INTRODUCTION: SHUIHU ZHUAN IN EARLY MODERN JAPAN
To say that the Chinese vernacular novel Shuihu zhuan 水滸傳 (Jp. Suikoden) was popular in Edo-period Japan would be an exercise in understatement. “Near-ubiquitous” is a more accurate assessment of the novel's influence on the literature and visual culture of Japan, particularly in the second half of the Edo period. This omnipresence, however, potentially occludes the fact that the novel was imported into Japan over one hundred years prior to its translation into Japanese in the second half of the eighteenth century. The colloquial language of the novel presented enormous difficulties for Japanese readers raised on classical Chinese texts, and originally, the novel was largely the preserve of a much smaller group of sinologists, professional translators, and connoisseurs of fiction capable of making their way through the difficult Chinese vernacular. One of the most intriguing developments in the history of Chinese vernacular fiction in Japan was the formation of study groups devoted to the close reading and explication of specific texts—an institution apparently inaugurated by the “Translation Society” (Yakusha 譯社) established by the classical scholar Ogyū Sorai 荻生徂徠 (1666–1728) in 1711.Footnote 1 Records from a number of such groups have been preserved, providing oblique insight into the hermeneutic approaches underlying the Japanese reception of Chinese narrative. This study focuses on the critical activities of Seita Tansō 清田儋叟 (1719–1785): a Kyoto-based historian and classical scholar, who lectured extensively on both Shuihu zhuan and its most famous Chinese commentator, Jin Shengtan 金聖歎 (1608–1661). Tansō's lecture notes, entitled Suikoden hihyōkai 水滸伝批評解 (A Discussion of “The Water Margin” and Its Commentary) provide a unique opportunity to analyze the Japanese interpretive strategies brought to bear on works of Chinese vernacular fiction. As the title of Tansō's commentary suggests, his analysis of Shuihu zhuan is not limited to the text of the novel itself, but extends to the fiction criticism of its most prolific commentator. Jin Shengtan's seventy-chapter edition of Shuihu zhuan presented a sophisticated method of reading the novel through the inclusion of prefatorial essays and copious interlineal commentary. Most editions of fiction and drama printed in late imperial China included some form of critical commentary, and scholars of Chinese fiction have increasingly noted the centrality of this material to the late imperial reading experience.Footnote 2
Through the example of Seita Tansō, I argue that the impact of Chinese critical commentaries must be addressed when discussing the reception of Chinese vernacular novels in Edo-period Japan as well. Novels and works of drama imported into Japan during this time frequently included Chinese commentaries, and even a cursory look at the Hakusai shomoku 舶載書目 reveals that different commentarial editions of the same novel were distinguished from each other in importation records.Footnote 3 Borrowing from Gérard Genette's discussion of prefaces, inscriptions, epigraphs, and other forms of paratexts, I interpret the use of these commentaries as both a way of presenting and “making present” foreign works of literature in a drastically different cultural and literary context.Footnote 4 Japanese connoisseurs of Chinese fiction were highly partisan in their selection and endorsement of particular editions of Chinese novels. Tansō, for example, accepted Jin Shengtan's Shuihu zhuan as the authentic version of the novel, and he was careful to justify his choice by demonstrating his familiarity with other commentaries available in Japan. As discussed earlier, enjoyment of vernacular novels like Shuihu zhuan was initially restricted to readers capable of making sense of the narrative's difficult language and unique syntax structures. Earlier guides to the novel focused primarily on the precise explication of exotic vocabulary and stock phrases common to Chinese vernacular fiction—approaching the novel less as a work of literature per se, than as a repository of philological data and information about contemporary China. According to the research of Takashima Toshio 高島俊男, it was not until the middle of the eighteenth century that deeper engagement with the themes and literary structure of Shuihu zhuan emerged in Japan: a development most likely related to the publication of a partial translation of the novel in 1757.Footnote 5
Tansō's guide constitutes a singular point of entry in terms of examining a poorly understood but pivotal era in the Japanese reception of Shuihu zhuan. In contrast to the earlier tradition of philological exegesis, Tansō followed Jin Shengtan's lead by mounting a sophisticated inquiry into the novel's narrative structure and thematic concerns. Tansō's extended exploration of Shuihu zhuan is predicated on the assumption that the pronouncements of Chinese critics like Jin were worth taking seriously, and my discussion of Tansō situates his critical activities in the larger history of transnational literary contact that has been examined by Atsuko Ueda, Emanuel Pastreich, Peter Kornicki, Atsuko Sakaki, and others. Instead of simply viewing the novel as a wellspring of linguistic data, Tansō attempted to inculcate a sensitivity to the aesthetic contours and ethical implications of a complex literary text. Tansō did not simply parrot Jin Shengtan's interpretation of the novel. Instead, he engaged both the body of the novel and Jin's pronouncements in a critical and systematic manner by alternately creatively appropriating and rejecting Jin's analysis as he deemed appropriate. The analysis generated from this critical engagement resulted not only in a highly original reading of Shuihu zhuan itself, but also in more widely applicable observations about narrative that could be applied to Japanese texts as well. In terms of emplotting a larger history of Japanese narratology, Suikoden hihyōkai should be recognized as a forerunner to a larger comparative discourse interrogating the relationship between Chinese and Japanese literary production—a text that assists in contextualizing the far better known projects of late Edo- and Meiji-period critics such as Kyokutei Bakin 曲亭馬琴 (1767–1848), Hagiwara Hiromichi 萩原広道 (1815–1863), and Tsubouchi Shōyō 坪内逍遥 (1859–1935), among many others.Footnote 6
SHUIHU ZHUAN AND THE MANY MEANINGS OF MARGINALITY
A largely overlooked contributor to the dissemination and popularization of vernacular Chinese narrative in eighteenth-century Japan, Seita Tansō hailed from a prestigious line of literati. His father, Itō Ryūshū 伊藤竜洲 (1683–1755), served as a scholar in the service of the daimyo of Echizen, and his older brother is far better known under his adopted name, Emura Hokkai 江村北海 (1713–1788): author of Nihon shishi 日本詩史 (A History of Japanese Kanshi Poetry), and a noted poet and arbiter of literary taste.Footnote 7 As a youth, Tansō corresponded with Yanada Zeigan 梁田蛻巌 (1672–1757), a famous kanshi poet and scholar living in Akashi. At some point, Tansō sent Zeigan an original composition entitled Okuni den 阿国伝. The subject matter and format of the work are unclear from Zeigan's response, but the hermeneutic he proposed is worth noting:
I received your Okuni den, and it is clear that half a lifetime's worth of energy went into it. Even before I was finished reading, I knew that it was the harbinger of a long-lasting literary elegance. Later, I'll draw up a commentary—sweeping aside everything else on the desk and lighting incense to summon [the Chinese fiction commentators] Jin Shengtan and Wang Wangru. They will discourse on its merits and explain it for all the talents in the realm.
又承足下阿國傳, 一生精力, 半在茲編, 未讀已, 知其為千百年風流嚆矢也, 異日亦使老居士評之, 則掃案熱香, 乃現聖歎望如身, 而為六十六州錦繡才子説法矣.Footnote 8
This playful promise to summon the Chinese fiction commentators Jin Shengtan and Wang Wangru demonstrates Zeigan's awareness of the tradition of Shuihu zhuan commentary, and suggests that he relied on this criticism in the interpretation of Chinese novels.Footnote 9
Despite his lifelong interest in Chinese vernacular fiction, Tansō made it clear in his writing that his paramount concern was history and historiography. Like many scholars of his generation, he was deeply interested in post-classical Chinese history, and in 1769, he published a Japanese-language translation of the Korean scholar Ch'oe Pu's 崔溥 (1454–1504) castaway narrative Kŭmnam p'yohae rok 錦南漂海録 (Record of My Time Adrift)—a text that remains an important source of information about Ming society and culture. The publication of Ch'oe Pu's account was indicative of a widespread desire to learn more about a China less removed in time than the archaic texts comprising a classical education. One source of information was the ever-increasing number of Chinese novels and short stories available in either Japanese translation or kundoku editions. The Kamigata literati Oka Hakku 岡白駒 (1692–1767) and Sawada Issai 沢田一斎 (1701–1782) published three sets of annotated Chinese stories in the 1740s and 1750s, and Tansō appears to have assisted in the preparation of a kundoku edition of the short story collection Zhaoshi bei 照世杯 (A Cup for Reflecting the World, Jp. Shōseihai) in 1765. Although Tansō's name did not appear on the kundoku Zhaoshi bei, his involvement has been assumed on the basis of his nom de plume Kujaku dōjin 孔雀道人 (The Peacock Adept) in the text's preface: a remarkable essay entitled “The Three Precepts for Reading the Vernacular” 読俗文三條 (Doku zokubun sanjō).Footnote 10
To my knowledge, “The Three Precepts” constitutes one of the earliest Japanese attempts at elucidating a systematic, non-linguistic rationale for engagement with Chinese vernacular fiction.Footnote 11 The text itself is illustrative of a more general transition from philological interest in Chinese vernacular texts to more figurative readings that looked for deeper narratological structures beyond the surface of the text. The first half of the essay focuses on the difficulties inherent in the study of the Chinese vernacular language and the potential dangers of reliance on Japanese kundoku as an interpretive crutch: both themes that were familiar to the point of cliché by the middle of the eighteenth century. In a passage that could easily have been penned by Ogyū Sorai or his contemporary, Itō Tōgai 伊藤東涯 (1670–1733) fifty years earlier, Tansō warned:
[The Chinese term] ganji can be translated as either “gratitude,” or “indebted.” The character yang can be translated as “to request,” and it can be translated as “to hire.” Laoshi can be “earnest and sincere,” or it can be translated as “docile and compliant.” Through raising just a few examples, you should be able to infer thousands more! But be certain of this: if you remain inflexible in your translation and are constrained by glosses, then there are many instances in which you will lose the central essence of the passage in question.
感激を有難しとも辱なしとも譯し, 央をたのむとも, やとうとも譯し, 老実をじつていとも,おとなしくとも譯す, 二三を舉て千万を推て知るべし, 必板定の譯をなし, 字訓に拘れば,本義を失う事多し.Footnote 12
Tansō's statement echoes Sorai's earlier insistence in Yakubun sentei 訳文筌蹄 that Japanese kundoku should best be interpreted as a form of translation or interpretation, and in the context of the mid-eighteenth century, the remark is uncontroversial to the point of being anodyne.Footnote 13 It is not until the second half of the “Three Precepts” that Tansō swerves from familiar terrain and provides a tantalizing glimpse of a theory he would develop in more detail in his other literary criticism. Although the text under scrutiny was ostensibly Zhaoshi bei, Tansō's preface raised the far better known example of Shuihu zhuan to argue that deeper meaning might be concealed beneath the surface of a fictional text:
It is true that there are many works written in the vernacular, but if you truly become conversant with Shuihu zhuan, then approaching the others is as easy as crushing bamboo. The reason for this is that, in the case of Shuihu zhuan, there is a “real water margin” beyond the surface of the text. In fact, the entire history of the Song dynasty is folded into the novel. [The bandit leader] Chao Gai is really [the Song emperor] Taizu. [Chao Gai's successor] Song Jiang is really emperor Song Taizong. [The bandit strategist] Wu Yong is really [the Song minister] Zhao Pu, Guan Sheng is [the southern Song general] Wei Sheng … and so on. Thus, we see that Shuihu zhuan can be a great help in reading the official histories.
俗文の書多しといえども, 水滸伝を能く取てまわせば, 其餘は破竹の勢ぞ. いかにといへば, 水滸は文面の外に真の水滸あり, 宋一代を一部中へ収入す, 晁蓋は太祖, 宋江は太宗, 呉用は趙普, 關勝は魏勝 . . . などの類は, 大に正史の助をなす.Footnote 14
While an uninitiated reader will merely see the novel Shuihu zhuan as an exciting tale of heroes and bandits, Tansō claims, a sophisticated reader recognizes that the story is actually an extended allegory describing the foundation and history of the Song dynasty (960–1279 CE). Tansō's interest in the “real” water margin to be glimpsed through a close reading of the novel provided a bedrock for his later discussion of Shuihu zhuan, which will be addressed below. For now, I would simply like to draw attention to his insistence upon a deeper significance to Shuihu zhuan and Chinese fiction in general. Tansō disagreed with Jin Shengtan on many issues, but as I will demonstrate, he took a cue from his Chinese predecessor by emphasizing the need to account for hidden themes and symbols not immediately apparent to superficial readers. This conception of Shuihu zhuan downplayed earlier Japanese interest in fiction as fodder for linguistic study, and substituted it with a far more literary focus on the language, imagery, and narrative themes of Chinese vernacular fiction.
Shuihu zhuan was a text that Tansō returned to repeatedly in his lifetime, and the text's title surfaces in a number of his extant works. Although clearly familiar with the many editions of Shuihu zhuan circulating in Japan by the middle of the eighteenth century, he was particular in his selection of recensions—taking Jin Shengtan's seventy-chapter edition of the novel as his base, and accepting many of Jin's pronouncements as orthodox. In order to discuss Tansō's interest in Shuihu zhuan, it is necessary to first introduce both the edition he was working with and the interpretations advanced by its iconoclastic commentator. A native of Suzhou, Jin Shengtan is best known in the present for his intense devotion to the marginalized field of vernacular fiction and drama. Shuihu zhuan was a work of particular importance to Jin, and he signaled his attachment to the novel by listing it alongside five other “works for men of genius” 才子書 (Ch. caizi shu): the poet Qu Yuan's 屈原 (ca. 340–278 BCE) Li Sao 離騷 (Encountering Sorrow), the Zhuangzi 莊子, Sima Qian's 司馬遷 (ca. 145–86 BCE) Shiji 史記, the poetry of the Tang poet Du Fu 杜甫 (712–770 CE), and the thirteenth-century dramatic work Xixiang ji 西廂記 (Romance of the Western Chamber). As a young man, Jin reached xiucai 秀才 status, the lowest of the three degrees in the imperial examination system, but he does not appear to have sat for more advanced degrees.Footnote 15 The absence of official responsibilities freed up time for Jin's preferred pastime of reading and explicating a wide range of literary texts. His annotated edition of Shuihu zhuan, the third preface of which is dated 1641, can fairly be considered his magnum opus. The edition he prepared contained five fascicles of introductory material, including three prefaces by Jin himself, a selection of excerpts from historical texts, a general “reading guide” to the novel (Ch. dufa 讀法), and a clearly spurious preface attributed to the novel's putative author, Shi Nai'an 施耐庵.Footnote 16 The bulk of Jin's textual analysis appeared in the body of the novel itself, where Jin copiously annotated each chapter through interlineal, marginal, and prechapter comments. Earlier editions of Shuihu zhuan had included this kind of paratextual commentary as well, but Jin's meticulous attention to detail and sheer garrulousness were unmatched in his day.
In addition to his commentary, Jin took editorial liberties with the body of the novel itself. He excised the vast majority of the text's poetry and parallel prose, and he rewrote a number of passages—often in ways that accorded with the points raised in his commentary. He justified these emendations by claiming that his edition of the novel was an “ancient edition” 古本 (Ch. guben) faithfully preserving the authorial intention of Shi Nai'an. He named his edition the Guanhuatang 貫華堂 edition after the studio in which it was allegedly discovered. This discovery of an ancient edition of Shuihu zhaun was patently untrue, but Jin stuck to his story. Throughout his commentary, he elevated his “original edition” through the castigation of various other “vulgar editions” 俗本 (Ch. suben) in circulation.
Jin's largest innovation was the excision of the second half of the novel. The longer recensions of Shuihu zhuan related two major stories. The first seventy-one chapters chronicled the individual adventures of the 108 bandit-heroes and described their gradual assemblage in the Liangshan marshes. In the second half of the novel, the rebellious bandits are pardoned by the Song emperor and make amends for their crimes by going to war with other groups of rebels and invaders. Jin's edition excised this second half and ended the novel on a note of suspense, in an ending clearly written by Jin himself. One of the bandits, the newly arrived “Jade Unicorn” Lu Junyi 玉麒麟盧俊義, has a portentous dream in which the Liangshan gang is put to death in a grisly execution scene—an ending meant to suggest that the bandits will ultimately be punished for their crimes. Jin justified this dramatic ending by claiming that it was the conclusion Shi Nai'an had originally written. The longer narrative, in which the bandits were pardoned by the Song court, Jin told his readers, was nothing more than an inferior sequel by the Yuan-dynasty playwright Luo Guanzhong 羅貫中 (ca. 1330–1400).Footnote 17 By excising the chapters in which the bandits are pardoned and providing Shi Nai'an's “original ending,” Jin insisted that he was restoring the text to a pristine condition.
Jin's excision of the second half of the novel was rooted in both artistic and thematic concerns. Of the many aspects of earlier editions he found objectionable, nothing sparked his ire more than his worry that readers might find the bandits sympathetic. The dream sequence Jin crafted for his ending made it clear that the rebels would be punished for their crimes, and he was highly critical of readers who read Shuihu zhuan as a tale of unappreciated valor. The Guanhuatang edition's full title, The Fifth Book for Men of Genius: Shi Nai'an's “The Water Margin” 第五才子書施耐庵水滸傳, was itself a polemical attempt at rebuking publishers who included the terms “loyal” 忠 (Ch. zhong) and “righteous” 義 (Ch. yi) in the title of the text. One such text was the 1610 100-chapter edition of the novel prepared by the Rongyutang 容與堂 publishing house. The text's full title was The Loyal and Righteous Chronicle of the Water Margin with Commentary by Mr. Li Zhuowu 李卓吾先生批評忠義水滸傳. As promised, the edition included a preface by the late Ming philosopher Li Zhuowu 李卓吾 (1527–1602), who discoursed at length on the meaning of the novel's title:
The Grand Historian Sima Qian wrote: “[The Warring States philosopher Han Feizi's] ‘Speaking of Difficulties’ and ‘Solitary Frustration’ were composed by a virtuous man who sought to express his righteous anger.” Looking at this, we see that in antiquity, men of virtue did not compose literary works unless they were overcome by indignation and frustration. If they were not frustrated and indignant but still wrote something, then this was like shivering without being cold, or groaning without being sick. You can do it, but who would ever watch such a thing! Shuihu zhuan is also a work produced by the author's indignation. The Song imperial line's decline meant that honors were given indiscriminately. Men of great virtue were forced to the bottom, and the unworthy rose to the top. Soon, barbarian invaders were in control, and the residents of the Central Plains were under their yoke.… Who were the ones venting their grief? The gallants who banded together in the water margins in former days. It is impossible not to call them men of loyalty and righteousness. For this reason, the authors Shi Nai'an and Luo Guanzhong called their work Chronicle of the Water Margin and capped it off with the words “loyal” and “righteous.”
太史公曰,說難孤憤賢聖發憤之所作也,由此觀之,古之賢聖不憤則不作矣,不憤而作譬如不寒而顫,不病而呻吟也,雖作何觀乎,水滸傳者,發憤之所作也,蓋自宋室不競冠屨倒施,大賢處下,不肖處上,馴致夷狄處上,中原處下 … 敢問洩憤者誰乎,則前日嘯聚水滸之強人也,欲不謂之忠義不可也,是故施羅二公傳水滸,而復以忠義名其傳焉.Footnote 18
By cutting off the narrative after the assemblage of the bandits and subjecting them to oneiric execution, Jin Shengtan denied the Liangshan rebels a chance at posthumous redemption. In the second preface to the Guanhuatang edition, Jin made his intentions explicit by directly attacking the proposals raised in Li Zhuowu's essay:
When one is observing phenomena, he investigates names; when he is discussing people's character, he distinguishes between intentions. Shi Nai'an chronicled the exploits of the bandit Song Jiang, and entitled his work The Water Margin. This is because he hated Song Jiang in the extreme, dismissed him, and did not want him sharing the Middle Kingdom. Later generations didn't understand that Song Jiang was a sower of discord and mistakenly affixed “loyal and righteous” to the title of the tale. Alas! Can there ever be such a thing as loyalty and righteousness in a place like the water margin? … What Shi Nai'an means by “the water margin” is this: the fringes of the empire are bounded by water, and what is beyond the water is called the “margin.” It is a term of distance. Those who are distanced in such ways are the loathsome creatures of the world: whom everyone in the realm should attack! They are the nauseating creatures of the world: whom everyone in the realm should forsake! If we say that loyalty and righteousness are in the water margin, then we are saying that loyalty and righteousness are loathsome, nauseating things. Moreover, if loyalty and righteousness are in the margins, then does that mean that the realm has none?
觀物者審名,論人者辨志,施耐庵傳宋江,而題其書曰水滸,惡之至,迸之至,不與同中國也,而後世不知何等好亂之徒,乃謬加以忠義之目,嗚呼,忠義而在水滸乎哉 … 若夫耐庵所云,水滸也者,王土之濱則有水,又在水外則曰滸,遠之也,遠之也者,天下之凶物,天下之所共擊也,天下之惡物,天下之所共棄也,若使忠義而在水滸,忠義為天下之凶物惡物乎哉,且水滸有忠義,國家無忠義耶.Footnote 19
Jin Shengtan's claims about the origin of the Guanhuatang Shuihu zhuan strain the modern reader's credulity, but he was a perceptive reader whose edition established a new standard for fiction criticism in late imperial China.Footnote 20 Despite his professed distaste toward the bandits' actions, Jin frequently distinguished between wen 文 (the style of writing) and shi 事 (the events related in the text)—separating the author's description of a particular event from his endorsement. For instance, in a grotesque scene in Chapter 54, the bandit Li Kui 李逵 murders a small child in order to convince his ward, Zhu Tong 朱仝, to join the gang. Jin shifted attention away from Li's shameful action by rebuking sentimental readers who conflated text and authorial intention:
Those who read to this point and sigh in consternation are fools. This is nothing more than a remarkable piece of writing by Shi Nai'an; how could something like this really happen?
讀至此句,失聲一歎者,癡也,此自耐庵奇文耳,豈真有此事哉.Footnote 21
In truth, Jin's most profound innovations in the field of narratology were necessitated by the radical nature of his claims. In crafting a new ending that unambiguously condemned the actions of the Liangshan bandits, Jin was left with the task of justifying the preceding sixty-nine chapters: a narrative in which many of the bandits are presented in a positive light. Some of the bandits are clearly the victims of government malfeasance, and as the Li Zhuowu preface suggests, many readers sympathized with their actions. An alternative to acknowledging the subversive expectations of the reading public was the assertion of subtly elucidated messages by the author, discernible only to an educated few. The best example of this interpretation is Jin's treatment of the bandit leader, Song Jiang. In contrast to his predecessors’ fulsome praise of the rebel leader, Jin heaped abuse on Song with a ferocity bordering on the self-parodic. It was true, Jin acknowledged in his commentary, that Song Jiang was frequently presented as a well-liked and heroic leader. By paying close attention to the language and structural principles of the novel as a whole, however, a perceptive reader would see that Shi Nai'an in fact detested Song Jiang and expressed his dislike through subtle clues. The fact that Song Jiang was often praised by his fellow bandits (and theoretically, by the implied author as well) was nothing more than a red herring. One must learn, to return at last to Seita Tansō's memorable phrase, to seek for the “real water margin” beyond the surface of the text. In Jin's words:
In this novel, it was simple to write about 107 of the heroes. It was writing about Song Jiang that was the most difficult. Therefore, when reading the novel, it is a simple matter to read about 107 of the heroes. Again, it is interpreting Song Jiang that is the most difficult. This is because Shi Nai'an uses “direct brushwork” to flesh out 107 of the heroes. If they are good, then he represents them as being good; if they are inferior, then he represents them as such. But when he came to Song Jiang, this isn't the case. If you rush through the text, then it seems like he's entirely good. Read it a second time, and it seems like he's about half good and half bad. Read it one more time, and the bad aspects outweigh the good. And the last time you read it, you see that he's entirely rotten with nothing to redeem him. And so, if someone reads Song Jiang's biography once, twice, three times, and once more again—finally realizing that he's all bad and no good at all—we can call this person a good reader!
一部書中,寫一百七人最易,寫宋江最難,故讀此一部書者,亦讀一百七人傳最易,讀宋江傳最難也,蓋此書寫一百七人處,皆直筆也,好即真好劣即真劣,若寫宋江則不然,驟讀之而全好,再讀之而好劣相半,又再讀之而好不勝劣,又卒讀之而全劣無好矣,夫讀宋江一傳,而至于再,而至于又再,而至于又卒,而誠有以知其全劣無好,可不謂之善讀書人哉.Footnote 22
This desire to train sensitive readers permeated the copious interlineal commentary and prefatorial essays in Jin's edition of the text. The rhetorical justification for the annotation stemmed from traditional concerns (the punishment of vice and allotment of historical blame), but the end result of Jin's labors was the enunciation of a sophisticated set of critical concepts and terminology—a complex mode of literary analysis focused on the recovery of authorial intention through close attention to textual detail.
SEITA TANSŌ AND THE GUANHUATANG EDITION IN JAPAN
This introduction to the Guanhuatang Shuihu zhuan is important to my discussion of the ways in which Seita Tansō approached both the novel and its extensive commentary. Tansō is a minor figure in Japanese literary history, but his surviving corpus provides a rare opportunity to reconstruct the interpretive strategies employed by an educated Edo-period reader. Tansō's writing on Shuihu zhuan is a fascinating work of Sino-Japanese literary engagement: both an analysis of a tremendously important novel and a judgment of the hermeneutic principles advanced by its most prolific Chinese commentator. Tansō admired Jin's powers of literary observation and agreed with many of his assertions, but he was intensely critical of his explication as well. Tansō's discussion of the novel represents an ambitious and original attempt at subjecting Chinese fictional texts to rigorous formal analysis—an attempt that builds on Chinese commentaries but engages them critically in ways that often move against precedent.
Tansō's understanding of Shuihu zhuan can be reconstructed through an analysis of three main sources: entries in his occasional notes Kujakurō hikki 孔雀楼筆記 (Notes from the Peacock Tower); the marginal comments inscribed in his personal copy of Shuihu zhuan (preserved in the library of Tokyo University's Institute for Advanced Studies on Asia [formerly the Institute of Oriental Culture]); and most importantly, a set of lecture notes entitled Suikoden hihyōkai. Suikoden hihyōkai is a lengthy, 300-page manuscript dealing with the terminology, structure, and proper interpretation of Jin Shengtan's edition of the novel. The manuscript is a student's transcription of Tansō's comments, and like many other vernacular dictionaries and reading guides compiled during the Edo period, the explication was never published. Two of the comments are dated to the mid-Meiwa period—1768 and 1769, respectively—providing a general idea of when Tansō's discussion of Shuihu zhuan took place. Suikoden hihyōkai is structured as a tripartite analysis made up of direct quotations from the Guanhuatang Shuihu zhuan (both the novel and Jin's comments), Tansō's original analysis, and clarification of that analysis by Tansō's transcriber: a certain Takada Jun 高田潤 from Hizen. Short phrases from the Guanhuatang Shuihu zhuan and Jin's commentary are quoted at the top of each page. Tansō's comments, written in kanbun, are provided next, indented below the level of the quoted text. These comments are generally terse, and rarely exceed ten characters or so in length. Takada Jun fleshes out these brief remarks in glosses that provide context for Tansō's remarks and generally clarify his pronouncements.
Although he was skeptical of many of Jin's arguments, Tansō appears to have accepted the Chinese commentator's claims about the authenticity of the Guanhuatang edition. In Kujakurō hikki, for instance, Tansō presented Shi Nai'an as the author of the original text:
The seventy chapters of Shuihu zhuan were written by Shi Nai'an. Luo Guanzhong wrote a continuation of fifty chapters, which brings the novel up to a total of 120 chapters.
水滸伝七十回, 施耐庵作ル, 羅貫中五十回ヲ続ナシ, 合テ百二十回トナス.Footnote 23
Similarly, throughout Suikoden hihyōkai, Tansō adopts Jin's custom of referring to the seventy-chapter Guanhuatang edition as the “ancient edition” 古本 (kohon) distinct from other “vulgar editions” 俗本 (zokuhon) in circulation. Tansō also agreed with Jin that the moral implications of the novel were fundamentally disquieting. Despite his considerable enthusiasm toward the novel's writing and structure, Jin had declared Shuihu zhuan “unworthy of use for [moral] instruction” 不可以訓,Footnote 24 and Tansō imported this comment directly in his first remark in Suikoden hihyōkai:
Shuihu zhuan is a popular novel. However, it speaks only of deception, treachery, and ruthless scheming. It is not a work that can be used for instruction.
水滸傳者,通俗之書,旦說詐偽機謀,不可以訓.Footnote 25
Demonstrating Tansō's awareness of the Rongyutang edition and critics who mistook the Liangshan bandits for loyal vassals, Takada Jun explained: “[Tansō's comment] was in response to Mr. Li [Zhuowu]'s titling the novel The Loyal and Righteous Water Margin” 李氏ガ忠義水滸伝ト云ニアタッテ云タルモノゾ.Footnote 26
With lip service to the dangers of reading Shuihu zhuan out of the way, both critics were free to engage in precisely that activity. Despite commonalities in their general rhetorical orientation toward the novel, Tansō immediately departed from Jin's analysis by firmly connecting the novel to his own interests in historiography. In his earlier preface to Zhaoshi bei, Tansō had described Chinese fiction as a complement (tasuke) to official historiography, and in a marginal note inscribed in his personal copy of Shuihu zhuan, he critiqued less-educated readers for lacking the historical training necessary for its comprehension:
The text Shuihu zhuan is of enormous scope. It gathers together three hundred years of the history of the Song, its lords, and its ministers without letting anything slip by. If one is not deeply conversant in the principles of historiography, then he will be unable to understand its outline.
水滸一書,關係甚廣,趙宋三百年君臣事跡,蒐集無遺,非深於史學者,無能得其綱領.Footnote 27
In Suikoden hihyōkai, Tansō built on his earlier assertion that Shuihu zhuan was a work of historical allegory in which different characters in the novel should be taken to represent figures at the Song court. This professed interest in the historical narrative embedded in the novel functioned as Tansō's central hermeneutic principle and justified his interest in the unorthodox and marginal field of Chinese vernacular fiction. For instance, Tansō insisted throughout his commentary that perceptive and well-trained readers would recognize the bandit leader Chao Gai 晁蓋 and his successor Song Jiang as fictional stand-ins for the Song emperors Taizu 太祖 (r. 960–976) and Taizong 太宗 (r. 976–997). Lest the identification appear far-fetched to the reader, Tansō showed how close attention to textual detail uncovered a key to the hidden narrative:
Seita Tansō: [The original bandit chief] Chao Gai is [Song] Taizu.
Takada Jun: Song Taizu's surname was Zhao. The surname Chao is pronounced the same as the character for “wave” (chao). “Zhao” is pronounced with ascending tone, and thus it is different from “Chao,” but the sound is absolutely identical. The term gai refers to something that covers over the realm. Taizu was the first lord of the Song dynasty, and so he was on top and akin to the lid on a vessel. Thus, the character Chao Gai is a veiled reference to Taizu.
Tansō: Song Jiang is Taizong.
Jun: Song Taizu established the great Song. [The next emperor] Taizong assassinated him and usurped the realm. Chao Gai made his abode in the marshes of Mt. Liangshan and was murdered by Song Jiang, who assumed his position.Footnote 28 Thus, we see that the two instances are very similar.
晁蓋之為太祖Footnote 29
宋ノ太祖,姓ハ趙ソ, 晁ハ音潮,趙ノ上声, 声ハ異ナレドモ, 音ハ全ク同シ, 蓋ハ天下ヲ掩蓋スルノ意ゾ, 太祖ハ宋ノ第一主ナルハ, 上ニ有テ, 器物ノフタニ比ス, 因テ晁蓋ハ太祖ヲ影写シタルトイウ.
宋江之為太宗
太祖宋ノ業ヲナス, 太宗コレヲ弑シ, ソノ天下ヲ奪フ, 晁蓋梁山泊ニ拠ル, 宋江コレヲ弑シ,ソノ交掎ニ坐スニ, 事相類ス.Footnote 30
A number of other characters are listed and identified as fictional avatars of figures at the Song court, before Tansō breaks off by writing that “the rest may be inferred” 可以類推其他 through close reading.
This allegorical interpretation departed dramatically from Jin Shengtan's own interpretation, but Tansō did not specify whether his reading was original or based on Chinese precedent. One possibility is that the interpretation was suggested by the characters of the novel themselves. In Chapter 40, during a gathering of the bandits, Li Kui exclaims:
“With all our cavalry and armaments, why should we be afraid of rebelling? Our brother Chao Gai will be ‘Senior Emperor of the Song,’ and our brother Song Jiang can be ‘Junior Emperor of the Song’!”
放着我們許多軍,便造反怕怎地,晁蓋哥哥便做大宋皇帝,宋江哥哥便做小宋皇帝.
Jin Shengtan merely chuckled at the irrepressible Li's comment, but the passage was dotted (the premodern equivalent of underlining) in Tansō's personal copy of the novel. It is tempting to see a particularly Edo-period fascination with historical camouflage at work in Tansō's analysis. The interpretation of fiction and drama as a masked reference to historical events would have been second nature to Tansō and his peers—all of whom faced prohibitions on the representation of contemporaneous events and would have been skilled at encoding and decoding subtle references. At roughly the same time Tansō was commenting on Shuihu zhuan, for instance, the jōruri play Kana dehon Chūshingura 仮名手本忠臣蔵 was being viewed and appreciated with similar double vision. With the large number of extant Chinese Shuihu zhuan glosses, it is highly possible that Tansō was building on earlier Chinese interpretations, but without information about sources, this must remain speculative.Footnote 31
Jin Shengtan and Li Zhuowu were perhaps the most famous commentators on Shuihu zhuan, but they were by no means the only ones. In the opening pages of Suikoden hihyōkai, Tansō bolstered his interpretive credentials by demonstrating his familiarity with other critics who had written on Shuihu zhuan in recent years:
Hu Yuanrui and Xie Zaihang have also touched on Shuihu zhuan in their writing many times. Yuanrui writes that in recent years there was a scholar who kept two books on his study-table: on the right, a copy of Nanhuajing [the Zhuangzi], and on the left, a copy of Shuihu zhuan.
胡元瑞謝在杭亦頗言及水滸傳,元瑞亦曰,近時一名公,書几右南華,左水滸.Footnote 32
The figures mentioned are the famous late-Ming literati Hu Yinglin 胡應麟 (1551–1602) and Xie Zhaozhe 謝肇淛 (1567–1624). Both Hu and Xie are known for championing the creative merit of vernacular fiction and classical tales, and in the words of Allan H. Barr, “for challenging the conventional assumption that narrative should record only what is known or believed to have happened.”Footnote 33 The anecdote Tansō recounts comes from Hu's miscellany Shaoshishanfang bicong 少室山房筆叢 (Collected Jottings from the Shaoshi Mountain Retreat), in which Hu tells the story as an example of Shuihu-mania in sixteenth-century China.Footnote 34 In the original text, Hu makes it clear that the tale was recounted “in jest” 滑稽 (Ch. huaji), but Tansō does not, or refuses, to get the joke. Takada Jun's explication—“This anecdote demonstrates the degree to which Shuihu zhuan is valued [in China]” 以上一段水滸伝ノ貴重セラルヲイウ—suggests that Tansō has misunderstood or even intentionally misrepresented the novel's prestige in contemporaneous China.
Tansō's allegorical interpretation of Shuihu zhuan further contradicted the other influential Chinese theorist named. Xie Zhaozhe's Wu za zu 五雜俎 (Five Assorted Offerings) provided support for fiction writers who wished to liberate their craft from the strictures of historical accuracy. As a connoisseur of Chinese-language fiction, Tansō was certainly aware of the following passage by Xie:
Now, in writing fiction and drama, historical truth and imaginative fancy should be equally present before a work can be called truly entertaining … There's no need to inquire as to whether such-and-such an event really occurred … [When plays are written,] authors feel compelled to go through each and every event so that dates and names accord to the dynastic histories. If they don't match up, they won't write it. If we approach drama like this, then just reading the dynastic histories should be enough—why even call it a “play”?
凡為小說及雜劇戲文,須是虛實相半,方為游戲三昧之筆 … 不必問其有無也 … 必事事考之正史,年月不合,姓字不同,不敢作也,如此則看史傳足矣,何名為戲.Footnote 35
Xie's text was well known in eighteenth-century Japan, and his discussion of the narrative's relation to the “actual” 實 and “fictional” 虛 (Ch. shi, xu; Jp. jitsu, kyo) found its way into Japanese critical writing on fiction and drama.Footnote 36 Although Tansō was clearly aware of Xie's corpus, there is no deeper analysis of his arguments about the nature of fiction in Suikoden hihyōkai. Indeed, Tansō seemed more interested in borrowing the names of Hu Yinglin and Xie Zhaozhe to promote Shuihu zhuan as cultural capital than engaging their theories of narrative in any depth. What is important for the purposes of this study, however, is Tansō's clear awareness of multiple commentaries to Shuihu zhuan: an indication of the degree to which eighteenth-century Japanese readers considered Chinese criticism in their readings, as well as a demonstration of Tansō's willingness to challenge or ignore Chinese interpretive precedent.
Although his allegorical reading provided a frame for analysis, the bulk of Suikoden hihyōkai focused on the aesthetic and literary structure of the novel itself by noting, often in critical terms very much indebted to Jin Shengtan, passages that were particularly well written, turns of phrase that clearly demonstrated the structural principles at work in Shi Nai'an's composition, and deceptively complex sections that even Jin had misinterpreted. A modern reader might fault Tansō for being overly concerned with minutiae, but that is precisely the justification for his work: an illustration of how subtle textual details contribute to larger structures of meaning. Just as Jin Shengtan's formal criticism was justified by his desire to recover the original authorial intention underlying the novel, Tansō's remarks on the structural features of the novel can be interpreted as similarly necessary in training readers to be more sensitive to hidden meanings.
The anxiety of influence plays a large role in Tansō's commentary, and Tansō alternates between enthusiastic endorsement and withering disdain of his predecessor, Jin Shengtan. At many points, Tansō appears to be trying to channel Jin's own unique commentarial voice. We see Jin's brash self-confidence reflected in the Japanese text when Tansō voices his approval by writing “[Jin's] comment hits the mark!” 評得, “[Jin is] correct” 是, or even applying Jin's personal favorite adjective “Marvelous!” 妙 back to Jin's own remarks. When Jin insisted that prospective readers must read the novel four times before realizing the author's hatred of Song Jiang, Tansō commented on his astuteness by remarking, “Jin Shengtan's comments on Shuihu zhuan are all refined and marvelous!” 聖嘆水滸傳只是論甚精甚妙.Footnote 37 Moments later, Tansō undercut this enthusiastic embrace by showing how even Jin could be led astray in his readings. He was perennially troubled that Jin remained unaware of the historical allegory encoded in the narrative, complaining that “Jin Shengtan was a man of great talent, but even he missed things among details” 聖歎才人,而亦失諸眉睫.Footnote 38 Tansō often accused Jin of over-reading or forced interpretation, and a number of the Chinese critic's more fanciful flights are dismissed as exaggerated or extraneous. For example, Jin praised Shuihu zhuan's use of “the unfinished clause method” 不完句法 (Ch. buwan jufa), in which the author creates a sense of conversational verisimilitude by having one character interrupt the other. When Jin claimed that the use of this technique was “unprecedented” 從古未有 (Ch. cong gu weiyou), Tansō quickly accused him of hyperbole by noting that “recent playwrights have all used this technique, and it's not remarkable in the least” 近時院本皆有,不足為奇.
Whereas Jin often treated Shi Nai'an as an infallible artistic genius and was loath to find fault with any aspect of Shuihu zhuan,Footnote 39 Tansō was far more willing to critique the text of the novel itself. Historical anachronisms and inconsistencies in the narrative were dutifully identified as “flaws in the stitching” 破綻 (Jp. hatan, Ch. pozhan), and Tansō resisted making the divisions between form (Ch. wen, Jp. bun) and content (Ch. shi, Jp. koto) that Jin occasionally espoused. For instance, Tansō abstained altogether from commenting on Chapters 23 and 24 of the novel, which related the story of the infamous seductress Pan Jinlian 潘金蓮, her illicit love affair with the wealthy playboy Ximen Qing 西門慶, and the murder of Pan's sexually deficient husband, Wu Da 武大. Despite his perfunctory condemnation of the lascivious Pan Jinlian and the chapters' racy content, Jin considered this cycle of stories to be an artistic highpoint of the novel, and his annotation of these chapters was carried out with particular gusto. The reader of Suikoden hihyōkai, on the other hand, finds a rather prim notice by Takada Jun:
These two chapters deal with squalid matters. It is not fitting that one read or hear about such things, and for this reason, Master Tansō left these out of his discussion and declined to comment.
以上二回, 猥雑残念, 目ニスベカラズ, 耳ニスベカラズ, 故ニ先生会誌ニ此二回ヲ闕ク, 因テ批評ナシ.Footnote 40
The development of critical acumen required analysis of concrete examples, and a number of Tansō's comments were devoted to the elucidation of textual patterns underlying the massive and seemingly sprawling novel. Structurally, Shuihu zhuan comprises a number of discrete story cycles focusing on the adventures of individual characters, and one of the most ambitious assertions Jin Shengtan made about Shuihu zhuan was that the episodic novel could be analyzed as a coherent whole. Rather than seeing the transference of narrative focus from one character to another as disjointed, Jin argued that the novel was largely ordered through recurrent motifs that surface at regular intervals throughout the text. Where the untrained eye would see chaos in Shuihu zhuan, the educated reader realized that a complex narrative scheme existed. As Jin claimed in a prefatorial essay:
In Shuihu zhuan, every passage has its method, every phrase has its method, and every character has its method. Children should be made to pore over it closely as soon as they can recognize characters. If you get a grasp on Shuihu zhuan, then other works of literature will be as easy as crushing bamboo.
水滸傳,章有章法,句有句法,字有字法,人家子弟稍識字,便當教令反覆細看,看得水滸傳出時,他書便如破竹.Footnote 41
This passage from Jin's essay is vigorously dotted in Tansō's personal copy of the novel, and throughout his commentary, Tansō followed Jin's example by alerting the reader to the variety of means by which the author sculpted the seemingly episodic plotline into a cohesive whole. Although he refused to follow Jin's lead by elevating the novel to the level of the Confucian classics and histories, he agreed that Shuihu zhuan demonstrated the same attention to method 法 (hō, nori) and structure 章 (shō) expected in more elevated genres.
An example is provided by Tansō's interest in “narrative structuring” 章法 (shōhō). Narrative structuring referred to the author's use of a particular motif to tie together passages, or create a seamless link between episodes. Once one learned to be a sensitive reader, Tansō implied, he would see that these links were ubiquitous. In the twenty-first chapter of the novel, for instance, Song Jiang—newly on the lam for the murder of his wife—takes shelter in the residence of the rebel-patron “Small Whirlwind” Chai Jin 小旋風柴進. When Song Jiang tipsily makes his way to bed after an evening of feasting, he upsets a charcoal brazier and burns a yet unseen invalid convalescing beside it. The singed sleeper turns out to be the soon-to-be famous tiger killer Wu Song 武松, who will take the leading role in the next cycle of chapters. The furious and malarial Wu Song breaks into a sweat, at which point Tansō tells the reader that “This [having Wu Song break out in a sweat] is structural” 章法. His student Takada Jun explains: “[The sweating] brings together Wu Song's fever with his killing of the tiger [in the next chapter]” 武松ガ瘧ト打虎トニ章ヲ合ス.Footnote 42 Though seemingly a mundane detail, the sweat acts as a linking motif by connecting the twenty-first chapter with the next, in which Wu Song will break into a cold sweat before dispatching a man-eating tiger with his bare hands. Similar instances, Tansō claimed, were scattered throughout the text, and he often critiqued Jin for failing to live up to his own high interpretive standards. Often, Tansō claimed, Jin allowed something to “slip by” 漏失 (rōshitsu) on the level of formal analysis. When the imperial executioner prepares to chop off the head of expert swimmer “White Streak in the Water” Zhang Shun 浪里白條張順 in the thirty-ninth chapter of the novel, for instance, Tansō chides Jin Shengtan for not realizing that the reader has been prepared for this moment in an earlier scene:
[The execution] resonates with the earlier scene in which Zhang Shun was [chopping up and] selling fish in the marketplace. Jin Shengtan let this slip by!
與張順賣魚作章法,聖嘆漏失.Footnote 43
Or again, when Song Jiang meets a seller of medicine en route to murdering his adulterous wife, Tansō chortles over the irony that Song meets a vendor of salubrious and life-extending elixirs only moments before ending the life of his wife:
Medicine is something that keeps people alive, but in this section of the text, the meeting with the medicine vendor segues into Song Jiang's murder of his wife!
薬ハ人ヲ活スモノゾ, 此処宋江売薬人ニ逢ニ因テ婆惜ヲ殺ス.Footnote 44
Although Tansō's interpretations often seem somewhat stretched to the modern reader, the important aspect to note is the similarity between his style of reading and Jin Shengtan's theories. Even in upbraiding his Chinese predecessor, Tansō demonstrates the degree to which he has been influenced by Jin's method of careful attention to detail. Compared to the earlier, cautious interpretive stabs at Shuihu zhuan made by the generation preceding him, Tansō's Suikoden hihyōkai is a confident and systematic attempt at using an earlier critic's method to produce an original reading.
CONCLUSION
It is with this observation that I bring my discussion of Seita Tansō and Shuihu zhuan to a close. Tansō rebuked Jin Shengtan for losing sight of major issues among details, and I will use my conclusion to avoid the same kind of myopia. My overall aim in this study has been to demonstrate the degree to which an educated Japanese connoisseur of imported fiction investigated Chinese theories of narrative in a deliberate and methodological way. Regardless of whether Tansō can be called a “typical” reader in eighteenth-century Japan, his engagement with Shuihu zhuan reminds us that works of Chinese fiction did not enter Japan as virginal, unmarked terrain. Most works of fiction and drama imported into Japan in the Edo period were commentarial editions that included interlineal annotation, prefaces, reading guides, and other paratextual features supplied by Chinese commentators. These commentarial editions were distinguished from each other in importation documents like the Hakusai shomoku, and educated Japanese readers considered these commentaries in their selection and interpretation of editions. Scholars of late imperial Chinese fiction have successfully argued that fiction commentary shaped the experience of leisure reading in late imperial China, and Japan studies would benefit immensely by considering how these interpretive strategies impacted the dissemination of Chinese vernacular fiction in Japan.
In the case of Seita Tansō and his literary sparring partner, Jin Shengtan, this issue of influence manifested itself in an ongoing process of creative appropriation and original synthesis. Tansō fundamentally agreed with Jin's assertion that the novel contained hidden meanings discernible only to well-trained readers. Both critics argued that the task of a thoughtful reader was to use textual criticism as a means of unearthing these secret truths. In the context of late imperial China, this assertion of formal complexity was a radical reevaluation of a genre hierarchy that relegated fiction and drama to the margins of critical discourse. In the case of Edo-period Japan, the type of figurative readings apparent in Tansō's work constitute a distinct rupture in the history of the reception of both Shuihu zhuan in particular, and Chinese vernacular narrative in general. What is immediately apparent in Tansō's work is his relative lack of interest in the philological dimensions of the novel. Earlier guides to the novel, such as Oka Hakku's 岡白駒 Suikoden yakkai 水滸伝訳解 (1728) or Suyama Nantō's 陶山南濤 Chūgi Suikodenkai 忠義水滸伝解 (1757), focused primarily on the precise definition of exotic vocabulary and stock phrases common to Chinese vernacular fiction. Tansō's analysis almost completely forgoes this type of philological explication in favor of a deeper focus on the narrative architectonics of a complex novel. Contemporary scholarship on the Japanese reception of Shuihu zhuan has focused almost entirely around two poles: its relation to early eighteenth-century classical and linguistic scholarship, and its use as a springboard for Japanese literary adaptation and rewriting in the second half of the Edo period. The relationship between these two poles has remained almost entirely unstudied, however, and Tansō's Suikoden hihyōkai provides a unique vista on a transitional literary episteme.
In a pioneering essay on the biography and literary accomplishments of Seita Tansō, the late Nakamura Yukihiko 中村幸彦 demonstrated that Tansō's extant corpus included discussions of Japanese fiction, kanshi poetry, and haikai in addition to his writing on Chinese texts. Footnote 45 It is clear that knowledge of these Japanese literary genres bled into his discussion of Shuihu zhuan, and as a topic for future consideration, I would argue that Tansō's interest in Shuihu zhuan impacted his understanding of the themes and narratological structure of Japanese literature as well. In contrast to earlier Japanese readers of Shuihu zhuan, who stressed the radical foreignness of the language and material artifacts presented in the novel, Tansō demonstrated an interest in relating the literary and structural features of Chinese texts to counterparts in the Japanese literary tradition. Although the literature of China and Japan differed from each other dramatically in terms of language, structure, and theme, Tansō suggested that certain composition techniques transcended these differences. By way of short example, Tansō drew the reader's attention to Chapter 5 of Shuihu zhuan, where the gallant Lu Zhishen 魯智深 engaged in hand-to-hand combat with both a corrupt Buddhist monk and Daoist priest. Jin Shengtan had also shown interest in the ways in which the author of Shuihu zhuan attempted to describe simultaneous action—pointing out, for instance, the “unprecedented” instances of interruption described earlier. In the case of Lu Zhishen's combat, however, Tansō makes no mention of Jin, and instead compares the author's technique to a genre that would be familiar to a Japanese reader. As Lu battles first the monk and then the priest, we are told, the author's use of alternating focus can best be compared to the respective roles played by the shite and waki in Japanese nō drama.Footnote 46
This interest in the comparative structural features of fiction and drama should be read in a larger context. During the period of Tansō's research on Shuihu zhuan, many scholars of Japanese classical literature were turning their attention to the Japanese classics—the paramount example being Murasaki Shikibu's Genji monogatari 源氏物語 (The Tale of Genji)—in an attempt at establishing new critical paradigms for their interpretation.Footnote 47 At the same time that the nascent kokugaku movement was attempting to move beyond didactic analyses of the text in favor of focus on the language, literary structure, and unique cultural essence of the work, Tansō made a remarkable comparison between the Genji and his favorite Chinese novel:
In both our own country and in China, what determines a well-written piece of writing more than anything else is an outstanding sense of creative structure. In parables and tales, this is even more the case—[otherwise] even if the words are chosen skillfully, they won't be worth reading. The Zhuangzi goes without saying, but even Shuihu zhuan—though it's not worthy of serious discourse—is still outstanding in its creative expression. In the works of our own country that I have encountered, there are none that surpass The Tale of Genji. It too can truly be termed a work of outstanding creative structure.
吾国も唐土も上手の作りたる書は, 何よりもまず心匠すぐれたり, 寓言の書は一入心匠都合せざれば, ことば匠なるとも, 見るにたらず, 荘子はいうまでもなし, 水滸伝は論ずるにたらざる書なれども, 心匠はすぐれたり, 吾国の寓言の書予が見る所にては, 源氏物語を第一とすべし, 所謂かの心匠甚すぐる.Footnote 48
Tansō's remark is followed by a brief sampling of criticism focused on different episodes in the Genji. Tansō's remarks are too brief to extract a sense of his larger interpretation of the novel, but the style of his analysis appears familiar to any reader of his writing on Shuihu zhuan. As with Shuihu zhuan, the miraculous craftsmanship of Genji, Tansō posits, lies in the author's attention to subtle detail—details that often reveal clues about the narrative's larger structure and uniqueness. For instance, Tansō informs the reader that the singularly chilling effect of the spirit possession in the “Yugao” chapter is a result of the fact that while spirits in “nine out of ten” ghost stories emerge during the hour of the ox (2:00 a.m.), the possessing spirit in Murasaki's text appears during the hour of the boar (10:00 p.m.)—lengthening the time of Genji's anxious wait for dawn, and thereby emphasizing the young man's solitude, immaturity, and terror during his vigil. Or, we are told, Kaoru's and Niou's inheritance of different aspects of Genji's manners, charm, and physical beauty in the Uji chapters, is actually the author's melancholy reminder that the all-encompassing brilliance of the departed Shining Prince was something irreplaceable in later generations.Footnote 49 These are terse and undeveloped examples to be sure, but Tansō's larger point is clear: in Genji, as in Shuihu zhuan, analysis of subtle detail and localized incident allows the careful reader to extract larger themes and hidden meanings in the text. And despite the radically different narratives, style, and genre features of Shuihu zhuan and Genji monogatari, Tansō analyzes both texts with reference to the same critical terminology. This cross-generic interrogation of literary structure characterizes not only Tansō's critical method, but also lies at the heart of the mid- and late-Edo florescence of adaptations, parodies, mashups, and sequels adapting exotic Chinese narratives into familiar Japanese settings. Finally, in terms of charting a broader history of early modern Japanese literature, this kind of exploration prefigures the formulation of more general theories of narratology that evaluated Japanese and Chinese (and later European) narrative with reference to common aesthetic principles. The comparative projects of Hagiwara Hiromichi and Kyokutei Bakin, who famously incorporated Chinese critical commentary into their criticism and fiction, have drawn a considerable amount of scholarly attention; the critical work of Seita Tansō pushes this history of transnational engagement back half a century further.
