Singapore Workshop on a Japanese Performing Arts Resource Center
シンガポール シンポジウム ・日本舞台芸術リソースセンター
June 24-26, 2005
National University of Singapore

Generously Funded by The Japan Foundation

Individual Presentation Abstract


Michael Watson
Meiji Gakuin University, Yokohama

Electronic databases for antiquarians? Extra-repertory noh play texts in the digital performing arts world

pdf version


Back to full Agenda

There are many good as well as practical reasons why the study of noh theatre has focussed on plays in the currently performed repertoire ( genkō kyoku 現行曲 ), works originating in the Muromachi period which have continued to be sung and acted for over five hundred years. We are fortunate indeed that the repertoire of noh theatre is large and varied, and contains so many plays that and that there is ample opportunity in Japan not only to study the textual elements of plays, but also to see and hear them in performance. Their texts are readily available in different forms, with summaries and translations into modern Japanese, and often in other languages as well. In the major series of classical Japanese literature, two or or three volumes are devoted to yōkyoku , with carefully annotated texts, notes on sources, allusions, and staging. For the study of song and dance, we can consult the utaibon and other texts produced by the five schools themselves. For each play in the repertoire, there are many resources—and people—we can consult to learn more about music, masks, costumes, stage movements, performance variations, even particular gestures.

When we venture outside the established repertory and attempt to read plays that are no longer performed ( bangai kyoku 番外曲 ), the familar apparatus of annotations, translations, and secondary resources is largely unavailable to us. Most of them have never been studied closely except by their editors and a limited number of specialists. This makes it all the more essential to build up a custom-made set of electronic tools: databases for “antiquarians” who find a joy in the challenge of texts that may not have been sung or performed for centuries. The discoveries await those who investigate bangai texts are of much more than mere antiquarian interest, however.

“Extra-repertory” noh plays outnumber plays in the current repertoire by a factor over over ten to one. To identify what plays are regarded as falling with the general performance tradition—leaving shinsaku 新作 from the Meiji to Heisei periods aside—one place to begin might be Sanari (1938-39) or Ōtani (1978), collections that contain 254 and 253 plays respectively. “Two hundred and fifty” some plays are usually said to represent the current repertoire, even if strictly speaking some of the plays are haikyoku 廃曲 , seldom performed by any school. A broader definition would add to this a few dozen plays performed by just one of five schools. We might also include few older plays that have been revived ( fukugen kyoku 復元曲 ). Even by the most generous of definitions, however, it would be hard to come to a figure that comes to more than a fraction of the three thousand or more of bangai kyoku that survived in manuscript form or woodblock printings up to the end of the Edo period.

This paper will reflect on my experience in carrying out a survey of this large body of texts to locate plays on a specific theme—the Genpei War—and to read through the plays found, comparing them with current plays on the same topic. I will address the following questions:

1. Where can one get an overview of this large corpus of bangai kyoku ?
2. What kind of editions and reference sources exist? How much help do they provide?
3. How does one start to track down plays of interest?
4. Just how many might one find? Setting limits to one's search.
5. What print and electronic resources can aid one in reading unannotated texts?
6. Is it really worth spending time and energy on “marginal” works?
7. What can a study of extra-repertory works contribute to our understanding of the performed repertory of plays? How can it contribute to our understanding of the historical development of noh?
8. If material concerning bangai texts were provided through GloPAD, what form might it take?


Printed and electronic resources cited

(1) Annotated editions (a) Sanari Kentarō 佐成謙太郎 , ed. Yōkyoku taikan 謡曲大観 , 6 vols. (Meiji Shoin, 1930-31).

Contains a total of 254 plays, all in the “current” repertoire. Arrangement of plays: gojûon order of titles (as written in historic kana). All plays accompanied by introductory matter, headnote annotation, and modern Japanese translation.

(b) Haga Yaichi 芳賀矢一 and Sasaki Nobutsuna 佐佐木信綱 , eds. Kōchū yōkyoku sōsho 校註謡曲叢書 , 3 vols. (Hakubunkan: 1913-15; reprint Rinsen shoten: 1987).

Annotated edition of 545 noh texts, some 290 of them bangai kyoku . One of the most accessible and readable collections of bangai plays. Headnote annotation. Arrangement of plays: gojûon order of titles.

(2) bangai editions by Tanaka Makoto

Tanaka Makoto, ed. Bangai kyoku, 2 vols. (Koten bunko, 1950-52) *vol. 33 and 50 田中允編 『番外謡曲(角淵本)』(古典文庫)

These editions are referred to as 正 and 続 .

Tanaka Makoto, ed. Mikan yōkyokushū , 52 vols. (Koten bunko, 1963-1998) 田中允編 『未刊謡曲集』 (古典文庫)

The first series ( 正 ) consists of 30 volumes, the second ( 続 ) of 22. The collection includes both bangai kyoku from early printed texts and plays written in the twentieth-century. Broadly speaking, the arrangement of volumes is chronological, but the order of texts differs from volume to volume. Texts contain notes on variant readings, but not headnotes, list of roles, or other form of annotation. Brief notes on textual matters are included at the beginning of volumes. Many variant texts are included. The final three volume of the series ( zoku 20-22) contain an invaluable index to all plays in gojūon order.

(3) Guide to plays within and without the repertory

(a) Maruoka Kei 丸岡桂 , Kokin yōkyoku kaidai 古今謡曲解題 (1918), reprinted ed. Nishino Haruo 西野春雄 (Kokin yōkyoku kaidai kankōkai, 1984)

A guide to a total of 2584 pieces, including 832 “complete” noh plays ( kankyoku 完曲 ), 241 utai-mono 謡い物 , as well as plays known under different titles, etc. Summaries are given for all of the complete plays, with a note of the identity of shite , waki , and other roles. In the 1984 edition, Nishino adds information about editions in which the plays are to be found. The plays are ordered by subject matter ( 「歌人及俊秀」「武人」「世話巷説」 etc.), making it possible to search for plays by theme. Good indices.

(b) Takemoto Mikio 竹本幹 and Hashimoto Asao 橋本朝生 , Nô kyôgen hikkei 能狂言必携 (Gakutôsha, 1995), p. 53-120 ( 能作品全覧 )

A clear guide to plays in gojūon order, with a short plot summary and information about probable date, authorship. The inclusion of bangai is selective, focussing on earlier works. Significant examples of modern plays ( shinsaku ) are listed, however.

(c) Nogami Toyoichirô 野上豊一郎 , ed., Nôgaku zensho 能楽全書 . Revised ed. (Tokyo Sôgensha, 1979-81), 3:235-279 ( 謡曲曲目総覧 )

(4) The UTAHI Hangyō bunko ( 半魚文庫 ) site

Carefully edited by team of scholars from print edition: Nonomura Kaizō 野々村戒三 , ed., Yōkyoku sanbyakugojōshū 謠曲三百五十番集 (Nihon meicho zenshū kankōkai, 1928). As of June 2005, all but 8 of 253 plays now online, though less than half have received final editing checks. Arrangement of plays: traditional (beginning with Okina, Takasago...). Details of the project and information about the texts and how they should and should not be used are given on the hanrei page. In a nutshell: free to use but not to sell, best used for SEARCHING rather than reading.

To display the texts correctly, you may have change the encoding manually to “Japanese (EUC-JP).” Mac users should use the Firefox browser rather than Safari.

The entire corpus can be searched using a single page: with patience you can copy it all to a single word processor file but it comes to over 2000 pages in Word. If you are patient, it is better to download the plays as individual files and search them locally on your hard disk, using GREP software. Or to put them in software that allows you to search all plays at once, but makes it clear in which plays the search string can be found. (I have been using HyperCard for Mac 9 and am now experimenting with DEVONThink for Mac OSX. I would welcome other ideas for appropriate software for any platform.)

(5) Print concordance

Ōtani Tokuzō 大谷篤蔵 , ed. Yōkyoku nihyakubanshū sakuin 謡曲二百五十番集索引 (vol. 1), Yōkyoku nihyakubanshū 謡曲二百五十番 (vol. 2). Kaidai sakuin sōkan 6. (Akaoshōbundō 赤尾照文堂 , 1988).

Concordance in one volume, with play/page/ dan reference to the unannotated text of 250 plays in the second volume. The text also contains small black-and-white photographs of performances. For reference rather than reading (the text of the plays is also somewhat muddy), but extremely useful. The concordance is well arranged and clearly laid out. Phrases are broken down by context: the entry for hajime yori (for example) gives seven phrases with initial or medial use of hajime yori .

Michael Watson <>

  Back to full Agenda