Articles by alphabetic order
A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
 Ā Ī Ñ Ś Ū Ö Ō
1 2 3 4 5 6 7 8 9 0


Is a Da zangjing 大藏經 a Canon?

From Tibetan Buddhist Encyclopedia
Jump to navigation Jump to search





Is a Da zangjing 大藏經 a Canon? On the Use of “Canon” with Regard to Chinese Buddhist Anthologies

A paper presented to the conference “study of Block-printed Edition of Tripitaka, Past, Present and Future, Centering around Jiaxing SutraHangzhou Jingshan Temple, May 10, 2015

Charles B. Jones (周文廣)

School of Theology and Religious Studies

The Catholic University of America


Introduction


This is a paper about scholarly categories in the discipline of religious studies. In particular, it is about the problems that arise when scholars from one cultural/linguistic group form the categories by which to study and analyze the religious traditions of other cultural/linguistic groups. Because this is a conference on the Jiaxing Canon, I want to take the opportunity to examine the use of the western term “canon” as a designation for collections of religious

literature in Chinese Buddhism in general, calling attention to special problems presented in the case of the Jiaxing collection. While I do not anticipate that the field will discontinue use of the wordcanon” as a result of this exercise, I hope at least to make scholars (especially those outside the west) aware of the connotations that the term brings with it, and to stimulate discussion on the things that make Chinese Buddhist anthologies distinct from western canons.


What is a “Canon”?


The wordcanon” goes far back in western linguistic history. Scholars speculate that it derives from a Sumerian word for a particular kind of reed plant that was used for measuring, and made its way into the Greek language via various Semitic languages. In Greek, the word came to mean both “rule” and “list” (synonymous with katalogos). This second meaning appears later in history and is rarer. While these two meanings may not always apply to phenomena known as “canons,” Einar Thomassen point out that in the case of lists of canonical books, they do; such lists are both normative and descriptive.


However, a closer look at the usage of the wordcanon” in all its contexts reveals a much broader semantic range. For instance, as Jonathan Z. Smith points out, we may use the wordcanon” to indicate the list of an author’s authenticated texts (as in the Shakespearean canon). The word has been used to describe a more informal list of works deemed desirable for basic cultural literacy. As Tomas Hägg notes, the Greeks first used the word in this sense to indicate works of drama, philosophy, and oratory that should be taught in school. In fact, the related word “classic” originally meant works to be read in class for the training of students. In both of these cases, the wordcanon” points only to a catalogue, not a published collection.


Canon” as a normative set of texts arose strictly from within the western religious tradition; Judaism, Christianity, Islam, and various Iranian religions acted on the need to create a closed set of canonical texts for use in specific contexts, sometimes liturgical (what books may be read during a religious service?) and sometimes juridical (what books serve as ultimate authorities for regulating the life of a community?). In this regard, it is interesting to note that not all sets of authoritative or normative texts receive the designation “canon”: no one refers to the U.S. Constitution and Bill of Rights as canonical, despite their dominant role in law and politics.


Within religious studies, the concept of “canon” has been much contested, in part because the category of “canon” is inadequate to the evidence when applied globally. Gregory Schopen has already shown convincingly that the western academic study of Buddhism was initially conducted within a Protestant framework, in which written scriptures reigned supreme over all else (sola scriptura). However, I would extend that claim and say that an essentially Protestant view of scripture has informed not just the study of Buddhism but the overall approach of religious studies to canons in general.


For example, Kendall Folkert, in his brief study of the concept of canon, pointed to a particular puzzle in the academic study of Jain scriptures. On the one hand, there was an officially recognized body of scriptural texts that constituted a formal canon. On the other hand, there were texts outside of this canon that clearly enjoyed high status and a prominent place within the ritual life of the Jain community. Folkert resolved this by proposing a two-fold typology of

canon, Canon I and Canon II. Canon I comprised texts that gained their authoritative status through their employment by the community; in this case, he meant liturgical texts that gained their privileged place through their use within ritual. Canon II, by contrast, comprises texts that are authoritative in the way that Protestant theology conceives of the Bible; they are inherently authoritative due to their source and content. The problem only arose, Folkert said, because scholars were mistakenly looking at a Canon I text as if it were Canon II, applying a strictly Protestant notion to a non-Christian phenomenon.

In a similar vein, Einar Thomassen notes that texts may become authoritative for a community either through revelation or tradition. If it is through revelation, then it is the status of the source that grants the text its authority. If tradition is the mechanism, then long usage becomes the basis for a text’s privileged status. While this vocabulary is different, one may detect some overlap with Folkert’s categories. Tradition and usage would correspond to Canon I, while revelation would align with Canon II. However, Thommassen goes on to say that the two categories may converge in a single text when a community

attributes a text long in circulation to revelation. At this point, the text exhibits features belonging to both categories of Canons I and II. When this occurs, the Thommassen claims that a text becomes sacred scripture. I wish to propose here that we also need to look at canons from two perspectives. Following Folkert’s dual model with Thommassen’s modifications, I will suggest that canonformation takes place in two stages. The first, corresponding to Folkert’s Canon II, takes place prior to canonization and involves the setting of criteria for the acceptance of texts as canonical, and that under this

heading, few of the texts of the Jiaxing and subsequent canons would qualify. The second takes place after the establishment of a set of texts within a catalogue, and corresponds to the kind of ritual usage that Folkert describes as leading to Canon I, and under this rubric I believe that the Jiaxing and later anthologies do qualify as canonical. In other words, I will describe the process that flows into the assembling of a set of privileged texts in terms of Folkert’s Canon II, and the continuing evolution of that assemblage’s privileging in terms of his Canon I.

The Formation of the Chinese Buddhist Canon


Tanya Storch has ably described the earliest phases of the process of canon-formation in Chinese Buddhism in her book The History of Chinese Buddhist Bibliography. At the outset there were no unified collections of Buddhist texts, only a set of manuscripts that increased as texts entered China and were translated. While the earliest products of this effort were what Jonathan Z. Smith would call simple “lists” of texts with no criteria of inclusion, the

critical lists of Buddhist texts produced by a series of monk-scholars beginning with Daoan 道安 (312-385) over several centuries correspond to what Smith has called “catalogues,” that is, ordered lists created through the use of stated criteria of inclusion and having internal organization. These catalogues could only correspond to Folkert’s Canon II and have their source in Thommassen’s revelation at the outset because Buddhism was new and had yet to build up a longstanding tradition.


Thus, it is not surprising that the only texts included in the earliest Buddhist bibliographies were those attributed to the Buddha and the small set of thinkers who composed the Abhidharma texts, and the means of authenticating these texts was verification that they had been put into Chinese by a known and trusted translator. It is also not surprising that the cataloguers omitted indigenous Chinese texts at this early stage. Such native productions lacked both

the claim of revelation and validation by tradition. Such catalogues, however, lacked two things: authority and closure. (1) It is not at all certain that any Chinese Buddhist individual or institution had any obligation to use these catalogues to regulate their own collection and use of manuscripts. (2) Also, new texts and translations continued to appear, precluding any catalogue from achieving finality at this early phase.


The first of these two limitations changed when the imperial court involved itself in the process. Emperor Wu of the Liang dynasty was the first to commission catalogues, and at first there were both officially-sanctioned catalogues serving imperial interests and independent catalogues serving the needs of the sangha. Eventually, the imperial court began assembling its own collection of manuscripts that served to provide official versions of Buddhist texts against which to check other manuscripts for accuracy, and catalogues changed their names accordingly from Zhongjing 眾經 to Inner Canon or Neidian 內殿, implying that the true canon resided within the palace, with the emperor as guardian.


The final evolution took place during the Tang dynasty (618-907). It was at this time that the court moved from sponsoring catalogues and accumulating manuscripts to publishing the

collection of scriptures as a unified set. The contents of the project were based on the Da tang kaiyuan shijiao mulu 大唐開元釋教目錄 (730), which formed the basis for all subsequent printed editions of the canon. The Tang dynasty also saw the first use of the title Da zangjing 大藏經, printed texts rather than manuscripts, the adoption of a call number system, and a precise system of classification. The advent of printed collections replaced the manuscript practice that had made catalogues necessary in the first place.13


From the Tang dynasty forward, the imperial court generally ordered, supervised, and underwrote the publication of printed Da zangjing. The Jiaxing Canon broke that pattern, though, with interesting consequences. As Lianbin Dai explains, the Jiaxing Canon was planned by Zibo Zhenke 紫柏眞可 (1543-1603) with the help of lay patrons, and they produced a rationale for its contents. However, when the project was taken over by the monk Xingcong 性琮 (1576-1669), accusations of profit-seeking came forward and editorial control over the contents loosened considerably. Works originally excluded were now incorporated into the project, and contemporary works were selected for inclusion, possibly to gain subventions from the authors. One example of which I am personally aware is the inclusion of the Xifang helun 西方合論 by

Yuan Hongdao 袁宏道 (1568-1610), a work completed in 1599, when the publication of the Jiaxing Canon was well underway. The consequence of this has been the expansion of subsequent Buddhist compendia such as the Taishō to include many more texts of less obviously canonical status than had been the case previously.


To review, so far we have seen that in the earliest phase of this process Buddhist scriptures entered China as Indian and Central Asian monks came in and, with the aid of teams of native literati, translated scriptures. Next, as the manuscripts began circulating, a few monks began putting together lists so as to create a kind of inventory of literature. Beginning with Daoan in the fourth century, monks and literati moved from making lists to making catalogues that both provided criteria for authenticating reliable literature and exposing spurious texts, and also began organizing the literature by genre. By the sixth century,

beginning with Emperor Wu of the Liang (r. 502–549), the royal court began taking over the function of creating catalogues, and later courts began accumulating collections of manuscripts in order not only to establish catalogues of reliable texts, but also to maintain authoritative editions. By the end of the Tang, the court moved into actual production of printed editions of authenticated Buddhist literature. By this time, nearly a thousand years had passed since Buddhism arrived in China, giving ample time for a tradition to develop. With this, the conditions were right for books to become canonical simply by communal acceptance and/or ritual usage over a long span of time. What had been a Canon II/revelation process of establishing texts by the authority of the source (the

Buddha) and validity of transmission into China (identification of reliable translators) now had room for a Canon I/tradition process to work as well. At this juncture, the collection of Chinese Buddhist literature does not yet fit the conditions for the genesis of a canon as outlined by most scholars. There is as yet no convergence of revelation and tradition (Thommassen). There is no closure, since new literature continually appeared, necessitating new bibliographies (Smith ). Only some of the texts within the catalogues were normative in the sense that they could serve as courts of appeal for settling disputed questions. The process has moved from lists to catalogues, but no farther.


In the next section, however, I will argue that the publication of printed anthologies under the title Da zangjing 大藏經 created the conditions for a true canon to appear.

Print Canons


What I described in the previous section may be regarded as the processes that flow into the creation of a canon; none of the products of these processes is a canon yet, but provide the conditions for canon-creation. Now I will turn to the phenomena that flow out of the formation of a canon.


I choose the appearance of printed anthologies as the pivot point between the flow in and the flow out because a published anthology of printed texts, regardless of the motives that brought it into being, produces unintended consequences.


First, as Lewis Lancaster argues, the mere existence of a printed set changes the way people regard a body of literature. “Once a name, yijiejing (All of the Jing [一切經]) came into common usage, we can assume that the idea of a canon in Chinese was fully established.”16 Second, once all the texts within the collection have been set in a uniform font on pages of uniform size and format and have been bound in uniform covers with uniform binding, the simple visual impression that the set conveys as a whole is one of uniformity of content and status. As Smith observed, canonization and canon are as much products of technology as theology; the invention of the codex matters. The codex and printing can turn a library into a single work. (See fig. 1)

Fig. 1: a modern edition of the Jiaxing Canon


In the case of the Chinese Buddhist Da zangjing, this homogenizing process gives rise to concrete effects. To understand what it did and did not do, we may turn to Miriam Levering’s 1989 contribution to the book Rethinking Scripture. In her discussion of scripture in Chinese Buddhism, she identified four different modes of reception and use: the informational, the transactional, the transformative, and the symbolic. In the first two of these modes, the production of an anthology with uniform production features does not seem to have made all the texts equal, while in the last two, it seems to have had just that effect.


(1) Reception may be described as informational when members of the community read texts for their ostensive meaning in order to inform themselves. Clearly, only a handful of Chinese Buddhists have ever read through an entire printed collection for this purpose. The modern monk Yinshun 印順 (1906-2005) did so, but I know of no others. The vast majority select from a few sutras that have achieved high status; Raoul Birnbaum notes that by the early 20th century, the Lotus Sutra, the Huayan Sutra, and the Śūraṅgama Sutra were preeminent, while my survey of late Ming dynasty literati conversation literature has revealed a few more. If one were to regard the entire Da zangjing as canonical, then these texts would comprise the “canon within the canon.”


(2) By transactional Levering refers to texts used primarily in rituals in order to do things such as create merit, obtain the empowerment (jiachi 加持) of a Buddha or bodhisattva, or produce a magical effect such as making rain. As with the first category, it would be rare for a practitioner to use the entire Da zangjing for this; in general one uses a particular ritual or dhāraṇī text for this purpose.


One exception to this is the practice of leafing through the entire Da zangjing with a glance at each page as a way of gaining merit. Otherwise, transactional practices such as recitation (without regard to whether or not the reciter understands), writing out scriptures in blood, and so on utilize only the sutra literature, and focus primarily on the same small set of scriptures mentioned above.


(3) In the transformative mode, one engages with texts not to gain information nor to transact exchanges of merit, but for personal transformation. Reception of scriptures in this mode pulls away from particular texts and can potentially extend to the entire collection since understanding of the words is not strictly necessary. Levering reports on elderly female devotees that came to sutra lectures but appeared not to be paying attention to the lecturer as they

sat quietly and recited Amitābha’s name or engaged in another practice. For them, the selection of the text did not matter, though within the lecture context it was certain to be a sutra. However, Levering reports another case in which a man, after having experienced enlightenment, took up the life of a hermit and developed the practice of reading through the entire collection over and over. He said that he did not always understand what he was reading, but that after his initial enlightenment he still drew inspiration from the practice. Significantly, he said he learned this practice from his teacher, who also read all 55 volumes of the Da zangjing.


(4) When we come to the symbolic approach, we encounter the mode of reception most likely to encompass the entirety of the published collection. According to Levering, this derives naturally from the early Mahāyānacult of the book,” in which written texts symbolized the ongoing presence of the Buddha and his dharma. In later Chinese Buddhism, a monastery or convent sought to acquire a complete set; From the dynastic period monasteries besought the court to bestow

the imperially-sponsored published anthology, and Holmes Welch tells how the abbot of the Guanzong Temple in Ningbo traveled to Beijing in order to raise the money necessary to purchase a set. This mode of reception most strikingly exhibits the tendency of published collections to take on a symbolic value by their mere existence as repositories of the entire Buddhist tradition. Their comprehensiveness and uniformity of appearance, and their enshrinement in a central part of the monastery, announces the centrality of the buddhadharma to all residents and visitors, whether or not anyone ever takes a volume out from behind the sliding glass doors to read.

Answering the Question


We return now to the question forming the title of this presentation: Is a Da zangjing a canon?


The answer is complicated. For one thing, the very process of identifying privileged texts within Chinese Buddhist history changed several times. Once a mass of texts in manuscript form began circulating, lists appeared, and after a while the lists became catalogues. The catalogues were compiled according to some standard, and the pre-Tang bibliographies, confining their contents to Indian and Central Asian Buddhist literature in translation, excluding Chinese productions, and presenting judgments on the reliability of this or that text, could be said to constitute a canon since they included texts taken as authentic and normative.


After some time, however, the imperial court took over the task of commissioning bibliographies, and once the catalogues served court interests rather than religious purposes, one might say the texts became more mixed. The advent of published collections continued this trend, with textual normativity deriving from imperial sanction and the association of less normative texts with highly valued ones simply by virtue of a common imprint and standardized format. The Jiaxing canon, in which editorial control was seemingly lost in a haze of commercial interest, could hardly be said to constitute a canon.


If we were to accept the definition of “canon” put forward by other western scholars, we would be hard pressed to accept later published Da zangjing as canons. From the Jiaxing Canon to the Taishō, we have now to do with quite a large and motley collection of sutras and sophisticated Indic śāstras (often in multiple translations) mixed with Chinese commentaries, dictionaries, epistolary exchanges, Chan “recorded sayings” literature, ritual texts, and even the bibliographies themselves. This hardly comports with a collection of texts that is (depending on the scholar putting forth the definition of “canon”) normative in its entirety, a resources for resolving disputed questions, a closed set, or an anthology sanctioned by its origin in revelation or longstanding tradition.


However, as I have tried to show, there is some room for regarding the various Da zangjing as canons. They contain texts that are unquestionably canonical such as the Lotus Sutra. The publication of Da zangjing with identical design and publishing specifications gives, as Smith noted, an appearance of uniform canonicity, which comes to expression in such practices as reading through the entire set (whether in earnest or by simple leafing through) and acquiring the entire set and putting it in a centralized display as a symbol of the Buddha’s presence. The very same set of texts, if purchased from difference sources with different cover designed, sizes, and fonts, would not give this same appearance.


I conclude, then, that the Da zangjing is not a canon in the commonly-accepted sense, with some small exceptions as noted. If that is so, then what should we call it? Miriam Levering, noting the problems with the term “canon,” for the most part simply transliterates Da zangjing without translation. I would suggest that the word zang could be translated as “treasury,” a word that in English can denote a curated collection of texts among other meanings. However, the term has become so habitually used that even in this paper I could not refrain from using the term “Jiaxing Canon,” so if this presentation has at least raised some reflection on the implications of that term, I will be satisfied.



Source