Skip to main content Accessibility help
Hostname: page-component-768dbb666b-k9l4c Total loading time: 2.53 Render date: 2023-02-07T03:30:48.989Z Has data issue: true Feature Flags: { "useRatesEcommerce": false } hasContentIssue true

7 - Animals as Text

Producing and Consuming ‘Text-Animals’

Published online by Cambridge University Press:  08 December 2018

Roel Sterckx
University of Cambridge
Martina Siebert
Berlin State Library
Dagmar Schäfer
Max-Planck-Institut für Wissenschaftsgeschichte, Berlin
Animals through Chinese History
Earliest Times to 1911
, pp. 139 - 159
Publisher: Cambridge University Press
Print publication year: 2018

In her reconstruction of the different enactments or ‘versions’ of cow’s milk and mother’s milk in early eighteenth-century discourse, Kristin Asdal called working with archival texts ‘the historian’s version of field work’, because ‘objects are enacted in written materials too’.Footnote 1 To their readers and writers, objects acquire an existence and reality through texts, which can be questioned or taken as fact and – after passing through another’s mind – might be quoted verbatim or adjusted and written down in a different author’s own words. Texts do not establish a neutral signifier:signified relationship but produce or ‘enact’ a version, an identity and reality of the thing they are describing. In this chapter, I consider reading and writing about animals in historical China from and into a textual form as comprising an active, ‘praxeological’ relation between the scholar and the animal.Footnote 2 This means that the animal in a text, which I call the ‘text-animal’, represents one actual instance, reality and ontology of an animal to the scholar reading it, which he then interacts with.

This chapter juxtaposes texts written from an unbridgeable distance between the author and the animal – and therewith seemingly also a distance between text-animal and physical animal – with texts that document close, hands-on encounters with animals in the real world. The first group of texts consider their animals as ‘remote’, approachable through philology and anecdote. Although most of these texts deal with wild animals such as tigers, snakes or crabs, there are also examples concerned with horses and cats, i.e. species with which the author had probably experienced multiple physical encounters. In contrast, the second type includes texts about breeding, caring for and training animals used in competitions, such as goldfish, fighting crickets and quails. The authors of these texts were literally ‘close’ to their animals, for example staring into a goldfish bowl to find new morphological features or tickling a cricket with a rat whisker. From our contemporary perspective, we might ask what happened when the words on a page differed from what readers were seeing in real life. This raises several questions: did these discrepancies actually matter to Chinese scholars?Footnote 3 Was it possible for contrasting ‘versions’ of an animal to co-exist peacefully? What was considered all-encompassing knowledge about ‘close’ or ‘remote’ animal species? How was it possible to write about animals at all?

To explore these points, this chapter begins by providing some reflections on ‘animal texts’ in historical China. This includes the genre of specialized monographs on natural studies and material culture labelled pulu 譜錄 ‘treatises and lists’, and the way that animals developed into one of their main subjects over time. The chapter then examines a selection of pulu topics as case studies to explore how scholars produced and consumed animals as texts – either by collecting stories and narratives which circumscribed the species from a distanced, ‘remote’ philological perspective, or by trying to identify ever-new phenotypic variations from their own personal interactions with animals. The chapter concludes with some general reflections on ‘text-animals’.

Animal Texts

As the chapters in this volume illustrate, text-animals or reflections/reproductions of animals in texts inhabit many Chinese genres: from poetry to scholarly anecdotes, from philosophical and religious writings to historical and agricultural treatises, from encyclopaedic entries to administrative regulations.Footnote 4 Pulu, however, created a node or hub where knowledge about one specific animal or group of animals extracted from other genres of literature was accumulated and interacted with the author’s own practical or scholarly experience. Moreover, pulu constitute stand-alone monographs. Whereas in the other genres animals are embedded into larger writing projects, such as a collection of an author’s brush notes or poems, provide just one aspect in an all-encompassing compendium of agricultural or encyclopaedic knowledge, or are considered from the expert viewpoint of veterinary medicine or administration, each pulu focuses on one ‘thing’ – which could comprise an animal or animal species. In short, only the monographic genre of pulu writes exclusively about animals per se.

The definition of pulu as a genre stems from both their format and subject matter: they form a ‘treatise’ with topical chapters, a ‘list’ arranged into categories, or a combination of both. Their topic is a defined ‘thing’ or group of ‘things’ from the artificial or the natural world: for example, there are pulu on bronze objects, ink stones and tea; on cats, snakes, goldfish, birds and quails; and on animals as a whole. Pulu on animal species are similar in structure and approach to those on orchids and bronze objects. Whereas writing a poem about an animal species, including it in a pharmaceutical, culinary or agricultural treatise, or providing a description and calligraphy in an animal painting were readily deemed worthwhile scholarly undertakings, pulu, by contrast, needed time to develop into a genre that scholars wrote with self-confidence. Pulu authors mutually referred to each other in their prefaces, to justify the notion that collecting knowledge about just one species or one group of species was a worthwhile endeavour. As pulu developed into a specialized genre and framework of scholarly inquiry, Chinese bibliographical classification schemes were amended to make room for the new genre. In the late twelfth century the bibliophile You Mao 尤袤 (1127–94) was the first to adapt the traditional classification scheme to accommodate pulu in a new bibliographical slot exclusively dedicated to them. After the late eighteenth-century Complete Library of the Four Branches (Siku quanshu 四庫全書) stated that ‘We cannot but establish a class for pulu’ (pulu zhi lei, bu ke bu li 譜錄之類,不可不立),Footnote 5 pulu became integral to catalogues and bibliographies, and still remain so in Chinese rare book collections today.

Pulu became a popular scholarly genre in the Song dynasty due to a number of developments: a growing awareness of the empire’s southern parts with its flora and fauna, a flourishing trans-regional and international market offering new species and products, and the ever higher esteem given to encyclopaedic knowledge. Scholars have also suggested that the growing urbanization of Chinese scholarly lives led to Song garden and Ming leisure culture which, in turn, produced several pulu authors.Footnote 6 In the wake of all this progression, an urgent demand to capture and fix objects in a textual form emerged, via nomenclature, descriptions of whatever was known about them, etc. But the development of pulu writing also drew on famous precedents that pre-dated the formative stages of the genre by some centuries, namely Dai Kaizhi’s 戴剴之 Zhu pu 竹譜 (Book on Bamboo) from the fifth century and the Tang dynasty Cha jing 茶經 (Classic of Tea) written by Lu Yu 陸羽 (733–804 CE).

In their quest for historical precedents, animal authors went back even further into the past. Thus, Table 7.1 lists a number of texts from before the Song, including the famous Qin jing 禽經 (Classic on Birds), which is believed to date as far back as the Zhou. However, most of these pre-Song titles either are lost or only exist in fragments with apocryphal authorship. None of these texts survives in a version that pre-dates the thirteenth century – at least not in the form of an independent monograph. For example, the Yang yu jing 養魚經 (Classic on Raising Fish) ascribed to Fan Li 范蠡 (fifth century BCE) is quoted in full in the sixth-century Qimin yaoshu 齊民要術 (Essential Techniques for the Common People), even though it may have been intended as a collection of metaphors on good rulership rather than a technical tract. But, since these early titles were included in bibliographies from the Tang onwards,Footnote 7 they became accepted as apposite bibliographical entities. This ensured that scholars in general became aware of books on animals and enabled pulu scholars to refer to them to justify their own repertoires. Works like the Classic on Birds became exemplars which served as reference points for writing monographic texts specializing in animals.

Table 7.1 Pulu about animals through time

All animalsZhouHanSan GuoJinLiangSuiTangSongYuanMingQing
Quadrupedsquadrupeds (shou)1135
other quadrupeds2a1a4b2c9
Quadrupeds total:221112141141654
Birdsbirds (qin)½d½d1248
birds of prey12126
other birds3e1f9g13
Birds total:½½2662338
Fishfish & sea animals1211822
Fish total:116141840
Insects total:261624
Pulu on animals in total:33120142814482169
Pulu in total:33264328224223525421189

a: pig, sheep; tortoise

b: tortoise, qilin, 猩猩 (chimpanzee?), dog

c: dragon, lion

d: Qin jing by Shi Kuang, commentary by Zhang Hua

e: dove, duck, chicken, goose

f: phoenix

g: thrush (2), martin (2), cock(fight), sparrow, dove

By the eleventh century, pulu was an established genre known and accepted by scholars, especially on topics such as bamboo, chrysanthemums, peonies, tea and wine, as well as ink slabs, ink and bronze objects. At the same time, pulu on animals started to appear. Nevertheless, the high number of animal titles in the column for the Song dynasty in Table 7.1 is rather misleading, as it consists largely of undated titles listed in state and private Song bibliographies.Footnote 8

Among the first full-blown and still extant treatises on animals from the Song era, three titles stand out. These are Fu Gong’s 傅肱 (Song) Xie pu 蟹譜 (Book on Crabs), written in 1059, and its ‘expansion and correction’ compiled in 1211 by Gao Sisun 高似孫 (1154–1212) under the name Xie lüe 蟹略 (Crab Survey), and the Cuzhi jing 促織經 (Classic on Crickets) attributed to Jia Sidao 賈似道 (1213–75). The latter probably dates to Ming times but is traditionally labelled as Song.Footnote 9 These examples proved that the structure of pulu could also be used without any modification to write about animal species. Whereas the Xie pu adopted the ‘list’ style of Dai Kaizhi’s Zhu pu (Book on Bamboo), i.e. assembling pieces of knowledge in an encyclopaedic manner from a wide range of literature and hearsay, the Cuzhi jing was supposedly written by an aficionado of cricket culture in close contact with the creatures and with hands-on, practical knowledge, following the Classic of Tea’s chapter format. This tension between ‘remote’ and ‘close’ topics runs through the whole genre of pulu writing, in particular those about animals, which include fewer examples of ‘close’ animal species. This latter type of text documents a fascination with variety and variation that only becomes apparent from close examination. I will explore this point in more detail in the ‘Animals in Variation’ section, and explore pulu dealing with ‘remote’ animals or describing animals in a ‘distanced’, narrative style in the ‘Animals in Narration’ section below.

The total number of pulu texts about animals increased by half between the Song and the Ming, and almost doubled between the Ming and Qing. As the genre took shape and began to flourish in the Song, most pulu texts were written on topics such as chrysanthemums and bamboo, with animals only comprising one-eighth of all pulu titles in the Song and Ming. This figure rose slightly to between one-seventh and one-sixth during the Qing.

Three developments can be identified when reviewing these numbers: firstly, pulu on ‘all animals’ and on ‘fish and sea animals’ increased in popularity; secondly, ‘aficionado’ animals used in gambling fights or competitions of beauty and exceptionality such as crickets, quails and goldfish, became more popular; and thirdly, an increasingly greater spectrum of species attracted the interest of pulu authors. While this last group mostly consists of short texts – some of which started life as stand-alone texts not until collated by a collectanea (congshu 叢書)Footnote 10 editor – books on ‘all animals’ were often voluminous and provided an extensive selection from the whole spectrum of Chinese animal classes, i.e. from ‘beasts’ (shou 獸) and ‘birds’ (niao 鳥) to ‘fish’ (yu 魚) and ‘insects’ (chong 蟲).Footnote 11 The most noteworthy examples of this include: 1. Yinshi ji 蟫史集 (Collection from the History of a Bookworm; preface dated 1586), consisting of over 280 entries; 2. Rufan 蠕範 (Master Plan of the Wriggling and Winding; written in 1791, printed in 1844) with 420 entries; 3. Chong hui 蟲薈 (Animal Florilegia; preface dated 1898) comprising quotations on 1,148 species; and 4. Niaoshou chongyu bianlüe 鳥獸蟲魚編略 (Arranged Summary of All Animals; an undated manuscript kept at the Chinese Academy of Science library), filling 110 juan. With nine juan missing, over 1,300 entries are still extant today – 383 on birds, 364 on quadrupeds, 377 on fishes and 188 in the fragmented chapter on insects. The title of the Sichong beilan 四蟲備覽 (Complete View of All Four [Types of] Beasts; printed in 1848) suggests that it should also belong to this list. But the copy housed today in the old branch of the Chinese National Library in Beijing only covers one type of animals, namely birds.Footnote 12 Titles focusing on the group of ‘fish and sea animals’ show a great awareness of the wealth of unknown species that might live in the obscure world beneath the sea – a darkness that Li Diaoyuan 李調元 (1734–1803) wanted to illuminate in his Ranxi zhi 然犀志 (Account of the Burning Rhinoceros [Horn]; preface dated 1779), by using a rhinoceros horn which was believed could burn underwater. Guo Bocang 郭柏蒼 (1815–90) assumed that the creatures in the sea’s abyss were the same as those on land. His Haicuo baiyi lu 海錯百一錄 (All Sea Animals in One Register; printed in 1886) therefore portrays sea-tigers, sea-horses, sea-humans, etc., which are all similar in shape to those on land but are ‘bigger, or with a different name or different use’.Footnote 13 In addition, authors compiled extensive anthologies of quotations and stories about horses, cats and tigers, which each strove to provide an abundance of stories about one particular species, instead of multiple species or morphological varieties.Footnote 14 The stories provided different perspectives on these animals and depicted the different contexts in which they played a role. Ming and Qing titles about horses (see below) confirm the growing popularity of this type of comprehensive pulu.

Animals in Narration

Many pulu authors combed through earlier works in search of relevant passages and exemplary stories to depict animals and existing knowledge about them. Some provided scraps of information in a disjointed manner as separate and apparently unconnected paragraphs; others included topical chapters to orientate their readers. While the use of quotations enabled an author to include knowledge that had already been attested by other texts (thereby acquiring some legitimacy as well as devolving full responsibility for its contents), this approach meant that the animals were depicted as not being directly approachable.

This chapter now considers horses, cats and tigers as subjects which exemplify different kinds of ‘remote’ animals and illustrate the different strategies which pulu authors adopted to approach them, using assemblages of quotations and stories. It seems appropriate to call some of these works ‘animal leishu’ 類書, i.e. the Chinese term for an encyclopaedic collection of quotations grouped into ‘categories’ (lei), although traditional Chinese bibliographies rarely classified these texts in this way. The genre of pulu overall does not fit the class of leishu because its focus is too narrow and its subject matter too marginal.Footnote 15


The multi-valence and importance of horses becomes particularly evident by comparing pulu on horses dating from Ming and Qing times with the administrative and epidemic concerns outlined by Schäfer and Han (see Chapter 8) and the horse as icon of both orderly state and free steppe, in a multicultural, multilingual Qing, as described by Aricanli (see Chapter 10). The Song and pre-Song era monographs on horses listed in Table 7.1 are nearly all called Xiang ma jing 相馬經 (Classic on Evaluating Horses According to Appearance), with no author cited, making it difficult today to distinguish the works from each other.Footnote 16 The title suggests that these texts were guides to evaluating horses’ good health and character by examining their fur whorls, colouring, body proportions, etc. After the xiang 相 ‘evaluation’ genre disappeared or, perhaps more accurately, was integrated into the government’s medical and managerial concerns, works on horses classified as pulu became more akin to other pulu titles.Footnote 17

Two comprehensive pulu on horses were compiled in the late sixteenth to early seventeenth century. Both investigate horses from a philological-historical angle by collating quotations about individual horses or horse-related events through history. They treat horses as individuals from China’s historical past, i.e. animals not observable by the author himself, thus constituting one form of ‘remote’ animal. The first text, Guo Zizhang’s 郭子章 (1543–1618) Pinyisheng Ma ji 蠙衣生馬記 (Pinyisheng’s Notes on Horses), contains eighty-eight entries on ‘famous horses of kings and lords from antiquity [i.e. including mythological figures such as Fuxi 伏羲] to contemporary times [i.e. the Ming dynasty]’, and seventeen entries with ‘anecdotes and other writings about ordinary horses and horse terminology’.Footnote 18 Chen Jiru 陳繼儒 (1558–1639) combined Guo’s entries with the eighty-two entries from Li Chengxun’s 李承勛 (Ming) sequel to Guo’s work into one work for his late sixteenth-century collection Baoyantang miji 寶顏堂秘笈, omitting Li’s name.Footnote 19 Hu Shi’an 胡世安 (1593–1663) took a similar, but even more comprehensive, approach to horses in his Long sheng 龍乘 (History of the [Two Kinds of] Dragons [i.e. Swords and Horses]; preface dated 1624) in sixteen juan. Hu dedicated five juan (juan 2–6) to the ‘dragons of the scabbard’ (xia long 匣龍), i.e. swords, and ten juan (juan 7–16) to the ‘dragons of the shed’ (jiu long 廄龍) – as he called his section on horses. One of his chapters lists and explains the names and stories of individual historical horses (kao ming 攷名). Other chapters are concerned with ‘mimicry’ (xingsi 形似) and simulacra (such as horses made from clay that could run 1,000 miles per day after being brought to life by magic), provide rhyming couplets (li ju 儷句), or explanations of terminology grouped into categories (lei gu 類詁).Footnote 20 Three juan are dedicated to quotations from ‘ancient scriptures’ (dian ji 典籍).Footnote 21 Interestingly, Guo Zizhang and Hu Shi’an both compiled a matching piece of writing on swords, which Hu made into the first part of his work. Both authors approached horses and swords using similar methods and framed them as remembered objects or actors with individual names and relationships comparable to a historically known person or place.Footnote 22


The three pulu on cats were written close together, in 1798, 1799 and 1852, although their authors – one a woman – did not know about or quote from each other. Nevertheless, all three used the same methodology, collecting quotations from a wide range of works, arranging them into thematic chapters and citing their sources. However, they also differ in certain ways. Whereas the Mao sheng 貓乘 (History of the Cat) by Wang Chutong 王初桐 (c. 1730–1810) consists only of quotations, authoress Sun Sunyi’s 孫蓀意 (1783–1820) Xianchan xiaolu 銜蟬小錄 (Short Account of Cicada-in-the-Mouth [the Cat]) and Huang Han’s 黃漢 (d. 1853) Mao yuan 貓苑 (Cat Garden) go slightly further. Sun Sunyi occasionally includes her own short commentaries as footnotes, while Huang Han and his friends contribute additional stories and information relating to a fact (or factoid, i.e. an accepted ‘truth’ constructed by prestigious authors of the past) established by the quotation.Footnote 23 These are appended as indented paragraphs, starting with ‘Han comments’ (Han an 漢按) or ‘(so-and-so) says’ (yun 云 or yue 曰), and thus are clearly distinguishable and separate from what is presented as the ‘main’ text.Footnote 24 These comments open up another level of discourse, not only optically by indentation, but also by reflecting on the cat as a pet of which the author and his friends have personal knowledge. Thus, their remarks occasionally reflect on the ‘remote’ or ‘textual’ idea of a cat presented in the quotation. For example, Huang Han and his friends add some practical advice to a quotation concerning the time dependency of a cat’s eating habits. Whereas the quote draws some cosmological parallels, stressing that cats prefer different body parts of a rat at certain parts of the month (in the first part they only eat the head, in the second the stomach, and in the third the legs), the comment comes from an experienced cat owner, who advises against disturbing a cat while it is eating a rat on top of one’s clothes or seating mats. One should wait until the cat has cleaned everything up. Otherwise, a bloody mess would be left behind. But the same entry also adds some hearsay, such as, if you watch a cat when it is eating a rat, its teeth will grow soft so it will never be able to kill another rat.Footnote 25 Often information given in the comments does not differ substantially from the main text quotations. But as they are presented in the personal voice of Huang Han or one of his friends, the comments draw the textual distanced, ‘remote’, factoid-based accounts of cats into the authors’ present, real world.


A similar web of quotations is constructed in textual assemblages around tigers. This animal’s actual spatial ‘remoteness’ has resulted in an even stronger reliance on ‘hearsay stories’. There are three extant monographs on tigers which – like the texts about cats – were published relatively close to one another. In contrast though, the authors clearly refer to each other, even citing their predecessor’s work as the incentive for compiling their own tiger pulu. The sequence starts with the Hu yuan 虎苑 (Tiger Garden) by Wang Zhideng 王稚登 (1535–1612), which dates from 1553 (the year that Wang states he decided to edit a book on tigers in his preface). In that year Wang had the rare opportunity to interview some professional tiger hunters who came to his town to chase a man-eating tiger. Chen Jiru’s Hu hui 虎薈 (Tiger Florilegia) followed in 1594/5. Chen’s inspiration to compile this weighty volume of 363 entriesFootnote 26 in six juan came directly from Wang Zhideng. Wang had personally given Chen a copy of his Hu yuan when he was suffering from malaria. The terror and excitement he experienced reading stories about this impressive, mysterious and dangerous animal brought his fever-struck senses back to normal, so he decided to produce an even more comprehensive book on the subject. The third author is Zhao Biaozhao 趙彪詔 (Qing), who mentions both Wang’s and Chen’s books as predecessors to his Tan hu 談虎 (Talking about Tigers; preface dated 1716).Footnote 27 Whereas the Hu yuan and Tan hu structure their content in thematically organized chapters, Chen’s Hu hui provides his stories without any obvious order. Zhao’s Tan hu sorts his stories into three loosely framed chapters: ‘Controlling and Subduing a Tiger’ (Zhi hu 制虎), ‘Encounters with Tigers’ (Hu shi 虎事) and ‘Miscellaneous Notes’ (Za ji 雜記). Wang constructs and frames the tiger from a more varied set of perspectives. In the first of his fourteen chapters, called ‘Virtuous Government’ (‘De zheng’ 德政), the stories depict tigers terrorizing the region, to indicate governmental mismanagement. The animals leave the country – often by swimming across a river – as soon as a virtuous official takes on leadership. These accounts stress that resources spent on hunting and trapping are wasted because only good government can bring permanent relief from a tiger invasion. The second chapter shows that tigers are capable of acting with a ‘Sense of Piety’ (‘Xiao gan’ 孝感). When a tiger sinks its teeth into the leg of a boy collecting firewood, the boy’s sister convinces the tiger not to eat him, by screaming and pulling its tail, explaining that she and her aged mother rely on the boy to survive. In the fifth and sixth chapters, tigers appear as prominent actors in events which show either their ability to ‘Repay Kindness’ (‘Dai yi’ 戴義) – for example bringing a deer to a human who had removed a thorn from a tiger’s foot – or their instinct to ‘Punish Cruelty’ (‘Ji bao’ 殛暴), by killing a man who planned to leave his old mother behind to improve the chances for himself and his wife to survive on the little money he had. The chapters ‘Strangeness’ (‘Ling guai’ 靈怪), ‘Taming and Training’ (‘Huan rao’ 豢擾), ‘Seizing and Shooting’ (‘Bo she’ 搏射), ‘Assistance to the Spirits’ (‘Shen she’ 神攝), ‘Human Transformation’ (‘Ren hua’ 人化) and ‘Providing Metaphors’ (‘Pang yu’喻) all follow the same format, collating short anecdotes or tales to convey ideas about specific aspects of the tiger. Each story states the place and name of the main actors as well as, occasionally, the era of the event. The chapters ‘Strangeness’ and ‘Seizing and Shooting’ deviate from the others, containing more stories of a longer length. The latter chapter shifts the focus onto humans’ courage and great skills in weaponry that enables them to defeat tigers. The chapters ‘Auspicious Signs’ (‘Zhen fu’ 貞符), ‘Omens’ (‘Zhan hou’ 占) and ‘Fierceness’ (‘Wei meng’ 威猛) have a different tone. Their information is not embedded in a narrative but is provided as factual/factoid statements such as ‘Tigers reach the high age of 1000 years; after 500 years they turn white’,Footnote 28 and ‘Tigers do not eat children, as they are foolish and without fear and thus cannot be eaten; they also do not eat men while drunk, but have to wait until they are sober. What they wait for is not their soberness but their fear.’Footnote 29

The last chapter of the Hu hui, ‘Miscellaneous Notes’ (‘Za zhi’ 襍志), contains a mixture of stories and quotations from books such as the Shijing, Erya and Han Feizi 韓非子.Footnote 30 It refers to tigers as one of the ‘four spirit [animals]’ (si ling 四靈), being traditionally called the ‘righteous beast’ (ren shou 仁獸 or yi shou 義獸).

The texts focusing on tigers almost exclusively describe the animal in relation to their encounters with humans. Only a few passages provide facts or factoids, such as, ‘Tigers get drunk when eating dogs, because dogs are the wine of tigers.’Footnote 31 These writings about tigers are a typical example of pulu, which relied on stories and anecdotes to enable readers to extract and abstract factual knowledge by themselves – sometimes with the help of a chapter heading. In his study of tiger lore in the Hu hui and elsewhere, Charles E. Hammond concludes: ‘Chinese beliefs concerning the tiger, like Western attitudes toward the tiger, are shot through with contradictions … [which in sum] demonstrates the pervasiveness of faith in an impartial divine power.’Footnote 32 I would argue that these contradictions do more than just point to a common explanatory node beyond human perceptions. Texts such as the Hu hui present discrepancies in the form of a monograph, portraying tigers as multifaceted animals with many realities, instead of one single entity of ‘the tiger’. Multiple versions of tigers exist side by side and neither one invalidates any other. This function of pulu stands in contrast to Chinese encyclopaedias, in which the keyword ‘tiger’, first, is one among many other keywords and, second, the material listed lays out a spectrum of quotations in order to provide a repertoire for scholars to draw from.

In all of these examples knowledge about horses, cats and tigers was wrapped inside a narrative. Hayden White (following Barthes) characterizes narratives as being situated in time and place with a sequence of events having a beginning and an end, which therefore in contrast to discourses lack an ego presenting a subjective perspective. Without a narrator, ‘the events seem to tell themselves’.Footnote 33 Information or knowledge that is embedded into a narrative can thus move between different time periods and social strata more easily. Yet, different readers might focus on different aspects of the tales and make diverse connections to their own contemporary reality. From this perspective, narratives like those in the pulu examples above can be seen as unprocessed, ‘raw’ samples which allowed readers to extract information by different means, from diverse perspectives and for different aims throughout time.

Animals in Variation

The closer one looks, the more differences can be detected amongst things which appear similar at first sight. In the case of decorative garden flowers, difference, variation and nomenclature were features of great interest to Chinese authors. Many pulu about flowers such as chrysanthemums, orchids and peonies merely consist of lists, with names and descriptions of varieties sorted into categories according to their blossom colour, beauty or region. The animal species Chinese authors observed most closely were crickets, quails and goldfish. Apart from Jia Sidao’s Song period work on crickets (see above), all pulu on hobby animals date from the late Ming and Qing and remained popular during the Republican period. They focus on differentiating between varieties (zhong 種) and evaluating qualities or grades (pin 品). These works include chapters on practical issues such as feeding and treating illnesses, as well as duties and taboos in caring for these species, along with rhyming verses to aid in memorizing the essentials. In addition, the books on crickets and quails contain chapters explaining how to train the animals for fighting. Training, feeding and care established an intimacy and communication between a cricket and its master, who spurred on or calmed down the insect’s fighting spirit by tickling and stroking it using a rat’s whisker. Goldfish manuals claimed that the fish could recognize their master’s voice or knocking as signs to come and beg for food.Footnote 34 Books about crickets and quails provide guidance on how to evaluate their morphology in order to predict their fighting spirit and chances of winning tournaments. In the case of goldfish, breeding new and exceptional phenotypes was considered an end in itself, so the ‘evaluation’ (xiang) sections aimed to define which name a fish with certain attributes deserved (or, conversely, what features a goldfish with a specific name should have). Nevertheless, both types of hobby animals – those for fighting and those for beauty – were textualized using strategies that were similar to each other and to the pulu on decorative flowers, i.e. fixing the sometimes ephemeral existence of a variety by giving it a name and description. These thereby could be reused by others who might not have seen that actual variety. The names attached to these evaluations may have also determined prices on the market.

The word ‘variety’ used here appears in the Chinese texts as zhong, or pin, and does not imply any Linnean or proto-Linnean meaning. Aficionados of goldfish (or chrysanthemums) were sometimes keen to create new varieties through breeding and, at other times, they were trying to preserve a variety. They were well aware that varieties of chrysanthemum and exceptional shapes in goldfish were products of human interventionFootnote 35 and that, if people stopped controlling breeding, their descendants might deteriorate and revert to an unornate, ‘original’ shape.


In Chinese literature, the history of goldfish is usually traced back to a Song poem by Su Shunqin 蘇舜欽 (1009–49). In the poem, the author describes looking for ‘gold carp’ (jinji 金鯽) – probably a mutation of the crucian carp – from a bridge near the Liuhe 六和 pagoda in Hangzhou. As early as 1548, the brush note collection Qixiu leigao 七修類稿 (Draft Arranged in Seven Categories) reported the enthusiasm with which people in Hangzhou reared ‘fire fish’ (huoyu 火魚), ‘competing for colour [variation] and striving for profit’.Footnote 36 What had once been an ordinary type of crucian carp had, over time, been developed by goldfish aficionados into a dazzling array of varieties in shape and colour. The transfer from harsh natural conditions to well-tended fish bowls in devoted hobbyists’ luxurious homes unlocked the species’ potential for diversification.

When the author of the Jinyu tupu 金魚圖譜 (Illustrated Treatise on Goldfish), the self-styled ‘peasant from Juqu 句曲 Mountain (in Jiangsu)’ (n.d.) compiled his work, he claimed to be the first person to produce a monograph on goldfish. He was only aware of one previous work, containing fifty-four illustrations of goldfish varieties by a Mr Xue 薛, which he felt lacked sufficient explanations. Thus, the author decided to scour Li Shizhen’s 李時珍 (1518–93) Bencao gangmu 本草綱目 (Compendium of Materia Medica), the Hua jing 花鏡 (Flower Mirror)Footnote 37 and other texts to gather the relevant information. The multicoloured print edition of the Jinyu tupu dated 1848 has two parts, each of which has a different decorative frame around the text.Footnote 38 The first part contains eleven short chapters on different practical aspects; the second part appends Mr Xue’s fifty-four names and illustrations. The first and last chapters, on ‘Origin’ (‘Yuan shi’ 原始) and ‘Usage’ (‘Zheng yong’ 徵用) discuss the history of the goldfish and its uses, namely as an emetic against opium poisoning and in fire prevention (by providing an additional use for firewater basins). The author dedicates the other chapters to detailed hands-on advice and explanations, such as how to control the ‘Biting of the Seed’ (‘Yao zi’ 咬子), i.e. the impregnation of females, how to identify and treat the fish ‘Sprouts’ (‘Miao’ 苗), how to feed and medicate fish, and how to evaluate their potential to develop a good colour and shape. He writes from the perspective of a professional.

Some 150 years before the Jinyu tupu, Jiang Zaiyong 蔣在雍 (fl. Qing period) had already written his Zhuyu pu 朱魚譜 (Treatise on Cinnabar Fish; preface dated 1699). Jiang names and describes fifty-six varieties or shapes of goldfish, some of which have identical names to those described in the Jinyu putu. Jiang also added some information about care and general guidelines on evaluation, but the main part of his manuscript was concerned with their varieties.Footnote 39 The first variety he describes is the ‘Buddha-head pearl’ (foding zhu 佛頂珠), which should be ‘completely white … without a single dot of red or other blend … where on top of the brain one red spot sticks out, round as a pearl, protruding and thick’. Where this ‘pearl’ was slanted or too small, the fish would not ‘meet the standard’ (bu ru ge 不入格). If there was a big, round, non-protruding spot instead of a ‘pearl’, the fish was called a ‘Buddha-head red’ (foding hong 佛頂紅), while those with a large but elongated spot were called ‘Number One Examinee’s red’ (zhuangyuan hong 狀元紅). Neither of these varieties was as valuable as a ‘Buddha-head pearl’ (for the original text, see Figure 7.2).

The illustration of a ‘Buddha-head pearl’ from Mr Xue’s earlier work that was reproduced in the second part of the Jinyu tupu nicely matches the description in Jiang’s Zhuyu pu, showing a goldfish with a prominent round ‘pearl’ on its head (see Figure 7.1).

Figure 7.1 ‘Buddha-head pearl’.

(Jinyu tupu, 626)

Figure 7.2 Description of a ‘Buddha-head pearl’.

(Zhuyu pu, 588)

Like the pulu on goldfish, those concerned with crickets and quails also focus on evaluating and precisely naming a ‘variety’ or ‘grade’, alongside practical information and advice. But there is a substantial difference between these books. Crickets and quails had to be caught from the wild, because those bred and raised by humans lacked fighting spirit. While their morphology when caught made them eligible for a certain name that indicated potential for being a victorious fighter, it was only by training and care that these insects or birds were transformed into an object of connoisseurship and gambling. In goldfish – as with chrysanthemums or orchids – the morphology was an aim in itself and was achieved by extensive care within the scholar’s own home.

In contrast to the many goldfish varieties or cricket qualities named in pulu writings, from the perspective of the pharmacopoeia Bencao gangmu they were each just ‘one’ form of materia medica that was of interest for their therapeutic value, for instance using goldfish as an emetic. Li Shizhen also mentions the history of goldfish and notes that there are also white, black and spotted types as well as those of a standard golden colour, observing that this variety in colouring does not give them any different therapeutic attributes. The ‘materia medica goldfish’ is framed much more loosely. Only from the close-up view of the collector-specialist and pulu author could fifty or more varieties be identified and portrayed in writing. Thus, these literary and praxeological contexts constructed vastly divergent versions of goldfish.


This chapter has explored pulu writing on animals, noting the division between ‘remote’ animals, which were constructed from other texts and hearsay, and ‘close’ animals, where a more direct translation took place – from actual observations of animals into textual portrayals of them. It has shown how both types of animals were created differently in writing and became established as different types of ‘text-animals’. Whereas the first was approached and represented mainly through narratives and from a safe, philological distance, scholars established a personal approach to, and intimate engagement with, the latter. As a consequence, each produced different kinds of enumerations or varieties. ‘Varieties’ such as ‘tigers showing piety’, ‘cruel and castigatory tigers’ and ‘tigers that are hexed humans’ were depicted in stories about human–tiger encounters; while ‘varieties’ of goldfish were designated by exploring and describing their diverse morphologies. Moreover, the latter texts paired their enumeration of varieties with hands-on knowledge gained from their authors’ own practical experience. This contrasts strongly with the accounts of tigers crossing paths with humans, which arose as lucky, ominous or blood-curdling stories but rarely provided any insights into real tigers’ habits.

In pulu about horses, the animals constructed from historical narratives had a slightly different twist from the texts on tigers. Hu’s Long sheng used excerpts from a variety of literature in order to, on the one hand, examine horses from several different perspectives (etymology, poetry, etc.) whilst, on the other hand, also trying to reconstruct individual horses, their ‘names’ and their roles in history – just as he did with swords.Footnote 40 Similar to the works on tigers, identifying and profiling an animal were achieved from textual narratives, not from its morphological features, habitat or biological needs. Yet, unlike horses, tigers were not individualized in these narratives but presented generically as ‘a’ tiger, as if each creature represented their whole species. Xiang Yu 項羽 (232–203 BCE) had ‘one’ individual horse named Zhui 騅. ‘Names’ of goldfish and quails had a different function: a goldfish called a ‘Buddha-head pearl’ or a quail named ‘Phoenix from the cinnabar mountain’ (danshan feng 丹山鳳) does not represent one individual creature, but many individuals of a similar type.

In compilations about animals that do not address one species but cover larger groups such as birds, sea animals or all animals, the entry for each species is generally limited to those features considered to be basic characteristics. In some works, only these ‘features’ were deemed important, not the ‘whole’ animal. This approach of bringing together characteristics from different species into contrasting or sequencing couplets was already evident in the Qin jing: ‘Birds that eat meat have pointed beaks; birds that eat grain have blunt beaks.’Footnote 41 Yuan Da 袁達 (juren 1513) made full use of this method in his Qinchong shu 禽蟲述 (Elaborations about Animals), composed around the mid-sixteenth century. As Yuan Da explains in his preface, his aim was not to provide a complete account of all animal species, because there are too many of them. He does not express his own views either, but merely ‘elaborates’ (shu 述) on those of others. He asserts that his contribution as an author is to ‘select them according to their categories to [allow] comparison’.Footnote 42 The section relating to animal eyes exemplifies the style of Yuan Da’s text:

Those [animals] born from the womb close their eyelid from the top, those born from eggs from below, those born from moist have no eyelid and those born by transformation have no eye-socket; the [fantastic] yu 蜮 has no eyes at all, the python’s (mang 蟒) eyes are round, those of the quail are turning … ; the heron copulates with its eyes, snakes listen with their eyes … ; dragons do not see stones, fish do not see water.Footnote 43

Here, Yuan Da shows how one common bodily feature exists differently in diverse species. With the focus moved from any particular species to one peculiar facet of different species, the animals are dis-assembled and reduced to certain properties and characteristics.

In the preface to Li Su’s Jian wu (Seeing Things), Lü Kun 呂坤 (1536–1618) wrote: ‘Animals are the “small selves” of humans’ (wu xiao wo ye 物小我也).Footnote 44 He explained that humans and animals had originally been connected, but human self-awareness had built a wall between them which broke the direct bond between human and animal. He claimed that they still touch at their ‘hub’ (ji 機), which is why an emotional echo can be triggered by pigeons in flight and why common characteristics can be detected in both humans and animals.

‘Writing the animal’ was a challenge. In his analysis of Greek authors writing on animals, von Staden identified three main difficulties. One of these was the choice and management of literary form, that is, deciding whether a scientifically distanced text or an expressive narrative seems more suitable.Footnote 45 Von Staden took Aristotle and Pliny the Elder as icons of these two contrasting approaches. Whereas Aristotle chose to present his texts with an ‘authorial distance combined with systematicity’, Pliny the Elder’s monumental Natural History is a highly personalized construction, in which the author guarantees his originality and scholarly soundness. Pliny quantified the scholarly achievement of this work by noting the number of books read and facts collected,Footnote 46 something that appears very similar to the Chinese pulu on horses, cats and tigers discussed above.Footnote 47

Animals were reconstructed from texts and hearsay and then regenerated into a new text, rejuvenating their existence as text-animals. Likewise, the works on goldfish still portrayed text-animals, despite their authors working from personal, practical interactions with them. In all cases, textual versions of the animals were constructed: either as a fixed form with a set name, such as ‘Buddha-head pearl’ with specific requisite features, or as a blurred, complex shape such as ‘the’ tigers which had – among other facets – a ‘sense of filial piety’ and were defined by a plethora of quotations and agendas.


1 Asdal (Reference Asdal2014), 311.

2 This idea of a praxeological relation between animal and text/author, both as actors and as acted upon in a web of mutual relations, draws on the concepts of ‘material semiotics’ outlined by Law and Mol (Reference Law, Mol, Knappett and Malafouris2008) and Law (Reference Law and Turner2009). See also Schlünder (Reference Schlünder2012) on the use of the term ‘praxeology’.

3 In Chapter 11 of this volume Zheng Xinxian discusses an outstanding example of this – the discrepancy between the canonical ‘Monthly Ordinances’ and the practical knowledge of the Qianlong emperor.

4 Two other genres which are often scoured for examples of Chinese natural studies are works in the tradition of the early second-century glossary Erya 爾雅 (e.g. the Piya 埤雅 by Song dynasty Lu Dian 陸佃), or works such as Lu Ji’s 陸璣 third-century Commentaries about Flora and Fauna in the Mao version of the Book of Songs (Mao shi caomu niaoshou chongyu shu 毛詩草木鳥獸蟲魚疏). These have influenced the tradition of pulu writing to some degree, but had different agendas.

5 Siku quanshu zongmu tiyao, 115.988.

7 The Classic on Raising Fish is said to have been mentioned in the lost sixth-century bibliography Qi lu 七錄 (Seven Registers). The bibliographic treatise of the Jiu Tangshu 舊唐書 (Old History of the Tang) lists it under ‘Agricultural writings’; see Jiu Tangshu, 47.2035. The Classic on Birds, attributed to Shi Kuang 師曠 (sixth century BCE), a music master of Zhou era kingdom Jin 晉, is shown in two halves in Table 7.1 since there does not appear to be any version of the text which omits Zhang Hua’s 張華 (232–300 CE) third-century CE commentary. The transmitted versions of these early titles mostly rely on the Song collection Baichuan xuehai 百川學海 compiled by Zuo Gui 左圭 (fl. thirteenth century) in 1273. The Xiang niu jing 相牛經 (Classic on Ox Physiognomy), also of supposedly Zhou ancestry, was listed as lost in the Suishu bibliography (fifth century CE), but reappeared in the Baichuan xuehai collection. Most of the titles in the pre-Song columns in Table 7.1 are xiang shu 相書, ‘books on evaluations according to appearance’ or books on ‘raising and care’ (yang 養), which both focus on horses and oxen. These titles can be called pulu ‘avant la lettre’ but, since Song times, they only existed as authorless entries in bibliographies. On xiang shu, see e.g. Li Ling (Reference Li2001), 84–7.

8 Pulu are mentioned in two Song dynasty bibliographies – the comprehensive private library catalogues Zhizhai shulu jieti 直齋書錄解題 (Commentated Book List of the Zhizhai studio) and You Mao’s Suichutang shumu 遂初堂書目 (Library Catalogue of the Suichu Hall), the ‘originator’ of the bibliographical category ‘pulu’. The large number of pulu in the bibliography of the Dynastic History of the Song (Songshu 宋書) derives from the compilers indiscriminately copying all titles from diverse official palace catalogues without actually seeing the books, identifying duplicates or more thoroughly searching for authors or dates. See Li Ruiliang (Reference Li1993), 174; Siebert (Reference Siebert, Raimund and Siebert2006), 60–1.

9 No bibliography before the Ming lists the Classic on Crickets; the earliest imprint bearing the title Chongkan dingzheng Qiuchong pu 重刊訂正秋蟲譜 (New and Corrected Imprint of the Book on the ‘Autumn Insect’) dates from 1546. See Xuxiu siku quanshu 續修四庫全書, vol. 1120, 239–56. With the inclusion of Jia’s cricket book in the Yimen guangdu 夷門廣牘 collection, edited by Zhou Lüjing 周履靖 (1549–1640) at the turn of the seventeenth century, the title became more widely known. History mainly remembers Jia Sidao as one of the Song dynasty’s treacherous officials, who only cared about his own pleasures and entertainment, including his personal favourite, cricket fights. Ascribing this title to Jia might be a case of yituo 依托 ‘assignment in support’, rather than weituo 偽托 ‘false assignment’ which, according to Li Ling’s definition, means writing and producing a compilation in the tradition of, and in support of, the assigned historic author – not a form of deception. See Li Ling (Reference Li2001), 28–30.

10 Collections such as the Shuofu 說郛 and Zhaodai congshu 昭代叢書 were particularly adept at turning essays into monographs. For instance, the Buji 哺記 (Notes on Hatching) by Huang Baijia 黃百家 (1643–1709) was originally an essay in Huang’s anthology Xueji chugao 學箕初稿 (First Drafts from the Winnowing Basket of Learning). See the facsimile in Sibu congkan, ji bu, vol. 341, 263–4. But it later appeared as a separate title in Zhaodai congshu, 3253. On the Buji, see Siebert (Reference Siebert, Raimund and Siebert2006).

11 See Siebert (Reference Siebert2012) on the different ways of grouping animals into morphological classes and hierarchies, and according to topics of scholarly interest.

12 This title, written by Ni Tingying 倪廷瑛 (n.d.), is listed in the bibliographies of the Qingshi gao 清史稿, 205 (‘Yiwen zhi’ 藝文誌) and the Baqianjuan lou shumu 八千卷樓書目, 12.12a.

13 Haicuo bai yi lu, preface, 1a.

14 The problematique of what counts as a species and what as only a variation was not a central concern for historical Chinese scholars. A ‘name’ was the anchor for identification to which in turn a number of descriptions of morphological features and narrative characterizations were attached. Of course, materia medica needed to be correctly identified and no one wanted to pay a high price for fake sable. But, in the textualized world of animals in historical China under study here, different levels of individualizations – i.e. what today’s zoology would consider a species, variety or individuum – appear next to each other. Some aspects of this are dealt with in their relation to the classification and grouping of animals in Siebert (Reference Siebert2012); a look into the ‘achievements’ and uses of knowledge about heredity is provided by, for example, Wang Zichun (1989).

15 There are two examples of pulu on ‘all animals’ that were occasionally classified by bibliographies as leishu or ‘encyclopaedias’. The Siku quanshu grouped the Yinshi ji by Mu Xiwen 穆希文 (preface 1586) into its leishu category (Siku quanshu zongmu tiyao, 138.1174); the Baqianjuan lou shumu listed the Jian wu 見物 (Seeing Things; preface 1581) by Li Su 李蘇 (Ming) under leishu (13.24a).

16 Whereas most Xiang ma jing had only one juan, Jiu Tangshu, 47.2035 and Xin Tangshu, 59.1538 record a lost Xiang ma jing with sixty juan that was attributed to Zhuge Ying 諸葛潁 (fl. late sixth century).

17 The Mazheng ji 馬政紀 (Records on Horse Administration) in twelve juan compiled by Yang Shiqiao 楊時喬 (d. 1609), for instance, appears in the bibliographical class of zhengshu 政書 ‘books on administration’ (see Siku quanshu zongmu tiyao, 82.711).

18 Ming ma ji xu 名馬記序, preface (n.d.) by Guo Zizhang, in Xuxiu siku quanshu, vol. 1119, 319.

19 The numbers given here are those mentioned in Guo’s preface and Li Chengxun’s epilogue (Ming ma ji 名馬記, in Xuxiu siku quanshu, vol. 1119, 319–24, 337–8). The Baoyantang miji edition adjusted the numbers in Guo’s preface to 160 for the first and forty for the second part, leaving thirteen entries that seem to have been added by Chen Jiru himself.

20 Long sheng, 8.1a–8a, 9.10a–b, 9.17a–18b.

21 Long sheng, 11–13.

22 A famous example is Zhui 騅, the warhorse whose owner, Xiang Yu 項羽 (232–202 BCE), sent him home before killing himself after being defeated by the Han army. The iconic pulu on swords, the Gujin daojian lu 古今刀劍錄 (Register of Swords Old and New), was compiled by Tao Hongjing 陶宏景 (456–536 CE), and also lists only historical swords, which Tao himself had never seen.

23 Huang Han gives a full list of all his aides and contributors in the ‘Instructions’ (Fan li 凡例) to the book. Mao yuan, 2 (‘Fanli’).

24 This structure of a ‘main text with commentary’ is also found in the Rufan (see above) and the Yiyu tuzan 異魚圖贊 (Illustrated Eulogies on Remarkable Fish; preface 1544) by Yang Shen 楊慎 (1488–1559) and its sequels by Hu Shi’an, i.e. Yiyu tuzan bu 異魚圖贊補, ~runji 閏集 and ~jian 箋 dated 1618 and 1630. Here ‘main text’ and ‘commentary’ were actually written by the same authors.

25 Mao yuan, 1.24b.

26 This is the number counted by Huang Dahong and Zhang Tianli (Reference Huang and Zhang1999).

27 While the most recent story in the book dates from 1699, the preface mentions the year 1716. See Tan hu, 1149.3. This text, together with a parallel text on snakes written by Zhao – the Shuo she 說蛇 (On Snakes) – was included in the Zhaodai congshu in 1783. See Zhaodai congshu, 1147–55. Zhao’s book has some overlaps with the Hu hui; for instance, both include the story about a man from Jingzhou 荆州 who was captured and transformed into a tiger by the ‘rude ghost’ of someone eaten by a tiger (chang gui 倀鬼). See Tan hu, 1151.3; Hu hui, 1.2b–3a.

28 Hu yuan, 1.5a (‘Zhen fu’).

29 Hu yuan, 1.9b (‘Wei meng’).

30 A certain overlap can be detected between the stories in the three tiger books and the eighty stories and historical anecdotes that constitute juan 82 of the Taiping guangji 太平廣記 (Extensive Records of the Taiping Era), an imperially sponsored encyclopaedic project from the tenth century. But, as the overlap is not exact and the sequence of the entries differs, it is likely that the pulu authors did not simply copy their information from the Taiping guangji but drew directly on the original sources and stories they claim were personal hearsay.

31 Hu yuan, 1.9b (‘Wei meng’).

32 Hammond (Reference Hammond1991), (Reference Hammond1996); quoted in Hammond (Reference Hammond1996), 211.

33 White (Reference White1980), 5–7.

34 See Jinyu tupu, 8a (‘Shi xing’ 識性).

35 Centuries of breeding new varieties made goldfish the ideal experimental organism for China’s emerging genetic research in the 1920s. Chen Zhen 陳楨 (1894–1957), one of the main figures, included historical texts and accounts of visits to markets and professional breeders in his investigations of the genetic history and current state of goldfish variances. See Jiang Lijing (Reference Lijing2016).

36 See, for example, the Jinyu tupu, 1b.

37 Some passages from the Hua jing section on goldfish are found verbatim in the Jinyu tupu. A significant difference between these books is that the latter’s author, Chen Haozi 陳昊子 (fl. late seventeenth century), depicts goldfish as just one species among many other decorative garden animals. See Hua jing, 6.17b–19a, in Gugong zhenben congkan, vol. 473.

38 While the first part has a floral decorative frame around the text (carved with the ascription ‘drawn by Yangzhi’ (Yangzhi xie 仰止寫), the list of names and images is bordered by a more maritime motif. Xuxiu siku quanshu, vol. 1120, 601–34.

39 Xuxiu siku quanshu, vol. 1120, 587–600. Jiang’s preface to this manuscript is dated 1699. The work appears never to have been printed.

40 The Long sheng does intersperse its lists of names with terms that designate certain types of horses and labels for particularly capable horses, but it mostly consists of the names of individual horses. This mixture also alludes to the idea that the most refined variant of a species is an exemplary individual. This blurred line between variety and individual is especially striking in the famous Song dynasty pulu on lychee, in which fruits from individual trees growing in a specific courtyard, such as ‘Chen’s purple’ (Chen zi 陳紫), and varieties of litchi, such as ‘tiger-fur’ (hupi 虎皮), are listed next to each other. See Cai Xiang 蔡襄, Lizhi pu 荔枝譜 (Book on Lychees; preface 1059), in Siku quanshu, vol. 844, 627–9.

41 Qin jing, 17a–b.

42 Qinchong shu, preface, 1a, in Xuxiu siku quanshu, vol. 1120.

43 Qinchong shu, 7a–b.

44 The title of Li Su’s book echoes Shao Yong’s ‘Observing Things’ (Guan wu 觀物), which is also mentioned in Lü Kun’s preface. The play on the word wo 我 would remind educated Chinese readers of Shao Yong’s request to observe things not from the perspective of the ‘self’ (bu yi wo guan wu 不以我觀物) but from the viewpoint of being a thing oneself (yi wu guan wu 以物觀物), or seeing them as being another ‘self’. This approach aims to copy the ‘reverse observation’ (fan guan 反觀) performed by the saints. See Huangji jingshi shu, 12.18a.

45 von Staden (Reference von Staden and Asper2013), 111–13, 120.

46 von Staden (Reference von Staden and Asper2013), 130–1.

47 The other two obstacles von Staden (Reference von Staden and Asper2013) identified were, firstly, an epistemological one, because human and non-human animals seem to have incommensurable cognitive structures, and, secondly, the hurdle of language, because animals cannot understand or respond to an author’s interrogation and human languages often lack terms to describe animals adequately.

Figure 0

Table 7.1 Pulu about animals through time

Figure 1

Figure 7.1 ‘Buddha-head pearl’.

(Jinyu tupu, 626)
Figure 2

Figure 7.2 Description of a ‘Buddha-head pearl’.

(Zhuyu pu, 588)
You have Access Open access

Save book to Kindle

To save this book to your Kindle, first ensure is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle.

Note you can select to save to either the or variations. ‘’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service.

Available formats

Save book to Dropbox

To save content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about saving content to Dropbox.

Available formats

Save book to Google Drive

To save content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about saving content to Google Drive.

Available formats