The Tale of Matsura

Front matter

The Tale of Matsura

Translated with an Introduction and Notes by Wayne P. Lammers

IJI

Contents

List of Charts Preface

Chapter One: Fujiwara Teika and Matsura no Miya Monogatari Chapter Two: The Aesthetic of Yen in a Narrative Context Chapter Three: The Manuscripts and Texts

The Tale of Matsura: A Translation of Matsura no Miya Monogatari Book One

Book Three

Appendix A: Evidence on Dating Matsura no Miya Monogatari Appendix B: The Authorship of Matsura no Miya Monogatari Bibliography

Indexes

List of Charts

Chart I: External Evidence for Dating 164 Chart II: Internal Evidence for Dating 166 Chart III: The Possible Existence of Different Versions 169

Preface

'This book derives from my doctoral dissertation, “A Poetic Ideal in a Narrative Context: Fujiwara Teika and The Tale of Matsura (Matsura no miya monogatari),” submitted to the University of Michigan in 1987. I have expanded chapter one to include brief remarks on aspects of the tale left untouched in my dissertation, and have added further elaboration at several places in chapter two. I have condensed former chapters three through six and combined them with the crucial analytical portions of the three original appendixes into two, much shorter, new appendixes; the analysis here supersedes the earlier in several respects but does not affect the essence of the argument. The translation has also seen minor adjustments throughout.

The translation is based on Hagitani Boku, ed., Massura no miya monogatari (Kadokawa, 1970), which is the only edition of the text to supply annotations and a complete modern Japanese translation. Further information on this text can be found in chapter three. My translation and notes owe a considerable debt to Hagitani’s work even where I have not provided explicit references; in general, I have given individual citations only where it seemed particularly important to underscore my adherence to, or signal my departure from, his reading of a passage.

Although I have striven to make the translation readable entirely on its own, without reference to notes, I have also made no effort to minimize the notes either in number or in size. The tale presumes a vast knowledge of prior Japanese and Chinese literature on the part of the reader, and it cannot be appreciated in its full complexity, even by the specialist, without supplementary information close at hand. My hope is that this has resulted in a work more useful to advanced students and specialists of Japanese literature.

In Romanizing Japanese I have followed the modified Hepburn system (z rather than m immediately before 4, p, and m), and for Chinese I have used Wade-Giles. All Japanese and Chinese names have been given in their native order, surname first. For Japanese, I have preserved the particle zo when it appears between surname and personal name in the translation, but in the introduction and notes I have followed the recent practice of retaining it only after surnames of one or two syllables—thus, Abe no Nakamaro, but Fujiwara Teika. For Chinese personal names and titles of works I have consulted The Indiana Companion to Traditional Chinese Literature (William H. Nienhauser, Jr., ed.) where possible.

xi xii PREFACE

I wish to acknowledge with deepest gratitude the generous encouragement and help I received from the late Robert H. Brower in preparing this translation and study. I benefited immeasurably from his learned counsel through many years of graduate study and dissertation research, and had hoped to profit from many more years as I made my way out from under his wing to establish myself as a full-fledged member of the scholarly community, getting to know him better as a colleague and friend and not just as a teacher. His sudden illness and death thus came as a particularly severe blow to me. His critical eye saved me from many an embarrassing mistake in writing my dissertation, and I do not doubt that it would have further improved this book had the manuscript been ready in time for him to review. In hope that what has resulted without benefit of his continued guidance is nevertheless worthy of him, itis to his memory that I dedicate this book.

I would also like to thank several other individuals for their various contributions: Tsai-fa Cheng, Tse-tsung Chow, and Joseph S. M. Lau for kindly answering questions about matters Chinese; James O'Brien for his invaluable comments on the translation; Robert Danly for his always helpful advice; and my wife Cheryl and son Michael for their patient and understanding support through all the years it has taken to bring this work to completion. Needless to say, I alone must bear the burden of any errors or infelicities that remain uncorrected.

Finally, I am most grateful for the support provided this study at the dissertation stage by grants from the Fulbright-Hays Doctoral Dissertation Research Abroad Program and the Horace H. Rackham School of Graduate Studies at the University of Michigan.

Chapter One: Fujiwara Teika and Matsura no Miya Monogatari ki

Matsura no miya monogatari (“The Tale of Matsura,” ca. 1190) is a classical Japanese tale or romance that belongs to the same category of courtly fiction as Murasaki Shikibu’s unsurpassed masterpiece, Genji monogatari (“The Tale of Genji,” ca. 1010). When compared with most of the best-known works of its genre, however, Matsura no miya monogatari stands out in striking contrast: Whereas the typical monogatari is set in the Heian period (794—1185) and in the Japanese capital of that time, Massura no miya monogatari is set in the period before Japan’s first “permanent” capital was established at present-day Nara in 710, and most of its action takes place in China. Whereas the typical monogatari centers almost entirely on affairs of the heart between men and women, nearly half of Matsura no miya monogatari is devoted to politics and government, including a violent succession struggle with armies marching halfway across China and back. And whereas the typical monogatari is fundamentally realistic, with only minor intrusions of the supernatural, the progress of Matsura no miya monogatari depends on supernatural interventions almost from beginning to end.

‘The work is in three maki (“scrolis” or “books”—in this case the latter), but, for purposes of outlining, the plot may be divided into five parts. First comes a sort of prologue, introducing the protagonist Ujitada as a child and in brief order taking him past the usual milestones in the early life of an idealized monogatari hero: he excels in Chinese poetry at the age of seven, demonstrates precocious mastery of musical instruments, undergoes a coming-of-age ceremony at the age of twelve, receives appointment to three highly demanding offices at the age of sixteen, and then experiences the awakening of first love. Ujitada’s hopes in love are stopped short, however, when the mother of the lady who holds his heart captive accedes to the emperor’s wish that she send her daughter to court. A short time later Ujitada is appointed deputy ambassador of an embassy scheduled to go to China the following year.

The second part tells of Ujitada’s early experiences in China, including the immediate and special favor he receives from the emperor for demonstrating his superior talents and great knowledge. On a night away from court, he meets an old man playing the kin, the seven-stringed Chinese koto, or zither,! and the

1. Two kinds of Chinese koto are commonly mentioned in monogatari: the seven-stringed kim, and the thirteen-stringed s#d. The kin was considered especially difficult to play, and at least in Japan had developed a certain association with mysterious supernatural forces (see Lammers 1982, 139-40 and passim, and Pollack 1983, 372-73 for discussions of the use of the kim in Utsuho monogatari and Genji monogatari, also see translation, n. 87).

old man tells him that he is destined to transmit the secrets of the instrument to Japan. Following the old man’s instructions, Ujitada goes to Mount Shang, a place protected by divine sages, to receive the teachings from the beautiful Princess Hua-yang. Predictably, he falls in love with her, and she returns the sentiment, but this love, too, is destined to be short-lived and unfulfilled. One night not long after the transmission of the teachings has been accomplished, the princess reveals that she will soon die, and tells Ujitada that when he returns to Japan he must go to Hatsuse and conduct a certain series of services before the image of Kannon there to ensure that they will meet again. Back in her own apartments, she sends her zither flying off into the heavens, then lies down as the life rapidly ebbs from her body. By morning she is dead.

In the meantime, the Chinese emperor, who has been ailing, has spoken to Ujitada prophetically about his own impending death, exhorting Ujitada not to shrink from the role that will be thrust upon him by subsequent events. The emperor’s death follows closely on the princess’s, and marks the beginning of the third part of the tale: a violent succession struggle between the emperor's brother and his son, the rightful successor. The new.emperor, who is still a child, is forced to flee the capital with his mother, the empress dowager, and they are pursued to the west by the rebel armies as their own loyalist armies dwindle rapidly along the way. Ujitada goes with the loyalists in this flight. When all looks hopeless, the empress dowager appeals to him to lead the imperial armies in a counterattack. He does so, and with the help of the Japanese god of Sumiyoshi, who suddenly appears at his side in the form of nine identical mounted warriors, vanquishes the previously invincible rebel general Yii-wen Hui, along with seven or eight of his stalwarts. Struck with terror at the sight, the entire rebel army takes to its heels, and the imperial train returns to the capital in triumph. The last rebel holdouts who would stand in the way of their entering the city are routed by Ujitada and his phantom assistants as summarily as the main rebel force had been.

Having escaped what seemed certain death at the hands of the rebels, Ujitada is now anxious to return home to his loved ones in Japan. But soon he becomes involved in a love affair with a mysterious lady who comes and goes like the morning clouds on the mountain peak and who will not tell him who she is or where she lives. His heart is torn between two desires: the desire to set sail for home as soon as possible, and the desire to extend his stay in China until he can learn who the lady is. After considerable delay—this fourth part of the tale, with its several dreamlike meetings between Ujitada and the lady, comprises nearly half the tale’s length—her identity is revealed as none other than the empress dowager herself, who is in fact not an ordinary mortal, but a lady sent from the Second Heaven of Buddhism to rescue China from disaster. The empress dowager tells Ujitada that he, too, was sent from heaven, to assist her in this charge.

In the final part of the tale, Ujitada returns to Japan to be reunited with his family as well as with the magically reincarnated Princess Hua-yang. They begin a happy life together, and before long the princess is with child. Then,

CHAPTER ONE 5 with a hint of stormier times to come as the princess concludes that Ujitada must be seeing another woman (he has been gazing through a magical mirror at the empress dowager, and her perfume has transferred to his robes), the tale breaks off with a note that the rest of the text is missing: “The manuscript states: ‘Here, too, the binding is damaged and the remaining pages have been lost.’”2

The title of the tale presumably derives from events surrounding Ujitada’s departure for China, when his mother arranges for a palace to be built on a mountainside in Matsura (the area of Saga and Nagasaki prefectures in Kyushu today) so that she can await her son’s return at the point in Japan closest to China. The word mya in the title may refer either to the palace or to the lady who goes to live there, but neither plays any significant role in the story; and though it might well have done so, especially in the long fourth part of the tale, the narrative does not develop or draw on the association between Ujitada’s waiting mother and the meaning “to wait” (matsu) embedded in the title.3 We may observe that the title refers to the episode commencing the main action of the tale, but even then we cannot escape the sense that the reference is pointless and the episode undistinguished.+ The puzzle of how and why Massura no

2. Another lacuna is noted some twelve manuscript pages earlier: “The manuscript states: “The binding is damaged and some pages are missing.” The page count is from the Go-Kogon manuscript. See chapter three for details on the manuscripts and texts.

3. Misumi (1975, 1—3) believes the title is indeed intended to suggest the presence of a lady who waits, but does not discuss how, if at all, the narrative draws upon that suggestion, or how the suggestion otherwise affects our understanding of the tale. Arguing by analogy with Hamamatsu Chiinagon monogatari (“Tale of the Hamamatsu Middle Counselor,” ca. 1060s), which in its alternate title of Afitsu no hamamatsu (“Waiting by the Pines on Mitsu Beach” —if matsu is rendered in both of its meanings, “pine” and “to wait”) undoubtedly refers to a lady who waits, Misumi asserts that naming tales after someone other than the hero or heroine was an accepted practice. The analogy misses the mark, however, for Ujitada’s mother has almost no role at all, while Taishé no Kimi in Hamamatsu Chinagon monogatari plays a prominent role throughout. In an apparent attempt to satisfy those who would insist on a title referring to the main character, Misumi also stretches for a way to construe “Matsura no Miya” as an indirect reference to Ujitada himself, but in this, too, the analogy he draws with Mitsu no hamamatsu is too loose to persuade.

Nishiki (1988, 139-40) observes that all of the ladies in Ujitada’s life must wait for him in one sense or another, and, by contrasting the experiences of the various ladies, attempts to find a special thematic place for Ujitada’s mother’s waiting, such as to justify a title reference to her. There are a number of problems with the reading he provides, but most fundamental among them are that he cannot alter the utterly minor role Ujitada’s mother plays in the narrative, and that he does not explain the failure of the narrative itself to draw on an association between her (or any of the other women who wait) and the title.

4. The title seems tangential even by comparison to Utsuko monogatari (“The Tale of the Hollow Tree,” ca. 983), which similarly takes its name from a minor episode early in the tale. The episode involving the hollow tree (see Kono Tama 1959-62, vol. 1, 78ff.; Harada 1969, vol. 1, 47ff.) is one of considerable charm in itself, and it plays an important role not only in introducing the hero of the subsequent action but also in establishing the tale’s most important motif—the magical kim music and its transmission. In fact, the episode may even be considered a thematic encapsulation of the tale as a whole. Nothing of this kind can be said about the events surrounding Ujitada’s departure for China in Matsura no miya monogatari.

miya monogatari gained such an opaque title will probably never be satisfactorily explained.

An intriguing four-part colophon is found at the end of all manuscripts of the tale, and reads as follows (I have added letters to permit easier reference to each of the four entries later):

A. This tale is about events that took place long ago, and, indeed, both the poetry and the language are pleasantly old-fashioned. Beginning with the flight to Mount Shu, however, the text appears to have been revised by some clever fellow of our own age, and contains many unsightly passages. I wonder what the truth of the matter is? And when the grand lady of China speaks of drifting onto the path of dreams—what a charming tale within a tale it is!

[new page]

B. Jogan 3, Fourth Month, Eighteenth Day.

Finished writing [copying] in the western wing of the Somedono Palace.

C. She seems a flower, but is not; she seems a mist, but is not. In the deep of night, she comes; as the heavens dawn, she departs.

She comes like a spring night’s dream, for but a fleeting moment; She departs like a morning cloud, leaving no trace to follow.

{blank page]

D. How true it is! But I am mystified why one so sober of heart—one who preferred not to meet beauty such as can topple cities— would have left such a poem behind him. Or is it that in China a mist like this really does exist?5

The date given in entry B, Jogan 3, corresponds to the year 861 on the Western calendar. If A and B were written by the same hand, that year must be a copy date, and the remarks in A suggest a date of original composition sufficiently earlier that the tale has had time to go through a revision that sets off the language in the early, unrevised part of the tale as distinctly “old fashioned” in contrast to the revised portions, which are written in the idiom of the colophon writers “own age,” ca. 861. The extant version of Matsura no miya monogatari then would be roughly contemporary with Taketori monogatari (“The Tale of the

5. The manuscript known as “the manuscript in imitation of the copy by Retired Emperor Go- Hanazono” (“Den Go-Hanazono In shinkan mohon”) carries the only variant colophon noted by any of the scholars who have pursued the questions associated with the manuscript tradition. In addition to the entries translated here, it bears the dated signatures of two copyists of later centuries. See Hachisuka 1935, 130-31. The reference to Retired Emperor Go- Hanazono in the traditional name for this manuscript is probably an error for Retired Emperor Go-Kogon. See chapter three.

Chapter One 7

Bamboo Cutter”), which scholars have placed variously between the early ninth century and the mid-tenth century, and which is called the “progenitor of all tales”—monogatari no idekihajime no oya—in the “Eawase” chapter of Genji monogatart.© The original Ur-Matsura no miya monogatari would antedate Taketori monogatari by as much as a century, perhaps even a little more.

Alternatively, if entry A is viewed as a later interpolation, 861 can be considered the date of original composition. In that year, the Somedono Palace was the residence of Chancellor Fujiwara Yoshifusa (804—72),7 and we may imagine that the western wing was occupied by his daughter, Meishi (or Akirakeiko, 829-900), consort of Emperor Montoku (827-58, r. 850-58) and mother of then-reigning Emperor Seiwa (850-80, r. 858-—76).8 The designation of the western wing of the Somedono Palace as the place of writing would therefore suggest that the Somedono Empress (as Meishi was known) or a lady in her service was the author of the manuscript.? We might also imagine that Yoshifusa played a sponsoring role, much as Fujiwara Michinaga (966—1027), in a later century, sponsored the literary endeavors of such women as Murasaki Shikibu and Izumi Shikibu, who were in service to his daughter, Shdshi (or Akiko, 988—1074), consort of Emperor Ichijo (980—1011, r. 986-1011). Because so few writings survive from the ninth and early tenth centuries, this view would not substantially change the historical position of the Ur-Matsura no miya monogatari in early Japanese literature: the only change would be in its historical

6. Efforts to determine the date of Taketori monogatari are complicated by the fact that parts of the tale must have existed in a different form, or perhaps even several different forms, before the tale was written down in its present redaction. Most evidence points to the late ninth century or very early tenth century for the tale as it has come down to the present.

For the relevant passage in Genji monogatari, see Abe et al. 1970-76, vol. 2, 370.

7. After crushing rival families vying with the Fujiwara for power at court, Yoshifusa became the first man not of royal blood to serve as chancellor and as regent to the emperor. He established, decisively, the ascendancy of che Fujiwara that was to continue virtually unchallenged for some two centuries.

9. Another choice might be to associate the task of writing with Meishi’s cousin, Koshi (or Takaiko, 842-910). With Yoshifusa as her adoptive father, Koshi apparently was close to Meishi and spent much time with her. In 866, at the age of twenty-four, she became consort to Meishi’s son, the sixteen-year-old Emperor Seiwa, and later came to be known as “the Empress from the Second Ward.” Her lasting fame, however, comes from a legendary liaison several years earlier—it could well have been around 861, though possibly as much as five years before— with Ariwara Narihira (825-80), undistinguished in rank but one of the most colorful lovers of the Heian court. According to one account, Narihira visited Koshi at Meishi’s residence and on one occasion actually managed to steal her away (lse monogatari | “Tales of Ise,” ca.935], episode 6). When the affair was discovered, Koshi was placed out of Narihira’s reach so as not to further jeopardize Yoshifusa’s plans to send her to court once Emperor Seiwa came of age. Although the historicity of the affair remains uncertain, it was given a lasting place in literature by Kokinsha (or Kokin wakashi, “Collection of Ancient and Modern Poetry,” ca. 905), in the headnotes to poems 632 and 747; and by /se monogatari, in episodes 3—6 (episode 4 is cited in chapter two, p. 30), 26, 29, 65, and 76. For a discussion of the affair and a translation of the account given in {se monogatari, see Helen McCullough 1968, 45—49 and the episodes indicated. Also see brief discussion of colophon entry B on pp. 24-25, below.

relationship to Taketori monogatari—if a particularly early date for that tale were assumed. On the other hand, the historical position of the extant version of the tale, with two of its three books dating from an indeterminate period after 861, would need to be evaluated quite apart from what is said in colophon entry B.

As anyone familiar with the historical development of classical Japanese fiction will immediately recognize upon reading the tale, the suggestion that Matsura no miya monogatari as it is known today might have been completed in or before the year 861 is beyond all credibility. Naive though the tale is in some of its features, it is in the end far too sophisticated a work to have been created in the eighth or ninth century. The reader might wonder momentarily if it perhaps represented a flash of genius far ahead of its time, but he would quickly realize that such a work could not possibly have existed for so many centuries without once being mentioned in any later work. Had the tale really been in circulation since 861 and before, surely it, instead of 7aketori monogatari, would have been the one called progenitor of the genre by the author of Genji monogatari.'°

Indeed, the earliest mention of the tale in any secondary source is found in Mumyozoshi (“Untitled Leaves,” ca. 1201), the first work of criticism to survey and discuss methodically, if impressionistically, the monogatari of the Heian period (794—1185) and the opening years of the Kamakura period (1185—1333). This work has long been placed by modern scholars at the turn of the thirteenth century, between 1196 and 1202; in a relatively recent study, Higuchi Yoshimaro has narrowed the span convincingly to between 1198 and 1201, and with some caution to a period of one year and several months running from 1200 to 1201.11 Scholars agree that the work was written by someone close to Fujiwara Shunzei (1114—1204) and his son Teika (1162-1241), and although some doubt remains, the likeliest candidate is the lady known as Shunzei’s Daughter (ca: 1171-after 1252)—in fact Shunzei’s granddaughter, brought up in her grandparents’ care from a young age. !2

10. Readers familiar with Genji monogatari will recall that the reference to Taketori monogatari as “progenitor of all tales” appears in the context of a picture contest in which illustrations of different tales are pitted one against another (see Abe et al. 1970-76, vol. 2, 370—72; Seidensticker 1976, 311-13). Whether we imagine Matsura no miya monogatari matched in such a contest with Ussuho monogatari, to compare an older and newer treatment of a similar plot and bring out the strengths of each, or matched with Zaketorr monogatari, to compare two older works of very different plots, Matsura no miya monogatari would in either case have to be the one given the distinction of “progenitor”—if any work be given it at all.

It is often emphasized (e.g., Nakano 1969; Akiyama et al. 1975, 107) that the remark about Taketort monogatari as progenitor appears in a contest between a tale of long ago and a tale that is more or less contemporary, and in this context should not be taken as literal truth. Murasaki Shikibu could not have had precise records indicating the dates of composition of the various early tales circulating among court circles in her day. Her characterization of Taketori monogatari is better taken as a critical statement (that the tale was the best—the most accomplished, impressive, and memorable—of the early tales) rather than a historical one (that the tale was necessarily the very first tale).

12. Although some doubt remains, for the sake of convenience I will refer to the author of Mumyoxoshi using feminine pronouns.

CHAPTER ONE 9 The passage of interest in Mumyozoshi reads as follows

As for all those tales that have been appearing one after another in recent years, alchough some of them do seem to improve on earlier tales in style and form, it is not possible to find any that compares to Nezame, Sagoromo, or Hamamatsu. In Ukinami, written by Takanobu, one is moved to a certain extent by the obvious effort and feeling the author has put into the writing, but the style is so awkward that one cannot really enjoy reading the story. The several works written by Minor Captain Teika are nothing but empty forms, lacking any semblance of reality. The only exception is the tale known as Matsura no miya: this tale is in the style of Man'yosha and reminds one of Utsuho; it is beyond anything a simpleminded person could have imagined. 13

Mumyozoshi is generally considered a reliable testament for such matters as are stated unambiguously in its pages,!* and there is no ambiguity here with regard to the approximate date its author assigns the tale. The passage cited, speaking of numerous tales from “recent years” (Aonogoro idekitaru mono amata), is itself part of a more extended discussion of tales belonging to the late Heian period. “Recent” is an elastic term, of course, but even in a survey of tales spanning several centuries, we would not normally expect it to extend beyond four or five decades. The other tale mentioned here, Ukinami by Fujiwara ‘Takanobu (1142-1205, half brother of Teika), has been dated between 1160 and 1180.15 The absence of a consistent chronological order for the tales discussed before and after this passage prevents the immediate conclusion that Matsura no miya monogatari must have appeared after these dates, but they can be taken as a rough measure of what the author of Mumyozoshi meant by “recent years.” In any case, it is quite clear that she lends no credence to the colophon date, for if she did, she would have discussed the tale in an altogether different section of her work, and would more likely have noted that Ussuho monogatari (“Tale of the Hollow Tree,” ca. 983) was reminiscent of Matsura no miya

13. Or, “it seems beyond the comprehension of a simpleminded reader (like myself).” Likewise, “nothing but empty forms” (keshiki-bakari) could be “nothing but atmosphere or flights of fancy,” or perhaps merely a reference to the brevity and insubstantiality of the stories.

The two ascriptions, zsukuritaru tote, might more literally be rendered as “said to be written by,” which potentially suggests uncertainty; but zoze, like the modern zo iz, often indicates indirect knowledge on which no doubt is to be cast, and given the discussion below a straightforward “written by” seems to be the appropriate translation here.

There are other ambiguities in the passage as well. See Kuwabara 1976, 97-98; Tomikura 1954, 248-50; Marra 1984, 418; Keene 1989, 4. Translations are my own unless indicated otherwise.

14, For example, to my knowledge no one has questioned the accuracy of the ascription of Usinami to Fujiwara Takanobu. Ishida (1940, 66-67) expresses his confidence in Mumyozoshi directly, saying it “does not seem at present to be a document whose word must be doubted.” Additionally, the general consensus that Mumydzdshi was written by Shunzei’s Daughter or by someone else close to Teika and Shunzei tends to instill even greater faith in its reliability for matters relating to Teika, since for such matters, presumably, whoever the author may be, his or her remarks are based on firsthand knowledge.

monogatari rather than vice versa. At the same time, the very mention of Matsura no miya monogatari in this passage provides the same terminus ad quem for the tale as for Mumydzdshi, namely 1201.16

In searching through the sources scholars have found no further secondary reference explicitly addressing the date of the tale, but the absence of any other secondary mention before Mumyozoshi, coupled with the relatively regular (though not frequent) mention thereafter, serves as strong corroboration of a late twelfth-century date. The date is also supported by a wide assortment of internal evidence, ranging from individual linguistic items, to general historical features, to poetic allusions and other textual precedents dating from the tenth through twelfth centuries.!7 So as not to overburden and delay the present discussion, a summary of all such evidence pertaining to the date of the tale has been placed in appendix A.