Out of the Margins Ge, Liangyan Published by University of Hawai'i Press Ge, L.. Out of the Margins: The Rise of Chinese Vernacular Fiction. Honolulu: University of Hawai'i Press, 2001. Project MUSE., https://muse.jhu.edu/. For additional information about this book https://muse.jhu.edu/book/8086 Access provided by University of Washington @ Seattle (14 Apr 2018 21:03 GMT) 2 Told or Written That Is the Question The field of early Chinese vernacular fiction has long been haunted by questions concerning the origins of the genre. How was each of the earliest fulllength vernacular novels—Shuihu zhuan, Sanguo yanyi, and Xiyou ji—related to the long oral tradition that preceded it? Did the popular story-cycles only provide the subject matter for the composition of the narrative, or did the oral model exert a shaping influence on the work in print on the level of narrative discourse as well? These questions are so hard to answer simply because we know so little about those popular traditions and about the textual evolution of the narratives themselves. Indeed, no words summarize our quandary better than these by W. L. Idema: “The prevailing uncertainty in these matters means that any view on the origin and role of Chinese colloquial fiction can only be advanced with diffidence.”1 In the case of Shuihu zhuan, a consensus has long been reached that there had been, before the narrative appeared in print, an oral complex of Shuihu stories. About that long tradition that started probably as early as the thirteenth century, the “hard facts” that we know are very few in number. And we know even less about the transition from the oral cycles to the book form of the narrative. With most of the historical information irrevocably lost, an investigation of the formative stage of Shuihu zhuan has to be, in a sense, a comparative study of different hypotheses. Yet if what Idema calls the “prevailing uncertainty” may daunt any attempt to probe the issue, it is also the very justification and motivation for doing so. This chapter revisits the oral Shuihu tradition, attempting to reclaim at least part of the evolutionary course of the Shuihu complex by examining the few ex- 36 Told or Written 37 tant texts or textual remnants that could be representative of different stages of that course. Next the chapter reviews different critical opinions on Shuihu zhuan, largely based on divergent assessments of the novel’s relationship to the oral tradition. My own belief is that the synthesis of the Shuihu materials from various popular genres may have culminated in a long narrative predominantly in oral prose, which did not merely become the source for the subject matter in Shuihu zhuan but bestowed on the novel much of the narrative discourse itself. This hypothesis is put forward here rather tentatively, before it is supported by the textual, philological, and historical studies of the narrative in later chapters. Early Stage of the Shuihu Tradition If the development of traditional Chinese narrative literature was characterized by a general shift from historicity to fictionality, Shuihu zhuan holds a pivotal position in the course of that transition.2 While Sanguo yanyi, which appeared in print probably slightly earlier, depends heavily on historiography in both the official (zhengshi ) and the popular versions ( yeshi), Shuihu zhuan’s relationship to history is rather tenuous. In Song shi (History of the Song), there are brief references to the historical Song Jiang and his band in the early twelfth century. In his memorial to the throne, Hou Meng, prefect of Dongping, referred to the bandits as “Song Jiang’s thirty-six men, for whom the thousands of official troops were no match.” He therefore suggested to the emperor that an amnesty be offered to Song Jiang in order to enlist his service in the campaign against another rebellion led by Fang La.3 In “Zhang Shuye zhuan,” it is recorded that Song Jiang and his band surrendered after being ambushed and defeated by the forces led by General Zhang Shuye.4 In Li Zhi’s Huang Song shi chao gang yao, Song Jiang’s name is listed among those who led the attack on Fang La, but judging from the accounts in Song shi of the defeat of Fang La, Song Jiang may not have been the main force in the historical campaign as is the case in Shuihu zhuan.5 Around such historical references, the novel agglomerates a large number of fictional tales that probably had first developed orally as separate short pieces before becoming assembled together. Luo Ye’s Zuiweng tanlu contains references to eight different types of oral stories under the rubric of xiaoshuo that were circulating during the Southern Song.6 For each type, a number of story titles are given as examples. Four stories, judging by their titles, may well have belonged to an oral complex centering on the Liangshan bandits, although none of them is now extant. “Shitou Sun Li” (Sun Li the Stony Man) is listed as a gong’an (court case) story. Most likely it was a story about the Liangshan chieftain Sun Li, and its classification as a gong’an story is consistent with Sun Li’s status in Shuihu zhuan as a police captain before joining the rebellion. “Qing- 38 Chapter 2 mian Shou” (Blue-faced Beast) is labeled as a podao (broadsword) tale. It may well have been a story about Yang Zhi, a Liangshan rebel nicknamed Qingmian Shou who was a skilled user of the broadsword. Among the ganbang (staff ) tales there are “Wu Xingzhe” (Wu, the Untonsured Monk) and “Hua Heshang” (The Tattooed Monk), which were almost certainly stories of the two monk-turned-bandit heroes who used staffs as their weapons: Wu Song and Lu Zhishen, who is nicknamed Hua Heshang.7 No other titles listed seem to suggest a similar relationship to the Shuihu complex. That, however, does not necessarily mean that these four were the only Shuihu stories circulating at the time, for the titles are listed only as examples. In fact, not only these four figures but most of the thirty-six who were to become major chieftains of the band in Shuihu zhuan were already popular with storytellers during the thirteenth century. This is evidenced by Gong Shengyu’s (1222–1304) “Song Jiang sanshiliu zan” (Encomiums to Song Jiang and His Thirty-Six), in which Gong celebrates each of the bandit heroes in a short piece of verse, which perhaps was originally attached to a portrait of each rebel either drawn by himself or by Li Song (fl. 1240), another painter one generation earlier.8 The thirty-six names in Gong Shengyu’s list are not completely identical with those of the thirty-six major chieftains, or tiangang xing (Stars of Heavenly Spirits), in Shuihu zhuan. Gongsun Sheng and Lin Chong, two prominent names among the thirty-six in the novel, are absent from Gong Shengyu’s list. On the other hand, among the names on Gong’s list, Chao Gai is not counted as one of the thirty-six in Shuihu zhuan, and Sun Li, despite his important role in the early days of the Shuihu complex, is relegated to the status of a minor chieftain, or a disha xing (Star of Earthly Fiends), in the novel. Although short, Gong’s prefatory note before the verses is remarkably informative about the state of the oral Shuihu complex at the time. It tells us at least two things. First, such figures as Yang Zhi, Sun Li, Wu Song, and Lu Zhishen, who had previously been protagonists in separate and individual stories, became by Gong Shengyu’s time regarded as among the thirty-six fellow members of the same group. This may mean that the process in which the short stories of individual bandits became strung together into a longer narrative may have started as early as the thirteenth century. Second, while he claims that “stories of Song Jiang are heard on the streets,” Gong Shengyu clearly states that he had not seen any such stories in written form apart from the brief notes about Song Jiang in wenyan historiographies.9 Since the artist had such a keen interest in the bandit heroes, any written literature about them, if there were any at the time, would probably not escape his notice. What Gong Shengyu says in the encomiums, therefore, makes it very likely that by the artist’s time no Shuihu texts had yet been produced from the oral storytelling—at least not for the general reading public. Told or Written 39 One text, a unified but very sketchy version of the nucleus of the Shuihu complex, finally appears in Xuanhe yishi, compiled anonymously probably in the early Yuan period.10 Part of the work is a chronicle of the last years of the Northern Song, with its military defeat by the Jin invaders, the abduction of its last two emperors, and the transfer of the nation’s capital from Kaifeng to the southern city of Lin’an. At other places, such as the section on the Shuihu rebels and that on Emperor Huizong’s dalliance with the courtesan Li Shishi, the tone of the narrator shifts from that of a historiographer to that of a storyteller. The embryonic Shuihu narrative in Xuanhe yishi consists of four episodes: (1) Yang Zhi, stranded in a small town, has to sell his broadsword and accidentally kills a rascal. He is sentenced to exile but is rescued by his friend Sun Li and others. The group of twelve afterward goes to become rebels in Taihangshan (Taihang Mountains). (2) Liang Shicheng, the prefect of Daming, sends an enormous amount of valuables to his father-in-law, the prime minister Cai Jing, as birthday gifts. Chao Gai and his cohorts drug the escorts and seize the gifts. Before they would have been arrested, Song Jiang, a county clerk, sends them a message that allows them to escape and join Yang Zhi and other outlaws. (3) To thank Song Jiang, Chao Gai and the band send him gold, which betrays Song Jiang’s association with the bandits to his adulterous mistress Yan Poxi. Song Jiang kills the woman and, while hiding in a monastery, acquires a “heavenly writ” that lists the names of the thirty-six rebels. Song and thirteen others join the band at Liangshan and, with the death of Chao Gai, bring the number of the chieftains to thirty-three. The number is afterward brought up to thirty-six with the arrivals of Lu Zhishen, Li Heng, and Huyan Chuo. (4) The court offers amnesty to the band, and Song Jiang is made a garrison commander ( jiedushi) after he helps quell the rebellion led by Fang La. In comparison with the immense length of Shuihu zhuan, the Shuihu segment in Xuanhe yishi is very short, occupying merely ten pages in the Shiliju congshu edition.11 It is incomplete even considered as an outline of the narrative, for it leaves out many elements that must have become part of the Shuihu complex by then. Stories about Lu Zhishen and Wu Song, for instance, seem to have already been in circulation during the Southern Song, judging from the titles of stories in Zuiweng tanlu, but neither story found its way into Xuanhe yishi. Even those episodes that are included are very different from their counterparts in Shuihu zhuan. Yet since the segment appears in a work that was intended to be a popular history, as the form of chronicle and the title of the book both indicate,12 the Shuihu complex could by then have started its evolutionary process from individual short tales into jiangshi, a different oral genre specializing in long narratives that “tell about history.” In general, Xuanhe yishi, like those texts of the Yuan under the rubric of pinghua, appears in a language medium that mixes wenyan with ingredients from 40 Chapter 2 baihua. There is no consistent stylistic register throughout the work; while elsewhere the language is often unadulterated wenyan, in the Shuihu segment it leans closer to baihua. As William O. Hennessey informs us, most of the wenyan sections in Xuanhe yishi have word-for-word parallels in seven classical sources that date from the Southern Song. Significantly, none of the vernacular sections, including the Shuihu segment, is found to have any textual parallels.13 On the other hand, it is quite obvious that the Shuihu segment was not written ab initio. In addition to its stylistic distinction from most other portions of the work, the Shuihu section starts and ends abruptly, with little or no connection with what proceeds or follows it. The entire Shuihu section is put under the heading of “Xuanhe sinian” (The Fourth Year of the Xuanhe Reign), even though the only thing in the section that could possibly have happened in that year was Song Jiang’s surrender to the throne, which the narrative mentions only in passing, while everything else is said to be in the second year of Xuanhe.14 This seems to indicate that the Shuihu segment was incorporated wholesale, with no attempt on the part of the compiler to assimilate it properly into the structure of Xuanhe yishi. Since there appear to be no written sources of Shuihu stories apart from some brief historical notes by Gong Shengyu’s time, which was perhaps only slightly earlier,15 we are led to believe that the source for the Shuihu segment in Xuanhe yishi may have been in some way associated to oral storytelling. This seems even more likely if we take into account the fact that the language in the Shuihu segment is notably more colloquial than elsewhere in Xuanhe yishi.16 Shuihu Plays in the Yuan and Early Ming Periods Unfortunately, the segment in Xuanhe yishi, which is the earliest extant text of a Shuihu narrative before Shuihu zhuan, happens to be the only such text; there are simply no extant texts of any Shuihu narratives that can be dated between Xuanhe yishi and the earliest known editions of the novel. Meanwhile, however, the Shuihu complex must have been vigorously active during that time, as evidenced by a large number of Shuihu zaju from the Yuan and the early Ming. Sixteen titles of Shuihu plays are listed in Zhong Sicheng’s (fl. 1321) Lu gui bu, all by early Yuan playwrights: eight by Gao Wenxiu, two by Li Wenwei, one by Yang Xianzhi, three by “Hongzi” Li Er, and two by Kang Jinzhi. Lu gui bu xubian, another bibliographic source attributed to Jia Zhongming (1343–?) on the late Yuan and early Ming drama, lists four titles of Shuihu plays without giving the authors’ names. Out of these twenty plays, five are extant and collected in Shuihu xiqu ji, which also includes another five extant Shuihu plays based on the seventeenth-century manuscripts from the Maiwangguan Library, whose titles do not appear in either Lu gui bu or Lu gui bu xubian. Apart Told or Written 41 from these, the two fifteenth-century Shuihu plays by Zhu Youdun, Hei Xuanfeng zhang yi shucai and Baozi heshang zi huansu, are also included in Shuihu xiqu ji. For three of these plays—Nao tongtai, Dongping fu, and Jiugong bagua zhen— the dating is problematic. Fu Xihua and Du Yingao designate them as plays from the late Yuan or early Ming.17 Y. W. Ma, noticing in them the emergence of minor chieftains of the rebellion and a closer parallel in their dramatic actions to Shuihu zhuan, believes them to be later than the novel. Actually, even for those plays attributed unequivocally to Yuan playwrights, the extant texts are all from late Ming editions. George A. Hayden has convincingly suggested that all the Yuan plays of the Shuihu zaju may have undergone extensive revisions in the Ming.18 Even for Zhu Youdun’s two plays, of which we know the exact date of the original publications, the situation is complicated by the existence of the late Ming manuscripts from the Maiwangguan Library, which feature a larger amount of dialogue and spoken verse than Zhu’s original versions. In Shuihu xiqu ji, the text of Zhang yi shucai is based on the late Ming manuscript, while that of the other play, Baozi heshang, appears in its original version. As I am discussing here the formation of the Shuihu saga, the focus of my interest is primarily on the dramatic plot of the plays. For that purpose one may still consider the plays attributed to Yuan or early Ming authors as early Shuihu plays, assuming that the basic structure of the dramatic action is not likely to have been radically reshaped by the subsequent revisions. Judging by the structure of these early Shuihu plays, the existence of a popular Shuihu complex during that period is beyond any doubt, as the audience’s background knowledge of the half-historical and half-legendary Liangshan rebellion is simply taken for granted. In most of the plays, usually at the very beginning, there is a monologue by Song Jiang summarizing how he has come to Liangshan and become the leader of the band, and the monologue mentions so many events in the band’s history in such a sketchy way that it would certainly confuse any viewer of the play who was not pre-informed of the tradition. It is striking that the monologue in different plays follows a largely standard line, which may suggest that some portions of the Shuihu complex had become established enough by Yuan times as to have such a strong centripetal impact on the plays.19 Apart from that, however, a Shuihu zaju is an autonomous work in itself, with its plot not necessarily consistent with the ongoing tradition of the Shuihu complex. An early Shuihu zaju typically focuses on an episode involving one or a few bandits, for a very limited period of time, in most cases a holiday. Except for Li Kui fujing,20 each of the extant Shuihu plays from the Yuan and the early Ming features a dramatic action that is not to be found in Shuihu zhuan.21 42 Chapter 2 The disparity between the plots of the early Shuihu plays and that of Shuihu zhuan may seem to suggest that the plays represent an evolutionary stage of the Shuihu complex that was vastly different from its final form. Sun Kaidi has even suggested that the Shuihu zaju and Shuihu zhuan may have belonged to two separate though not mutually insulated traditions, one in the north and the other in the south.22 If that were the case, one would expect the plots of the early Shuihu plays to be basically consistent among themselves. This is, however, not true. On the ranking of the Liangshan chieftains, an issue that carried particular significance throughout the entire history of the Shuihu tradition, the plays are irreconcilably contradictory to one another. The same position of Number 13, for instance, is assigned to three different characters in as many plays: to Li Kui in Shuang xian gong, to Hua Rong in Sanhu xiashan, and to Ruan Jin in Baozi heshang. An early Shuihu play, as mentioned earlier, is typically about a short excursion of one or a few Liangshan figures from the lair—after the assemblage of the major chieftains has reached the plenary number of thirty-six,23 but before the band’s surrender to the throne.24 The dramatic actions, however divergent they might appear to us today, did not really encroach on the core of the Shuihu complex—namely, the aggregation of the heroes under the banner of rebellion. In Yan Dunyi’s words, such dramatic plots are “amplifications of the story” but not the “making of the blood and bones of the story itself.”25 By the time of the early Shuihu plays, one of the elements that did make “the blood and bones” of the Shuihu tradition was Song Jiang’s killing of his mistress Yan Poxi, as the event is uniformly mentioned in Song Jiang’s monologue in several different plays. Yet even though the episode is obviously very dramatic and therefore easily adaptable into a play, none of the early Shuihu zaju, extant or not, was based on it.26 Indeed, the playwrights might have to strike a balance: On the one hand the dramatic plot had to be kept within a tradition familiar to the audience, while on the other hand commercial competition among theatrical troupes would necessitate innovations. They might be consciously shunning the materials from the Shuihu complex where the authority of tradition was too intimidating and prohibitive of any modification, turning instead to a zone where the tradition was less dominant and therefore more promising for further experiments. The bandits’ activities after their final ranking (pai zuoci) and before the imperial amnesty may well have been such a zone, which allowed a compromise between tradition and novelty.27 To some extent, the Shuihu plays can be considered a few more examples of those numerous literary works that— driven by an anxiety of the influence from a powerful tradition—have to struggle for their own territory on the flanks of that tradition. Indeed, many of the early Xiyou zaju and Sanguo zaju are also character- Told or Written 43 ized by a similar deviation from the established narrative cycles. Judging by the titles listed in Lu gui bu, some of the nonextant Sanguo plays attributed to Yuan authors, such as Sima Zhao fuduo shoushantai (Sima Zhao Retakes the Rostrum for Receiving the Abdication of the Crown), Guan Dawang san zhuo hongyi guai (General Guan Thrice Captures the Demon in Red), and Mang Zhang Fei da’nao xiang fu yuan (The Reckless Zhang Fei Creates an Uproar at the Prime Minister’s Residence), might contain plots that have no parallels in either Sanguo zhi pinghua or Sanguo yanyi.28 The plots of the two extant Sanguo plays collected in Yuanqu xuan—Gejiang douzhi (The Battle of Wits across the River) and Lianhuan ji (The Interlocking Ruses)—are closer to Sanguo zhi pinghua than to Sanguo yanyi. The former play is about Zhou Yu’s unsuccessful strategy to retake Jingzhou by marrying Sun Quan’s younger sister to Liu Bei. In the play, the bride goes to Jingzhou for the wedding, rather than the groom going to the Wu as is the case in the novel. The latter play is based on the episode in which Wang Yun promises the beautiful girl Diao Chan to both the powerful and treacherous minister Dong Zhuo and Dong’s adopted son Lü Bu and, by doing so, has Dong killed by Lü. In the play, Diao Chan turns out to be Lü Bu’s separated wife, a twist it shares with Sanguozhi pinghua but not Sanguo yanyi. As for the Xiyou plays, Lu gui bu lists the title Tang Sanzang xitian qujing (Tripitaka of the Tang Travels to the West for the Sutras), a nonextant zaju attributed to the Yuan playwright Wu Changling. The Tianyige edition of Lu gui bu gives the subtitle of the play, Lao Huihui donglou jiaofo (The Old Hui on the Eastern Tower Invokes the Buddha),29 which suggests that at least part of the plot of the play was quite alien to the novel Xiyou ji. The same can be said of an extant play entitled Xiyou ji, which may or may not be Yang Jingxian’s Xiyou ji listed in Lu gui bu xubian.30 While the title of the play is identical to that of the novel, Sun Wukong in that play is actually more like the ape spirit in the story “Chen Xunjian meilin shiqi ji” (in Qingping shantang huaben) than the heroic monkey in the full-length fiction.31 We therefore need not have too many doubts about the continuity of the Shuihu complex only because of the disparity between the early Shuihu plays on the one hand and the two narratives, Xuanhe yishi and Shuihu zhuan, on the other. This is, of course, by no means to downplay the pertinence of the Shuihu zaju to the study here. Even those dramatic plots that are apparently different from the plot in Shuihu zhuan could have a latent kinship with the novel. In several early extant Shuihu plays, part of the plot follows a stereotyped pattern: An honest man is framed by a treacherous wife/concubine and her lover, usually a yanei (man with powerful connections), when one or a number of Liangshan bandit heroes, who are the husband’s friends or sworn brothers, come to rescue the wronged man and mete out justice to the adulterous couple. Plays 44 Chapter 2 whose plots can at least partly fit this pattern include Hei Xuanfeng shuang xiangong, Yan Qing puyu, Huan laomo, and Zheng bao’en. In the last play, however, it is the wife who is framed by an adulterous concubine and who is subsequently rescued by three Liangshan chieftains. This pattern bears a strong resemblance to two long narrative stretches in Shuihu zhuan, in each of which the hero, Wu Song and Shi Xiu respectively, rebuffs the sexual advances of a sister-in-law and subsequently kills the adulterous woman and her lover in order to avenge a wronged/murdered brother (chapters 24–26, 44–45). Another Yuan zaju, although not a Shuihu play, features a plot that parallels the two stretches in Shuihu zhuan even more closely, especially the one about Wu Song. In that play, entitled Gengzhi Zhang Qian ti sha qi (The Righteous Zhang Qian Kills a Sister-in-law on His Brother’s Behalf ),32 the protagonist Zhang Qian relentlessly resists seduction by his sworn brother’s wife in order to honor his fraternal loyalty. In the end, he has to kill the woman to stop her attempted murder of her husband, and he is consequently thrown into jail. It is, however, hard to say whether the two stretches in Shuihu zhuan were already part of the Shuihu complex during the Yuan and served as models for the zaju, or whether they represent the narrative’s adaptations of those Yuan zaju plays. Whichever was the case, it seems fair to say that the Wu Song and Shi Xiu stories might have once belonged to the same repertoire in popular orality as those Yuan plays featuring similar plots. The Hypothetical Shuihu cihua Despite the contributions the Shuihu zaju may have made to the Shuihu complex, their plots could have mostly belonged to the periphery rather than the nucleus of the tradition. As the actions in most of the plays are paralleled neither in Xuanhe yishi nor in Shuihu zhuan, we can hardly consider the Shuihu zaju the only link between the two narratives of the Shuihu saga. The major part of the tradition must have been carried on in other genres of popular orality, of which, unfortunately, no textual relics of any extended length are now available. Zheng Zhenduo asserts that in the Yuan period, side-by-side with the Shuihu zaju, “there must have been a Shuihu zhuan,” by which he means a possible earlier version of the narrative not necessarily sharing that title.33 Sun Kaidi, who puts forward the hypothesis that the Yuan antecedent of Shuihu zhuan might be in the form of a Shuihu zhuan cihua, attempts to explain some of the contradictions in Shuihu zhuan in terms of its duplicate incorporation of similar incidents from two different versions of the cihua, one from the south and the other from the north.34 Our knowledge of cihua as a form of oral narrative was deficient until very recently. One extant text that contains the word “cihua” in its title is Da Tang Told or Written 45 Qin Wang cihua, a narrative about the founding of the Tang dynasty compiled in the late Ming by Zhu Shenglin.35 We have reason to believe that by the late Ming that form of storytelling might have long passed its heyday, for the word “cihua” had by then lost its stringency as a generic label. Jin ping mei cihua, which also contains the word in its title, does not have the formal features that would distinguish it from other vernacular narratives of the time, even though cihua might well have been one of the sources for the work. Again, in the story “Jiang Xingge chonghui zhenzhu shan” ( Jiang Xingge Reencounters the Pearl Shirt), collected in Gujin xiaoshuo, the narrator claims to be telling a cihua.36 But the story, according to Patrick Hanan, may have been written by Feng Menglong himself, and indeed it bears little formal distinction from Feng’s other stories.37 If by the late Ming the conception of cihua had become so murky, it is reasonable to surmise that the genre must have belonged to a period considerably earlier, possibly the Yuan and the early Ming.38 The sixteen texts of shuochang cihua printed in the Chenghua period (1465–1487) and unearthed in 1967,39 which appeared in 1973 in the volume Ming Chenghua shuochang cihua congkan, have substantially enhanced our knowledge of cihua as a form of storytelling.40 The woodblock editions of the shuochang cihua texts date from the 1470s, and some of them may be reprints of Yuan editions over one hundred years earlier.41 The dating of the prints and the possibility of even earlier textual prototypes agree with our conjecture on cihua’s prime time. Except for “Bao Longtu duan baihu jing zhuan” (Lord Bao Judges the Case of the White Tiger Demon), which consists entirely of verse, all the cihua texts are in prosimetric form, alternating between prose and verse. The absence of prose in that particular cihua, however, does not necessarily suggest that it was performed in a different manner. Most likely, as in so many other texts in early vernacular literature, vernacular prose was not registered or regulated textually but left for the performers to improvise. Despite the word “ci” in the name of the genre, which often suggests a diversity of metric patterns, the verses in these cihua appear overwhelmingly in heptasyllabic lines, which occasionally vary into decasyllabic ones. Unlike zhugongdiao, these shuochang cihua texts do not indicate the tunes for the songs; in fact, no names of any tunes are given anywhere.42 Additionally, in contrast to bianwen, there are no verse-introducing phrases. The prose passages are rather parsimonious, in some cases presenting merely one or two exchanges in the characters’ dialogue. The story is thus narrated mainly by the verses, which are heavily formulary and therefore could be orally improvisable.43 The dearth of prose in the shuochang cihua texts contrasts strikingly with what we see in Da Tang Qin Wang cihua, where prose, although interspersed with verses, is the more important vehicle of narration. 46 Chapter 2 To these shuochang cihua, a stanza of verse in Shuihu zhuan seems quite akin in nature. In chapter 48 of the Rongyutang edition, when Song Jiang and his forces are launching a second assault on Zhujia Zhuang (Zhu Family Village), the narrative shifts into these verse lines: Lone Dragon Cliff before Lone Dragon Mount, Atop Lone Dragon Mount is Zhu Family Village. Surrounding the mount is a flowing stream, Circling the village are lines of trailing willows. Within the walls bristle swords and halberds, Before the gate spears and lances are in array. For a foe, people here are all valiant soldiers, In a battle, every man is in his prime vigor. Zhu Long, on the battlefield, is hard to rival, Zhu Hu, in a fight, nobody can equal. And Zhu Biao is an even greater warrior, Bellowing in anger like Xiang Yu, the Hegemon. Lord Zhu is a man full of wily strategies, And his treasures fill a thousand chests. Before the gate stand two white banners in a pair, Strikingly writ on them are these two lines: “Level up the Marsh and seize Chao Gai, Trample flat Liangshan and capture Song Jiang.” (3: 1587–1588)44 In the Chinese text these are all heptasyllabic lines, which are the standard metric form for shuochang cihua. The phrases describing the military array and the martial prowess of the Zhu brothers are clearly formulaic, which can be easily applied to any number of other battle scenes. More important, the stanza forwards the action of the story, rather than merely commenting on the characters or events in the narrative as the verses do elsewhere in Shuihu zhuan. Indeed, these lines may well be a residue of an earlier cihua, as Sun Kaidi has suggested.45 Sun Kaidi, however, obviously considered the textual inheritance to be on a much larger scale. He suggests in the same article that the long stretch in Shuihu zhuan on Song Jiang’s exile to Jiangzhou (chapters 36 to 41), the episodes of Dame Wang’s instigation of Ximen Qing’s lust for Pan Jinlian (chapter 25), and that of Lu Zhishen’s encounter with the evil Daoist priest Qiu Xiaoyi may all have been taken over from the cihua.46 Additionally, Sun Kaidi cites from Shuihu zhuan words and phrases from the southern dialects as evidence of his hypothetical southern Shuihu zhuan cihua.47 But since most of those southern Told or Written 47 expressions do not occur in sections that Sun identifies as derived from the cihua, Richard Irwin notes the dilemma: Either the scope of the cihua needs to “be enlarged to include them,” or they suggest instead “the locale in which the novel itself was compiled.”48 Indeed, Sun Kaidi has largely disregarded the formal differences between the verse-dominated cihua and the prose narrative of the novel and proposed a relationship of textual derivation so immediate and so extensive that the text of Shuihu zhuan and his hypothesized Shuihu zhuan cihua almost become identical. Irwin must have noted this as a possible problem with Sun Kaidi’s hypothesis, for he tried to straighten out the contradiction with a hypothesis of his own: “As a matter of fact, such a short time elapsed between the writing of the missing cihua and the drafting of the novel that they may have been done by the same group of men.”49 Our knowledge of another formal feature of the shuochang cihua also contradicts Sun Kaidi’s theory of one single southern cihua as Shuihu zhuan’s antecedent, even though he acknowledges possible supplements from another cihua from the north. The cihua, as evidenced by the chenghua texts, was a form of rather limited length,50 which tells one coherent and integrated story.51 Given the episodicity and exuberance of narrative actions in Shuihu zhuan, one Shuihu zhuan cihua as the bedrock for the voluminous and panoramic narrative is hardly conceivable. Nevertheless, the cihua hypothesis remains useful for us. Among the sixteen shuochang cihua, twelve belong respectively to two clusters: four on the life of Guan Suo, allegedly son of Guan Yu, a prominent general of the ThreeKingdoms period; and eight about Judge Bao, who is called Bao Longtu or Bao Shizhi in these cihua, a legendary upright official of the eleventh century. The cluster of Judge Bao cihua particularly deserves our attention. Each of the Judge Bao cihua tells a story of the incorruptible official, a story that is self-contained and noncontingent on any other cihua in the same cluster. Except for the plot in “Bao Shizhi chushen zhuan” (The Family Origins of Lord Bao), which tells of Judge Bao’s entrance into officialdom and for that reason has to precede all other Judge Bao stories, there is no established temporal sequence for all the stories in the cihua cluster. Yet clearly all the Judge Bao cihua belong to the same story complex, as there are frequent cross-references in one cihua to the events narrated in another. In “Bao Longtu Chenzhou tiaomi ji” (Lord Bao Sells Rice in Chenzhou), Judge Bao, immediately before he leaves for Chenzhou on a famine-relief mission, fines Empress Cao one hundred ounces of gold for illegally lending her chariot to another imperial consort of a lower rank.52 That incident, along with Lord Bao’s executions of four of the emperor’s corrupt kinsmen during his mission in Chenzhou, is referred to in “Bao Longtu duan Cao Guojiu gong’an zhuan” (Lord Bao Judges the Case of Empress Cao’s 48 Chapter 2 Brothers).53 And the executions of the unscrupulous Cao brothers, along with Judge Bao’s many other feats, are in turn enumerated in the introductory lines of “Renzong renmu zhuan” (Emperor Renzong Acknowledges His Mother), just before Judge Bao meets the banished empress dowager in a broken kiln on his way back from Chenzhou to the capital.54 As we can see, no unified temporal order can possibly be established for all the events in these different stories—or rather, such a temporal order may have been considered totally irrelevant. That implies that these Judge Bao cihua, while belonging to the same story complex, remained as separate and independent short pieces, not ready to be synthesized into a long Judge Bao narrative. The implication of such a cihua cluster is relevant to our discussion of the role that cihua possibly played in the evolution of the Shuihu narrative. If there had been any Shuihu cihua—which is likely, as the textual vestiges in Shuihu zhuan suggest—it must have existed in the form of multiple and relatively short pieces, each telling a self-contained Shuihu story. Since the stanza of verses that could be a remnant from a cihua occurs amid the narrative of Song Jiang’s campaign against Zhujia Zhuang, one might hypothesize a cihua about Liangshan bandits’ three assaults on the village. Given the immense base of the Shuihu complex, however, it might not be the only Shuihu cihua; rather, a cluster of such cihua is highly possible, although textual traces from other Shuihu cihua have been lost. For instance, there might be—to speculate a little further—a cihua on Lin Chong, another one on Lu Zhishen, and still another one on Wu Song. Among these cihua there could be a strong affinity, yet—as in the case of the cluster of Judge Bao cihua—a mechanism was lacking that could coordinate between different cihua and then integrate them into a longer narrative of epic dimensions. If indeed such a cluster of Shuihu cihua existed, the stories had to belong to a relatively early stage in the evolution of the Shuihu cycle, most likely in the Yuan, before they were later assimilated into a much longer Shuihu narrative where prose instead of verse was the primary form of storytelling. This conjecture tallies with the rough dating of the cihua’s heyday and corroborates the delineation in chapter 1 of the general process in which the vernacularization of written prose lagged considerably behind that of verse. In addition, it helps explain the narrative structure in some portions of Shuihu zhuan, especially in the first seventy chapters, where the narrative looks rather like a collage of the personal sagas of individual bandit heroes. At this point, a couple of clarifications are in order. The discussion of a possible cluster of Shuihu cihua does not mean that the cihua was the only form of storytelling that carried the tradition. The seemingly focused attention on the cihua is only because it was, apart from the Shuihu zaju, the only genre in popular orality with which some textual relations can be traced in Shuihu zhuan. Told or Written 49 Indeed, even those textual vestiges are rather tenuous. In any case, the evolution of the Shuihu complex should not be taken as a unilinear process. There might very well be other genres of oral performance existing contemporaneously with the Shuihu cihua and Shuihu zaju, all contributing to the final form of the narrative in Shuihu zhuan. Furthermore, although I suggested that the Shuihu stories told in the cihua had to be prosified later in a different form of storytelling, oral narratives that were primarily in prose did not necessarily belong only to the later stages of the tradition. Some of them might have existed early. The Southern Song stories of Yang Zhi, Wu Song, and Sun Li in the genre of xiaoshuo, as one may recall, might well be prose narratives, as the word “xiaoshuo” (small talk) would suggest. The Shuihu segment in Xuanhe yishi, possibly based on storytellers’ notes, points to some Shuihu narratives of more extended length in the late thirteenth century. Since different forms of shuoshu where prose is the primary medium—especially those devoted to historical themes such as jiangshi and pinghua55—have continued to prosper since the Song times down to the present day, it would be highly improbable that no such genres played any role in the Shuihu complex when Shuihu zaju and Shuihu cihua were flourishing.56 We may well believe that the different oral forms in the Shuihu story-cycles conditioned and supplemented each other as well as contended with each other for a niche, all enriching and enlarging the Shuihu complex as a result. In this conglomeration, zaju and cihua, as determined simply by their own formal features, had to remain in pieces of limited length, each independent and self-contained. For them there was never such a need—nor did they ever have the capacity—to absorb materials from different sources and synthesize them into a single work of a mammoth size. That task had to be fulfilled by storytelling in prose, which enjoyed the elasticity to incorporate subplots into a basically unified temporal-spatial scale and prosify materials from theatrical and prosimetric sources to fit in its own narrative discourse.57 The development of that long prose narrative itself had to be a gradual process. Each appropriation of new material would inevitably necessitate an adjustment of the preexisting narrative structure. And the augmentation of the narrative was not simply an indiscriminate quantitative expansion, as materials that would potentially jeopardize the integrity of the narrative had to be barred out. Other factors for the selection and remaking of the narrative materials, as will be considered later, could include the dynamic interaction between the raconteurs and the audiences and the particular interests of those men of letters who helped transmit the story-cycles. Eventually the prose-dominated oral narrative reached its mature form, ready to be edited and compiled for its debut in print. 50 Chapter 2 Premodern Chinese Views on Shuihu zhuan and Early Vernacular Fiction The conviction that the rise of Chinese vernacular fiction was historically related to the tradition of oral storytelling started early among Chinese scholars, although many aspects of that relationship have remained murky up to the present day. In his preface (dated 1589) to Zhong yi Shuihu zhuan, the pseudonymous writer Tiandu Waichen explicitly associates the emergence of the new narrative genre with the flourishing tradition of storytelling during the Song period:58 The ascendance of xiaoshuo [fiction] started during the reign of the Renzong Emperor of the Song Dynasty. That was an age when people enjoyed a peaceful and affluent life, with the security of the national boundaries unchallenged. After taking care of state affairs, the emperor had the leisure to have the officials in charge of court music [ jiaofang yuebu] collect and compile popular stories, set them to music, and have them performed alternately with court dramas. Since then, it has flourished both at the court and among the commoners.59 Tiandu Waichen suggests the existence of a certain kind of texts to be used for the performance. We do not know whether they were detailed scripts or sketchy promptbooks, but since they were first “collected” and then “compiled,” they could be both notational and compositional. In any case these texts were intimately tied to orality, although since they were “set to music” they were probably some species of chantefables, like guzici, taozhen, or changzhuan, and did not contain extended stretches of vernacular prose. Such observations by a scholar not too remote from the formative period of vernacular fiction seem to lend considerable weight to the theory on the relationship of the incipience of vernacular fiction to oral storytelling. Since they appear in a preface to an edition of Shuihu zhuan, obviously the writer must have believed that Shuihu zhuan exemplified that relationship. Hu Yinglin (1551–1602), who also lived early enough to witness some of the earliest known editions of Shuihu zhuan, made similar remarks that seem to confirm the origins of vernacular fiction, especially those of Shuihu zhuan, in popular orality:60 “Nowadays the talks of the streets [ jietan xiangyu] are in circulation. What are called yanyi are usually inferior to chuanqi and zaju, but Shuihu zhuan, compiled by someone with the name of Shi, a native of Wulin [present-day Hangzhou] in the Yuan times, is extraordinarily popular.”61 Yet such remarks, on the other hand, were largely responsible for an elitist contempt for ver- Told or Written 51 nacular fiction in general and Shuihu zhuan in particular, a critical attitude prevailing among conservative literati scholars of the Ming-Qing times. The new narrative genre was considered a literary form congenitally deficient, being inextricably involved with the lower strata of the culture and contaminated by the vulgarity of public entertaining. While poetry and classical prose were acclaimed as imparting Confucian values and convictions and therefore indispensably beneficial in one’s education and cultivation—paving the way in some cases ultimately to officialdom—vernacular fiction was usually dismissed as something morally uncouth, aesthetically crude, and therefore a mere pastime for vulgar tastes. “It was most painful to have nothing to do in the spare time, especially when one was satiated with wine and tea,” Lu Xun wrote sarcastically about the prevailing contempt and condescension toward vernacular fiction during the Ming and Qing periods. “And this was made worse by the fact that there were no dancing halls at that time. So people needed something to while the time away.”62 To many literati, vernacular fiction was an improper form of literature, deviating from the rules long established in the literary tradition in classical Chinese. As a result, early vernacular fiction was for a long time kept at the bottom of the hierarchy of letters. For Shuihu zhuan, the disdain for the new literary form was compounded by a moral condemnation. The Ming scholar Tian Rucheng (1503–?), for instance, claims that Luo Guanzhong, the purported compiler of Shuihu zhuan, had his offspring of three generations turned dumb as a punishment for “all the treachery and deception” in the narrative.63 In an imperial edict by the Qianlong Emperor (r. 1735–1796) of the Qing, Shuihu zhuan, along with Xixiang ji, was banned both because of its “vulgar and indecent language” ( bili zhi ci) and because of its possible “pernicious” influence on readers.64 At odds with the general disparagement of vernacular fiction, a group of scholars in the Ming and Qing periods stepped forward as champions of the new genre. Prominent among them were Li Zhi (1527–1602), Jin Shengtan (1608–1661), Ye Zhou (?–1625), Feng Menglong (1574–1645), Mao Zonggang (1632–1709+), and Zhang Zhupo (1670–1698). Their common ground for arguing for the literary excellence of vernacular fiction was a readiness to welcome a pivotal shift in the course of Chinese literature, a shift away from the unchallenged dominance of expressive genres toward a generic diversity with a considerable portion of the nation’s literature becoming more mimetic oriented.65 After the long estrangement of narrative literature from the terrestrial daily life of the common people, these critics hailed the advent of vernacular fiction, in which narrative was finally brought into much closer contact with external reality. 52 Chapter 2 Li Zhi’s philosophy, which laid the foundation for a new critical approach to narrative literature, argues in favor of truth from direct life experience as against abstract Confucian principles. According to Li Zhi, what governs a person’s relations with others is the most fundamental activities of material life. “Eating and clothing are all human ethics are about. Apart from eating and clothing, there can be no talk of ethics.”66 Naturally, this philosophical stance led to a new critical view that considered narrative literature as a representation of material reality, including all the commonplaces of the day. This new mimetic view found a most vigorous expression in the criticism on Shuihu zhuan by the commentary ( pingdian) critics.67 In a preface to the Rongyutang edition of Shuihu zhuan (1610), for instance, the writer argues that the narrative, which he attributes to Shi Nai’an and Luo Guanzhong, was modeled upon a larger Shuihu zhuan, which was nothing other than life itself:68 There was a Shuihu zhuan in life before Shi Nai’an and Luo Guanzhong made it appear in ink. . . . There were adulteresses in life, then they were instanced by Yang Xiong’s wife and Wu Song’s sisterin-law; there were procuresses in life, then they were instanced by the Old Woman Wang; and there were liaisons between mistresses and their man-servants in life, then they were instanced by Lu Junyi’s wife and Li Gu. . . . If there were no such things in real life in the first place, how could the writer have accomplished all this even if he shut himself in his study for years and worked his heart out?69 Naturally, for the critics who were mindful of literary imitation of real-life models, verisimilitude became a major principle for critical assessment. When Jin Shengtan in his commentaries on Shuihu zhuan comes to a passage that he considers a successful description of the action or character portrayal, he often bursts in exclamation: “Just like it! Very much like it!” ( Xiang! Ji xiang!) The mission of a narrative as seen by Jin Shengtan is not to record a historical or allegedly historical event, but to create a varisemblance to real life. If narrative is supposed to present a vivid and lifelike world, then, among other things, its characters should speak as people do in real life. Jin Shengtan recognized the role of individualized speech of the characters in Shuihu zhuan. To him, one of the greatest merits of the novel was its use of colloquial language, getting rid of those particles from classical Chinese such as zhi, hu, zhe, and ye. “Each individual character is made to speak in his own individual way.”70 In a similar manner, the commentator in the Rongyutang edition observes that after reading Shuihu zhuan one sees the images of various characters and hears Told or Written 53 their various voices, while “forgetting there is any mediation of language and writing.”71 This representational view of narrative literature with its emphasis on lifelike character portrayal led these commentators to a hospitable acceptance of Shuihu zhuan and its vernacular prose. The Rongyutang commentator made a good point in noting the spontaneity of the language of Shuihu zhuan. As writing approached the living voice, its mediation naturally became less noticeable than that in wenyan, which had long been divorced from the living tongue. However, the commentators usually did not associate this merit of Shuihu zhuan with the narrative’s origins in popular orality. Instead, they tended to attribute every achievement of the vernacular narrative solely to the creative mind of a writer. Jin Shengtan, for instance, considered all the effects of the narrative’s colloquial language entirely derived from the pen at the command of a literary genius: If one has in his bosom unusual talents, he must have an unusually effective pen; and if he has an unusually effective pen, he must have unusual vigor. Without unusual talents, there is no way to conceive the plot; without an unusually effective pen, there is no way to give full play to the talents; and without unusual vigor, there is no way to support the pen.72 Shuihu zhuan, according to Jin Shengtan, was a piece of writing based on a topic of the writer’s own choice, like an eight-legged essay ( bagu wen).73 He even envisaged the scene of the composition: The writer, whom he chose to believe to be Shi Nai’an, “spread out the paper and picked up a brush, selected a topic [timu], and then wrote out his fine thoughts and polished phrases [ jinxin xiukou].”74 Like writing an examination essay of the eight-legged style, the selection of a topic was said to be most important: “As long as the topic is a good one, the book will be well done.”75 The various events and divergent voices of the characters in the book were all said to center around the chosen topic. They “are used as methods to begin [qi ], continue [cheng], change direction [zhuan], and sum up [he] the composition,” in order to constitute a structural pattern that was well conceived in the first place.76 And, as “each chapter has its own principle of organization [zhang fa], so does each sentence [ jufa] and each word [zifa].”77 This eulogy suggests that everything in the narrative, including the use of colloquial speech, is governed by an ingeniously conceived writing plan. Little wonder that Jin Shengtan calls Shuihu zhuan the “fifth book of genius” and ranks it on a par with Sima Qian’s Shiji.78 Thus there were among Chinese scholars of the Ming and Qing periods 54 Chapter 2 two conflicting critical attitudes toward Shuihu zhuan. While many literati disdained it as something aesthetically inferior and morally pernicious, especially because of its association with oral storytelling, the commentators represented by Jin Shengtan endeavored to establish it as a respectable literary work by severing its ties to popular orality and explaining everything in terms of the literary imagination of a writing genius. To put Jin Shengtan’s hermeneutic strategy in perspective, the following passage by Fredric Jameson, although on a different cultural tradition, may be helpful: The great traditional systems of hermeneutic . . . sprang from cultural need and from the desperate attempt of the society in question to assimilate monuments of other times and places, whose original impulses were quite foreign to them, and which required a kind of rewriting—through elaborate commentary, and by means of the theory of figures—to take their place in the new scheme of things. Thus Homer was allegorized, and both pagan texts and the Old Testament itself refashioned to bring them into consonance with the New.79 As a champion for vernacular fiction, Jin Shengtan knew perfectly well that the only way to surmount the orthodox literati’s disdain and to elevate the status of the new genre was to distance it from its low and humble origins. The oral-derived Shuihu zhuan would not be accepted as such by the established literary world, to which its “original impulses were quite foreign.” With his “elaborate commentary,” Jin Shengtan wanted to introduce Shuihu zhuan into the pantheon of great literature, but not before he tried to strip the novel of all the traces left by its oral antecedents. Modern Western Disparagement of Shuihu zhuan The “Literary Reform” movement launched by Hu Shi (1891–1962) and other Chinese intellectuals in the early twentieth century brought Ming-Qing vernacular fiction to the foreground of literary studies. Since then, much has been written by both Chinese and Western scholars on Shuihu zhuan and other early Chinese vernacular novels.80 In the last few decades Western scholars have offered many critical insights that have vastly enhanced our understanding of the genre. When it comes to the relationship of the rise of vernacular fiction to the oral and popular traditions, however, their opinions bifurcate in a way similar to what happened among the Chinese literati scholars of the Ming-Qing periods. A slight oversimplification helps make it clear. Modern critical views on the rise of Chinese vernacular fiction have developed along two divergent Told or Written 55 lines: (1) The close ties to popular storytelling were largely responsible for both the moral and aesthetic “limitations” of early vernacular fiction; and (2) vernacular fiction is a refined art form, for it is a phenomenon of the literati culture, and its connections to oral traditions are not essential to its nature. While prevalent in its own time, the first trend of critical thinking on early Chinese vernacular fiction is now largely out of vogue, but a brief engagement with that trend is pertinent to our discussion of the incipience of the genre in general and the formative period of Shuihu zhuan in particular. Claiming to be disappointed by the formal as well as moral limitations of the early Chinese vernacular narratives,81 scholars of this camp compare these works unfavorably with their counterparts in the Western novel.82 The formal limitations of the Chinese works are attributed to what was considered as the detrimental influence from popular storytelling.83 The devices and conventions of oral performance, as the theory goes, survived the evolutionary process from orality to writing, persisted in “versions designed to be read,” and became undesirable “literary clichés” as a result.84 Such formal flaws are found to be particularly annoying in Shuihu zhuan, where those formulary phrases usually associated with the raconteur’s art are dismissed as “conventional hyperbole that add nothing to our actual apprehension,” each being “a storyteller’s cliché that could as well be omitted.”85 The “most disturbing” influence of the raconteurs on the vernacular narratives is believed to be the “heterogeneous and episodic quality of plot,” resulting in the vernacular narratives’ preoccupation with “surface reality” and failure to offer a coherent pattern of meaning.86 Again, this conviction of Chinese vernacular fiction being episodic in plot would be particularly applicable to Shuihu zhuan, a narrative that may be, as we have seen, an agglomeration of materials originally from different popular sources. Early Chinese vernacular fiction is also believed faulty in its moral standing, for which popular orality is found once again to be responsible. With most of their thematic material inherited from oral traditions, the “moral ambiguity” in some vernacular narratives is considered almost as a symptom of a congenital malformation, as much of the material from the popular sources is said to be “frankly pornographic or immoral in nature.”87 Again, it is Shuihu zhuan that has to bear the brunt of the censure, for it is called a book “full of scenes of brutality and sadism, representing only one extremity of Chinese culture.”88 Admittedly, the criticism on the formal features of the early Chinese vernacular narratives is basically true, for the narrative structure is often episodic in nature and the formulary phraseology does at its extreme appear irritating to those allergic to “literary clichés.” What one may wish to contradict is therefore not the descriptions of the characteristics of the narrative form but the declaration or suggestion that such a narrative form represents artistic inferi- 56 Chapter 2 ority. The Homeric epics, crystallized from a long oral tradition, and the numerous literary epics in Western literature from the Aeneid on were no longer meant to be sung but to be read in their book forms. They can all be called— to different degrees—“episodic” in their narrative structure, and they are full of traditional phraseology in their versification, but few people would challenge their artistic excellence. The reason is that literary critics always have to maintain a generic differentiation, by which the epics are distinguished from later forms of narrative literature and therefore not subject to the critical principles governing later narrative art. Even within that more recent genre of Western narrative literature—the novel—generic differentiation has never ceased to function as a critical principle. Works such as Gil Blas, Moll Flanders, and Roderick Random are usually accepted as good literature, and their episodicity, under the protection of the generic label of the “picaresque novel,” has not suffered much from unfavorable comparisons with a Clarissa or a Liaisons Dangereuses. If generic diversity is taken for granted and no all-governing norms can be established even within Western narrative tradition itself, then there should be no justification for imposing features of some of the Western narrative works as standards of quality upon a form of narrative literature in a totally different culture. W. L. Idema’s criticism of this type of Eurocentrism is forceful and pertinent: “We should not compare the totality of Chinese ‘fiction’ with some selected summits of western ‘fiction,’ without going into their different places in their perspective literatures, and without a thorough awareness of the difference in the concepts of literature involved. This is like comparing champagne to Shaoxing without being aware of the difference between wine and jiu.”89 Calling for an enhancement of our awareness of literary multiplicity, Eugene Eoyang adopts a different but equally vivid analogy: While we eat the peach, we should “taste the apricot” as well.90 As for the moral censure of the early Chinese vernacular fiction, especially of Shuihu zhuan, one may, once again, concede that those reprehensions are not totally unjustified. A modern reader can hardly read without a shudder the passage in Shuihu zhuan where Wu Song, in an act of vengeance, struggles to cut off a woman’s head with a dagger that has been “blunted by too much killing” (chapter 31; 2: 974). Similarly, the penalty for the adulteress Pan Qiaoyun, a brutal dismemberment executed by a cuckolded husband and his wronged friend, is to the modern sensibility definitely out of proportion to her crime (chapter 46; 3: 1527–1528). Even Song Jiang himself, who is often depicted as a man of generosity and magnanimity, can be so carried away by the malicious desire for vengeance that he has his enemy dissected alive and orders that his heart be prepared as an hors d’oeuvre (chapter 41; 3: 1342–1343). We might, however, compare this spirit of vengeance with that in the Told or Written 57 Iliad, where part of the narrative is about Achilles’ terrible anger ignited by Hektor’s killing of his beloved companion Patroklos. When Achilles has finally taken his revenge by killing Hektor, he and his comrades-in-arms ruthlessly mutilate the Trojan warrior’s body before tying it to a chariot and letting the running horses drag it with the head tumbling in the dust. To get a sense of the spirit of vengeance that is as irrational and violent as anything in Shuihu zhuan, let us quote the verses at some length: He spoke, and pulled the brazen spear from the body, and laid it on one side, and stripped away from the shoulders the bloody armor. And the other sons of the Achaians came running about him, and gazed upon the stature and on the imposing beauty of Hektor; and none stood beside him who did not stab him; and thus they would speak one to another, each looking at his neighbor: “See now, Hektor is much softer to handle than he was when he set the ships ablaze with the burning firebrand.” . . . In both of his [Hektor’s] feet at the back he [Achilles] made holes by the tendons in the space between ankle and heel, and drew thongs of ox-hide through them, and fastened them to the chariot so as to let the head drag, and mounted the chariot, and lifted the glorious armor inside it, then whipped the horses to a run, and they winged their way unreluctant. A cloud of dust rose where Hektor was dragged, his dark hair was falling about him, and all that head that was once so handsome was tumbled in the dust. (XXII, ll. 367–403)91 It is significant to note that such details did not go unnoticed in subsequent ages. In eighteenth-century England, when Henry Fielding was trying to create a new type of fiction modeled on the classical epic, some of his fellow writers found the moral aspect of the Homeric poems unacceptable. Alexander Pope, despite the homage he paid to the Greek minstrel in the famous line that “Nature and Homer were . . . the same,” found a “most shocking” thing in the Homeric poems, “that spirit of cruelty which appears too manifestly in the Iliad.”92 Samuel Richardson even held the Iliad responsible for the belligerent spirit of subsequent times: “I am afraid this poem, noble as it truly is, had done some infinite mischief for a series of ages; since to it, and its copy the Eneid, is owing, in a general measure, the savage spirit that has actuated, from the earliest ages to this time, the fighting fellows, that, worse than lions and tigers, have ravaged the earth, and made it a field of blood.”93 I am not suggesting that Shuihu zhuan is a work derived from a kind of orality similar to that which produced the Homeric verses. As I stated earlier, 58 Chapter 2 popular orality in China that led to the rise of vernacular fiction was enormously different from what Walter Ong terms the “primary orality” of Homeric Greece. The comparison between the violent spirit in Shuihu zhuan with the vengeful wrath in the Iliad is meant to show that, if the early Chinese vernacular fiction can sometimes appear wanting in civility to our modern sensibility, it is in that respect not very different from the Iliad, which has been a major source of inspiration for Western narrative tradition. This is not to suggest that oral people are inevitably more violent and brutal than the literate. Yet as we have learned from scientific research, people in an oral culture are more susceptible to “states of confused excitement” due to “a lack of systematized fancy or delusions acting as ego defenses.”94 Achilles’ wrath may have derived from such a “blind frenzy,” which, due to “confused cultural memories,” became a persistent convention in literary epics all the way down to Orlando furioso, even when it “had grown less understandable and more palpably histrionic as social conditions and personality structures changed with the growing effects of literacy.”95 In a similar way, it is possible that the violent revenge in Shuihu zhuan, often presented in hyperbolic words, is a convention that can be traced very far back in the oral tradition. Like the furious violence of Achilles or of Orlando, Wu Song’s wrathful killing gives him an epic stature. Such characters are products of oral noetic processes, which operate most effectively with outsize figures or “ ‘heavy’ characters, persons whose deeds are monumental [and] memorable.”96 Hyperbolic rendering of the violence would help create a type of such “heavy” characters and achieve an effect of “sensationalism” in the delivery of the story. So, just as the vengeful savagery in the Western epics should be viewed in the context of the conventions they inherited from their previous oral mode of existence, so should that in Shuihu zhuan. Recent Western Reassessment of Shuihu zhuan In the last two or three decades, the study of Chinese vernacular fiction in the West has reached an unprecedented level of sophistication. Indeed, all students of the genre today, in the West as well as in China itself, have to be grateful for the brilliant scholarship of a new generation of Western sinologists who have promoted Chinese vernacular fiction to the status of respectable literature that it fully deserves, washing clean the stigma of “limitations” and revealing instead its limitless artistic possibilities. Amid the new boom of the study of Chinese vernacular fiction, there emerged another trend of critical thinking on the origins of the genre, which started with the much-needed redress for the failure to consider the best achievements in vernacular fiction as part of the mainstream culture. The scholars have challenged, forcefully and rightly, the older view that regarded vernacular fiction indiscriminately as a form of popular entertainment and put the Told or Written 59 genre virtually in the category of subculture. Andrew Plaks, for instance, complains that scholars of Chinese vernacular fiction in the past overemphasized its connection with the popular tradition to such an extent that “the more crucial role” of historiography has been neglected. From there he argues that the great works of Ming vernacular fiction can “lend themselves to most meaningful interpretation when they are treated not as examples of a ‘popular’ counter-culture, but rather as major documents in the mainstream of Ming and Qing literati culture.”97 In his monumental Four Masterworks of the Ming Novel, Plaks has formally established the concept of the “literati novel” (wenren xiaoshuo) with the argument that the masterworks in vernacular fiction, like the “literati painting” (wenren hua), display many of the presumptions, aesthetic expectations, and a sense of self-realization that were typical of the literati culture.98 W. L. Idema also calls for a more discriminate treatment of the works in vernacular fiction. He argues for a division of the colloquial fictional narratives in the Ming and Qing periods into two different categories, which he designates as “chapbooks” and “literary novels” respectively. While putting works like Sanxia wuyi and Shuo Tang in the first category, Idema uses the second label for those more esteemed works in the genre.99 In this critical context, the role of popular orality in the formative period of vernacular fiction has often to be downplayed. Indeed, to the older trend of disparagement of Chinese vernacular fiction, the discount of the pertinence of orality may be an essential and necessary antidote. Yet one has to wonder whether the repudiation or disregard of its pretextual life is indeed the right price that early vernacular fiction should have to pay in order to earn due respectability. To be sure, nobody really denies the connection of early vernacular fiction to popular orality, but there seems to be a reluctance to acknowledge that connection as a shaping force on the new literary genre, as the ties of a text to its possible oral predecessors cannot always be proven by tangible evidence. Understandably, people feel compelled to focus instead on the text-reader relationship in literary reception, attenuating the relevance of an early vernacular narrative’s possible preprint existence. The argument is plausible: Since the text was meant to be read, it should be considered as written literature. And that argument seems to be corroborated by the rightful abrogation of the traditional theory that huaben and pinghua were scripts or promptbooks used in actual storytelling.100 Although Shuihu zhuan is not called a pinghua, this argument can easily be applied to the novel as well: Not only was the work meant for reading, the different recensions of the work—the so-called fanben (full editions) and jianben (simplified editions)—might be catering to different types of readership! It is indisputable that the printed texts of early vernacular narratives were 60 Chapter 2 meant for a reading public from the very start. But to reject the old belief that a huaben or a pinghua was a storyteller’s scenario is one thing; to repudiate the kinship of early huaben or vernacular sections in pinghua to the oral tradition is quite another. A huaben or a pinghua does not have to be a storyteller’s promptbook to be related to an oral tradition. The text could be situated at any point along the line of transition from the oral to the written without being directly used as a promptbook in an actual oral presentation.101 Indeed, the text of an orally derived narrative is in itself a paradox, from which no modern student of any oral literature from an age predating tape recording would be able to find an escape. For however deeply the object of one’s study was once immersed in an oral tradition, one would still have to have some written or printed text on the desk that has evolved one way or another from what was once told by word of mouth, and that text is, of course, meant to be read. Still, while it is not only justified but also necessary to keep in mind the difference between the text and the voice in terms of the mode of transmission and the psychological dynamics of reception, to sever the ties of the text from its oral precursor on the ground that the text was meant to be read is simply to annihilate altogether any orality studies that predate acoustic recording. The printed texts of the Iliad and the Odyssey—let us return to the most convenient examples—have been for many centuries meant for a reading public, of course; but if we stress the distinction between the text and the singing of the minstrel to an undue extent, we will deprive ourselves of the only access to the Homeric oral world. That is, the original oral mode of “production” of the poems cannot be denied simply on the grounds of the subsequent literate/literary mode of “consumption.” Some early vernacular texts, especially the vernacular sections in the pinghua, feature a prose that is obviously uneven and awkward, which means that they are not quite textual equivalents to the stories once told in oral prose. One important reason was the predominance of wenyan. In the tentative writing of the vernacular prose, the influence of wenyan would loom large and its interference could sometimes be irresistible. In the case of pinghua, the limited vernacular prose is often outweighed by the textual incorporations from classical sources. Yet while such borrowings contribute to the pinghua, which are generally considered vernacular works, they actually remain in the classical language and contribute nothing to those sections where the language is more vernacular. In other words, while most of the wenyan segments of the pinghua texts were compiled on the basis of a variety of textual sources, the same cannot be said of the vernacular sections, which are much more likely to have had some kind of kinship to orality. Scholars have argued, quite justly and properly, that Shuihu zhuan and other Ming masterworks in vernacular fiction brought Chinese narrative literature Told or Written 61 to an unprecedented artistic level. Recent critical analyses, especially those by Andrew Plaks in his Four Masterworks of the Ming Novel, have greatly enhanced our knowledge of the textual intricacy and artistic sophistication of these narratives. We have come to realize that these works, once dismissed by some literati readers as coarse and crude, are in fact pregnant with ironic and allegorical meanings.102 Yet the exposition of the textual richness, especially the rhetorical stance of irony, sometimes carries with it the implication—if not explicit proposition—that the possible pretextual life of the early vernacular narrative no longer matters. The textual sophistication of Shuihu zhuan, for instance, can be taken as a sign that the work, by thoroughly transforming the “source material,” had far distanced itself from its oral past, which was too primitive to bear much relevance to the work’s textual existence.103 Since what is interpreted as an ironical rhetorical stance is taken as counterevidence for the kinship to popular orality, one may wish to raise the question whether a narrative of oral or popular origins is categorically incapable of being interpreted ironically or allegorically. A case in Western literature that may come to mind at this point is the Neo-Platonist Porphyry’s reading of the Odyssey, in which Homer’s verse describing a cave in Ithaca with one entrance for men and the other for gods104 is interpreted as the poet’s allegorical presentation of the cosmos itself.105 But a weightier example will be the tradition of the Christian exegeses following the Paulian aphorism that “The letter kills, but the spirit gives life.”106 Origen, for instance, admonishes the readers of The Song of Songs that it will be detrimental to understand the text literally in terms of the flesh: For if he does not know how to listen to the names of love purely and with chaste ears, he may twist everything he has heard from the inner man to the outer and fleshly man and be turned away from the Spirit to the flesh. Then he will nourish in himself fleshly desires, and it will seem because of the divine Scriptures that he is impelled and moved to the lusts of the flesh.107 Similarly, St. Augustine draws distinctions between things and signs and between the things to be used and the things to be enjoyed,108 on the basis of which he proceeds to exhort: “He who follows the letter takes figurative expressions as though they were literal and does not refer to the things signified to anything else.”109 In other words, the Scriptures, according to the early Church fathers, must not be read straightforwardly. Rather, they are also “roundabout writing[s] with deep seated meanings” (shenwen qubi), to adopt Jin Shengtan’s terminology.110 62 Chapter 2 If we accept the well-established theories that both the Homeric epics and much of the Christian Scriptures are of oral origin (although the modes in which they were tied to the oral traditions differed from each other), it will seem that the ironical and allegorical interpretations are by no means proofs of the literary nature of the origin of any text. Nothing prevents an orally derived text from being interpreted into something different from its most obvious meaning. Inversely, a text’s capacity for being interpreted ironically or allegorically does not preclude the possibility of its oral provenance. We should not take for granted any connection between the type of the textual meaning and the mode of the textual origination. If a work, at a certain point in the history of its reception, acquires an elite status in literature, it is made so by the intellectual climate of the time. Many a work of oral or popular origins in Western literature has found its way into what is called the “literary canon.” In Chinese literature, the best-known example is, of course, the Shijing, a collection of folk lyrics and ceremonial songs of the court during the Spring and Autumn period, which was later appropriated by the mainstream culture and became a classic for the literati. After that, some of the poems received many rounds of allegorical interpretations. That, however, does not contradict the fact that many poems in the Shijing are of oral provenance. On the Bifurcated Critical Attitudes toward Shuihu zhuan and Its Oral Provenance This chapter has presented an overview of the formative stage of Shuihu zhuan in some genres of oral performance and discussed divergent views on the oral provenance of early vernacular fiction in general and that of Shuihu zhuan in particular. I have noted that the two different attitudes toward Shuihu zhuan’s preprint life that were prevalent during the Ming-Qing periods have largely been taken over by our own age. While some scholars denigrate the work and other early vernacular narratives for their inceptions in popular orality, others praise them for being able to break away from oral sources or for being less involved with popular materials than generally supposed. Despite their different assessments of the narrative art, both groups seem to agree that the oral tradition belonged to a humble and primitive past, a mire from which Shuihu zhuan would have to lift itself before reaching the solid ground of fine literature. This bifurcation of the critical views on Shuihu zhuan coincides with such a binary logic: The work should be considered either a work derived from a tradition of popular storytelling or an artifact produced by sophisticated literary sensibility. It cannot be both. Like the theory of orality once held by A. B. Lord, which refuses to see any intermediate zone between oral and literate Told or Written 63 modes of composition and transmission, this binary logic puts Chinese literary tradition and the tradition of popular orality from the Southern Song to the Ming period into two mutually exclusive categories. Yet in the social conditions of premodern China, different cultural forces were constantly reshaped and reshuffled owing to the drastic social mobility. Especially, when Chinese narrative literature needed to explore new possibilities beyond wenyan, the forces of popular orality happened to be firmly allied with the forces for change within the literary culture. Under these historical circumstances, Shuihu zhuan as the earliest vernacular novel was a great literary innovation launched by men of letters, and it could be such precisely because much of its narrative discourse had originated in the spoken words of the storytellers. It is therefore both an orally derived narrative to a great extent and a work conceived by a literary sensibility, and it was this dual nature of its origins that made the narrative both truly vernacular and truly “literary,” marking the inauguration of vernacular fiction as a new literary genre. As I noted earlier, in more recent scholarship on early Chinese vernacular fiction, the downplaying of the impact from the oral tradition was often meant to be a rectifying measure for an earlier tendency to slight the genre. To correct a wrong, overshooting the target is perhaps inevitable. The seeming negligence of the role of the oral tradition is therefore quite understandable. It is even more so when we take into account the fact mentioned earlier: Whatever we say about the oral antecedents of the earliest vernacular narratives, the facts that we know about the evolution of those narratives in the popular genres remain meager. In the case of Shuihu zhuan, that the narrative developed from an oral story complex seems beyond any doubt, but many of the details about that development—without discovery of further evidence—will remain hypothetical. The brief discussion of the relevant oral genres earlier in this chapter is a start, but if the study of the impact from orality is based exclusively on the tenuous evidence of those few historical facts, one might end up making much ado about nothing. To certify the role of oral storytelling in the making of the narrative, and especially to substantiate my conjecture of a process of synthesis of the Shuihu stories in an oral narrative mostly in prose, I now have to turn to the text of Shuihu zhuan itself. Out of the Margins Ge, Liangyan Published by University of Hawai'i Press Ge, L.. Out of the Margins: The Rise of Chinese Vernacular Fiction. Honolulu: University of Hawai'i Press, 2001. Project MUSE., https://muse.jhu.edu/. For additional information about this book https://muse.jhu.edu/book/8086 Access provided by University of Washington @ Seattle (14 Apr 2018 21:02 GMT) Introduction In both the West and the East, the relationship of writing to speech has a direct bearing on literature, especially narrative literature. Around the eighth century b.c., the Greeks adopted the writing shapes from the Phoenician syllabary and invented what is known to be the earliest alphabetic system. The advent of the Greek writing system marks the beginning of a new era of Western civilization, with written literature as one of the most immediate results. The Homeric epics, hitherto existing only orally, now became written, although there is no consensus how that was accomplished. The significance of this process has been most succinctly summarized by Eric Havelock in the title of one of his books, The Muse Learns to Write. Writing, as the newly arrived visitor, knocked at the door of the oral world and was hospitably received by the Muse, daughter of the goddess of Memory and mistress of oral literature. Only with the integration of the old and the new did it become possible for the Homeric verses to appear as literature, in the literal sense of the word. In a literate age, there is another kind of relationship between writing and orality that has received less scholarly attention. The positions of the old and the new are now reversed. Writing, by nature resistant to changes, can become the agent for what is old and conservative. It necessarily lags behind the more dynamic and fluid development of orality and therefore only represents yesterday’s language. Once the gap grows excessively wide, writing becomes senile and decrepit, to be rejuvenated only by a drastic dose of orality. On the other hand, the rapidly changing oral speech persistently seeks new surrogates in written script. Voice’s never-ending quest for a chirographical body results in constant reinvigoration of writing. Such changes in writing have been a driving force behind the evolution of literary forms. Very often, when the linguistic medium in literature was realigned with the latest living speech, new genres and styles were born, keeping literature fresh and vigorous. Histori- 1 2 Introduction cally, therefore, what happened between writing and orality was by no means a one-way street. If the Muse had to learn to write in order to bring ancient Greece out of the preliterate “dark age,” the forces of literate culture afterwards also had to draw vitality from spoken words: The Muse-turned-writer had to return to the world of orality and learn—or relearn—to “babble.” In the history of Western literature, a significant moment of the writer learning to babble was in the thirteenth and fourteenth centuries, when men of letters in Europe who had written exclusively in Latin—which was predominantly a literary language seldom spoken out of the church—turned to orality and brought writing more in line with speech. The results were vernacular works such as Giovanni Boccaccio’s Decameron and Geoffrey Chaucer’s Canterbury Tales, which heralded the ascendance of European national literatures written in the vernaculars. In China, due to particular social and cultural conditions and especially to the unique nature of the Chinese script, vernacularization was a much longer and more arduous process. While the bulk of Chinese literature up to the end of the nineteenth century was written in classical Chinese, or wenyan, vernacular literature had existed stubbornly and persistently for many centuries. Yet the written vernacular, or baihua, did not become a full-fledged literary language until the emergence of voluminous novels in the Ming period (1368–1644), especially the works generally known as the “Four Masterworks” (si da qishu), namely Shuihu zhuan (Water Margin/ Outlaws of the Marsh), Sanguo yanyi (Romance of the Three Kingdoms), Xiyou ji ( Journey to the West), and Jin ping mei (Golden Lotus/The Plum in the Golden Vase). To be sure, the dominance of wenyan was to continue in most of China’s cultural and literary sectors until the early twentieth century, when it was finally replaced by the vernacular as the nation’s standard written language—but in narrative literature that revolutionary change had started a few centuries earlier. After the “Four Masterworks,” many narrative pieces, especially short tales such as those by Pu Songling (1640–1715) and Ji Yun (1724–1805), continued to be written in the classical language, but the use of the vernacular became increasingly a genre convention in prose fiction, especially in the novel.1 Among the “Four Masterworks,” it was Shuihu zhuan that played the most crucial role in establishing the written vernacular as the new literary language. Two separate pages from an early edition of Shuihu zhuan, with the title of Jingben Zhong yi zhuan, were found accidentally in the Shanghai Library in 1975. Most scholars agree that the fragments are from a Zhengde (1506–1521)-Jiajing (1522–1566) edition, if not earlier, possibly the earliest among all extant editions of Shuihu zhuan, full or incomplete. From a different Jiajing edition entitled Zhong yi Shuihu zhuan, five chapters are extant, previously in the possession of Zheng Zhenduo (1898–1958) and now housed in the Beijing Library. Introduction 3 The earliest known reference to a printed edition of Shuihu zhuan occurs in the catalogue entitled Baichuan shuzhi, compiled by the Jiajing scholar Gao Ru, which lists a hundred-chapter Zhong yi Shuihu zhuan.2 Whether it was of the same edition as the Zhong yi Shuihu zhuan fragments in the Beijing Library remains unknown. Since Baichuan shuzhi carries a preface dated 1540 (the nineteenth year of the Jiajing reign), we can safely mark 1540 as the latest possible year for the earliest edition of Shuihu zhuan. The earliest reference to Jin ping mei was made by Yuan Hongdao (1568–1610) in his letter dated 1596 to the famous painter and calligrapher Dong Qichang (1555–1636). It is indicated in the letter that Yuan had obtained part of a manuscript copy of the novel from Dong.3 During the following two decades, manuscript copies of the work may have been circulating among literati readers, as Shen Defu (1578–1642) reports in Wanli yehuo bian that he had copied the entire manuscript from Yuan Zhongdao (1570–1623), Hongdao’s younger brother, in 1609.4 The work’s earliest known edition in print is Jin ping mei cihua, with one of the two prefaces dated 1617–1618 by a pseudonymous Dongwu nongzhu ke. The year 1617 is therefore the earliest possible date for the edition, which was antedated by the Jiajing editions of Shuihu zhuan by at least two-thirds of a century. When Jin ping mei was written, the writer may have looked to Shuihu zhuan for a model, as evidenced by the fact that the story in Jin ping mei itself sprouts from an episode in the earlier work. Both Xiyou ji and Sanguo yanyi had a long process of textual evolution similar and, for the most part, temporally parallel to that of Shuihu zhuan. However, extant fragments from budding textual precursors notwithstanding, the earliest exemplar of Xiyou ji in its fully developed form—the Shidetang edition— is dated 1592, at least half century later than the Jiajing editions of Shuihu zhuan. The earliest known edition of Sanguo yanyi, entitled Sanguo zhi tongsu yanyi, carries two prefaces, one by Jiang Daqi and the other by Zhang Shangde, which are dated 1494 and 1522 respectively. In all likelihood, the edition was contemporaneous with or slightly predated its Shuihu counterpart.5 Yet while Shuihu zhuan appears in a language that is distinctly vernacular, Sanguo yanyi is called a vernacular novel only when the word vernacular is more generously defined, as it employs a linguistic medium that mingles baihua with simple wenyan.6 Shuihu zhuan is, therefore, China’s earliest full-length fictional narrative in the true vernacular prose. If we take into account the fact that the earliest anthology containing vernacular short stories, Liushi jia xiaoshuo (Sixty Short Stories) was not published until around 1550,7 indeed we may call Shuihu zhuan the trailblazer for Chinese vernacular fiction at large. To say this, of course, is not to repudiate the efforts in vernacularization prior to Shuihu zhuan; on the 4 Introduction contrary, as we will see, Shuihu zhuan is to be considered in the present study as part of a larger vernacularizing process. It is in Shuihu zhuan, however, that vernacular prose extends to an unprecedented length and the degree of vernacularity ascends to an unprecedented level. Writing and speech, divorced in wenyan, are now brought much closer to each other in the full-fledged vernacular prose of this novel. The last statement may call for a little qualification here. Chinese characters, whether employed in wenyan or baihua, are not a phonetic system of writing like alphabetic scripts. The peculiar relationship of Chinese characters to speech has long been a focus of the scholarly debate. The terminology that has been applied to Chinese characters has almost become a nomenclatorial kaleidoscope. They are called variously “sinographs,” “pictographs/pictograms,” “ideographs/ideograms,” “logographs,” “lexigraphs,” “morphographs,” “phonograms,” “phonosemantics,” “phonoideograms,” “logo-syllabics/word-syllabics,” and many others.8 This messy situation reflects the formidable difficulty of defining the Chinese script in terms of its relationship to the semantic and phonetic aspects of the language. Contending that no writing system represents ideas without regard to sound, John DeFrancis argues that terms such as “ideographic” should be consigned to “the Museum of Mythological Memorabilia along with unicorn horns and phoenix feathers.” DeFrancis designates the Chinese script, which he believes to be basically phonetic, as “morphosyllabic,” but even he has to admit that “the Chinese syllabary is only partially reliable in representing the pronunciation of the Chinese characters.”9 More recently, William Hannas tries to reconcile the “aphonic” and the phonetic views of the Chinese writing system by proposing that a character is part of a morpheme, which he defines as the conventional interface between meaning and a phonological form.10 Hannas, too, is quite aware of the inefficiency and inaccuracy of the phonetic denotation by the characters. “Lacking a systematic relationship between symbol and sound, or even an efficient protocol for describing the structure of its units,” Chinese writing “places an enormous burden on the user in comparison to alphabetic systems,” as Hannas puts it.11 The written vernacular, which inherits many—in fact most—of its characters from classical Chinese, may be no more phonetic than the latter. But what is relevant here is not each written symbol’s representation of an isolated sound, but the correspondence between a string of written symbols and a string of sounds in a speech utterance. A character in wenyan, of course, was and is pronounceable, and one of the most important pedagogical methods in traditional Chinese education was making students read aloud writings in wenyan until they could recite them from memory. Most of the characters, the so-called xingshengzi (semantic-phonetic composites), even have a component as a cue Introduction 5 for pronunciation.12 But if one reads aloud a line of characters in typical wenyan, the string of sounds will usually represent a highly stylized utterance. On the other hand, the written vernacular can be much more accommodating to common speech, although it can be stylized in different ways as well. Characters in the written vernacular can therefore correspond to speech sounds, usually one character to one syllable, although each character is not phonetic in the same way as a letter or a syllabogram in an alphabetic script. It is in this sense that we can say that writing was finally realigned with the speaking voice in Shuihu zhuan, bringing to fruition the repeated and tenacious attempts of vernacularization for several centuries. This book is devoted to that historic moment when written vernacular prose—the product of a gradual acclimatization of writing to speech—became established in Shuihu zhuan as the new literary language for Chinese narrative literature. The argument here is that Shuihu zhuan, in its fullest and artistically most complex form, is the product of long-term interaction of oral and written traditions, reflective of the fascination with oral language used by generations of professional storytellers. Although the thorny issue of recensions will be addressed, this is not a study of the textual history of Shuihu zhuan punctuated by the different fanben (full-version) and jianben (simple-version) editions in post-Jiajing times. Rather, it is more about the poesis of the text, or the pre-Jiajing life of the narrative, before the fanben-jianben scramble ever had a chance to start. Chapter 1 is a review on the long and arduous process of vernacularization in literary prose during the few centuries prior to the fanben editions of Shuihu zhuan. Vernacular texts associated with various oral and performative genres in different historical periods will be discussed. Such a survey is necessary, as it informs us of the general linguistic conditions for the evolution of Shuihu zhuan, in both oral and textual forms. As the discussion will demonstrate, what happened to many other works in early vernacular literature may have been quite similar to the cumulative textualization of Shuihu zhuan. This makes the evolution of the Shuihu zhuan not an anomaly but representative of the general trend of vernacularization in literary prose. In chapter 2, efforts will be made to reclaim at least part of the formative process of the narrative in Shuihu zhuan. A number of oral and performative genres and their possible roles in the evolution of the narrative will be discussed. After that, the chapter will provide a discussion of the historical and contemporary critical views on the formation of the narrative, focusing on the divergent opinions on the relationship of the narrative discourse to the oral tradition. I will argue against the either-or binary logic that often lurks behind the divergent critical judgments on Shuihu zhuan’s connection to popular orality. 6 Introduction And I will propose that Shuihu zhuan, as the earliest vernacular novel wrought by the long-term orality-writing interplay, should be considered both a work of oral provenance and a great literary innovation by men of letters. The subsequent chapters will substantiate that proposition. Chapter 3 discusses the oral provenance of the narrative discourse in Shuihu zhuan. One conceivable way to determine the oral derivation of the narrative would be to track down in reverse chronological order all its textual precursors, but that approach is obviously unfeasible, as the only extant text of Shuihu narrative before the fanben editions of Shuihu zhuan itself is the sketchy account of the Liangshan rebellion contained in Xuanhe yishi (Unrecorded Events of the Xuanhe Period), a historical narrative probably published in the early Yuan dynasty (1279–1368). Everything else was lost, including the earliest stories featuring Shuihu figures that circulated during the Southern Song period (1127–1279). Consequently, it is “difficult to show any direct textual connections to these materials, even to the Xuanhe yishi segments that are sometimes taken as a kind of blueprint for the novel,” as Andrew Plaks rightly points out.13 But the difficulty goes even deeper. Even if we could collect all necessary textual clues that would enable us to trace all the way back to the very earliest textual prototype, we would still have to face the same question about the nature of that version: Was it written or derivative of the oral? In the present study, therefore, what are considered as features inherited from the novel’s oral antecedents will be demonstrated through other means. Chapter 3 examines the thematic patterns of the narrative that are frequently recurrent, with numerous stretches paralleled either within the novel itself or elsewhere in early vernacular literature. The recurrence of the thematic patterns will be discussed in terms of the dialectic between the uniform and the multiform in the economy of the oral mode of story making. A major factor in the twentieth-century study of oral and oral-derived literature is the Parry-Lord theory on the making of oral epics. The universality of the theory has, however, met considerable resistance. One of the controversial issues is the very definition of orality itself. Parry-Lord studies seem to suggest a “pure” type of oral culture—what Walter Ong calls the “primary orality”—as opposed to an equally “pure” type of literate society. A singer of tales, as Albert Lord believed in his earlier works, can never be a literate or semiliterate man; if so, he would cease to be a singer. Therefore, a text is “either one or the other . . . either oral or written.”14 This categorical separation of orality from literacy, where the oral sensibility and the mode of composition-in-performance precludes literate mentality and the mode of writing, may be an accurate description of the Yugoslav oral world, where Lord did the field study, but the absolute polarization of orality to writing may not be univer- Introduction 7 sally applicable. For instance, Ruth Finnegan, with the support of her fieldwork in the South Pacific, forcefully challenges w...
Purchase answer to see full attachment