room]. Nearing it, it is still a wall. There was no linear perspective [xianfa] in antiquity: because it is as precise as this, I regret that the ancients did not see it, so I record it in particular.”66 In mistakenly regretting that people in the past did not have the opportunity to experience such paintings, Yao demonstrates just how profound Chinese art-historical amnesia was regarding its own illusionistic mural tradition.
Korean visitors to these churches made similar comments, demonstrating a consistent response to the deceptively three-dimensional quality of the paintings as well as the openness of the churches to receiving visitors. In 1720, Yi Kiji (1690–1722), son of the Joseon mission envoy to the Kangxi court, recorded seeing scientific instruments, optical devices, a book on natural history, and Western-style paintings. He initially mistook these paintings for real images and was subsequently embarrassed, but he also described his general impression of Beijing’s Jesuit churches and their paintings as “another world” (bie shijie).67 Hong Daeyong (1731–83), the Joseon envoy to Beijing in 1766, also marveled at his own sensory response and the emotions he experienced because of the paintings in these churches, specifically mentioning the method of spatial recession and calculation underpinning the illusions. Only after inquiring directly did he discover that one of the paintings “was not a real door but a picture on the wall in order to show the skillfulness to onlookers. I approached the wall doubtfully and found that it was not real but painted on the wall. This was enough for me to imagine the skillfulness of Western painting.”68 Although these paintings might not have been considered true art in the classical literati definition of the term, their effects on these highly educated Korean viewers were nonetheless considerable and remarkably consistent.
It is precisely this sort of skill and its ensuing deceptive effects on the viewer that Nian takes up in the eighth section (53v–56v) of The Study of Vision. It directly follows the section on illusionistic ceilings, the implication being that the next set of paintings would create the same sort of illusion directly in front of the viewer as the previous section of paintings did overhead. Using a detailed (if entirely idiosyncratic) method for depicting spatial recession (figure 2.11), the multipart illustration demonstrates how to depict a richly appointed room full of young boys actively engaged in riding hobbyhorses, brandishing toy weapons, tugging small animals on wheels behind them, and amusing themselves with a variety of other toys (figure 2.12). A canopied ceiling recedes toward a mountainous landscape visible beyond the background doorways in the far distance, potentially suggesting a garden area. This subject matter would have been immediately legible to a Chinese viewer. The multitude of boys and their playthings identifies the image as a “boys-at-play picture” (yingxitu), an auspicious traditional subject for paintings and woodblock prints that offered the viewer or recipient of such a work with a wish for
many sons to carry on the family line, while the lanterns also suggest the traditional theme of a New Year’s image celebrating the lantern festival.69 Nian divides this image into six layers (ceng) of overlapping images, each with the same eponymous grid (quadro) that aided the European quadraturisti in transferring their original cartoons to walls.70 The implication of Nian’s grid is that the contents of each image layer corresponded with various squares in the grid and could therefore be recreated simply by scaling the contents of any given square in the grid onto a larger square.
Each image occupies progressively less space relative to the layer that precedes it. An elaborate architectural frame provides the frontmost
