Introduction

Chinese medicine today is frequently said to rest on the foundation of a group of texts known as the Yellow Emperor’s Inner Classic (Huangdi neijing 黃帝內經, hereafter the Inner Classic, Neijing 內經). This point of view is found in medical texts as far back as the Song (960–1279) and is even cited in modern scholarly discussions of Chinese medicine as well. But is it accurate?

The Inner Classic is the oldest text in the received tradition of East Asian medicine and has been cited since the Song to explain and justify medical practices. There are, however, many ways in which a text can be authoritative or canonical, and although we now know a great deal about the formation of the Inner Classic, we still know very little about how it acquired its current status at the centre of the Chinese medical canon.

The very concept of canonicity is somewhat problematic. There is no clear Chinese equivalent to the term. The notion of a classic text (jing 經) comes closest; however, throughout China’s history, a variety of texts and groups of texts have been selected as authoritative in various fields, and many of these ‘canonical’ texts were never given the label ‘classic’ (an excellent survey of these is found in De Weerdt 1999: 91–4). Even in English, the term ‘canon’ is far from precise. It can refer to closed canons, such as the Bible and open canons, such as ‘the canon’ of English literature. No single model of textual authority can be applied to all of these cases. However, rather than lament the imprecision of this term, I wish to make its vagueness a virtue. A more relaxed analytical gaze may enable us to better recognise the various types of textual authority the Inner Classic has possessed throughout its long history. Therefore, in this chapter, I do not take the meaning of canonicity as a given but rather as a topic for investigation.

The earliest mention of a Yellow Emperor’s Inner Classic is found in the bibliographic section (Yiwen zhi 藝文志) of the Hanshu 漢書 (History of the Han) 6.30, alongside a Yellow Emperor’s Outer Classic 黃帝外經 (Huangdi waijing) and Inner and Outer Classics by the legendary physician Bian Que (扁鵲, sixth to fifth century BCE) and a Mr White (Baishi 白氏), who is otherwise unknown (Hanshu, 30.1776). None of the other texts in this list is still extant. Surviving literature mentions six titles associated with the Yellow Emperor’s Inner Classic (Table 7.1). ¹ A great deal of ink has been spilled arguing over which of these was the original text mentioned in the Hanshu. A combination of archaeological discoveries and careful investigation of extant texts, however, has led scholars to reject this question entirely and radically rethink our understanding of this group of texts.

The formation of the Inner Classic corpus

The various texts titled Yellow Emperor’s Inner Classic share not only a title, but many passages as well. These shared passages vary little from one Inner Classic text to another and are generally quite short, in modern terms no more than a few paragraphs in length. Although their wording varies little from one text to another, they are located in radically different positions within each text – different chapters, different positions within a chapter, etc. (Keegan 1988: 64–6). The explanation of these passages began to become clear after the archaeological discovery of similar passages in excavated texts. In 1973, archaeologists working at a site called Mawangdui (馬王堆) in Hunan province discovered a collection of texts buried along with the occupant of a tomb (Chapter 3 in this volume). Several medical texts were included among them. Since this tomb was sealed in 168 BCE, these texts provide an unparalleled look into the medical world of the early Han. Four of the medical texts deal specifically with the channels (jing 經) used in acupuncture and moxibustion, and some of their passages are clearly parallel to parts of the Numinous Pivot (Lingshu 靈樞) (Harper 1997).

In his seminal dissertation, David Keegan analysed these parallels and demonstrated that our understanding of the Inner Classic was hampered by the anachronistic imposition onto it of our concept of a book. Keegan argues that in its earliest stages, the Inner Classic is best thought of as a large collection of very small primary texts – the shared passages noted above – which were rearranged in various orders to compose the chapters and books that we know as the Yellow Emperor’s Inner Classic. The Inner Classic is therefore not a single book, and the various extant books bearing that title are actually compilations produced by selection and rearrangement of the primary texts. The primary texts were thus conserved with few changes across all compilations, but how they were put together varied greatly. The question of which extant compilation is the ‘original’ Yellow Emperor’s Inner Classic is therefore moot. None of them is; there never was any such creature (Keegan 1988: 110–13).

Having unravelled the structure of the Inner Classic, Keegan went further. Drawing on Nathan Sivin’s discussion of the transmission of medical learning in the Han and medieval periods, he posited a social mechanism by which the primary texts were compiled in such different orders. Physicians learned medicine by apprenticeship to a master. Initially, learning consisted of observing the master, receiving oral instructions, and being guided in the memorisation of texts. When the master decided the student was ready, he would transmit a text to the student. The student would copy it and the master would impart an oral explanation. The texts thus transmitted were not necessarily whole books, but rather short texts like the primary texts found in the Inner Classic. Students copied texts in the order in which they received them and might receive the same text in slightly different forms from different teachers. The resulting compilation of primary texts was therefore unique to that student. The various Inner Classic compilations we have today were produced by editing this sort of ad hoc compilation, and the order of primary texts in each compilation therefore differs (Keegan 1988: 219–47; Sivin 1995a).

Keegan’s analysis also enabled a more accurate dating of the Inner Classic corpus by making it clear that the individual primary texts were composed at different times and predate the earliest compilation of an Inner Classic text. Comparison of the Mawangdui channel texts with parallel primary texts in the Numinous Pivot reveals that the Numinous Pivot texts are substantially more developed (Harper 1997: 86–90). Given that the Mawangdui texts cannot date from later than 168 BCE and that Records of the Historian (Shiji 史記) – the earliest portions of which were completed c. 100 BCE – makes no mention of the Inner Classic, we can safely take 100 BCE as the earliest possible date for its formation. If we take the date of the Hanshu as its terminus ante quem – the Numinous Pivot’s passages must date from between 100 BCE and 92 CE. This evidence has led most scholars to conclude that the primary texts of the Inner Classic corpus must have reached something approaching their current state in the first century BCE, though individual texts may have earlier or later dates (Unschuld 2003: 1–3).

The extant compilations of the Inner Classic differ considerably in their structure and textual history. These differences have been discussed at length by Nathan Sivin (1993) and Paul Unschuld (2003) and will not be repeated here except as they bear on the question of the Inner Classic corpus’s textual authority.

The Han dynasty (202 BCE–220 CE)

The relative importance of the Inner Classic before the late Han is difficult to reconstruct. We can gain some sense of its relationship with the broader medical context, however, by examining the list of medical books possessed by the Granary Master (Canggong 倉公) Chunyu Yi 淳于意 (fl. mid-second century BCE), as recorded in his biography in Records of the Historian. Although his biography does not mention the Inner Classic, eight of the ten books he received from his teachers are also mentioned in extant Inner Classic compilations. The Granary Master presents his possession of these texts as proof of his medical skill and, in the case records included within the biography, is depicted as relying on these texts to justify his treatments.

We must, however, use this source with care. Miranda Brown has pointed out that this biography, along with most of the other sources used in reconstructing early medical history in China, was written by non-physicians for purposes unrelated to medical practice. Furthermore, the sources used in constructing this biography have been shown to be heterogeneous, some of them clearly not the work of a historical Chunyu Yi. Brown argues that the practice of justifying medical treatment by textual reference is not seen elsewhere in the early medical literature. Such justification was, however, part of the record-keeping practices of government officials. Recording and treating illness were among the duties of officials, and the case records in this biography were most likely derived from such records. This makes it impossible to say with certainty that the use of textual justification was common among doctors in this period (Brown 2015: 63–86). Nevertheless, the fact that medical texts are mentioned at all and that many of the same texts are also referred to in the Inner Classic does support Keegan’s thesis that there was a group of medical texts circulating in the early Han that were seen – by some at least – as authoritative.

In addition to revealing their provenance, the use of a governmental model in the Granary Master’s case records may also tell us something about the kind of authority these medical texts – and the Inner Classic – claimed. The presence of terms and metaphors related to government has long been recognised in both the Inner Classic and medical literature more broadly. By the end of the Warring States, there was widespread acceptance of a homology between the cosmos at large, the state, and the human body (Sivin 1995b). It is possible that medical practitioners adopted a governmental style of justification as part of their effort to bolster their authority. Just as officials could cite law codes in deciding legal cases, doctors could cite medical texts in deciding on diagnosis and treatment. The choice of the Yellow Emperor as an interlocutor in the Inner Classic may also reflect this governmental sensibility. The Yellow Emperor was believed to have taught humans the art of government in the distant past and is considered by some scholars (Yates 1997: 10–16) to have been a key figure in the Huang-Lao (黃老) movement, which was purportedly influential in the early Han and made explicit links between ruling the state and caring for the body. Yamada Keiji first put forward the ‘working hypothesis’ that this choice might reflect a connection between the Inner Classic and the Huang-Lao movement (Yamada Keiji 1979: 89). Subsequent research has tended to support this hypothesis and has shown that the Inner Classic relies heavily on a cluster of concepts – the Way (Dao 道), compliance (shun 順), and rebellion (ni 逆) – that was fundamental to much of Huang-Lao thought (Peerenboom 1993: 51–3, 64–6; Boyanton 2006). In medical texts like the Inner Classic, shun and ni can refer to compliance with or variance from the natural pattern of things, or to going with or against the normal flow of bodily fluids – such as qi along the channels. In Huang-Lao texts, the same terms reflect positions in relation to authority (i.e. of the state).

By the late Han, we have more evidence for the perceived authority of the Inner Classic. Its mention in the Hanshu bibliography attests to its significance, but the mention of other medical texts side by side with it indicates that it was not the only esteemed medical classic. Likewise, although the titles Inquiry into the Fundamental (Suwen 素問) and Nine Fascicles (Jiujuan 九卷) appear in the preface to the Treatise on Cold Damage ( Shanghan lun 傷寒論, c. 206 CE), they are accompanied by a number of other texts (Shanghan lun, preface.305). ² In the late Han, the Inner Classic appears to have been one of a large number of important medical texts, highly valued, but lacking its later pre-eminence.

In addition to considering the bibliographic evidence and the intellectual context within which the Inner Classic made its earliest claims to canonical status, it is also important to bear in mind its physical aspect as a book. During the Han, texts were generally written on bamboo slips that were then tied together and rolled up. These bamboo slip ‘books’ were difficult to produce, very heavy, and prone to falling apart if the ties holding them together were not maintained (Nylan 2000: 244). Given the technical limitations of book production, it is unlikely that physicians could produce and carry about large texts like the extant compilations of the Inner Classic, and the production of standardised texts would have been all but impossible. The type of authority the Inner Classic could claim was limited by the type of book into which it could be formed.

The period of division, Sui, and Tang (220–907)

We have abundant evidence for the Inner Classic’s importance throughout the period of division following the fall of the Han and up through the Tang. It is clear from the bibliographic literature and the surviving compilations that it was not only circulating, but also circulating in several compilations and various editions of these compilations.

The earliest surviving compilation of the Inner Classic corpus is the Systematic Classic of Acumoxa (Zhenjiu jiayi jing 針灸甲乙經), or to use its full title, the Yellow Emperor’s Systematic Classic of Acumoxa in Three Parts ( Huangdi sanbu zhenjiu jiayi jing 黃帝三部針灸甲乙經). Compiled by Huangfu Mi 皇甫謐 (215–282) between 256 and his death, it is of particular historical importance. Huangfu Mi became an imperial physician (taiyi 太醫) under the Jin (265–420) but was already an accomplished and respected scholar and author in other fields. According to his preface, he became interested in medicine after suffering an illness that was treated unsuccessfully (Huangdi sanbu Zhenjiu jiayijing, preface, 16–21). ³ The stature of the Inner Classic at this time is to some degree attested by the fact that disappointment in the quality of contemporary doctors would lead Huangfu Mi to examine it. More enlightening, however, is how the preface justifies the authority of the Inner Classic. It attributes the Inner Classic to the Yellow Emperor’s investigation of medicine and roots its authority in the fact the he ‘… inwardly investigated the five viscera and the six bowels; outwardly, surveyed the channels and network vessels, the qi, blood, and the complexion; consulted heaven and earth and examined people and things; took the nature allotted by heaven as the basis; and exhaustively studied the marvellous and thoroughly inquired into transformation’ (Huangdi sanbu Zhenjiu jiayijing, preface, 16). For Huangfu Mi, the authority of the Inner Classic lay in its transmission of the knowledge of the sages, a wisdom beyond the ken of contemporary humans.

When he investigated the Inner Classic, however, he found it in disarray. The texts were not ‘in order’ and contained many ‘repetitions’ and ‘misplacements’. He therefore rearranged the texts in a logical order and removed repetitions and unnecessary verbiage to produce a new text (Huangdi sanbu Zhenjiu jiayijing, preface, 20). Huangfu Mi’s statement has often been taken as an indication that the Inner Classic was on the verge of being lost (Unschuld 2003: 22) – and he intended it that way – but given the nature of textual transmission among physicians, described above, it is far more likely that he was simply looking at a doctor’s text with the eyes of a literatus accustomed to a completely different book culture. Huangfu Mi approached the Inner Classic expecting a ‘book’ in which all of the contents had fixed locations and logical relationships. The repetitions and variations inevitably produced by physicians’ textual practices therefore struck him as textual errors, and he set out to do what any good scholar would under the circumstances: produce a reliable edition. ⁴

The next two known compilations of the Inner Classic unfortunately provide us with little information about its claims to canonical status save the fact that they were produced by and circulated among the social elite. The Quan Yuanqi 全元起 (fl. sixth century) edition of Inquiry into the Fundamental (Suwen 素問) has been lost since the Song – fragments quoted in other texts are all that survive – and the preface is missing from the rediscovered copy of Grand Fundamental (Taisu 太素), compiled by Yang Shangshan 楊上善 (fl. early Tang). The third, however, reaffirms Huangfu Mi’s view of the Inner Classic’s origins and authority. In the preface, dated 762, to his edition of Inquiry into the Fundamental, Wang Bing 王冰 (fl. eighth century) affirms that the Inner Classic is the work of the Yellow Emperor and that without such sagely advice one cannot hope to ‘be released from the grievous bondage [of illness], complete and guide the true qi, enable the common people to attain a long life, or aid the frail and weak to obtain peace’.

Like Huangfu Mi, Wang Bing complains in his preface that the Inquiry into the Fundamental was in poor condition when he started. He took a nine-fascicle edition of Quan Yuanqi’s edition as his base text but found it necessary to edit it extensively. He removed redundant chapters, corrected passages he deemed corrupt, altered the wording of the dialogues to maintain proper etiquette between ruler and minister, divided chapters he felt combined two originally separate discourses, and added chapters to replace those that had been lost (Chongguang buzhu Suwen, preface, 5–6). Since Quan Yuanqi’s edition is no longer available, we cannot be certain what type of text Wang Bing was examining, but it seems likely that once again we are witnessing the reaction of an elite scholar to the very different writing and reading habits of non-elite physicians.

In spite of the sagely pedigree claimed for it, throughout this period, the Inner Classic remained one among many esteemed medical texts. In Chen Yanzhi’s 陳延之 (c. late fourth to early fifth centuries) Short Essay on Classical Formulae ( Jingfang xiaopin 經方小品), a Methods of the Yellow Fundamental (Huangsu fang 黃素方) is mentioned in a list of important medical texts; however, even if we assume that this text is related to Inquiry into the Fundamental, it is only one text in a list of sixteen (Yan Shiyun and Li Qizhong, 2009: 785). Likewise, in Sun Simiao’s 孫思邈 (d. 682) famous list of texts a doctor should study, Inquiry into the Fundamental, the Systematic Classic of Acumoxa, and the Needle Classic are listed first, but they are followed by twelve other medical texts and then by a variety of non-medical texts and methods of divination (Qianjin yaofang, 1.1–2).

By the end of the Tang, although the position of the Inner Classic vis-à-vis other medical texts was largely unchanged, the nature of the Inner Classic as a physical text had changed greatly. What had been a corpus of texts extant only in ad hoc compilations produced by and for physicians now also existed in at least four compilations produced by and structured according to the norms of the literati elite. These compilations fixed the form of the text in ways that decreased its mutability and made it more easily shared. The shift from writing on bamboo slips to writing on paper also vastly increased its portability. They altered both the nature of its claims to authority – by limiting that authority to particular books as opposed to the Inner Classic corpus as a whole – and the scope of its claims – by allowing larger numbers of people to share the same edition of the text. These developments continued and reached unprecedented levels during the following Song dynasty.

The Song dynasty (960–1279)

The Song was a period of intense transformation in many aspects of Chinese society, economy, government, and thought. The changes most relevant to this discussion are the development of printing, an unprecedented involvement by the government in medical affairs, and the changing status of medical practice as an occupation (Hinrichs 2013; Boyanton 2015: 76–115; Hymes 2015).

Although the technology of woodblock printing was invented during the Tang, it was only in the Song that it became the basis of a large publishing industry. The ability to rapidly produce large numbers of books drove the cost of books down, making literacy and book ownership affordable to a larger segment of the population than ever before. The Song book market generated a wide range of books, from expensive editions that are considered among the finest examples of printing ever produced to cheap editions of poor quality. Books were published on a wide variety of topics, including medical books of many types (Hymes 2015: 542–68).

In addition to increasing the availability of books, printing allowed the circulation of texts in identical editions. Copying texts by hand tended to introduce errors, and when working with manuscripts, it was a standard practice to compare several copies and produce a new edition for one’s own use (Hymes 2015: 562). This process meant that for all but the most common texts, any two copies were unlikely to be identical. Mass printing created the possibility of readers widely separated in space or time working from truly identical editions.

The Song government capitalised on this possibility in one of its most innovative interventions in medicine: large-scale, government-sponsored medical publishing. Before the Song, it was rare for the court to commission medical texts. A total of five medical books were produced for the court prior to the Song. From its inception, the Song broke with this precedent, and the state continued to produce and distribute medical books throughout the Northern Song (960–1127). The height of Song government medical publishing occurred between 1057 and 1069, when the government established the Bureau for Editing Medical Texts (Jiaozheng yishu ju 校正醫書局). The court was concerned that the most significant medical texts were in limited circulation in editions that were often erroneous. The Bureau was tasked with correcting this problem by editing and publishing accurate editions of these books. It was the first and almost the only time a Chinese court used medical publication as a means of benefiting the people by improving medical practice (Hinrichs 2013: 102–8).

Over the course of twelve years, the Bureau produced fifteen books (Table 7.2), the last eight of which rapidly became among the most widely cited medical texts in Chinese history.

The influence of these texts appears to have been due to multiple factors that are not easily disentangled. Medical texts produced by the government certainly carried a type of prestige that other medical texts lacked. It is difficult to discern, however, whether their influence derived from this prestige or from their ready availability. The government saw to it that its texts were widely distributed, and private reprints spread them even further (Goldschmidt 2008: 87–93; Miyashita 1967: 135–8).

If we cannot fully explain the success of the Bureau’s publications, the prefaces its editors wrote allow us to reconstruct their purpose and reasoning to a far greater extent than is possible for earlier editors of the Inner Classic. An analysis of these prefaces reveals an arbitrariness in the process of canonisation that differs strikingly from the aura of inevitability and independent value that surrounds canonical texts.

Although its members changed over the course of its history, a core group of six high-status, well-educated men was responsible for producing the last eight texts. With two exceptions – who participated in the editing of only two of these eight books – the members of this core group were all connected to one another through their association with a man named Gao Ruona 高若訥 (997–1055). He was the father of one editor, the father-in-law of another, and taught medicine to a third, whose brother, in turn, was the fourth. Gao Ruona himself passed the civil service examination in 1024 and held several high positions in the Song government but also nourished an interest in medicine and wrote a text on the Han-dynasty medical text Treatise on Cold Damage (Goldschmidt 2008: 79–83). In addition to their connection with Gao Ruona, the core group of editors all shared an active interest in medicine. The two brothers were practising physicians, and their father, like Gao Ruona, had written books on the Treatise on Cold Damage.

The original imperial mandate that created the Bureau appears to have included a list of texts to be edited and published. The editors’ memorial on submitting the Jiayou Materia Medica (Jiayou bencao 嘉祐本草) to the throne states that they were entrusted with editing:

… the Divine Farmer’s Materia Medica (Shennong bencao 神農本草), the Numinous Pivot, the Grand Fundamental, the Systematic Classic [of Acumoxa], and Inquiry into the Fundamental, as well as the formularies Expansive Aid (Guangji 廣濟), [Formulae worth] a Thousand Gold (Qianjin 千金), and Secret Essentials of the Outer Terrace (Waitai miyao 外臺秘要) …

Conspicuously absent from this list are the works of Zhang Ji 张机 (stylename Zhang Zhongjing; 150–219) – author of the Treatise on Cold Damage, three of whose books the Bureau published under the direction of these four editors – and the Classic of the Pulse, which they published in 1069. Furthermore, they never published the formulary Expansive Aid, which was included in the list. There is no evidence that the imperial mandate was ever revised. The final decision about what to edit and publish was left in the hands of the editors, and they chose to publish texts they personally felt were important. The authority of imperial publication was thus conferred on texts that were chosen neither by imperial decree nor because they were universally acknowledged as worthy of canonisation, but merely because a small group of men was able to bend the imperial apparatus to their own purposes.

Among the texts they chose to anoint with imperial blessing were two compilations of the Inner Classic corpus: Inquiry into the Fundamental and the Systematic Classic of Acumoxa. Unlike previous editors, they do not lament finding the texts in a deplorable condition, though editing was still necessary, particularly for the Systematic Classic of Acumoxa (Chongguang buzhu Suwen, preface, 4–5; Zhenjiu jiayi jing, preface, 12–13). The Song publication likely saved these compilations of the Inner Classic from being lost. Both Grand Fundamental and the Needle Classic were extant at the time the Bureau was working, but were lost shortly afterwards. Ironically, however, the success of the Bureau’s publications may have contributed to this loss. As the new, authoritative editions became widely available in cheap, printed editions, there was little incentive to copy and preserve manuscripts of other editions. Printing and canonisation thus acted as a two-edged sword, ensuring the survival of the chosen texts and hastening the demise of those not chosen.

The final eight texts published by the Bureau both survived and thrived. They remain the standard editions of these texts even today. An exception to the hegemony of the Bureau’s texts, however, is Numinous Pivot. As noted in Table 7.2, the Numinous Pivot produced by the Bureau in 1057 was not published. In their preface to Inquiry into the Fundamental, the Bureau’s editors lament the poor quality of the Numinous Pivot available to them, and this may account for their failure to produce an edition of it. The extant editions of the Numinous Pivot derive from a private edition submitted to the court in 1155 by Shi Song 史崧 (fl. twelfth century), which he produced by comparing an edition of the Numinous Pivot held by his family to other texts with parallel passages (Sivin 1993: 203). Once again, we see an element of arbitrariness in the process of canonisation. Although the Song government had desired to publish an edition of the Numinous Pivot, the edition which finally received its blessing and became canonical was produced by a single individual with no direct connection to the court.

The Song editors of the Inner Classic continued to justify its canonicity through reference to the Yellow Emperor. It would be a mistake, however, to assume that the meaning of this claim was the same as it was during the Han or even the Tang. Even in the earliest periods, the myth of the Yellow Emperor was polysemous. Charles LeBlanc has identified twenty different themes in the stories surrounding the figure of the Yellow Emperor in the ancient sources. He was the founding ancestor of Chinese civilisation, an icon of good governance, a devotee of esoteric practices of sexual self-cultivation, and an immortal transcendent, among many other roles (LeBlanc 1985). Nor was that the end of his transformations: the meanings associated with him continued to grow and develop throughout the imperial period and into modern times (Jochim 1990; Yates 1997: 17–19; Harper 1999; Marsili 2003; Leibold 2006; Shinno 2007). The Song editors’ support for the claim that the Inner Classic conveyed the wisdom of the Yellow Emperor is particularly significant since the Song witnessed serious challenges to that claim (Brown 2015: 146) alongside a widespread conviction that the ideal method of governance was to found in the ‘Way of the sages’ (shengren zhi dao 聖人之道) (Bol 1994: 126–31, 156–71, 181). The stakes were high. Successfully claiming the Yellow Emperor as the founder of medicine would provide a strong basis for assertions that it was a proper calling for the literati elite. Failure to establish a suitable pedigree could result in it being seen as a profession whose respectability was dubious at best (Boyanton 2015: 90–7, 130–4).

The Song editions of the Inner Classic contributed to a revolution in its status. Prior to the Song, the practice of medicine as an occupation was considered beneath the dignity of the elite, literati stratum of society. During the Song, this changed. The Bureau’s publications were part of an effort by the imperial government to encourage more elite men to become physicians, an effort which the Bureau’s members and an increasing number of literati supported (Boyanton 2015: 78–82; Goldschmidt 2008: 42–57; Hinrichs 2013: 99–117). Although the transformation of medicine into an elite occupation was still incomplete during the Song, increasing numbers of literati became physicians, bringing with them their assumptions about learning, reading, and writing. Previously, medical authors did not so much cite other medical texts as excerpt from them. The medical compendia of the Tang, for example, are compilations of treatment methods from a wide range of earlier texts. Whole sections are excerpted, but they are not used to justify the compiler’s assertions. They are simply presented as possible approaches to treatment. Starting in the Song, however, medical authors cited texts as support for their own arguments. They drew on a far smaller range of texts – in particular the publications of the Bureau – but they did so much more frequently. The Inner Classic was no longer one of a large group of important medical texts; it was now one of a smaller group of essential medical texts, reference to which was almost mandatory. The canon had narrowed, resulting in a more intense engagement with the texts that composed it.

Conclusion: canonicity as a historical process

In this chapter, I have shown that the nature of the authority claimed for the Inner Classic and received varied greatly over the roughly 1,200 years discussed here. These differences resulted primarily from changes in the technology of book production, the types of authority available to supporters of the Inner Classic, and the varying assumptions they made about the nature and use of books. The last two are inextricably tied to the changing social status of medicine as an occupation. My findings, therefore, largely parallel Richard Ohmann’s first three conclusions in his classic study of literary canon formation, namely:

(1) A canon – a shared understanding of what literature is worth preserving – takes shape through a troubled historical process. (2) It emerges through specific institutions and practices, not in some historically invariant way. (3) These institutions are likely to have a rather well-defined class base.

His fourth conclusion – that the ideology of the ruling class is reproduced in canon formation by a subordinate class acting in its perceived self-interest – is, I think, also applicable to the canonisation of the Inner Classic. However, limitations of space make it impossible to demonstrate this point here.

Taken as a case study of canonisation, the example of the Inner Classic offers several further insights. First, the canonicity of long-lived canonical texts is not static. Changes in social conditions, economic structures, cultural values, and technology all impact the claims that can be made for a text and the types of authority it can be granted. If a text remains important over a long period of time, these changes will inevitably be reflected in the way its canonicity is understood and even in the structure and physical form of the text itself. Second, the history surveyed here suggests that such changes will tend to be obscured by assertions that canonical texts possess eternal value and therefore unchanging form. Third, although canonisation is a restrictive phenomenon that operates by means of exclusion, it serves, somewhat counterintuitively, to enlarge the field of discourse by focusing discussion on a limited number of shared texts. This capacity of canonisation becomes particularly powerful after the advent of printing and the development of a thriving book market.

This chapter concludes with the Song dynasty because the medical canon formed by the Bureau for Editing Medical Texts has remained the core of the Chinese medical canon from that time up to today. This does not, however, indicate that either the canon or the canonicity of the Inner Classic has been static since that time. Research has shown that during the last thousand years, other texts have risen to and fallen from prominence and the relative importance of the texts within the canon has fluctuated greatly (Chapter 9 in this volume; Hanson 2003; Leung 2003; Simonis 2015; Vigouroux 2015). The process of asserting, disputing, and renegotiating the Inner Classic’s canonicity continues to this day. Far from indicating that its canonical status is in jeopardy, this engagement with the text is precisely the proof of the Inner Classic’s continuing significance for East Asian medicine. Canonicity is only fixed when the texts involved are dead; the canonicity of a living text is always an ongoing process.