Catalogues in the Electronic Era: CBETA and The Korean Buddhist Canon: A Descriptive Catalogue.

  Lewis Lancaster

University of California , Berkeley


It is very pleasing to have The Korean Buddhist Canon: A Descriptive Catalogue reused in digital form on the internet. CBETA is to be congratulated for the great contribution they are making to our study of the Buddhist canon. When the Korean catalogue was published in codex form, it represented a traditional reference work for library collections.[1] That is to say, it gave information that could be used to go to other books and documents in the library to find data. In the format that is used by CBETA, we see a whole new way of referencing. The catalogue is no longer limited to simply indicating sources, it is being linked directed to the sources. This marks the great difference between the current Reference Room of most libraries and the needs and potential for referencing digital resources. Just as CBETA has moved the catalogue into an entirely new format, so too, librarians are beginning to build similar methods for the internet resources. It will no longer be sufficient to give bibliographies or guides that require the user to go out and search for each item. The linking of the data itself with the reference is a basic aspect of the power of information technology. CBETA is becoming a world class model for this procedure and I am grateful that the creators of the resource have decided to make use of my catalogue.


The compilation of The Korean Buddhist Canon: A Descriptive Catalogue was started soon after my arrival at the University of California , Berkeley in 1967 to teach in what was then known as the Department of Oriental Languages. My predecessor Professor F. Lessing had established Buddhist Studies within the department in the 1930s and it continues today under the Department of East Asian Languages and Culture, headed by Professor Robert Sharf. At the time of my assuming the post at Berkeley , the East Asiatic Library (now named the C.V. Starr East Asian Library) had just acquired a set of rubbings from the 13 th century printing blocks housed at Hae-in Monastery in Korea . The sheets were sewn into 1428 volumes and represented a distinct challenge to the library staff in terms of cataloguing and housing such a large collection. Since I was the new resident Buddhist specialist, the library appealed for help and I agreed to produce a shelf list of all the titles contained in the volumes. Fortunately, in 1970, I was able to visit Korea and see the printing blocks at Hae-in Monastery. The wonder of how more than 83,000 blocks could have been produced during a time of great duress in the 13 th century and have been preserved through wars and the constant danger of fires, increased my interest and appreciation for the rubbings. As the work progressed, I enlisted graduate students to assist me in making a fuller description of each text. The names of these individuals are mentioned in the original preface that is available online. My co-editor was Professor Sungbae Park , now teaching at SUNY Stony Brook. Eventually, the project took seven years to complete, working without the aid of computers and spreadsheets. It was done manually or with regular typewriters. When the time came to publish the catalogue, Phillip Lillienthal of the University of California Press took special interest and managed to find the funding necessary to produce the 723 page volume. Typesetting and printing was done in Korea , handset in metal font. Proofs of the pages had to be shipped back and forth by air-freight. In the end, we had to conclude the process and print, even though there were still unresolved problems.

Over the years, I have been asked why I did such a huge task for what must be a limited audience. The real reason was that I wanted to have the data for my own research and in many ways it was a personal endeavor. To the surprise of the press, the volumes did find buyers and in a reasonable period of time were all sold. We never considered doing a second printing since it was assumed that most libraries that would stock the book were among the initial buyers and would not repurchase. The copyright was returned to me, but I did not try to find a new publisher.

The Electronic Version of the Catalogue

It was a happy surprise when Professor Charles Muller contacted me and asked if I would give permission to have a digital version made. I was pleased at the prospect of the new format and I encouraged him to proceed. My only request was that the online version be made freely available without cost to users and without any restrictions for educational use. He agreed to these terms and in short order produced the new edition and has made it available online for some years. His work in providing this and other research tools to the scholarly community is unparalleled and we all owe him a great debt of gratitude.

After completing the catalogue, I became very involved in digitization of the canon. Working with the Tripitaka Korean Institute, the content of the Hae-in blocks was put into the computer and published on CD ROMs. The work of CBETA in Taiwan was the next major step in the evolving technology of Buddhist Studies, when they made the Taisho Edition available on the internet with search and retrieval software. This has revolutionized the use of the canonic material and changed the way in which Chinese Buddhism is studied. Now, the CBETA team incorporates the catalogue with this online data and I am honored to grant such use. A fuller description of how all these efforts have influenced Buddhist studies can be found in my articles contained in the Encyclopedia of Buddhism.[2]

The Ordering of the Texts

It has been nearly thirty years since the hard copy of The Korean Buddhist Canon: A Descriptive Catalogue appeared in print. During these years, I have continued to explore the arrangement of the printing blocks at Hae-in Monastery in order to understand the way in which the texts were ordered. A number of patterns can be discerned by looking carefully at the information contained in the catalogue.

(1) Dates of translations of the texts.

There are hundreds of different dates listed in the ancient catalogues of the Buddhist texts. While some dates are indicated in the first century C.E., these are hard to justify. The first firm date is in the mid-second century, 148 C .E. It is from that date that the translation bureau of An Shigao began to produce the Chinese versions. If we outline the pattern of translations over time we find two distinct periods:

(1) 148 -798 C .E.

(2) 982 -1158 C .E.

As the catalogue data informs us, the first stage of translations lasted for approximately 650 years and came to an end. From 798 C .E. to 982 C .E., we find no listing for a translation. Some texts were compiled and other composed in China during those years, but there are no recorded translations. The work of translating Indic and Central Asian materials, a task that once held the highest importance for Buddhist in China apparently did not proceed during this period of 184 years. It was the Northern Sung dynasty that revived the enterprise in 982 and the activity during this second period of translations lasted until ca. 1158 C .E. To be sure, over the centuries that followed a few more texts were translated but the effort was limited and made little impact.

(2) Ordering of the text.

The problem of understanding the methods that have been used to make an ordering of the hundreds of translations is a complex one. We can note that attempts were made to group certain texts together:

(a) Multiple translations of the same text are often put together. Sometimes these translations are listed by reference to the time of completion of the work. However, in the Korean block prints, there is no consistent use of putting each translation in temporal ordering.

(b) Grouping of texts that belong to a particular “school” of Buddhism has been attempted in many of the catalogues and in the printing block editions. This is a difficult task since the definition of the various schools was based on interpretation that could vary over time and between scholars.

(c) Grouping of texts by subject matter, such as Vinaya, Yogacara, Abhidharma, and Commentaries is reflected in some of the catalogues. This can be spotted in the Koryo arrangements but similar texts are found scattered throughout the set.

(d)  Grouping of texts by order of acquisition

In making a catalogue of an edition such as the Koryo, a new view of the ordering emerged. We can now identify four distinct units within the canon that can only be explained by reference to the way in which the texts were acquired. In other words, the ordering seems to maintain the original shelf listing of the materials that were arriving in the Koryo court in a sequence of acquisitions. Therefore, we might say that the arrangement is the work of librarians who took the prints coming from the Northern Sung and decided how to house them. The process did not involve a rearrangement of individual texts. There are four distinct units

Segment One

Materials completed before 730 C .E.

(1) K 1-1046

These texts most probably represent the original Kaipao print edition that was carved in Chengdu and housed in Kaifeng . They are the translations made before 730 C .E. Once the collection arrived in Koryo and a set of printing blocks was made from the Northern Sung prints, the order of the material was preserved. Even when the Koryo blocks of Hae-in Monastery were prepared after the destruction of the first set of blocks during the Mongol invasion, we believe the old order was maintained.

(2) K 1047-1087

These forty texts are placed immediately after the translations. They are authored texts and compilations made by Chinese, dated from 502 to 730 C .E. This collection of East Asian writings constitutes an important part of the canon and indicates the scriptural tradition was by no means limited to translations or texts attributed to Indic or Central Asian sources. It is interesting to note that none of these texts bears the word jing in the title. Since the Kaipao printed edition resulted from a command of the emperor that all the jing in the da tsang (i.e. library) be put on the printing blocks, it would seem that these East Asian generated documents were not included in the original Kaipao.

Segment Two

Materials produced before 1000 C .E.

(1) K 1088-1256

After the resumption of translations in the Northern Sung, a supplement containing new material was sent to Koryo. It represented the translations completed between ca 973 C .E. and 999 C .E.

(2) K 1257-1261

Similar to the pattern we see in Segment One, the collection of translations was followed by compilations and authored texts completed during the same period of the 10 th century (i.e. 940 -997 C .E. in the texts of this unit).

Segment Three

Supplement of materials done after 730 C .E.

(1) K 1262-1397

This appears to be a supplement of translations done after 730 C .E. and therefore not found in the original Kaipao.

(2) K 1398-1406

As we see in both Segment One and Segment Two, the East Asian texts are placed after the translations.

Segment Four

Second Supplement of Northern Sung translations

(1) K 1407-1496

A second supplement of new translations came from the Northern Sung. These were completed between 1000 and 1072 C .E.

(2) K 1497-1514

Following the pattern above, we have a fourth set of compilations and authored texts. This is a miscellany of materials that was written between 602 -1259 C .E. A few of the texts appearing in this segment were written or compiled in Korea .

In these four segments, we can note that the librarians arranged the documents by designating two types. In each segment, we find the translations come first followed by compilations and authored texts. Thus, the arrangement is:

Segment One

Translations K 1-1046

Compilations K 1047-1087

Segment Two

Translations K 1088-1256

Compilations K 1257-1261

Segment Three

Translations K 1262-1397

Compilations K 1398-1406

Segment Four

Translations K 1407-1496

Compilations (and miscellany) K 1497-1514

This charting of the arrangement shows us that the Koryo II version used for the The Korean Buddhist Canon: A Descriptive Catalogue, was first done by acquisition and there was no attempt to intermingle additions with the earlier collections of texts.

The careful distinction of translations and East Asian compilations and authored works needs to be studied in greater detail. First, we note that the East Asian material was considered to be canonic and was included in the editions, even though grouped after the translations. The earliest dated compilations were done sometime between is 502 -557 C .E. in the Liang dynasty (K 1047, 1049, 1050, 1051, 1053, 1080).

Second, the arrangement allows us to trace the movement of the printed editions from Northern Sung to Koryo. Histories record that the work of carving the Kaipao, which is probably represented by K 1-1046, took place between 971 and ca. 983 C .E. Immediately after the completion of those carvings, the Northern Sung court undertook the resumption of translation of Sanskrit texts. Therefore, the first supplement to the canon represented translations made from 983 -1000 C .E. This came to Koryo and is preserved as K 1088-1256. These two segments were missing all of the translations made from 730 -798 C .E. and so blocks were carved to capture the products of those years. We find this supplement in K 1262-1397. It is hard to know whether both of these supplements came to Koryo at the same time. If they had done so, it would have been more logical, in terms of acquisition housing, to have put the earlier translations before the later Northern Sung. Since, the Northern Sung translations of 983 -1000 C .E. became Segment Two, it is tempting to assume that Segment Three translations of 730 – 798 C .E. were acquired later. Here we can chart the acquisition history:

K 1-1046 (translations up to 730 C .E) (arrived in Koryo in 991 C .E.)

K 1088-1256 (translations up to 940 -1000 C .E.) (may have arrived in Koryo in 1021 C .E.)

K1262-1397 (translations 730 -798 C .E.) (perhaps arrived after 1021 C .E.)

K 1407-1496 (translations up to 1072 C .E.) (arrived after 1073 C .E. but the compilations that follow carry the late date of 1090)

As stated above, the ordering of Koryo II indicates that the Northern Sung prints were received in this order and housed accordingly. The order was not done strictly by date of translation. As we see from the chart, the first segment only takes in material produced before 730 C .E., the year of the publication of the catalogue of the Kaiyuan era (K 1062) compiled by Zhi-sheng. This raises the question of why the carvers of the Kaipao plates done between 971 -983 C .E. did not include the translations done between 730 -798 C .E. It would appear that they limited their content to the listings of Zhi-sheng's catalogue rather than doing all of the jing in the da tsang . Therefore, the court in Kaifeng was required to mount a special project to make prints for these post-730 translations.

The compilations that were attached at the end of each translation segment, also indicate the history of acquisition. While segments three and four contain a miscellany of texts and even a few translations, the chart below only indicates the dates of those works written in China .

K 1047-1087 (written between ca 502 -730 C .E.)

K1257-1261 (written between 940 -997 C .E.)

K 1398-1406 (written between ca 627 -945 C .E.)

K 1497-1514 (written between 788 -1090 C .E.)

When the material was rearranged in Japan in the 20 th century, this structure of the Koryo was removed and cannot be seen within the Taisho numbering. Because the Taisho edition was a new method of organization of the individual titles, the editors removed the references to the “case” characters that had long been the guide based on the “Thousand Character Classic.” In making the catalogue of the Koryo II rubbings, I decided to retain the information regarding the “case” characters. Here is the outline of how the case characters relate to the structure

(1) Original Kaipao K 1-1046 Case characters 1-437

(2) Compilations/authored texts done before 730 C .E. K 1047-1087 Case characters 438-480

(3) First Northern Sung Translations K 1088-1256 Case characters 481-510

(4) Compilations/authored texts done in 10 th century K 1257-1261 Case characters 511-520

(5) Supplement of texts translated after 730 C .E. K 1262-1397 Case characters 521-555

(6) Compilations/authored texts K 1398-1406 Case characters 556-585

(7) Second Northern Sung Translations K 1407-1496 Case characters 586-628

(8) Compilations/Authored texts K 1497-1514 Case characters 629-639

The “case” characters were the primary tool for keeping the order of shelving or storing the texts. The indicators are especially important for noting the comparisons between editions. For example, the Koryo and the Jin versions follow identical patterns for the “case” characters from K 1-1256. The identical listing of the two is attributed to a common source, the rubbings coming from the Kaipao and the Northern Sung first supplement. From K 1262, the “case” characters are different in the two and it indicates that the number of rolls differed and so the “case” numbering. In order to understand the relationship between Jin and Koryo, we need to look at the ordering of the first set of printing blocks, now referred to as Koryo I. Fortunately, a sizable number of Koryo I rubbings have been discovered in Korea , Tsushima, and Nanzenji in Japan . These examples are now being scanned for display on the internet through the Tripitaka Koreana Institute in Seoul .

The attempt was to put 10 scrolls into each “case.” However, the rule was not followed if the titles of a text had a few more scrolls. For example “case” number 88 had eleven scrolls contained eight texts (i.e. K 48-55). Within each “case” it was difficult to keep consistent ordering of the scrolls. When they belonged to one title, then each scroll was numbered and could be arranged in proper order. However, when multiple titles were found within the “case” it was not easy to give ordering. “Case” 169 is a case in point.

For example, we find K 334-K354 all contained in “Case” 169.
(This chart gives the K number, translation date, Taisho number, Nanjio number and the Zhengyuan lu number[3])

K 334 Translated 388-407T. 1342Nj 374Zy 391
K 335 Translated 562-565T. 1343Nj 375Zy 392
K 336 Translated 525-539T. 1344Nj 373Zy 395
K 337 Translated 587T. 1345Nj 372Zy 396
K 338 Translated 587T. 1334Nj 367Zy 407
K 339 Translator Unknown (317-420)T. 1358Nj 339Zy 402
K 340 Translator Unknown (317-420) T. 1357Nj 338Zy 401
K 341 Translator Unknown (317-420) T. 1044Nj 340Zy 404
K 342 Translator Unknown (327-420) T. 1137Nj 369Zy 411
K 343 Translator Unknown (327-420)T. 1138aNj 370Zy 412
K 344 Translation 222-228T. 1356Nj 337Zy 403
K 345 Translator Unknown (502-557) T. 1045b Nj 341Zy 405
K 346 Translator Unknown (502-557)T. 1333Nj 368Zy 406
K 347 223-253T. 1351 Nj 364 Zy 408
K 348 381-395T. 1352 Nj 365 Zy 409
K 349 585-600T. 1353Nj --- 
K 350 585-600T. 1354 Nj 366 Zy 410
K 351 693T. 1139Nj 371Zy 413
K 352 704T. 1024Nj 380Zy 414
K 353 419T. 1043Nj 326Zy 415

If we analyze the list in terms of content, we see the following:

K336 and K 337 are translations of the same text and they are placed next to each other. However, K 339, K340, K 344 are also the same content but are not grouped in Koryo. They are put together in the Nanjio listing, Nj 337, 338, 339 and in order of time of translation and this is followed by the Taisho numbering 1356, 1357, 1358. In another instance of multiple translations, we find K 342, K 343, and K 351 for the same text. The ordering is not consistent with title or date of translation. Again, Nj 369, 370, 371 puts the grouping in better order and is followed by Taisho. By looking at the Zhengyuan lu done in the Tang dynasty, we see that the ordering was unsettled even in those earlier days. Zhengyuan lu does not agree with Koryo, Taisho, or Nanjio in ordering K 339, 340, and 344. The listing using the Zhengyuan lu order would be as follows:

Zyl Koryo Nanjio Taisho
401 340 338 1357
402 339 339 1358
403 344 337 1356

It might be argued that the Koryo rubbings should be rearranged to follow the more logical order. However, the Berkeley set was the standard for the catalogue, even though inconsistencies are noted. With a mixed signal from other catalogues, it is difficult to solve all of these micro-listings with ease or assurance.

There is still much to be done in working with catalogues and ordering of the canonic texts. Even though 4175 Buddhist texts have been noted in twenty two catalogues, there are missing materials from Korea and Vietnam . We are not keeping pace with modern publications of compiled lectures by contemporary monks and nuns. The Dainihon zokuzokyo[4] has been little studied in terms of its ordering and unique example of texts from Fang Shan stone carvings need study. The CBETA listings of comparative catalogue information for each text is a major step forward. As we move forward with such research, the contextualization of each Buddhist text will be more precise and even the content of the text better understood. I look forward to using my own data in the new ways that are made available on the CBETA site. My thanks to all who make this technology so cogent to our research.


[1] Lewis R. Lancaster and Sung-bae Park , The Korean Buddhist Canon: A Descriptive Catalogue. University of California Press: Berkeley/Los Angeles/London. 1979.

[2] See my articles entitled: “Electronic Publications,” “Cyber Buddhism”, “Digital research resources”, “Digital input of Buddhist texts.” And “Technology and Buddhism” in Encyclopedia of Buddhism edited by Damien Keown and Charles S. Prebish. Routledge: London / New York . 2007.

[3] See K 1401 (T. 2157)

[4] Dainihon zokuzokyo. Zokyo shoin : Kyoto . 1902-05.